Glenn - Urologic Surgery 5th ed

Glenn's Urologic Surgery 5th edition (September 15, 1998): By by Sam D. Graham (Editor), James F. Glenn (Editor) By Lippincott Williams & Wilkins Publishers

By OkDoKeY

Glenn's Urologic Surgery
CONTENTS
Editors Contributors Preface Acknowledgments

Section I: Adrenal
Thomas E. Keane
Chapter 1 Cushing’s Disease and Syndrome David P. O’Brien III Chapter 2 Adrenal Adenoma and Carcinoma Muta M. Issa and Thomas E. Keane Chapter 3 Primary Aldosteronism James F. Glenn Chapter 4 Pheochromocytoma Thomas E. Keane, Pablo J. Santamaria, and Muta M. Issa

Section II: Kidney
Jerome P. Richie
Chapter 5 Simple Nephrectomy W. Holt Sanders and Cragin Anderson Chapter 6 Partial Nephrectomy Andrew C. Novick Chapter 7 Radical Nephrectomy Michael S. Cookson Chapter 8 Intracaval Tumors Thomas J. Polascik and Fray F. Marshall Chapter 9 Transplant Nephrectomy J. Thomas Rosenthal Chapter 10 Renovascular Disease John A. Libertino Chapter 11 Anatrophic Nephrolithotomy Michael L. Paik and Martin I. Resnick Chapter 12 Renal and Retroperitoneal Abscesses J. Quentin Clemens and Anthony J. Schaeffer Chapter 13 Renal Trauma Allen F. Morey and Jack W. McAninch Chapter 14 Renal Allotransplantation Bruce A. Lucas Chapter 15 Ureteral Complications Following Renal Transplantation Rodney J. Taylor Chapter 16 Renal Autotransplantation Philip Ayvazian and Mani Menon

Section III: Ureter and Pelvis
Charles B. Brendler
Chapter 17 Nephroureterectomy Gary D. Steinberg Chapter 18 Pyelolithotomy John M. Fitzpatrick Chapter 19 Ureterolithotomy Michael Marberger Chapter 20 Ureteral Reconstruction Bernd J. Schmitz-Dräger and Rolf Ackermann Chapter 21 Ureteral Stricture Glenn S. Gerber

Section IV: Bladder
James E. Montie
Chapter 22 Simple and Partial Cystectomy Paul LaFontaine and John A. Petros Chapter 23 Radical Cystectomy in Men Mohamed A. Ghonheim Chapter 24 Radical Cystectomy in Women James E. Montie Chapter 25 Bladder Diverticulectomy J. M. Gil-Vernet Chapter 26 Bladder Augmentation R. Duane Cespedes and Edward J. McGuire Chapter 27 Vesicovaginal Fistula

Hubert G. W. Frohmüller Chapter 28 Vesicoenteric Fistula Luis Gonzalez-Serva Chapter 29 Vesical Trauma and Hemorrhage Badrinath R. Konety, Michael P. Federle, and Robert R. Bahnson Chapter 30 Interstitial Cystitis Gary J. Faerber

Section V: Prostate
Joseph A. Smith, Jr.
Chapter 31 Open Prostatectomy Ray E. Stutzman Chapter 32 Pelvic Lymphadenectomy Ralph W. deVere White and Andrew Huang Chapter 33 Radical Retropubic Prostatectomy Joseph A. Smith, Jr. Chapter 34 Radical Perineal Prostatectomy Sam D. Graham, Jr. Chapter 35 Brachytherapy for Localized Prostate Cancer Haakon Ragde Chapter 36 Prostatic Ultrasound and Needle Biopsy Johan Braeckman and Louis J. Denis

Section VI: Urethra
Shlomo Raz
Chapter 37 Stamey and Gittes Bladder Neck Suspension David A. Ginsberg, Eric S. Rovner, and Shlomo Raz Chapter 38 Abdominal Approaches to Surgery for Female Incontinence E. P. Arnold and Peter Gilling Chapter 39 Anterior Vaginal Wall Sling Lynn Stothers Chapter 40 Pubovaginal Fascial Slings R. Duane Cespedes and Edward J. McGuire Chapter 41 Injections for Incontinence in Women and Men Rodney A. Appell and Randy A. Fralick Chapter 42 Pelvic Floor Relaxation David A. Ginsberg, and Eric S. Rovner, and Shlomo Raz Chapter 43 Rectus Muscle Sling Procedure for Severe Stress Urinary Incontinence Niall T. M. Galloway Chapter 44 Cystocele Eric S. Rovner,David A. Ginsberg, and Shlomo Raz Chapter 45 Transvaginal Enterocele Repair Victor W. Nitti Chapter 46 Vaginal Hysterectomy Eric S. Rovner, David A. Ginsberg, and Shlomo Raz Chapter 47 Vaginal Repair of Vesicovaginal Fistula David A. Ginsberg, Eric S. Rovner, and Shlomo Raz Chapter 48 Female Urethral Diverticula Kumaresan Ganabathi and Gary E. Leach Chapter 49 Closure of Bladder Neck in the Male and Female Scott E. Litwiller and Philippe E. Zimmern Chapter 50 Reconstruction of the Severely Damaged Female Urethra Jerry G. Blaivas Chapter 51 Urethral Stricture and Disruption E. James Wright and George D. Webster Chapter 52 Surgery for Urethral Trauma David M. Nudell, Allen F. Morey, and Jack W. McAninch Chapter 53 Artificial Genitourinary Sphincter Implantation James A. Dugan and David M. Barrett Chapter 54 Urethral Cancer in Women John Naitoh, William J. Aronson, and Jean B. DeKernion Chapter 55 Carcinoma of the Male Urethra William J. Aronson, John Naitoh, and Jean B. deKernion

Section VII: Vas Deferens and Seminal Vesicle
Jon L. Pryor
Chapter 56 Seminal Vesicle and Ejaculatory Duct Surgery Paul J. Turek Chapter 57 Vasectomy Jon L. Pryor and Douglas A. Schow Chapter 58 Vasoepididymostomy Anthony J. Thomas, Jr. Chapter 59 Vasovasostomy William Forbes Hendry Chapter 60 Varicocele Alain Jardin

Section VIII: Testes
David A. Swanson

Chapter 61 Simple Orchiectomy Sherri M. Donat Chapter 62 Inguinal Orchiectomy David A. Swanson Chapter 63 Retroperitoneal Lymphadenectomy Michael A. S. Jewett Chapter 64 Torsion of the Testicle Giovanni Grechi and Vincenzo Li Marzi Chapter 65 Scrotal Trauma and Reconstruction Gerald H. Jordan

Section IX: Penis and Scrotum
Tom F. Lue
Chapter 66 Penectomy for Invasive Squamous Cell Carcinoma of the Penis James L. Mohler and John A. Freeman Chapter 67 Inguinal Lymphadenectomy for Penile Carcinoma John A. Freeman and James L. Mohler Chapter 68 Peyronie’s Disease Kenneth S. Nitahara and Tom F. Lue Chapter 69 Priapism Kenneth S. Nitahara and Tom F. Lue Chapter 70 Penile Prosthesis John J. Mulcahy Chapter 71 Penile Venous Surgery Mark R. Licht and Ronald W. Lewis Chapter 72 Penile Arterial Reconstruction (Penile Revascularization) John Mulhall and Irwin Goldstein Chapter 73 Penile Trauma David M. Nudell, Allen F. Morey, and Jack W. McAninch Chapter 74 Penile Replantation Farhad Parivar, Allen F. Morey, and Jack W. McAninch Chapter 75 Hydrocele and Spermatocele Theodros Yohannes and James I. Harty

Section X: Urinary Diversion
George D. Webster
Chapter 76 Ureterosigmoidostomy and the Mainz Pouch II Margit Fisch and Rudolf Hohenfellner Chapter 77 Conduit Urinary Diversion Lesley K. Carr and George D. Webster Chapter 78 Kock Pouch Continent Urinary Diversion John A. Freeman Chapter 79 Right Colon Reservoir Jorge L. Lockhart Chapter 80 Mitrofanoff Continent Urinary Diversion Hubertus Riedmiller and Elmar Werner Gerharz Chapter 81 Orthotopic Urinary Diversion Using an Ileal Low-Pressure Reservoir with an Afferent Tubular Segment Hansjörg Danuser and Urs E. Studer Chapter 82 Orthotopic Urinary Diversion Using a Colonic Segment Daniela Schultz-Lampel and Joachim W. Thüroff Chapter 83 Orthotopic Bladder Replacement in Women John P. Stein and Donald G. Skinner

Section XI: Pediatric Urology
Edmond T. Gonzales, Jr., and Stephen A. Kramer
Chapter 84 Neuroblastoma Yves L. Homsy and Paul F. Austin Chapter 85 Wilms’ Tumor O. Lenayne Westney and Michael L. Ritchey Chapter 86 Renal Fusion and Ectopia Ross M. Decter Chapter 87 Transureteroureterostomy Anthony J. Casale Chapter 88 Pyeloplasty Eugene Minevich and Jeffrey Wacksman Chapter 89 Megaureter Edmond T. Gonzales, Jr. Chapter 90 Triad Syndrome David B. Joseph Chapter 91 Supravesical Urinary Diversions Byron Joyner and Antoine Khoury Chapter 92 Surgery for Childhood Rhabdomyosarcoma Bruce Broecker Chapter 93 Vesicoureteral Reflux Anthony Atala and R. Dixon Walker Chapter 94 Ureterocele Sami Arap and Amilcar Martins Giron

Chapter 95 Urachal Anomalies and Related Umbilical Disorders H. Norman Noe Chapter 96 Vesical Neck Reconstruction Stuart B. Bauer Chapter 97 Considerations in Pediatric Endoscopy Scott A. Berkman and Edwin A. Smith Chapter 98 Pediatric Urethral Diverticulum Hrair-George O. Mesrobian Chapter 99 Posterior Urethral Valves Alan B. Retik Chapter 100 Megalourethra Mark P. Cain Chapter 101 Hypospadias Warren Snodgrass Chapter 102 Exstrophy and Epispadias Stephen A. Kramer Chapter 103 Congenital Anomalies of the Scrotum David R. Roth Chapter 104 Cryptorchidism and Pediatric Hydrocele/Hernia Stanley J. Kogan and Bhagwant Gill Chapter 105 Imperforate Anus and Cloacal Malformations W. Hardy Hendren III Chapter 106 Ambiguous Genitalia Richard I. Silver and John P. Gearhart Chapter 107 Pediatric Vesical Diversion Steven J. Skoog Chapter 108 Urinary Undiversion Andrew L. Freedman and Ricardo Gonzalez Chapter 109 Circumcision Jay B. Levy and Stephen A. Kramer

Section XII: Endoscopy
Culley C. Carson III
Chapter 110 Cystoscopic Stone Basket Extraction Stevan B. Streem Chapter 111 Cystoscopic Treatment of Bladder Tumors Keith A. Harmon and Michael J. Droller Chapter 112 Transurethral Resection, Incision, and Ablation of the Prostate Richard P. Santarosa, Alexis E. Te, and Steven A. Kaplan Chapter 113 Endoscopic Laser Surgery Ken Koshiba and Toyoaki Uchida Chapter 114 Ureteroscopy Anuar Ibrahim Mitre, Jose Luis Chambo, and Sami Arap Chapter 115 Percutaneous Lithotomy Joseph W. Segura Chapter 116 Endopyelotomy for Ureteral Pelvic Junction Obstruction Culley C. Carson III Chapter 117 Endoscopic Ablation of Upper Urinary Tract Tumors Mantu Gupta and Arthur D. Smith Chapter 118 Internal Urethrotomy Joseph M. Khoury Chapter 119 Transurethral Cystolitholapaxy Marshall L. Stoller and Donald L. Gentle Chapter 120 Extracorporeal Shock Wave Lithotripsy James E. Lingeman and John W. Dushinski Chapter 121 Ureteral Stents and Endoscopic Treatment of Ureteral Obstruction Gerhard J. Fuchs, Kamil Noordin, and Anup Patel

Section XIII: Laparoscopy
Leonard G. Gomella
Chapter 122 Basic Laparoscopy: Transperitoneal and Extraperitoneal Approaches Leonard G. Gomella and David M. Albala Chapter 123 Laparoscopic Pelvic Lymph Node Dissection: Transperitoneal and Extraperitoneal Techniques Blake D. Hamilton and Howard N. Winfield Chapter 124 Laparoscopic Varix Ligation James F. Donovan and Eduardo Sanchez de Badajoz Chapter 125 Transperitoneal Laparoscopic Nephrectomy and Nephroureterectomy Inderbir S. Gill and Sakti Das Chapter 126 Retroperitoneoscopic Nephrectomy and Nephroureterectomy Inderbir S. Gill and Sakti Das Chapter 127 Laparoscopic Retroperitoneal Renal Procedures John B. Adams II Chapter 128 Laparoscopic Pyeloplasty Robert G. Moore and Jeffrey A. Cadeddu Chapter 129 Laparoscopic Bladder Neck Suspension J. Stuart Wolf, Jr. and Elspeth M. McDougall Chapter 130 Laparoscopic Management of Lymphoceles

Blake D. Hamilton and Howard N. Winfield Chapter 131 Laparoscopic Management of the Impalpable Undescended Testicle Gerald H. Jordan Chapter 132 Laparoscopic Adrenalectomy H. Tazaki Chapter 133 Renal Cysts Michael P. O’Leary

Section XIV: Frontiers
R. Ernest Sosa
Chapter 134 Robotics, Telepresence, and Virtual Reality in Urologic Surgery Roland N. Chen and Louis R. Kavoussi Chapter 135 Cryosurgical Ablation of the Prostate Harry S. Clarke Chapter 136 Transurethral Microwave Thermotherapy Aaron P. Perlmutter Chapter 137 Interstitial Laser Therapy of Benign Prostatic Hyperplasia Rolf Muschter Color Plate

Editors
Editor-in-Chief Sam D. Graham, Jr., M.D. Louis McDonald Orr Professor of Urology Emory University School of Medicine The Emory Clinic 1365 Clifton Road NE Atlanta, Georgia 30322 Consultant Editor James F. Glenn, M.D. Professor of Surgery (Urology) University of Kentucky College of Medicine P.O. Box 1390 Lexington, Kentucky 40536 Associate Editors Charles B. Brendler M.D. Professor and Chief Section of Urology, MC6038 University of Chicago Pritzker School of Medicine University of Chicago Medical Center 5841 South Maryland Avenue Chicago, Illinois 60637 Section: Ureter and Pelvis Culley C. Carson III, M.D. Professor and Chief Division of Urology University of North Carolina School of Medicine Division of Urology 427 Burnett-Womack CB7235 Chapel Hill, North Carolina 27599 Section: Endoscopy Leonard G. Gomella, M.D. The Bernard W. Godwin, Jr. Associate Professor of Prostate Cancer Department of Urology Thomas Jefferson University School of Medicine 1025 Walnut Street, Suite 1102 Philadelphia, Pennsylvania 19107 Section: Laparoscopy Edward T. Gonzales, Jr., M.D. Chief of Pediatric Urology Department of Urology Baylor College of Medicine Scott Feigin Center, Suite 270 6621 Fannin Street Houston, Texas 77030-2399 Section: Pediatric Urology Thomas E. Keane, M.D. Associate Professor of Urology Division of Urology Emory University 1365 Clifton Road NE Atlanta, Georgia 30322 Section: Adrenal Stephen A. Kramer, M.D. Head, Section of Pediatric Urology Urology Professor in Honor of Dr. Utz Mayo Clinic 200 First Street SW Rochester, Minnesota 55905 Section: Pediatric Urology Tom F. Lue, M.D. Professor of Urology Department of Urology, U-575 University of California San Francisco, Box 0738 San Francisco, California 94143-0738 Section: Penis and Scrotum James E. Montie, M.D. Interim Head Section of Urology George F. and Nancy P. Valassis Professor of Urologic Oncology University of Michigan Medical Center 1500 East Medical Center Drive Ann Arbor, Michigan 48109-0330 Section: Bladder Jon L. Pryor, M.D. Associate Professor and Director Center for Men’s Health and Infertility Departments of Urologic Surgery, Cell Biology and Neuroanatomy, and Obstetrics and Gynecology

University of Minnesota P.O. Box 394 UMHC 420 Delaware Street SE Minneapolis, Minnesota 55455 Section: Vas Deferens and Seminal Vesicle Shlomo Raz, M.D. Professor of Surgery Department of Urology U.C.L.A. School of Medicine Box 951738 Los Angeles, California 90024 Section: Urethra Jerome P. Richie, M.D. Elliot C. Cutler Professor of Surgery Chair, Department of Urology Brigham and Women’s Hospital Harvard Medical School 45 Francis Street, ASB II-3 Boston, Massachusetts 02115 Section: Kidney Joseph A. Smith, Jr., M.D. Professor and Chairman Department of Urologic Surgery Vanderbilt University Medical Center A-1302 Medical Center North Nashville, Tennessee 37232-2765 Section: Prostate R. Ernest Sosa, M.D. Associate Professor of Surgery/Urology New York Hospital—Cornell Medical Center 525 East 68th Street, Box 94 New York, New York, 10021 Section: Frontiers David A. Swanson, M.D., F.A.C.S. Professor and Chairman (Ad Interim) Department of Urology The University of Texas M.D. Anderson Cancer Center 1515 Holcombe Boulevard, Box 110 Houston, Texas 77030 Section: Testes George D. Webster, M.B., F.R.C.S. Professor of Urology Surgery Department of Surgery Division of Urology Duke University Medical Center Box 3146 Durham, North Carolina 27710 Section: Urinary Diversion

Contributors
Rolf Ackermann, M.D. Professor of Urology and Medicine Department of Urology Heinrich-Heine-University Moorenstrasse 5 D-40225 Düsseldorf, Germany Chapter 20 John B. Adams II, M.D. Assistant Professor of Surgery Head of Endourology and Laparoscopy Department of Surgery Section of Urology Medical College of Georgia 1120 15th Street Augusta, Georgia 30912-4050 Chapter 127 David M. Albala, M.D. Associate Professor of Urology Department of Urology Loyola University Medical Center 21065 First Avenue Maywood, Illinois 60153, and Attending Urologist Hines Veteran’s Hospital Hines, Illinois Chapter 122 Cragin Anderson, M.D. Atlanta Urological Institute 217 Upper Riverdale Road Riverdale, Georgia 30274 Chapter 5 Rodney A. Appell, M.D. Head Section of Voiding Dysfunction and Female Urology Department of Urology The Cleveland Clinic Foundation 9500 Euclid Avenue Cleveland, Ohio 44195 Chapter 41 Sami Arap, M.D. Professor and Chairman Division of Urology Hospital das Clínicas University of Sao Paulo School of Medicine Caixa Postal 11-273-9 CEP 05422-970 Sao Paulo, SP, Brazil Chapters 94 and 114 E. P. Arnold, M.B., Ch.B. (NZ), Ph.D. (U.K.), F.R.C.S. (U.K.), F.R.A.C.S. Associate Professor of Urology Department of Urology Christchurch Hospital Riccarton Avenue Christchurch, New Zealand Chapter 38 William J. Aronson, M.D. Assistant Clinical Professor Department of Urology U.C.L.A. School of Medicine Box 951738 Los Angeles, California 90095-1738 Chapters 54 and 55 Anthony Atala, M.D. Assistant Professor of Surgery Department of Surgery Harvard Medical School, and Assistant in Urology Department of Urology Children’s Hospital 300 Longwood Avenue Boston, Massachusetts 02115 Chapter 93 Paul F. Austin, M.D. Fellow in Pediatric Urology Indiana University School of Medicine James Whitcomb Riley Hospital for Children Indiana University Medical Center 702 Barnhill Drive, Suite 1739 Indianapolis, Indiana 46202 Chapter 84 Philip Ayvazian, M.D. Department of Urology University of Massachusetts

55 Lake Avenue North Worcester, Massachusetts 01655 Chapter 16 Robert R. Bahnson, M.D. Louis Levy Professor and Director Division of Urology Ohio State University Medical Center 456 West 10th Avenue Columbus, Ohio 43210-1228 Chapter 29 David M. Barrett, M.D. Anson L. Clark Professor of Urology Department of Urology Mayo Clinic 200 First Street SW Rochester, Minnesota 55905 Chapter 53 Stuart B. Bauer, M.D. Associate Professor of Urology Department of Urology Harvard Medical School, and Department of Urology The Children’s Hospital 300 Longwood Avenue Boston, Massachusetts 02115 Chapter 96 Scott A. Berkman, M.D. Department of Surgery Emory University School of Medicine Egleston Children’s Hospital Atlanta, Georgia 30322 Chapter 97 Jerry G. Blaivas, M.D. Department of Urology New York Hospital-Cornell Medical Center 400 East 56th Street New York, New York 10022 Chapter 50 Johan Braeckman, M.D. Assistant Head Department of Urology Vrije Universiteit Brussels Laarbeeklaan 101 1090 Brussels, Belgium Chapter 36 Bruce Broecker, M.D. Department of Surgery Medical College of Virginia Virginia Commonwealth University Richmond, Virginia 23298 Chapter 92 Jeffrey A. Cadeddu, M.D. Resident Department of Urology Johns Hopkins Hospital 600 North Wolfe St. Baltimore, Maryland 21287 Chapter 128 Mark P. Cain, M.D. Department of Pediatric Urology Indiana University School of Medicine Indianapolis, Indiana 46202 Chapter 100 Lesley K. Carr Department of Surgery Division of Urology University of Toronto Wellesley Hospital Toronto M4Y 1J3 Ontario, Canada Chapter 77 Culley C. Carson III, M.D. Professor and Chief Division of Urology University of North Carolina School of Medicine 427 Burnett-Womack CB7235 Chapel Hill, North Carolina 27599 Chapter 116 Anthony J. Casale, M.D. Associate Professor

Department of Urology Indiana University School of Medicine, and Pediatric Urology Division James Whitcomb Riley Hospital for Children 702 Barnhill Drive, Room 1739 Indianapolis, Indiana 46202 Chapter 87 R. Duane Cespedes, M.D. Department of Urology/MKCU Wilford Hall Medical Center 2200 Bergquist Drive, Suite I Lackland AFB, Texas 78236 Chapters 26 and 40 Jose Luis Chambo, M.D. Assistant Attendent Division of Urology Hospital das Clínicas University of Sao Paulo School of Medicine Caixa Postal 11-273-9 CEP 05422-970 Sao Paulo, SP, Brazil Chapter 114 Roland N. Chen, M.D. Department of Urology The Cleveland Clinic Florida 3000 West Cypress Creek Road Fort Lauderdale, Florida 33309 Chapter 134 Harry S. Clarke, M.D. Division of Urology Emory University School of Medicine Atlanta, Georgia 30322 Chapter 135 J. Quentin Clemens M.D. Department of Urology Northwestern University Medical School 303 East Chicago Avenue, T299 Chicago, Illinois 60611-3008 Chapter 12 Michael S. Cookson, M.D. Division of Urology Department of Surgery University of Kentucky 800 Rose Street, MS 273 Lexington, Kentucky 40536 Chapter 7 Hansjorg Danuser, M.D. Department of Urology University of Berne 3010 Berne, Switzerland Chapter 81 Sakti Das, M.B.B.S., F.R.C.S. Chief Department of Urology Kaiser Permanente Medical Center Associate Professor of Urology University of California, Davis School of Medicine Walnut Creek, California 94596 Chapters 125 and 126 Ross M. Decter, M.D. Associate Professor Department of Surgery Section of Urology Pennsylvania State Geisinger Health System 500 University Drive Hershey, Pennsylvania 17033 Chapter 86 Jean B. deKernion, M.D. Department of Urology U.C.L.A. School of Medicine Room 66-133 CHS, 10833 LeConte Avenue Los Angeles, California 90095-1738 Chapters 54 and 55 Louis J. Denis, M.D. Professor and Director Department of Urology Oncology Centre Antwerp Lindendreef 1 2020 Antwerp, Belgium Chapter 36 Ralph W. deVere White, M.D. Professor and Chair

Department of Urology University of California, Davis School of Medicine 4301 X Street Suite 2210 Sacramento, California 95817 Chapter 32 Sherri M. Donat, M.D. Assistant Attending Surgeon Department of Surgery Division of Urology Memorial Sloan-Kettering Cancer Center 1275 York Avenue, Room M-606 New York, New York 10021 Chapter 61 James F. Donovan, Jr., M.D. Professor of Urology University of Oklahoma Health Science C 920 Stanton L. Young Boulevard, WP3150 Oklahoma City, Oklahoma 73190 Chapter 124 Michael J. Droller, M.D. Professor and Chair Department of Urology The Mount Sinai Medical Center One Gustave L. Levy Place, Box 1272 New York, New York 10029 Chapter 111 James A. Dugan, M.D. Department of Urology Mayo Clinic 200 First Street SW Rochester, Minnesota 55905 Chapter 53 John W. Dushinski, M.D. Assistant Clinical Professor Department of Surgery University of Calgary, and Rockyview Professional Center 1011 Glenmore Trail, SW Calgary, Alberta T2V 4R6, Canada Chapter 120 Gary J. Faerber, M.D. Assistant Professor of Surgery Department of Urology University of Michigan 1500 East Medical Center Drive Ann Arbor, Michigan 48109-0330 Chapter 30 Michael P. Federle, M.D. Professor of Radiology Department of Radiology University of Pittsburgh Medical Center Presbyterian University Hospital Pittsburgh, Pennsylvania 15213 Chapter 29 Margit Fisch, M.D. Associate Professor of Urology Department of Urology Johannes Gutenberg University Medical School Langenbeckstrasse 1 55131 Mainz, Germany Chapter 76 John M. Fitzpatrick, M.Ch., F.R.C.S.I. Professor of Surgery Department of Surgery/Urology Mater Hospital and University College Dublin 47 Eccles Street Dublin 7, Ireland Chapter 18 Randy A. Fralick, M.D. Fellow Section of Voiding Dysfunction and Female Urology Department of Urology The Cleveland Clinic Foundation Cleveland, Ohio 44195, and Holy Family Memorial Medical Center 1020 Maritime Drive Manitowoc, Wisconsin 54220 Chapter 41 Andrew L. Freedman, M.D.

Pan-Pacific Pediatric Urologic Institute Los Angeles, California 90095 Chapter 108 John A. Freeman, M.D. Assistant Professor of Urology Division of Urology University of North Carolina School of Medicine 428 Burnett Womack Building CB 7235 Chapel Hill, North Carolina 27599-7235 Chapters 66, 67, and 78 Hubert G. W. Frohmüller, M.D., M.S., F.A.C.S. Emeritus Professor of Urology Department of Urology University of Würzburg School of Medicine Josef Schneidederstrasse 2 Würzburg D-97080, Germany Chapter 27 Gerhard J. Fuchs, M.D. Professor of Urology Section Chief Endourology, Laparascopic Surgery, and Stone Disease Director, Stone Treatment Center Box 951738, Room BU-183 U.C.L.A. Medical Center Los Angeles, California 90095-1738 Chapter 121 Niall T. M. Galloway, M.B., F.R.C.S., F.R.S.C.(E) Associate Professor of Surgery/Urology Department of Surgery Section of Urology Emory University School of Medicine The Emory Clinic 1364 Clifton Road, NE Atlanta, Georgia 30322 Chapter 43 Kumaresan Ganabathi, M.B., B.S., M.S. (Gen Surg), Dip. Urol. (London), F.R.C.S. (E) Brookville, Punxsutawney, and Clarion Hospitals 240 Allegheny Boulevard Suite C Brookville, Pennsylvania 15825 Chapter 48 John P. Gearhart, M.D. Professor and Director of Pediatric Urology Department of Pediatric Urology Johns Hopkins University School of Medicine The James Buchanan Brady Urological Institute Marburg, 149 600 North Wolfe Street Baltimore, Maryland 21287-2101 Chapter 106 Donald L. Gentle, M.D. Department of Urology University of California San Francisco Urology Clinic 400 Parnassus Avenue Room A605 San Francisco, California 94143 Chapter 119 Glenn S. Gerber, M.D. Associate Professor Department of Surgery The University of Chicago Hospitals 5841 South Maryland Avenue, MC 6038 Chicago, Illinois 60637 Chapter 21 Elmar Werner Gerharz, M.D. Department of Urology Julius Maximilians-University Medical School 97080 Wurzburg, Germany, and The Institute of Urology and Nephrology University College London Medical School London W1P 7PN, United Kingdom Chapter 80 Mohamed A. Ghoneim, M.D. Professor of Urology Director, Urology and Nephrology Center Mansoura, Egypt Chapter 23 J. M. Gil-Vernet, M.D. Catedra de Urologia Facultad de Medicina

C. Casanova 143 Urologia 08036 Barcelona, Spain Chapter 25 Bhagwant Gill, M.D. Eastchester Professional Center 1695 Eastchester Road Suite 501 Bronx, New York 10461-2330 Chapter 104 Inderbir S. Gill, M.D., MCh Head Section of Laparoscopic and Minimally Invasive Surgery Department of Urology/A100 The Cleveland Clinic Foundation 9500 Euclid Avenue Cleveland, Ohio 44195 Chapters 125 and 126 Peter J. Gilling Consultant Urologist Promed House P.O. Box 56 Tauranga, New Zealand Chapter 38 David A. Ginsberg, M.D. Assistant Professor of Urology Department of Urology University of Southern California/Norris Cancer Center 1441 Eastlake Avenue, Suite 7414 Los Angeles, California 90033 Chapters 37, 42, 44, 46, and 47 Amilcar Martins Giron, M.D. Associate Professor Division of Urology Hospital das Clínicas University of Sao Paulo School of Medicine Caixa Postal 11-273-9 CEP 05422-970 Sao Paulo, SP, Brazil Chapter 94 James F. Glenn, M.D. Professor of Surgery Department of Surgery Division of Urology University of Kentucky Medical Center 800 Rose Street Lexington, Kentucky 40536 Chapter 3 Irwin Goldstein, M.D. Boston Medical Center Boston, Massachusetts 02118 Chapter 72 Leonard G. Gomella, M.D. The Bernard W. Godwin, Jr. Associate Professor of Prostate Cancer Department of Urology Thomas Jefferson University School of Medicine 1025 Walnut Street, Suite 1102 Philadelphia, Pennsylvania 19107 Chapter 122 Edmond T. Gonzales, Jr., M.D. Chief of Pediatric Urology Department of Urology Baylor College of Medicine Scott Feigin Center, Suite 270 6621 Fannin Street Houston, Texas 77030-2399 Chapter 89 Ricardo Gonzalez, M.D. Professor and Chief Department of Pediatric Urology Children’s Hospital of Michigan Wayne State University School of Medicine 3901 Beaubien Boulevard Detroit, Michigan 48201 Chapter 108 Luis Gonzalez-Serva, M.D. Jet International M-211 P.O. Box 0200-11 Miami, Florida 33102 Chapter 28 Sam D. Graham, Jr., M.D. Louis McDonald Orr Professor of Urology

Department of Urology Emory University School of Medicine The Emory Clinic 1365 Clifton Road NE Atlanta, Georgia 30322 Chapter 34 Giovanni Grechi, M.D. Professor of Urology Department of Urology Clinica Urologica II Università degli Studi di Firenze Viale Pieraccini, 18 50139 Firenze, Italy Chapter 64 (Deceased) Mantu Gupta, M.D. Department of Urology Columbia-Presbyterian Medical Center New York, New York 10032 Chapter 117 Blake D. Hamilton, M.D. Division of Urology University of Utah Salt Lake City, Utah 84132 Chapters 123 and 130 Keith A. Harmon, M.D. Department of Urology The Mount Sinai Medical Center One Gustave L. Levy Place, Box 1272 New York, New York 10029 Chapter 111 James I. Harty Division of Urology University of Louisville School of Medicine Louisville, Kentucky 40292 Chapter 75 W. Hardy Hendren III, M.D. Department of Surgery Children’s Hospital 300 Longwood Avenue Boston, Massachusetts 02115 Chapter 105 William Forbes Hendry, M.D. Consultant Urologist St. Bartholomew’s Hospital and Royal Marsden Hospital 149 Harley Street London W1N 2DE, United Kingdom Chapter 59 Rudolf Hohenfellner, M.D. Department of Urology Johannes Gutenberg University Medical School Lagenbeckstrasse 1 55131 Mainz, Germany Chapter 76 Yves L. Homsy, M.D., F.R.C.S.C., F.A.A.P. Director Department of Pediatric Urology University of South Florida College of Medicine 2727 West Dr. Martin Luther King Boulevard Tampa, Florida 33607 Chapter 84 Andrew Huang, M.D. Chief Resident Department of Urology University of California, Davis School of Medicine 4301 X Street Suite 2210 Sacramento, California 95817 Chapter 32 Muta M. Issa, M.D., F.A.C.S. Assistant Professor of Urology Department of Surgery Division of Urology Emory University School of Medicine, and Atlanta Veterans Affairs Medical Center 1365 Clifton Road NE Atlanta, Georgia 30322 Chapters 2 and 4 Alain Jardin, M.D. Professor and Chairman

Hopital de Bicêtre Université Paris Sud 78 Rue General Leclerc 94275 Kremlin Bicêtre, France Chapter 60 Michael A. S. Jewett, M.D. Division of Urology University of Toronto 200 Elizabeth Street, EN14-205 Toronto, Ontario M5G 2C4, Canada Chapter 63 Gerald H. Jordan, M.D. Professor of Urology Department of Urology Eastern Virginia Medical School 400 West Brambleton Avenue Suite 100 Norfolk, Virginia 23510 Chapters 65 and 131 David B. Joseph, M.D. Chief of Pediatric Urology Division of Urology Children’s Hospital of Alabama, and Professor Department of Surgery The University of Alabama 1600 7th Avenue, South Birmingham, Alabama 35233 Chapter 90 Byron Joyner, M.D. Division of Urology The Hospital for Sick Children 555 University Avenue Toronto, Ontario M5G 1X8, Canada Chapter 91 Steven A. Kaplan, M.D. Professor and Vice-Chairman Department of Urology Columbia Presbyterian Medical Center 161 Fort Washington Avenue New York, New York 10032 Chapter 112 Louis R. Kavoussi, M.D. Associate Professor of Urology Department of Urology The James Buchanan Brady Urological Institute Chief of Urology Department of Urology Johns Hopkins Bayview Medical Center 4940 Eastern Avenue Baltimore, Maryland 21224 Chapter 134 Thomas E. Keane, M.B., F.R.C.S.I., F.A.C.S. Associate Professor of Urology Section of Urology Emory University School of Medicine 1365 Clifton Road NE Atlanta, Georgia 30322 Chapters 2 and 4 Antoine E. Khoury, M.D. Associate Professor University of Toronto Department of Surgery/Division of Urology The Hospital for Sick Children 555 University Avenue Toronto, Ontario M5G 1X8, Canada Chapter 91 Joseph M. Khoury, M.D. Associate Professor of Urology Medical Director, Urodynamics and Reconstructive Urology Division of Urology University of North Carolina CB 7235 Burnett-Womack Building, Room 428 Chapel Hill, North Carolina 27599 Chapter 118 Stanley J. Kogan, M.D. 311 North Street Suite 310 White Plains, New York 10605 Chapter 104 Badrinath R. Konety, M.D. Chief Resident

Department of Urology University of Pittsburgh 3471 Fifth Avenue, Suite 700 Pittsburgh, Pensylvania 15213 Chapter 29 Ken Koshiba, M.D. Professor and Chairman Department of Urology Kitasato University School of Medicine Kitasato, Sagamihara Kangawa, 228 Japan, and Director Kitasato University Medical Center Kitamoto, Saitama, 364 Japan Chapter 113 Stephen A. Kramer, M.D. Head Section of Pediatric Urology Urology Professor in Honor of Dr. Utz Mayo Clinic 200 First Street SW Rochester, Minnesota 55905 Chapters 102 and 109 Paul LaFontaine, M.D. Section of Urology The Emory Clinic 1365 Clifton Road NE Atlanta, Georgia 30322 Chapter 22 Gary E. Leach, M.D. Director Tower Urology Institute for Continence Cedar Sinai Medical Center, and Department of Urology University of Southern California 8631 West 3rd Street, Suite 915E Los Angeles, California 90048 Chapter 48 Jay B. Levy, M.D. Assistant Clinical Professor Division of Urology University of North Carolina School of Medicine, and Presbyterian Hospital Carolinas Medical Center, and 1718 East Fourth Street Suite 805 Charolotte, North Carolina 28204 Chapter 109 Ronald W. Lewis, M.D. Chairman Section of Urology Medical College of Georgia Room BA-8412 1120 15th Street Augusta, Georgia 30912 Chapter 71 John A. Libertino, M.D. Assistant Clinical Professor of Surgery Department of Surgery, Harvard University Medical School Boston, Massachusetts 02115, and Clinical Professor of Urology Tufts University School of Medicine Lahey Clinic Medical Center 41 Mall Road Burlington, Massachusetts 01805 Chapter 10 Mark R. Licht, M.D. Head Section of Sexual Dysfunction and Prosthetic Surgery Department of Urology The Cleveland Clinic Florida 3000 West Cypress Creek Road Fort Lauderdale, Florida 33309 Chapter 71 Vincenzo Li Marzi, M.D. Department of Urology Clinica Urologica II Università degli Studi di Firenze Viale Pieraccini, 18 50139 Firenze, Italy Chapter 64 James E. Lingeman, M.D. Director of Research Methodist Hospital of Indiana Institute for Kidney Stone Diseases, and Associate Clinical Instructor Department of Urology Indiana University School of Medicine 1801 North Senate Boulevard

Suite 655 Indianapolis, Indiana 46202 Chapter 120 Scott E. Litwiller, M.D. Assistant Professor Department of Urology University of Texas Southwestern Medical Center 5323 Harry Hines Boulevard Dallas, Texas 75235-9110 Chapter 49 Jorge L. Lockhart, M.D. Director Division of Urology Harborside Medical Center Suite 730 4 Columbia Drive Tampa, Florida 33606 Chapter 79 Bruce A. Lucas, M.D. Professor of Surgery Director Kidney Transplant Program Transplant Section, Department of Surgery University of Kentucky Medical Center 800 Rose Street Lexington, Kentucky 40536 Chapter 14 Tom F. Lue, M.D. Professor of Urology Department of Urology, U-575 University of California Box 0738 San Francisco, California 94143-0738 Chapter 68 and 69 Michael Marberger, M.D. Professor and Chairman Department of Urology University of Vienna Wahringer Gurtel 18-20 Vienna A-1090, Austria Chapter 19 Fray F. Marshall, M.D. Professor of Urology and Oncology Department of Urology The Johns Hopkins Medical Institutions 600 North Wolfe Street Baltimore, Maryland 21287-2101 Chapter 8 Jack W. McAninch M.S., M.D. Professor of Urological Surgery Department of Urology University of California, and San Francisco General Hospital 1001 Potrero Avenue, Room 3A20 San Francisco, California 94110 Chapter 13, 52, 73, and 74 Elspeth M. McDougall, M.D. Associate Professor of Urologic Surgery Washington University Medical School 10130 Wohl Clinic 4960 Children’s Place St. Louis, Missouri 63110 Chapter 129 Edward J. McGuire, M.D. Division of Urology University of Texas 6431 Fannin, Suite 6018 Houston, Texas 77030 Chapters 26 and 40 Mani Menon, M.D. Chairman Department of Urology Henry Ford Hospital 2799 West Grand Boulevard Detroit, Michigan 48202 Chapter 16 Hrair-George O. Mesrobian, M.D. Chief Division of Pediatric Urology Medical College of Wisconsin Medical Director, Section of Urology Children’s Hospital of Wisconsin

Suite 3035 8701 Watertown Plank Road Milwaukee, Wisconsin 53226 Chapter 98 Eugene Minevich, M.D. Instructor of Clinical Surgery Department of Surgery University of Cincinnati College of Medicine Division of Pediatric Urology Children’s Hospital Medical Center 3333 Burnet Avenue Cincinnati, Ohio 45229 Chapter 88 Anuar Ibrahim Mitre, M.D. Associate Professor Division of Urology Hospital das Clínicas University of Sao Paulo School of Medicine Caixa Postal 11-273-9 CEP 05422-970 Sao Paulo, SP, Brazil Chapter 114 James L. Mohler, M.D. Associate Professor of Surgery Adjunct Associate Professor of Pathology Division of Urology University of North Carolina School of Medicine CB 7235 Chapel Hill, North Carolina 27599-7235 Chapters 66 and 67 James E. Montie, M.D. George F. and Nancy P. Valassis Professor of Urologic Oncology Interim Head Section of Urology University of Michigan Medical Center 1500 East Medical Center Drive Ann Arbor, Michigan 48109-0330 Chapter 24 Robert G. Moore, M.D. Associate Professor of Surgery St. Louis University, and Department of Surgery Division of Urology St. Louis University Hospital 3635 Vista Avenue at Grand Boulevard P.O. Box 15250 St. Louis, Missouri 63110-0250 Chapter 128 Allen F. Morey, M.D., F.A.C.S. Clinical Assistant Professor of Surgery (Urology) Department of Surgery (Urology Service) Uniformed Services University of the Health Sciences, and Attending Urologist Brooke Army Medical Center Fort Sam Houston, Texas 78258 Chapters 13, 52, 73, and 74 John J. Mulcahy, M.D., Ph.D. Professor of Urology Department of Urology Indiana University Medical Center 1001 West 10th Street Indianapolis, Indiana 46202 Chapter 70 John Mulhall Department of Urology Loyola University Medical Center Maywood, Illinois 60153 Chapter 72 Rolf Muschter, M.D., PhD Associate Professor Department of Urology Grosshadern Hospital of Ludwig-Maximilians University of Munich Marchioninistrasse 15 D-81377 Munich, Germany Chapter 137 John Naitoh, M.D. Clinical Instructor Department of Urology U.C.L.A. School of Medicine Los Angeles, California 90095-1738 Chapters 54 and 55 Kenneth S. Nitahara, M.D.

Department of Urology University of California Box 0738 San Francisco, California 94143-0738 Chapters 68 and 69 Victor W. Nitti, M.D. Assistant Professor Director of Neurology and Female Urology Department of Urology New York University Medical Center 540 First Avenue New York, New York 10016 Chapter 45 H. Norman Noe Department of Urology LeBonheur Children’s Medical Center Memphis, Tennessee 38120 Chapter 95 Kamil Noordin, M.D. Universiti Kebangsaan Malaysia 43600 UKM, Bangi Selangor Darul Ehsan, Malaysia Chapter 121 Andrew C. Novick, M.D. Chairman Department of Urology The Cleveland Clinic Foundation 9500 Euclid Avenue Cleveland, Ohio 44195 Chapter 6 David M. Nudell, M.D. Department of Urology University of California 533 Parnassus Avenue, U-575 San Francisco, California 94110 Chapters 52 and 73 David P. O’Brien III, M.D. Professor of Surgery (Urology) Section of Urology Emory University School of Medicine The Emory Clinic 1365 Clifton Road NE Atlanta, Georgia 30322 Chapter 1 Michael P. O’Leary, M.D. Department of Surgery Division of Urology Harvard University Medical School, and Division of Urology Brigham and Women’s Hospital 45 Francis Street ASB 11-3 Boston, Massashusetts 02115 Chapter 133 Michael L. Paik, M.D. Resident Department of Urology University Hospitals of Cleveland/Case Western Reserve University School of Medicine 11100 Euclid Avenue Cleveland, Ohio 44106 Chapter 11 Farhad Parivar, M.D., F.R.C.S.(Ed) Chief Resident Department of Urology University of Calfornia, San Francisco San Francisco, California 94143, and Kaiser Permanente Medical Center 27400 Hesperian Boulevard Hayward, California 94545 Chapter 74 Anup Patel, M.D., M.S., F.R.C.S. Clinical Instructor in Urology University of California, Los Angeles Los Angeles, California 90095, and 58 Whiteadder Way Clippers Quay London E14 9UR, United Kingdom Chapter 121 Aaron P. Perlmutter, M.D., Ph.D. Director Brady Prostate Center Department of Urology The New York Hospital-Cornell Medical Center

525 East 68th Street New York, New York 10021 Chapter 136 John A. Petros, M.D. Assistant Professor and Director of Urologic Research Department of Surgery Division of Urology Emory University School of Medicine 1365 Clifton Road, NE Atlanta, Georgia 30322 Chapter 22 Thomas J. Polascik, M.D. Instructor in Urology The James Buchanan Brady Urological Institute The Johns Hopkins Medical Institutions 600 North Wolfe Street Baltimore, Maryland 21287 Chapter 8 Jon L. Pryor, M.D., M.S. Associate Professor Departments of Urologic Surgery, Cell Biology and Neuroanatomy, and Obstetrics and Gynecology University of Minnesota P.O. Box 394 UMHC 420 Delaware Street SE Minneapolis, Minnesota 55455, and Director Center for Men’s Health and Infertility Reproductive Health Associates 360 Sherman Street, Suite 160 St. Paul, Minnesota 55102 Chapter 57 Haakon Ragde, M.D. Director Department of Prostate Brachytherapy Urological Services Pacific Northwest Cancer Foundation Northwest Hospital 1560 North 115th Street Seattle, Washington 98133 Chapter 35 Shlomo Raz, M.D. Professor of Surgery Department of Urology U.C.L.A. School of Medicine Box 951738 Los Angeles, California 90024 Chapters 37, 42, 44, 46, and 47 Martin I. Resnick, M.D. Lester Persky Professor and Chairman Department of Urology University Hospitals of Cleveland Case Western Reserve University School of Medicine 11100 Euclid Avenue Cleveland, Ohio 44106 Chapter 11 Alan B. Retik, M.D. Department of Urology Children’s Hospital 300 Longwood Avenue Mailstop HU-216 Boston, Massachusetts 02115 Chapter 99 Hubertus Riedmiller, M.D. Professor of Urology Department of Urology Julius Maximilians-University Medical School Josef Schneider Strasse 2 97080 Wurzburg, Germany Chapter 80 Michael L. Ritchey, M.D. Division of Surgery and Pediatrics University of Texas Medical School 6431 Fannin, Suite 5.258 Houston, Texas 77030 Chapter 85 J. Thomas Rosenthal, M.D. Professor of Urology Department of Urology U.C.L.A. School of Medicine 10833 LeConte Avenue Los Angeles, California 90095-1731 Chapter 9

David R. Roth, M.D. Associate Professor Scott Department of Urology Baylor College of Medicine, and Department of Pediatric Urology Texas Children’s Hospital MC 3-3430 6221 Fannin Street Houston, Texas 77030-2399 Chapter 103 Eric S. Rovner, M.D. Assistant Professor of Urology Division of Urology Department of Surgery Hospital of the University of Pennsylvania 1 Rhoads 3400 Spruce Street Philadelphia, Pennsylvania 19104 Chapters 37, 42, 44, 46, and 47 Eduardo Sanchez de Badajoz, M.D. Profesor Titular de Urologia Universidad de Malaga Strachan 4 29015 Malaga, Spain Chapter 124 W. Holt Sanders, M.D. Assistant Professor Department of Surgery Section of Urology Emory University School of Medicine 1365 Clifton Road NE Atlanta, Georgia 30322 Chapter 5 Pablo J. Santamaria, M.D. Clinical Assistant Professor Section of Urology Emory University School of Medicine Atlanta, Georgia 30322, and Middle Georgia Urology Associates, PC Erin Office Park Suite 11 Dublin, Georgia 31021 Chapter 4 Richard P. Santarosa, M.D. J. Bentley Squier Urological Clinic Columbia University College of Physicians and Surgeons New York, New York 10032 Chapter 112 Anthony J. Schaeffer, M.D. Professor and Chairman Department of Urology Northwestern University Medical School Tarry Building 1-715 303 East Chicago Avenue Chicago, Illinois 60611-3008 Chapter 12 Bernd J. Schmitz-Dräger, M.D., Ph.D. Professor of Urology Department of Urology Heinrich-Heine-University Moorenstrasse 5 D-40225 Düsseldorf, Germany Chapter 20 Douglas Schow, M.D. Department of Urologic Surgery, University of Minnesota 360 Sherman Street, Suite 160 P.O. Box 394 UMHC 420 Delaware Street, SE Minneapolis, Minnesota 55455, and Center for Men’s Health and Infertility Reproductive Health Associates 360 Sherman Street, Suite 160 St. Paul, Minnesota 55102 Chapter 57 Daniela Schultz-Lampel, M.D. Adult Department of Urology and Pediatric Urology Klinikum Wuppertal GmbH University of Witten/Herdecke Medical School Heusnerstrasse 40 Wuppertal, Germany Chapter 82 Joseph W. Segura, M.D.

Professor of Urology Department of Urology Mayo Clinic 200 First Street SW Rochester, Minnesota 55905 Chapter 115 Richard I. Silver, M.D. Assistant Professor of Urology and Pediatrics Department of Pediatric Urology Thomas Jefferson University Philadelphia, Pennsylvania 19107-5083 Chapter 106 Donald G. Skinner, M.D. Department of Urology University of Southern California/Norris Comprehensive Cancer Center 1441 Eastlake Avenue, Suite 7414 Los Angeles, California 90033 Chapter 83 Steven J. Skoog, M.D. Professor of Surgery and Pediatrics Division of Urology The Oregon Health Sciences University 3181 South West Sam Jackson Park Road Portland, Oregon 97201 Chapter 107 Arthur D. Smith, M.D. Professor Albert Einstein College of Medicine Bronx, New York 10461, and Chairman Department of Urology Long Island Jewish Medical Center 270-05 76th Avenue New Hyde Park, New York 11040 Chapter 117 Edwin A. Smith, M.D. Department of Surgery Emory University School of Medicine Egleston Children’s Hospital Atlanta, Georgia 30322 Chapter 97 Joseph A. Smith, Jr., M.D. Professor and Chairman Department of Urologic Surgery Vanderbilt University Medical Center A-1302 Medical Center North Nashville, Tennessee 37232-2765 Chapter 33 Warren Snodgrass, M.D. Methodist Children’s Hospital 3606 21st Street, Suite 207 Lubbock, Texas 79410 Chapter 101 John P. Stein, M.D. Department of Urology, MS 74 University of Southern California Norris Cancer Center 1441 Eastlake Avenue, Suite 7414 Los Angeles, California 90033 Chapter 83 Gary D. Steinberg, M.D. Clinical Assistant Professor Department of Surgery Section of Urology The University of Chicago Hospitals MC 6038 5841 South Maryland Avenue Chicago, Illinois 60637 Chapter 17 Marshall L. Stoller, M.D. Department of Urology University of California San Francisco Urology Clinic 400 Parnassus Avenue Room A605 San Francisco, California 94143 Chapter 119 Lynn Stothers, M.D., M.H.S.C. Department of Surgery St. Paul’s Hospital Vancouver, British Columbia V52 4E3, Canada Chapter 39

Stevan B. Streem, M.D. Head Section of Stone Disease and Endourology Department of Urology The Cleveland Clinic Foundation/A100 9500 Euclid Avenue Cleveland, Ohio 44195 Chapter 110 Urs E. Studer, M.D. Professor and Chairman Department of Urology University of Berne 3010 Bern, Switzerland Chapter 81 Ray E. Stutzman, M.D. Associate Professor of Urology Department of Urology The Johns Hopkins University 601 North Caroline Street Baltimore, Maryland 21287 Chapter 31 David A. Swanson, M.D., F.A.C.S. Professor and Chairman (Ad Interim) Department of Urology The University of Texas M.D. Anderson Cancer Center 1515 Holcombe Boulevard Box 110 Houston, Texas 77030 Chapter 62 Rodney J. Taylor, M.D. Professor of Surgery Section of Urologic Surgery University of Nebraska Medical Center 600 South 42nd Street Omaha, Nebraska 68198-2360 Chapter 15 H. Tazaki Department of Urology New York Medical College Valhalla, New York 10595 Chapter 132 Alexis E. Te, M.D. J. Bentley Squier Urological Clinic Columbia University College of Physicians and Surgeons New York, New York 10032 Chapter 112 Anthony J. Thomas, Jr., M.D. Head Section of Male Infertility Department of Urology The Cleveland Clinic Foundation 9500 Euclid Avenue Cleveland, Ohio 44195 Chapter 58 Joachim W. Thüroff, M.D. Professor and Chairman Department of Urology Johannes Gutenberg University Medical School Langenbeckstrasse 1 55131 Mainz, Germany Chapter 82 Paul J. Turek, M.D. Assistant Professor In-Residence Department of Urology University of California Box 0738, Room U-575 533 Parnassus Avenue San Francisco, California 94143-0738 Chapter 56 Toyoaki Uchida, M.D. Assistant Professor Kitasato University School of Medicine 1-15-1, Kitasato, Sagamihara Kanagawa 228, Japan Chapter 113 Jeffrey Wacksman, M.D. Associate Professor Department of Surgery University of Cincinnati Medical Center, and Division of Pediatric Urology Children’s Hospital Medical Center

3333 Burnet Avenue Cincinnati, Ohio 45229-3039 Chapter 88 R. Dixon Walker, M.D. Professor of Surgery and Pediatrics Department of Surgery Division of Urology University of Florida College of Medicine Box 100247 JHMHC Gainesville, Florida 32610 Chapter 93 George D. Webster, M.B., F.R.C.S. Professor of Urology Surgery Department of Surgery Division of Urology Duke University Medical Center Box 3146 Durham, North Carolina 27710 Chapters 51 and 77 O. Lenayne Westney, M.D. Division of Urology University of Texas Health Science Center 6431 Fannin, Suite 6.018 Houston, Texas 77030 Chapter 85 Howard N. Winfield, M.D. Chief of Urology Department of Urology VA Palo Alto Health Care System, and Palo alto, California 94304-1290, and Associate Professor Department of Urology, S-287 Stanford University School of Medicine Stanford University Medical Center Stanford, California 94305-5118 Chapters 123 and 130 J. Stuart Wolf, Jr., M.D. Chief Section of Urology Ann Arbor Veterans Affairs Medical Center, and Assistant Professor of Surgery (Urology) University of Michigan Hospital 2916 Taubman Center 1500 East Medical Center Drive Ann Arbor, Michigan 48109-0330 Chapter 129 E. James Wright, M.D. Assistant Professor of Surgery Department of Surgery Division of Urology University of Kentucky 800 Rose Street Lexington, Kentucky 40536 Chapter 51 Theodros Yohannes, M.D. Resident Division of Urology University of Louisville School of Medicine Louisville, Kentucky 40292 Chapter 75 Philippe E. Zimmern, M.D. Associate Professor Department of Urology University of Texas Southwestern Medical Center 5323 Harry Hines Boulevard Dallas, Texas 75235-9110 Chapter 49

Preface
It has been nearly 30 years since the first edition of Urologic Surgery was published in 1969. Since then, the field of urology has undergone massive transformation. We have witnessed advances and improvements in virtually every aspect of our surgical craft. An exhaustive survey of improved capabilities would include reconstructive techniques, endourology, laparoscopic surgery, mastery of multiple energy sources for transurethral surgery, and tremendous strides in pediatric urology. However, certain fundamentals remain constant. As stated in the preface to the first edition, “Urologists are—first and foremost—surgeons.” The second edition expressed the hope that the volume “constitute the basis for further advances and that it be rendered obsolete by progress in urology.” In the third edition, it was acknowledged that progress in urology was paralleled by “advances in anesthesia, antibiosis, medical techniques, and diagnostic capability.” The fourth edition reaffirmed that “although many textbooks of urology and a number of excellent atlases dealt with surgical procedures, no single volume combined the virtues of text and illustrations that amplify the fundamental considerations and technical aspects” of urologic surgery. In retrospect, these thoughts are verities. We are still seeing rapid advances in our specialty. Perhaps there is no greater compliment to a medical publication than to admit to obsolesence in only a matter of a few months or years, as the result of our expanding capabilities. However, it is our expressed hope that this volume will serve as a ready reference for medical students, residents in training, and even our colleagues with the most advanced surgical skills. On a personal note, it is with great satisfaction that the editorship of Urologic Surgery passes on to Sam D. Graham, Jr., M.D., my capable friend, respected colleague, and former resident. He has assembled an outstanding group of authors for the fifth edition, and I anticipate with pleasure the prospect of further editions. James F. Glenn, M.D. Lexington, Kentucky

Acknowledgments
The editors would like to thank the following for their tireless efforts in bringing this book to completion: Tawn Edwards for organization and editing, Sandra Spruill for editing and formatting, and Jennifer Smith for editing the artwork. Finally, we are greatly indebted to Craig Percy at Lippincott–Raven Publishers for his guidance and help in bringing this project to fruition.

Chapter 1 Cushing's Disease and Syndrome Glenn’s Urologic Surgery

Chapter 1 Cushing's Disease and Syndrome
David P. O'Brien III

D. P. O'Brien III: Section of Urology, Emory University School of Medicine, The Emory Clinic, Atlanta, Georgia 30322.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

The association of pituitary lesions in patients with hirsutism, proximal muscle weakness, round plethoric faces, increased supraclavicular and infrascapular fat pads, thin skin, and other less frequent signs such as acne, purple abdominal striae, and psychiatric symptoms (Cushing's syndrome) has been known as Cushing's disease since the original description relating the illness to pituitary lesions by Cushing in 1912. It was many years later that the syndrome was found to be caused by cortisol excess, and later still it was found that there were multiple etiologies for this excess, including ectopic production of ACTH in adrenal tumors and tumors arising in other organs, including some that may be ectopic sources for corticotropin-releasing factor. The first planned operation for adrenal tumor was performed in 1914, with removal of a 17-cm adrenal adenoma and subsequent cure of hyperadrenocorticism. Initial attempts at curative pituitary surgery for Cushing's disease were short lived for lack of the necessary technology, but with modern microsurgical techniques, transsphenoidal pituitary microsurgery has become the treatment of choice. Adrenal surgery for Cushing's syndrome has in the past varied from partial to total adrenalectomy, depending on the availability of supplemental glucocorticoids. The etiologies of Cushing's syndrome are summarized in Table 1-1 and Fig. 1-1. In general, the pathophysiology of these disorders involves the production of excessive ACTH from pituitary adenomas or from ectopic sources, benign adrenal tumors and macronodular and micronodular adrenal hyperplasia, which usually produce excessive glucocorticoids only, whereas many adrenal malignancies also produce excessive androgens and mineralocorticoids. Iatrogenic administration of glucocorticoids is also a common etiology of Cushing's syndrome.

TABLE 1-1. Etiology of Cushing's syndrome

FIG. 1-1. Causes of Cushing's syndrome: (A) adrenocortical tumor; (B) adrenocarotical hypertension of hypothalamic origin; (C) adrenocortical hyperplasia caused by ectopic ACTH production; and (D) exogenous cortisol administration.

DIAGNOSIS
The sine qua non for the diagnosis of Cushing's syndrome is an abnormality of the plasma or urinary cortisol and/or ACTH. Because these studies are extremely variable, sometimes fluctuating daily and frequently negative, a high degree of suspicion usually leads the clinician to evaluate the patient with an overnight dexamethasone suppression test, a metapyrone stimulation test, or corticotropin-releasing factor stimulation. Imaging studies such as CT scanning, adrenal arteriography and venography, MRI (with and without gadolinium), and scintigraphy have also been used with some success. Because of the lack of sensitivity and specificity of these studies, it would also be common to perform repetitive evaluations in many patients.

INDICATIONS FOR SURGERY
Indications for surgery of the adrenal gland in patients with Cushing's syndrome include adrenal adenoma, adrenal hyperplasia, and adrenal carcinoma. Bilateral adrenalectomy has been suggested for those patients with micronodular adrenal hyperplasia, macronodular hyperplasia, patients with unknown sources of ACTH, and those with incurable pituitary Cushing's syndrome. 3,6

ALTERNATIVE THERAPY
Treatment of the aforementioned causes of Cushing's disease all require surgical management except for the iatrogenic administration of glucocorticoids.

SURGICAL TECHNIQUE
The posterior approach to the adrenal glands is described here; the other surgical approaches to the adrenals are described in Chapter 2, Chapter 3, Chapter 4 and Chapter 132. The posterior approach to the adrenal gland was first described by Young in 1936. Most authors would reserve this technique for small adrenal

adenomas or adrenal hyperplasia, i.e., noncancerous states with small lesions. 1,6 It is also the ideal method for bilateral adrenal exploration because the patient does not have to be repositioned. The patient is placed in the prone position under general endotracheal anesthesia. Appropriate padding is used to pad the chest, anterior pelvis, and legs (Fig. 1-2).

FIG. 1-2. Posterior approach for total adrenalectomy showing position of patient on operating table and bilateral incisions over 11th ribs. The medial end of the incision should be extended superiorly along the paravertebral musculature.

Several incisions have been described, and we prefer the hockey-stick incision, which begins just lateral to the midline at the ninth or tenth rib and extends downward and then laterally over the 11th or 12th ribs. Alternatively, an 11th rib incision or supracostal incision is used. 4,6 In the supracostal incision, the rib is spared, the intercostal muscles are divided, the pleura is swept away from the rib, and the retroperitoneum is entered ( Fig. 1-3). In the other approaches, the latissimus dorsi and sacrospinalis muscles are divided and retracted medially. The incision is extended through the periosteum of the 12th rib, which is resected close to the vertebral body (Fig. 1-4). The deep periosteum is incised, avoiding the neurovascular bundle, while laterally the abdominal muscles are divided and the pleura dissected from the diaphagm, exposing Gerota's fascia. As an alternative, in the transthoracic approach, the exposed pleura and diaphragm are incised, again exposing Gerota's fascia. It should be noted that the decision on which rib space to utilize (10, 11, or 12) depends on the position of the adrenal as estimated by the imaging studies, and it would be rare to be too high with the placement of the incision. In bilateral adrenalectomy, a Finochetti retractor can be placed for exposure ( Fig. 1-5).

FIG. 1-3. (A) Mobilization of the rib in the supracostal approach. The intercostal muscle is divided, with a finger behind to protect the deep structures. (B) Extrapleural fascia dissects away from the posterior surface of the rib with exposure of the insertion of the diaphragm and the pleura.

FIG. 1-4. Rib resection in any of the surgical approaches to the adrenal involves incision and elevation of the periosteum (A) with mobilization and resection of the rib as medially as is convenient (B).

FIG. 1-5. Simultaneous bilateral exposure of the adrenals is facilitated by use of a self-retaining retractor.

Gerota's fascia is incised, and the perinephric fat is swept away or incised superiorly, exposing the adrenal. Inferior retraction 3,6 of the kidney aids this portion of the dissection (Fig. 1-6). On the left side, the resection of the adrenal proceeds from laterally and superiorly to medially and inferiorly, where the main veins are ligated, the largest draining into the renal vein while the major adrenal artery arises from the main renal artery. On the right, the dissection is similar, but care is taken medially where the short right adrenal vein (and occasional accessory veins) empties into the vena cava. The largest adrenal artery usually arises from the main renal artery.

FIG. 1-6. With the subcostal approach (A), the kidney must be retracted inferiorly (B) to give access to the adrenal, permitting application of metal clips (C) for control and division of vessels.

If the pleura is entered, a temporary “pull-out” Robinson catheter (14 to 18 F) is placed, the pleura sutured, and the musculature approximated. While deep inspiration is maintained by the anesthesiologist, and the catheter is placed in an underwater seal, the catheter is quickly removed after all air bubbling in the water ceases. The remainder of the wound closure is completed, and a dressing applied.

OUTCOMES
Complications Surgical complications following adrenal surgery for Cushing's syndrome include not only those that pertain to routine retroperitoneal surgery, e.g., blood loss and infection, but also those complications specific to patients with hormonal imbalances. It should be mentioned that a chest x-ray in the recovery room is essential after any flank surgery in which the patient has been placed in the lateral or prone position, to evaluate the patient's pulmonary status for atelectasis and/or pneumothorax when the pleura has been violated. The occurrence of adrenocortical insufficiency should be kept uppermost in the clinician's mind even in the patient who has had a unilateral adrenalectomy. The use of supplemental glucocorticoids and mineralocorticoids is commonplace in these complex patients, whereas those in whom adrenalectomy is not curative need further evaluation, looking for ectopic sites of disease, either benign or malignant. Postoperative wound healing may be impaired, and the infection rate has been described to be between 4% and 21%. Other complications, e.g., thromboembolism, may be related to Cushing's syndrome or the associated obesity. Results The operative mortality for adrenalectomy in patients with Cushing's syndrome has been reported to be 2% to 6%, and the occurrence of Nelson's syndrome (the development of invasive pituitary tumors after adrenalectomy) seems minimal. Most patients with pituitary Cushing's syndrome who have poor results from pituitary surgery are cured with bilateral adrenalectomy, and a successful outcome should occur after adenalectomy in the patient with adrenal hyperplasia or an adrenal adenoma correctly diagnosed. In the small number of patients with adrenal malignancy, sugery may be curative if the tumor is localized, but metastatic disease responds poorly to the combination of adrenalectomy, radiation, and chemotherapy. It still remains, however, that management of these complicated endocrinologic patients is a continuing challenge for the urologic surgeon. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Angermeier KW, Montie JE. Perioperative complications of adrenal surgery. Urol Clin North Am 1989;16(3):597–606. Bennett AH, Cain JP, Dluhy RG, et al. Surgical treatment of adrenocortical hyperplasia: 20 year experience. J Urol 1973;109:321–324. Bertagna C, Orth DN. Clinical and laboratory findings and results of therapy in 58 patients with adrenocortical tumors admitted to a single medical center. Am J Med 1981;1:855–875. Bloom LS, Libertino JA. Surgical management of Cushing's syndrome. Urol Clin North Am 1989;16(3):547–565. Cushing H. The pituitary body and its disorders. Philadelphia: JB Lippincott, 1912. Glenn JF. Adrenal surgery. In: Glenn JF, ed. Urologic surgery, 4th ed. Philadelphia: JB Lippincott, 1993;1–21.

Chapter 2 Adrenal Adenoma and Carcinoma Glenn’s Urologic Surgery

Chapter 2 Adrenal Adenoma and Carcinoma
Muta M. Issa and Thomas E. Keane

M. M. Issa: Department of Surgery, Division of Urology, Emory University School of Medicine, Atlanta, Georgia 30322. T. E. Keane: Department of Urology, Emory University Hospital, Atlanta, Georgia 30322.

Epidemiology and Natural History Diagnosis Biochemical Evaluation Radiologic Evaluation Fine Needle Aspiration Cytology Other Evaluation Indications for Surgery Surgical Technique Surgical Approaches to the Adrenal Gland General Considerations The Posterior, Modified Posterior, and Posterior Transthoracic Approaches Lateral Flank Approach Anterior Transabdominal and Thoracoabdominal Approaches Laparoscopic Approach Outcomes Complications Results Chapter References

Adrenal tumors present either as a result of their clinical symptoms or as incidental findings during radiologic imaging studies. The nomenclature “adrenal incidentaloma” describes adrenal tumors discovered inadvertently by radiologic imaging in the absence of clinical indication. The objectives of this chapter are to discuss the various aspects of adrenal incidentaloma and adrenocortical carcinoma with regard to the epidemiology, natural history, investigation, diagnosis, and treatment with particular emphasis on their surgical management. Aldo-steronoma, pheochromocytoma, Cushing's disease and syndrome, and laparoscopic adrenalectomy have been purposely excluded from this chapter because they are reviewed in detail in Chapter 1, Chapter 3, Chapter 4, and Chapter 132.

EPIDEMIOLOGY AND NATURAL HISTORY
The prevalence of adrenal incidentalomas is estimated to approach 2%, 3 similar to the 1.9% figure of autopsy series, considering the increasing and widespread use of various radiologic images such as ultrasonography (US), computed tomography (CT), and magnetic resonance imaging (MRI). 1,3,10 The incidence of adrenocortical carcinoma, on the other hand, is extremely rare, with an estimated annual rate of 0.00006% to 0.00017% in the population, i.e., approximately one in a million. 4 Because the majority of adrenal carcinomas present late, and approximately half are hormonally active, it is calculated that the chance of detecting an adrenocortical carcinoma is one in 1,700 adrenal incidentalomas. Adrenocortical carcinomas occur in all age groups but are most common in the fifth to seventh decades of life. The vast majority of adrenal incidentalomas are benign, remain asymptomatic, and have favorable outcome. On the other hand, the opposite is true in adrenocortical carcinoma, which is an aggressive malignant disease with ominous prognosis. The majority present late with local invasion, regional lymph node involvement, or distant metastases (stages III and IV) rather than early, i.e., confined to the adrenal gland (stages I and II). Approximately half of these tumors are hormonally active and release excessive secretions of glucocorticoids, estrogens, androgens, and, rarely, mineralocorticoids.

DIAGNOSIS
Biochemical Evaluation Incidentally discovered adrenal tumors found by radiologic imaging can exhibit hormonal activity that initially may not be clinically apparent. Subsequent detailed endocrine clinical assessment may reveal, for the first time, clinical features suggestive of hormonal activity. In these patients, further evaluation with appropriate biochemical testing is indicated according to the clinical suspicion. On the other hand, 15% of adrenal incidentalomas may be totally asymptomatic in a setting of subclinical hormonal activity. Much debate currently exists regarding the recommended biochemical evaluation in this latter group. It is to be emphasized that before embarking on an extensive and exhaustive biochemical evaluation, one must consider the expected detection rate and cost/benefit ratio of such an empirical approach. The overall low prevalence of hormonal activity in these adrenal incidentalomas and the type of hormonal dysfunction likely to be encountered call for a more selective approach. The calculated possibility of uncovering a pheochromocytoma or aldosteronoma is one per 15 adrenal incidentalomas, and for a glucocorticoid-producing adenoma or adrenal carcinoma, it is one per 1,700 to 2,800. 9 It therefore seems reasonable to initiate biochemical screening for primary hyperaldosteronism with serum potassium first and for pheochromocytoma with urinary VMA, catecholamines, and metanephrines. Biochemical assessment for glucocorticoid and sex hormones should be reserved for patients with clinical features suggestive of such hormonal dysfunction. Radiologic Evaluation Computerized tomography (CT) is the standard radiologic imaging modality for adrenal tumors. It is utilized to delineate the anatomy and characterize the morphology of adrenal tumors. Although CT is reliable in the diagnosis of certain benign adrenal masses such as myelolipoma ( Fig. 2-1) and simple cysts, its dependability in accurately differentiating between benign and malignant tumors is limited. Certain CT features suggestive of a malignant process include a tumor >6 cm, inhomogeneity, irregular contours, thickened walls, and calcification, as illustrated in Fig. 2-2.

FIG. 2-1. Computerized tomography appearance of right myelolipoma (marked by white arrow), which has combined tissue characterization of muscle and fat.

FIG. 2-2. Computerized tomographic appearance of locally advanced left adrenocortical carcinoma (marked by white arrows), illustrating its huge size, irregular posterior borders, and inhomogeneity of its anteromedial portion.

Magnetic resonance imaging (MRI) is increasingly becoming a valuable radiologic tool in the evaluation of adrenal tumors. The MRI provides better visual clarity and resolution than CT imaging. Furthermore, the utilization of coronal planes in MRI ( Fig. 2-3 and Fig. 2-4) offers superior images for understanding anatomy and for assessing vena cava involvement. More recent advances in the MRI technique have revealed encouraging results in distinguishing between benign and malignant adrenal tumors. 2,5,6,7 and 8 This is achieved by the manipulation of fat and water MRI signals (chemical shift technique), which allows for the detection of microlipids within the tumor; these are a hallmark for benign adrenal adenomas. A significant subtraction of microlipids from the adrenal mass tissue on MRI as illustrated in Fig. 2-5 implies a benign adrenal adenoma. Conversely, lack of microlipid subtraction from the adrenal tissue, i.e., an area of hyperintensity, correlates with the presence of malignancy (Fig. 2-6). Further clinical experience and future refinements of this technique hold a real potential for MRI to become the imaging modality of choice in the evaluation of adrenal masses.

FIG. 2-3. Magnetic resonance imaging (coronal view) of a left adrenocortical carcinoma (marked by white arrow) illustrating its huge size, lobular contour, irregular borders, inhomogeneity, and invasion of the upper pole of the left kidney.

FIG. 2-4. Magnetic resonance imaging (coronal view) of a left adrenocortical carcinoma (marked by white arrow) illustrating its huge size, inhomogeneity, and local invasion both superiorly and inferiorly. The entire uninvolved length of the inferior vena cava (marked by black arrows) can be clearly visualized.

FIG. 2-5. (A) Magnetic resonance imaging appearance of an adrenal adenoma illustrating its isointense appearance on T 1-weighted image and (B) its hypointense appearance following the dramatic substraction of its microlipid signal. White arrows point to the adrenal mass, and the black arrow to the inferior vena cava. (Courtesy of Roger Y. Shifrin, M.D.)

FIG. 2-6. (A) Magnetic resonance imaging appearance of a small isointense malignant lesion (marked by white arrow) in a hypointense right adrenal adenoma. This image illustrates the value of MRI in accurately detecting a small metastatic deposit in an adenomatous gland. (B) Cross-section gross appearance of the adrenal gland following surgical excision illustrating the lesion (marked by black arrow). The lesion was a deposit from primary bronchogenic adenocarcinoma. (Courtesy of Roger Y. Shifrin, M.D.)

Fine Needle Aspiration Cytology When clinical, biochemical, and radiologic evaluation fail to provide sufficient diagnostic information for an appropriate management decision, fine needle aspiration (FNA) cytology may play a role. Difficulty in tissue sampling, preparation, and interpretation can lead to false-negative results and limit its reliability. Potential indications for FNA cytology include atypical adrenal cysts, i.e., with thick, irregular walls or inhomogeneous fluid content, and differentiation between primary and metastatic deposits in individual cases. Other Evaluation Recent advances in noninvasive radiologic imaging such as CT and MRI have replaced the need for scintigraphy as well as NP-59 and MIBG, venography, and arteriography. The previously recommended use of venography in the assessment of inferior vena caval involvement has been superseded by the less invasive MRI. In rare circumstances, preoperative arteriography may have a role in delineating the vascular supply of large adrenal masses to aid in planning surgical resection.

INDICATIONS FOR SURGERY
Generally, there are two definitive indications for surgical excision: (a) symptomatic and hormonally active adrenal tumor and (b) adrenal carcinoma. The dilemma arises in patients with asymptomatic adrenal tumors and unconfirmed diagnosis despite extensive evaluation. These constitute a gray zone into which fall the majority (>80%) of adrenal incidentalomas. Conservative management of these patients by observation and serial imaging poses certain concerns about the possibility of delayed or missed diagnosis of a biologically active lesion. To date, the controversies continue with no clear answers to the primary concern of whether or not a tumor is malignant. As a result of the available epidemiologic data, the influence of radiologic features, and FNA cytology, certain guidelines are becoming available for management decisions in an attempt to minimize unnecessary surgical excisions without compromising the final outcome. Surgical resection continues to be the treatment of choice for adrenal masses with features suggestive of malignancy such as large size (>6 cm), inhomogeneity, irregular contours, thickened walls, calcification, regional lymphadenopathy, lack of microlipids on chemical-shift MRI, and suspicious FNA cytology. Conservative management with observation and serial imaging is recommended for small tumors (<3 cm) with benign radiologic features and negative FNA cytology ( Fig. 2-7).

FIG. 2-7. Algorithm for management strategy of incidentally discovered adrenal masses.

In the midst of this spectrum reside the 3- to 6-cm tumors that pose the greatest diagnostic difficulties, especially when their imaging features and FNA cytology are equivocal. The decision whether to observe or operate should be carefully tailored in each individual case, keeping in mind the possibility of small adrenocortical carcinoma. In patients with a metastatic deposit to the adrenal gland, the decision for surgical excision should be meticulously evaluated, and surgical excision should be limited to patients with solitary metastatic deposits in whom the primary malignancy has been adequately treated.

SURGICAL TECHNIQUE
Surgical Approaches to the Adrenal Gland Adrenalectomy can be performed through a number of surgical approaches. These include posterior (with and without rib resection); posterior transthoracic; lateral flank (with and without rib resection); anterior transabdominal through subcostal, transverse, Chevron, or midline incisions; thoracoabdominal; and laparoscopic approaches. The choice of surgical approach is influenced by the tumor pathology, size of the adrenal tumor, and patient's habitus as well as by the surgeon's preference, familiarity, and experience with the surgical technique. This chapter focuses primarily on the lateral flank approach because the remaining surgical approaches are discussed in detail in the other chapters. General Considerations Regardless of the surgical approach, it is of paramount importance for the urologist to fully understand the anatomic position and relation, the blood supply of the adrenal glands, and the difference between the two sides. Each adrenal is a delicate and friable gland located superior to the upper pole of the kidney on the right but in a more superomedial position on the left. They lie adjacent to their respective great vessels, with the right posterolateral to the inferior vena cava and the left adrenal lateral to the aorta ( Fig. 2-8 and Fig. 2-9). Other structures in close anatomic proximity include the duodenum on the right and the stomach, spleen, and pancreas on the left ( Fig. 2-9). The sympathetic chain and ganglia are also closely located posteromedially to the adrenal glands, particularly on the left side.

FIG. 2-8. The anatomic relationship of the adrenal glands to the aorta and inferior vena cava. Multiple arterial vessels entering the glands indicate the rich arterial supply, while a single central adrenal vein illustrates the limited and relatively constant venous drainage.

FIG. 2-9. An anterior view of the abdomen illustrating the anatomic relationship of the adrenal glands to surrounding gastrointestinal tract and organs.

The arterial blood supply originates from three sources: the inferior phrenic artery superiorly, the aorta centrally, and the renal artery inferiorly ( Fig. 2-8). This rich blood supply enters the adrenal gland in a stellate fashion, primarily superomedially, while the base and the posterior surface of the gland often lack vasculature. Each adrenal has one central vein; the one on the right is short and drains directly into the inferior vena cava, and that on the left is longer and drains into the left renal vein. The Posterior, Modified Posterior, and Posterior Transthoracic Approaches The posterior surgical approach is performed through a subcostal incision, though resection of a rib (12th or 11th) is frequently required for better exposure. This approach was favored in the past in bilateral adrenal surgery for diagnostic exploration and total ablative adrenalectomy. However, with better preoperative radiologic diagnosis and localization of adrenal masses and the decline of bilateral surgical adrenalectomy in cancer therapy, the use of this approach has diminished. The posterior approach provides a limited surgical field, and the incidence of pleurotomy increases with rib resection, which may prove challenging and unsuitable for excision of large adrenal masses and in overweight patients. Nonetheless, it provides a relatively direct access to the adrenal gland, is potentially less traumatic, is well tolerated by patients, and causes less postoperative ileus. In our opinion, the posterior approach to adrenalectomy is best reserved for thin patients with small well-localized benign adrenal tumors in anatomically favorably positioned glands. Modifications of the standard posterior approach have been utilized for excision of larger adrenal masses. Such modifications include an upward extension of the medical end of the standard posterior oblique skin incision, rib resection, and transthoracic accesses. A superior surgical exposure is achieved at the expense of more extensive dissection and the need for chest tubes postoperatively. Lateral Flank Approach The lateral flank approach offers better and larger operative exposure for excision of larger tumors in adrenal glands positioned in less favorable anatomic locations. In general, urologists tend to be more familiar with this approach because of its frequent utilization in renal surgery. Following the administration of general anesthesia with endotracheal intubation, a Foley catheter is used to drain the bladder, and sequential compression devices (SCD) are placed on the legs. The patient is then placed in the lateral position with the flank over the flexion/kidney rest site of the operating table with the patient's back close to the edge. Both arms are kept extended, with the contralateral one resting on an arm rest with an axillary roll and the ipsilateral arm on a stand adjusted to the appropriate height and angle. A pillow is placed between the legs, keeping the contralateral leg flexed, and the ipsilateral leg is allowed to remain straight. The kidney rest is then raised, and the table is flexed to 30 to 45 degrees ( Fig. 2-10). The patient position is then further secured by the use of wide adhesive tape over the hip and shoulder.

FIG. 2-10. The lateral flank position for adrenalectomy with skin marking of the surface anatomy of the 11th and 12th ribs. With the patient in the lateral position, the operating table is flexed at 30 to 45 degrees, and the kidney rest elevated (indicated by the black arrow) to facilitate exposure. The dotted line marks the midline on the back.

Following skin preparation and draping, a flank skin incision is made along the rib extending from sacrospinalis muscle posteriorly to lateral border of the rectus abdominis muscle anteriorly. The incision is deepened through the subcutaneous fat layer, exposing the external oblique muscle anterolaterally and the latissimus dorsi muscle posteriorly, under which lies the internal oblique muscle and inferior part of the serratus posterior muscle. The muscles are incised with diathermy along the line of the incision to expose the rib, thoracodorsal fascia, and the transversus abdominis muscle. Two spring retractors are placed to retract the muscle edges and facilitate the exposure. The periosteum over the rib is incised and elevated using a periosteal elevator or diathermy. The tip of the rib is freed and gently retracted outward with the aid of a Kocher clamp to expose its undersurface. The rib is dissected off its bed along its entire length, starting at the tip. It is then transected at its proximal end with the aid of a guillotine rib resector. The use of diathermy instead of conventional periosteal elevators during rib dissection allows for better control of hemostasis and decreases the chance of accidentally injuring the neurovascular bundle or the pleura. The lumbodorsal fascia and the transversus abdominus muscle

are incised, exposing the peritoneum and its preperitoneal fat. The peritoneum is bluntly dissected off the abdominal wall using the index finger in gentle sweeping motions, and the dissection is continued to free the undersurface of the rib bed. Starting anteriorly, the rib bed is carefully incised to complete the access into the flank. Careful attention should be practiced to avoid injury to the neurovascular bundle inferiorly, pleura, and diaphragm superiorly, and a Finochietto self-retaining retractor is then placed. Once in the flank, the kidney is partially mobilized by blunt dissection of the overlapping peritoneum and colon off Gerota's fascia. Crossing vessels along the dissection plane are coagulated or ligated to secure hemostasis as the dissection is continued medially toward the renal hilum. The renal vein is visualized laterally and inferiorly to further mobilize the lower pole of the kidney. The superior pole of the kidney and the adrenal gland are purposely kept attached together to aid in the dissection later; this allows the adrenal gland to be brought down by gentle inferior traction of the kidney and thus facilitates the identification of its vessels and surrounding attachments. The vessels and the surrounding attachments of the adrenal are then secured with 3-0 silk ligatures, surgical clips, or electrocautery, resulting in mobilization of the adrenal gland ( Fig. 2-11 and Fig. 2-12). It is to be stressed that the adrenal vessels are small, short, and easily traumatized, especially on the right, thus demanding gentle retraction and careful dissection. Finally, unless simultaneous nephrectomy is not indicated, the adrenal gland is dissected off the upper pole of the kidney and removed ( Fig. 2-13 and Fig. 2-14).

FIG. 2-11. Intraoperative photograph of the upper pole of the left kidney with its attached left adrenal gland (marked by the white arrows) as visualized through a left flank incision. A nontraumatic clamp is placed on the perirenal fat to aid in the downward traction (marked by the black arrow).

FIG. 2-12. Intraoperative photograph of left adrenalectomy through a flank approach. The adrenal gland has been mobilized except for its left adrenal vein seen ligated and clipped (marked by the left black arrow). A communicating branch of inferior phrenic vein (marked by the right black arrow) is seen ligated and clipped along its path as it drains into the left adrenal vein.

FIG. 2-13. Intraoperative photograph of the left adrenal fossa following adrenalectomy through a flank approach.

FIG. 2-14. A cross section of the left adrenal gland with well-defined adrenal incidental tumor measuring 1.8 cm in diameter. Surgical excision was performed because of the hormonal activity of the tumor, which proved to be a benign aldosteronoma on pathological examination.

In the treatment of adrenocortical carcinomas, surgical treatment also necessitates additional en bloc radical nephrectomy. The renal vessels and ureter are therefore identified, ligated with 2-0 silk, and transected. The kidney is removed en bloc with its Gerota's fascia and attached adrenal gland. Surgical excision of the regional lymph nodes is performed by removing the retroperitoneal tissue surrounding the adjacent great vessels. Drains are not used unless there is some residual bleeding. The flank musculature and fascia including the periosteum of the rib bed are closed in layers using running 0 polyglycolic (Dexon) or polygalactic polymer (Vicryl) sutures. Subcutaneous tissue is closed with interrupted or running fine (3-0) plain catgut sutures, and the skin is closed with standard skin staples. Anterior Transabdominal and Thoracoabdominal Approaches

For large adrenal tumors, where a posterior or flank approach offers a limited and relatively inadequate exposure, the transabdominal or transthoracic surgical approach is a preferred alternative because of its superior operative exposure. The anterior abdominal approach can be performed through subcostal, transverse, chevron, or midline incisions. The use of these approaches is dictated by the size and pathology of the adrenal tumor as well as patient habitus and operator preference. In general, adrenocortical carcinomas are excised through such approaches to ensure ample exposure, control, and the ability to accurately stage the disease intraoperatively. Laparoscopic Approach The demand for minimally invasive alternative surgical therapies and for cost savings in our current health care system has initiated the concept of laparoscopic adrenalectomy. To date, the experience with this approach is limited to a few medical centers with specific interest in laparoscopic surgery. Technical difficulties, limited instrumentation, and the low volume of surgical adrenalectomies have restricted the utilization of this approach. Laparoscopic adrenalectomy is discussed in detail in Chapter 132.

OUTCOMES
Complications Preoperatively, patients with hormonally active adrenal tumors should undergo careful preparation to control their hormonal dysfunction and optimize their fluid and electrolyte status. These are essential prerequisites in pheochromocytoma, aldosteronoma, and Cushing's syndrome. In pheochromocytoma, a blockers are used to decrease vascular tone and control hypertension while intravenous fluid hydration is used to counteract the potential of vasodilation and vascular collapse on excision of the tumor. In aldosteronoma, hypokalemia is treated with spironolactone and potassium supplements. Such preparation should be instituted for a number of weeks to ensure adequate correction and recovery of the suppressed zona glomerulosa of the contralateral adrenal gland. In glucocorticoid-secreting adrenal tumors, glucocorticoid is given intraoperatively and continued postadrenalectomy because the zona fasciculata of the contralateral adrenal function may be suppressed by negative feedback from the excessive glucocorticoid secreted by the tumor. Careful monitoring during the postoperative period is mandatory to recognize and correct such potential adverse events. It should also be emphasized that healing in patients with Cushing's syndrome may be slow because of the compromised tissue status from excessive glucocorticoid and associated glucose intolerance and the increased potential risks for wound infection, dehiscence, and the development of postoperative incisional hernia. The use of prophylactic antibiotics, meticulous wound closure, and careful postoperative wound care should decrease the risk for such adverse events. Early ambulation and aggressive pulmonary toilet are encouraged to minimize the development of postoperative respiratory complications as well as venous thrombosis and pulmonary embolism. Intraoperatively, injury to adjacent structures may result, including a pneumothorax, which can be recognized by the formation of bubbles at the site of pleurotomy. Treatment includes closure of the pleural tear with running 4-0 chromic catgut suture and removal of the air and fluid from the pleural cavity through a soft catheter at time of closure. The use of suction or an underwater seal system during hyperinflation of the lung by the anesthesiologist aids in accomplishing this task. Rarely, a chest tube is required for larger pneumothorax (greater than 10%) or when respiratory compromise results. Careful follow-up with serial chest x-rays is needed to ensure resolution of the pneumothorax. Other structures at risk for injury during adrenalectomy include the kidney, spleen, liver, and pancreas. A tear of the renal capsule may result from forceful retraction of the kidney during dissection and may be treated with gentle tamponading in minor injuries; larger tears may require repair by suturing. Splenic capsular injuries may also occur as a result of retraction or direct dissection, and the use of pressure tamponade is usually sufficient, though occasionally hemostatic gel packing, splenic repair, cauterization with an argon beam coagulator, or even splenectomy may be indicated. Liver injuries may also occur through the same mechanism and should be handled by hemostatic gel packing or repair. Injuries to the pancreas and subsequent pancreatic inflammation may occur during dissection at the region of the upper pole of the left kidney, leading to the postoperative pancreatitis. Rarely, however, a more substantial injury may result in fulminant pancreatitis and significant pancreatic leak and fistula. Avulsion of the adrenal vessels, especially the delicate central veins, can lead to significant bleeding. Hemostasis is secured by packing and suture ligation. With all such injuries in which there is a potential of delayed bleeding, the use of postoperative drains is recommended. Results Generally, convalescence of patients undergoing adrenal surgery is surprisingly smooth. Careful preoperative preparation of the patient, meticulous intraoperative surgical technique, and proper postoperative care are mandatory for successful outcome and complication-free recovery. The prognosis of adrenalectomy in benign adrenal disease is most favorable once the tumor is completely excised and recovery is uneventful. Complete cure from the disease process and return to normal function are expected in the majority of patients. The prognosis of adrenalectomy in malignant disease, however, is variable depending on the stage of the disease and on whether or not complete surgical excision is achieved. The potential for a cure can be achieved only in early-stage adrenocortical carcinoma on complete excision of the disease without tumor spillage or positive surgical margins. The 5-year survival in early (stage I or II) adrenocortical carcinoma is 50%, which drops to 5% to 10% in the advanced (stage III or IV) disease. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Abecassis M, McLoughlin MJ, Langer B, Kudlow JE. Serendipitous adrenal masses: prevalence, significance, and management. Am J Surg 1985;149:783–788. Bilbey JH, McLoughlin RF, Kurkjian PS et al. MR: Imaging of adrenal masses: value of chemical shift imaging for distinguishing adenomas from other tumors. Am J Roentgenol 1995;164:637. Caplan RH, Strutt PJ, Wickus CG. Subclinical hormone secretion by incidentally discovered adrenal masses. Arch Surg 1994;129:291. Copeland PM. The incidentally discovered adrenal mass. Ann Surg 1984;199:116. Leroy-Willig A, Bittoun J, Luton JP, et al. In vivo MR spectroscopic imaging of the adrenal glands: distinction between adenomas and carcinomas larger than 15 mm based on lipid content. Am J Roentgenol 1989;153:771. Mitchell DG. Chemical shift magnetic resonance imaging: applications in the abdomen and pelvis. Top Magnet Reson Imag 1992;4(3):46. Mitchell DG, Crovello M, Metteucci T, et al. Benign adrenocortical masses: diagnosis with chemical shift MR imaging. Radiology 1992;185:345. Reinig JW, Doppman JL, Dwyer AJ, Johnson AR, Knop RH. Distinction between adrenal adenomas and metastases using MR imaging. J Comput Assist Tomogr 1985;9(5):898. Ross NS, Aron DC. Hormonal evaluation of the patient with an incidentally discovered adrenal mass. N Engl J Med 1990;323:1401. Russi S, Blumenthal HT, Gray SH. Small adenomas of the adrenal cortex in hypertension and diabetes. Arch Intern Med 1945;76:284.

Chapter 3 Primary Aldosteronism Glenn’s Urologic Surgery

Chapter 3 Primary Aldosteronism
James F. Glenn

J. F. Glenn: Department of Surgery, Division of Urology, University of Kentucky Medical Center, Lexington, Kentucky 40536.

Epidemiology Pathophysiology Diagnosis Laboratory Localization Techniques Indications for Surgery Alternative Therapy Surgical Technique Outcomes and Complications Chapter References

EPIDEMIOLOGY
Primary aldosteronism was first described by Jerome Conn in 1955. 3 The clinical characteristics and manifestations, frequently referred to as Conn's syndrome, include hypertension, hypokalemia, hypernatremia, and alkalosis with increased urinary potassium excretion and decreased urinary sodium. Many patients with aldosteronism exhibit certain forms of diabetes. 1 The most common cause of primary aldosteronism is a solitary aldosterone-secreting adrenal adenoma. However, there are a large group of patients who will exhibit primary aldosteronism as a result of bilateral focal or nodular adrenal hyperplasia. In addition, a few patients have been reported with aldosteronism as a manifestation of adrenocortical carcinoma; the syndrome may be exaggerated in these rare instances. In children, aldo-steronoma is rare, with most cases secondary to adrenocortical hyperplasia.

PATHOPHYSIOLOGY
The electrolyte and acid–base disorders of aldosteronism are readily understood as a consequence of excessive production of aldosterone. However, the exact mechanism of the hypertension, generally indistinguishable from essential hypertension, is not easily explained. On the other hand, hypertension in the presence of hypokalemia (serum potassium less than 3.5 mEq/liter), alkalosis (serum bicarbonate more than 30 mEq/liter), and hyperkaluria (urinary excretion of more than 30 mEq/24 hr) is most certainly caused by aldosteronism. The diagnosis can be confirmed by measurement of plasma aldosterone levels, remembering that there will be a diurnal variation and postural responses. A more reliable method of establishing elevated aldosterone levels is the measurement of 24-hour urinary excretion values (normal range 5 to 19 µg/24 hr). 10

DIAGNOSIS
Laboratory The routine laboratory determinations noted above are fundamental to the diagnosis of primary aldosteronism. Measurement of plasma and urinary aldosterone can generally be accomplished in hospital and commercial laboratories. Adrenal venous sampling for measurement of plasma adosterone is almost always accurate in determining elevated levels of aldosterone as well as aiding in the localization of solitary adenomas. The postural stimulation test (PST) may be used not only to confirm the diagnosis of primary aldosteronism but also to distinguish between aldosterone-producing adenoma and idiopathic hyperplasia as the cause of the syndrome. Serum aldosterone and cortisol levels are measured after overnight recumbency, patients are ambulated for 4 hours, and the tests are then repeated. Patients with adenoma exhibit elevated base levels of aldosterone that will either fall or increase minimally after upright activity, whereas the high base level of aldosterone in patients with hyperplasia will generally rise signficantly, more than 30%. Demonstration of a reduced or unchanged cortisol after ambulation confirms the validity of the test, reflecting the usual morning fall in ACTH. Further diagnostic confirmation of primary aldosteronism may be gained through measurement of plasma renin. 4 Renin production is suppressed, usually below normal levels, in aldosteronism as compared with renin levels in normal subjects or those with hypertension of other causes. This is the consequence of the homeostatic mechanism by which blood volume and hence blood pressure is maintained under normal circumstances. Figure 3-1 graphically depicts this mechanism in which a decreased blood volume or flow to the kidney stimulates the stretch receptor mechanism, invoking increased production of renin. Renin then acts on angiotensin I to convert it to angiotensin II, which in turn stimulates adrenocortical production of aldosterone. This generates increased excretion of urinary potassium with conservation of sodium, leading to fluid retention, which, in turn, increases the blood volume. Clearly, the presence of increased aldosterone will lead to some degree of hypervolemia, which will result in decreased production of renin.

FIG. 3-1. Interrelationship between renal blood flow and the renin–angiotensin pathway.

Aldosteronoma may also be differentiated from idiopathic hyperplasia by serum assay for 18-hydroxycorticosterone (18-OHB). Young and Klee 10 observed that patients with adenoma usually have 18-OHB levels more than 100 ng/dl after overnight recumbency, whereas patients with hyperplasia have values less than 100 ng/dl. It has been reported that 18-OHB assay accuracy is increased by intravenous saline infusion, and it has also been observed that the ratio of 18-OHB to cortisol is greater than 3 in patients with aldosteronoma but not in patients with hyperplasia. The diagnosis of primary aldosteronism may be confirmed or excluded using the fluorocortisone (0.1 mg q6h for 4 days) suppression test. Similarly, familial hyperaldosteronism type 1 may be diagnosed or excluded by use of the dexamethasone (0.5 mg q6h for 4 days) suppression test, although identification of the hybrid gene in peripheral blood DNA may also be diagnostic. Localization Techniques A variety of imaging techniques may be employed for localization of aldosterone-producing adenomas or adrenal hyperplasia, though actual adrenal enlargement with

patients with focal micronodular hyperplasia is uncommon. Large adrenal tumors greater than 3.0 cm in diameter may be visualized on a plain film of the abdomen or by intravenous pyelography. However, it is extremely unusual for aldosterone tumors of the benign type to present as a large mass. In contrast, the rare malignancy may achieve significant size. In years past, retroperitoneal carbon dioxide insufflation studies were employed for visualization of the adrenals. This invasive technique is no longer necessary because both CT scanning and magnetic resonance imaging are extremely effective in identifying hyperplastic adrenals or aldosterone-producing adenomas as small as 6 or 7 mm.6 The anatomic relationship of the adrenal glands to the kidneys or other viscera as delineated by CT or MRI is illustrated in Fig. 3-2.

FIG. 3-2. Anatomic relationships of the adrenal glands as revealed by CT scan.

Adrenal venography is not recommended as a routine procedure. The retrograde injection of contrast material into the adrenal may produce an infarction on occasion. If such injury is incurred by the normal adrenal, removal of the contralateral adrenal with an aldosteronoma could result in adrenal insufficiency. However, adrenal venography may be accomplished on some occasions, particularly if adrenal vein hormone sampling is desired. Adrenal scintigraphy is one of the most useful noninvasive studies in functional evaluation and anatomic localization of aldosteronomas. 5 Adrenocortical scanning is performed with the radionuclide 131I-6-b-iodomethyl-19-norcholesterol (NP-59). All antihypertensive medications should be discontinued for 1 week, allowing normal dietary sodium intake with the addition of dexamethasone treatment for 3 to 7 days. Oral iodide is given to block tracer uptake by the thyroid, and imaging is performed 2 to 5 days after administration of NP-59. Prompt unilateral adrenal uptake and activity are evidence of aldosteronoma, whereas bilateral uptake is evidence of adrenal hyperplasia. The test has about a 75% accuracy.

INDICATIONS FOR SURGERY
It is imperative that the cause of primary aldosteronism be established because the etiology will determine treatment and management. 7 Patients with adrenal carcinoma most certainly should have adrenalectomy, preferably by open surgery and usually through the transabdominal approach. Patients with idiopathic hyperplasia as a cause of aldosteronism are best managed medically because surgical removal of the adrenals rarely will result in control of hypertension. Medical management consists of administration of spironolactone, diuretics such as hydrochlorothiazide, and antihypertensive medications if necessary. Removal of a discrete aldosteronoma will result in cure of hypertension and conversion of biochemical abnormalities in 80% to 90% of patients. Accordingly, adrenalectomy or partial adrenalectomy is the procedure of choice in patients with an aldosterone-producing adenoma. Because bilateral and ectopic aldosteronomas are essentially unheard of, the unilateral removal of the adenoma or the adrenal is preferred.

ALTERNATIVE THERAPY
It must be stated that aldosterone-producing adenomas of the adrenal are essentially ideal for laparoscopic management. 2 This technique is amply described in Chapter 132. Laparoscopic adrenalectomy will become the procedure of choice for the single adenoma producing aldosteronism, but there will be instances in which open surgery is desirable. 9

SURGICAL TECHNIQUE
General endotracheal anesthesia is required. Position of the patient demands control of the airway, and forceful ventilation at times during the procedure is necessary for identifying and protecting the diaphragm and pleura and in closing any pleural rents that may be incurred during the exposure. There are no special anesthetic requirements, such as nitroprusside, as with pheochromocytoma. The preferred approach to the involved adrenal in cases of primary aldosteronism secondary to adenoma is posterior ( Fig. 3-3). Incision may be the classical hockey stick, as described by Hugh Hampton Young, or a rib incision may be utilized. Personal preference is for an oblique incision over the 11th rib, resecting the rib as far medially as possible but preserving the periosteum, which affords excellent substance for closure. An alternative is the supracostal incision of Turner-Warwick, cutting the rib at its posterior angle to allow for inferior deflection. The supracostal incision avoids the infracostal vessels and nerves. Personal preference is for rib resection (Fig. 3-4).

FIG. 3-3. With the patient properly positioned and supported, (A) the posterior approach through a hockey-stick, 11th rib, or Turner-Warwick incision (B) permits extrapleural and extraperitoneal exposure of the adrenal.

FIG. 3-4. Rib resection in any of the surgical approaches to the adrenal involves incision and elevation of the periosteum (A) with mobilization and resection of the rib as medially as convenient (B).

Entry into the pleural cavity can be avoided by carefully reflecting the pleura upward. This is best accomplished by beginning the dissection laterally at the extreme lateral margin of the incision. By blunt and sharp dissection, the pleura and most of the diaphragmatic muscle fibers can be swept medially and superiorly, literally moving the pleural cavity away from the diaphragm to give access to the retroperitoneal space. By using this technique and proceeding from lateral to medial, it is usually possible to avoid entry into the pleural cavity. Once the retroperitoneal space has been entered, Gerota's fascia can be identified. This should be opened as far superiorly as possible. The perinephric fat is readily identified, having slightly different texture than the surrounding retroperitoneal fat deposits. The peri-nephric fat is dissected medially and superiorally to expose the upper pole of the kidney, which is then freed of surrounding fat. A padded Deaver rectractor is then introduced over the upper pole of the kidney, which is retracted inferiorly. This maneuver will almost always bring the ipsilateral adrenal into view in the operative field. The adipose tissue surrounding the adrenal is delicately teased away from the adrenal gland. No significant vasculature is encountered in this dissection, but small bleeders may be controlled by electrocautary. The adrenal is thoroughly exposed before an attempt is made to gain access to the vessels. The blood supply of the adrenals is variable from side to side and from patient to patient. It is relatively independent of renal circulation ( Fig. 3-5). The arterial supply to each adrenal consists of a multitude of small branches derived from the renal artery, the aorta, the inferior phrenic, and occasionally from the splenic artery on the left. The venous return is much more constant than the arterial supply, the right adrenal vein arising from the hilum of the gland medially and emptying directly into the vena cava, whereas on the left, the main adrenal vein joins the inferior phrenic to empty into the left renal vein. Control of the vasculature is generally achieved with small metal surgical clips as the dissection proceeds.

FIG. 3-5. The usual blood supply of the adrenal glands is pictured here (anterior view), but a highly variable pattern of vascular supply and drainage can be anticipated.

Injury to contiguous structures must be avoided. On the right, the adrenal is in close proximity to the liver superiorly and anteriorly, the vena cava medially, and the kidney inferiorly. Great care should be taken to identify the vena cava because the gland is in intimate contact. On the left, the adrenal is somewhat closer to the renal pedicle, and care must be taken in exposing the inferior surface of the gland to prevent damage to the renal artery and vein. The posterior aspect of the stomach and the tail of the pancreas are in close approximation to the left adrenal anteriorly. Once the gland has been isolated, it is removed from the wound, and the bed is inspected for bleeding, which is controlled by cautery in most instances. Occasionally it is useful to employ a square of Gelfoam or Oxycel in the bed of the adrenal. Surgical drains are unnecessary in virtually all cases. When retraction is discontinued, the kidney will ascend into its normal position, virtually obliterating the space from which the adrenal gland was removed. Gerota's fascia is closed using multiple running and interrupted sutures of absorbable materials, usually catgut or polyglycolic acid (PGA). Closure of the rib bed consists of approximating the two portions of the periosteum with either absorbable or nonabsorbable suture material, usually 2-0 catgut, taking care to avoid injury to the infracostal neurovascular bundle. There is scanty musculature overlying the rib, but this may be closed with interrupted absorbable sutures. Subcutaneous tissue is approximated with running or interrupted plain catgut, and skin may be closed with metal clips or, preferably, with interrupted vertical mattress sutures of 3-0 silk. With the imaging studies available, aldosterone tumors will almost always have been localized preoperatively. Rarely, it may be necessary to accomplish bilateral adrenal exploration before making a decision about adrenalectomy. If this is the case, simultaneous and identical exposures of the adrenal area may be achieved. The Finochietto thoracic retractor may be utilized with the blades reversed to compress the paraspinialis musculature in the fashion shown in Fig. 3-6.

FIG. 3-6. Simultaneous bilateral exposure of the adrenals is facilitated by the use of a self-retaining retractor.

A typical and characteristic aldosteronoma of approximately 1 cm diameter is depicted in Fig. 3-7.

FIG. 3-7. Adrenal gland with well-defined aldo-steronoma. Multiple areas of focal microtubular hyperplasia are also seen.

OUTCOMES AND COMPLICATIONS
Entry into the pleura occurs with great frequency simply because of the deep inferior reflection in many patients. Indeed, such entry into the pleural cavity cannot be regarded as a complication but rather as the anatomic consequence in most instances. Opening into the pleura can be managed near the conclusion of the case. A soft rubber catheter of about 18 F caliber with multiple openings cut near its tip is introduced into the pleural cavity. The pleura with diaphragm is then closed with running sutures incorporating the catheter. The anesthesiologist is requested to fully expand the lungs for a period of about 30 seconds while suction is applied to the catheter, which is then quickly withdrawn, and the suture is tied securely while the lung is inflated. A postoperative chest x-ray obtained in the recovery room will assure that there is no residual pneumothorax. Intraoperative bleeding is virtually never a problem. On the left, the aorta is readily identified by palpation of its pulsation, which permits avoiding any injury in dissection. A bit more caution must be employed on the right because the vena cava is not as readily identified. If the vena cava is opened during the process, it is best closed with running suture of atraumatic 3-0 vascular silk. Variations in blood pressure during surgery are rarely encountered. Whereas the patient with pheochromocytoma may experience wide fluctuations in blood pressure, the hypertensive patient with aldosteronism does not exhibit this same lability. As a consequence, preventive measures such as administration of blood volume expanders or whole blood are rarely necessary. Postoperative infection is also a rarity in surgery for aldosteronoma. Although the patient with Cushing's disease may suffer immune suppression and a tendency to infection, this is not the case in primary aldosteronism. Ordinarily, prophylactic postoperative antibiotic therapy is not employed, but an elderly person with poor ventilation and potential pulmonary compromise may justify utilization of anticipatory antibiotic therapy. Early ambulation is optimal. To facilitate patient mobility, a Marcaine block of the intercostal nerves attendant to the resected rib and the ribs above and below the line of incision may be useful. A simple dry dressing is applied in the operating room but may be removed the day after surgery. Discharge can be effected on the second postoperative day in some cases. The most significant complication of surgery of aldosteronism is the persistence of hypertension. This is common in patients with idiopathic adrenal hyperplasia but uncommon in patients who have a solitary adenoma as the etiology of aldosteronism. 8 If a normal adrenal gland is present contralaterally, there should be no problem with adrenal insufficiency. However, in congenital absence of the opposite adrenal, supplemental corticosteroids and mineralocorticoids will be necessary. CHAPTER REFERENCES
1. Blevins LS Jr, Wand GS. Primary aldosteronism: An endocrine perspective. Radiology 1992;184:599–600. 2. Brunt LM, Doherty GM, Norton JA, Soper NJ, Quasebarth MA, Molet JF. Laparoscopic adrenalectomy compared to open adrenalectomy for benign adrenal neoplasms. J Am Coll Surg 1996;183:1–10. 3. Conn JW. Primary aldosteronism: A new clinical syndrome. J Lab Clin Med 1955;45:3–17. 4. Conn JW, Cohen EL, Rovner DR. Suppression of plasma renin activity in primary aldosteronism. JAMA 1964;190:213–221. 5. Conn JW, Beierwaltes WH, Lieberman LM, et al. Primary aldosteronism: Preoperative tumor visualization by scintillation scanning. J Clin Endocrinol Metab 1971;33:713–716. 6. Ikeda DM, Francis IR, Glazer GM, Amendola MA, Gross MD, Aisen AM. The detection of adrenal tumors and hyperplasia in patients with primary aldosteronism: Comparison of scintigraphy, CT and MR imaging. Am J Roentgenol 1989;153:301–306. 7. Irony I, Kater CE, Biglieri KG, Shackleton CHL. Correctable subsets of primary aldosteronism: Primary adrenal hyperplasia and renin responsive adenoma. Am J Hypertens 1990;3:576–582. 8. Obara T, Ito Y, Okamoto T, et al. Risk factors associated with postoperative persistent hypertension in patients with primary aldosteronism. Surgery 1992;112:987–993. 9. Weisnagel SJ, Gagner M, Breton G, Pomp A, Pharand D, Lacroix A. Laparoscopic adrenalectomy. Endocrinologist 1996;6:169–178. 10. Young WF Jr, Klee GG. Primary aldosteronism: Diagnostic evaluation. Endocrinol Metab Clin North Am 1988;17:367–395.

Chapter 4 Pheochromocytoma Glenn’s Urologic Surgery

Chapter 4 Pheochromocytoma
Thomas E. Keane, Pablo J. Santamaria, and Muta M. Issa

T. E. Keane: Section of Urology, Emory University School of Medicine, Atlanta, Georgia 30322. P. J. Santamaria: Section of Urology, Emory University School of Medicine, Atlanta, Georgia 30322, and Middle Georgia Urology Associates, Dublin, Georgia 31021. M. M. Issa: Department of Surgery, Division of Urology, Emory University School of Medicine, and Atlanta Veterans Affairs Medical Center, Atlanta, Georgia 30322.

Diagnosis Symptoms Biochemical Imaging Techniques Indications for Surgery Alternative Therapy Surgical Technique Preoperative Management Anesthesia Surgical Approach Thoracoabdominal and Transabdominal Approaches Postoperative Care and Specific Complications Outcomes Complications Results Chapter References

Pheochromocytomas are tumors that arise from chromaffin cells of the adrenal medulla. When such tumors arise at an extra-adrenal site, they are called paragangliomas, which can be located from the neck to the base of the pelvis ( Fig. 4-1). These tumors have an incidence of one to two per 100,000 adults and represent a curable cause of hypertension in 0.1% to 1% of hypertensive patients. 1,2 The malignancy rate is thought to be between 10% and 20%, though paragangliomas have a higher reported rate of malignancy, with over 50% reported malignant in some series. 3,4 Pheochromocytomas may be familial in 10% of cases and are associated with a variety of other conditions including the syndromeof multiple endocrine neoplasia (MEN) type II-A,II-B, von Hippel–Lindau disease, and von Reckling-hausen's disease.

FIG. 4-1. Pheochromocytoma (paraganglioma) may occur wherever neural crest (chromaffin) tissue is found. The most common sites are indicated.

The diagnosis and medical and surgical management of this condition have been evolving continuously since the first surgical resections of these lesions by C. H. Mayo in the United States and Roux in Europe. 5,6 The nature of the biochemical pathways and the diagnostic urinary studies for catecholamines and their metabolites were established by Crout et al. in the early 1960s. 7 The development of computed tomography in the 1970s provided an accurate, noninvasive method of imaging the adrenal glands and localizing these tumors, 8 as has the later development of magnetic resonance imaging. Throughout the 1980s new medications and surgical techniques were developed to control intraoperative hemodynamics, which allowed a variety of surgical approaches including laparoscopy to be utilized for resection of these lesions.

DIAGNOSIS
Symptoms Pheochromocytomas are rarely asymptomatic, though the symptoms may be varied and frequently mimic those of other conditions. Paroxysmal symp-toms (headache, diaphoresis, pallor, palpitations, andapprehension) are present in over 50% of patients. Hy-pertension, present in more than 90% of patients,may be paroxysmal in 25% to 50%. Other symptomsinclude nausea, trembling, weakness, epigastric pain, and syncope. Biochemical Pheochromocytoma or paraganglioma can be confirmed by demonstration of elevated urinary catecholamine levels (>100 µg total catecholamines per 24 hours overall, with epinephrine >20 µg and norepinephrine >80 µg, per 24 hours, respectively) or catecholamine degradation products such as metanephrines (>1.3 mg per 24 hours) and vanillymandelic acid (>6.5 mg per 24 hours); a value of >9.0 mg/24 hours determines over 90% of patients with pheochromocytomas. 1 Total catecholamines, such as epinephrine, norepinephrine, and dopamine, can also be measured in the blood. Although urinary catecholamines have a higher specificity than plasma catecholamines, either may give misleadingresults because of other medical conditions (acute alco-holism, hypothyroidism, or volume depletion) or in-terfering medications. A combined free plasma norepinephrine and epinephrine level in excess of 950 pg/ml has a diagnostic sensitivity of 94% and a specificity of 97%. The radioenzymatic assay is sensitive to the circumstances in which the blood was collected, and this should be carefully controlled by having the patient in a fasting state and supine for at least 30 minutes before blood sampling through a large-bore needle placed at least 20 minutes beforehand to avoid a spurious catecholamine elevation caused by pain or apprehension. Baseline plasma norepinephrine and epinephrine levels of more than 2,000 pg/ml (norepinephrine >2,000pg/ml, epinephrine >200 pg/ml) indicate a pheochromocytoma. Failure of these levels to decline to less than 500 pg/ml after oral clonidine is also indicative of tumor. Stimulation or provocation tests of patients suspected of this diagnosis can be extremely hazardous and are not recommended. Despite the reliance on diagnostic measurements of urinary catecholamines and their degradation products, newer agents have been tried to delineate borderline patients. Plasma catecholamines are measured after administration of clonidine or pentolinium tartrate. Patients with pheochromocytoma experience no significant decrease in circulating catecholamines with either agent, as opposed to normal patients, who do. Imaging Techniques A variety of imaging techniques are available for the detection of pheochromocytoma. Several decades ago, angiography and venography were the imaging

techniques of choice. However, along with a low sensitivity, these modalities carried a significant morbidity and a risk of provoking a hypertensive crisis if the possibility of pheochromocytoma had not been considered. The most frequent initial diagnostic modality currently is the abdominal CT scan, which, as a result of its widespread usage, has led to a significant increase in the diagnosis of asymptomatic adrenal lesions. This test has an accuracy rate for diagnosis of pheochromocytomas of over 90% and can be performed in patients who have not previously undergone a blockade, although unenhanced CT has been recommended as the initial localizing study to avoid even the small risk of precipitating a hypertensive crisis during the intravenous injection of contrast medium. 9 Computed tomography has largely replaced nephrotomography, ultrasonography, selective angiography, and venography with venous sampling. However, CT does not differentiate among adrenal lesions and benign and malignant disease. Magnetic resonance imaging (MRI) appears to be as accurate as CT in identifying adrenal lesions and also has a characteristically bright, light bulb image on T2-weighted study (Fig. 4-2). It is also very useful in the detection of recurrent local tumors in patients with metal clips and may have indications in pregnant patients. Sagittal and coronal imaging can give excellent definition of the surrounding anatomic and vascular relationships.

FIG. 4-2. Magnetic resonance appearance of pheochromocytoma showing axial T 2-weighted, axial, sagittal, and coronal T 1-weighted images.

An alternative in the search for residual or multiple pheochromocytoma is the meta-iodobenzylguanidine scan ( 131I-MIBG). This compound is taken up by adrenergic granules and adrenal medulla cells and causesvirtually no pharmacologic effects. Because MIBG is concentrated in catecholamine storage vesicles, it is quite specific for pheochromocytoma rather than just an adrenal mass: MIBG scans have a 78.4% sensitivity in primary sporadic lesions, 92.4% in malignant lesions, and 94.3% in familial cases, giving an overall 87.4% sensitivity with 99% specificity. Although it provides no anatomic detail, this test is extremely useful when CT and MRI findings are confusing (Fig. 4-3).

FIG. 4-3. 131I-MIBG scan demonstrating bilateral adrenal lesions.

INDICATIONS FOR SURGERY
Indications for surgery are an adrenal mass or extra-adrenal mass that meets the biochemical criteria for a pheochromocytoma. Additional indications include a positive MIBG scan or MRI with borderline biochemical criteria.

ALTERNATIVE THERAPY
There is no acceptable alternative therapy to the management of pheochromocytoma except surgery. Alternative approaches to the adrenal would include laparoscopic surgery, though with the high perioperative risks associated with this procedure, open surgery is by far the preferred route. In pregnant patients, oral blocking agents may be utilized until the fetus has matured, and cesarean section and tumor excision can be safely performed as one procedure, avoiding the potential stress of vaginal delivery.

SURGICAL TECHNIQUE
Preoperative Management The localization of a pheochromocytoma is essential in planning definitive therapy and is influenced by whether it is sporadic (80% solitary adrenal lesion) or familial (50% bilateral) and whether it occurs in children or in adults. Multiple or extra-adrenal lesions occur in 10% of cases in adults and up to 30% in children. Localization techniques usually involve one or more of the imaging techniques listed above. Adequate preoperative pharmacologic blockade provides a smoother and safer procedure for the surgeon, patient, and anesthesiologist. Phenoxybenzamine hydrochloride is an a-adrenergic blocker with both postsynaptic (a 1) and presynaptic (a 2) blocking capabilities. An initial divided dose of 30 to 60 mg orally is commenced. The dose is increased by 10 to 20 mg per day until the blood pressure has stabilized (maximum 100 mg/day). Recently, newer blocking agents have become available that are more selective and avoid some of the associated side effects. Patients are usually adequately blocked when they complain of postural hypotension and nasal stuffiness. b-Blockers protect against arrhythmias, control tachyphylaxis from a-blockers and permit a decrease in the amount of a-blocker necessary to control blood pressure. These agents should be used only after a-blockade has been established because, alone, they may precipitate a rise in total peripheral vascular resistance through unopposed a-adrenergic activity, and used only when cardiac arrhythmias are expected. a-Methylparatyrosine has been utilized in addition to phenoxybenzamine and/or propranolol. This agent decreases the rate of catecholamine synthesis and is particularly useful in patients who are resistant to a-blockers or have multiple paragangliomas. Preoperative preparation also requires intravenous fluid replacement to ensure adequate hydration because many patients will have a depleted intravascular volume. Unless active fluid expansion is planned, the pharmacologic blockade should be of at least 2 weeks' duration in order to allow the patient's own homeostatic

mechanisms to compensate for the recently expanded intravascular space. Crystalloids and in some cases blood transfusions may be required to accommodate the expanded intravascular volume produced by blockade, and an extra fluid load of 1 to 2 liters should be administered the night before surgery. Anesthesia The primary focus of anesthetic management of a pheochromocytoma patient is hemodynamic control. Close monitoring of the blood pressure, EKG, urinary output, and central venous pressure is essential in all phases of the procedure. An arterial line and Swan–Ganz catheter are frequently utilized. Sodium pentobarbital is usually used for induction, and virtually all inhalational agents have been administered for maintenance of anesthesia; these are usually combined with the neuromuscular blocking agents succinylcholine, d-tubocurarine, or pancuronium. The two inhalational agents of choice appear to be enflurane and isoflurane, with the latter decreasing myocardial contractility but being more resistant to metabolism and consequently less toxic. Either agent can also be combined with phentolamine or nitroprusside to control hypertension. Intraoperative arrhythmias can be controlled with either lidocaine or propranolol, although they frequently resolve following blood pressure normalization. Following interruption of the venous drainage of a pheochromocytoma, profound hypotension may ensue, and the surgeon should inform the anesthesiologist before performing this maneuver. Volume replacement is the treatment of choice, which may be augmented by the addition of vasopressors (Levophed) if necessary until the situation stabilizes. Surgical Approach There are numerous approaches to the adrenal gland, and the appropriate choice of access is governed by the size, multiplicity, and site of the lesion and the underlying pathology. In situations where a paraganglioma is a possibility, a midline abdominal incision allows a full assessment of the abdomen, retroperitoneum, and pelvis. The one essential is a detailed knowledge of the surgical anatomy of the adrenal glands ( Fig. 4-4 and Fig. 4-5). Other factors such as the body habitus of the patient and the preference and experience of the surgeon must also be considered. Therefore, in view of all these variables, it is apparent that each case should be approached individually, with account taken of the various preferred guidelines for individual diseases.

FIG. 4-4. When viewed anteriorly, the adrenals lie behind the colon, stomach, duodenum, and pancreas.

FIG. 4-5. Arterial blood supply of the adrenals may vary with multiple small arteries, whereas venous drainage is relatively constant.

The left adrenal gland is supplied by multiple small arteries superiorly originating from the inferior phrenic artery. Medially, multiple arteries arise directly from the aorta. Inferiorly, a constant artery arises either directly from the aorta just above the left renal artery or from the proximal renal artery itself. The venous drainage of the left gland is mainly through the inferior adrenal vein, which drains into the superior aspect of the left renal vein, usually just lateral to the aorta. There are virtually no blood vessels entering or draining the lateral aspect of the left adrenal under normal circumstances. Gerota's fascia can be dissected anteriorly off the posterior aspect of the pancreas and the splenic vein and artery. If an anterior approach is used, the splenorenal ligaments must be divided, and the colon reflected medially, before Gerota's fascia can be dissected, following which the spleen and pancreas can be elevated superiorly to expose the underlying anterior surface of the adrenal gland. A virtually avascular plane also exists posteriorly between Gerota's fascia and the paraspinous muscles such that the gland can be mobilized from the surrounding structures before its blood supply. The right adrenal gland is supplied superiorly by branches of the inferior phrenic artery that are often obscured by the overlying liver and inferior vena cava. Medially, small direct branches from the aorta course beneath the vena cava, and inferiorly a fairly constant branch of the proximal renal artery enters the gland. The venous drainage is again mainly through one vessel, which is short and enters directly into the vena cava just below the hepatic veins. Securing this vein is perhaps the most challenging aspect of adrenal surgery, as it is nearly always higher and shorter than expected, and adjacent vascular and fascial structures may have to be divided beforehand. In cases involving a large right-sided tumor, following reflection of the colon and duodenum it is not unusual to have to divide the caudate lobe veins entering directly into the vena cava in order to gain adequate exposure anteriorly. Once again, there are relatively avascular planes anteriorly, posteriorly, and laterally. Care should be exercised when freeing the inferior aspect of the gland, which can have vascular attachment to the upper pole of the kidney. Small well-localized adrenal lesions may be approached by either a posterior, modified posterior, or flank approach. Larger lesions including pheochromocytomas, both single and multiple, may be approached by either an abdominal or thoracoabdominal incision. These latter two options are dealt with below, and the other approaches are described elsewhere in this section. Thoracoabdominal and Transabdominal Approaches The thoracoabdominal eighth, ninth, or tenth intercostal approach is the incision preferentially utilized for large right-sided pheochromocytomas. 10 This approach offers the advantage of excellent adrenal exposure and an opportunity to palpate the thoracic sympathetic chain in the case of a rare associated paraganglioma metastasis. The peritoneal cavity may be widely opened for laparotomy, and in cases in which the contralateral gland must be explored, the incision is extended, though closure may be tedious. There is usually less requirement for this incision, with its extrapulmonary morbidity, when dealing with left-sided lesions. There is generally less vascularity to deal with, especially on the superior, lateral, and posterior aspects of the tumor, and the pancreas and spleen can be mobilized away from the lesion with ease. The patient is placed in a semioblique position on a bean bag that is rolled to elevate the relevant flank and hemithorax. For this description the lesion is assumed to be right-sided, but the approach is similar for left-sided lesions. The right arm is then draped across the chest over a Mayo stand with careful padding and positioning to avoid a stretch or pressure injury. The left axilla is protected with a pad. The pelvis should lie almost parallel, with the contralateral knee flexed 90 degrees and lying under the straight ipsilateral leg, with padding between the two g Figure 4-6). The table is then hyperextended, and the patient is fixed in position with adhesive tape. The incision is made in the eighth, ninth, or tenth intercostal space at the angle of the rib and extended across the costal margin to curve inferiorly to the

midpoint of the opposite rectus muscle. Once the latisimus dorsi, posterior inferior serratus, and the external oblique have been divided, the internal oblique is divided on the upper border of the rib itself, which need not be resected; the costvertebral ligament is divided, allowing the rib to swing down after division of the intercostal muscles. The pleura is then carefully entered, and the lung protected by a padded retractor. The diaphragm with overlying pleura will then be visualized and should be divided at the periphery about 2 cm from the chest wall in a circumferential pattern from anterior to posterior to allow for later reconstruction and to avoid damage to the phrenic nerve.

FIG. 4-6. Thoracoabdominal incision. (A) The patient is placed in a semirecumbent position using sandbags with the chest angled at 40 to 45 degrees and the pelvis almost flat. If the chest is entered through the ninth intercostal space, the incision extends from the midaxillary line across the costal margin at the intercostal space to the midline or across it just above the umbilicus. (B) The anterior rectus sheath and the external oblique and latissimus dorsi muscles are divided. (C) The intercostal muscles parallel the directions of the three abdominal layers and are divided. The costal cartilage and the internal oblique and rectus muscles are incised. (D) The pleural reflection (shaded areas) lies progressively closer to the costal margin in the more cephalic intercostal spaces. (E) The pleura, reflecting as the costophrenic sinus near the costal margin, is exposed beneath the intercostal muscles. The diaphragm can be seen inferior and dorsal to the pleura. The pleura is opened with care to avoid injuring the lung, which comes into view with inspiration. After the lung is packed away, the diaphragmatic surface of the pleura is seen. The diaphragm is incised avoiding damage to the phrenic nerve. (F) The transversus muscle is divided, exposing the peritoneum with the liver beneath it. (G) The peritoneum is divided, and a rib-spreading retractor is inserted, enabling upward displacement of the liver (or spleen on the left) into the thoracic cavity and giving wider access to the posterior peritoneum than in an anterior abdominal incision.

Heavy scissors are used to divide the costochondral junction, the peritoneum is opened, and the underlying liver can then be retracted upward. Because this incision is usually employed in large pheochromocytomas, the right triangular and coronary ligaments are divided, thus mobilizing the right lobe of the liver, which can be further retracted upward, providing excellent exposure of the suprarenal vena cava and the adjoining right adrenal vein. A fixed retractor (either Omnitract or Bookwalter) is preferred for this procedure. With the liver well protected and retracted into the chest, the posterior peritoneum lateral to the right colon is incised, and the incision is carried up along the vena cava to the level of the retracted liver edge (at the level of the hepatic veins) ( Fig. 4-7). The right colon and duodenum are mobilized medially, and the kidney is gently retracted downward to bring the adrenal into view. The attachment of the kidney to the adrenal should be preserved until the gland has been completely mobilized, as this facilitates exposure and prevents direct manipulation of the tumor. At this point care must be exercised to avoid trauma to the small veins draining directly into the vena cava from the caudate lobe. If necessary, these veins may be divided between Adson clamps and sutured below the clamps with 5-0 Proline. At this point, full retraction is instituted to expose the entire operative field.

FIG. 4-7. Anterior approach to the right adrenal gland. (A) The posterior peritoneum lateral to the right colon is incised, and the mobilization is carried along the vena cava to the level of the hepatic veins. (B) With the duodenum and colon reflected medially, the liver and gallbladder are retracted upward. Gentle downward retraction on the kidney brings the anterior surface of the right adrenal gland into view. Small veins draining the caudate lobe of the liver to the vena cava may be injured with excessive retraction, and a self-retaining ring retractor is ideal at this stage. (C) The adrenal vein should be ligated early in the procedure. In many cases, the vein may lie high and enter the cava posterolaterally. Extensive dissection of the arterial supply medially and laterally may be necessary to adequately expose the vein. Exposure is facilitated by medial and downward traction on the cava. The remaining lateral and inferior attachments are readily divided to complete the procedure. (D) In rare cases where the kidney is invaded, nephrectomy and adrenalectomy are the treatments of choice.

In pheochromocytomas, it is essential that the blood supply be isolated as soon as possible, with the adrenal vein ligated with either silver clips or 2-0 silk sutures. In cases where the vein lies far superior, extensive dissection and division of the arterial supply medially and inferiorly may be necessary to allow safe and satisfactory exposure of the vein. When there are large tumors extending medially below the vena cava, mobilization of the vena cava with division of the relevant lumbar veins, if necessary, facilitates exposure and ligation of the medial vessels between vascular clips. The vena cava can be retracted by passing vascular tapes below it to provide the necessary medial traction. If the tumor is confined to the adrenal gland, the remaining lateral and inferior attachments of the gland are mobilized and divided to complete the adrenalectomy. An alternative approach to large pheochromocytomas is the anterior transperitoneal approach. This approach also allows early vascular control and provides the opportunity for a thorough laparotomy. The optimal approach involves a bilateral subcostal or chevron incision. A midline incision is used only when an extra-adrenal pheochromocytoma is suspected in either the pelvis or retroperitoneum. This approach is particularly suitable for left-sided large pheochromocytomas. A number of approaches are available to expose the left adrenal gland. Exploration through the lesser sac or the avascular plane in the transverse mesocolon is ideal for small tumors. However, for large tumors, following laparotomy, the posterior peritoneum lateral to the left colon is incised, and the incision is extended to include the lienorenal ligament. Splenic injury is a risk during this maneuver. Once again, a fixed ring retractor is optimal in this procedure. The initial dissection involves exposure of the left adrenal vein, which is easier to approach on this side and is ligated at its entry into the renal vein with 2-0 silk ligatures. The anterior and posterior adrenal planes are then developed, and the spleen and pancreas are gently retracted superiorly. The left colon and duodenum are reflected medially. Leaving the distal ligature long permits the ligated adrenal vein to be utilized to provide retraction to facilitate ligation of the inferior adrenal artery ( Fig. 4-8). The gland is then mobilized with gentle blunt dissection on its lateral and posterior margins. Downward traction on the kidney facilitates exposure of the superior vascular ligaments, which may be tied with 3-0 silk or ligated with vascular clips. Lateral traction is then employed to expose the medial arteries and lymphatic vessels, which are ligated. Finally, the remaining inferior attachments are divided, and the adrenal gland is removed ( Fig. 4-9, Fig. 4-10 and Fig.4-11).

FIG. 4-8. Anterior approach for left adrenalectomy. (A) The adrenal vein is divided and can be used for traction, although excessive traction may provoke a sharp elevation in the blood pressure of even adequately blocked patients. (B) The kidney can be used to provide excellent traction of the adrenal, allowing the lateral attachments to be divided.

FIG. 4-9. (A) Intraoperative photograph showing IVC retraction with mobilization of the medial aspect of the right gland. (B) Postoperative appearance.

FIG. 4-10. (A) Intraoperative photograph showing ligation of the left adrenal vein as it enters the renal vein. (B) Postoperative appearance.

FIG. 4-11. Resected specimen, gross appearance, showing multiple areas of necrosis.

The adrenal fossa is carefully inspected for bleeding and, after electrocautery, packed while the surrounding viscera are carefully examined. Persistent oozing may be controlled with Surgicel or Gelfoam. The incisions are closed in a standard fashion, with no drainage utilized. Nasogastric suction may be employed for 48 hours to minimize postoperative distention. Postoperative Care and Specific Complications In the recovery room, vital signs and mental status are closely monitored. Patients with a flank, posterior, or thoracoabdominal incision should have a chest radiograph to rule out a pneumothorax or document proper positioning of a chest tube. Pain control is a major contributing factor to reduce atelectasis and promote early ambulation. Intensive care monitoring for the initial 24-hour period is prudent following removal of a pheochromocytoma.

OUTCOMES
Complications Specific complications seen during both the intra- and postoperative period include profound hemodynamic instability, which requires precise monitoring and adequate preoperative preparation. Postoperatively, large boluses of intravenous fluids with pressor support may be necessary to maintain stability. Vasospasm sufficient to reduce enteral blood flow may be encountered and should be considered along with possible neurologic complications in inadequately blocked or unrecognized cases. When dealing with large lesions, the surgeon needs to bear in mind that the renal vascular anatomy may be distorted and out of position, putting it at risk of inadvertent injury. Significant hemorrhage secondary to vena caval or renal vein lacerations may occur and require repair with 4-0 or 5-0 Proline once adequate control has been established. Left-sided lesions may be associated with pancreatic or splenic injuries resulting in postoperative bleeding and hypotension (which may be attributed to metabolic causes) or a fistula. In cases where such an injury has occurred, placement of a drain may be prudent.

Results The Lahey Clinic reported a 0% mortality in 62 patients treated for pheochromocytoma and a 25% postoperative morbidity rate in the 41 patients whose records were available. 11 However, overall, convalescence of patients undergoing adrenal surgery is reasonably benign. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Chute R, Soutter L, Kerr WS. The value of the thoracoabdominal incision in the removal of kidney tumors. N Engl J Med 1949;241:951–960. Crout JR, Pisano JJ, Sjoerdsma A. Urinary excretion of catecholamines and their metabolites in pheochromocytoma. Am Heart J 1961;61:375–381. Malone MJ, Libertino JA, Tsapatsaris NP, Woods OB. Preoperative and surgical management of pheochromocytoma. Urol Clin North Am 1989;16(3):567–582. Mayo CH. Paroxysmal hypertension with tumor of retroperitoneal nerve. JAMA 1927;89:1047–1050. Radin DR, Ralls PW, Boswell WD, et al. Pheochromocytoma: detection by unenhanced CT. Am J Urol 1986;146:741–744. Roux C. Thesis Lausanne. Cited in Welbourne RB. Early surgical history of phaeochromocytoma. Br J Surg 1987;74:594–596. Samaan NA, Hickey RC, Shutts PE. Diagnosis, localization, and management of pheochromocytoma: pitfalls and follow up in 41 patients. Cancer 1988;62:2451–2460. Sheps SG, Jiang N-S, Klee GG. Diagnostic evaluation of pheochromocytoma. Endocrinol Metab Clin North Am 1988;17:397–414. St John Sutton MG, Sheps SG, Lie JT. Prevalence of clinically unsuspected pheochromocytoma: Review of a 50 year autopsy series. Mayo Clin Proc 1981;56:354–360. Stewart BM, Brevo EL, Haaga J. Localization of tumor by computed tomography. N Engl J Med 1978;299:460–461. Whalen RK, Althausen AF, Daniels GH. Extra-adrenal pheochromocytoma. J Urol 1992;147:1.

Chapter 5 Simple Nephrectomy Glenn’s Urologic Surgery

Chapter 5 Simple Nephrectomy
W. Holt Sanders and Cragin Anderson

W. H. Sanders: Department of Surgery, Section of Urology, Emory University School of Medicine, Atlanta, Georgia 30322. C. Anderson: Atlanta Urological Institute, Riverdale, Georgia 30274.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Anatomy Preoperative Evaluation Subcostal Flank Incision Eleventh and 12th Rib Incision Subcostal Abdominal Incision Vertical Abdominal Incision Nephrectomy Subcapsular Approach Closure Outcomes Complications Results Chapter References

The term “simple nephrectomy” describes the technique of removing the kidney from within Gerota's fascia and in no manner is meant to indicate that the operation is technically easy. This procedure is usually performed in the setting of a nonneoplastic disease state but is the end operation after other therapies (including surgery) have failed, making this operation technically challenging.

DIAGNOSIS
Usually the indications for simple nephrectomy are for (a) trauma that is so severe that reconstruction is not possible, (b) a nonfunctioning kidney associated with hypertension or nephrolithiasis, or (c) a severe infectious process that cannot be cleared medically. The diagnostic studies, therefore, will depend on the clinical setting. Renal trauma and the diagnostic studies are discussed in Chapter 13. The diagnosis of a nonfunctioning kidney is made by the absence of a nephrogram on excretory urogram or CT scan or by the absence of signal on a radionuclide scan in a kidney that is not obstructed. It is occasionally necessary to place a ureteral stent or a nephrostomy tube to relieve an obstruction so that lack of function can be documented by contrast or radionuclide study. Severe infections such as xanthogranulomatous pyelonephritis or emphysematous pyelonephritis are often treated with simple nephrectomy. Patients with xanthogranulomatous pyelonephritis, an uncommon form of chronic renal infection, present with flank pain, fever, persistent bacteriuria, and flank mass. The CT reveals a renal mass associated with a calcification in the renal pelvis. The parenchyma is replaced by small water-density masses. Xanthogranulomatous pyelonephritis is often misdiagnosed preoperatively as renal cell carcinoma and correctly identified only after nephrectomy, by the pathologist. Emphysematous pyelonephritis is a complication of acute pyelonephritis in patients with diabetes. Patients with this condition present with severe pyelonephritis that does not resolve within the first 3 days of treatment. The diagnosis is made by a plain abdominal radiograph, which reveals small gas bubbles radiating in a radial distribution through the renal parenchyma.

INDICATIONS FOR SURGERY
Simple nephrectomy is indicated for nonneoplastic diseases of the kidney. The more common specific indications include severe trauma, renal infections (e.g., xanthogranulomatous pyelonephritis and emphysematous pyelonephritis), nonfunctioning kidneys with stru-vite stones or obstruction, renal vascular hypertension (when all attempts at medical and surgical therapy have failed), and renal fistula.

ALTERNATIVE THERAPY
Alternatives to simple nephrectomy include partial nephrectomy, renal embolization, laparoscopic nephrectomy, and radical nephrectomy.

SURGICAL TECHNIQUE
Anatomy The right kidney is lower than the left kidney in 90% of patients as a result of downward displacement by the liver. The renal arteries run posterior to the renal veins, and the right renal artery runs posterior to the inferior vena cava. The renal arteries divide into segmental branches at the junction of the middle and final third of their course. This anatomic point is important to keep in mind if the vessels are divided close to the renal hilus, as in a subcapsular nephrectomy. The left renal vein receives tributaries from the phrenic vein, the adrenal vein, the gonadal vein, and occasionally the lumbar vein. The right renal vein usually receives no tributaries. Anomalies of the renal vasculature are present in three-quarters of all patients. 2 Departures from the anatomy presented in the textbooks usually involve supernumerary arteries rather than the veins. These additional arteries commonly supply the lower pole of the kidney. 5 Preoperative Evaluation The preoperative evaluation has two purposes: to minimize risk and to determine the optimal incision. These two purposes are interrelated, and occasionally the primary reason for choosing one type of incision over another is to minimize risk to the patient. It is essential to document function in the contralateral kidney. For most cases, a serum creatinine and an IVP are adequate. In indeterminate cases, a nuclear renogram may be required to demonstrate sufficient renal function. The scout film of the IVP is useful for determining the level of a flank incision. Note which rib is superimposed on the middle of the lateral border of the kidney. An appropriate flank incision should be made at the level of this rib or above. The surgeon should inquire about a history of pulmonary disease. The decubitus position with an elevated kidney rest can decrease the vital capacity by 20%. In the decubitus position, there is also preferential ventilation of the upper lung and perfusion of the lower lung, creating a ventilation–perfusion mismatch. A history of severe pulmonary disease will therefore favor an abdominal approach. Exploration of the kidney in the setting of traumatic injury must be done through an abdominal approach. In an obese patient, a flank approach optimizes exposure and minimizes wound complications. Previous abdominal surgery also favors a flank approach. An extraperitoneal flank approach is preferable in a patient with a chronically infected kidney. In other cases, the choice of incision depends largely on the surgeon's preference.

Subcostal Flank Incision The patient is placed on the operating table so that the kidney rest is just cephalad to the anterior superior iliac spine. The patient is turned to the lateral decubitus position with his or her back toward the edge of the table. The contralateral leg is flexed and padded at the knee and ankle. The ipsilateral leg is appropriately padded with pillows and kept only gently flexed. The table is then flexed, and the kidney rest elevated. The patient should then be secured to the table with 2-inch tape over the patient's hip. The patient should have an axillary roll placed to avoid brachial plexus injury, and upper extremities should be secured to arm board and sling support or Mayo stand. Some find a bean bag to be useful in holding the patient's position. It is important to remember that the patient should be positioned with the table flexed before the bean bag is inflated. The incision is made approximately 2 cm inferior to the 12th rib starting posterior to the angle of the 12th rib or at the inferior border of the paraspinous muscles. The incision usually is gently curved toward the umbilicus to the lateral edge of the rectus muscle. The latissimus dorsi and external oblique are divided with cautery, exposing the serratus posterior inferior and the internal oblique, which are then divided Figure 5-1). Often the subcostal nerve emerges from the fibers of the lumbodorsal fascia to extend along a course that is superficial to the transversus muscle. Careful proximal and distal dissection of this nerve will minimize the risk of injury. A small incision in the lumbodorsal fascia provides access to the retroperitoneum. The peritoneum is dissected medially off the transversalis fascia with blunt dissection. The transversus can then be divided with cautery or bluntly divided between the muscle fibers. Gerota's fascia is identified beneath the paranephric fat and is then incised. The kidney is dissected free from the surrounding perinephric fat using blunt and sharp dissection. The renal pedicle can be approached anteriorly or posteriorly.

FIG. 5-1. (A) The subcostal incision begins below the tip of the 12th rib and extends toward the umbilicus, ending at the rectus and running parallel to the rib. (B) The external oblique edges of the latissimus dorsi muscles are divided in the direction of the incision. The 12th nerve usually is seen beneath this layer as it penetrates the lumbodorsal fascia. The nerve can be freed from its subcostal tunnel medially by sharp dissection until it is slack enough to be drawn out of the way (see D). (C) The internal oblique muscle is incised and can be seen to originate from the posterior layer of the lumbodorsal fascia. (D) The transversus abdominis muscle has the same origin and can be slit in the direction of its fibers. Care is taken to separate the peritoneum bluntly from the transversalis fascia, which limits the transversus internally. (E) Retroperitoneal (paranephric) fat lies under the lumbodorsal fascia. Beneath is a smooth envelope of paranephric fascia (Gerota's). Paranephric fat is less lobular and lighter yellow. The kidney can be dissected with Gerota's fascia widely opened. If a tumor is present, fat and fascia are taken with the kidney.

If additional exposure is required, the costovertebral ligament of the 12th rib can be divided bluntly or sharply, allowing for greater cephalad retraction. Eleventh and 12th Rib Incision If the kidney appears high in relation to the thoracic cage on preoperative radiographic studies, an 11th or 12th rib resection may be preferred. The patient is positioned as detailed above for the flank subcostal approach. The incision is made over the selected rib from the costovertebral angle over the tip of the rib medially to the edge of the rectus muscle (Figure 5-2).

FIG. 5-2. Technique of 11th-rib resection. (A and B) The incision is made over the 11th rib.

Once the rib is exposed, the periosteum is incised along the length of the rib. The periosteum is dissected off the rib using the periosteal elevator and the Alexander periosteotome. The Doyen periosteal elevator is guided beneath the rib to complete the dissection posteriorly. Once free, the rib can be divided with a rib cutter, and the edges smoothed with a rongeur (Fig. 5-3).

FIG. 5-3. Rib resection technique. (A) After exposure of the rib, the intercostal muscles are stripped from the upper and lower rib surfaces with an Alexander Farabeuf costal periosteotome. (B) The rib is freed from the periosteum by subperiosteal resection or from the pleura with a periosteum elevator. (C) A Doyen costal elevator is slipped beneath the rib to free it. The proximal and distal portions of the rib are immobilized with Kocher's clamps, and the rib is divided proximal to its angle with a right-angled rib cutter. (D) The costal cartilage is cut free with scissors. The cut surface of the rib is inspected for spicules, which are removed with a rongeur.

The posterior periosteum is divided, exposing the fascial attachments of the pleura to the diaphragm Figure 5-4). These attachments are sharply incised so that the pleura can be reflected superiorly. The peritoneum is bluntly dissected from the deep surface of the transversalis fascia by sweeping it medially with the fingers. The medial extent of the incision, including the external oblique, the internal oblique, and the transversus, can now be completed. Gerota's fascia is incised, and the kidney is dissected free from the surrounding peri-nephric fat using blunt and sharp dissection ( Fig. 5-5).The flank incisions enhance the ease of a posterior approach to the pedicle without increasing the difficulty of the anterior approach.

FIG. 5-4. (A) After division of the latissimus and external oblique muscles, subperiosteal resection of the rib is performed. (B) The incision is carried through the periosteum posteriorly and the internal oblique and transversus muscles medially, exposing the paranephric space. A tongue of pleura lies in the upper portion of the wound. Diaphragmatic slips that come into view are divided, and the pleura can be retracted upward. (C) The paranephric fat is dissected bluntly.

FIG. 5-5. (A) Gerota's fascia is incised and entered. In nephrectomy for renal donation, a sharp technique is used to dissect the paranephric fat. (B) The renal vein is exposed to its entrance into the vena cava anteriorly. The tissue between the ureter laterally and the vena cava medially is dissected free, with care taken to preserve the periureteral blood supply. The hilar region of the kidney is avoided in dissection. (C) The kidney is rotated anteromedially, and the renal artery is isolated as far as possible. The ureter is divided as far inferiorly as possible.

Subcostal Abdominal Incision The subcostal abdominal incision is preferred by some surgeons because of: 1. Early exposure of the renal pedicle 2. Lower risk of inadvertent pleurotomy 3. Decreased effect on ventilation in patients with pulmonary disease The patient is positioned with the table break at the level of the 12th rib, and the operative side is elevated with a rolled sheet. The table is then flexed to maximize exposure. The incision is typically two fingerbreadths below the costal margin with its medial extent being approximately two fingerbreadths below the xyphoid process. After the skin incision, the anterior rectus fascia is divided along with the rectus muscle and the external oblique. The superior epigastric artery is divided. The internal oblique is divided. The lumbodorsal fascia is incised laterally, and the peritoneum can be opened or bluntly stripped off the anterior abdominal wall. The transversus can then be divided with cautery or bluntly divided between the muscle fibers. If peritoneum is opened, one must reflect the colon medially to expose Gerota's fascia, which is then incised. The kidney is dissected free from the surrounding perinephric fat using blunt and sharp dissection. The anterior approach to the renal pedicle is easier than the posterior approach when a subcostal abdominal incision is used. Vertical Abdominal Incision If a vertical abdominal incision is required for the evaluation of intraperitoneal structures (such as in a patient who has had abdominal trauma) or for a combined procedure, a simple nephrectomy can be performed through a midline incision. This incision is typically from the xyphoid process to the pubic symphysis. After incision of the skin and subcutaneous fat, the linea alba is identified and incised. The peritoneum can be identified beneath preperitoneal fat and is incised sharply and carefully to avoid bowel injury. The colon is reflected medially to expose Gerota's fascia. In a patient who has suffered renal trauma, it is important to obtain early vascular control by dissecting along the aorta for a left renal injury and along the inferior vena cava for a right renal injury. The dissection is carried superiorly to the level of the renal vessels. Vessel loops are placed around the renal artery and vein before exploration of the injured kidney. The paramedian incision is helpful if an attempt will be made to stay extraperitoneal or if a two-layer closure is preferred. It is typically two fingerbreadths lateral to midline. The rectus muscle fibers are dissected off the linear alba and retracted laterally. If peritoneum is opened, one must reflect the colon medially. Gerota's fascia is then incised, and the kidney is dissected free from the surrounding perinephric fat using blunt and sharp dissection. Nephrectomy After Gerota's fascia is incised and the kidney is dissected free from surrounding perinephric fat, the renal artery should be identified. One must keep in mind possible aberrant vessels, particularly lower-pole branches. The renal artery can usually be identified during posterior dissection of the pedicle. Ligation of the artery before the vein prevents renal congestion and is thus preferred. Two size-0 silk ties are placed proximally, and a single silk is placed distally. The artery is divided with scissors; a scalpel is used when there is minimal distance between the proximal and distal ligatures. To minimize the possibility that the proximal tie will slip off the arterial stump, some surgeons place a suture ligature distal to the 0 silk ties. The ureter is quickly identified by blunt dissection in the fat inferior to the kidney. It is divided between ligatures or clips. The connective tissue and lymphatics are dissected off the kidney, revealing the renal vein. On the left, particular attention is paid to the gonadal vein, inferior adrenal vein, and lumbar venous branches. These branches are divided between silk ties if distal to the area dissected. The renal vein is doubly ligated, as was the artery. The adrenal gland can be dissected off with sharp dissection, taking care to clip all vessels. If the nephrectomy is secondary to an infectious process, a drain is left in

the posterior flank. Subcapsular Approach In patients undergoing simple nephrectomy for stone disease or for infection, previous surgery or chronic inflammation can make dissection very difficult. In these cases, it is advantageous to come down to the capsule, incise it, and continue the dissection under the capsule to the hilus. It is important to remember that the renal vessels have already divided into several branches once they reach the renal hilum and to continue searching for additional arterial branches once the apparent main branch has been divided. Closure There are differing opinions on the best technique for closure of a flank wound, although there is general agreement that the abdominal portion of a flank wound should be closed in two layers. The bean bag is deflated, the kidney rest is lowered, and the flexion is taken out of the table. The closure should be initiated at each end of the incision and continued toward the middle of the incision. Anteriorly, the internal oblique is closed with a running PDS suture. In the posterior portion of the wound, the inferiorly reflected periosteum is approximated to the periosteum and intercostal muscle of the superior rib. When the rib has been resected, the periosteum and intercostal muscles above and below the rib are approximated. The latissimus dorsi fascia is then closed in continuity with the external oblique fascia using a running PDS suture. A single-layer closure is often sufficient over the ribs. A single running PDS suture closing the fascia of the external oblique and continuing posterior to close the fascia of the latissimus dorsi has resulted in one hernia in approximately 700 donor nephrectomies at our institution.

OUTCOMES
Complications The operative mortality of nephrectomy for benign disease is less than 1%. 4 The most common intraoperative problem is hemorrhage, especially of the renal vein and vena cava. It is important to avoid blind clamping and suture ligatures that can lead to an arteriovenous fistula. 6 The proper strategy is to gain control by direct pressure on the vena cava with sponge sticks, followed by optimization of exposure. The surgeon can then use a running 5-0 vascular suture to repair the vessel. Blind clamping can also lead to duodenal injury in a right nephrectomy. It is important to reflect the duodenum medially before division of the vessels from an anterior approach. The superior mesenteric artery is vulnerable to injury during a left nephrectomy. Unintentional laceration of the parietal pleura is common and can usually be repaired without placement of a chest tube. The edges of the pleura are approximated with a running absorbable suture. A red rubber catheter is placed through the laceration into the pleural cavity before the suture is tied. The end of the catheter is placed in a basin of water. The anesthesiologist gives the patient a deep breath; air is expelled from the pleural cavity; the catheter is removed; and the suture is tied. After such a maneuver, it is important to obtain a chest x-ray in the recovery room. A small pneumothorax will be reabsorbed without sequelae. A larger pneumothorax may benefit from aspiration of air from the pleural cavity using a large luer lock syringe, a stopcock, and an intravenous angiocath. Because the pulmonary parenchyma is not injured, it is rarely necessary to insert a chest tube. Inability to reapproximate the pleural edges in an airtight fashion may necessitate placement of a chest tube. Atelectasis is common after a simple nephrectomy even if the pleural cavity has not been entered. A flank bulge is common after a nephrectomy through the flank approach, especially if the subcostal nerve has been injured. The nerve lies below the internal oblique muscle and above the transversus abdominus muscle. Careful identification, proximal and distal dissection, and gentle retraction of the nerve can minimize this problem. Flank bulges must be distinguished from incisional hernias, which are rare. A fascial defect is usually palpable in patients with a hernia. Results Generally, the patients recover from simple nephrectomies uneventfully and remain in the hospital for less than a week, depending on the indication for the nephrectomy, the comorbidities, and the patient's preoperative status. Success rates with improved control of hypertension are as high as 86% in patients with unilateral atherosclerotic disease of the renal artery. 1 Many of these patients continue to require antihypertensive medications, although at lower doses. The success in the treatment of xanthogranulomatous pyelonephritis approaches 100%. In contrast, patients with emphysematous pyelonephritis have a mortality rate as high as 43% despite aggressive intervention with nephrectomy. 3 CHAPTER REFERENCES
1. Andersen GS, Gadsboll N, McNair A, et al. Treatment of renovascular hypertension by unilateral nephrectomy. A follow-up study in patients above 60 years of age. Scand J Urol Nephrol 1986;20:51–56. 2. Anson BJ, Kurth L. Common variations in the renal blood supply. Surg Gynecol Obstet 1955;100:157–160. 3. Freiha FS, Messing EM, Gross DM. Emphysematous pyelonephritis. J Contin Educ Urol 1979;18:9. 4. Scott RF Jr, Selzman HM. Complications of nephrectomy: review of 450 patients and a description of the modification of the transperitoneal approach. J Urol 1966;95:307–312. 5. Sykes D. The arterial supply of the human kidney with special reference to accessory renal arteries. Br J Surg 1963;50:368–370. 6. Yeates WK. Post-nephrectomy arteriovenous fistula. Proc R Soc Med 1967;60:112–115.

Chapter 6 Partial Nephrectomy Glenn’s Urologic Surgery

Chapter 6 Partial Nephrectomy
Andrew C. Novick

A. C. Novick: Department of Urology, The Cleveland Clinic Foundation, Cleveland, Ohio 44195.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Anatomic Considerations Timing of Surgery in Bilateral Tumors General Operative Considerations Segmental Polar Nephrectomy Wedge Resection Transverse Resection Simple Enucleation Extracorporeal Partial Nephrectomy and Autotransplantation Partial Nephrectomy in Duplex Collecting Systems Postoperative Follow-Up Outcomes Complications Results Chapter References

Recent interest in partial nephrectomy or nephron-sparing surgery for renal cell carcinoma has been stimulated by advances in renal imaging, improved surgical techniques, the increasing number of incidentally discovered low-stage renal cell carcinomas, and good long-term survival in patients undergoing this form of treatment. Partial nephrectomy entails complete local resection of a renal tumor while leaving the largest possible amount of normal functioning parenchyma in the involved kidney.

DIAGNOSIS
Evaluation of patients with renal cell carcinoma for partial nephrectomy should include preoperative testing to rule out locally extensive or metastatic disease. For most patients, preoperative renal arteriography to delineate the intrarenal vasculature aids in excising the tumor with minimal blood loss and damage to adjacent normal parenchyma. This test can be deferred in patients with small peripheral tumors. Selective renal venography is performed in patients with large or centrally located tumors to evaluate for intrarenal venous thrombosis secondary to malignancy. The latter, if present, implies a more advanced local tumor stage and also increases the technical complexity of tumor excision.

INDICATIONS FOR SURGERY
Accepted indications for partial nephrectomy in malignancies include situations in which radical nephrectomy would render the patient anephric with subsequent immediate need for dialysis. 1 This encompasses patients with bilateral renal cell carcinoma or renal cell carcinoma involving a solitary functioning kidney. The latter circumstance may be present as a result of unilateral renal agenesis, prior removal of the contralateral kidney, or irreversible impairment of contralateral renal function and is seen in patients with unilateral renal cell carcinoma and a functioning opposite kidney, when the opposite kidney is affected by a condition that might threaten its future function such as calculus disease, chronic pyelonephritis, renal artery stenosis, ureteral reflux, or systemic diseases such as diabetes and nephrosclerosis. Partial nephrectomy is also indicated in selected patients with localized benign pathology of the kidney. The indications include (a) hydronephrosis with parenchymal atrophy or atrophic pyelonephritis in a duplicated renal segment, (b) calyceal diverticulum complicated by infection or stones or both, (c) calculus disease with obstruction of the lower-pole calyx or segmental parenchymal disease with impaired drainage, (d) renovascular hypertension from segmental parenchymal damage or noncorrectable branch renal artery disease, (e) traumatic renal injury with irreversible damage to a portion of the kidney, and (f) removal of a benign renal tumor such as an angiomyolipoma or oncocytoma.

ALTERNATIVE THERAPY
Alternatives to partial nephrectomy include simple nephrectomy and radical nephrectomy.

SURGICAL TECHNIQUE
Anatomic Considerations Figure 6-1 illustrates the normal renal arterial supply. The kidney has four constant vascular segments, which are termed apical, interior, posterior, and basilar. Each of these segments is supplied by one or more major arterial branches. Though the origin of the branches supplying these segments may vary, the anatomic positions of the segments are constant. All segmental arteries are end arteries with no collateral circulation; therefore, all branches supplying tumor-free parenchyma must be preserved to avoid devitalization of functioning renal tissue.

FIG. 6-1. Normal arterial supply for the right kidney with anterior and posterior views.

The normal renal venous anatomy is depicted in Fig. 6-2 (for the left kidney). The renal venous drainage system differs significantly from the arterial blood supply in that the intrarenal venous branches intercommunicate freely among the various renal segments. Ligation of a branch of the renal vein, therefore, will not result in segmental infarction of the kidney because collateral venous blood supply will provide adequate drainage. This is important clinically because it enables one to obtain surgical access safely to tumors in the renal hilus by ligating and dividing small adjacent or overlying venous branches. This allows major venous branches to be

completely mobilized and freely retracted in either direction to expose the tumor with no vascular compromise of uninvolved parenchyma ( Fig. 6-3).

FIG. 6-2. Renal venous anatomy depicted here for the left kidney.

FIG. 6-3. Mobilization of the left renal vein to obtain better exposure of the renal hilus by ligating and dividing small renal venous branches.

Timing of Surgery in Bilateral Tumors In patients with bilateral synchronous renal cell carcinoma, the kidney more amenable to a partial nephrectomy is usually approached first by the author. Then, approximately 1 month after a technically successful result has been documented, radical nephrectomy or a second partial nephrectomy is performed on the opposite kidney. Staging surgery in this fashion obviates the need for temporary dialysis if ischemic renal failure occurs following nephron-sparing excision of renal cell carcinoma. General Operative Considerations It is usually possible to perform partial nephrectomy for malignancy in situ by using an operative approach that optimizes exposure of the kidney and by combining meticulous surgical technique with an understanding of the renal vascular anatomy in relation to the tumor. We employ an extraperitoneal flank incision through the bed of the 11th or 12th rib for almost all of these operations; we occasionally use a thoracoabdominal incision for very large tumors involving the upper portion of the kidney. These incisions allow the surgeon to operate on the mobilized kidney almost at skin level and provide excellent exposure of the peripheral renal vessels. With an anterior subcostal transperitoneal incision, the kidney is invariably located in the depth of the wound, and the surgical exposure is simply not as good. In partial nephrectomy for benign disease, the preferred surgical approach is usually through an extraperitoneal flank incision except for cases of renal trauma, which are best approached anteriorly. When in situ partial nephrectomy is performed for malignancy, the kidney is mobilized within Gerota's fascia while the perirenal fat around the tumor is left intact. For small peripheral renal tumors, it may not be necessary to control the renal artery. In most cases, however, partial nephrectomy is most effectively performed after temporary renal arterial occlusion. This measure not only limits intraoperative bleeding but, by reducing renal tissue turgor, also improves access to intrarenal structures. In most cases, we believe that it is important to leave the renal vein patent throughout the operation. This measure decreases intraoperative renal ischemia and, by allowing venous back bleeding, facilitates hemostasis by enabling identification of small transected renal veins. In patients with centrally located tumors, it is helpful to occlude the renal vein temporarily to minimize intraoperative bleeding from transected major venous branches. When the renal circulation is temporarily interrupted, in situ renal hypothermia is used to protect against postischemic renal injury. Surface cooling of the kidney with ice slush allows up to 3 hours of safe ischemia without permanent renal injury. An important caveat with this method is to keep the entire kidney covered with ice slush for 10 to 15 minutes immediately after occluding the renal artery and before commencing the partial nephrectomy. This amount of time is needed to obtain core renal cooling to a temperature (approximately 20°C) that optimizes in situ renal preservation. During excision of the tumor, invariably large portions of the kidney are no longer covered with ice slush, and, in the absence of adequate prior renal cooling, rapid rewarming and ischemic renal injury can occur. Cooling of the kidney by perfusion with a cold solution instilled via the renal artery is not recommended because of the theoretical risk of tumor dissemination. Mannitol is given intravenously 5 to 10 minutes before temporary renal arterial occlusion. Systemic or regional anticoagulation to prevent intrarenal vascular thrombosis is not necessary. A variety of surgical techniques are available for performing partial nephrectomy in patients with malignancy. These include simple enucleation, polar segmental nephrectomy, wedge resection, transverse resection, and extracorporeal partial nephrectomy with renal autotransplantation. All of these techniques require adherence to basic principles of early vascular control, avoidance of ischemic renal damage, complete tumor excision with free margins, precise closure of the collecting system, careful hemostasis, and closure or coverage of the renal defect with adjacent fat, fascia, peritoneum, or Oxycel. Whichever technique is employed, the tumor is removed with a surrounding margin of grossly normal renal parenchyma. Special equipment that is utilized in partial nephrectomy may include intraoperative ultrasound which is very helpful in achieving accurate tumor localization, particularly for intrarenal lesions that are not visible or palpable from the external surface of the kidney. 3 The argon beam coagulator is a useful adjunct for achieving hemostasis on the transected renal surface. If possible, the renal defect created by the excision is closed as an additional hemostatic measure. A retroperitoneal drain is always left in place for at least 7 days. An intraoperative ureteral stent is placed only when major reconstruction of the intrarenal collecting system has been performed. In patients with renal cell carcinoma, partial nephrectomy is contraindicated in the presence of lymph node metastasis because the prognosis for these patients is poor. Enlarged or suspicious-looking lymph nodes should be biopsied before the renal resection is begun. When partial nephrectomy is performed, after excision of all gross tumor, absence of malignancy in the remaining portion of the kidney should be verified intraoperatively by frozen-section examinations of biopsy specimens obtained at random from the renal margin of excision. It is unusual for such biopsies to demonstrate residual tumor, but, if so, additional renal tissue must be excised. Segmental Polar Nephrectomy In a patient with malignancy confined to the upper or lower pole of the kidney, partial nephrectomy can be performed by isolating and ligating the segmental apical or basilar arterial branch while allowing unimpaired perfusion to the remainder of the kidney from the main renal artery. This procedure is illustrated in Fig. 6-4 for a tumor confined to the apical vascular segment. The apical artery is dissected away from the adjacent structures, ligated, and divided. Often, a corresponding venous branch is present, which is similarly ligated and divided. An ischemic line of demarcation will then generally appear on the surface of the kidney and will outline the segment to be excised. If this area is not obvious, a few milliliters of methylene blue can be directly injected distally into the ligated apical artery to better outline the

limits of the involved renal segment. An incision is then made in the renal cortex at the line of demarcation, which should be at least 1 cm away from the visible edge of the cancer. The parenchyma is divided by sharp and blunt dissection, and the polar segment is removed. In cases of malignancy, it is not possible to preserve a strip of capsule beyond the parenchymal line of resection for use in closing the renal defect.

FIG. 6-4. Technique of segmental polar nephrectomy for a tumor confined to the apical vascular renal segment.

Often a portion of the collecting system will have been removed with the cancer during a segmental polar nephrectomy. The collecting system is carefully closed with interrupted or continuous 4-0 chromic sutures to ensure a watertight repair. Small transected blood vessels on the renal surface are identified and ligated with shallow figure-of-eight 4-0 chromic sutures. The edges of the kidney are reapproximated as an additional hemostatic measure, using simple interrupted 3-0 chromic sutures inserted through the capsule and a small amount of parenchyma. Before these sutures are tied, perirenal fat or Oxycel can be inserted into the defect for inclusion in the renal closure. If the collecting system has been entered, a Penrose drain is left in the perinephric space. When an apical or basilar partial nephrectomy is performed for benign disease, the segmental apical or basilar arterial branch is secured, and the parenchyma is divided at the ischemic line of demarcation, without the need for temporary renal arterial occlusion. More complex transverse or wedge renal resections are best performed with temporary renal arterial occlusion and ice slush surface hypothermia. The technical aspects of partial nephrectomy for benign disease are otherwise the same as those described for malignancy, with adherence to the same basic principles of appropriate vascular control, avoidance of ischemic renal damage, precise closure of the collecting system, careful hemostasis, and closure or coverage of the renal defect. In benign conditions necessitating partial nephrectomy, however, the renal capsule is excised and reflected off the diseased parenchyma for subsequent use in covering the renal defect. Wedge Resection Wedge resection is an appropriate technique for removing peripheral tumors on the surface of the kidney, particularly ones that are larger or not confined to either renal pole. Because these lesions often encompass more than one renal segment, and because this technique is generally associated with heavier bleeding, it is best to perform wedge resection with temporary renal arterial occlusion and surface hypothermia. In performing a wedge resection, the tumor is removed with a 1-cm surrounding margin of grossly normal renal parenchyma ( Fig. 6-5). The parenchyma is divided by a combination of sharp and blunt dissection. Invariably, the tumor extends deeply into the kidney, and the collecting system is entered. Often, prominent intrarenal vessels are identified as the parenchyma is being incised. These may be directly suture-ligated at that time, while they are most visible. After excision of the tumor, the collecting system is closed with interrupted or continuous 4-0 chromic sutures. Remaining transected blood vessels on the renal surface are secured with figure-of-eight 4-0 chromic sutures. Bleeding at this point is usually minimal, and the operative field can be kept satisfactorily clear by gentle suction during placement of hemostatic sutures.

FIG. 6-5. Technique of wedge resection for a tumor on the midlateral aspect of the kidney.

The renal defect can be closed in one of two ways. The kidney may be closed upon itself by approximating the transected cortical margins with simple interrupted 3-0 chromic sutures, after placing a small piece of Oxycel at the base of the defect. If this is done, there must be no tension on the suture line and no significant angulation or kinking of blood vessels supplying the kidney. Alternatively, a portion of perirenal fat may simply be inserted into the base of the renal defect as a hemostatic measure and sutured to the parenchymal margins with interrupted 4-0 chromic. After closure or coverage of the renal defect, the renal artery is unclamped, and circulation to the kidney is restored. A Penrose drain is left in the perinephric space. Transverse Resection A transverse resection is done to remove large tumors that extensively involve the upper or lower portion of the kidney. This technique is performed using surface hypothermia after temporary occlusion of the renal artery ( Fig. 6-6). Major branches of the renal artery and vein supplying the tumor-bearing portion of the kidney are identified in the renal hilus, ligated, and divided. If possible, this should be done before temporarily occluding the renal artery to minimize the overall period of renal ischemia.

FIG. 6-6. Technique of transverse resection for a tumor involving the upper half of the kidney.

After the renal artery has been occluded, the parenchyma is divided by blunt and sharp dissection, leaving a 1-cm margin of grossly normal tissue around the tumor. Transected blood vessels on the renal surface are secured as previously described, and the hilus is inspected carefully for remaining unligated segmental vessels. An internal ureteral stent may be inserted if extensive reconstruction of the collecting system is necessary. If possible, the renal defect is sutured together with one of the techniques previously described. If this suture cannot be placed without tension or without distorting the renal vessels, a piece of peritoneum or perirenal fat is sutured in place to cover the defect. Circulation to the kidney is restored, and a Penrose drain is left in the perirenal space. Simple Enucleation Some renal cell carcinomas are surrounded by a distinct pseudocapsule of fibrous tissue. The technique of simple enucleation implies circumferential incision of the renal parenchyma around the tumor simply and rapidly at any location, often with no vascular occlusion, and with maximal preservation of normal parenchyma. Initial reports indicated satisfactory short-term clinical results after enucleation with good patient survival and a low rate of local tumor recurrence. However, most recent studies have suggested a higher risk of leaving residual malignancy in the kidney when enucleation is performed. These latter reports include several carefully done histopathologic studies that have demonstrated frequent microscopic tumor penetration of the pseudocapsule that surrounds the neoplasm. These data indicate that it is not always possible to be assured of complete tumor encapsulation before surgery. Local recurrence of tumor in the treated kidney is a grave complication of partial nephrectomy for renal cell carcinoma, and every attempt should be made to prevent it. Therefore, it is the author's view that a surrounding margin of normal parenchyma should be removed with the tumor whenever possible. This provides an added margin of safety against the development of local tumor recurrence and, in most cases, does not appreciably increase the technical difficulty of the operation. The technique of enucleation is currently employed only in occasional patients with von Hippel Lindau disease who have multiple low-stage encapsulated tumors involving both kidneys. Extracorporeal Partial Nephrectomy and Autotransplantation Extracorporeal partial nephrectomy for renal cell carcinoma with autotransplantation of the renal remnant was initially described to facilitate excision of large complex tumors involving the renal hilus. Reconstruction of kidneys with renal cell carcinoma as well as renal artery disease may also be facilitated with this approach. The advantages of an extracorporeal approach include optimum exposure, a bloodless surgical field, the ability to perform a more precise operation with maximum conservation of renal parenchyma, and a greater protection of the kidney from prolonged ischemia. Disadvantages of extracorporeal surgery include longer operative time with the need for vascular and ureteral anastomoses and an increased risk of temporary and permanent renal failure; the latter presumably reflects a more severe intraoperative ischemic insult to the kidney. Although some urologic surgeons (including the author) have found that almost all patients undergoing partial nephrectomy for renal cell carcinoma can be managed satisfactorily in situ, others have continued to recommend an extracorporeal approach for selected patients. Extracorporeal partial nephrectomy and renal autotransplantation are generally performed through a single midline incision. The kidney is mobilized and removed outside Gerota's fascia with ligation and division of the renal artery and vein as the last steps in the operation. Immediately after division of the renal vessels, the removed kidney is flushed with 500 ml of a chilled intracellular electrolyte solution and is submerged in a basin of ice slush saline solution to maintain hypothermia. Under these conditions, if warm renal ischemia has been minimal, the kidney can safely be preserved outside the body for as much time as is needed to perform extracorporeal partial nephrectomy. If possible, it is best to leave the ureter attached in such cases to preserve its distal collateral vascular supply, particularly with large hilar or lower renal tumors, in which complex excision may unavoidably compromise the blood supply to the pelvis, ureter, or both. When this procedure is done, the extracorporeal operation is performed on the abdominal wall. If the ureter is left attached, it must be occluded temporarily to prevent retrograde blood flow to the kidney when it is outside the body. Often, unless the patient is thin, working on the abdominal wall with the ureter attached is cumbersome because of the tethering and restricted movement of the kidney. If these are observed, the ureter should be divided, and the kidney placed on a separate workbench. This practice will provide better exposure for the extracorporeal operation, and, as this is being done, a second surgical team can be simultaneously preparing the iliac fossa for autotransplantation. If concern exists about the adequacy of ureteral blood supply, the risk of postoperative urinary extravasation can be diminished by restoring urinary continuity through direct anastomosis of the renal pelvis to the retained distal ureter. Extracorporeal partial nephrectomy is done with the flushed kidney preserved under surface hypothermia. The kidney is first divested of all perinephric fat to appreciate the full extent of the neoplasm ( Fig. 6-7A,B). Because such tumors are usually centrally located, dissection is generally begun in the renal hilus with identification of major segmental arterial and venous branches. Vessels clearly directed toward the neoplasm are secured and divided, and those supplying uninvolved renal parenchyma are preserved. The tumor is then removed by incising the capsule and parenchyma to preserve a surrounding margin of normal renal tissue (Fig. 6-7C,D). Transected blood vessels visible on the renal surface are secured, and the collecting system is closed as described for in situ partial nephrectomy.

FIG. 6-7. Technique for extracorporeal resection of large central renal neoplasm. (A) The kidney is removed to outside Gerota's fascia with its surrounding perinephric fat. (B) The flushed kidney is divested of all perinephric fat, and the gross margins of the tumor in relation to uninvolved renal parenchyma are defined. The dashed line indicates the area of tumor to be excised with a margin of surrounding normal renal tissue. (C) The capsule and parenchyma are incised, and vessels directed toward the neoplasm are secured and divided. (D) The renal remnant following extracorporeal excision of the tumor is shown. (E) The renal remnant is placed on the pulsatile perfusion unit and is alternatively perfused through the renal artery and vein. (F) The defect created by the partial nephrectomy is closed by suturing the kidney to itself.

At this point, the renal remnant may be reflushed or placed on the pulsatile perfusion unit to facilitate identification and suture ligation of remaining potential bleeding points (Fig. 6-7E). Alternatively, the kidney can be perfused through the renal artery and vein to ensure both arterial and venous hemostasis. Because the flushing solution and perfusate lack clotting ability, there may continue to be some parenchymal oozing, which can safely be ignored. If possible, the defect created by the partial nephrectomy is closed by suturing the kidney on itself to further ensure a watertight repair ( Fig. 6-7F). Autotransplantation into the iliac fossa is done employing the same vascular technique as that in renal allotransplantation. Urinary continuity may be restored with ureteroneocystostomy or pyeloureterostomy, leaving an internal ureteral stent in place. When removal of the neoplasm has necessitated extensive hilar dissection of vessels supplying the renal pelvis, an indwelling nephrostomy tube is also left for postoperative drainage. After autotransplantation, a Penrose drain is positioned extraperitoneally in the iliac fossa away from the vascular anastomotic sites. Partial Nephrectomy in Duplex Collecting Systems Occasionally, heminephrectomy in a kidney with a duplicated collecting system is indicated because of hydronephrosis and parenchymal atrophy of one of the two segments. In these cases, the demarcation of the tissue to be removed is usually very evident. The atrophic parenchyma lining the dilated system can be further

delineated by blue pyelotubular backflow if the ureter is ligated and the affected collecting system is distended by blue dye under pressure. In such cases, there is also often a dual arterial supply with distinct segmental branches to the upper and lower halves of the kidney. Segmental arterial and venous branches to the diseased portion of the kidney are ligated and divided. After preserving a strip of renal capsule, the parenchyma is divided at the observed line of demarcation. There is usually minimal bleeding from the renal surface, and temporary occlusion of the arterial supply to the nondiseased segment is often unnecessary. There should be no entry into the collecting system over the transected renal surface, which is then closed or covered as described above. Postoperative Follow-up Patients who undergo a partial nephrectomy for renal cell carcinoma are advised to return for initial follow-up 4 to 6 weeks postoperatively. At this time, a serum creatinine measurement and intravenous pyelogram are obtained to document renal function and anatomy; in patients with impaired overall renal function, a renal ultrasound study is obtained instead of an intravenous pyelogram.

OUTCOMES
Complications Complications of partial nephrectomy include hemorrhage, urinary fistula formation, ureteral obstruction, renal insufficiency, and infection. Significant intraoperative bleeding can occur in patients who are undergoing partial nephrectomy. The need for early control and ready access to the renal artery is emphasized. Postoperative hemorrhage may be self-limiting if confined to the retroperitoneum, or it may be associated with gross hematuria. The initial management of postoperative hemorrhage is expectant with bed rest, serial hemoglobin and hematocrit determinations, frequent monitoring of vital signs, and blood transfusions as needed. Angiography may be helpful in some patients to localize actively bleeding segmental renal arteries, which may be controlled via angioinfarction. Severe intractable hemorrhage may necessitate reexploration with early control of the renal vessels and ligation of the active bleeding points. Postoperative urinary flank drainage after a partial nephrectomy is common and usually resolves as the collecting system closes with healing. Persistent drainage suggests the development of a urinary cutaneous fistula. This diagnosis can be confirmed by determination of the creatinine level of the drainage fluid or by intravenous injection of indigo carmine with subsequent appearance of the dye in the drainage fluid. The majority of urinary fistulas resolve spontaneously if there is no obstruction of urinary drainage from the involved renal unit. If the perirenal space is not adequately drained, a urinoma or abscess may develop. An intravenous pyelogram or retrograde pyelogram should be obtained to rule out obstruction of the involved urinary collecting system. In the event of hydronephrosis or persistent urinary leakage, an internal ureteral stent is placed. If this is not possible, a percutaneous nephrostomy may be inserted. The majority of urinary fistulas resolve spontaneously with proper conservative management, although this may take several weeks in some cases. A second operation to close the urinary fistula is rarely necessary. Ureteral obstruction can occur after partial nephrectomy because of postoperative bleeding into the collecting system with resulting clot obstruction of the ureter and pelvis. This obstruction can lead to temporary extravasation of urine from the renal suture line. In most cases, expectant management is appropriate, and the obstruction resolves spontaneously with lysis of the clots. When urinary leakage is excessive, or in the presence of intercurrent urinary infection, placement of an internal ureteral stent can help to maintain antegrade ureteral drainage. Varying degrees of renal insufficiency often occur postoperatively when partial nephrectomy is performed in a patient with a solitary kidney. This insufficiency is a consequence of both intraoperative renal ischemia and removal of some normal parenchyma along with the diseased portion of the kidney. Such renal insufficiency is usually mild and resolves spontaneously with proper fluid and electrolyte management. Also, in most cases, the remaining parenchyma undergoes compensatory hypertrophy that serves to further improve renal function. Severe renal insufficiency may require temporary or permanent hemodialysis, and patients should be made aware of this possibility preoperatively. Postoperative infections are usually self-limiting if the operative site is well drained and there was no preexisting untreated urinary infection at the time of surgery. Unusual complications of partial nephrectomy include transient postoperative hypertension and aneurysm or arteriovenous fistula in the remaining portion of the parenchyma. A recent study detailed the incidence and clinical outcome of technical or renal-related complications occurring after 259 partial nephrectomies for renal tumors at The Cleveland Clinic. 2 In the overall series, local or renal-related complications occurred after 78 operations (30.1%). The incidence of complications was significantly less for operations performed after 1988 and significantly less for incidentally detected versus suspected tumors. The most common complications were urinary fistula formation and acute renal failure. A urinary fistula occurred after 45 of 259 operations (17%). Significant predisposing factors for a urinary fistula included central tumor location, tumor size >4 cm, the need for major reconstruction of the collecting system, and ex vivo surgery. Only one urinary fistula required open operative repair, and the remainder resolved either spontaneously ( n=30) or with endoscopic management (n=14). Acute renal failure occurred after 30 of 115 operations (26%) performed on a solitary kidney. Significant predisposing factors for acute renal failure were tumor size >7 cm, >50% parenchymal excision, >60 minutes ischemia time, and ex vivo surgery. Acute renal failure resolved completely in 25 patients, nine of whom (8%) required temporary dialysis; five patients (4%) required permanent dialysis. Overall, only eight complications (3.1%) required repeat open surgery for treatment, and all other complications resolved with noninterventive or endourologic management. Surgical complications contributed to an adverse clinical outcome in only seven patients (2.9%). These data indicate that partial nephrectomy can be performed safely with preservation of renal function in most patients with renal tumors. Results We recently completed a detailed analysis of tumor recurrence patterns after partial nephrectomy for sporadic localized renal cell carcinoma (RCC) in 327 patients at The Cleveland Clinic. 4 The purpose of this study was to develop appropriate guidelines for long-term surveillance after partial nephrectomy for RCC. Recurrent RCC after partial nephrectomy occurred in 38 patients (1 1.6%) including 13 patients (4.0%) who developed local tumor recurrence (LTR) and 25 patients (7.6%) who developed metastatic disease (MD). The incidence of postoperative LTR and MD according to initial pathologic tumor stage was as follows: 0% and 4.4% for T1 RCC, 2.0% and 5.3% for T2 RCC, 8.2% and 11.5% for T3a RCC, and 10.6% and 14.9% for T3b RCC. The peak postoperative intervals for developing LTR were 6 to 24 months (in T3 RCC patients) and >48 months (in T2 RCC patients). The above data indicate that surveillance for recurrent malignancy after partial nephrectomy for RCC can be tailored according to the initial pathologic tumor stage. The recommended surveillance scheme is depicted in Table 6-1. All patients should be evaluated with a medical history, physical examination, and selected blood studies on a yearly basis. The latter should include serum calcium, alkaline phosphatase, liver function tests, blood urea nitrogen, serum creatinine, and electrolytes. A 24-hour urinary protein measurement should also be obtained in patients with a solitary remnant kidney to screen for hyperfiltration nephropathy. 9 Patients who have proteinuria may be treated with a low-protein diet and a converting enzyme inhibitor, which appears to be beneficial in preventing glomerulopathy caused by reduced renal mass.

TABLE 6-1. Recommended postoperative surveillance after NSS for sporadic localized RCC

The need for postoperative radiographic surveillance studies varies according to the initial pT stage. Patients who undergo partial nephrectomy for pT1 RCC do not require radiographic imaging postoperatively in view of the very low risk of recurrent malignancy. A yearly chest x-ray is recommended after partial nephrectomy for pT2 or pT3 RCC because the lung is the most common site of postoperative metastasis in both groups. Abdominal or retroperitoneal tumor recurrence is uncommon in pT2 patients, particularly early after partial nephrectomy, and these patients require only occasional follow-up abdominal CT scanning; we recommend that this be done every 2 years in this category. Patients with pT3 RCC have a higher risk of developing LTR, particularly during the first 2 years after partial nephrectomy, and they may benefit from more frequent follow-up abdominal CT scanning initially; we recommend that this be done every 6 months for 2 years and every 2 years thereafter. The technical success rate with partial nephrectomy for renal cell carcinoma is excellent, and several large studies have reported 5-year cancer-specific survival rates of 87% to 90% in such patients (Table 6-2). These survival rates are comparable to those obtained after radical nephrectomy, particularly for low-stage renal cell carcinoma. The major disadvantage of partial nephrectomy for renal cell carcinoma is the risk of postoperative local tumor recurrence in the operated kidney, which has been observed in 4% to 6% of patients. These local recurrences are most likely a manifestation of undetected microscopic multifocal renal cell carcinoma in the renal remnant. The risk of local tumor recurrence after radical nephrectomy has not been studied, but it is presumably very low.

TABLE 6-2. Results of partial nephrectomy for renal cell carcinoma

Recent studies have clarified the role of partial nephrectomy in patients with localized unilateral renal cell carcinoma and a normal contralateral kidney. The data indicate that radical nephrectomy and partial nephrectomy provide equally effective curative treatment for such patients who present with a single, small (<4 cm), and clearly localized renal cell carcinoma. 1,5 The results of partial nephrectomy are less satisfactory in patients with larger (>4 cm) or multiple localized renal cell carcinomas, and radical nephrectomy remains the treatment of choice in such cases when the opposite kidney is normal. The long-term renal functional advantage of partial nephrectomy with a normal opposite kidney requires further study. CHAPTER REFERENCES
1. Butler B, Novick AC, Miller D, et al. Management of small unilateral renal cell carcinomas: Radical versus nephron-sparing surgery. Urology 1995;45:34–41. 2. Campbell SC, Novick AC, Streem SB, et al. Complications of nephron-sparing surgery for renal tumors. J Urol 1994;151:1177–1180. 3. Campbell SC, Fichtner J, Novick AC, et al. Intraoperative evaluation of renal cell carcinoma: Prospective study of the role of ultrasonography and histopathological frozen sections. J Urol 1996;155:1191. 4. Hafez KS, Novick AC, Campbell SC. Patterns of tumor recurrence and guidelines for follow-up after nephron-sparing surgery for sporadic renal cell carcinoma. J Urol 1997;157:2067–2071. 5. Lerner SE, Hawkins CA, Blute ML, et al. Disease outcome in patients with low-stage renal cell carcinoma treated with nephron-sparing or radical surgery. J Urol 1996;155:1868. 6. Licht MR, Novick AC, Goormastic M. Nephron-sparing surgery in incidental versus suspected renal cell carcinoma. J Urol 1994;152:39–42. 7. Licht MR, Novick AC. Nephron-sparing surgery for renal cell carcinoma. J Urol 1993;149:1–7. 8. Morgan WR, Zincke H. Progression and survival after renal-conserving surgery for renal cell carcinoma: Experience in 104 patients and extended follow-up. J Urol 1990;144:852–858. 9. Novick AC, Gephardt G, Guz B, et al. Long-term follow-up after partial removal of a solitary kidney. N Engl J Med 1991;325:1058–1062. 10. Steinbach F, Stockle M, Muller SC, et al. Conservative surgery of renal cell tumors in 140 patients: 21 years of experience. J Urol 1992;148:24–30.

Chapter 7 Radical Nephrectomy Glenn’s Urologic Surgery

Chapter 7 Radical Nephrectomy
Michael S. Cookson

M. S. Cookson: Division of Urology, Department of Surgery, University of Kentucky, Lexington, Kentucky 40536.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Flank Incision Thoracoabdominal Incision Transabdominal (Chevron or Anterior Subcostal) Radical Nephrectomy Outcomes Complications Results Chapter References

Renal cell carcinoma (RCC) is the most common malignancy of the kidney and accounts for about 3% of all adult neoplasms. The estimated number of new cases of renal cell carcinoma in the United States in 1997 is 28,000 with a projected 11,300 deaths, and this incidence is expected to increase as a result of the expanded use of radiographic imaging. 7 Because renal cell carcinoma is relatively refractory to chemotherapy and radiation therapy, surgery in general and radical nephrectomy in particular has evolved as the primary treatment in patients with clinically localized and locally advanced disease. Radical nephrectomy is defined as resection of Gerota's fascia and its entire contents including the kidney, perinephric fat and lymphatics, and ipsilateral adrenal gland. In theory, complete surgical excision of all tumor with negative surgical margins would offer the best opportunity for cure in patients with renal cell carcinoma. This argument would favor radical nephrectomy over simple nephrectomy, given the frequent propensity of the tumor to extend microscopically outside of the renal capsule and into perinephric fat. Thus, although no randomized trial has demonstrated the superiority of radical nephrectomy over simple nephrectomy, multiple series have documented improved survival in patients with renal cell carcinoma treated with radical nephrectomy over the past 30 years. 2,3,4 and 8

DIAGNOSIS
Between 85% and 90% of all solid renal masses are renal cell carcinoma, and, therefore, the diagnosis of renal cell carcinoma should be considered in all patients with a suspected solid renal mass. A renal mass detected on either intravenous pyelography or ultrasound is usually confirmed by computed tomography (CT scan). Typically, renal cell carcinomas are characterized on CT scan by a solid parenchymal mass with a heterogeneous density and enhancement with intravenous contrast injection (between 15 and 40 Hounsfield units). However, despite modern imaging, some benign tumors of the kidney may be indistinguishable and confirmed only after surgical excision. Additionally, metastatic deposits from a variety of malignancies including lung and breast cancers may involve the kidney and should be considered in patients with a known primary. The role of percutaneous biopsy or needle aspiration in differentiating an indeterminate renal mass remains controversial, and the absence of malignant cells on biopsy does not rule out the possibility of a neoplasm. For this reason, percutaneous renal biopsy for the purpose of diagnosis should be used only in selected cases. Clinical staging in patients suspected of renal cell carcinoma usually includes a contrast-enhanced CT scan of the abdomen; however, MRI is used occasionally and is particularly useful in patients with a history of a contrast allergy, renal insufficiency, or a suspected vena caval thrombus. From these imaging modalities, a number of factors can be determined, including the size and resectability of the primary, the presence or absence of lymphadenopathy or metastasis, involvement of adjacent structures, and the status of the contralateral kidney. A chest x-ray is obtained to rule out lung metastasis. Bone scans are performed in any patients with symptoms referable to the bone, as well as an elevation in serum alkaline phosphatase or hypercalcemia. In cases of suspected vena caval involvement, Doppler ultrasound is a useful screening tool. If results are equivocal, or if a vena caval thrombus is confirmed, a vascular phase MRI is usually able to determine the level of extension of the tumor thrombus, which allows the surgeon to properly plan an operative strategy.

INDICATIONS FOR SURGERY
The indication for radical nephrectomy is a clinically localized solid renal mass in a patient with a normal contralateral kidney. Patients with solitary kidneys, renal insufficiency, and bilateral renal masses should be considered candidates for nephron-sparing surgery. A thorough preoperative history and physical examination should be performed before the procedure. If significant comorbidities are suspected, consultation with the appropriate physician is recommended. The patient should be expected to physically withstand the operation and have a reasonable overall performance status and a 5-year life expectancy. In general, radical nephrectomy in patients with metastatic disease is performed for palliation, such as those patients with intractable pain or life-threatening hemorrhage who fail conservative treatment. Also, radical nephrectomy may be performed in the setting of an approved investigational protocol. Radical nephrectomy as a potential adjunct to enhance the effectiveness of biological response modifiers in patients with metastatic renal cell carcinoma remains experimental. The role of radical nephrectomy in patients with a solitary metastatic site is controversial; however, 5-year survival rates of 30% have been reported in selected patients, with best results reported in patients with solitary pulmonary metastases. 2 Although local extension of primary renal cell carcinoma into the perinephric fat, vena cava, or ipsilateral adrenal gland may portend a worse prognosis, in the absence of metastatic disease these factors alone should not dissuade the surgeon from attempting a radical nephrectomy. Additionally, radical nephrectomy has been successfully performed in the setting of direct extension of the tumor into adjacent organs such as the liver, colon, or psoas muscle. However, surgical removal in this setting is technically difficult and is associated with a higher morbidity and a potentially poor prognosis. Therefore, it should be attempted only in selected patients without obvious nodal or metastatic disease and in cooperation with appropriate surgical consultants. The role of regional lymphadenectomy at the time of radical nephrectomy remains controversial, and presently most patients should undergo only a limited unilateral lymphadenectomy for the purpose of staging. 2

ALTERNATIVE THERAPY
Surgery remains the only effective and potentially curative form of therapy for primary RCC. Along this line, the main challenge to radical nephrectomy in the near future appears to be from more conservative surgical approaches. Partial nephrectomy, enucleation, and wedge resection have recently been proposed in small, clinically localized RCC, with excellent early results ( Chapter 6).6 Arguments against partial nephrectomy include the potential for local recurrence, tumor multifocality, and the potential for increased complications. The routine removal of the ipsilateral adrenal gland at the time of radical nephrectomy has also been questioned, particularly in small tumors and tumors not involving the upper pole. 9 The use of laparoscopic techniques is currently expanding, and recently laparoscopic nephrectomy has been reported in patients with small tumors, although concerns over potential for tumor spillage and alteration of pathologic staging remain to be addressed. 5 Currently, chemotherapy and radiation therapy have proven to be inadequate treatment in primary renal cell carcinoma. 2

SURGICAL TECHNIQUE
There are a variety of factors that influence the choice of incision during radical nephrectomy. These include location of the affected kidney, tumor size and characteristics, body habitus, and physician preference. There are advantages and disadvantages to each incision, and it is important to be familiar with several approaches to the kidney, as no one incision is appropriate in all settings. The most commonly used incisions for radical nephrectomy are the flank,

thoracoabdominal, and transabdominal (subcostal or chevron) ( Fig. 7-1).

FIG. 7-1. Types of incisions during radical nephrectomy.

Flank Incision The flank approach is an excellent choice for a variety of reasons. First, it allows direct access to the retroperitoneum and kidney, and the entire procedure can often be performed in an extrapleural and extraperitoneal fashion. Additionally, the incision is anatomic in that it follows the track of the intercostal nerves with minimal risk of denervation. However, in large tumors, tumors involving the upper pole, or in situations where vena cava access is critical, a flank approach may be suboptimal. Although a flank approach may be performed through a subcostal incision, an 11th or 12th rib incision is superior for exposure of the upper pole and ipsilateral adrenal gland during radical nephrectomy. The patient is positioned on an inflatable mattress in the lateral decubitus position with the upper chest at about a 45 degree angle. An axillary roll is placed under the patient to cushion against pressure on the brachial plexus, and the elbows are padded to prevent ulnar nerve injury. The upper arm is draped across the body and placed on a Mayo stand or a padded support. The lower leg is flexed at 90 degrees, and the upper leg is extended over one or two pillows. The kidney rest is raised and the table is flexed to elevate the flank, and the table is adjusted to make the flank horizontal to the floor. The inflatable mattress is then activated, and the patient is secured with wide tape. An 11th or 12th rib incision is made based on several factors including the kidney position, the cephalad extent of the tumor, and the patient's body habitus. A general rule is to incise over the rib that, when extended medially, will position the incision over the renal hilum. The incision is then made over the rib from the posterior axillary line to the tip and extended medially as far as necessary, which usually stops short of the lateral border of the rectus abdominis Figure 7-2). The latissimus dorsi is divided, and the upper portion of the incision is carried down to the rib. At this point a partial rib resection may be accomplished as shown in ( Fig. 7-3). An Alexander periosteal elevator is used to deflect the periosteum from the bone to avoid injury to the intercostal bundle located under the inferior portion of the rib. A Doyen elevator is then used to strip the periosteum from the entire undersurface of the rib to be resected. Next, a rib cutter is used to divide the proximal segment of the rib. Alternatively, the incision may be created between the ribs in the intercostal space.

FIG. 7-2. Technique of 11th-rib resection. (A and B) The incision is made over the 11th rib. (C) After division of the latissimus dorsi and external oblique muscles, subperiosteal resection of the rib is performed. (D) The incision is carried through the periosteum posteriorly and the internal oblique and transversus muscles medially, exposing the paranephric space. A tongue of pleura lies in the upper portion of the wound. Diaphragmatic slips that come into view are divided, and the pleura can be retracted upward. (E) The paranephric fat is dissected bluntly.

FIG. 7-3. Rib resection technique. (A) After exposure of the rib, the intercostal muscles are stripped from the upper and lower rib surfaces with an Alexander Farabeuf costal periosteotome. (B) The rib is freed from the periosteum by subperiosteal resection or from the pleura with a periosteum elevator. (C) A Doyen costal elevator is slipped beneath the rib to free it. The proximal and distal portions of the rib are immobilized with Kocher's clamps, and the rib is divided proximal to its angle with a right-angled rib cutter. (D) The costal cartilage is cut free with scissors. The cut surface of the rib is inspected for spicules, which are removed with a rongeur.

The posterior layer of the periosteum is then incised carefully, and the pleura is protected superiorly. Anteriorly, the external and internal oblique muscles are divided, and the transversus abdominis muscle is split in the direction of its fibers, taking care not to enter the peritoneum. The peritoneum is swept medially, and the intermediate stratum of the retroperitoneal connective tissue is incised sharply to expose the paranephric space. Approaching this in a posterior fashion with early identification of the psoas muscle helps to keep proper orientation. A self-retaining retractor such as a Finochetto or Balfour helps to maintain exposure. A radical nephrectomy is then performed. The wound is closed after checking to ensure that no injury to the pleura has occurred (see complications). The table flex is released, and the kidney rest is lowered. The posterior layer consisting of the fascia of the transversus abdominis and the internal oblique is closed in a running fashion with #1 PDS or Prolene. The anterior layer of external oblique fascia is closed with a running #1 PDS or Prolene. Alternatively, interrupted figure-of-eight sutures of #1 Vicryl can be used for both layers. The skin is closed in accordance with surgeon preference.

Thoracoabdominal Incision The thoracoabdominal approach allows for excellent exposure of large tumors as well as upper-pole tumors, particularly on the left. Additionally, it affords easy access to the adrenal gland and thoracic cavity. The patient is positioned with the hips flat and with the break of the table located just above the iliac crest. The pelvis can be torqued up to about 30 degrees if necessary. The patient's ipsilateral shoulder is rotated 45 degrees, and the ipsilateral arm is extended over the table and properly supported on a Mayo stand or padded arm rest Figure 7-4). It is important to properly pad all pressure points including between the legs and the contralateral shoulder. The kidney rest may be elevated to accentuate the proper extension, and the break in the table is made to optimize the incision. After positioning, the patient is secured with wide adhesive tape.

FIG. 7-4. Thoracoabdominal incision. (A) The patient is placed in a semirecumbent position using sandbags. If the chest is entered through the ninth intercostal space, the incision extends from the midaxillary line across the costal margin at the intercostal space to the midline or across it just above the umbilicus. (B) The anterior rectus sheath and the external oblique and latissimus dorsi muscles are divided. (C) The intercostal muscles parallel the direction of the three abdominal layers and are divided. The costal cartilage and the internal oblique and rectus muscles are incised. If more exposure is desired, the linea alba and opposite rectus can be divided. (D) The pleural reflection (shaded areas) lies progressively closer to the costal margin in the more cephalic intercostal spaces. (E) The pleura, reflecting as the costophrenic sinus near the costal margin, is exposed beneath the intercostal muscles. The diaphragm can be seen inferior and dorsal to the pleura. The pleura is opened with care to avoid injuring the lung, which comes into view with inspiration. After the lung is packed away gently, the diaphragmatic surface of the pleura is seen. The diaphragm is incised on its thoracic surface, avoiding the phrenic nerve. (F) The transversus abdominis muscle is divided, exposing the peritoneum with the liver lying beneath it. (G) The peritoneum is incised, and a rib-spreading retractor (Finochletto) is inserted, enabling upward displacement of the liver (or the spleen on the left) into the thoracic cavity and giving wider access to the posterior peritoneum than in an anterior abdominal incision.

The thoracoabdominal incision is made over the bed of the eighth, ninth, or tenth rib, depending on the surgeon's preference based on patient and tumor characteristics. The incision may be made between the ribs, or a portion of the rib may be removed. The incision is made over the rib beginning at the posterior axillary line. The incision is carried medially across the costal cartilage margin to the midline and then carried down the midline to the umbilicus. Alternatively, the medial portion of the incision may be carried across the midline or combined with a low midline to form a “T.” The latissimus dorsi is divided, and the upper portion of the incision is carried down to the rib. At this point, a rib resection can be performed as previously described ( Fig. 7-3). The peritoneum may be entered by incising the external and internal obliques, the transversus abdominis, and the ipsilateral rectus belly. Next, the costochondral cartilage at the inferior portion of the upper thoracic incision is divided, and the chest is entered along the entire length of the periosteal bed. The pleural space is entered, and care should be taken not to injure the lung. The lung is protected by pads, and the diaphragm is divided in the direction of the muscle fibers, which helps to avoid injury to the phrenic nerve. A self-retaining retractor such as a Finochetto or a Balfour is properly padded and placed to maintain exposure. A radical nephrectomy is then performed. After completion of the radical nephrectomy, the table flex is removed, and the diaphragm is closed with interrupted 2-0 silk sutures with knots placed on the inferior side. After a #32 chest tube has been inserted through a separate incision and properly positioned, the ribs are reapproximated with 2-0 chromic pericostal sutures. The thoracic portion of the incision is closed with interrupted figure-of-eight 1-0 Vicryl sutures through all layers of the chest wall. The medial portion of the intercostal muscle closure should include at least a small portion of the diaphragm. An intercostal nerve block is administered before closure and may be accomplished by injecting approximately 10 ml of 0.5% lidocaine or bupivicaine hydrochloride into the intercostal space of the incision and two interspaces above and below. The costal cartilage can be reapproximated with 0 chromic suture. The peritoneum is closed with a running 2-0 chromic, although this is optional. The posterior rectus fascia, the fascia of the transversus abdominis, and the internal oblique muscles are closed with a running or interrupted #1 PDS suture. The anterior rectus and the external oblique fascia are closed with either a running or interrupted #1 PDS or Maxon suture. Skin closure is determined by surgeon preference. The chest tube is secured in place with a 0 silk and taped securely in place. Transabdominal (Chevron or Anterior Subcostal) Anterior incisions offer several advantages including excellent exposure of the renal pedicle and access to the entire abdomen and contralateral retroperitoneum. With the patient in the supine position, the operative side is elevated slightly with a flank roll, and the patient hyperextended to accentuate the line of incision. An incision is made from near the tip of the 11th or 12th rib on the ipsilateral side two fingerbreadths below the costal margin and extended medially to the xyphoid process. The incision is then gently curved across the midline and as far laterally as necessary for exposure up to near the tip of the contralateral 11th rib. Occasionally, only a portion of the contralateral side will be incised just across the rectus abdominis. The incision is carried down to the anterior rectus fascia, which is then divided (Fig. 7-5). Next, the external and internal oblique fascia and muscles are divided, and the fibers of the transversus abdominis split. The rectus muscle and posterior rectus sheath are divided with electrocautery by placing a straight clamp or army–navy retractor underneath and gently elevating it. The superior epigastric artery is ligated with 2-0 silk and divided when encountered. The peritoneal cavity is then entered, and the falciform ligament is ligated between two Kelly clamps, divided, and tied with 0 silk suture. To facilitate exposure, the lower aspect of the incision is rotated caudally with a rolled towel placed underneath the skin, and the fascia sutured to the lower portion of the abdominal wall with two #2 nylon sutures. Use of a self-retaining retractor such as a Wishbone Omni-Tract is helpful. A radical nephrectomy is then performed.

FIG. 7-5. Transabdominal chevron incision. (A) With the patient in the supine position and slightly hyperextended, an incision is made two fingerbreadths below the costal margin to just below the xyphoid process and then curved gently down across to the tip of the opposite 11th rib. (B) Divide the subcutaneous tissue and the anterior rectus sheath bilaterally. Insinuate a Kelly or an army–navy retractor under the rectus muscle, and the muscle is divided with electrocautery. (C) Divide the external oblique and internal oblique muscles and split the transversus abdominis. Enter the peritoneal cavity in the midline by tenting up on the peritoneum and incising sharply with Metzenbaum scissors.

Closure of the wound is performed after the table is returned to the horizontal position. The wound is then closed in two layers. The posterior layer consisting of the fascia of the transversus abdominis and the internal oblique laterally along with the posterior rectus fascia medially is closed with two running #1 PDS sutures, each starting at the lateral aspect and running medially to the midline. The anterior layer of external oblique and anterior rectus fascia is closed in a similar fashion with #1 PDS. Alternatively, the layers can be closed with interrupted #1 Vicryl. Occasionally, it is helpful to place a U stitch of #1 Prolene at the apex of the chevron incision before closure, which includes the rectus fascia on either side of the midline, securing this suture after the anterior fascia has been approximated. The skin is then closed according to the surgeon's preference. Radical Nephrectomy Irrespective of the choice of incision, certain caveats are universal for the safe and successful completion of a radical nephrectomy. This includes a systematic approach with careful mobilization of Gerota's fascia and early vascular control. For a flank approach, the posterior peritoneum lateral to the colon is incised along the length of the descending colon (left side) or ascending colon (right side) and reflected medially. For left-sided exposure, the lienorenal ligament is incised to mobilize the spleen cephalad. On the right side, the hepatic flexure of the colon is mobilized. The ureter is identified and encircled with a vessel loop. The gonadal vein is ligated and divided. The plane between the mesentery of the colon and Gerota's fascia is then developed using a combination of sharp and blunt dissection. On the right side, the vena cava is exposed by Kocherizing the duodenum. Using blunt dissection, the retroperitoneal fat overlying the renal vessels is separated, exposing the renal hilum. It is often helpful to ligate and divide the ureter before this to allow for mobilization and upward displacement of the lower pole of the kidney. The dissection is then carried cephalad along the vena cava (right side) or aorta (left side). On the right side, the right renal vein is identified exiting from the vena cava, isolated, and encircled with a right-angle clamp and a 0 silk suture and tagged. After identification of the renal artery (exposure may be enhanced by the use of a vein retractor on the renal vein), the artery is dissected free and cleaned for a distance of approximately 2 to 3 cm. With a right-angle clamp, the renal artery is encircled, and 2-0 silk ties are passed ( Fig. 7-6). The sutures are then separated and tied, allowing a safe distance for division of the artery. A small hemoclip or a 3-0 silk suture ligature may be placed on the proximal aspect of the artery before division. A right-angle clamp is placed under the artery to be divided and gently elevated, and the artery is cut with either a knife (#15 blade) or Metzenbaum scissors. The right renal vein is then ligated in a similar fashion with 0 silk sutures.

FIG. 7-6. The right renal artery is identified by palpation beneath the vein. After identification with a right-angle clamp, the artery is cleaned in the same manner as the vein. With a right-angle clamp beneath the artery, a suture is passed on a tonsil clamp to the mouth of the right-angle clamp, and the suture is passed around the artery. (From Donohue RE. Radical nephroureterectomy for carcinoma of the renal pelvis and ureter. In: Crawford ED, Borden TA, eds. Genitourinary cancer surgery. Philadelphia: Lea & Febiger. 1982;101.)

On the left, the renal vein is isolated as it courses over the aorta. The left adrenal and gonadal veins are identified emanating from the left renal vein, and, if present, a posteriorly directed lumbar venous tributary is noted. A right-angle clamp is passed around the renal vein, followed by a 0 silk suture proximal to the tributaries, and tagged. The venous tributaries are then individually ligated and divided with 2-0 or 3-0 silk and small hemoclips where necessary, leaving the 2-0 silk suture on the main renal vein tagged ( Fig. 7-7). The left renal artery and vein are then ligated similarly to the technique described above for the right side.

FIG. 7-7. The same technique for ligation is employed on the left side. A suture is passed around the lumbar vein proximally and distally after the vein has been cleaned, and the vein is ligated. Sutures are passed around the main renal vein but are not tied until after the branches have been ligated. The artery is identified above the vein and cleaned, isolated, and ligated, and the proximal end is sutured with a 5-0 cardiovascular silk suture. (From Donohue RE. Radical nephroureterectomy for carcinoma of the renal pelvis and ureter. In: Crawford ED, Borden TA, eds. Genitourinary cancer surgery. Philadelphia: Lea & Febiger, 1982;101.)

Gerota's fascia is then mobilized posteriorly and superiorly using a combination of sharp and blunt dissection. Hemoclips along the superior and medial border are useful to control any potential bleeding during this portion of the procedure. The adrenal hilum is then dissected from caudal to cranial with the aid of either hemoclips or straight clamps and ties. On the right side, the short posteriorly located right adrenal vein should be anticipated as it exits directly from the vena cava. When encountered, the right adrenal vein is isolated, ligated, and divided. The specimen is then delivered, and meticulous hemostasis is achieved.

OUTCOMES
Complications The potential for bleeding during radical nephrectomy necessitates careful patient preparation and preoperative planning to significantly reduce the chances. Any medications that interfere with platelet function or clotting should be discontinued, and patients should be type and cross matched for 2 units of packed red blood cells. The patient should have either two large-bore peripheral intravenous lines or a central venous line to allow for rapid infusion of fluids or blood products. Bleeding during radical nephrectomy may be from a variety of locations including the renal hilum, collateral tumor vessels, or adjacent structures. Venous bleeding is usually the most problematic. The first maneuver is to apply direct pressure to the area of bleeding. The point of bleeding is then carefully exposed and controlled by a suture ligature. In the case of venal caval injury, a Satinsky clamp is placed, and a vascular 5-0 or 6-0 Prolene suture is used to oversew the defect. Lumbar veins should be exposed by gentle retraction of the vena cava, appropriately clamped, ligated with vascular silk suture, and divided. Renal artery bleeding may be controlled by direct pressure on the aorta proximally until adequate exposure can be obtained, and the artery is then ligated. Only in rare circumstances will a pedicle

clamp or mass ligature be necessary. Adrenal tears may result in significant hemorrhage during radical nephrectomy, particularly on the right side, where the short adrenal vein enters into the vena cava directly posterior. Control of the right adrenal vein should be attempted only after control of the vena cava, adequate exposure, and proper suction. The vein is then ligated with a 2-0 or 3-0 silk tie or a vascular Prolene. Venous bleeding from a torn adrenal gland can be oversewn with a running suture or stopped by placement of surgical clips. However, removal of the ipsilateral adrenal may be the most expeditious method of controlling bleeding. Failure to recognize a rent in the pleura during flank incision will result in a pneumothorax. Small openings in the pleura may be recognized by filling the flank wound with sterile water and administering a deep inspiratory breath. Small tears recognized intraoperatively can be managed by closing the pleura with a 3-0 chromic pursestring suture over a 12 Fr or 14 Fr Robinson catheter. Before it is removed, all air is aspirated from the pleural cavity either by suction or by placing the Robinson catheter under water and administering a deep inspiratory breath. The air is evacuated from the pleural space, and the tube is removed while the pursestring suture is simultaneously tied in place. Alternatively, the Robinson catheter can be temporarily left in place with the chromic suture secured and the fascial layers closed around the catheter, which exits from the corner of the wound. Just before skin closure, after all air has been evacuated under water seal as described above, the catheter is removed. The latter technique is helpful when the pleura is attenuated or contains multiple small holes that are not easily closed. Alternatively, a 22 Fr or 24 Fr chest tube may be placed and left to suction. An upright end-expiratory chest x-ray is obtained after all flank incisions to ensure that no significant pneumothorax exists. A small (usually less than 15%), asymptomatic pneumothorax can be followed conservatively with serial chest x-rays and oxygen therapy. In a symptomatic or large pneumothorax, aspiration of the pleural space using a needle or a central venous catheter (Seldinger technique) introduced just over the rib in the anterior fourth or fifth interspace can be therapeutic. However, if these attempts are not successful, a chest tube should be inserted and placed on suction. Injuries to the colon during radical nephrectomy are uncommon. In locally advanced tumors suspected of extension into either the colon or mesentery, patients should undergo a mechanical and antibiotic bowel preparation. Segmental colon resection and primary anastomosis should be possible in most cases. Inadvertent injury to the colon during radical nephrectomy can usually be repaired primarily; however, in situations where there is gross spillage of fecal contents or a devascularized segment, a diverting colostomy should be considered, and a general surgery consultation is advisable. Defects in the mesentery of colon should be closed to prevent internal herniation of peritoneal contents. Right radical nephrectomy is also associated with the potential for injury to the duodenum and liver. The duodenum must be carefully mobilized, and care must be taken to properly pad retractors to prevent injury to the bowel and adjacent structures including the head of the pancreas. The second portion of the duodenum may be injured during a right radical nephrectomy. Duodenal hematomas should only be observed, but rapidly enlarging hematomas will require control of the bleeding, and an intraoperative general surgery consultation should be obtained. Duodenal lacerations should be repaired in multiple layers with interrupted nonabsorbable sutures for the mucosal and serosal layers. 10 When possible, an omental wrap may provide additional support, and all patients should be managed with a nasogastric tube during the postoperative period. Superficial liver lacerations are repaired with absorbable horizontal mattress sutures utilizing a Surgicel or Hemopad bolster. Deep liver lacerations, which may involve the hepatic ducts, could result in bile leakage and should be drained following repair. Direct invasion of the liver by renal cell carcinoma is rare; however, resection including en-bloc removal is possible in selected cases. If a major lobectomy or a partial hepatectomy is to be performed because of either direct extension or major hemorrhage, a general surgeon should be present. Splenic injury is one of the most common intraoperative complications during a left nephrectomy, with an incidence as high as 10% in some series. Most superficial lacerations or tears can be managed conservatively without the need for splenectomy. Although minor tears may require only some gentle pressure and the application of a Hemopad or Surgicel with Avitene application, closure of a moderate splenic capsular tear is facilitated through the use of nonabsorbable sutures over bolsters of Surgicel. Major hemorrhage secondary to severe splenic lacerations may require splenectomy. The splenic artery and vein are controlled by compressing these structures, located in the splenic hilum near the tail of the pancreas. Initially, this can be accomplished manually by compressing the tail of the pancreas between the thumb and the forefinger. Once bleeding has been temporarily controlled, the spleen is mobilized by dividing the splenocolic and splenorenal ligaments as well as taking down the peritoneal attachments to the diaphragm. The short gastric vessels are then ligated, and the hilum of the spleen is dissected free from the tail of the pancreas. The splenic artery and vein are ligated and divided. The pancreas should be inspected closely to rule out inadvertent injury. Following splenectomy, patients will have a reduced resistance to pneumococcal organisms and should receive Pneumovax and Hibtiter on a yearly basis. Results Surgical excision remains the only effective and potentially curative therapy for clinically localized RCC. Pathologic staging remains the best prognostic variable in terms of patient survival, and the two most commonly used staging systems are the Robson classification and the American Joint Committee on Cancer recommendations (TNM) classification. 1,8 Both staging systems have demonstrated an inverse relationship between survival and increasing stage, but the TNM is generally thought to be more accurate because it more precisely defines the extent of disease. In patients treated with radical nephrectomy and found to have tumors confined to the kidney (Robson stage I), the 5-year survival is between 60% and 90% compared with 47% to 67% in patients whose RCC is confined to Gerota's fascia (Robson stage II). Survival for patients with distant metastases is poor with 5-year survival of between 5% and 10%. 2 Under the TNM staging system, the 5-year survival for patients with organ-confined tumors treated with radical nephrectomy for T1N0M0 tumors is between 80% and 91%, whereas that for T 2N 0M0 tumors is 68% to 92%.3,4 For those patients with T 3aN0M0 (tumor invading into the adrenal gland) and T3bN0M0 (tumor invading into the renal vein) carcinomas, the 5-year survival is 77% and 59%, respectively. Finally, patients with node-positive disease (N 1–3M0) have a 5-year survival between 15% and 52%. Although radical nephrectomy remains the standard of care for unilateral renal cell carcinoma, more conservative surgical options have been proposed. Recently, excellent results have been seen in patients treated with nephron-sparing surgery, with 5-year cancer-specific survivals of greater than 90% for small, unilateral, stage I tumors.9 The ultimate choice of surgical treatment in patients with these favorable clinical features, and in particular the risk of local recurrence and cancer-specific survival, remains to be determined through long-term follow-up. Currently, radical nephrectomy remains the treatment of choice in patients with clinically localized renal cell carcinoma and the standard against which future alternative surgical strategies will be measured. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. American Joint Committee on Cancer. Manual for staging of cancer, 4th ed. Philadelphia: JB Lippincott, 1992. deKernion JB, Belldegrun A. Renal tumors. In: Walsh PC, Retik AB, Stamey AB, Vaughan ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;1053–1093. Giuliani L, Giberti C, Martorana G, Rovida S. Radical extensive surgery for RCC: long-term results and prognostic factors. J Urol 1990;143:468–474. Hermanek P, Schrott KM. Evaluation of a new tumor, nodes and metastases classification of RCC. J Urol 1990;144:238–242. Kerbl K, Clayman RV, McDougall EM, Kavoussi LR. Laparoscopic nephrectomy: the Washington University experience. Br J Urol 1994;73:231–236. Novick AC. Renal sparing surgery for RCC. Urol Clin North Am 1993;20:277–282. Parker S, Tong T, Bolden S, Wingo PA. Cancer statistics, 1997. CA 1997;47:5–27. Robson CJ, Churchill BM, Andersen W. The results of radical nephrectomy for RCC. J Urol 1969;101:297–301. Sagolowsky AI, Kadesky KT, Ewalt DM, Kennedy TJ. Factors influencing adrenal metastasis in RCC. J Urol 1994;151:1181–1184. Smith RB. Complications of renal surgery. In: Smith RB, Ehrlich RM, eds. Complications of urologic surgery: Prevention and management, 2nd ed. Philadelphia: WB Saunders, 1990;128–159.

Chapter 8 Intracaval Tumors Glenn’s Urologic Surgery

Chapter 8 Intracaval Tumors
Thomas J. Polascik and Fray F. Marshall

T. J. Polascik, Central Medical Park, Durham, North Carolina 27704. F. F. Marshall: James Buchanan Brady Urological Institute, The Johns Hopkins Medical Institutions, Baltimore, Maryland 21287.

Diagnosis Indications Alternative Therapy Surgical Technique Cardiopulmonary Bypass, Hypothermia, and Temporary Cardiac Arrest Outcomes Complications Results Chapter References

Tumor thrombus extending from the renal vein into the vena cava has been reported to occur in 4% to 10% of patients with renal cell carcinoma. 5 The extent of tumor thrombus involving the inferior vena cava can vary from involvement of the renal vein only to extension into the right atrium, which occurs in some of these patients. The majority of patients with vena caval tumors have right-sided renal primaries because of the short right renal vein. In the absence of metastatic disease, numerous centers have demonstrated long-term cancer-specific survival rates comparable to early-stage renal cell carcinoma following complete surgical excision. Improvements in surgical technique have allowed the surgeon to safely perform a radical nephrectomy and vena cavotomy. Several centers have documented reduced morbidity and mortality associated with these procedures. We have removed tumor from both renal veins, lumbar veins, hepatic veins, the right atrium, and the right ventricle. Technical difficulties and complications (excessive bleeding, coagulopathy, and postoperative renal failure) can accompany these procedures, especially with extensive intra- or suprahepatic caval neoplastic extension.

DIAGNOSIS
Today, the majority of patients presenting with a renal mass and intracaval tumor extension are diagnosed with computerized tomography (CT). In the past, patients presenting with advanced disease had clinical signs and symptoms related to vena caval occlusion including bilateral lower extremity edema, a recently enlarging varicocele, or dilated abdominal wall veins. Patients may also present with proteinuria, hepatic dysfunction with hepatomegaly, or pulmonary embolus. Patients should have a thorough evaluation for metastatic disease because, if present, we do not typically recommend proceeding with surgery. Further radiologic imaging typically includes CT of the chest and abdomen and a bone scan if applicable. The renal vein and vena cava can be noninvasively imaged using magnetic resonance imaging (MRI). The MRI can usually define the superior limit of the caval thrombus unless the distal thrombus is mobile, thus limiting its accuracy. The MRI is also effective when total caval occlusion is present. Vena cavography can be used to define a caval tumor; however, its invasive nature, its false-positive and -negative results, and a decreased ability to define the superior extent of the tumor limit its use. To fully delineate the extent of a large caval tumor, the combination of MRI and intraoperative transesophageal sonography provides the best results. 8 All patients should have a complete medical evaluation and be deemed candidates to withstand an extensive surgery.

INDICATIONS
The primary indication for nephrectomy and vena cavotomy is a renal mass with intracaval tumor extension in the absence of metastatic disease. The patient should also be medically able to tolerate an extensive surgical procedure.

ALTERNATIVE THERAPY
To date, complete surgical excision of tumor is the only curative treatment. Expectant therapy or systemic protocols may be applicable if the patient is a candidate.

SURGICAL TECHNIQUE
In addition to a general anesthetic, a thoracic epidural can be utilized and is often effective with postoperative pain management with most flank incisions. For the majority of tumors, standard intraoperative monitoring includes central venous pressure, arterial pressure tracings, electrocardiography, and urinary output. Additional monitoring is used for extensive vena cava tumors, including a Swan–Ganz catheter, esophageal and rectal temperature probes, oxygen and carbon dioxide measurements, and transesophageal sonography. A hypothermic blanket is used to maintain body temperature. Elastic stockings and sequential compression devices are placed to prevent lower extremity venous stasis. An intravenous cephalosporin is generally sufficient as prophylactic antibiotic coverage. The patient's body habitus and extent of both the primary and intracaval tumor direct the surgical approach. For renal tumors with neoplasm extending minimally into the inferior vena cava, a supra-11th-rib or standard thoracoabdominal approach with rib excision is ideal, especially in obese patients. For left-sided tumors and more extensive caval tumors, an anterior incision will provide good exposure. We have used a thoracoabdominal incision extending from the tip of the scapula across the costal margin to the midline halfway between the umbilicus and the xyphoid process for right-sided tumors with intrahepatic and supradiaphragmatic intracaval tumor extension. In this approach, the patient should be positioned with the right shoulder rotated toward the contralateral side; the hips remain in the supine position, and the table is slightly extended. Although this incision provides both intra-abdominal and intrathoracic exposure, the infradiaphragmatic dissection is easier for the urologist while cannulating the aortic arch for cardiopulmonary bypass is more difficult. We typically use a median sternotomy extending into either a midline abdominal or a chevron incision when the intracaval neoplasm extends into or beyond the liver and cardiopulmonary bypass is considered ( Fig. 8-1).2 The chevron incision is useful in patients with a wide abdominal girth. Although these extensive incisions provide excellent exposure, allowing for additional operations to be performed, we recommend limiting the procedure to nephrectomy and caval thrombectomy.

FIG. 8-1. Incision for radical nephrectomy with removal of vena caval thrombus. Reprinted with permission from Marshall FF, Reitz BA. Radical nephrectomy with excision of vena caval tumor thrombus. In: Marshall FF, Reitz BA (eds.), Marshall's textbook of operative urology. Philadelphia: WB Saunders, 1996.

The operation is commenced by utilizing the entire incision including the median sternotomy, as this approach gives the best exposure. The abdomen is inspected for metastatic disease, and if discovered, the procedure is usually stopped, as cancer-specific survival has not been demonstrated to be improved in the long term. In the absence of overt metastasis, the renal tumor is approached first. For a right renal tumor, the right colon is mobilized along the line of Toldt and retracted medially to gain access to the retroperitoneum. For significant tumors via a midline approach, incision of the root of the mesentery up to the ligament of Trietz with placement of the bowel into an intestinal bag retracted onto the chest provides additional exposure. We use the OmniTract retractor (Minnesota Scientific Inc.), as it provides excellent superficial and deep exposure of the surgical field. The kidney and Gerota's fascia are mobilized, first by a posterolateral approach developing the plane between the quadratus/psoas muscles and Gerota's fascia. After the kidney has been mobilized posteriorly, the renal artery is ligated early to keep blood loss to a minimum. Anteriorly, the mesocolon is then reflected medially from the anterior surface of Gerota's fascia until the vena cava is visualized. A Kocher maneuver provides additional medial exposure near the vena cava. Superiorly, dissection above the adrenal is undertaken with Ligaclips, and the adrenal vein is ligated. Inferiorly, the kidney is mobilized along with ligation of the gonadal vein and ureter. Mobilization of the primary tumor is complete when the kidney remains attached to the vena cava by the renal vein. A left-sided renal tumor with caval thrombus requires dissection on both sides of the abdomen to access both the vena cava and the left kidney. A midline incision usually provides sufficient exposure. The descending colon is reflected medially by incising the line of Toldt. In a dissection similar to that for a right-sided tumor, the kidney and Gerota's fascia is mobilized until only the left renal vein remains. The ascending colon is then mobilized medially by incising the line of Toldt, and the duodenum is reflected by the Kocher maneuver. Once adequate exposure to the vena cava is obtained, the remainder of the procedure is similar to that for a right-sided renal primary tumor. The extent of the intracaval tumor dictates the length the vena caval needs to be isolated. Dissection should proceed directly on the vena cava with care taken to prevent potential dislodgment of caval tumor. If the intracaval tumor extends slightly beyond the ostium of the renal vein into the vena cava, a Statinsky vascular clamp can be placed on the caval sidewall beyond the tumor. This segment of caval wall can be excised with the nephrectomy specimen en bloc, and the cava can be oversewn with a 4-0 polypropylene on a cardiovascular needle. With a more extensive infrahepatic intracaval tumor, control of the vena cava must be obtained above and below the extent of the caval tumor thrombus. During mobilization of the vena cava, one or more posterior lumbar veins may require ligation to prevent unexpected bleeding. Inferiorly, a Rummel tourniquet (umbilical tape passed through a 16-Fr red rubber catheter) is placed loosely below the tumor thrombus and both renal veins. For a right-sided tumor, a Rummel tourniquet is placed loosely around a segment of the left renal vein to secure control of this vessel. Additional exposure to the vena cava can be gained superiorly by dividing the posterior attachments of the liver and rotating the liver medially. Depending on the superior extent of the caval tumor, variable venous branches draining the caudate lobe of the liver may need to be ligated and divided ( Fig. 8-2). If these veins are short, they can be controlled using suture ligatures placed into the liver parenchyma. Cardiopulmonary bypass can be obviated when vascular control using a vascular clamp or Rummel tourniquet can be gained above the superior extent of the tumor. Division of the diaphragm may aid in gaining vascular control above the superior extent of the tumor thrombus.

FIG. 8-2. Ligation and division of venous tributaries of the caudate lobe of the liver. Reprinted with permission from Marshall FF, Reitz BA. Radical nephrectomy with excision of vena caval tumor thrombus. In: Marshall FF, Reitz BA (eds.): Marshall's textbook of operative urology. Philadelphia: WB Saunders, 1996.

After adequate mobilization of the vena cava superior and inferior to the tumor thrombus with ligation of any lumbar veins, all vascular clamps or Rummel tourniquets are secured. A narrow elliptical incision circumscribing the ostium of the involved renal vein is made. If the tumor is inseparable from the caval endothelium superior to the renal veins, the involved cava is excised. The renal primary and caval tumor is removed en toto under direct vision. On occasion, we have used a dental mirror to inspect the hepatic veins or the flexible cystoscope to inspect the cava to ensure complete removal of tumor. If additional verification is necessary, transesophageal echography can be used to evaluate the superior extent of the cava, or direct intraoperative sonography can be used to evaluate the extent of the cava. 8 To close the vena cava, a 4-0 or 5-0 cardiovascular polypropylene suture is used. Before the cavotomy closure is completed, the inferior tourniquet is released to allow trapped air to escape through the cavotomy site. If excision of the cava decreases the vascular diameter by more than 50%, reconstruction of the vena cava is recommended to prevent caval thrombosis (Fig. 8-3). We prefer to reconstruct the vena cava using pericardium because it is less thrombogenic, although prosthetic grafts can be employed.3 Venous drainage of the right kidney must always be preserved to prevent venous infarction. In some instances, the cava has been oversewn to prevent subsequent embolism if the thrombus below the renal veins is adherent to the caval endothelium.

FIG. 8-3. Reconstruction of the vena cava is recommended if the cross-sectional diameter of the vena cava is reduced by more than 50% after resection of the tumor thrombus and vena caval wall. Reprinted with permission from ref. 3.

Cardiopulmonary Bypass, Hypothermia, and Temporary Cardiac Arrest Cardiopulmonary bypass, hypothermia, and temporary cardiac arrest greatly facilitate the resection of a suprahepatic caval thrombus. 4 It is best to dissect as much of the kidney and the vena cava as possible before cardiac bypass. Following isolation of the renal tumor, the pericardium is opened and retracted with stay sutures. Typically, the right atrial appendage is cannulated with a 32-Fr venous cannula, and the aorta is cannulated with a 22-Fr Bardic cannula. Heparin is then administered to maintain an activated clotting time greater than 450 seconds. The patient is placed on bypass with flow rates maintained between 2.5 and 3.5 liters/min. A core temperature of 18° to 20°C is attained within 30 minutes while an 8° to 10°C gradient is maintained between the perfusion and the patient's core temperature. When a rectal temperature of 20°C is reached, the aorta is cross-clamped, and 500 cc of cardioplegic solution is administered. Once cardiac arrest is achieved, bypass is terminated, and the patient is temporarily exsanguinated into an oxygen reservoir. The patient's brain is protected by placing ice bags around the head. At this point

there is no anesthesia, ventilation, or circulation. To reduce the incidence of complications, circulatory arrest time is best limited to 45 minutes. An elliptical incision is made around the ostium of the renal vein and carried superiorly along the length of the vena cava. The incision can extend into the right atrium or ventricle, depending on the superior extent of the thrombus. Cardiopulmonary bypass and deep hypothermic circulatory arrest permit the thrombus to be removed in a bloodless field and the interior of the vena cava and heart to be inspected under direct vision ( Fig. 8-4). It is not uncommon to find some degree of adherence of the tumor to the endothelium. In this case, the tumor thrombus can be “endarterectomized” from the interior of the vena cava or atrium. Reconstruction of the vena cava is as previously described.

FIG. 8-4. (A) The ostium of the renal vein is circumferentially incised, and the right atrium is opened. (B) Following removal of the tumor thrombus, the atriotomy and vena cavotomy incisions are closed. Reprinted with permission from Novick AC, Montie JE. Surgery for renal cell carcinoma. In: Novick AC, Streem SB, Pontes JE (eds.), Stewart's operative urology, Vol. 1, 2nd ed. Baltimore: Williams and Wilkins, 1989.

Following closure of the vena cavotomy, cardiopulmonary bypass is begun. The patient is slowly warmed using a 10°C gradient between the bypass machine and a warming blanket. Mannitol (12.5 g) is given along with 1 g of CaCl 2 when core temperature reaches 25°C. Electrical defibrillation is necessary if the heart does not resume spontaneous beating. Following resumption of cardiac activity, blood is returned to the patient from the oxygen reservoir. Following the rewarming process, which can take up to 1 hour, heparin is neutralized with protamine. The patient is returned to the cardiac ICU intubated.

OUTCOMES
Complications Intraoperative complications include excessive bleeding and coagulopathy. Coagulopathy is more common with prolonged cardiopulmonary bypass and cardiac arrest times. Intraoperatively, red blood cells, platelets, fresh frozen plasma, and calcium chloride are routinely administered. Furosemide and/or mannitol is given if urine output remains low. Transient hypotension can occur when the vena cava is clamped. This can be managed with volume expansion and is less of a problem if venous collaterals have developed with a completely occluded vena cava. Embolization of a segment of tumor thrombus can be a potentially lethal intraoperative complication, and extreme care should be taken when handling the vena cava to prevent such an occurrence. Postoperatively, several complications can occur because of the magnitude of the surgical procedure or the use of cardiopulmonary bypass. Potential complications include caval thrombosis, deep venous thrombosis, pulmonary embolus, postoperative bleeding, or coagulopathy. Patients may also develop hepatic dysfunction, renal failure, sepsis, or myocardial infarction. Although the mortality rate associated with this procedure is tolerable, most patients who die of complications within the first postoperative month succumb to multisystem organ failure. Results The 5-year survival rates in most reported large series vary from 14% to 68% following complete surgical removal of the renal tumor and caval extension. 1,6,7 Differences in reported survival may reflect several factors, including local extension of the primary tumor, presence of lymphatic or visceral metastases, level of caval tumor extension, or invasion into the vascular wall. lt is generally agreed that patients with metastatic disease and significant perinephric fat involvement tend to have a poor prognosis. The majority of patients eventually dying of their disease succumb to metastases, which suggests that occult metastatic disease is frequently present at the time of surgery. 6 We believe that patients with good performance status who have tumors confined to the renal capsule and are without evidence of metastatic disease are ideal candidates for this surgery and have improved long-term survival. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Glazer AA, Novick AC. Long-term follow up after surgical treatment for renal cell carcinoma extending into the right atrium. J Urol 1996;155:448. Marshall FF, Dietrick DD, Baumgartner WA, Reitz BA. Surgical management of renal cell carcinoma with intracaval neoplastic extension above the hepatic veins. J Urol 1988;139:1166. Marshall FF, Reitz BA. Supradiaphragmatic renal cell carcinoma tumor thrombus: indications for vena caval reconstruction with pericardium. J Urol 1985;133:266. Marshall FF, Reitz BA. Technique for removal of renal cell carcinoma with suprahepatic vena caval tumor thrombus. Urol Clin North Am 1986;13:551–557. Marshall VF, Middleton RG, Holswade GR, Goldsmith EI. Surgery for renal cell carcinoma in the vena cava. J Urol 1970;103:414. Polascik TJ, Partin AW, Pound CR, Marshall FF. Radical nephrectomy with intrahepatic or supradiaphragmatic intracaval thrombectomy for renal cell carcinoma: long-term outcome analysis. (submitted) 7. Skinner DG, Pfeister RF, Colvin R. Extension of renal cell carcinoma into the vena cava: the rationale for aggressive surgical management. J Urol 1972;107:711. 8. Treiger BFC, Humphrey LS, Peterson JCV, et al. Transesophageal echocardiography in renal cell carcinoma: an accurate diagnostic technique for intracaval neoplastic extension. J Urol 1991;145:1138.

Chapter 9 Transplant Nephrectomy Glenn’s Urologic Surgery

Chapter 9 Transplant Nephrectomy
J. Thomas Rosenthal

J. T. Rosenthal: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90095-1731.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Success rates for kidney transplantation continue to improve. Graft survival at 1 year is now 83% for all cadaver transplants done in the United States and 93% for living related transplants. 9 Some programs report greater than 90% survival even for cadaver grafts, and greater than 85% survival at 2 years. 1 Despite these good results, they are not yet 100%, so a percentage of patients transplanted will fail. The fate of the kidney after transplant failure is the subject of this chapter.

DIAGNOSIS
The diagnosis of a failed renal transplant is not difficult because the patients are azotemic and require dialysis. The common causes of graft failure are technical, such as arterial or venous thrombosis, and rejection, acute or chronic. Irreversible rejection can occur despite optimum immune suppression. Sometimes rejection results from the necessity to reduce or stop immune suppression to treat a life-threatening infection. Acute rejections occurring more than 1 year after the transplant are often a result of patient noncompliance with drugs. In any of these situations, patients will usually have undergone some radiologic investigation such as Doppler ultrasound or renal scan to ascertain whether there was a treatable cause for graft dysfunction. Renal biopsy is often performed once it is clear that there is no technical cause of graft dysfunction to determine the degree of acute or chronic rejection present. Once it is certain that renal failure is irreversible, immune suppression is withdrawn, chronic dialysis is reinstituted, and a decision is made as to whether or not it is necessary to remove the allograft.

INDICATIONS FOR SURGERY
Grafts failing within the first 12 months are almost always removed prophylactically regardless of the cause of graft loss and whether or not there are specific symptoms present. This is because if the graft is left in place, significant symptoms necessitating its removal will almost always occur. 3 Most of these are either technical failures, which occur in the first few weeks after transplant, or acute rejection, which is manifest in the first 2 to 3 months. Those grafts failing after 12 months are left in place initially, immunosuppression is stopped, and patients are followed. Approximately 50% of these patients will develop symptoms consistent with acute rejection such as fever, malaise, graft tenderness, or gross hematuria. These symptoms can be confused with an infection and can be difficult to distinguish from rejection. onetheless, these symptoms occurring in a patient who has recently had a failed kidney transplant are usually a result of rejection, and so a prolonged workup, searching for the source of fever, is not usually necessary or productive. These symptoms can occur many months after the patient is back on dialysis and many after the transplantation, which can make the diagnosis more difficult. Sometimes proceeding with nephrectomy is the only way to distinguish the symptoms. There has been some suggestion that repeat transplants may do worse in patients who have had the primary graft removed. 1 The data on this point are not absolutely convincing, although they further support the notion of doing the nephrectomy only for specific indications. A rare indication for allograft nephrectomy is the presence of a mass lesion of the transplant kidney. We have seen this twice in approximately 1,500 transplants. The diagnosis is usually made by ultrasound done for routine evaluation of the transplant, which detects the mass, followed by CT scan and biopsy to confirm the diagnosis. This may involve lesions unintentionally transmitted from the donor or lesions arising de novo in the transplant kidney. In the early era of transplantation, nephrectomy was often required because of technical complications such as urinary fistula or graft hemorrhage, which were not lethal to the kidney but could not be repaired. This almost never occurs now.

ALTERNATIVE THERAPY
One alternative to allograft nephrectomy is watchful waiting, as described earlier. In those whose graft is lost beyond 1 year from the transplant, about half the time no symptoms occur, and the kidney shrinks and sometimes will calcify. In those chronic cases where symptoms do occur, a short course of oral steroids is sometimes given over a week or two with prednisone, 15 mg/day initially and quickly tapered. This will ameliorate symptoms and, in a rare case, may eliminate the need for nephrectomy. Another alternative is radiologic embolization, which we have performed in a few cases. The drawbacks are having a large kidney occluded and uncertainty as to whether this would make a subsequent transplant on that side more difficult in the future. Another drawback is that it may be hard to angiographically identify the renal artery(ies) and catheterize it (them) in the small chronic kidney, which commonly has significant intimal thickening from the rejection process. One group has reported the combination of ethanol and stainless steel coil transvenous catheter ablation in 14 patients. 4 Postembolization syndrome occurred in 11 of 14 patients, and one patient had an abscess develop in the graft.

SURGICAL TECHNIQUE
The patient is placed on the operating room table in the supine position. In most cases the previous transplant incision is a lower quadrant incision ( Fig. 9-1). The nephrectomy incision is over that incision and extended laterally if necessary. If the transplant had been done through a midline transperitoneal incision, it may be necessary to reopen that incision and go transperitoneally to get to the kidney. When the transplant has been transperitoneal, it is often swollen and palpable in the lower abdomen, and thus, it is possible to make the incision directly over the kidney, incise the external and internal obliques, and reach the kidney capsule. Because the kidney is usually stuck to its surrounding structures, it is possible in this situation to stay extraperitoneal and remove the kidney. Routinely, in reopening a lower quadrant incision, the scars in the external and internal obliques are reincised. Once the internal oblique is incised, the kidney should be approached as laterally as possible because the peritoneum is often draped over the kidney in the line of the incision. If the peritoneum is opened and the patient is on peritoneal dialysis. this creates a problem to use the P-D catheter until the peritoneum reseals, necessitating temporary vascular access for dialysis. This is particularly problematic in children, and it is helpful to avoid this circumstance.

FIG. 9-1. The patient is positioned in the supine position, and the incision is made over the previous lower quadrant transplant incision.

In cases where the kidney has rejected, it is sometimes swollen to two to three times its normal size. The kidney is palpable superficially, but the bulk of the size is hidden in the flank. This may be the situation even where there has been chronic rejection because acute rejection is often superimposed. A very long incision may not create the amount of access needed to dissect the kidney under direct vision because of the lie of the kidney in the abdomen and flank. The subcapsular approach, in addition to helping with the kidney being stuck to surrounding structures, helps with these very large kidneys because it allows the mobilization of the kidney to be done blindly and delivers the large kidney into the incision so the pedicle can be dissected under direct vision. This is not unlike the dissection of the prostate in a retropubic or suprapubic prostatectomy for BPH. Kidneys that have been lost in the first few days after transplant for technical reasons can be removed in toto, including the whole capsule, because these kidneys are not usually stuck to the surrounding tissues. Kidneys that have undergone rejection are usually stuck to the pelvic side wall, iliac vessels, and peritoneum. Attempts to remove these extracapsularly can result in injury to vessels or bowel in addition to being very difficult to do. 6 A subcapsular approach simplifies the operation considerably, reducing risk of damage to adjacent structures. 7 The renal vessels are ligated well into the hilum. After the internal and external obliques are opened, the surface of the kidney can be balloted and often has a bluish tint. A tiny incision in the capsule is made with the knife, confirming the presence of renal parenchyma ( Fig. 9-2). The Metzenbaum scissors are used to dissect bluntly under the capsule and enlarge the capsulotomy. A finger sliding under the capsule should meet a plane that dissects easily ( Fig. 9-3). An exception is the rare circumstance where the kidney is small and partly calcified, where the subcapsular plane may be quite difficult to appreciate. In the more usual circumstance, the plane between capsule and parenchyma is extended around the entire kidney except the hilum, and the kidney is delivered into the wound. It may be necessary to extend the fascial incisions to allow this depending on the size of the kidney. The kidney will now be tethered by the vessels and the ureter. Care must be taken not to avulse these in delivering the kidney into the wound.

FIG. 9-2. The incision is carried through the external and internal oblique muscles to the capsule of the kidney, where a small capsular incision is made.

FIG. 9-3. The renal parenchyma is enucleated subcapsularly.

Breaking the connections between parenchyma and capsule will usually result in a moderate amount of bleeding from the raw internal capsular surface, and it is difficult to control this bleeding until the kidney is out. The reflected capsular surface overlying the hilum is then incised sharply on either the medial or lateral surface, depending on which is more accessible( Fig. 9-4).

FIG. 9-4. The hilar vessels are exposed by making an incision close to the renal parenchyma (A) and incising the overlying capsule (B).

The peritoneum can be entered here, so it is best to make this incision close to the renal parenchyma. The other advantage is that ligating the vessels well into the hilum minimizes the risk of damage to the iliac artery. A theoretical disadvantage to this approach is that donor material is left in situ. Blowout of this remnant is a possibility but appears to occur only if the suture line itself becomes infected. 8 Therefore, leaving this material in place usually causes no problem. The only other drawback is that if a third transplant is ever required on that side, it may be necessary to dissect this scar out at the time of the transplant. The potential problems that this presents are outweighed by the complexity of the nephrectomy and risk of trying to control the proximal and distal iliac artery and vein in the presence of a large, friable failed allograft and can sometimes be avoided by going higher on the vessels, to the common iliacs or aorta and vena cava. The hilar structures may be densely adherent to one another, but it is usually possible to dissect arteries, veins, and ureters free for individual ligation ( Fig. 9-5). It may be possible to obtain only enough length to handle the proximal vessels without ligating the kidney side. A finger over the kidney side may be sufficient until all the vessels have been ligated. Ligation plus suture ligation with 2-0 or 3-0 cardiovasculars wedged-on sutures provides security that there will be no major bleeding postoperatively. Silk, Prolene, or some other nonabsorbable suture can be used. The vessels themselves are often friable, and care must be taken not to saw through them. Occasionally a vessel is torn flush with the dense scar at the base of the kidney overlying the iliac artery or vein, which can not be seen. A figure-of-eight stitch can control it, but care must be taken not to pass the needle too deeply into the scar because the iliac vessels may be very superficial.

FIG. 9-5. The renal artery and vein are suture ligated (A) and divided (B).

In the event that it is elected to remove all the donor vessels, as might be done in the early technical failure case, it may be necessary to repair the defect left in the iliac artery using a saphenous vein patch or some nonautologous graft to avoid narrowing the iliac artery ( Fig. 9-6). Repair of the vein is not usually necessary other than to oversew the venotomy. Side-biting vascular clamps are helpful to control the iliacs for these repairs.

FIG. 9-6. If the renal allograft artery is completely removed, it may be necessary to cover the defect in the iliac artery with a patch graft.

Once all the hilar structures have been ligated, hemostasis of the capsule must be obtained. This can be done by using the flat part of the electrocautery or an argon-beam coagulator. The entire capsular surface should be inspected. The unacceptable alternative, unless it is absolutely impossible to obtain a dry field, is to drain the space, which runs the risk of potentiating infection of the space, even if a closed drain is used. Spray thrombin can be liberally applied after coagulation. Care should be taken not to rub off the carefully obtained surface hemostasis. Antibiotic irrigation is carried out, although the evidence that this prevents infection in this situation is lacking. The wound is closed in one or two layers, depending on what is possible. I prefer interrupted 2-0 Prolene, but absorbable and running sutures are possible. Usually there is no value in specifically closing the remaining renal capsule.

OUTCOMES
Complications Complications reported in earlier series included major perioperative bleeding, wound infections, and intraoperative injury to adjacent structures. 5 The incidence of hemorrhage following transplant nephrectomy is 5% to 6%. 2 Routine use of the subcapsular approach can reduce these complications to less than 1%. 6 The other type of complication is less related to the operation than it is to immune suppression. Even though patients undergoing allograft nephrectomy have had their immune suppression stopped before the nephrectomy, it may have been only a few days before. The sequelae of immune oversuppression may take days or weeks to manifest themselves. In the UCLA series, two perioperative deaths were from B-cell lymphoma related to immune oversuppression. Persistent fever after nephrectomy should be carefully evaluated. Results Transplant nephrectomy can be performed with minimal morbidity, though it can be extremely challenging in the postrejection setting. The issue of the value of transplant nephrectomy to reduce the possibility of rejection of subsequent renal allografts remains controversial. CHAPTER REFERENCES
1. Abouljoud MS, Deiehoi MH, Hudson SLL, Diethelin AGL. Risk factors affecting second renal transplant outcome, with special reference to primary allograft nephrectomy. Transplantation 1995;60:138–144. 2. Chiverton SG, Mufic JA, Allen RD, Morris P. Renal transplant nephrectomy. Surg Gynecol Obstet 1987;164:324–328. 3. DiSesa VJ, Tilney NL. Conservative management of the failed renal allograft: indications for transplant nephrectomy. Curr Surg 1982;39:417–418. 4. Lorenzo V, Diaz F, Perez L, et al. Ablation of irreversibly rejected renal allograft by embolization with absolute ethanol. Am J Kidney Dis 1993;22:592–595. 5. O'Sullivan D, Murphy DM, McLean P, Donovan MG. Transplant nephrectomy over 20 years; factors involved in associate morbidity and mortality. J Urol 1994;151:855–858. 6. Rosenthal JT, Peaster ML, Laub D. The challenge of kidney transplant nephrectomy. J Urol 1993;149:1395–1397. 7. Sutherland DER, Simmons RL, Howurti R. Intracapsular technique of transplant nephrectomy. Surg Gynecol Obstet 1978;146:950–952. 8. Starnes HF, McWhinnie DL, Bradley JA, et al. Delayed major arterial hemorrhage after transplant nephrectomy. Transplant Proc 1984;16:1320–1323. 9. United Network for Organ Sharing. Center specific report. Richmond, VA: Author, 1997.

Chapter 10 Renovascular Disease Glenn’s Urologic Surgery

Chapter 10 Renovascular Disease
John A. Libertino

J. A. Libertino: Department of Surgery, Harvard University Medical School, Boston, Massachusetts 02115, and Department of Urology, Tufts University School of Medicine, Lahey Clinic Medical Center, Burlington, Massachusetts 01805.

Diagnosis Indications for Surgery Alternative Therapy Surgical Techniques Aortorenal Bypass Graft Procurement of Saphenous Vein Technique of Insertion of Saphenous Vein Graft Alternative Arterial Bypass Grafts Splenorenal Arterial Bypass Hepatorenal Bypass Graft Renal Autotransplantation andEx Vivo Bench Surgery Techniques for Autotransplantation Outcomes Complications Results Conclusion Chapter References

Although the true incidence of renovascular hypertension is unknown, it is estimated that between 5% and 10% of all hypertensive patients suffer from renovascular hypertension. 4 During the past two decades, there have been dramatic changes in the diagnosis and treatment of renovascular hypertension. There is clearly a better understanding of the renin–angiotensin system, and newer, more potent antihypertensive medications are available. In addition, newer diagnostic radiologic procedures, such as digital subtraction angiography, captopril renal scans, balloon angioplasty, and newer surgical techniques, have dramatically changed the ways in which we diagnose and treat renovascular hypertension today.

DIAGNOSIS
In the past, the clinician's task of identifying potentially curable patients in a safe, cost-effective, and reliable manner was difficult. Recently we have been given the means to reliably identify patients with a physiologically significant renal artery stenosis that, in the past, might have eluded the physician. A single-dose captopril test is reported by some investigators to be a reliable screening test and is well suited for outpatient use. Although it is less reliable in patients with a degree of renal insufficiency, the peripheral plasma renin response to a single dose of oral captopril has proved to be a simple and sensitive test. 7 In patients with a functional renal artery stenosis, ACE inhibitors lead to a disproportionate increase in peripheral plasma renin activity as a result of the disappearance of the inhibitory effect of angiotensin II on renin secretion. This phenomenon can be used diagnostically to detect unilateral renal artery stenosis. In cases of bilateral stenosis, however, the test cannot be used reliably as an indicator of renal artery stenosis. 1 However, Postma et al.8 have reported that the captopril test is not a reliable screening test, and as a consequence, the value of this procedure as a screening test in detection of renal artery disease is at present unsettled. Another test that has recently been analyzed for the detection of renal artery stenosis is the renal scintigram with isotopic nephrography following the administration of captopril.9 In a functional renal artery stenosis when ACE is inhibited, the glomerular infiltration rate decreases as a consequence of the decreased inhibition of angiotensin II on the vasa efferens. This can be demonstrated by technetium DTPA scintography. The sensitivity of this means of identifying renal artery stenosis varies from 71% to 92% with a specificity of 72% to 97%. In the hipuran scintigram, the patient with renal artery stenosis showed a continual enrichment of the isotope in the renal cortex, probably related to a decreased excretion of hipuran because of the ACE inhibitor. Digital venous subtraction angiography (DSA), in my opinion, remains the most definitive and reliable screening test available. It has a sensitivity and specificity of nearly 90%. Intravenous DSA is susceptible to artifacts such as crossing vessels and from intestinal motility. In these patients, digital arterial subtraction angiography achieves an excellent view of the renal circulation using less contrast dye than conventional angiography. Because this method uses a comparatively small 4- to 5-Fr catheter, groin hematoma is rare, and the technique can be used in an outpatient setting. Several other new modalities have appeared recently. Duplex Doppler ultrasound scanning is now a recognized way of demonstrating and locating focal renal artery stenosis. This noninvasive test is gaining acceptance in the diagnosis of both atherosclerosis and fi-bromuscular hyperplasia. In a recent study, Ferdinandi et al. 3 report the ability to detect renal artery stenosis with a success rate of over 90%. Further studies will be necessary to determine the ultimate role of duplex sonography and color Doppler evaluation of renal ar-tery stenosis. At present, digital subtraction angiography in conjunction with divided renal vein renin assays, which demonstrate contralateral suppression, remains the best screening test and predictor of treatment outcome. In those patients who have azotemia, in whom contrast is contraindicated, MRI angiography and CO 2 digital subtraction angiography are useful 12 These studies avoid the use of contrast and obviate the occurrence of contrast toxicity.

INDICATIONS FOR SURGERY
The surgical management of renovascular hypertension has changed dramatically in the last two decades. In the early 1970s, we demonstrated that renal function could be preserved or restored by renal revascularization of nonfunctioning kidneys with totally occluded renal arteries. This contribution led to the notion that if patients who had totally occluded renal arteries and nonfunctioning kidneys could have restoration of renal function, then we could treat patients with renal artery stenosis who had azotemia with the expectation that they could also have preservation or improvement of renal function. 5 Progressive azotemia in the elderly atherosclerotic patient population is now one of the indications for renal revascularization. Secondly, based on our observations, the use of alternative bypass procedures has significantly reduced the morbidity and mortality of high-risk patients undergoing renovascular surgery. The use of the hepatic artery, the gastroduodenal artery, and other alternative procedures instead of the aortorenal saphenous vein bypass graft has not only reduced the morbidity and mortality of surgery but, in doing so, has dramatically changed the nature of our patient population.

ALTERNATIVE THERAPY
Converting enzyme inhibitors prevent the conversion of angiotensin I to angiotensin II. These drugs in conjunction with calcium channel blockers have greatly improved the medical management of patients who suffer from renovascular hypertension. Unfortunately, even if adequate blood pressure control is maintained by pharmacologic means, progression of renal artery disease is not prevented, and renal ischemia and renal damage may clearly progress. 11 When medical management fails or azotemia progresses, then balloon angioplasty and surgical treatment must be considered. The choice between angioplasty and surgery relies on a well-defined set of criteria established by published results. Angioplasty is indicated in the treatment of fibrous dysplasia and atherosclerosis of the mid–main renal artery. Surgery is indicated for the treatment of osteal atherosclerosis and for branch lesions of the renal artery. Renal artery aneurysms are a different problem. Surgery is indicated when they are the cause of hypertension. Also aneurysms larger than 2 cm in diameter that are noncalcified, especially in gestational women, should be repaired, as they are prone to rupture during pregnancy.

It is our feeling that patients who develop recurrent disease following balloon angioplasty are probably best subjected to surgical management, as repeat balloon angioplasty is associated with a significant complication rate. Use of the thoracic aorta may also be a viable alternative on the left side because the thoracic aorta is usually less atherosclerotic than the abdominal aorta. 6

SURGICAL TECHNIQUES
Aortorenal Bypass Graft The widespread popularity of the bypass graft for renal artery disease was attained by virtue of its technical ease of insertion and the favorable short- and long-term patency rates achieved. Bypass grafts are applicable to almost any disease process involving the main renal artery or its branches. This procedure also eliminates the more hazardous and tedious dissection of the juxtrarenal portion of the aorta required in endarterectomy. Bypass grafts are particularly suitable for fibrous lesions that affect long and multiple segments of the renal artery and its branches ( Fig. 10-1). Dacron, autogenous artery (hypogastric and splenic), and autogenous saphenous vein may be chosen as aortorenal bypass grafts in properly selected patients.

FIG. 10-1. (A) An aortogram shows a double right renal artery with stenoses at the ostia of both trunks. (B) A postoperative aortogram with the vein graft making a side-to-side anastomosis to the stenotic lower renal artery and an end-to-end anastomosis to the distal stump of the upper renal artery.

Dacron has been applied extensively in renal artery reconstruction but has been associated with a relatively high rate of early thrombosis. Excellent long-term patency rates have been reported with a segment of autogenous hypogastric artery. Such a graft matches the size of the renal artery and is sutured more simply than the Dacron prosthesis. Autogenous hypogastric artery is the most favorable graft material for children with renal artery disease because the saphenous vein is usually too small and is more prone to aneurysmal dilation than in adults. The major disadvantage is that the hypogastric artery is often the first to be involved with generalized atherosclerosis and therefore is not suitable graft material in older patients. It is also a short vessel and occasionally is technically more difficult to insert between the renal arteries and aorta. During the past two decades the autologous saphenous vein has emerged as our preferred graft material and is the most common source for restoration of renal blood flow at our hospital. Saphenous vein is readily available and closer in size to the lumen of the renal artery than other vascular conduits. Its intima is less thrombogenic than prosthetic material and accommodates the creation of a precise contoured anastomosis with a delicate thin-walled distal renal artery. Patent anastomoses can be achieved with the most challenging 2- to 3-mm-lumen branches beyond the major bifurcation. Because of its inherent properties and the favorable surgical results obtained, saphenous vein has become the conduit of choice for aortorenal bypass at most major renovascular centers. If the saphenous veins are not available, we use cephalic vein and Gore-Tex graft, in that order, as substitutes. Procurement of Saphenous Vein The procurement of an adequate segment of the long saphenous vein is critical to the success of the graft procedure. Meticulous technique in exposure and excision of the vein is essential to prevent mural trauma and ischemia. Improper harvesting of the vein may result in the delayed complications of stenosing intimal hyperplasia and aneurysmal dilation. Removal of the saphenous vein should be performed by an experienced surgeon. The saphenous vein is usually obtained from the thigh opposite the renal lesion so that two surgeons may simultaneously expose the renal vessels and mobilize the graft, shortening the operative time. The vein is mobilized through a single long incision in the upper thigh ( Fig. 10-2), which begins parallel to and below the groin crease over the palpable femoral pulses and is extended toward the knee after the junction of the saphenous and femoral veins has been exposed. The incision should be made directly over the vein to avoid producing devascularized skin flaps that can result in necrotic edges and wound sepsis. Finger dissection between the trunk of the vein and the skin is helpful to ensure accurate placement of the incision and, thus, to avoid development of these flaps (see Fig. 10-2). On the day before operation, the course of the saphenous vein is outlined with an indelible pen while the patient is standing.

FIG. 10-2. (A) Position of patient for harvesting of saphenous vein graft. (B) Line of incision for saphenous vein graft harvest. (C) Exposure of saphenous vein.

A 20-cm-long vein graft with an outside diameter of 4 to 6 mm is usually adequate for reconstruction of the renal artery. Excess vein should always be available for revision of any intraoperative technical problems that may occur during anastomosis. The vein is handled gently without stretching or tearing its branches. The tributaries are tied in continuity with fine silk before they are divided. The areolar tissue is dissected from the specimen, and the adventitia is left undisturbed. To decrease transmural ischemia, the vein graft remains4 in situ until the renal vessels are mobilized and it is ready to be used. If the graft is inadvertently removed prematurely, it is placed in cold Ringer's lactate solution or autologous blood, even if only a short period of time will ensue. The distal end of the vein is transected, cannulated with a Marks needle, and secured with a silk tie ( Fig. 10-3). A dilute heparinized solution of autologous blood distends the vein graft before the proximal is transected. This step helps to identify any untied tributaries or unrecognized leakage and washes out any residual blood clots. The vein is distended to a minimal diameter of 5 to 6 mm by exerting gentle pressure on the syringe. The proximal end of the vein is transected, and the vein graft is now ready for use. The thigh incision is not closed until the bypass procedure has been completed to ensure that any delayed bleeding caused by the heparinized state is identified and controlled.

FIG. 10-3. Harvest of saphenous vein graft.

Technique of Insertion of Saphenous Vein Graft Heparin is initially given systemically after the surgical dissection has been completed and approximately 30 minutes before the arteries are clamped. The saphenous vein graft should be oriented properly to avoid misalignment during implantation. Either an end-to-end or an end-to-side anastomosis can be accomplished, depending on the anatomic situation encountered. An end-to-end anastomosis is preferred under usual circumstances because it permits the best laminar flow. The aorta, which has already been mobilized and exposed from the renal arteries to the level of the inferior mesenteric artery, is carefully palpated to determine a suitable soft location for the anastomosis that is relatively free of atherosclerotic plaque. A medium-sized DeBakey clamp is placed on the anterolateral portion of the infrarenal aorta in a tangential manner. A vertical 13- to 16-mm aortotomy is made without excising any of the aortic wall or attempting to perform a localized endarterectomy (Fig. 10-4), which may dislodge intimal plaque fragments that can form emboli to the lower extremities when the clamp is released.

FIG. 10-4. The bypass graft is placed along the lateral aortic wall to determine the best position for its placement. (From Novick AC, Streem SB, Pontes JE, eds. Stewart's operative urology. Baltimore: Williams & Wilkins, 1989.)

Excision of the aortic wall is not necessary because intraluminal aortic pressure spreads the edge of the linear aortotomy to the appropriate dimensions when the clamp is released. The vein graft is anastomosed to the aorta with continuous 5-0 Proline suture after it has been satisfactorily spatulated ( Fig. 10-5). A microvascular Schwartz clamp is placed on the end of the saphenous vein graft, and the aortic clamp is released. The graft is allowed to lie anterior to the vena cava on the right side or anterior to the renal vein on the left side. Although it is preferable to leave the vein too long than too short, it should not be so long as to bend into an acute angle at any point. The renal artery is secured distally with a smooth-jawed Schwartz microvascular clamp placed on either the distal main renal artery or its branches. The proper site for the arterial anastomosis is selected. An end-to-end anastomosis is performed utilizing a continuous 6-0 Proline suture or interrupted sutures of the same material, depending on the diameter of the anastomosis (Fig. 10-6). When the saphenous vein graft is being anastomosed with two branches 3 mm or less in size, interrupted sutures are chosen. An interrupted suture line is also selected in children to prevent a pursestring effect with growth of the vessels when the patients become older. This effect may also occur with running synthetic monofilament sutures when too much tension is applied during the creation of the anastomosis. The pursestring effect can be avoided by placing sutures at four quadrants in the arterial wall before beginning the anastomosis. Operating loupe magnification and fiberoptic headlamps are very helpful at this point in the operation to allow precise placement of the sutures, particularly when exposure in the renal artery is difficult.

FIG. 10-5. Following partial aortic occlusion, an oval aortotomy is made for end-to-side anastomosis with the spatulated bypass graft. (From Novick AC, Streem SB, Pontes JE, eds. Stewart's operative urology. Baltimore: Williams & Wilkins, 1989.)

FIG. 10-6. (A) Anastomosis of the graft to the aorta is performed with interrupted vascular sutures. (B) After completion of the aortic anastomosis, the renal artery is prepared for anastomosis with the graft. (C) A spatulated end-to-end anastomosis of the graft and distal renal artery is performed. (From Novic AC, Streem SB, Pontes JE, eds. Stewart's operative urology. Baltimore: Williams & Wilkins, 1989.)

The single most important factor responsible for long-term patency is a wide flawless anastomosis with the renal artery. After completion of the anastomosis, the microvascular bulldog clamps are removed from the distal renal circulation and the saphenous vein graft, permitting reconstitution of the renal circulation ( Fig. 10-7).

FIG. 10-7. Completed aortorenal bypass operation. (From Novic AC, Streem SB, Pontes JE, eds. Stewart's operative urology. Baltimore: Williams & Wilkins, 1989.)

Alternative Arterial Bypass Grafts Extensive atherosclerosis, previous aortic surgery, and complete thrombosis of the aorta may preclude the use of the aortorenal bypass procedure for renal artery reconstruction. When the surgeon is treating a patient with stenosis of the right renal artery in association with these pathologic limitations on the aorta, a splenorenal or hepatic-to-renal artery saphenous vein bypass or gastroduodenal-to-renal artery bypass procedure can be selected. Splenorenal Arterial Bypass Splenorenal arterial bypass has many desirable features as a substitute for aortorenal bypass in patients with stenosis of the left renal artery. It is particularly suitable for patients who have diffuse atherosclerotic disease or thrombosis of the aortic lumen and for those who have previously undergone difficult aortic reconstructions. The splenic artery has the advantages of being an autogenous artery that has not been separated from its nutrient vaso vasorum, of being exposed without difficulty by a relatively uncomplicated anatomic dissection, and of requiring only one vascular anastomosis. Carefully monitored oblique and lateral angiography of the celiac axis is required to determine the patency of this artery because atherosclerosis can affect the arterial lumen early in the patient's life. Surgical exploration and intraoperative evaluation by palpation and measurement of splenic blood flow are also helpful in establishing its suitability for renal revascularization. If the blood flow is less than 125 ml/min, the splenic artery should probably not be utilized for renal artery bypass. We now prefer to expose the splenic artery through a supracostal 11th-rib flank incision ( Fig. 10-8). The dissection is continued along the upper border of the rib. The overlying latissimus dorsi, the serratus posterior inferior, and the intercostal muscles are divided. Division of the intercostal ligament permits the rib to move freely. The external, internal oblique, and transversus abdominis muscles are divided, and the intercostal muscle attachments on the distal 1 inch of the rib are divided carefully until the corresponding intercostal nerve is identified. The investing fascia around the nerve is entered. Dissection in this plane allows an extrapleural approach and generally avoids entry into the pleural cavity. This approach also allows excellent exposure, for the ribs are free to pivot downward in a “bucket-handle” fashion (Fig. 10-9).

FIG. 10-8. Supracostal 11th-rib incision: (A) posterior view; (B) anterior view. (C) The costovertebral ligament must be divided to allow the rib to pivot inferiorly. (D) Closure of incision, taking care to spare the intercostal nerves. The diaphragm is not incorporated in the closure.

FIG. 10-9. Bilateral subcostal “bucket handle” incision.

The plane between Gerota's fascia and the adrenal gland posteriorly and the pancreas anteriorly is entered. The splenic artery is identified at the upper border of the pancreas. Its enveloping fascia is entered, and the splenic artery is mobilized by a purely retroperitoneal approach. Several small pancreatic branches are identified, isolated, ligated, and divided. The splenic artery can usually be mobilized from the splenic hilum to the celiac axis without difficulty, and it provides sufficient length to reach the left renal artery. After the splenic artery is mobilized, a sponge soaked with papaverine is placed on it to permit it to dilate. The artery is divided just proximal to its primary bifurcation in the hilum of the spleen, after a suitable vascular clamp has been applied to the origin of the artery. If necessary, the artery may be dilated with a Gruntzig balloon or Fogarty catheter intraoperatively to obtain maximum caliber. Removal of the spleen is not necessary because it continues to receive adequate blood flow from the short gastric arteries. The left kidney is approached posteriorly, and the left renal artery is identified and mobilized ( Fig. 10-10). The renal artery is ligated at the aorta, and an end-to-end anastomosis between the splenic artery and the distal renal artery is carried out using continuous or interrupted 6-0 Proline sutures ( Fig. 10-10).

We have employed this approach in nearly 100 patients and now prefer it to the traditional transabdominal technique.

FIG. 10-10. Technique of splenorenal bypass. Note that the pancreas is lifted cephalad in order to expose the splenic artery.

On rare occasions, a sufficient length of splenic artery cannot be achieved. In this instance, an interposition saphenous vein graft from the splenic artery to the renal artery can be utilized. This maneuver enables the creation of a tension-free anastomosis ( Fig. 10-11).

FIG. 10-11. An aortogram shows a splenorenal end-to-side bypass.

Splenic artery disease, the risk of pancreatitis, and the formation of a pancreatic pseudocyst are some of the limitations that have restricted splenorenal bypass as a routine procedure in the management of disease of the left renal artery. Hepatorenal Bypass Graft Arising from the celiac axis and continuing along the upper border of the pancreas, the hepatic artery reaches the portal vein and divides into an ascending and a descending limb. The ascending limb is a continuation of the main hepatic artery upward within the lesser omentum; it lies in front of the portal vein and to the left of the biliary tree. The descending limb forms the gastroduodenal artery. In the porta hepatis, the hepatic artery ends by dividing into the right and left hepatic branches, which supply the corresponding lobes of the liver ( Fig. 10-12). The anatomic variations in the hepatic circulation must be appreciated before this procedure can be utilized. The right hepatic artery is more variable than the left. It may be anterior (24% of patients) or posterior (64% of patients) to the common bile duct, and in 12%, this artery arises from the superior mesenteric artery ( Fig. 10-13). The hepatic artery lies anterior (91% of patients) or posterior (9% of patients) to the portal vein. In addition, the left hepatic artery arises from the left gastric artery in 11.5% of patients.

FIG. 10-12. Normal course of the main hepatic artery and its various branches. (From Novick AC. Diminished operative risk and improved results following revascularization for atherosclerotic renal artery disease. Urol Clin North Am 1984;11:435.)

FIG. 10-13. Separate origins of the left and right hepatic arteries from the celiac and superior mesenteric arteries, respectively. (From Novick AC. Diminished operative risk and improved results following revascularization for atherosclerotic renal artery disease. Urol Clin North Am 1984;11:435.)

Careful dissection of the porta hepatis is essential, and the common hepatic, gastroduodenal, and right and left hepatic arteries should be identified before an anastomotic procedure is attempted. Vascular elastic loops are placed about these vessels, and the common bile duct and portal vein are identified. After careful dissection and mobilization of the renal artery, clamps are placed on the proximal portion of the common hepatic artery and its distal branches. The

gastroduodenal artery is divided ( Fig. 10-14). The inferior surface of the hepatic artery is mobilized from the underlying portal vein and the common bile duct. An arteriotomy, 10 to 12 mm in length, is made in the anterior inferior wall of the common hepatic artery, beginning at the ostium of the gastroduodenal artery. A reversed autogenous saphenous vein is inserted with an end-to-side anastomosis between the vein graft and the hepatic artery. This maneuver is usually accomplished with a continuous 6-0 Proline suture. A microvascular clamp is placed on the vein graft after it has been filled with heparin and after the proper alignment and length for the renal artery anastomosis has been determined. The clamps are removed from the hepatic circulation, and a small Schwartz microvascular clamp is placed on the distal renal artery. The vein graft is anastomosed to the right renal artery in an end-to-end fashion. When the gastroduodenal artery is used, it is divided, and an end-to-end anastomosis between the gastroduodenal artery and the renal artery is accomplished.

FIG. 10-14. Use of the gastroduodenal artery to perform hepatorenal revascularization through direct end-to-end anastomosis with the right renal artery. (From Novick AC, Streem SB, Pontes JE, eds. Stewart's operative urology. Baltimore: Williams & Wilkins, 1989.)

We have employed this procedure in approximately 50 patients with good results. Postoperative angiography has demonstrated the absence of a renal–hepatic steal syndrome. Liver function has not been compromised in any of our patients to date. We no longer advocate the use of the gastroduodenal artery in adult patients, but it is a perfectly acceptable bypass procedure in the pediatric patients. We have also utilized the superior mesenteric-to-renal artery saphenous vein bypass as a “bailout procedure” as well with good results ( Fig. 10-15). An iliac-to-renal bypass graft has been done as an alternative to the aortorenal bypass procedure in ten of our patients, with favorable results ( Fig. 10-16).

FIG. 10-15. Hepatorenal bypass performed with an interposition saphenous vein graft anastomosed end to side to the common hepatic artery and end to end to the right renal artery. (From Novick AC. Diminished operative risk and improved results following revascularization for atherosclerotic renal artery disease. Urol Clin North Am 1984;11:435.)

FIG. 10-16. Iliorenal bypass with a saphenous vein graft anastomosed end to side to the common iliac artery and end to end to the renal artery. (From Novick AC, Streem SB, Pontes JE, eds. Stewart's operative urology. Baltimore: Williams & Wilkins, 1989.)

Renal Autotransplantation and Ex Vivo Bench Surgery On rare occasions, kidneys with lesions of the renal artery or its branches are not amenable to in situ reconstruction. In these circumstances, temporary removal of the kidney, ex vivo preservation, microvascular repair (bench surgery), and autotransplantation may permit salvage. Autotransplantation developed as an outgrowth of the technique in renal transplantation. The simultaneous development of an apparatus that could preserve kidneys extracorporeally for long periods of time and of preservation solutions also led to the technique of extracorporeal renal repair. Autotransplantation and ex vivo repair should be considered in patients with traumatic arterial injuries, when disease of the major vessels extends beyond the bifurcation of the main renal artery into the segmental branches, and when multiple vessels supplying the affected kidney are involved. Bench surgery may also be required in patients who have very large aneurysms, arteriovenous fistulas, or malformations ( Fig. 10-17).

FIG. 10-17. (A) An arteriogram shows complex involvement of the right renal artery by disease extending into the primary branches. (B) A postoperative arteriogram shows patent anastomoses.

Other indications for autotransplantation that usually do not require ex vivo repair include abdominal aortic aneurysms that involve the origin of the renal arteries and extensive atheromatous aortic disease, when an operation on the aorta itself may prove hazardous. In the last case, the patients usually have extensive internal iliac artery disease that precludes utilization of this artery for autotransplantation. However, we have noted that in these instances, the external iliac artery is spared extensive atherosclerosis and is suitable for autotransplantation, with an end-to-side renal artery anastomosis or an iliac-to-renal bypass graft. Techniques for Autotransplantation Autotransplantation can be accomplished through a large single midline incision or two separate flank and iliac fossa incisions. When the kidney is removed, care is taken to preserve the maximum length of renal vessels and ureter. If the transabdominal approach is selected, ureteral continuity can be retained, necessitating only vascular anastomosis afer the kidney is flipped over. If ex vivo surgery requires transection of the ureter, ureteroneocystostomy is necessary in addition to vascular anastomosis. When ureteral continuity is preserved, autotransplantation is performed as illustrated in Fig. 10-18. When the kidney has been excised completely, the standard techniques for renal homotransplantation are used.

FIG. 10-18. Ipsilateral autotransplantation of the kidney with end-to-side anastomosis of renal vein to common iliac vein and end-to-end anastomosis of hypogastric artery to renal artery. Ureter is taking a redundant course to bladder, and kidney is placed in upside-down position.

During dissection of the iliac vessels, meticulous care is taken to ligate the lymphatics in this area to prevent the development of a lymphocele. The external iliac vein is freed to the point where it is crossed by the internal iliac artery ( Fig. 10-19A). The renal vein is anastomosed end-to-side to the external iliac vein using 5-0 Proline sutures (Fig. 10-19B). If the renal artery is free of atherosclerotic disease, it is then anastomosed end-to-end to the internal iliac artery, employing 6-0 Proline sutures (Fig. 10-19C,D). If the internal iliac artery is diseased, the renal artery is anastomosed end-to-side to the external iliac artery.

FIG. 10-19. (A) Dissection of lymphatic and areolar tissue from iliac vessels. (B) End-to-side anastomosis of renal vein to external iliac vein. (C) End-to-end anastomosis of renal artery to internal iliac artery.

When the ureter requires reimplantation, we prefer a modification of the Politano–Leadbetter ureteroneocystostomy. Saline solution, 2 to 3 ml, is injected submucosally, raising a mucosal bleb (Fig. 10-20A). A small segment of mucosa is removed from the inferior portion of the bleb ( Fig. 10-20B). A right-angle clamp is inserted into this opening, and a 3-cm-long submucosal tunnel is created ( Fig. 10-21A). At the apex of the tunnel, the right-angle clamp is rotated 180 degrees to pierce the detrusor muscle. The ureter is brought to lie in the submucosal tunnel ( Fig. 10-21B). The distal ureter is cut at a 45-degree angle, and the ureter is anastomosed to the bladder with interrupted 4-0 or 5-0 Dexon or Vicryl sutures ( Fig. 10-21C,D).

FIG. 10-20. (A) Injection of saline submucosally to dissect mucosa from muscularis before reimplantation. (B) Creation of tunnel.

FIG. 10-21. (A) Ureteroneocystostomy: opening in posterolateral bladder wall for ureter. (B) Submucosal tunnel for ureteroneocystostomy. (C) Ureter in position in submucosal tunnel. (D) Elliptic anastomosis of ureter to bladder wall.

In the future, the use of endovascular prostheses in maintaining the effect of luminal balloon dilatation is a very promising technique if long-term evaluation confirms the preliminary results.

OUTCOMES
Complications Complications of renal vascular surgery can be classified as early or delayed. Early complications include bleeding, thrombosis of the artery, embolization of the branch vessels, subintimal dissection, false aneurysms, and loss of the kidney. Postoperative bleeding requiring operative intervention is to some extent a technical failure but may also be a function of the structural integrity of the arterial wall in diseased segments. It is important to recognize the enlarged perihilar vessels that are seen in high-grade stenosis and to be cognizant of the adrenal venous channels. Delayed bleeding may be the result of false aneurysm formation or erosion of the graft anastomosis into the duodenum or other bowel. Renal artery thrombosis is the most common complication of renal vascular reconstruction and is most common after either placement of a dacron graft or endarterectomy. Predisposing factors include small dacron grafts, renal atrophy associated with thin-walled diseased arteries and high intrarenal vascular resistance, hypotension, or hypovolemia. It has been shown that the thrombosis of both venous and synthetic grafts is partially affected by the adequacy of the peripheral runoff as well as the adequacy of resection of the endothelial plaques in atheroschlerotic vessels. Embolization of plaque to the distal extremities or aortic thrombosis is rare. Results Balloon angioplasty is primarily used to treat patients with mural dysplasia. 2,10 This modality of treatment is 80% to 85% effective in the management of these patients at our institution. Balloon angioplasty has a very limited role in the management of atherosclerotic renovascular disease at our institution. The combination of balloon angioplasty for the younger, healthier patients suffering from mural dysplasia, the advent of alternative bypass procedures, and the concept of revascularization for preservation and restoration of renal function have dramatically changed the nature of the patient population being referred to our institution for renal revascularization. We are now frequently being called on to revascularize more elderly, higher-risk patients with diffuse atherosclerosis who have failed aggressive antihypertensive therapy in order to improve their renal function. We have recently reported a series of more than 100 patients who have undergone renal revascularization for preservation and restoration of renal function with an 85% success rate in this very high-risk patient population. 5

CONCLUSION
Renovascular disease is a rapidly changing clinical entity. Considerable progress is being made in screening and diagnosis, primarily as a result of the development of less- or noninvasive studies. Renovascular disease is a potentially curable cause of hypertension and one of the few curable or preventable causes of renal failure. The indications for angioplasty and surgical revascularization are better known, and interventional therapy is justified when anesthesia and surgery represent an acceptable risk. We look forward to the short- and long-term results of renal artery atherectomy and wall stenting with anticipation. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Distler A, Spies KP. Diagnostic procedures in renovascular hypertension. Clin Nephrol 1991;36(4):174–180. Englund R, Brown MA. Renal angioplasty for renovascular disease. J Cardiovasc Surg 1991;32:76–80. Ferdinandi A, Pavlica P, Lupattelli L, et al. Duplex sonography and color Doppler evaluation of renal artery stenosis—angiographic correlation. Scand J Urol Nephrol [Suppl] 1991;137:67–72. Libertino JA. Surgery for renovascular hypertension. In: Walsh PC, Retik AB, Stamey TA, Vaughan ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;2521–2551. Libertino JA, Bosco PJ, Ying C, et al. Renal revascularization to restore and preserve renal function. J Urol 1992;147:1485–1487. Novick AC. Management of renovascular disease: a surgical perspective. Circulation 1991;83(Suppl 2):1167–1171. Pickering TG. Diagnosis and evaluation of renovascular hypertension: indications for therapy. Circulation 1991;83(Suppl 2):1147–1154. Postma CT, Dernout HA, Van Oljen MD, et al. The value of tests predicting renovascular hypertension in patients with renal artery stenosis treated by angioplasty. Arch Intern Med 1991;151:1531–1535. Setaro JF, Saddler MC, Chen CC, et al. Simplified captopril renography in diagnosis and treatment of renal artery stenosis. Hypertension 1991;18:289–298. Tegtmeyer CJ, Selby JB, Hartwell GD, et al. Results and complications of angioplasty in fibromuscular disease. Circulation 1991;83(Suppl 2):1155–1161. Tollefson DF, Ernst CB. Natural history of atherosclerotic renal artery stenosis associated with aortic disease. J Vasc Surg 1991;14:327–331. Weaver F, Pentecost MJ, Yellin AE, et al. Clinical applications of CO 2 digital subtraction angiography. J Vasc Surg 1991;13:266–273.

Chapter 11 Anatrophic Nephrolithotomy Glenn’s Urologic Surgery

Chapter 11 Anatrophic Nephrolithotomy
Michael L. Paik and Martin I. Resnick

M. L. Paik and M. I. Resnick: Department of Urology, University Hospitals of Cleveland/Case Western Reserve University School of Medicine, Cleveland, Ohio 44106.

Diagnosis Indications for Surgery Alternative Treatments Surgical Technique Outcomes Complications Results Chapter References

Anatrophic nephrolithotomy is a procedure that has been used by urologists for nearly 30 years in the removal of large renal calculi, specifically branched or staghorn calculi. These stones are often associated with urinary tract infections, and the coexistence of these two conditions makes it difficult to eradicate either. Definitive treatment of these stones is generally advocated because of the significant morbidity and mortality associated with untreated staghorn calculi. Blandy and Singh found that patient survival is reduced with untreated staghorn calculi, with a mortality rate of 28% at 10 years. 3 Since the early 1980s, with the development of less invasive approaches such as extracorporeal shock wave lithotripsy and percutaneous nephrolithotomy, the role of anatrophic nephrolithotomy and other open stone operations has certainly diminished. 2 However, anatrophic nephrolithotomy remains the gold standard for the treatment of staghorn calculi and thus maintains a role in the treatment of these large complex stones. The original description of anatrophic nephrolithotomy was by Smith and Boyce in 1968. 9 The operation they described was based on the principle of placing the nephrotomy incision through a plane of the kidney that was relatively avascular. This approach would avoid damage to the renal vasculature with resulting atrophy of the renal parenchyma, hence the term anatrophic. The operation also involves reconstruction of the intrarenal collection system to eliminate anatomic obstruction, thus improving urinary drainage, reducing the likelihood of urinary tract infection, and preventing recurrent stone formation.

DIAGNOSIS
The diagnosis of staghorn calculi is usually established in a similar fashion as other forms of urolithiasis. Patients may have the typical symptoms of flank pain, fever, and hematuria, or they may be asymptomatic. The diagnosis of chronic urinary tract infection is common in patients with these types of stones. Urine culture is often positive, and typical organisms include urea-splitting organisms such as Proteus, Klebsiella, Providencia, and Pseudomonas. Useful radiographic studies traditionally include plain abdominal radiographs, nephrotomograms, and excretory urograms to identify the stones, the collecting system, and, if present, to define the degree of obstruction. Computed tomography can be helpful for detection of radiolucent or poorly calcified stones. Retrograde pyelography is usually performed in cases of equivocal findings on excretory urography. Nuclear renal scans can help to determine differential renal function when such information might affect the surgical approach. Renal arteriography is usually not indicated unless there is suspicion of anomalous arterial anatomy such as in renal fusion anomalies. Before elective surgery, a metabolic evaluation is recommended to attempt to determine an etiology for stone formation and to aid in preventing a recurrence. For instance, it is important to determine the presence of hypercalciuria, hyperuricosuria, hyperoxaluria, cystinuria, hyperparathyroidism, and renal tubular acidosis. The measurement of serum and urine calcium, phosphorus, creatinine, uric acid, and electrolytes should be routine. A 24-hour urine collection for creatinine clearance as well as urinary calcium, phosphorus, oxalate, citrate, cystine, and uric acid is also an integral part of the workup.

INDICATIONS FOR SURGERY
Anatrophic nephrolithotomy should be performed for the removal of branched or staghorn calculi, usually complete staghorn stones, or those associated with infundibular stenosis or other intrarenal anatomic obstruction, for the combined goals of removing all calculi and open surgical correction of the anatomical obstruction. This procedure may also be preferred in the treatment of a staghorn calculus in a kidney with a small intrarenal pelvis, making access to the renal pelvis difficult, or in a patient who has undergone prior renal surgery to avoid a more risky renal sinus dissection. This operation is also indicated for the treatment of staghorn calculi in patients who would benefit from or prefer a single therapeutic procedure versus multiple, less invasive, procedures such as extracorporeal shock wave lithotripsy and/or percutaneous nephrolithotomy. The goals of the procedure should be to remove all calculi and fragments, to improve urinary drainage of any obstructed intrarenal collecting system, to eradicate infection, to preserve and improve renal function, and to prevent stone recurrence. 10

ALTERNATIVE TREATMENTS
Most staghorn calculi can now be preferentially treated with percutaneous nephrolithotomy, with or without extracorporeal shock wave lithotripsy. The stone-free rates reported are approaching comparability with traditional anatrophic nephrolithotomy, and there is probably an advantage to be gained in shorter convalescent periods following the less invasive methods. These alternative odalities can sometimes require multiple different procedures to accomplish a stone-free state. The American Urological Association Nephrolithiasis Clinical Guidelines Panel recommended as a guideline that initial percutaneous nephrolithotomy followed by extracorporeal shock-wave lithotripsy and/or further percutaneous procedures should be the treatment for most standard patients with staghorn calculi. Open surgery is recommended as an appropriate option in unusual cases when a stone is not expected to be removed with a reasonable number of the less-invasive procedures. 8 Despite impressive advances with the less-invasive techniques, anatrophic nephrolithotomy remains a treatment option for large complete staghorn calculi or staghorn stones associated with anatomic obstruction and requiring open surgical correction.

SURGICAL TECHNIQUE
After administration of general anesthesia and placement of a Foley catheter, the patient is placed in the standard flank position with elevation of the kidney rest and flexion of the operating table to achieve adequate spacing between the lower costal margin and the iliac crest. Three-inch-wide adhesive tape applied at the shoulders and hips can be used to secure the patient to the table. Adequate padding should be used to protect pressure points. A standard flank approach is used. The incision can be placed through the bed of either the 11th or 12th rib, depending on the estimated position of the kidney. If a previous flank incision has been made for renal surgery, it is preferable to place the incision above the old scar, ensuring that access to the kidney can be achieved through unscarred tissue. After rib resection, when access has been gained into the retroperitoneal space, Gerota's fascia is identified overlying the kidney. Gerota's fascia is incised in a cephalad–caudal direction, which facilitates returning the kidney to its fatty pouch at the end of the operation. The kidney is then fully mobilized, and the perinephric fat is carefully dissected off the renal capsule with care taken not to disrupt the renal capsule. Should the capsule become inadvertently incised, it can be closed at that time with chromic catgut sutures. The kidney is now free to be suspended in the operative field by utilizing two 1-inch umbilical tapes as slings. At this point a preliminary portable plain radiograph can be obtained. The renal hilar dissection is the next step. The main renal artery and its branches are carefully dissected and identified ( Fig. 11-1A). The avascular plane, or Brodel's line, can be identified by temporarily clamping the posterior segmental artery and injecting 20 ml of methylene blue intravenously, which results in the blanching of the posterior renal segment while the anterior portion turns blue 5 (Fig. 11-1B). This allows identification of this avascular plane. Placing the nephrotomy incision through this plane will achieve maximal renal parenchymal preservation and minimize blood loss. The avascular plane can also be identified with the use of a Doppler

stethoscope to localize the area of the kidney with minimal blood flow.

FIG. 11-1. Anatrophic nephrolithotomy. (A) Main renal artery and branches are isolated. (B) The posterior segmental artery is occluded, and methylene blue is administered intravenously. The resulting demarcation between pale ischemic and bluish perfused parenchyma defines a relatively avascular nephrotomy plane.

Extensive renal hilar dissection can be avoided by utilizing a modification of the original procedure described by Smith and Boyce. Redman and associates relied on the relatively constant segmental renal vascular supply in advocating placing the incision at the expected location of the avascular line after clamping the renal pedicle with a Satinsky clamp, seeking to prevent vasospasm of the renal artery and warm ischemia. 6 This modification can be time-saving and spare extensive dissection of the renal hilum. However, we continue to advocate precise identification of the avascular plane to minimize parenchymal loss. At this point, 25 g of intravenous mannitol is administered. This promotes a postischemic diuresis and prevents the formation of intratubular ice crystals by increasing the osmolarity of the glomerular filtrate. The main renal artery can now be occluded with a noncrushing bulldog vascular clamp ( Fig. 11-2). A bowel bag or barrier drape is placed around the kidney, and hypothermia is initiated with the placement of iced saline slush within the barrier surrounding the kidney. Dry laparotomy sponges are used to pack away the peritoneal contents and to protect and insulate them from hypothermia. The kidney should be cooled for 10 to 20 minutes before the nephrotomy incision is made. This should allow achievement of a core temperature in the 5° to 20°C range, which will allow safe ischemic times from 60 to 75 minutes and minimize renal parenchymal damage. 4

FIG. 11-2. The main renal artery is clamped and a bowel bag or rubber dam is placed around the kidney. Dry gauze packs are placed anterior to the kidney to protect the intra-abdominal organs from hypothermia.

The renal capsule is then incised sharply over the previously identified avascular plane, and the renal parenchyma can be bluntly dissected with the back of the scalpel handle (Fig. 11-3). Blunt dissection minimizes injury to the intrarenal arteries that are traversed. Small bleeding vessels can be controlled with 5-0 or 6-0 chromic catgut suture ligature. If renal back bleeding continues to be a problem despite these measures, the main renal vein can be occluded. As the incision is angled toward the midportion of the renal hilum, the nephrotomy should theoretically remain close to the avascular plane ( Fig. 11-4).

FIG. 11-3. A superficial incision is made in the renal capsule through the avascular plane.

FIG. 11-4. The parenchyma is bluntly dissected with the back of a scalpel handle. The incision closely approximates the avascular plane.

As the nephrotomy incision proceeds toward the renal hilum, the ideal location to enter the collecting system is at the base of the posterior infundibula. Occasionally, with large posterior calyceal calculi, a dilated posterior calyx will be entered initially. The remainder of the collecting system can then be identified and opened with a probe or stone forceps. If a posterior infundibulum is entered first, the incision is then carried toward the renal pelvis ( Fig. 11-5). The stone is palpated, and the

remainder of the infundibula are incised in a similar fashion. Attention is then turned towards the anterior infundibula. All of the calyceal extensions should be identified and incised. In order to minimize stone fragment formation and retained calculi, the stone should not be manipulated or removed until all of the calyceal and infundibular extensions are appropriately identified and incised, allowing for complete visualization and mobilization of the collecting system and calculi. Ideally, the stone or stones should be removed without fragmentation; however, often it is inevitable that there will be some piecemeal extraction ( Fig. 11-6). After removal of all stone fragments, the renal pelvis and calyces are copiously irrigated with cold saline and carefully inspected for retained fragments. A plain radiograph is obtained at this point to rule out residual calculi or fragments.

FIG. 11-5. The collecting system is carefully incised.

FIG. 11-6. After the collecting system is opened, calculi are extracted and total removal is confirmed radiographically.

At this time, a “double-J” stent is passed from the renal pelvis into the bladder ( Fig. 11-7). The routine use of internal ureteral catheters is encouraged. They provide good urinary drainage and protect the freshly reapproximated collecting system and minimize postoperative urinary extravasation. The stents also prevent intraoperative migration of smaller calculi into the ureter.

FIG. 11-7. A Silastic stent is passed in an antegrade fashion from the pelvis to the bladder. Traction sutures are placed to mark the walls of adjacent calyces before suturing them together with a running 6-0 chromic suture.

The next step in the procedure is the reconstruction of the intrarenal collecting system. Infundibular stenosis or stricture, which results in obstruction promoting urinary stasis and recurrent stone formation, should be corrected with calyorrhaphy or calycoplasty. The former is the repair of a single narrowed calyx, achieved by incising the calyx along its appropriate margin (anterior margin for posterior calyces and posterior margin for anterior calyces) and suturing those margins to the renal pelvis, resulting in a shorter, wider calyx ( Fig. 11-8). The infundibulum can also be incised longitudinally and then closed transversely in a Heinecke–Mickulicz fashion. Calycoplasty is the repair of adjacent stenotic calyces by suturing the adjacent walls of the neighboring calyces, thus forming a single structure ( Fig. 11-9 and Fig. 11-10). All intrarenal reconstructive suturing should be accomplished with 5-0 or 6-0 chromic catgut sutures. When suturing the mucosal edges, it is important to avoid incorporation of underlying interlobular arteries, thus preventing ischemia.

FIG. 11-8. Technique of repairing strictured infundibula. (1) Narrowed elongated infundibulum. (2) Incision into calyx forms an inverted Y. (3) Pelvic flap is advanced into infundibulotomy. (4) Incision in calyx is closed transversely.

FIG. 11-9. Adjacent infundibula are sutured together starting in the renal pelvis. Peripelvic fat is depressed during this closure.

FIG. 11-10. The collecting system is completely reconstructed.

The renal pelvis is then closed with a running 6-0 chromic catgut suture ( Fig. 11-11). The renal capsule is closed with a running 4-0 chromic suture ( Fig. 11-12). The use of mattress-type parenchymal sutures can lead to tissue ischemia and should be avoided if possible. One should inspect closely for further parenchymal bleeding points and ensure good hemostasis before closing the renal capsule. After the capsule is closed and hemostasis has been achieved, the slush surrounding the kidney is removed, and the renal artery unclamped. The kidney is observed for good hemostasis and return of pink color and good turgor after unclamping. The kidney is then returned into Gerota's fascia, and the kidney and proximal ureter are covered with some perirenal fat to minimize the postoperative scar formation. If Gerota's fascia is unavailable because of prior surgery, omentum can be mobilized through a peritoneal opening and wrapped around these structures. The peritoneal opening should be sutured to the omentum to prevent herniation of the abdominal viscera.

FIG. 11-11. The renal pelvis is closed with a running 6-0 chromic suture.

FIG. 11-12. The renal capsule is closed with a running 4-0 chromic suture.

A Penrose or suction-type drain is placed within Gerota's fascia and brought out through a separate stab incision. This drain is left in place until minimal drainage occurs, usually by the third or fourth postoperative day. Nephrostomy tubes are generally avoided because of their potential for causing infection or further renal damage. The flank musculature and skin are closed in the standard fashion. Postoperative management after anatrophic nephrolithotomy should follow the same principles that guide management after other major operations. Intravenous fluids are maintained to achieve brisk urine output and until the patient is able to tolerate a clear liquid diet. Broad-spectrum intravenous antibiotics are administered perioperatively and continued postoperatively. Antibiotic coverage is guided by preoperative urine culture and sensitivity results. The patient is usually converted to appropriate oral antibiotics and maintained on a 14-day course. The ureteral stent is removed cystoscopically at approximately 6 weeks after the operation in uncomplicated cases.

OUTCOMES
Complications Pulmonary complications are perhaps the most common following anatrophic nephrolithotomy, especially atelectasis. Patients with a history of pulmonary disease should probably undergo preoperative evaluation with pulmonary function testing and initiation of vigorous pulmonary toilet prior to surgery. Postoperatively, patients should be encouraged to breathe deeply, and use of an incentive spirometer should be routine. Early ambulation will also be beneficial.

Pneumothorax should occur in fewer than 5% of patients. 10 Inadvertent opening of the pleura, usually during incision and resection of a rib, should be readily identified intraoperatively. The defect should be closed immediately with a running chromic catgut suture. The lung is hyperinflated just before the final suture is placed to ensure reexpansion of the lung. Chest tubes are not routinely used but may be necessary if any question remains regarding the reliability of the pleural closure. A chest radiograph should be obtained in the recovery room for any patient who undergoes repair of a pleural defect. Pulmonary embolism remains a potential complication of any major surgery. Routine use of elastic support hose and sequential-compression stockings can lower the risk of deep venous thrombosis. Encouragement of early ambulation is also an important preventative measure. Significant postoperative renal hemorrhage should occur in fewer than 10% of patients. Assimos and associates reported an incidence of 6.4%. 1 Bleeding usually occurs immediately or about a week postoperatively. Extensive intrarenal reconstruction, older age, worse renal function, and presence of blood dyscrasias were found to be significant risk factors. Slow bleeding will usually resolve on its own; management includes correction of any bleeding abnormalities and replacement with blood products as necessary. Oral e-aminocaproic acid can be successful in certain cases. Bleeding that is brisk or cannot be adequately treated conservatively will require a more aggressive approach. A renal arteriogram can help identify the lesion, and an attempt at arteriographic embolization can be considered. Reexploration may be required in the remainder of the cases, with reinstitution of hypothermia and suture ligation of the bleeding vessel(s). Persistent hematuria 1 to 4 weeks postoperatively should alert the clinician to the possibility of renal arteriovenous fistula formation. 1 Stone recurrence rates following anatrophic nephrolithotomy have been reported from 5% to 30%. 10 Inspection, intraoperative plain radiographs, intraoperative ultrasound, and nephroscopy can all aid in the identification and treatment of retained calculi. Recurrent calculi usually form in those with persistent urinary tract infections, persistent urinary drainage impairment, and those with previously unidentified or refractory metabolic disturbances. 7 Urinary drainage or extravasation should occur infrequently with the routine use of perinephric drains and internal ureteral catheter drainage. Should drainage recur or persist following removal of the drain and/or ureteral stent, replacement of the ureteral stent should be considered to decompress the system and relieve any obstruction. Results When performed for appropriate indications and with meticulous technique, anatrophic nephrolithotomy can achieve successful removal of all calculi, preservation of renal function, improved urinary drainage, and eradication of infection. Stone-free rates greater than 90% should be achieved. We believe that for large complex staghorn calculi and those associated with some anatomic abnormality leading to impaired urinary drainage, anatrophic nephrolithotomy remains superior to percutaneous nephrolithotomy or combination therapy with respect to both stone-free rates and the achievement of a stone-free state with a single operative procedure. In the long term, treatment of these staghorn calculi with anatrophic nephrolithotomy should preserve renal function in the involved kidney and, in a majority of patients, eradicate stone disease and chronic urinary infection. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Assimos DG, Boyce WH, Harrison LH, Hall JA, McCullough DL. Postoperative anatrophic nephrolithotomy bleeding. J Urol 1986;135:1153–1156. Assimos DG, Boyce WH, Harrison LH, McCullough DL, Kroovand RL, Sweat KR. The role of open stone surgery since extracorporeal shock wave lithotripsy. J Urol 1989;142:263–267. Blandy JP, Singh M. The case for a more aggressive approach to staghorn stones. J Urol 1976;115:505–506. McDougal WS. Renal perfusion/reperfusion injuries. J Urol 1988;140:1325–1330. Myers RP. Brodel's line. Surg Gynecol Obstet 1971;132:424–426. Redman JF, Bissada NK, Harper DL. Anatrophic nephrolithotomy: experience with a simplification of the Smith and Boyce technique. J Urol 1979;122:595–597. Russell JM, Harrison LH, Boyce WH. Recurrent urolithiasis following anatrophic nephrolithotomy. J Urol 1981;125:471–474. Segura JW, Preminger GM, Assimos DG, et al. Nephrolithiasis Clinical Guidelines Panel summary report on the management of staghorn calculi. J Urol 1994;151:1648–1651. Smith MJV, Boyce WH. Anatrophic nephrotomy and plastic calyrhaphy. J Urol 1968;99:521–527. Spirnak JP, Resnick MI. Anatrophic nephrolithotomy. Urol Clin North Am 1983;10(4):665–675.

Chapter 12 Renal and Retroperitoneal Abscesses Glenn’s Urologic Surgery

Chapter 12 Renal and Retroperitoneal Abscesses
J. Quentin Clemens and Anthony J. Schaeffer

J. Q. Clemens and A. J. Schaeffer: Department of Urology, Northwestern University Medical School, Chicago, Illinois 60611-3008.

Classification Pathogenesis Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Percutaneous Drainage Open Surgical Drainage Ancillary Procedures Subcapsular Nephrectomy Outcomes Complications Results Special Considerations Renal Tuberculosis Renal Echinococcosis Chapter References

Renal and retroperitoneal abscesses are uncommon clinical entities that often pose a significant diagnostic challenge. Nonspecific signs and symptoms frequently lead to a delay in diagnosis and treatment. Consequently, they are associated with significant morbidity, and mortality rates approaching 50% have been reported. An understanding of the anatomy of the retroperitoneal space is essential for classification, diagnosis, and management of renal and retroperitoneal abscesses.

CLASSIFICATION
The retroperitoneal space is bounded by the posterior parietal peritoneum and transversalis fascia ( Fig. 12-1 and Fig. 12-2). It is divided into the perirenal space and the pararenal space.

FIG. 12-1. Right sagittal view showing the anterior pararenal, perirenal, posterior pararenal and retrofascial spaces. (From Simons GW, Sty JR, Starshak RJ. Retroperitoneal and retrofascial abscess. J Bone Joint Surg 1983;65A:1041.)

FIG. 12-2. The three retroperitoneal compartments. The striped and crosshatched areas correspond to the perirenal and posterior pararenal space, respectively. (From Meyers MA. Dynamic radiology of the abdomen. In: Normal and pathologic anatomy, 2nd ed. New York: Springer-Verlag, 1982;107–110.)

The perirenal space surrounds the kidney and is bounded by the renal (Gerota's) fascia. It contains a lemon-yellow layer of fat, which is thickest posteriorly and laterally. The anterior and posterior leaves of the renal fascia fuse above the adrenal gland, becoming continuous with the diaphragmatic fascia. 1 A thinner, more variable layer meets between the adrenal gland and the kidney. Laterally, the fascial layers join to form the lateroconal fascia, which becomes continuous with the posterior parietal peritoneum. Medially, the posterior layer fuses with the psoas muscle fascia, and the anterior layer fuses with the connective tissue surrounding the great vessels and organs of the anterior retroperitoneum (i.e., the pancreas, duodenum, and colon). Because the perirenal space rarely crosses the midline, perirenal abscesses usually remain unilateral. 16 Inferiorly, the renal fascial layers do not fuse but, rather, become continuous with the psoas and ureteral coverings. 1,11 This opening inferiorly allows spread of perirenal infections to the pararenal space, to the pelvis, to the psoas muscle, and, in some cases, to the contralateral retroperitoneum. The pararenal space is divided into two compartments: the anterior compartment, which is bounded by the posterior parietal peritoneum and the anterior renal fascia; and the posterior compartment, which is bounded by the posterior renal fascia and transversalis fascia. The pararenal space contains pale adipose tissue, which fills much of the retroperitoneal space. Because the anterior pararenal space extends across the midline, infection arising in one space may become bilateral. The posterior pararenal space does not cross the midline, and infection within it remains unilateral. 16 The retrofascial compartment lies posterior to the transversalis fascia. It is important only in development of the rare retrofascial abscess from abscesses of the psoas, iliacus, and quadratus muscles.

PATHOGENESIS
Before the advent of antimicrobial therapy, most renal abscesses occurred as a result of hematogenous spread of gram-positive organisms, usually Staphylococcus aureus. These abscesses, which were called renal carbuncles, may still be seen in intravenous drug users and in patients with dermatologic disorders. They may

resolve with aggressive antimicrobial therapy if treated before frank suppuration. Presently, most renal and retroperitoneal abscesses are caused by retrograde ascent of gram-negative bacteria from the bladder. The most common organisms include Escherichia coli, Proteus, Klebsiella, and Pseudomonas.6,12 Anaerobes may be isolated in abscesses associated with gastrointestinal and respiratory infections. 2,4 Abscesses caused by opportunistic organisms such as Candida and Aspergillus may occur in immunosuppressed patients. Other uncommon pathogens include Mycobacterium tuberculosis and Echinococcus (see below). A renal abscess is generally preceded by pyelonephritis, which progresses to abscess formation in the presence of a virulent uropathogen, a damaged or obstructed urinary tract, or a compromised host. Renal abscesses have a predilection for the cortical medullary region and may drain spontaneously through the renal collecting system. When renal infection is complicated by obstruction, a purulent exudate collects in the renal collecting system. Pyonephrosis refers to infected hydronephrosis with suppurative destruction of the parenchyma of the kidney, with total or near total loss of renal function. The most frequent cause of obstruction is calculous disease. 3,20 A previous history of urinary tract infection or surgery is also common. Perirenal abscesses usually occur by erosion of abscesses or pyonephrosis into the perirenal space. 14 Because of gravity, the resulting perirenal suppuration tends to localize dorsolaterally to the lower pole of the kidney. Posterior pararenal abscesses may arise from perirenal abscesses or from anterior pararenal abscesses tracking into the pelvis, where the anterior and posterior pararenal spaces communicate. Occasionally they result from hematogenous spread. Anterior pararenal abscesses are rarely urologic in origin. They arise from infection involving the organs within the anterior pararenal space, namely the ascending and descending colon, appendix, duodenal loop, and the pancreas. Abscesses arising from the gastrointestinal tract usually harbor a mixture of microorganisms, of which E. coli is the most prevalent. Extension of anterior pararenal abscesses into the perirenal space is uncommon.

DIAGNOSIS
The diagnosis of renal and retroperitoneal abscesses requires a high index of suspicion, as they typically present with insidious, nonspecific signs and symptoms. 6,17 Presenting symptoms may include fever, chills, abdominal or flank pain, irritative voiding symptoms, nausea, vomiting, lethargy, or weight loss. Symptoms have been present for more than 5 days in the majority of patients with renal and retroperitoneal abscesses, compared with 10% of patients with pyelonephritis. Over one-third of patients may be afebrile. The majority of patients diagnosed with renal and retroperitoneal abscesses have underlying, predisposing medical conditions. These include diabetes mellitus, urinary tract calculi, previous urologic surgery, urinary tract obstruction, polycystic kidney disease, and immunosuppression. A palpable flank or abdominal mass is present in about half of the cases. The mass may be better appreciated by examination of the patient in the knee–chest position. There may also be signs of psoas muscle irritation with flexion of the thigh. Laboratory tests are helpful but nondiagnostic. Leukocytosis, elevated serum creatinine, and pyuria are common. Blood and urine cultures are frequently negative; when positive, they usually correlate with culture results from the abscess. Excretory urography may aid in the diagnosis of renal or retroperitoneal abscesses by showing diminished mobility on inspiratory–expiratory films. A renal abscess causes a decrease in function and enlargement of the nephrogram during the acute phase. Retroperitoneal abscesses may cause displacement of the kidneys or ureters by a mass, scoliosis of the spine, and free air or fluid in the retroperitoneal space. Computed tomography (CT) is highly sensitive for the diagnosis of renal and retroperitoneal abscesses. It precisely localizes and assesses the size of an abscess so that the type of intervention and its anatomic approach can be determined. The presence of gas within a lesion is pathognomonic for an abscess. Additional CT findings characteristic of an abscess include a mass with low attenuation, rim enhancement of the abscess wall after contrast, obliteration of tissue planes, and displacement of surrounding structures. Ultrasonography is less sensitive than CT but useful for monitoring response to therapy. Arteriography and radioisotope scanning rarely add significant information.

INDICATIONS FOR SURGERY
Renal and retroperitoneal abscesses are generally lethal if untreated. Therapeutic options include antimicrobial therapy, percutaneous catheter drainage, and open surgical drainage.

ALTERNATIVE THERAPY
Antimicrobial therapy as the sole treatment is an option, yet most abscesses cannot be cured without drainage. Small renal abscesses may resolve, however, if they are treated early with aggressive antimicrobial therapy. Prolonged antimicrobial therapy without drainage is indicated only if favorable clinical response and radiologic confirmation of abscess resolution indicate that the therapy is effective. If antimicrobial therapy is not effective, prompt percutaneous or open surgical drainage of the pus is mandatory. Progression of a renal abscess leads to perinephric abscess or perforation into the collecting system and results in signs and symptoms of urinary tract infection. Antimicrobial therapy should be instituted after the urine has been Gram-stained and urine and blood cultures have been obtained. Broad-spectrum coverage should be guided by the presumptive diagnosis and the presumed pathogen. An aminoglycoside for gram-negative rods and ampicillin for gram-positive cocci are preferred. Anaerobic coverage with a drug such as clindamycin is warranted when Gram stain reveals a polymicrobial flora or when a gastrointestinal source is suspected. If the abscess may be of staphylococcal origin, a penicillinase-resistant penicillin, such as nafcillin, should be added. Antimicrobial therapy should be reevaluated when the results of culture and sensitivity tests are available. Unfortunately, urine and blood cultures are frequently sterile, and empirical therapy must be modified on the basis of clinical response and changes in imaging studies.

SURGICAL TECHNIQUE
Percutaneous Drainage Most renal and retroperitoneal abscesses are treated with empirical antimicrobial therapy and immediate percutaneous drainage. When successful, minimally invasive therapy minimizes operative morbidity and allows for preservation of renal tissue. The abscess must be confirmed by CT-guided or ultrasonography-guided needle aspiration and must be drainable without injury to other organs. Immediate surgical drainage must be instituted if the procedure fails. After a multiport drainage catheter (8 to 12 Fr) is positioned, the abscess should be drained, and adequate evacuation should be confirmed by CT or ultrasonography. The catheter should then be connected to low intermittent suction, and drainage outputs should be monitored daily. If drainage stops abruptly, occlusion of the catheter should be suspected, and it should be irrigated gently with small amounts of normal saline. Computed tomography or ultrasonography should be performed periodically to monitor catheter position and size of the abscess. Direct instillation of contrast through the drainage tube may be helpful to confirm the catheter position or to rule out a fistula. To avoid bacteremia, prophylactic antimicrobial coverage should be given, and the contrast should be instilled under gravity or by gentle injection. Instillation of 2,500 units of urokinase in 50 ml of normal saline on a daily basis may be successful in evacuating an organizing infected hematoma. Routine abscess irrigation with antimicrobials is of questionable benefit and may promote overgrowth of resistant bacteria. The catheter should be withdrawn gradually as the abscess cavity shrinks and the drainage decreases. The usual duration of drainage is 1 to 3 weeks. The catheter is removed when drainage stops and CT and ultrasonography show complete resolution. Open Surgical Drainage The incision should be smaller than that used for routine nephrectomy, and usually a posterior flank muscle-splitting incision below the 12th rib is sufficient. When the retroperitoneal abscess is entered, the pus should be cultured, and the space gently but thoroughly explored to ensure that all loculated cavities are drained. Thorough irrigation of the cavity is essential. Multiple Penrose drains should be inserted into the space through separate stab wounds, and the ends of the drains should be sutured to the skin and tagged with safety pins. Fascial and muscular closure may be performed with chromic catgut suture, but skin and subcutaneous tissue should be left open to prevent the formation of a secondary body wall abscess. The wound can be left to heal from within, or skin sutures may be placed and left untied for dermal approximation 5 to 7 days postoperatively after drainage has ceased. The wound should be packed with gauze, and the packs should be changed daily. The drains should be left in place until purulent drainage has decreased, and then they can be removed slowly over several days.

ANCILLARY PROCEDURES

If a perinephric abscess is due to long-standing obstruction and there is no functioning renal tissue, a nephrectomy at the time of drainage is theoretically attractive. Drainage of a perinephric abscess should usually be performed as a primary procedure, however, with nephrectomy performed at a later date if necessary. Patients are frequently too ill for prolonged general anesthesia and surgical manipulation. Furthermore, nephrectomy is usually difficult to accomplish technically, and preoperative information is usually not sufficient to determine accurately the amount of functioning of salvageable renal tissue. After drainage of the abscess, removal of obstruction, and appropriate antimicrobial therapy, many kidneys may regain sufficient function to obviate future nephrectomy. Nephrectomy, if indicated, can be performed using a standard nephrectomy approach or a subcapsular nephrectomy technique outlined later. A small renal abscess confined to one pole of the kidney may be managed by partial nephrectomy. If the infection extends beyond the apparent line of cleavage, however, it is essential to remove all infection, and the line of excision should extend through healthy tissue. If multiple abscesses are present, internal drainage is difficult, and nephrectomy may be required. Subcapsular Nephrectomy When a kidney is so adherent to surrounding tissues that dissection is difficult and hazardous, a subcapsular nephrectomy is indicated. These conditions are usually seen after multiple or chronic infections or previous operations have caused scarring to adjacent organs. Blunt dissection results in tearing of structures such as bowel wall. Sharp dissection when there is no definable tissue plane often results in lacerations of the vena cava, aorta, duodenum, spleen, and other structures. In subcapsular nephrectomy, dissection beneath the renal capsule enables one to avoid these vital structures. Subcapsular nephrectomy should not be performed for malignant disease and is undesirable in tuberculosis. The main difficulty with subcapsular nephrectomy is that the capsule is adherent to the vessels in the hilum, and one usually must go outside the capsule to ligate the renal pedicle. In this setting, the renal hilum usually is involved in the inflammatory reaction, and separate identification of the vessels is difficult. Kidney exposure is accomplished through the flank using a 12th rib incision. For low-lying kidneys, a subcostal incision may be satisfactory. When the kidney is reached, the capsule is incised and is freed from the underlying cortex ( Fig. 12-3). The capsule is stripped from the surface of the kidney, and an incision is made carefully in the capsule where it is attached to the hilum ( Fig. 12-4). The vessels may be protected by placing a finger in front of the pedicle when cutting the capsule. The dense apron of capsule can usually be incised best on the anterior aspect. Control of bleeding can be difficult in this procedure. Frequently all landmarks are obscured, and the renal artery and vein cannot be identified. Sharp dissection is usually required, and major vessels may be entered before they are recognized. Fortunately, the dense fibrous tissue tends to prevent their retraction. Frequently, several chromic suture ligatures can be placed through the pedicle between a proximally placed pedicle clamp and the kidney. To avoid damage to the duodenum or major vessels, pieces of capsule may be left behind. However, prolonged drainage can ensue, and as much of the infected tissue should be removed as possible. After ligation and cutting of the pedicle, the ureter is identified and cut, and the distal end is ligated. If distal ureteral obstruction has caused pyonephrosis, a small, 8 to 10 Fr red Robinson catheter may be placed in the distal ureter to allow postoperative antimicrobial irrigation. Multiple drains should be placed and brought through separate stab wounds.

FIG. 12-3. Subcapsular secondary nephrectomy showing freeing of capsule from anterior surfaces of kidney.

FIG. 12-4. Subcapsular secondary nephrectomy showing incision in and removal of a portion of renal capsule to expose and ligate renal pedicle.

OUTCOMES
Complications Complications associated with percutaneous drainage include the formation of additional abscesses that communicate with the renal collecting system and may require temporary urinary diversion via percutaneous nephrostomy drainage to affect a cure. Sepsis, the most frequent complication of percutaneous drainage, occurs in fewer than 10% of patients. Other complications, such as transpleural puncture, vascular or enteric injury, and cutaneous fistula, are rare. Additional complications to open or percutaneous drainage include prolonged purulent drainage, which may indicate a retained foreign body, calculus, or fistula. Results Cure rates for percutaneous drainage of renal and retroperitoneal abscesses range from 60% to 90%. 8,15 Multiloculated, viscous abscesses and abscesses in immunocompromised hosts are associated with lower cure rates. Large abscesses may require more than one percutaneous access procedure to completely drain them. In the past, mortality rates were reported to be as high as 50% in patients with retroperitoneal or perinephric abscesses. More recent reports indicate a significant improvement in mortality (approximately 10%), in large part because of more accurate diagnosis from improved imaging techniques, more effective antimicrobial therapy, and better supportive care. 4,6,17

SPECIAL CONSIDERATIONS

Renal Tuberculosis Renal tuberculosis is caused by hematogenous dissemination from an infected source somewhere else in the body. Both kidneys are seeded with tuberculosis bacilli in 90% of cases. Clinically apparent renal tuberculosis is usually unilateral, however. The initial lesion involves the renal cortex, with multiple small granulomas in the glomeruli and in the juxtaglomerular regions. In untreated patients who fail to heal spontaneously, the lesions may progress slowly and remain asymptomatic for variable periods, usually 10 to 40 years. As the lesions progress, they produce areas of caseous necrosis and parenchymal cavitation. Large tumor-like parenchymal lesions or tuberculomas frequently have fibrous walls and resemble solid mass lesions. Once cavities form, spontaneous healing is rare, and destructive lesions result, with spread of the infection to the renal pelvis and development of a parenchymal or peri-nephric abscess. Indications for Surgery Surgery was once commonly used in the treatment of renal tuberculosis, but since the advent of effective antituberculosis chemotherapy, it is reserved primarily for management of local complications, such as ureteral strictures, or for treatment of nonfunctioning kidneys. If surgery is warranted, it is wise to precede the operation with at least 3 weeks and preferably 3 months of triple-drug chemotherapy. Use of isoniazid, 300 mg/day; pyrazinamide, 25 mg/kg to a maximum of 2 g, once daily; and rifampicin, 450 mg/day is recommended. If segmental renal damage is obvious and salvage of the kidney is possible, a drainage procedure or cavernostomy can be performed. 7 Removal of a nonfunctioning kidney is usually indicated for advanced unilateral disease complicated by sepsis, hemorrhage, intractable pain, newly developed severe hypertension, suspicion of malignancy, inability to sterilize the urine with drugs alone, abscess formation with development of fistula or inability to have appropriate follow-up. 9,10,13 Alternative Therapy Prophylactic removal of a nonfunctioning kidney to prevent complications, remove a potential source of viable organisms, and shorten the duration of convalescence and requirement for chemotherapy is advocated by some authors. 5,19 Others, who followed a large series of patients treated with medical therapy alone, concluded that, because the frequency of late complications is only 6%, routine nephrectomy should not be performed for every nonfunctioning kidney. 9 These authors, however, treated patients for at least 2 years. The merits of short-term therapy and prophylactic nephrectomy versus long-term 2-year chemotherapy and selective nephrectomy warrant further study. Modern percutaneous drainage techniques have largely replaced open cavernostomy for treatment of closed pyocalyx. Surgical Technique Cavernostomy Renal tuberculosis sometimes results in caliceal infundibular scarring, causing a closed pyocalix. Unroofing of a pyocalix is called cavernostomy. If the calix still communicates with the renal pelvis, or if it is connected to significant functioning parenchyma, a cavernostomy should not be done because a urinary fistula or urinoma may result. To minimize wound contamination and tuberculous spread, thorough needle aspiration of purulent material and saline irrigation of the abscess cavity should be performed using a large-bore needle and syringe ( Fig. 12-5). The abscess cavity is then unroofed, and the edge is sutured with a running suture for hemostasis. Any unsuspected connection with the renal pelvis by an open infundibulum must be closed using 5-0 chromic catgut suture to prevent fistula or urinoma formation. After thorough wound irrigation, multiple drains are placed, and closure is undertaken. Drains are managed as previously described for perinephric abscess.

FIG. 12-5. Cavernotomy drainage of tuberculous renal abscess. (From Hanley HG. Cavernotomy and partial nephrectomy in renal tuberculosis. Br J Urol 1970;42:661.)

Nephrectomy When unilateral tuberculosis causes more extensive parenchymal destruction or nonfunction, a partial or total nephrectomy, respectively, should be performed. For partial nephrectomy, a guillotine incision is made 1 cm beyond the abscess. If the renal pedicle can be freed and polar vessel located and occluded, the incision can be made at the line of demarcation of the ischemia. In partial nephrectomy, it is important to try to save the capsule (if it is not involved with the infection) to cover the raw surface for hemostasis. Alternatively, fat can be used for hemostasis. The amputated calyx is carefully ligated with a 4-0 chromic catgut suture to prevent urinary fistula or urinoma formation. After nephrectomy, the distal ureter can be ligated and in most cases does not need to be brought to the skin because tuberculosis of the ureter generally heals with chemotherapy after nephrectomy. If renal tuberculosis is associated with severe tuberculosis cystitis, ureteral catheterization for 7 days postoperatively to minimize subsequent ureteral stump abscess formation should be considered. 18 Renal Echinococcosis Echinococcosis is a parasitic infection caused by the canine tapeworm E. granulosus. Echinococcal or hydatid cysts occur in the kidney in some 3% of patients with this disease. The hydatid cyst gradually develops at a rate of about 1 cm/year and is usually single and located in the cortex. Diagnosis The symptoms are those of a slowly growing tumor; most patients are asymptomatic or have a dull flank pain or hematuria. Excretory urography typically shows a thick-walled cystic mass, which is occasionally calcified. Ultrasonography and CT usually show a multicystic or multiloculated mass. Confirmation of the diagnosis is most reliably made by diagnostic tests using partially purified hydatid antigens in a double diffusion test. 13 Complement fixation and hemagglutination are less reliable. Diagnostic needle puncture is associated with significant risk of anaphylaxis as a result of leakage of toxic cyst contents. Indications for Surgery Cyst removal is indicated when an enlarging cyst threatens renal function or produces obstruction. Surgical Technique The cyst should be removed without rupture to reduce the chance of seeding and recurrence. In cases where cyst removal is impossible because of its size or involvement of adjacent organs, marsupialization is required. The contents of the cyst initially should be aspirated, and the cyst should be filled with a scolecidal agent

such as 30% sodium chloride, 2% formalin, or 1% iodide for about 5 minutes to kill the germinal portions. Complete evacuation of all hydatid tissue and thorough postmarsupialization irrigation are critical to preventing systemic effects. Penrose drains are left in the cystic cavity until drainage ceases. If large amounts of renal tissue have been damaged, partial or simple nephrectomy may be required. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Amin M, Blandford AT, Polk HC. Renal fascia of Gerota. Urology 1976;7:1–3. Brook I. The role of anaerobic bacteria in perinephric and renal abscesses in children. Pediatrics 1994;93:261–264. Doughney KB, Dineen MK, Venable DD. Nephrobronchial colonic fistula complicating perinephric abscess. J Urol 1986;135:765–767. Edelstein H, McCabe RE. Perinephric abscess: modern diagnosis and treatment in 47 cases. Medicine 1988;67(2):118–131. Flechner SM, Gow JG. Role of nephrectomy in the treatment of nonfunctioning or very poorly functioning unilateral tuberculous kidney. J Urol 1980;123:822–825. Fowler JE, Perkins T. Presentation, diagnosis and treatment of renal abscesses: 1972–1988. J Urol 1994;151:847–851. Hanley HG. Cavernostomy and partial nephrectomy in renal tuberculosis. Br J Urol 1970;42:661–666. Lambiase RE, Deyoe L, Cronan JJ, Durfman GS. Percutaneous drainage of 355 consecutive abscesses: results of primary drainage with 1-year follow-up. Radiology 1992;184:167–179. Lattimer JK, Wechsler MW. Editorial comment: surgical management of nonfunctioning tuberculous kidneys. J Urol 1980;124:191. Lorin MI, Hsu KHF, Jacob SC. Treatment of tuberculosis in children. Pediatr Clin North Am 1983;30:333–348. Mitchell GAG. The renal fascia. Br J Surg 1950;37:257–266. Patterson JE, Andriole VT. Bacterial urinary tract infections in diabetes. Infect Dis Clin North Am 1995;9:25–51. Schaeffer AJ. Urinary tract infections. In: Gillenwater JY, Grayhack JT, Howards SS, Duckett JW, eds. Adult and pediatric urology. Chicago: Year Book Medical Publishers, 1996;289–351. Sheinfeld J, Ertuk E, Spataro RF, Cockett ATK. Perinephric abscess: current concepts. J Urol 1987;137:191–194. Siegel JF, Smith A, Moldwin R. Minimally invasive treatment of renal abscess. J Urol 1996;155:52–55. Simons GW, Sty JR, Starshak RJ. Retroperitoneal and retrofascial abscesses. J Bone Joint Surg 1983;65A:1041–1058. Thorley JD, Jones SR, Sanford JP. Perinephric abscess. Medicine 1974;53:441–451. Wechsler M, Lattimer JK. An evaluation of the current therapeutic regimen for renal tuberculosis. J Urol 1975;13:760–761. Wong SH, Lou WY. The surgical management of non-functioning tuberculous kidney. J Urol 1980;124:187–191. Yoder JC, Pfister RC, Lindfors KK, et al. Pyonephrosis: imaging and intervention. Am J Roentgenol 1983;141:735–739.

Chapter 13 Renal Trauma Glenn’s Urologic Surgery

Chapter 13 Renal Trauma
Allen F. Morey and Jack W. McAninch

A. F. Morey: Department of Surgery (Urology Service), Uniformed Services University of the Health Sciences, and Brooke Army Medical Center, Fort Sam Houston, Texas 78258. J. W. McAninch: Urology Department, University of California, and San Francisco General Hospital, San Francisco, California 94110. The opinions expressed herein are those of the authors and are not to be construed as reflecting the views of the Armed Forces or the Department of Defense

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Renal injuries can be some of the most complex and challenging cases a urologist or trauma surgeon may face. The vast majority of renal injuries occur as a result of blunt trauma, and most of these are amenable to nonoperative management. Penetrating renal trauma usually occurs in conjunction with injuries to associated abdominal organs, which require urgent laparotomy. Systematic renal reconstruction at the time of laparotomy provides excellent functional results in the majority of cases.

DIAGNOSIS
Signs, symptoms, and laboratory findings that suggest renal injury should prompt immediate radiologic evaluation in stable patients. Gross hematuria after blunt trauma should warrant renal imaging in all cases. Adults with microhematuria in the presence of shock, deceleration injuries, or signs of significant abdominal, flank, or chest injuries after blunt trauma should also be imaged. 3 Pediatric patients with significant microhematuria or signs of multiple injuries after blunt trauma should be radiographically evaluated. 4 Penetrating wounds of the abdomen or flank with any degree of hematuria also warrant urgent renal imaging. The best study for assessing the injured kidney in a stable patient is a renal CT scan. Renal images can be obtained in conjunction with an abdominal CT when trauma surgeons need this study to evaluate the extent of associated intra-abdominal injuries. When unstable patients are taken emergently for laparotomy and renal injuries are suspected, a one-shot intraoperative IVP is extremely useful. The intraoperative IVP consists of a high-dose (2 cc/kg) intravenous bolus injection of radiographic contrast; a single film is taken at 10 minutes. No scout film is necessary. This technique provides important information regarding the degree of injury of the kidney in question and the status of the contralateral kidney without delaying resuscitation. 4

INDICATIONS FOR SURGERY
The decision to surgically repair the traumatized kidney is based on consideration of the patient's mechanism of injury, hemodynamic stability, associated injuries, and accurate radiographic staging of the injury. The vast majority of blunt traumatic renal injuries are clinically insignificant. At San Francisco General Hospital, fewer than 3% of patients with blunt renal trauma require renal exploration ( Fig. 13-1). Penetrating renal injuries, on the other hand, should usually be explored. Approximately 70% of patients with penetrating renal trauma are treated surgically at our trauma center. Only when radiographic staging clearly defines a penetrating injury as minor can a nonoperative approach be used successfully. 8

FIG. 13-1. Abdominal CT reveals left renal laceration after blunt trauma (grade 3). Even major renal lacerations occurring after blunt trauma are usually amenable to nonoperative management. Renal CT provides detailed information regarding the depth of laceration, size of perirenal hematoma, tissue viability, urinary extravasation, and the status of the contralateral kidney.

Persistent renal bleeding is an absolute indication for renal exploration. Relative indications for renal surgery include extensive urinary extravasation, nonviable renal tissue in association with a parenchymal laceration, incomplete clinical or radiographic staging, and arterial thrombosis. 3 Also, if a trauma surgeon elects to perform an exploratory laparotomy to manage an associated abdominal injury, we will usually repair significant renal injuries at that time in order to prevent late complications. Nearly all renal lacerations occurring from gunshot wounds require immediate repair. In the absence of severe vascular injury or hemodynamic instability, renal reconstruction may safely be attempted. Successful reconstruction can be undertaken despite spillage from bowel injury, pancreatic injury, or other associated injuries. 7

ALTERNATIVE THERAPY
Nephrectomy, when required after renal trauma, usually occurs when an injury is deemed irreparable or in the setting of hemodynamic instability. Although nephrectomy is clearly a life-saving maneuver in these instances, it is only necessary in about 10% of cases. In general, patients requiring nephrectomy are much more seriously injured, are frequently in shock, and cannot be managed conservatively. 5 Renal stab wounds are successfully managed nonoperatively in about 50% of cases at San Francisco General Hospital. The types of stab wounds most amenable to an observational approach are those occurring posteriorly or in the flank, where intra-abdominal organs are unlikely to be involved. For those stab-wound patients in whom nonoperative management is being contemplated, renal CT provides excellent information regarding the depth of laceration, extent of urinary extravasation, and size of perirenal hematoma. 8

SURGICAL TECHNIQUE
Renal exploration in the trauma setting should be carried out through a standard midline abdominal incision. This approach provides complete access to the intra-abdominal viscera and vasculature, and it also gives the greatest flexibility to assess and repair a variety of genitourinary injuries. Major bleeding noted on opening the abdominal cavity should be controlled immediately with laparotomy packs followed by surgical control and repair. Associated injuries to other abdominal organs are usually addressed before examination of the kidneys if the patient is stable. The bowel, liver, spleen, pancreas, and other organs should be inspected

systematically and carefully. The renal vasculature is routinely isolated before a retroperitoneal hematoma surrounding an injured kidney is entered. This creates a safety net for reconstruction and reduces the risk of uncontrolled renal bleeding and subsequent nephrectomy. To facilitate access to the retroperitoneum, the transverse colon is lifted out of the abdomen superiorly and placed on moist laparotomy packs. The small bowel is placed in a bowel bag and lifted anteriorly to the right. An incision is made in the retroperitoneum over the aorta from the level of the inferior mesenteric artery to the ligament of Treitz, which can be divided for additional exposure. If hemorrhage obscures the aorta, the inferior mesenteric vein is identified, and the retroperitoneal incision is placed just medial to this important landmark. Once the aorta is identified in the lower part of the incision, it is followed superiorly to the left renal vein, which reliably crosses anteriorly. The renal arteries can be found just posterior to the left renal vein on either side of the aorta. If the right renal vein is difficult to isolate through this approach, an alternative method of exposure is to mobilize the second portion of the duodenum off the vena cava. With lateral retraction on the vena cava, the right renal artery can then be isolated in its interaortocaval location. The ipsilateral renal artery and vein are individually isolated with vessel loops. These vessels are not occluded initially unless bleeding is heavy, which occurs in approximately 12% of cases in our experience. 1 Because the vessels are not routinely clamped, renal perfusion is continuous, and warm ischemia is avoided. Patients most likely to require temporary vascular occlusion are those in shock from active, uncontrolled renal bleeding. Vascular occlusion, when necessary for reconstruction, does not increase the incidence of postoperative complications when the warm ischemia time is kept around 30 minutes. After vascular control, the kidney is exposed by incising the retroperitoneum just lateral to the colon. The colon is reflected medially, and dissection through the hematoma allows renal exposure. After the kidney has been bluntly and sharply mobilized, the entire renal surface, renal vasculature, and upper ureter are routinely inspected for the presence of exit wounds or multiple injured areas ( Fig. 13-2 and Fig. 13-3). If heavy bleeding ensues, Rummel tourniquets can be applied to the vessel loops for vascular occlusion. First, the renal artery alone is occluded. If bleeding persists, the renal vein is then occluded to eliminate back bleeding.

FIG. 13-2. After preliminary vascular control, the colon is reflected, and the kidney explored. Here a small gunshot entrance wound on the anterior aspect of left kidney is identified.

FIG. 13-3. When a renal injury is identified, the entire kidney must be examined for associated wounds. Here, a complex gaping exit wound is identified on the posterior surface of the kidney. Nonviable tissue is debrided, the collecting system is closed, and segmental vessels are individually suture ligated with 4-0 chromic suture. Capsular sutures of 3-0 Vicryl are used to reapproximate wound edges.

For major polar injuries, partial nephrectomy offers the best management. Nonviable tissue is sharply debrided from the injured area. Manual compression of the adjoining normal renal parenchyma, rather than formal vascular occlusion, is extremely useful during partial nephrectomy as an adjunct during control of moderate renal hemorrhage. Arcuate arteries are individually suture-ligated with 4-0 chromic suture to control hemorrhage. The collecting system is then closed watertight with a running 4-0 Vicryl suture. Methylene blue may be injected into the renal pelvis with simultaneous compression of the ureter to elucidate any leaks in the collecting system, which may then be oversewn. The renal parenchymal defect should be covered with thrombin-soaked Gelfoam to enhance hemostasis and then covered with renal capsule, if possible. Typically, after partial nephrectomy for polar injuries, the remaining renal capsule is insufficient to allow for primary closure. In this case, an omental pedicle flap can be brought around the colon or through a window in the colon mesentery and attached with interrupted suture to the existing renal capsule for coverage of the defect. Its excellent vascular supply and lymphatic drainage make omentum an excellent tissue choice for coverage of renal injuries, especially in the setting of concomitant bowel or pancreatic injury. A retroperitoneal drain is placed routinely. Major injuries to the midportion of the kidney are more difficult to repair than polar injuries, but the same surgical principles apply. Nonviable tissue is removed sharply. Sites of bleeding are individually ligated with fine absorbable sutures, and the collecting system is closed watertight. Interrupted 3-0 chromic sutures placed superficially are ideal for renal capsule approximation. Capsular sutures are best placed without incorporating the underlying parenchyma, as that tissue is extremely friable. Thrombin-soaked Gelfoam bolsters in the defect enhance hemostasis, prevent urinary leakage, and stabilize capsular closure ( Fig. 13-4). Again, omentum should be used if primary capsular closure cannot be achieved. We frequently place a row of small titanium staples in the renal capsule near the closure to visualize the operative site on subsequent imaging studies. A retroperitoneal Penrose drain is brought out through a separate incision in most cases. Suction-type drains may initiate or prolong urinary leakage.

FIG. 13-4. Closure of parenchymal defect after central renal injury. Capsular sutures of 3-0 Vicryl may be used to sew gelatin foam bolsters into repair site. Titanium clips may be placed along the repair line to identify the area of reconstruction on subsequent imaging studies. Alternatively, if primary renal closure cannot be achieved, an omental flap may be tacked over the defect using small interrupted chromic sutures.

Renal stab wounds may be repaired using the same methods detailed above. As discussed, many may be amenable to nonoperative management. If laparotomy is performed for associated injuries, renal reconstruction should be done concomitantly. Tissue destruction is frequently much less than that seen with gunshot injuries. Frequently, entrance and exit wounds may be simply oversewn (Fig. 13-5 and Fig. 13-6).

FIG. 13-5. Technique of renorrhaphy after stab wound. Gelfoam bolsters are laid into the capsular defect, and overlying 3-0 chromic sutures are placed superficially to approximate the adjoining renal capsule, thus sealing the reconstructed area.

FIG. 13-6. Completed renal reconstruction after stab wound. The entire kidney has been mobilized and evaluated for associated wounds. Titanium clips along the capsular sutures denote the area of repair.

Renal vascular injuries are a major cause of renal loss and may coexist with parenchymal lacerations. Main renal artery or complex renal vein injuries frequently lead to total nephrectomy.5 Venous injuries may occur along the main renal vein or in segmental branches. In either case, the first step is to temporarily occlude the main renal artery. Vascular clamps are then placed proximal and distal to the venous laceration. A running suture of 5-0 vascular silk is then used to close the venous defect. Segmental arterial injuries are best repaired in a similar fashion. Smaller segmental veins can safely be ligated because of the internal collateral circulation of the venous system. Also, the left main renal vein may be ligated proximally because there is extensive collateral flow through the adrenal, lumbar, and gonadal branches. Gross blood in the urine usually clears within 24 hours, and patients should be observed at bed rest during this time. Ambulation is resumed once the urine is clear. Serial hematocrits should be monitored because delayed bleeding is possible. Renal angiography and selective embolization may be considered in the event of continued hemorrhage. Retroperitoneal drains are normally removed within 48 to 72 hours. If drainage is excessive, an aliquot may be checked for creatinine; a level similar to that of serum suggests peritoneal fluid rather than urine. Blood pressure is checked before discharge. A radionuclide study is usually obtained around the time of discharge to assess function, and a renal imaging study is again obtained at about 3 months.

OUTCOMES
Complications Small amounts of urinary extravasation are usually not clinically significant as long as they do not become infected. Large urinomas are best treated with percutaneous drainage. Delayed renal hemorrhage is most likely within the first 2 weeks, and this complication is best treated initially with percutaneous embolization and supportive therapy. Hypertension occurs rarely after renal injuries, and it is usually easily controlled by medical therapy alone. Delayed urinary bleeding may be a sign of a vascular fistula to the collecting system: this complication is frequently difficult to reconstruct and may often be best treated with nephrectomy. Results Renal reconstruction has achieved adequate preservation of function in 83% of patients at our institution. concomitant bowel or pancreatic injuries. 7 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Carroll PR, Klosterman P, McAninch JW. Early vascular control for renal trauma: a critical review. J Urol 1989;141:826–829. McAninch JW. Surgery for renal trauma. In: Novick AC, Streem SB, Pontes JE, eds. Stewart's operative urology, 2nd ed. Baltimore: Williams & Wilkins, 1989;237–245. Miller KS, McAninch JW. Radiographic assessment of renal trauma: our 15-year experience. J Urol 1995;154:352–355. Morey AF, McAninch JW. Efficacy of radiographic imaging in pediatric blunt renal trauma. J Urol 1996;156(6):2014–2018. Nash PA, Bruce JE, McAninch JW. Nephrectomy for traumatic renal injuries. J Urol 1995;153:609–611. Wessels HB, Deirmenjian JM, McAninch JW. Quantitative assessment of renal function after renal reconstruction for trauma: radiographic scintigraphy results in 52 patients. J Urol 1997;157:1583–1586. 7. Wessels HB, McAninch JW. Effect of colon injury on the management of simultaneous renal trauma. J Urol 1996;155:1852–1856. 8. Wessels HB, Meyer A, McAninch JW. Criteria for conservative management of penetrating renal trauma: comparison of non-operative and surgical treatment of grade 2–4 renal lacerations due to gunshot and stab wounds. J Urol 1997;157:24–27.
6

We have found renal salvage to be safe in the presence of

Chapter 14 Renal Allotransplantation Glenn’s Urologic Surgery

Chapter 14 Renal Allotransplantation
Bruce A. Lucas

B. A. Lucas: Kidney Transplant Program, Transplant Section, Department of Surgery, University of Kentucky Medical Center, Lexington, Kentucky 40536.

Indications for Surgery Alternative Therapy Surgical Technique Preparation of the Patient Incision and Iliac Fossa Dissection Allograft Positioning and Vascular Anastomoses Multiple Renal Vessels Ureteroneocystostomy Pediatric Kidneys Pediatric Transplantation Wound Closure Chapter References

Transplantation of a kidney allograft and subsequent immunosuppression in patients with renal failure demand surgical precision and zero tolerance for errors of judgment or technique. The devastating consequences of vascular, urologic, and infectious wound complications in renal transplantation, with their associated morbidity, mortality, and graft loss, are well documented. Fortunately, strict adherence to techniques and principles outlined in this chapter can reduce the incidence of these problems to very low levels.

INDICATIONS FOR SURGERY
Indications for surgery include patients with chronic renal failure. Contraindications to renal allograft transplantation include a history of cancer (especially hematopoietic, renal cell carcinoma, or melanoma), active infections, and patients who are a poor operative risk. Relative contraindications include oxalosis and other metabolic disorders, psychological instability, and focal glomerulosclerosis.

ALTERNATIVE THERAPY
The alternative to renal allotransplantation is chronic dialysis.

SURGICAL TECHNIQUE
Preparation of the Patient The prospective transplantation recipient should be in metabolic, fluid, and electrolyte balance to avoid perioperative hyperkalemia, unstable blood pressure, pulmonary edema, dehydration, or difficult operative hemostasis associated with inadequate dialysis. When dialysis can be scheduled in advance, as with living related donor transplantation, it should be performed on the day before surgery. The patient's cardiopulmonary status needs to be well documented, and central venous pressure monitoring is routine. Swan–Ganz monitoring is often useful. The entire abdomen is shaved and prepped after the induction of anesthesia and insertion of an indwelling 16 or 18 Fr Foley catheter. Any urine present in the bladder is submitted for culture. Then the bladder is distended with 150 ml or more of a saline solution containing Neosporin GU Irrigant. This greatly facilitates the anterior cystotomy later in the procedure and, in addition, protects against possible wound contamination when the bladder is opened. After instillation of the antibiotic solution, the catheter is clamped. The clamp is removed only after cystotomy closure is completed. Incision and Iliac Fossa Dissection A lower quadrant curvilinear incision is extended from the symphysis pubis passing 2 cm medial to the anterior superior iliac spine and up to about 4 to 5 cm below the lower costal margin (Fig. 14-1A). The upper half of the incision is extended through the external oblique, internal oblique, and transversus abdominis muscles; in the lower half of the incision, the anterior rectus fascia is incised. The rectus muscle can then be dissected inferiorly to its tendinous insertion on the symphysis pubis and retracted medially. In thin patients we prefer to keep the cephalad portion of the incision also within the lateral border of the rectus muscle, thereby obviating any transection of muscle and simplifying the closure. The inferior epigastric vessels are identified as they pass across the incision and are preserved for possible use later. Next, an anterolateral retroperitoneal fascial plane is developed, permitting extraperitoneal entry into the iliac fossa.

FIG. 14-1. (A) The incision is depicted for the right abdomen, and subsequent illustrations represent graft implantation in the right iliac fossa. The renal transplantation, however, can be performed on either the right or left side. (B) The iliac vessels are best exposed with a self-retaining retractor. Sequential separation, ligation, and division of perivascular tissue containing lymphatics are essential and must precede skeletonization on the iliac vessels.

With medial retraction of the peritoneum, the spermatic cord in the male patient or round ligament in the female patient is easily identified. In men, some of the connective tissue around the cord is freed to permit easier retraction. Usually, cord ligation should be avoided to prevent hydrocele formation, testicular atrophy, or infertility. In women, the round ligament is divided and ligated. Further development of the extraperitoneal space in the iliac fossa is accomplished with exposure of the distal common and external iliac artery. The insertion of a self-retaining retractor at this point assures adequate exposure for the subsequent iliac vessel dissection and vascular anastomoses. The dissection and skeletonization of the iliac vessels must be performed in a manner that allows secure ligation of the divided lymphatics passing along and across these vessels. Usually, this process is best approached on the medial aspect of the external iliac vein, working cephalad with a right-angle clamp toward the internal iliac artery, which crosses the vein. In some cases, especially when the donor kidney is large or has a short vein, the internal iliac artery must be sacrificed in order to achieve sufficient mobilization of the underlying vein. The iliac vein can be skeletonized as far cephalad as the vena cava if necessary. Posterior venous tributaries must be divided to permit maximum anterior mobility of the iliac vein. It is best to ligate all tributaries doubly with 2-0 or 3-0 silk in continuity before division because a

double-clamping maneuver may sometimes result in injury or avulsion of a poorly accessible stump during ligation. Hemostasis then can be achieved only with difficulty and with risk of obturator nerve injury. Unless the internal iliac artery already has been selected for an end-to-end allograft anastomosis, right-angle clamp dissection is used to partially skeletonize the common and external iliac arteries ( Fig. 14-1B). The tissue overlying the arteries and containing the lymphatics is sequentially separated, doubly ligated with 3-0 silk, and divided, a strategy that greatly reduces the incidence of lymphocele. Again, this tissue should be doubly ligated before it is incised, in contrast with double clamping and division of the tissue. Just as with the vein, the anterior separation of tissue over the iliac artery is more easily performed in a cephalad direction. At this point, palpation of the common iliac bifurcation and internal iliac artery determines the suitability of the internal iliac artery for an end-to-end anastomosis with the renal artery and the need for endarterectomy. If there is moderate or severe atherosclerosis extending into the bifurcation, or great size disparity, the internal iliac artery is usually not used. If an endarterectomy can be performed safely, or if there is little evidence of atheroma in the internal iliac vessel, skeletonization of this vessel prepares it for end-to-end anastomosis. Before skeletonization is begun, the lymphatics on the medial aspect of the iliac bifurcation should be doubly ligated and divided. If the internal iliac artery is to be used, it may be clamped proximally with a Fogarty clamp and divided distal to its bifurcation with appropriate ligation of the distal stumps deep in the pelvis. The mobilized internal iliac artery is irrigated with heparinized saline solution. Allograft Positioning and Vascular Anastomoses Before recipient vessel anastomotic sites are selected, visualization of the ultimate resting place for the allograft lateral or anterior to the iliac vessels should be considered, with all anatomic factors taken into account. The iliac vein is prepared for the end-to-side renal vein anastomosis by placement of clamps proximal and distal to the proposed venotomy. Fogarty clamps usually serve this purpose well. Excision of a thin ellipse of vein produces an ideal venotomy. The isolated segment of the iliac vein is irrigated with heparinized saline. After this, four 6-0 cardiovascular sutures are placed at the superior and inferior apices and at the midpoints of the medial and lateral margins of the venotomy. These sutures later are passed through corresponding points on the donor renal vein or vena cava patch for a four-quadrant end-to-side anastomosis. If a cadaveric kidney is used, the allograft is removed from cold storage or perfusion preservation at this point. With living related transplantation, the flushed and cooled graft is obtained from the live donor in an adjacent operating room. The kidney is secured in a sling or a 3-inch stockinette 4 containing ice slush and held in position for the vascular anastomosis by the assistant. A clamp is used to secure the sling to relieve the assistant from holding the kidney in position with the hands, which might accelerate warming of the kidney during the performance of the vascular anastomoses. The previously placed four sutures through the iliac vein are passed through the corresponding points of the donor renal vein, Carrel patch, or vena cava conduit and secured, bringing the renal vein into juxtaposition with the iliac vein ( Fig. 14-2A). The medial and lateral sutures are retracted to separate the venotomy opening and facilitate rapid anastomosis without inadvertent suturing of the back wall. With the table rotated laterally, the superior suture is used as a running suture down the medial side of the renal vein to meet the inferior suture running up. The lateral suture line is then run in similar fashion after the table has been rotated medially. The clamps on the iliac vein may be left in place until completion of the arterial anastomosis, but application of a finger Fogarty or a bulldog clamp across the renal vein at this time allows for removal of the iliac vein clamps and earlier restoration of venous return from the lower extremity.

FIG. 14-2. (A) The renal vein is brought into exact juxtaposition with the iliac vein phlebotomy by previously placed four-quadrant sutures. A running suture anastomosis will follow. (B) The renal artery is positioned to the end of the internal iliac artery by superior and inferior apical sutures. Subsequent placement of interrupted sutures completes the anastomosis. Note the occluding bulldog clamp on the renal vein. (C) The completed venous and arterial anastomoses are demonstrated.

If the internal iliac artery is to be used for the arterial anastomosis, an end-to-end anastomosis is then performed with the renal artery ( Fig. 14-2B). The two vessels are positioned to allow a gentle upward curve from the iliac bifurcation to the kidney by fixating the superior and inferior arterial apices with interrupted 6-0 cardiovascular suture. The anastomosis is completed with continuous or interrupted sutures. With the kidney resting in the iliac fossa or suspended in a sling, the initial interrupted suture may be placed midway between the apical sutures on the anterior vessel walls facing the operator, thus allowing better approximation of the opposing arterial margins, particularly when a discrepancy exists in the size of the vessels. Subsequently, the remaining sutures are placed to approximate each anterior quadrant. Next, the previously placed apical sutures are used to rotate the arteries so that the posterior vessel walls are now in the anterior position for subsequent interrupted or running suture placement. Just as before, a suture placed midway between the apical sutures again divides the rotated posterior vessel walls into quadrants for subsequent suture placement. A preference for interrupted sutures instead of a running suture in this end-to-end anastomosis prevails when one needs to avoid absolutely any pursestring effect that might occur from a running suture or to achieve optimal accommodation of the two vessels to each other when a size or thickness discrepancy exists. In most cases, the internal iliac artery is left intact to preserve potency in men as well as gluteal and pelvic blood supply in the elderly. Therefore, end-to-side anastomosis of the renal artery to the external or common iliac artery is chosen more commonly than the end-to-end procedure just described. This anastomosis usually is placed cephalad to the level of the venous anastomosis. The location of clamp placement must be carefully selected so as not to disrupt existing arteriosclerotic plaques and precipitate embolization or thrombosis. A longitudinal incision is made on the anterior or anterolateral portion of the iliac artery segment with a #11 blade knife, and a 4.8- or 5.6-mm aortic punch is used to prepare an ideal oval arteriotomy. After the incision is made, regional heparinization of the lower extremity may be accomplished by instilling about 80 to 100 ml of heparinized saline (1,000 units/100 ml) into the distal iliac limb. Systemic heparinization is usually not necessary. This anastomosis is also performed with 6-0 cardiovascular continuous or interrupted sutures after initial fixation of the end of the renal artery to an apex of the arteriotomy with suture cinched down by parachute technique. The previously placed sling around the kidney is removed. It is desirable to have obtained preoperative assessment of the recipient for existing cold agglutinins, because moderate to high titers of these agglutinins require warming of the kidney before the circulation is reestablished. 1 The vascular clamps are released after IV infusion of mannitol and methylprednisolone, venous clamps before arterial. At this point, the patient should be judiciously overhydrated with saline and albumin, and a dopamine drip should be ready to optimize renal blood flow if needed. Multiple Renal Vessels Although the Carrel patch may frequently be used with single arteries and veins, a cadaveric kidney with multiple renal arteries perfused through the aorta is especially well suited to an end-to-side anastomosis of a Carrel patch encompassing the multiple arteries ( Fig. 14-3).2 If the vessels are close to each other, a single Carrel patch is sufficient. If the vessels are more than 2 cm apart, we prefer two Carrel patches. The Carrel patch of donor aorta is fashioned to accommodate the multiple vessels, and its anastomosis to the common or external iliac artery is performed with continuous 5-0 or 6-0 cardiovascular sutures after an arteriotomy that accommodates the width and length of the Carrel patch. This anastomosis is best performed by fixating the patch at the superior and inferior apices of the arteriotomy or by parachute technique. Each suture limb runs away from the apex.

FIG. 14-3. A donor aorta Carrel patch encompassing two renal arteries is positioned by apical sutures to an iliac arteriotomy fashioned to accommodate the length and width of the patch.

The presence of multiple arteries in related donor transplantation is known in advance because all living related donors have preoperative arteriograms. Most donors have at least one kidney with a single artery, but, at times, a donor kidney with double arteries or triple arteries must be used. These arteries cannot be taken with a Carrel patch because of the risk to the donor. In these instances, several strategies for arterial anastomoses are possible: double end-to-side renal arteries to iliac artery, end-to-end superior renal artery to internal iliac artery with end-to-side inferior renal artery to external iliac artery, and implantation of an accessory artery end-to-side into the larger main renal artery, with the larger renal artery anastomosed to the internal, external, or common iliac artery. If two renal arteries are of similar diameter, the spatulation edges of the renal arteries can be joined with a running 6-0 or 7-0 cardiovascular suture to create a single bifurcating artery. 3 An accessory artery to main renal artery anastomosis should be performed with ex vivo bench technique in cold ice slush before the renal vein anastomosis is done. Finally, some recipients have a deep inferior epigastric artery that is suitable for end-to-end 7-0 suture interrupted anastomosis of a small lower-pole artery, which may be essential for ureteral viability. 8 Our experience in more than 30 cases with this technique has been excellent; no ureteral ischemia or necrosis has occurred. Ureteroneocystostomy Some patients are prepared for kidney transplantation by creation of an ileal loop or isolated ileal stoma to divert urine from a dysfunctional or absent bladder. These techniques are beyond the scope of this discussion. In addition, when the donor ureter is absent or damaged, the recipient ureter may be used for ureteroureterostomy or ureteropyelostomy to the allograft. 10 Various modifications of the Politano-Leadbetter, Paquin, and Lich techniques are used for allograft ureteral implantation into the bladder. In our experience, when the bladder is very small or the donor ureter is very short, an extravesical technique is best. 6 Otherwise, we prefer the ease and reliability of a transvesical approach without a formal submucosal tunnel. 7 In either case, previous filling of the bladder facilitates a longitudinal anterior cystotomy with minimal trauma to the bladder wall. In the transvesical approach, the bladder dome is packed and retracted cephalad, exposing the bas fond. An oblique tunnel is created in the bladder floor using a tonsil clamp directed toward the trigone from outside the bladder. This maneuver prevents subsequent angulation of the ureter when the bladder is distended. An 8 Fr Robinson catheter or heavy silk is passed through the tunnel in retrograde fashion and secured to the donor ureter ( Fig. 14-4A).

FIG. 14-4. Ureteroneocystostomy. (A) A small Robinson catheter or heavy silk suture with donor ureter attached is brought into the bladder through an oblique hiatus. (B) The completed transplant ureteroneocystostomy is demonstrated. Four interrupted sutures secure the spatulated ureteral orifice.

The ureter is pulled down and brought into position in the bladder by gentle traction. This maneuver avoids any handling of the ureter, which is important because the ureter of the transplanted kidney receives its blood supply exclusively from the renal vessel branches that course in its adventitia. In male patients, it is important to pass the ureter beneath the spermatic cord. Intravesically, the ureter is hemitransected about 4 cm from its entrance site into the bladder and spatulated about 1 cm. Four sutures of 4-0 chromic catgut are usually sufficient for an anastomosis incorporating bladder mucosa and muscularis ( Fig. 14-4B) as the ureteral transection is completed. When the apical stitch also catches ureteral adventitia 1 to 2 cm above the apex, a nice everted ureteral nipple may be produced. This eversion is especially desirable with patulous ureters. The ureter is not stented routinely. A no-touch technique is essential to avoid producing vascular insufficiency, ureteral necrosis, and urinary extravasation from injury to the adventitial vascular network of the ureter. The oblique bladder tunnel and muscle hiatus must accommodate the ureter comfortably to avoid postoperative obstruction from edema, and a gentle oblique course of the ureter must be ensured so that no kinks, twists, or obstructions occur. This attention is important because the ureter of a transplanted kidney crosses the iliac vessels in a much more caudal position than the native ureter. A little redundancy of the ureter is established outside the bladder to ensure that the ureteroneocystostomy is done without tension and that postoperative allograft swelling will not unduly stretch or angulate the ureter. Patency of the ureteroneocystostomy is confirmed by gently passing a 5 Fr feeding tube or an 8 Fr or smaller soft Robinson catheter toward the renal pelvis. Kidneys with a double ureter can also be transplanted successfully. These ureters should be dissected en bloc within their common adventitial sheath and periureteral fat so that the ureteral blood supply is protected. The technique of ureteroneocystostomy is essentially the same as with a single ureter, except that the ureters are brought through together side by side in a nonconstricting tunnel. The distal end of each ureter is spatulated, and the adjacent margins are approximated with 5-0 chromic catgut. To ensure a watertight closure, the cystotomy incision is closed in three layers. The first 3-0 chromic running suture secures the full thickness of the bladder near the bladder neck and closes the mucosal layer. The second 2-0 chromic running suture is an inverting layer of muscularis. The third 2-0 chromic layer inverts the adventitia. Each layer should overlap the immediately underlying layer about 0.5 cm at each end of the cystotomy closure to avoid urinary extravasation at these two points. Pediatric Kidneys Although en bloc transplantation of kidneys from very young children is often desirable, 5 it is not necessary to transplant both kidneys from young children en bloc each kidney can be used for a different recipient, as is the case with adult cadaveric donors, using Carrel patches of donor aorta and vena cava ( Fig. 14-5).9 A Carrel patch is mandatory in these cases because direct implantation of a small vessel into a much larger or diseased vessel may result in thrombosis or produce functional stenosis as the kidney grows. When the en bloc technique is used, the two ureters are implanted separately and stented. Pediatric kidneys have proven to be excellent donor grafts for carefully selected adults and children. Avoidance of older recipients or diabetics with advanced arteriosclerosis will minimize the potential for

thrombosis. Rapid growth and hypertrophy occur in the immediate posttransplantation period. If early rejection can be avoided, these allografts achieve adult size and function in adult recipients within several weeks.

FIG. 14-5. Small pediatric cadaver renal vessels are anastomosed to larger recipient iliac vessels using Carrel patches of donor aorta and vena cava.

Pediatric Transplantation In small children, the iliac fossa is not large enough to accommodate a kidney from an adult donor, and the pelvic vessels in a small child are so small that the disparity between the donor renal vessels and the recipient vessels precludes the technique described for adults. In these small children, graft implantation must use the recipient aorta and vena cava, which is best accomplished through a right-sided retroperitoneal or transperitoneal midline abdominal incision that provides ready access to the great vessels as well as the urinary bladder. After the right colon is reflected medially, the right kidney is usually removed to make room for the allograft. The vena cava is then freed from the level of the right renal vein inferiorly to its bifurcation or beyond. Posterior lumbar veins are doubly ligated with 5-0 silk and divided. Mobilization of the vena cava is important to facilitate the end-to-side anastomosis of the renal vein, which is performed with running 6-0 ardiovascular sutures, as described for the adult ( Fig. 14-6). Performing the venous anastomosis superiorly allows room for an end-to-side anastomosis of the renal artery to the inferior abdominal aorta. Aortic mobilization should be limited to its distal portion, from the level of the inferior mesenteric artery, and including both common iliac arteries. The segment of the aorta to be used for the end-to-side renal artery anastomosis can be isolated by a superior pediatric vascular clamp and by two inferior clamps or silastic loops on the common iliac arteries. The end-to-side anastomosis is performed with interrupted 6-0 cardiovascular sutures.

FIG. 14-6. Anatomic relationships of an adult donor kidney in a small child are shown with renal vessel anastomoses to the inferior vena cava and aorta.

Important to the revascularization of an adult kidney in small children is the need to anticipate the impending consumption of several hundred milliliters of effective blood volume by the renal allograft. Initiation of blood transfusion before beginning the vascular anastomoses will avoid hypotension after release of the vascular clamps. When the vascular anastomoses are completed, the superior aortic clamp must be kept loosely in place until it is clear that hypotension is not a problem. Immediately after establishing circulation in the graft, the anesthesiologist must obtain blood pressures at 30-second or 1-minute intervals until stabilization is assured. The ureteral implantation is carried out as described except that the ureter must be passed retroperitoneally behind the bladder near the midline. Wound Closure Except in unusual cases, the allograft ureter is not stented, and the space of Retzius and iliac fossa are not drained. Jackson–Pratt suction may be employed, but Penrose drains are never used. If good hemostasis has been obtained, and if the principles of implantation as outlined in this chapter have been followed, there is no need for postoperative drainage other than a urethral catheter. The optimal period of Foley catheter drainage is debatable. We prefer to remove the catheter within 48 hours unless the patient has worrisome hematuria, large diuresis, or poor bladder function. Before wound closure, the wound is thoroughly irrigated with saline. The wound is then closed using a 1 Maxon running suture to approximate transversus abdominis and internal oblique muscles in a single-layer closure; the adjacent fascia is included inferiorly at the tendinous insertion of the rectus muscle. Next, the rectus fascia anteriorly and the fascia of the external oblique are approximated with 1 Prolene running suture. The subcutaneous tissue is thoroughly irrigated with saline and then may be approximated with interrupted 2-0 or 3-0 sutures. These sutures are placed about 2 to 3 cm apart and include both Scarpa's fascia and the underlying fascia superficially. In this manner, one can obliterate dead space in the subcutaneous area in which a seroma in an immunosuppressed patient might become secondarily infected. The skin is approximated with interrupted fine nylon sutures or staples. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Belzer FO, Kountz SL, Perkins HA. Red cell cold autoagglutinins as a cause of failure of renal allotransplantation. Transplantation 1971;11:422–424. Belzer FO, Schweizer RT, Kountz SL. Management of multiple vessels in renal transplantation. Transplant Proc 1972;4:639–644. Codd JE, Anderson CB, Graff RJ, Gregory JG, Lucas BA, Newton WT. Vascular surgical problems in renal transplantation. Arch Surg 1974;108:876–878. Gill IBS, Munch LC, Lucas BA. Use of a cotton stockinette to minimize warm ischemia during renal transplant vascular anastomoses. J Urol 1994;152:2053–2054. Kinne DW, Spanos PK, DeShazo MM, Simmons RL Najarian JS. Double renal transplants from pediatric donors to adult recipients. Am J Surg 1974;127:292–295. Konnak JW, Herwig KR, Finkbeiner A, Turcotte JG, Freier DT. Extravesical ureteronecocystostomy in 170 renal transplant patients. J Urol 1975;113:299–301. Lucas BA, McRoberts JW, Curtis JJ, Luke RG. Controversy in renal transplantation: Antireflux versus non-antireflux ureteroneocystostomy. J Urol 1979;121:156–158. Merkel FK, Straus AK, Anderson O, Bannett AD. Microvascular techniques for polar artery reconstruction in kidney transplants. Surgery 1976;79:253–261. Salvatierra O Jr, Belzer FO. Pediatric cadaver kidneys: their use in renal transplantation. Arch Surg 1975;110:181–183 Welchel JD, Cosimi AB, Young HH, Russell PS. Pyeloureterostomy reconstruction in human renal transplantation. Ann Surg 1975;181:61–66.

Chapter 15 Ureteral Complications Following Renal Transplantation Glenn’s Urologic Surgery

Chapter 15 Ureteral Complications Following Renal Transplantation
Rodney J. Taylor

R. J. Taylor: Section of Urologic Surgery, University of Nebraska Medical Center, Omaha, Nebraska 68198-2360.

Urinary Leaks Ureteral Obstruction Diagnosis Indications for Surgery Alternative Therapy Description of the Procedure Outcomes Complications Results Chapter References

Historically, the incidence of urologic complications following kidney transplantation, manifested primarily as ureteral leaks or obstruction, was as high as 10%. 1,5 The complications often resulted in significant morbidity, graft loss, and occasional patient death. Improvements in surgical techniques, immunosuppression, and methods for diagnosing and treating the complications have led to a significant decline in the rate of urologic complications to the current reported incidence of 2% to 2½%. 4,7,9 This has resulted in lower morbidity and rare loss of a kidney or patient to urologic complications. However, despite these changes, the need for diligence in diagnosing these complications and quickly addressing them remains as true today as in the past. The most common cause for ureteral complications following kidney transplantation is technical error. 1,5,7 Damage to the ureteral blood supply during graft harvest or transplantation can result in ureteral ischemia and subsequent leak or obstruction. Additional technical errors such as excessive tension at the ureteroneocystostomy site or hematoma development within the tunnel may also cause problems. 4,7 With careful attention to detail, most of these problems can be minimized, especially in the early postoperative setting. Long-term or delayed ureteral obstruction may be the result of ischemic changes secondary to chronic rejection or a continuation of the spectrum of damage associated with the organ harvest and transplantation, and although not all are preventable, the incidence can be markedly reduced with good surgical technique. 1,4

URINARY LEAKS
In current practice most urinary leaks are the result of ureteral problems, s a majority of surgeons now employ an extravesical ureteroneocystostomy technique for implantation of the ureter. This results in a shorter ureter, decreased likelihood of ischemia, and a limited cystostomy that rarely leads to leakage from the bladder. 4,10 The majority of leaks occur early after transplantation and are manifested by either drainage from the wound, unexplained graft dysfunction, or a pelvic fluid collection. Signs and symptoms can also include fever, graft tenderness, and lower extremity edema. 8 Early urinary leaks can be divided into two types according to the timing of presentation. The first usually occurs within the first 1 to 4 days and is almost always related to technical problems with the implantation. In this case, the ureter has usually pulled out of a tunnel. This is likely caused by excessive tension at the anastomosis. This complication appears to be more common with the extravesical ureteroneocystostomies. 8 Some investigators have recommended use of a ureteral stent to lessen the likelihood of this complication. 4,5 The second type of early ureteral leak is associated with distal ureteral ischemia, which may be a consequence of injury during the donor recovery, technical causes such as tunnel hematoma, or distal stripping of the blood supply. This type usually presents between 5 and 10 days posttransplant. 7 To correct the early leak caused by excessive tension, it is often possible to do a repeat ureteroneocystostomy. In most other cases, especially with the current techniques of extravesical reimplantation, a different operative procedure is often more suitable. 5,7

URETERAL OBSTRUCTION
Ureteral obstruction can also be the result of ureteral ischemia but occurs later than ureteral leaks and usually presents as graft dysfunction. It can occur years after the transplant and in this situation may represent vascular injury associated not only with the technical complications but also with chronic rejection. 1,7,8 The spectrum of ureteral ischemic injury extends from early necrosis and urinary leakage to delayed ureteral obstruction, presenting months to years after the actual transplantation.

DIAGNOSIS
Urinary leaks are often suspected because of increased drainage from the wound. The fluid should be tested for BUN/creatinine to see if it is urine. Radiographic tests of help include an abdominal ultrasound and nuclear renal scan. A renal scan demonstrating extravasation is the most sensitive method to differentiate a urine leak from other fluid collections such as lymphoceles or hematomas. 2 A cystogram should be performed if a bladder leak is suspected. Ureteral obstruction, usually manifested by graft dysfunction, requires evaluation, and again an ultrasound and nuclear renal scan are the most common screening studies. Additional radiographic studies such as a CT scan may be of assistance in some cases. With both ureteral leaks and obstruction, endourologic techniques can be both diagnostic and therapeutic.

INDICATIONS FOR SURGERY
Anything that causes graft dysfunction or results in disruption of the urinary tract in a renal transplant patient is of utmost concern and requires rapid diagnosis and treatment. In the case of ureteral leakage or obstruction, the goals of treatment include careful and accurate diagnosis of the exact cause and site. If the problem has a physical cause such as a leak or an obstruction and is not associated with an acute rejection episode, then treatment is directed at stabilization of the renal function, minimization of morbidity, and a restoration of the continuity and function of the urinary tract. If there is concomitant rejection, then definitive operative therapy is withheld pending the treatment of rejection. 7,8

ALTERNATIVE THERAPY
The need for immediate open operative surgical intervention has been replaced, to a large extent, by early endourologic intervention. 1,6,7 and 8 The placement of a percutaneous nephrostomy can divert a leak or relieve obstruction and allow more definitive diagnosis. As described by Streem et al., endourologic management algorithms can select patients for whom the likelihood of successful nonoperative management is good. Depending on the selection criteria, the results of management of distal ureteral leaks with stenting and a nephrostomy tube show that approximately one-third of patients do well long term and require no additional treatment. For ureteral strictures or stenoses, approximately 45% of patients, carefully selected, will avoid an open operative repair. 8 For the other patients, percutaneous access can allow stabilization of renal function and a more critical assessment before open surgical repair is carried out. In a few cases, percutaneous access can offer long-term treatment with chronic stent management. This choice, in my opinion, is of limited application in most patients with a well-functioning graft because of the long-term risks (i.e., stone formation, infection, etc.) and inherent costs. However, in patients who are not operative candidates and for some patients with marginal graft function, chronic endourologic treatment can be an alternative to definitive repair. 5

DESCRIPTION OF THE PROCEDURE
There are many procedures available to restore the continuity of the urinary tract. 1,2 and 3,5,6 and 7 In our experience dealing with a difficult ureteral stenosis or a leak from significant ureteral ischemic necrosis, we favor the use of the native ureter to replace the transplant ureter. Advantages of this repair include: the native ureter is usually nonrefluxing, the results are reliable, there is a low likelihood of recurrence of the primary problem, and a tension-free anastomosis with good blood supply is easily attained. The focus of this operative description is on that surgical choice. Surgical access to the transplanted kidney and ureters (transplant and native) is usually achieved by reopening the old incision. Occasionally, if extensive mobilization of transplanted kidney is anticipated or access to the contralateral native ureter is planned, a midline incision is an option. 7 Surgical access to repair an early ureteral leak is usually simplified because dense fibrosis has not yet occurred, the fascial layers are easily opened, the peritoneum and its contents are freely mobilized medially and cephalad, and the kidney and ureter are identified without much difficulty. A primary repair can often be performed, and in most cases, a repeat ureteroneocystostomy at a new site in the bladder is the best choice. Use of a mechanical retractor greatly simplifies exposure and allows excellent access to the pelvis. If the repair has been delayed because of attempted endourologic management or because of delay in presentation or diagnosis, then access to the ureter and kidney can be much more challenging and hazardous. In these cases, mandatory preoperative preparation includes a review of the operative note, especially if the operation was performed by someone else. It is important to know whether the kidney to be operated on was the donor's right or left kidney. It is critical to know the position of the ureter and renal pelvis in relation to the renal vessels (below or above), and this depends on which kidney was used and into which side of the recipient's pelvis it was transplanted. Additional information to be sought includes the type of vascular anastomosis performed (end-to-end versus end-to-side, etc.) and whether or not the iliac vessels (especially the iliac vein) were mobilized. All of this information can help to determine the likely position of the kidney in relation to the transplanted and native ureter and the anticipated ease in gaining access to these structures. Figure 15-1 demonstrates the relationship of the transplanted kidney, vessels, and ureter to the recipient's iliac vessels and ureter. Note that this depicts a donor right kidney on the right side, as the renal pelvis is posterior to the renal vessels.

FIG. 15-1. Relationship of the transplanted kidney and its vasculature to the recipient's iliac vessels and ureter.

In terms of the recipient, it is critical to know the status of the recipient's urinary tract. This is especially true if the recipient had a history of ureteral reflux or had undergone nephroureterectomy and might not have a suitable native ureter available to use for repair. Finally, the status of the recipient's urinary bladder in terms of capacity, compliance, and function can be important in determining which other repair options are available. Additional preoperative preparation involves stabilization of the patient and function of the graft. It is important to delay any open operative repair until concurrent rejection episodes have been adequately treated and renal function stabilized. All patients should be treated with preoperative antibiotics based on anticipated contaminants or cultures obtained from the urine. If there is a likelihood that bowel might be needed (a very unusual circumstance) to repair the urinary tract, then a full bowel prep is indicated. The goals of surgery are to repair the ureteral defect, reestablish continuity of the urinary system, get rid of all foreign bodies as quickly as possible, and avoid graft or patient loss. With a well-planned and executed procedure, these goals should be easily obtained in essentially all cases. Delayed surgical repair because of attempted endourologic management, delayed diagnosis, or late presentation of obstruction makes surgical exposure of the kidney and ureter very challenging. As noted earlier, access is almost always achieved through the old transplant incision, and cephalad extension of the incision is often needed in these cases because of perinephric fibrosis, the increased size of the kidney posttransplant, and to achieve access to the iliac vessels and native ureter. It is usually possible to extend the incision several centimeters cephalad. Additional exposure, if needed, can also be obtained by extending the inferior aspect of the incision across the midline, though this is rarely needed and should be delayed until the need is present. With delayed repair, the normal tissue planes are obliterated, and a dense fibrosis has occurred around the graft. This makes it very easy to violate the “renal capsule” and get into significant bleeding. As a routine, it is preferable to operate from a position of “known to unknown” with good exposure. The surgeon should also plan to gain vascular control proximally and distally if it appears that the kidney may need to be mobilized in order to permit access to the renal pelvis. A three-way Foley catheter should always be placed into the bladder before the start of the surgery to allow for irrigation and filling with an antibiotic solution. In order to assure a safe and adequate exposure, I usually open the peritoneum early in cases where there is dense fibrosis. This allows better cephalad exposure, protects the bowel, and gives good access to the bladder. Because the transplant ureter usually crosses the external iliac vessels below the renal vessels, one should take care to avoid these structures while gaining access to the ureter. This is a critical feature of this operative procedure because exact visualization of the renal vascular structures is often difficult, and many times one is operating based on the expected, not visualized, location of these structures. In some cases a percutaneous nephrostomy tube will be placed as well as a ureteral stent. If present, the nephrostomy tube should be accessible during a procedure as injection of saline or methylene blue may aid in identifying the ureter and renal pelvis. In some cases, because of the dense fibrosis, the ureter is identified only when it is actually cut. The routine placement of a ureteral stent is of limited value in most cases because the fibrosis is so dense, it is hard to discern the presence of the catheter. If the ureter is not in dense fibrosis, then access is usually easy. Once access to the bony pelvis is obtained, careful dissection along the lateral wall of the bladder usually leads to the ureter. Once it is identified, care must be used in mobilizing the ureter to avoid any further vascular injury. When the site of leakage and/or obstruction has been identified, the most commonly used repairs include (a) a repeat ureteroneocystostomy, (b) use of the bladder (Boari flap or bladder hitch) to help bridge the gap, or (c) use of a native ureter to perform a ureteroureterostomy or ureteropyelostomy. Repeat ureteroneocystostomies are indicated only to repair early leaks when the problem was from tension at the anastomosis or distal ureteral ischemia and a well-vascularized minimally fibrosed ureter is present. In most circumstances, especially late, with a lot of periureteral reaction or ischemia, the preferred option is the use of the ipsilateral native ureter if it is present and of adequate caliber. If not, then a Boari flap is an excellent choice. Access to the native ureter is obtained by identifying it as it crosses the common iliac vessels. Care must be used in mobilizing the ureter down into the pelvis to the level of the superior vesical artery to avoid injury to the ureter blood supply. The ureter is divided well above the iliac vessels, and the proximal end of the ureter is doubly ligated. In our experience of over 30 cases, this has not resulted in problems with the native kidney or ureter requiring any further intervention. Figure 15-2 shows the native ureter mobilized distally and doubly ligated proximally in preparation for a ureteropyelostomy.

FIG. 15-2. Mobilization of the native ureter distally with the proximal segment ligated.

The operative positioning of the native ureter depends on access to the transplant ureter and/or pelvis. In addition, whether a side-to-side ureteral anastomosis or a ureteropyelostomy is to be performed may make a difference to the exact positioning of the native ureter. All of these factors relate to the extent of fibrosis and the appearance of the transplant ureter. To prevent any additional future problem, a tension-free, widely spatulated anastomosis of well-vascularized ureter to either transplant ureter or renal pelvis is critical ( Fig. 15-3). The anastomosis is performed using 5-0 Maxon (Davis and Geck, Danbury, CT) or Polydioxanone (PDS, Ethicon, Somerville, NJ) in a watertight single layer. The critical aspect is to obtain a mucosa-to-mucosa approximation avoiding tension, devascularization, and urinary leak. A 12-cm 4.7 double-J stent is routinely used on all anastomoses. The anastomosis may be additionally wrapped in omentum or peritoneal flap, if available, to decrease further the risk of leak. The wound is well irrigated with antibiotic solution, and if no preoperative infection was present, we close the wound without a drain. If there is concern about urinary leak, lymphatic leak, or possible infection, one or two Jackson Pratt drains are indicated. The fascia is closed in layers with a 0 or #1 permanent monofilament suture. The subcutaneous tissue is not closed. The skin is usually closed with staples. A nephrostomy tube, if present, is removed at day 5 to 7 after an antegrade nephrostogram has been obtained to be sure that there is no leak. The ureteral catheter is left in for 4 to 6 weeks.

FIG. 15-3. Anastomosis of spatulated native ureter to (A) transplant ureter and to (B) transplant renal pelvis.

OUTCOMES
Complications Complications that can occur postprocedure include infection, urinary leak, bleeding, recurrence of the stricture, and possible loss of graft. In all series, these are very uncommon complications. 1,7 Results We have performed over 30 native-to-transplant ureteroureterostomies or ureteropyelostomies to treat ureteral obstruction or ureteral leaks or to deal with damaged ureters at the time of the transplant. In our experience, all kidneys involved have been “salvaged,” and none lost to urologic complications. There have been no significant postoperative complications and no patient deaths. We have not had to repeat any procedures in any of the patients we have treated and have not had any recurrence of either leak or stricture. As noted earlier, we routinely tie off the proximal native ureter, do not do a nephrectomy, and have not had any problems related to the native kidney. We feel that routine native nephrectomy is not indicated, and if one is ever subsequently indicated, a laparoscopic nephrectomy would be our choice. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Banowsky LHW. Surgical complications of renal transplantation. In: Glenn JF, ed. Urologic surgery, 4th ed. Philadelphia: JB Lippincott, 1991;252–266. Bretan PN Jr, Hodge E, Streem SB, et al. Diagnosis of renal transplant fistulas. Transplant Proc 1989;21:1962–1966. Gerridzen RG. Complete ureteral replacement by Boari bladder flap after cadaveric renal transplant. Urology 1993;41:154–156. Gibbons WS, Barr JM, Hefty TR. Complications following unstented parallel incision extravesical ureteroneocystostomy in 1,000 kidney transplants. J Urol 1992;148:38–40. Khauli RB. Surgical aspects of renal transplantation: New approaches. Urol Clin North Am 1994;27:321–341. Martin DC, Mims MM, Kaufman JJ, Goodwin WE. The ureter in renal transplantation. J Urol 1969;101:680–687. Rosenthal JT. Surgical management of urological complications after kidney transplantation. Semin Urol 1994;XII(2):114–122. Streem SB. Endourological management of urological complications following renal transplantation. Semin Urol 1994;XII(2):123–133. Taylor RJ, Rosenthal JT, Schwentker FN, et al. Factors in urologic complications in 400 cadaveric renal transplants. J Urol 1984;131:336A. Thrasher JB, Temple DR, Spees EK. Extravesical versus Leadbetter–Politano ureteroneocystostomy: A comparison of urological complications in 320 renal transplants. J Urol 1990;144:1105–1109.

Chapter 16 Renal Autotransplantation Glenn’s Urologic Surgery

Chapter 16 Renal Autotransplantation
Philip Ayvazian and Mani Menon

P. Ayvazian: Department of Urology, University of Massachusetts, Worcester, Massachusetts 01655. M. Menon: Department of Urology, Henry Ford Hospital, Detroit, Michigan 48202.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Nephrectomy Renal Preservation Autotransplantation Outcomes Complications Results Chapter References

Renal autotransplantation is a safe and effective procedure to reconstruct the urinary tract. The first successful surgery was performed by Hardy in 1963 in a patient with severe ureteral injury following aortic surgery. The advent of microvascular techniques and renal preservation extends the scope of the procedure, allowing for successful extracorporeal (bench) surgery and subsequent autotransplantation. Current indications of autotransplantation include renal-vascular disease, severe ureteral damage, tumors of the kidney and ureter, complex nephrolithiasis, and retroperitoneal fibrosis. The procedure is technically demanding and is contraindicated in the setting of severe occlusive atherosclerosis of the iliac arteries. The advantages of autotransplantation include optimal surgical exposure, bloodless surgical field, and hypothermic protection of the kidney from ischemia. In cases of malignancy, there is less risk of tumor spillage and better assessment of tumor margins than by in vivo renal reconstruction. It is possible that this procedure may be underutilized because a good proportion of urologists are unfamiliar with the principles of renal homotransplantation.

DIAGNOSIS
Preoperative renal and pelvic arteriography should be performed to define the renal artery anatomy and ensure disease-free iliac vessels. In cases where autotransplantation is performed for the management of ureteral disease, ureteral involvement can be assessed by intravenous or retrograde pyelography. A CT scan of the pelvis may be beneficial in cases of retroperitoneal fibrosis to assess pelvic extension of disease.

INDICATIONS FOR SURGERY
Renal autotransplantation is particularly attractive for a variety of vascular lesions affecting the aorta and renal artery. These include traumatic arterial injuries, renal artery stenosis with extension into the segmental branches (fibromuscular disease), large aneurysms, or arteriovenous fistulas. Other vascular indications include aortic aneurysms involving the renal arteries (Marfan's syndrome) and occlusive aortic disease. In patients with central, intrarenal tumors or multiple tumors in a solitary kidney, renal autotransplantation with extracorporeal surgery is a useful technique. Following radical nephrectomy and exterior hypothermic renal perfusion, the kidney is dissected beginning in the hilum. The vasculature to the tumor is ligated. After tumor-free margins are achieved, autotransplantation is carried out. Renal autotransplantation allows for a direct anastomosis of the renal pelvis to the bladder. Therefore, it can be used in cases of ureteral damage or long ureteral lesions such as iatrogenic ureteral injuries, ureteral strictures, ureteral tumors, ureteral tuberculosis, failed urinary diversions, and retroperitoneal fibrosis. The procedure can also be used to facilitate stone passage in patients with complex nephrolithiasis. Renal autotransplantation has been effective in controlling the symptoms related to loin-pain hematuria syndrome.

ALTERNATIVE THERAPY
Replacement of the ureter for reconstruction of the urinary tract may be performed with a segment of ileum. The advantages of an ileal ureter over autotransplantation are threefold: (a) the procedure is technically less demanding; (b) vascular anastomosis is not necessary; and (c) bladder argumentation with bowel can be done simultaneously. Disadvantages include mucus production, metabolic and electrolyte imbalance, propensity for bacteriuria, and the need for indefinite radiologic surveillance of the ileal segment. Contraindications include intestinal diseases, hepatic dysfunction, and renal insufficiency (serum creatinine greater than 2.0 mg/dl). For severe renovascular disease, the first surgical options typically include in situ reconstruction. This may involve endarterectomy or aortic–renal or splenorenal bypass grafting. When these techniques are not possible and microvascular reconstruction is required, autotransplantation becomes the procedure of choice.

SURGICAL TECHNIQUE
Perioperatively, a brisk diuresis should be induced by IV hydration and 12.5 g mannitol given 1 hour before surgery. This will minimize ischemic injury to the kidney and hasten restoration of renal function. A broad-spectrum antibiotic is also administered 1 hour before surgery. During the operation, an adequate central venous pressure should be maintained with fluid boluses as needed. Renal autotransplantation is a two-step procedure: first, the kidney is removed; then, it is transplanted. The surgical approach to removing the kidney is similar to that of living-donor nephrectomy. However, the operation may be complicated by the particular disease process necessitating the surgery. Typically, two incisions are needed: the first, either a subcostal transperitoneal or extrapleural–extraperitoneal flank to remove the kidney; and the second, a lower-quadrant curvilinear or midline incision to access the iliac fossa ( Fig. 16-1). In thin patients, an alternative approach is a single midline incision from the xiphoid to the symphysis pubis, although the exposure to the kidney is not optimal.

FIG. 16-1. Location of flank and inguinal incisions.

Nephrectomy After the peritoneum and colon are reflected medially, Gerota's facia is incised. The perinephric fat is sharply dissected off the renal capsule with minimal spreading using Metzenbaum scissors. Excessive retraction of the kidney should be avoided because that may result in subcapsular hematomas or capsular tears. The adrenal gland should be carefully separated from the upper pole of the kidney. In cases where ureteral continuity is not preserved, the distal ureter is isolated and transected before the dissection of the renal hilum. The ureteral stump can be tied with #0 chromic catgut suture. The ureter should be maintained with its vascularity and the gonadal vein. To maximize ureteral viability, the tissue between the lower pole of the kidney and the ureter should be kept intact. Urinary output can then be assessed before the renal pedicle is approached. On the right side, the vena cava is carefully isolated from the surrounding tissue. The gonadal vein can be ligated at its insertion on the vena cava. The renal vein should be identified anterior to the renal artery. Accessory renal veins can be ligated, but accessory renal arteries must be maintained. Careful dissection of the renal artery is performed toward the aorta by slightly retracting the vena cava with a closed forceps or vein retractor. After a further 12.5 g of mannitol has been given, a right-angle clamp is placed on the renal artery, and it is transected. A Satinski vascular clamp is then placed proximal to the renal vein ostium, and the renal vein is transected. The renal artery can be tied with #0 silk ties or, alternatively, one #0 silk tie and a #0 silk ligature. We have found that the renal vein retracts after transection and that placement of a second larger Satinski clamp behind the first allows for a less stressful closure of the renal ostium. A 5-0 Proline suture is tied at one end, run down to the other end, back to the first, and retied. On the left side, the artery is divided at the aorta; the vein is divided anterior to the aorta and tied with two #0 silk sutures. Renal Preservation Once the kidney is removed, it is immersed in ice-cold slush at 4°C. The renal artery is flushed with either a Collins intracellular electrolyte solution or a lactated renal solution with 10,000 units/liter of heparin and 50 mEq/liter of sodium bicarbonate. Flushing is continued until the effluent from the renal vein is clear. An adaptor is used to hold a good seal during flushing. Adequate flushing allows about 4 to 6 hours of renal preservation. If extracorporeal renal surgery is required, a second team can close the flank incision. In closure of the flank, a strong monofilament absorbable suture such #1 PDS or Maxon is used. Anteriorly, the transversus abdominis, internal oblique, and external oblique muscles are closed separately. Often the transversus and internal oblique are closed together. Closure should begin after flexion is removed from the operating room table. Posteriorly, the intercostal muscles and latissimus dorsi are closed as separate layers. Autotransplantation The patient is placed in the supine position and draped. A Foley catheter carries 150 cc of 1% neomycin sulfate solution into the bladder, and the catheter is clamped. A curvilinear incision is made, extending from one fingerbreadth above the pubis to two fingerbreadths above the anterior superior iliac spine. The incision is carried down to the rectus facia. The rectus facia, the external oblique, internal oblique, and transversus abdominis muscles are opened along the line of the incision. The lateral edge of the rectus muscle is transected off the pubis to get better exposure to the pelvis. The epigastric vessels are transected beneath the transversus abdominis muscle and tied with two 2-0 silk ties. The round ligament or the spermatic cord is doubly ligated. In young men, the spermatic cord is preserved and displaced inferomedially. The peritoneum is reflected medially to expose the iliac vessels and bladder. A Buckwalter retractor is placed into the wound to provide optimal exposure. The next part of the procedure is similar to that used for renal homotransplantation. The iliac vessels are evaluated for potential size for anastomosis with the renal artery. If the caliber of the internal iliac artery is sufficient, and there is not significant plaque formation, then this vessel is selected. It is mobilized from the common iliac to the first branch, the superior gluteal artery. A bulldog vascular clamp is placed just beyond the origin of the internal iliac artery, and a right-angle clamp is placed distally ( Fig. 16-2). After transection of the vessel, the distal portion is tied with #0 silk tie. The bulldog vascular clamp is opened to test for flow. The proximal portion of the vessel is flushed with 2,000 units of heparin mixed as follows: 10,000 units per 100 cc normal saline. If a plaque is discovered, it can be trimmed back, an endarterectomy can be performed, or it can be tacked down with 6-0 silk. Further dissection of the common iliac and external iliac artery is not required. The external iliac vein is mobilized for 5 to 7 cm with special care to ligate any lymphatic vessels with 4-0 silk to prevent lymphocele formation.

FIG. 16-2. (A) Ipsilateral autotransplantation of kidney with end-to-side anastomosis of renal vein to common iliac vein and end-to-end anastomosis of hypogastric artery to renal artery. (B) The completed venous and arterial anastomoses are demonstrated.

When the internal iliac artery is unavailable, the external iliac artery is selected. After the external iliac artery is mobilized for 4 to 5 cm, an end-to-side anastomosis is performed between it and the renal artery ( Fig. 16-3). Wide mobilization of the external iliac artery may result in kinking of the vessel. Vascular clamps are placed proximally and distally, and an arteriotomy is performed. Typically only a slit is needed, and an ellipse of the anterior wall need not be removed. The distal artery is flushed with 2,000 units of dilute heparin with a red rubber catheter. A Satinski vascular clamp is placed distally on the external iliac vein, and a bulldog is placed proximal to the venotomy. The iliac vein is then carefully incised with a #11 blade scalpel to accommodate the renal vein. Four 5-0 Proline sutures are placed on the external iliac vein in an outside-to-in fashion, one at each apex and one at the midpoint on each side of the venotomy.

FIG. 16-3. (A) Renal autotransplantation with end-to-side anastomoses of both renal and external iliac arteries and renal and external iliac veins. (B) Small pediatric cadaver renal vessels are anastomosed to larger recipient iliac vessels using Carrel patches of donor aorta and vena cava.

The kidney is placed in the operative field. An assistant holds the kidney with a surgical sponge in an anatomic position with the ureter inferiorly. To minimize warm ischemia while the anastomosis is accomplished, the kidney is irrigated with cold saline. The four Proline sutures in the external iliac vein are then brought through

the renal vein in an inside-to-out fashion. The kidney is lowered into the wound, and each suture is tied with the knots on the outside. The two apex sutures are used to close the venotomy. The midpoint sutures are placed with mild traction to keep the back wall from being incorporated in the running suture. The internal iliac artery is then anastomosed to the renal artery end to end with 6-0 siliconized silk suture. The artery should be placed posterior to the renal vein to preserve anatomic relationships. The first two sutures of the arterial anastomosis are placed at either apex with double-armed needles such that the knots are on the outside. The remainder are placed with single-armed needles. We prefer an anastomosis with interrupted sutures when the internal iliac artery is used. Sutures should be placed close enough to avoid any gaps, especially at the apex. After one side is complete, the apical sutures are rotated to give exposure to the back wall. If the external iliac artery is used, the arteriotomy should be staggered with the venotomy to avoid kinking of the vessels. The anastomosis is performed with two to four continuous 6-0 Proline sutures. After completion of the anastomosis, oxidized cellulose is wrapped in small pieces around the arteriotomy and venotomy. The venous clamps are then removed, followed by the arterial clamps. It is important to maintain adequate intravascular volume with colloid or blood, especially when the clamps are removed, so that the kidney is well perfused. If this produces an excessively elevated central venous pressure, intravenous furosemide (Lasix) should be administered. Occasionally, renal autotransplantation can be performed with the ureter left intact. Although it will follow a redundant course to the bladder, normal peristalsis will provide effective drainage from the kidney. Care must be taken to avoid positioning the kidney so as to produce an obstruction of the ureter. If the ureter is transected, the urinary system can be reconstructed by a ureteroneocystostomy, ureteroureterostomy, pyeloureterostomy, or a pyelovesicostomy ( Fig. 16-4 and Fig. 16-5). We prefer an extravesical ureteroneocystostomy when there is adequate length of nondiseased ureter for a tension-free anastomosis. The Buckwalter retractor is repositioned to provide better exposure to the lateral wall of the bladder. A 2- to 3-cm tunnel is made in the bladder wall by incising the posterior lateral serosa and detrusor muscle. After the margins of the detrusor are retracted with 3-0 chromic stay sutures, the mucosa is mobilized and allowed to bulge. An ellipse of the mucosa is removed from the apex of the tunnel, and the spatulated ureter is anastomosed with the bladder mucosa using a continuous 50 chromic catgut suture. Two sutures are used, each of which incorporates 180 degrees of the anastomosis, and are tied without tension. The anastomosis is performed over a 4.8 Fr double-J ureteric stent, which is positioned into the bladder after the bladder mucosa has been opened. The stent will be removed in the postoperative period.

FIG. 16-4. Lich extravesical ureteroneocystostomy. Gregoir and Campos Freire techniques are similar.

FIG. 16-5. Alternative options for reconstruction of urinary tract: ureteroureterostomy, pyeloureterostomy, or Boari flap to renal pelvis.

The detrusor is closed over the ureter with interrupted 3-0 chromic catgut suture. The tunnel should allow passage of a right-angle clamp between the ureter and overlying muscle. The wound is irrigated with a 1% neomycin solution. No external drains are required if the ureteral reimplantation is watertight. In cases where the upper ureter is diseased, the area is removed, and the proximal ureter or renal pelvis is anastomosed to the normal lower ureter. The ureteroureterostomy or pyeloureterostomy is performed over a ureteric double-J stent by end-to-end anastomosis of the spatulated lower ureter to either the spatulated upper ureter or the renal pelvis ( Fig. 16-5). If the entire ureter is not viable, or for recurrent stone disease, a pyelovesicostomy is performed. This technique can be performed with a Boari flap and end-to-end anastomosis of the renal pelvis to tubularized bladder. The Boari flap should be secured to the psoas muscle to avoid tension on the anastomosis. The wound is then closed in layers. The rectus muscle is approximated back to the tendinous insertion at the pubic bone with a #0 Proline suture. The internal oblique and transversus abdominis are closed with a #0 Proline suture. The external oblique is closed with continuous #0 Proline. The subcutaneous layer is closed with 3-0 Dexon and the skin is closed with 3-0 nonabsorbable suture or clips. For optimal renal perfusion during the immediate postoperative period, the central venous pressure should be maintained adequately, and the diastolic blood pressure kept at 85 mm Hg or higher. Mild hypertension is preferred over normotension or mild hypotension. Aspirin can be started postoperatively to reduce the risks of graft thrombosis. A renal scan is obtained on the first postoperative day to document renal perfusion and again about postoperative day 7. Broad-spectrum antibiotics are administered during the immediate postoperative period to maintain sterile urine and help prevent infection of the vascular grafts. The ureteral stent is left in place for 2 to 3 weeks and is removed during outpatient cystoscopy. The Foley catheter is removed on postoperative day 5. It may be removed sooner if a ureteroureterostomy or pyeloureterostomy is performed, but it should be kept in place for 5 days following a pyelovesicostomy or ureteroneocystostomy. An intravenous pyelogram or a cystogram is obtained 1 to 2 weeks after surgery to evaluate ureteral integrity.

OUTCOMES
Complications Early postoperative complications include bleeding from the vascular anastomosis, renal artery or vein thrombosis, distal extremity embolization, or urinary extravasation. Bleeding from a disrupted anastomosis is a rare event but requires immediate exploration. It is usually associated with anastomosis to diseased vessels or errors in surgical technique. Peripheral collateral vessels from the renal hilum can attain significant size if there is stenosis of the renal artery or vein and can be a source of postoperative bleeding. Renal artery or vein thrombosis occurs in fewer than 2% of cases and should be ruled out in cases of oliguria following autotransplant of a solitary kidney. The diagnosis is made by renal scan; if it is made without delay, salvage of the autotransplant should be attempted. Any significant hypotension or hypovolemic event in the postoperative period or error in surgical technique can predispose to this threat. Distal extremity embolization as a result of dislodging of plaque during aortic clamping or unclamping can occur, especially with diseased blood vessels. Heparinization at the time the vessels are prepared aids in preventing this problem, but the distal pulses and color of the legs should be assessed after arterial clamps are opened. Deep venous thrombosis can result in propagation of clot from the renal vein. Intimal injury, low-flow states, and venous obstruction can predispose to

this condition. Urinary extravasation is the most common complication from autotransplantation. Placement of a ureteric double-J stent diminishes this risk. If a leak occurs, it should be treated by a percutaneous nephrostomy. In circumstances when these conservative measures fail, such as when the distal ureter is ischemic, operative repair is required. The most common late complications include renal artery stenosis, ureteral stricture, and ureterovesical reflux. Renal artery stenosis may be manifested by hypertension or impaired renal function. Diagnosis is made by renal scan and digital subtraction angiography. Initial management should be percutaneous angioplasty. Obstruction of the urinary system demonstrated by pain or impaired renal function can be managed by dilation and stenting. Results Bodie et al. reported on 24 autotransplanted kidneys in 23 patients in whom the primary indication was to replace all or a major portion of the ureter. There were no operative deaths reported. Of the 24 autografts, three were ultimately lost (12%). The function of the remaining grafts was stable or improved postoperatively. 1 Novick reported successful outcomes in 29 of 30 patients who underwent autotransplantation for the management of intrarenal branch arterial lesions. 7 Van der Valden reported on six cases of renal carcinoma treated by extracorporeal surgery and autotransplantation. Dialysis was not required, and the patients' blood pressure improved or remained within normal limits. Mean follow-up time was 54 months, with three patients dying during this period. 9 Zincke and Sen performed extracorporeal surgery and autotransplantation in 15 kidneys. Of these, 11 had renal cell carcinoma, and four had transitional cell carcinoma. Three autografts were lost because of venous and arterial thrombosis in two and necrosis of the renal pelvis and ureter in one. The remaining patients were dialysis-free with stable creatinine values. Other complications cited included a caliceal fistula requiring closure in one patient and an intimal injury requiring partial replacement of the external iliac artery with a Gore-Tex graft. 10 Novick et al. observed an increased incidence of temporary and permanent renal failure for extracorporeal compared to in situ partial nephrectomy for renal cell carcinoma. 8 Postoperative initial nonfunction occurred in five of 14 patients (36%) undergoing autotransplantation but in only two of 86 patients (2.3%) who underwent an in situ procedure. Permanent renal failure occurred in two of 14 (14.3%) autotransplanted patients and in one of 86 managed in situ(1.2%). 8 Renal autotransplantation is a rare procedure that is technically demanding with several potentially serious complications. However, in a variety of instances, it may be of great utility for organ salvage and should be included in the armamentarium of the urologist. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Bodie B, Novick AC, Rose M, Straffon RA. Longterm results with renal autotransplantation for ureteral replacement. J Urol 1986;136:1187–1189. Brunetti DR, Sasaki TM, Friedlander G, et al. Successful renal autotransplantation in a patient with bilateral renal artery thrombosis. Urology 1994;43(2):235–237. Khauli RB, Menon M. Ileal ureter and renal autotransplantation. In: Fowler JE Jr, ed. Mastery of surgery: Urologic surgery. Boston: Little, Brown, 1992;183–191. Libertino JA. Renovascular surgery. In: Walsh PC, et al, eds. Campbell's urology, 5th ed. Philadelphia: WB Saunders, 1986;2546–2551. Libertino JA, Zinman L. Renal transplantation and autotransplantation. In: Libertino JA, ed. Pediatric and adult reconstructive urologic surgery, 2nd ed. Baltimore: Williams & Wilkins, 1987;176–177. Novick AC. Technique of renal transplantation. In: Novick AC, et al, eds. Stewart's operative urology, 2nd ed. Baltimore: Williams & Wilkins, 1989;324–340. Novick AC. Microvascular reconstruction of complex branch renal artery disease. Urol Clin North Am 1984;11(3):465–475. Novick AC, Streem S, Montie JE, et al. Conservative surgery for renal cell carcinoma: a single-center experience with 100 patients. J Urol 1989;141:835–839. Van der Valden JJ, Van Bockel JH, Zwartendijk J, Van Krieken JH, Terpstra JL. Longterm results of surgical treatment of renal carcinoma in solitary kidneys by extracorporeal resection and autotransplantation. Br J Urol 1992;69 Zincke H, Zen SE. Experience with extracorporeal surgery and autotransplantation for renal cell and transitional cell cancer of the kidney. J Urol 1988;140:25–27.

Chapter 17 Nephroureterectomy Glenn’s Urologic Surgery

Chapter 17 Nephroureterectomy
Gary D. Steinberg

G. D. Steinberg: Department of Surgery, Section of Urology, The University of Chicago Hospitals, Chicago, Illinois 60637.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Two-Incision Approach Outcomes Complications Results Chapter References

Malignant tumors of the upper urinary tract are uncommon and account for only 5% to 10% of all urothelial malignancies. The peak incidence is in the sixth and seventh decade of life with a male predominance of 2:1. 3 Most upper tract tumors are transitional-cell carcinoma (TCC, 85% to 90%), with 10% to 15% squamous cell carcinoma or mixed TCC and squamous. Adenocarcinoma of the renal pelvis is extremely rare, accounting for only 1% of upper tract tumors. Cigarette smoking is the major risk factor for development of TCC of the renal pelvis. It has been reported that there is a three- to sevenfold increased risk of carcinoma associated with cigarette smoking and that cessation of smoking is associated with a decreased risk. Phenacetin abuse is also associated with an increased risk of TCC of the renal pelvis. Although the specific mechanism of tumorigenesis is unknown, the phenacetin metabolite 4-acetoaminoprophenol is thought to cause chronic inflammation and papillary necrosis. The combination of papillary necrosis and chronic inflammation has been associated with a 20-fold increased risk of cancer development. 5 Balkan nephropathy, also known as Danuvian endemic familial nephropathy, is a condition strictly associated with TCC of the upper tracts. This endemic disease is confined to the Balkan states that lie on the Danube river. Cancer of the renal pelvis in these states accounts for 42% of renal tumors. The specific cause is unknown, although the drinking water has been suggested. The tumors are typically low grade, multifocal, and slow growing. Bilateral tumors occur 10% of the time. Occupational risk factors have also been correlated with TCC of the renal pelvis, including exposure to chemicals in the rubber, petroleum, plastics, and aniline dye industries. Forty to eighty percent of patients with upper tract tumors will have urothelial carcinomas at some time elsewhere in the urinary tract, usually in the bladder. About 3% of patients with transitional cell cancer of the bladder develop upper tract tumors; however, patients with urothelial tract tumors of the prostate or urethra have approximately a 30% risk of developing upper tract tumors. 1

DIAGNOSIS
Approximately 80% of patients present with hematuria. Some patients present with flank pain or constitutional symptoms. Intravenous pyelography (IVP) is the initial study of choice in the evaluation of a patient suspected of having a renal pelvic or ureteral tumor. Assessment of the entire urinary tract is important in evaluating patients diagnosed with a renal pelvic or ureteral tumor because the upper urinary tract has a high potential of developing multiple tumors as described by the field-change theory. Grabstald reported that approximately 50% of patients with renal tumors have coexisting tumors in the ipsilateral ureter and bladder, and 3% to 4% of those patients have tumors in the contralateral upper urinary system. A retrograde pyelogram is usually indicated if the collecting system of the affected kidney is not completely visualized or in the case of renal insufficiency or contrast allergy. Additional urothelial assessment may include renal pelvic and/or ureteral washing for cytology, brush biopsies, cystoscopy, and bladder washing for urinary cytology. The role of ureteroscopy in the diagnosis of upper tract tumors is complementary and may confirm the findings of the IVP, retrograde pyelography, and cytology. Ureteroscopy may aid in visual identification and biopsy of tumors for grading and staging. Additional staging evaluation for the detection of metastatic disease should include a chest radiograph and/or computed tomography (CT) of the chest, abdomen, and pelvis. A bone scan may be obtained in patients with an elevated serum calcium, alkaline phosphatase, or bony abnormalities seen on CT scan.

INDICATIONS FOR SURGERY
Nephroureterectomy with excision of a cuff of bladder is the classic surgical procedure for carcinoma of the renal pelvis or ureter. However, conservative surgery may be indicated in those patients diagnosed with a small, solitary, well-differentiated papillary tumor. Current staging techniques, however, may make accurate preoperative staging and grading of tumors difficult. In addition, half of all cases of ureteral tumors involve at least the musculature. Furthermore, there is a high incidence of multiple ipsilateral tumors. Last, recurrent tumors in the remaining ureteral stump have been reported in more than 30% of patients treated by nephrectomy and partial ureterectomy. Although patients with solitary distal ureteral tumors may be successfully treated with distal ureterectomy and ureteroneocystostomy, in general, a conservative surgical approach should be reserved for the highly selected patient pop-ulation in whom nephron sparing is essential, i.e., pa-tients diagnosed with bilateral tumors, Balkan nephropathy, patients with a solitary kidney, renal insufficiency, and patients with comorbid health problems. Patients treated with a conservative approach are at increased risk of local recurrence and require frequent and careful follow-up including IVPs, retrograde pyelograms, and endoscopies.6

ALTERNATIVE THERAPY
Alternatives to nephroureterectomy include (a) endoscopic resection and/or fulguration, in either a retrograde or antegrade fashion, (b) topical chemo- or immunotherapy via either a nephrostomy tube or ureteral stent, (c) external beam radiotherapy, or (d) laparoscopic nephroureterectomy. Lesions in the ureter may be treated with resection of the ureteral tumor and ureteroureterostomy, replacement with ileal interposition, and ureteral reimplantation. These operations require a careful assessment of the entire urothelium and careful follow-up. Because of the “field change” effect of the urothelium and multiplicity of tumors, these operations may not be appropriate in patients with high-grade or -stage tumors.

SURGICAL TECHNIQUE
In performing a nephroureterectomy, technical considerations include the choice of incision, whether it is appropriate and to what extent the surgeon should perform a lymph node dissection, and excision of bladder cuff and distal ureter via an intravesical versus extravesical approach. Two-Incision Approach An intrathoracic, extrapleural, extraperitoneal approach, removing the kidney within Gerota's fascia without removing the adrenal gland, is our preferred procedure. In order to gain proper exposure, the incision can never be too high, and thus, a tenth interspace or supra-11th-rib incision is generally utilized. The patient is placed in a modified flank position (approximately 60 degrees rotated) with the table flexed and the kidney rest elevated. The patient is taped into position with wide adhesive tape, and an arm rest is utilized. The patient is adequately padded with an axillary roll, pillows, and sheets and is prepped and draped from nipples to the symphysis pubis in the usual sterile fashion. The 11th rib and the tenth intercostal space are identified, and a supra-11th-rib incision is made in the tenth intercostal space. The incision extends from the edge of the erector spinae muscle and courses obliquely and medially to the lateral border of the rectus fascia to incise the external and internal oblique muscles, exposing the transversalis fascia, and the latissimus dorsi and serratus muscles, exposing the intercostal muscles. The lumbodorsal fascia is then incised at the level of the tip of the 11th rib to avoid inadvertent division of the peritoneum or pleura ( Fig. 17-1).

FIG. 17-1. The patient is rotated on the table and flexed in the flank position. The incision is made through the transversalis abdominis muscle off the end of the 11th rib to avoid the pleura and peritoneum. The lower border of the pleura is shown. (From Marshall FF. Radical nephrectomy. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;18–25.)

The peritoneum is mobilized off the posterior aspect of the transversalis muscle, moving the peritoneum medially and inferiorly. Primarily by use of blunt dissection with minimal sharp dissection, Gerota's fascia is mobilized superiorly from the diaphragm and posteriorly and inferiorly from the psoas and quadratus musculature (Fig. 17-2). The intercostal muscles are then incised, carefully avoiding the pleural membrane. The plane between the pleura and chest wall is identified with careful blunt dissection along the tenth rib using a Kitner dissector. The diaphragmatic attachments to the 11th and 12th rib are transected sharply down to their insertion between the quadratus and psoas muscles, avoiding the intercostal nerve and vessels. Sharp dissection is continued posteriorly until the intercostal ligament is divided, allowing the rib to hinge posteriorly ( Fig. 17-3).

FIG. 17-2. As the left radical nephrectomy continues, Gerota's fascia is mobilized off the abdominal side of the diaphragm. The diaphragm can be seen after the transversalis muscle has been divided. Pleura is identified in the space between the transversus abdominis muscle and the diaphragm. It can then be mobilized with the Kitner dissector superiorly. (From Marshall FF. Radical nephrectomy. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;18–25.)

FIG. 17-3. The transverse costal (intercostal) ligament is divided after the pleura has been reflected superiorly and the diaphragm has been divided. Division of the intercostal ligament allows the 11th rib to hinge inferiorly and improves the exposure. (From Marshall FF. Radical nephrectomy. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;18–25.)

A self-retaining retractor is placed into the wound for optimal exposure. A Balfour or Finochietto retractor may be used, but a multibladed ring retractor that is secured to the operating table such as a Buckwalter retractor is preferable. The renal mass within Gerota's fascia is rotated medially, and the dissection is carried posteriorly off the psoas and quadratus musculature. The iliohypogastric and ilioinguinal nerves and 12th thoracic neurovascular bundle can usually be identified ( Fig. 17-4).

FIG. 17-4. The transversalis fascia and Gerota's fascia are mobilized from the psoas and quadratus musculature posteriorly. The iliohypogastric and ilioinguinal nerves and 12th thoracic neurovascular bundle are seen. (From Marshall FF. Radical nephrectomy. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;18–25.)

The colon is then held medially and superiorly, an avascular plane between the colonic mesocolon and Gerota's fascia is developed, and the renal mass is sharply separated from the peritoneum. By use of sharp and blunt dissection, the superior and inferior aspects of the kidney are dissected free of the adrenal gland and surrounding tissues, respectively. There may be several vessels between the adrenal gland and the kidney that should be ligated with ligaclips. The kidney is dissected posteriorly to the level of the renal hilum. Attention is directed to the main renal vessels. The pulsating renal artery is identified by palpation, double ligated

as it exits the aorta with 0 silk sutures and then divided ( Fig. 17-5).

FIG. 17-5. The aorta is located, and the renal artery can be identified, ligated, and divided. (From Marshall FF. Radical nephrectomy. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;18–25.)

On the right side, especially with a large tumor mass, the artery may be approached anteriorly or in the interaortocaval region, though the preferred approach to the right renal artery is posteriorly. On the left side, the gonadal and adrenal veins are identified anteriorly, as is the renal vein. The gonadal and any lumbar veins are ligated before double ligating the renal vein with 2-0 silk suture ( Fig. 17-6 and Fig. 17-7). On the right side, the inferior vena cava is identified as well as the renal vein. Careful palpation for a second renal artery is important before ligation of the renal vein. The remaining soft tissue attachments to the kidney should be divided so that the only remaining attachment is to the ureter.

FIG. 17-6. The medial dissection includes mobilization of the colon and mesocolon from Gerota's fascia. Gerota's fascia remains with the specimen. The aorta is identified, and an en bloc periaortic node dissection is commenced. Care is taken not to injure the femoral branch of the genitofemoral nerve on the psoas muscle. (From Marshall FF. Radical nephrectomy. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;18–25.)

FIG. 17-7. The gonadal vessels and ureter are ligated inferiorly. The regional node dissection is carried along the aorta, and the renal vein is ligated and divided. Additional vessels to the adrenal gland are ligated and divided. Adrenal gland is usually not removed with nephroureterectomy for renal pelvic or ureteral tumors. (From Marshall FF. Radical nephrectomy. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;18–25.)

Attention is then directed to the inferior aspect of the kidney. The ureter is identified and dissected free to a level distal to the bifurcation of the iliac vessels. The ureter is ligated distally with a 0 silk suture, making sure not to include any surrounding tissue in the ligature. A large straight clip is placed proximally to prevent urine spillage, and the ureter is divided. The specimen is then removed. Transitional-cell carcinoma may spread by direct extension or metastasis by hematogenous or lymphatic routes. Therefore, a regional lymphadenectomy should be performed as part of the surgical procedure. A lymph node dissection is performed by identifying the midline of the aorta for a left-sided tumor, and the vena cava for a right-sided tumor. Starting from just cephalad to the renal hilum to the level of the inferior mesenteric artery, the lymphatic tissue is dissected using a “split and roll technique” with ligaclips placed on the lymphatics to avoid a lymphocele. Hemostasis is obtained using electrocautery. The diaphragm is not repaired if only the lateral attachments have been taken down. If a pleurotomy has been made, a red rubber catheter with additional side holes cut out is placed into the pleural space, and the pleurotomy is closed with a running 3-0 chromic suture. The kidney rest is lowered, the table is taken out of flexion, and the wound is closed in two layers using a continuous suture of #1 PDS. The skin is closed using staples. The pleural cavity is then bubbled out with the red rubber catheter in a basin of saline. When fluid and bubbles cease to emerge from the catheter, it is removed, and additional skin staples are applied. Auscultation of the chest at the apex of the lung as well as a chest x-ray should be performed postoperatively to diagnose a pneumothorax. If there are any concerns, a temporary chest tube may be placed. The patient is taken out of flank position, placed in supine position over the break of the operating room table with the table flexed, and prepped and draped in the usual sterile fashion. A 20-Fr Foley catheter is passed into the bladder, and the bladder is then filled with 200 to 300 cc of normal saline. A lower midline abdominal incision is made and carried down through the rectus and transversalis fascia. A Balfour retractor is placed. The bladder is identified and opened longitudinally between two laterally placed 2-0 Vicryl stay sutures. Additional stay sutures are placed at the apex of the incision in the bladder. The ureteral orifices are identified, the bladder is packed with several sponges, and the bladder blade is placed in the dome of the bladder. A 5-Fr feeding tube is placed in the ipsilateral ureteral orifice and sewn in place with a 4-0 chromic suture. The ureteral orifice is circumscribed sharply, including a 1-cm cuff of bladder. The ureter is dissected from its orifice using a pinpoint electrocautery and sharp dissecting scissors ( Fig. 17-8).

FIG. 17-8. A catheter is placed in the left ureteral orifice and sutured. A wide circumferential incision around the ureteral orifice and periureteral dissection free the intravesical ureter. (From Lange PH. Carcinoma of the renal pelvis and ureter. In: Glenn JF, ed. Urologic surgery, 4th ed. Philadelphia: JB Lippincott, 1991;273.)

The entire distal ureter is dissected free to the level of the 0 silk tie, removed with a cuff of bladder, and passed off the table as a specimen. In most cases the remaining stump of distal ureter may be removed entirely with an intravesical approach; however, in some cases additional extravesical dissection is required in which the superior and middle vesicle pedicles are divided. A two-layer closure of the posterior bladder wall is performed using 2-0 Vicryl suture to close the muscle and serosa and 5-0 chromic to close the bladder mucosa. A 3-0 Vicryl continuous suture and subsequently 2-0 Vicryl figure-of-eight Lembert sutures are used to close the bladder incision in two layers. A Davol drain is placed in the pelvis and secured with a 3-0 nylon suture. The abdomen is closed with a continuous #1 PDS suture. The skin is closed with staples.

OUTCOMES
Complications Early complications include hemorrhage, wound infection, pneumothorax, atelectasis, and pneumonia. Meticulous dissection around the renal vessels, aorta, and vena cava will aid in decreasing intraoperative blood loss. The supra-11th-rib incision provides excellent exposure to the great vessels and kidney, thus reducing the chance of inadvertent injury to the vasculature. Later complications include “flank sag,” which may be related to division of more than one intercostal nerve. Results The survival rate after nephroureterectomy is dependent on the stage and grade of the tumor. Superficial low-grade tumors rarely metastasize and when adequately treated rarely decrease life expectancy. Invasive lesions have a higher metastatic rate and are associated with a poorer prognosis. Patients with low-grade and high-grade tumors have approximately 80% and 20% survival at 5 years, respectively. 4 Patients with pT2–3a renal pelvic and ureteral tumors have a 75% and 15% survival at 5 years, respectively, and patients with pT3b–4, N+ tumors have approximately a 5% survival at 5 years. Interestingly, in patients with ureteral tumors, survival may be more dependent on the stage and grade of tumor than the surgical approach. 2 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Abercrombie GF, Eardley I, Payne SR, et al. Modified nephroureterectomy: long-term follow-up with particular reference to subsequent bladder tumors. Br J Urol 1988;61:198. Badalament RA, O'Toole RV, Kenworthy P, et al. Prognostic factors in patients with primary transitional cell carcinoma of the upper urinary tract. J Urol 1990;144:859. Kleer E. Transitional cell carcinoma of the upper tracts. In: Soloway MS, ed. Problems in urology, vol 6, no. 3. Philadelphia: JB Lippincott, 1992;531. Nielson K, Ostri P. Primary tumors of the renal pelvis: evaluation of clinical and pathological features in a consecutive series of 10 years. J Urol 1988;140:19. Ross RK, Paganini-Hill A, Landolph J, et al. Analgesics, cigarette smoking, and other risk factors for cancer of the renal pelvis and ureter. Cancer Res 1989;49:1045. Teffens J, Nagel R. Tumors of the renal pelvis and ureter: observations in 170 patients. Br J Urol 1988;61:277.

Chapter 18 Pyelolithotomy Glenn’s Urologic Surgery

Chapter 18 Pyelolithotomy
John M. Fitzpatrick

J. M. Fitzpatrick: Department of Surgery/Urology, Mater Hospital and University College Dublin, Dublin 7, Ireland.

Diagnosis Indications for Surgery Alternative Procedures Surgical Technique Surgical Access to the Kidney Access to the Renal Pelvis Simple Pyelolithotomy Extended Pyelolithotomy Additional Nephrotomies Wound Closure Outcomes Complications Results Chapter References

Pyelolithotomy is an operation that is now uncommonly performed. The advent of percutaneous nephrolithotomy with contact lithotripsy (PCN) and extracorporeal shockwave lithotripsy (ESWL) has reduced the indications for pyelolithotomy, which is a considerably more invasive procedure. It is interesting to note that when surgeons were first performing pyelolithotomy, there was considerable disagreement as to which was the preferred approach to stone removal. Clearly, people liked to argue even then, when one had to do so mainly by letter or book rather than by published article, conference, phone call, telefax, or Internet. Vincenz Czerny probably performed the first pyelotomy, with Sir Henry Morris performing a similar operation to remove a stone in the same year, 1880. Further developments took place over the years, with many incisions through the thorax and abdomen being introduced, and then many incisions through the renal pelvis and renal parenchyma following. The introduction of radiologic visualization of the kidney completed the picture. It is clear, however, that the landmark contributions to open surgical removal of stones from the kidney were made in recent years by Gil Vernet, 5 Marshall,7 Boyce,1 and Wickham. 9

DIAGNOSIS
The usual presenting symptom for renal calculi is radiating colicky flank pain, usually associated with hematuria. Larger stones, however, may be relatively asymptomatic or present with persistent infection and/or hematuria. The diagnosis of renal calculi is generally made radiographically. Currently, the most common radiologic method of diagnosis is via a KUB and intravenous pyelography, though some centers are investigating the use of ultrasound and computed tomography.

INDICATIONS FOR SURGERY
Although its use is limited because of this relative invasiveness, it sometimes has a role to play, particularly when the stone burden is large or when problems with body shape or habitus prevent percutaneous access to renal calculi or focusing on the stone by ESWL.

ALTERNATIVE PROCEDURES
Alternatives to pyelolithotomy include ESWL, percutaneous stone extraction/destruction, ureteroscopic stone destruction, chemolysis (uric acid or struvite stones), or anatrophic nephrolithotomy.

SURGICAL TECHNIQUE
Surgical Access to the Kidney The urologist may consider five possible approaches to the kidney for open stone removal: 1. Flank approach a. Subcostal b. Costal (11th or 12th rib) c. Intercostal (above the 11th or 12th rib) 2. Transabdominal 3. Posterior lumbotomy The advantages of the flank approach are described after I explain why, in my opinion, the transabdominal and posterior lumbotomy incisions are rarely required in what is, after all, a relatively uncommon procedure. Transabdominal and transperitoneal access may be required if the patient has spinal deformities and very occasionally after several previous surgical procedures. It is the most invasive of all the approaches, and recovery is delayed postoperatively. It should not be used as the standard approach for pyelolithotomy. The posterior lumbotomy incision ( Fig. 18-1) has its advocates, particularly because postoperative pain is minimal, and recovery is quick with shortened hospital stay. The patient is placed either in the lateral decubitus position or prone, with pillows under the upper abdomen. The incision is made about 2.5 cm lateral to the erector spinae muscle from the 12th rib down to the superior border of the iliac crest. The incision is extended through fat and fascia and then through the aponeurotic fibers of the latissimus dorsi. If further access is required, a small part of the 12th rib can be removed, or the lower end of the incision can be curved inferolaterally along the iliac crest. The advantages of this approach have been listed above, but a major disadvantage is that access to the upper pole is difficult, as is access to the ureter below its upper portion. I feel that when an open operation is being performed today for stone removal, it is unlikely to be a simple procedure but rather a more complex one for which greater exposure may be required than is afforded by this incision.

FIG. 18-1. Gil Vernet incision. (A) Line of posterior vertical lumbotomy with patient in Murphy's position. (B) Plane of dissection. (C) With sacrolumbar and quadratus lumborum muscles retracted, the kidney is rotated to present the hilar surface.

Of the flank incisions, I prefer the costal route of access. The subcostal approach is usually too low for renal surgery of any complexity. In considering the incision, it is worth remembering that although it is possible to be too low, preventing complete visualization of each step of the subsequent dissection, it is never possible to be too high. For this reason, the skin incision should be made on top of or superior to one of the ribs. In this way, damage to the subcostal or infracostal nerves is prevented, and the likelihood of a wound hernia is minimized. The intercostal approach requires division of the posterior costotransverse ligament in order to allow the rib to “bucket-handle.” Otherwise, access between the ribs may be suboptimal, and, indeed, the ribs may break when spread apart by a self-retaining retractor. When an incision is based on a rib (the costal approach), removal of the end of the rib is required. A careful review of the preoperative x-ray films and examination of the patient with the table broken will clarify whether the 12th, 11th, or even, on some occasions, the tenth rib should be the line of the skin incision. The patient should be placed on the table in the lateral decubitus position with the side to be operated on facing directly upward ( Fig. 18-2). The patient can be stabilized by inserting three T-pieces along the sides of the table or by fastening tape to the upper thorax and over the hip, thereby fixing the patient on the table. The table is broken, thus opening up the space between the ribs and the iliac crest, and then tilted 20 degrees laterally toward the surgeon. The surgeon and assistant are both positioned behind the patient, and the surgeon can sit down throughout the procedure.

FIG. 18-2. Classic loin approach. (A) Patient positioned and table flexed. (B) The entire table is tilted about 20 degrees toward the surgeon, who may then be seated for the operation.

The incision is made in the skin over the distal 6 cm of the rib, extending medially for another 10 to 12 cm. It should be deepened down to the rib before any muscles are cut. Once the rib can be clearly seen, the skin, fat, and fascial layers are retracted on both sides to give a better view. The intercostal muscles are divided above the rib with a knife until the diaphragm can be seen; this is then divided with a scissors until the distal 6 cm of rib is cleared. Then the same approach is made below the rib, with the intercostal muscles being divided if the 11th rib is being used, or the latissimus dorsi if it is the 12th. The diaphragm will not be divided inferior to the rib, but it is advisable to identify the nerve bundle and sweep it inferiorly. Once the rib has been dissected free of the surrounding muscles, the distal 6 cm is removed with a rib shears. I do not approach the rib subperiosteally, as leaving the periosteum does not confer any advantage. Once the rib has been removed, Gerota's fascia can be visualized, and through it the kidney can be palpated. Two fingers should be introduced under the abdominal muscles through this incision, and the peritoneum swept away from their under surface. The incision is then deepened through the external oblique, internal oblique, and transversus abdominus muscles using the knife or cutting diathermy. The incision should not extend as far medially as the rectus sheath. At this stage, a body wall retractor should be inserted, preferably the Wickham retractor, which is self-retaining and shaped to the body wall. Gerota's fascia is then opened, and this incision is extended upward toward the diaphragm and inferiorly toward the pelvic brim. The peritoneum can then be mobilized medially away from the ureter, which is visualized exiting from the perirenal fat, and the ureter is encircled with a loop or tape. The perirenal fat is then grasped over the lateral border of the kidney, elevated by two Babcock forceps, and incised, revealing the capsule of the kidney ( Fig. 18-3). The degree of mobilization of the kidney required depends on how large the stone is. If full mobilization is required, it can be performed easily but should always be carried out by sharp dissection with a Metzenbaum scissors under direct vision. Remember that the main renal vein is always best accessed from anterior to the kidney (although there may be tributaries lying posteriorly). The main renal artery is best approached from above and posterior to the kidney, although there may be other branches, particularly an upper pole or lower pole branch directly from the aorta. The artery need not be isolated in the unusual situation that small stones are being removed, unless a parenchymal incision is being contemplated. When it is isolated, a loop or tape should be put around it.

FIG. 18-3. Babcock clamps are used to grasp the perirenal fat, which is incised to reveal the renal capsule

In the case of surgical access after previous surgery and after many previous surgical procedures, great care must be taken ( Fig. 18-4). After incision of the muscles, the fascial and perirenal areas are likely to be greatly thickened and indurated. It is very helpful to isolate the ureter first (prestented if required) and to trace the ureter upward to the ureteropelvic junction. The kidney can then be dissected free from the surrounding tissues with the scissors. Care must be taken not to incise the renal capsule because considerable hemorrhage can occur under these circumstances. Because the kidney is likely to be encased in dense fibrous tissue, the perirenal anatomy will be difficult to define accurately, and upper and lower pole arteries can be damaged. In addition, identification of the renal artery can be somewhat more difficult; palpation of the tissues medial to the kidney will reveal its position.

FIG. 18-4. Standard pyelothotomy with sinus retraction.

Access to the Renal Pelvis In general terms, it is preferable to open the renal pelvis posteriorly rather than anteriorly. This approach will avoid the renal vein, which often runs along the upper part of the anterior surface of the pelvis. Damage to the renal parenchyma can be avoided if only the renal pelvis is incised. The degree of dissection around the renal pelvis will not affect renal function. 2 The subparenchymal and intrasinusal pyelotomy 5 has made easier the removal of even the most complex calculi. Simple Pyelolithotomy This method of opening the renal pelvis would only be considered if the stone to be removed is only 1 to 2 cm in diameter in the renal pelvis or in a calyx, or if there were a number of such sized calculi in several calyces. After opening Gerota's fascia and the perirenal fat and putting a tape around the upper ureter, as described above, the amount of dissection in the region of the renal pelvis that is required is not extensive. The ureteropelvic junction and the pelvis itself should be clearly defined, but a subparenchymal dissection is not required unless the pelvis is intrarenal ( Fig. 18-5). After placing two stay sutures of 4-0 polyglycolic acid or chromic catgut, make a longitudinal incision in the renal pelvis using a scalpel. The incision must not extend through or into the ureteropelvic junction because of the risk of subsequent scarring.

FIG. 18-5. Exposure of the renal sinus in the presence of severe inflammatory erection. (A) Dissection begins along the upper ureter and proceeds superiorly. (B) Peripelvic fat is mobilized and then incised. (C) Near the hilus, peripelvic tissue is dissected bluntly. (D) Excess adipose tissue may be excised. (E) With hilar retractors in place, intrasinusal fat is dissected away by blunt dissection using surgical gauze.

When the urologist is certain that all stones have been removed, the pelvis should be closed with continuous 4-0 polyglycolic acid or chromic catgut suture. The attempt is to make the closure watertight, but even making the suture continuous does not guarantee this, so the peripelvic tissues should be drained. Extended Pyelolithotomy In most cases, the reason for performing open pyelolithotomy will be the complexity of the stone in the renal pelvis and its multiple extensions into the calyces. In many cases it will be possible to remove the stone completely by extending the dissection under the parenchyma and exposing the renal pelvis and calyceal infundibula in the manner described by Gil Vernet. 5 In this way, incisions into the renal parenchyma can be avoided, thus reducing the potential for renal injury. The anatomy of the renal hilum allows for extensive exposure of the renal pelvis, but care must be taken that the correct planes of dissection are adhered to. There is a thin layer of connective tissue extending from the renal capsule into the fat in the renal hilum and then onto the renal pelvis. This closes off the renal hilum, and it is this layer that must be incised in order to gain access to the infundibula in carrying out an extended pyelolithotomy. Once this layer of connective tissue has been incised, the dissection is continued by inserting specially designed retractors under the parenchyma. The dissection is carried out by inserting and spreading a fine scissors or by the use of a Küttner dissector. This dissection is carried out between the fatty layer in the hilum and the pelvis itself ( Fig. 18-3). If, mistakenly, the surgeon enters the layer between the fat and the parenchyma, considerable hemorrhage can be encountered because of many venous channels in this area. Even if there is perihilar inflammation, or if there has been previous surgery, it is possible to develop this plane. Sharp dissection is required, but early insertion of Gil Vernet retractors moves the vessels in the hilum out of the way so that damage to important structures is avoided. Even if veins in this area are opened, they can be compressed by the insertion of small sponges between the retractors and the hilum, and bleeding kept to a minimum. The subparenchymal dissection can be extended to the infundibula without damaging the superior or inferior apical branches of the renal artery. An incision is then made with a scalpel into the renal pelvis, directly down onto the main bulk of the stone. It is extended in a curved fashion with angled scissors into the necks of the superior and inferior calyces. Alternatively, a straight incision is made in the parenchyma from side to side, and perpendicular extensions are made into the necks of the individual calyces ( Fig. 18-6).

FIG. 18-6. Alternative extended incisions of the renal pelvis.

In general, the large central bulk of the stone is removed first. The best way to do this is to pass a stone dissector around the stone and lever it out of the pelvis. This is preferable to grasping the stone with a forceps, as the stone may break. Once the main fragment is out, fine Turner–Warwick stone forceps, either straight or curved, can be inserted into the calyces, and individual fragments can be removed. After the surgeon feels that all of the stone has been removed, it is advisable to irrigate the renal pelvis and flush smaller fragments out of the calyces. This is done by inserting a wide-bore tube into the calyces, through which a high pressure jet of saline can be passed; the high flow of the saline is essential for effectiveness, and this is best induced by a pressure cuff around a bag of saline attached to an infusion cannula. Contact radiography should then be performed by putting a kidney film behind the kidney. It is helpful to put the kidney into an elastic net sling and then tie the sling to the retractor or to a gantry on the retractor. Ligaclips can then be clipped onto the sling, thereby facilitating the location of even small residual fragments on the x-ray film. The renal pelvis is then closed by using a continuous 4-0 polyglycolic acid or chromic catgut suture. Sometimes it may be easier to put one or two interrupted sutures in the apical parts of the necks of the infundibula, and this will facilitate closure. Again, the closure may not be watertight, and drainage of the area is thus required. Unless there is noticeable intrarenal hemorrhage, no nephrostomy tube is necessary. Additional Nephrotomies On some occasions, it may not be possible to remove the entire stone through a pyelotomy, and additional transparenchymal access is required; the anatrophic nephrolithotomy 1 would be excessive if the renal pelvis has been opened in the manner described above, and multiple radial paravascular nephrotomies 9 are a relatively atraumatic method of access. The method of doing this is to make a small (1 cm) radial incision over the stone, which can be localized either by palpation with a needle through the parenchyma or by intraoperative ultrasonography. The parenchyma is then separated by spreading with two MacDonald's stone dissectors until the calyx is opened and the stone removed. This can be done in a number of positions with a minimal effect on renal function. 2 The nephrotomies are closed with a continuous 4-0 polyglycolic acid or chromic catgut suture, which is placed superficially, incorporating only capsule and a thin layer of parenchyma. A nephrostomy tube should be placed into the most dependent calyx opened; a 12-Fr whistle-tip catheter is satisfactory. This should be brought out through a separate stab incision in the skin. If a radial paravascular incision is to be made, the renal artery must be located and either a silastic loop or a cotton tape passed around it in order to gain control. A single nephrotomy may not require vascular occlusion, and occlusion can be avoided altogether by the use of the Doppler ultrasound, which is especially valuable in kidneys with decreased function and thinned parenchyma. 4 If the renal artery must be occluded, renal function should be preserved during the period of ischemia. Renal hypothermia is achieved by surface cooling with sterile crushed ice 6 or by external cooling coils. 8 A less complex method of protecting against renal ischemic damage is the use of intravenous inosine. This can be injected into a peripheral vein and is valuable in protecting renal function, particularly if the ischemic period is less than 60 minutes and overall preoperative renal function is good. 3 Wound Closure After complete stone removal and adequate hemostasis are ensured, the wound is closed. A Robinson drain is brought out through a separate stab incision. A gravity drainage system such as this is preferable to a suction drain, which may cause a urinary fistula to develop. The wound is closed using a series of interrupted #1 polyglycolic acid sutures. These are passed through all muscular layers at 2-mm intervals and are left untied until all are placed. The table is then unbroken, which brings the wound edges closer together and allows the sutures to be tied without tension. A continuous layer of #1 polyglycolic acid suture is then passed through the outer layer of the external oblique muscle and the fascial layers. The skin can be closed with 3-0 monofilament nylon or with skin clips.

OUTCOMES
Complications Complications from open renal stone surgery are significant and include hemorrhage, urinary fistula, recurrent stones, and actual or functional loss of the renal unit. The risk of these complications is dependent on the associated findings of chronic infection, prior surgery, and surgeon's expertise. There is a small chance that the pleura may be opened when a costal or intercostal incision is made, and the probability of this increases the higher the incision is made. A pleurotomy is readily identified by hearing the sound of air being sucked into the thorax and by seeing the lung on inspiration. The diaphragm should be dissected free from the ribs and used to strengthen the closure of the pleura, which is itself too thin and fragile to hold a suture. The 3-0 chromic catgut suture should include the diaphragm, pleura, and intercostal muscles, and the anesthesiologist should inflate the lung before the last suture is put in. This usually prevents a pneumothorax. A postoperative chest x-ray must be performed, and, in the relatively uncommon event of a persistent pneumothorax, a chest tube should be inserted. Results Stone-free rates in patients undergoing pyelolithotomy are variable, depending on the number of stones, the composition of the stone, and the presence of calyceal stones or obstruction. Solitary stones have virtually a 100% stone-free rate, whereas staghorn stones (stru-vite) or patients with multiple stones scattered among the calyces may have an incidence of retained stones of 10% or more. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Boyce WH, Elkins IB. Reconstructive renal surgery following anatrophic nephrolithotomy: follow-up of 100 consecutive cases. J Urol 1974;111:307. Fitzpatrick JM, Sleight MW, Braack A, Marberger M, Wickham JEA. Intrarenal access: effects on renal function and morphology. Br J Urol 1980;52:409. Fitzpatrick JM, Wallace DMA, Whitfield HN, Watkinson LE, Fernando AR, Wickham JEA. Inosine in ischaemic renal surgery: long-term follow-up. Br J Urol 1981;53:524. Fitzpatrick JM, Murphy DM, Gorey TF, Alken P, Thuroff J. Doppler localization of intrarenal vessels: an experimental study. Br J Urol 1984;56:614. Gil Vernet JM. New surgical concepts in removing renal calculi. Urol Int 1965;20:255. Graves FT. Renal hypothermia: an aid to partial nephrectomy. Br J Surg 1968;50:362. Marshall VF, Lavengood RW Jr, Kelly DG. Complete longitudinal nephrolithotomy and the Shorr regimen in the management of staghorn calculi. Ann Surg 1965;162:366. Wickham JEA. A simple method for regional renal hypothermia. J Urol 1968;99:246. Wickham JEA, Coe N, Ward JP. One hundred cases of nephrolithotomy under hypothermia. J Urol 1974;112:702.

Chapter 19 Ureterolithotomy Glenn’s Urologic Surgery

Chapter 19 Ureterolithotomy
Michael Marberger

M. Marberger: Department of Urology, University of Vienna, Vienna A-1090, Austria.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Anterior Supracostal Approach Posterior Lumbotomy Suprainguinal Approach Ureterolithotomy Outcomes Complications Chapter References

For centuries, “cutting for stone” was synonymous with urology, and just over a decade ago it still made up at least one-fourth of the surgical activity in the field. The development of extracorporeal shockwave lithotripsy (SWL) and endoscopic stone surgery shattered this tradition, and the change becomes most obvious in the indications for ureterolithotomy. Once one of the most common procedures in urology, it all but vanished in the last years in spite of the fact that almost 50% of all patients with upper tract urolithiasis coming to treatment today have stones impacted in the ureter. 8 Specifically, the development of ultrathin semirigid and flexible ureteroscopes with effective laser, electrohydraulic, or ballistic lithotripsy, 1,3,5,8 laparoscopic ureterolithotomy, and third-generation lithotriptors with ultrasonic and fluoroscopic stone localization and small focal zones, which can be pinpointed onto ureteric stones even in infants, 9 have closed the last gaps in the spectrum of minimally invasive therapy of ureteric calculi.

DIAGNOSIS
With less invasive methods of stone removal, a sudden change of the position of the calculus can be met without major problems, even when noticed only during the intervention. In open stone surgery, this could result in a catastrophe, with failure to remove the stone and the need for further procedures. The time-honored rule of precise delineation of the size, number, and shape of all calculi and their topography within the collecting system before an incisional procedure remains as valid as ever. In general, ureteric stones can be located precisely with a good intravenous pyelogram with appropriate oblique, delayed, and postvoiding films. To differentiate radiolucent stones from tumors, clots, or papillae, a nonenhanced abdominal computed tomography may be helpful, but significant stones may be missed with this technique, even when 5-mm cuts are obtained at the level of interest. Retrograde ureterography and, if needed, diagnostic ureteroscopy immediately before surgery will clarify the situation. Urinary infection should always be treated with appropriate antibiotics before surgery. With severe obstruction and any evidence of infection, it is prudent to first drain the kidney by percutaneous nephrostomy for about 48 hours until any pathogen is cultured and adequately treated. A plain abdominal roentgenogram is always obtained immediately before surgery, before anesthesia is initiated. Even the largest calculus seemingly incapable of changing its position may do so, and this may necessitate a completely different surgical strategy.

INDICATIONS FOR SURGERY
In general, ureterolithotomy today becomes necessary only where ESWL or endoscopic techniques fail. Usually, these failures are concomitant with a complication of previous therapeutic interventions, in particular endoscopic manipulation. Urinary extravasation, an impacted ureteral basket, ureteral avulsion, and an obstructing stone are the typical scenarios. At the author's institution, incisional surgery was required in only six of 3,123 patients subjected to a therapeutic intervention to remove ureteric stones in a 7-year period. Two patients had suffered ureteric avulsion, one patient had a basket trapped around a stone, in two patients stones could not be reached endoscopically, and one patient, pregnant in the fourth week of gestation, required rapid removal of a very large stone impacted in the lumbar ureter. Ureteric reconstruction is beyond the scope of this chapter (see Chapter 20), but the latter three patients demonstrate that there is still an occasional, anecdotal need for ureterolithotomy. Stones can of course also be trapped above congenital or acquired ureteric strictures. Where these require surgical correction, the stone is removed at the time of reconstructive surgery, but the underlying pathology dictates the surgical strategy and technique.

ALTERNATIVE THERAPY
Alternatives to open ureterolithotomy are observation, which is indicated in small (<5 mm) stones with no signs of sepsis or extravasation, or one of many forms of minimally invasive surgeries including SWL, endoscopic extraction (via cystoscope, nephroscope, or ureteroscope), percutaneous stone surgery including laparoscopic removal, endoscopic destruction of the stone, or, in certain stones such as struvite or uric acid, chemolysis.

SURGICAL TECHNIQUE
In the difficult situations in which ureterolithotomy is still indicated today, ample exposure is usually needed. Many of the minimally muscle-splitting incisions designed for specific stone situations in the past, such as the Foley incision through the lumbar triangle for high ureteral stones, the gridiron incision for midureteral stones, and the transvesical or transvaginal approach for intramural stones, have become obsolete. They provide only limited access to a small segment of the ureter and should be avoided in difficult situations, especially if the surgeon has limited experience with them. The entire proximal half of the ureter is best approached by a modified 12th-rib supracostal incision, which is carried anterior to the tip of the rib (anterior supracostal incision). Large, firmly embedded stones in the area of the ureteropelvic junction can be removed with minimal morbidity through a posterior lumbotomy. The distal half of the ureter is best reached by a suprainguinal extraperitoneal access. Anterior Supracostal Approach The patient is placed in a lateral jackknife position ( Fig. 19-1). The skin incision runs parallel to the upper margin of the 12th rib and distally in the line of the rib. Its length depends on the precise nature of the procedure. For a standard ureterolithotomy in the lumbar ureter, an incision along the distal half of the rib extending 5 to 7 cm into the abdominal muscles suffices. It can be extended anteriorly as required to reach lower stones and posteriorly so that the entire kidney can be mobilized if necessary.

FIG. 19-1. Position of the patient for anterior 12th-rib supracostal incision. (Reprinted with permission from Marberger M, Fitzpatrick JM, Jenkins AD, Pak CLYC. Stone surgery. Edinburgh: Churchill Livingstone, 1991.)

After the subcutaneous fat has been divided, the fibers of the abdominal musculature are incised with cutting diathermy immediately beyond the tip of the 12th rib ( Fig. 19-2). The transversus abdominis muscle blends here with the deep leaf of the thoracolumbar fascia and should be divided along the same line. The second and third fingers of each hand are now used to sweep the peritoneum off the underside of the abdominal wall muscles before the muscle incision is extended medially as required. The incision should be kept strictly in line with the extension of the 12th rib so as to keep well clear of subcostal vessels and nerves. Once it has been carried as far medially as needed, dissection can proceed in the opposite direction along the 12th rib. The latissimus dorsi and intercostal muscles are divided by diathermy moving backward along the upper margin of the rib. As the rib is progressively mobilized, the insertion of the diaphragm and the pleural reflection come into view. The subcostal nerve is carefully preserved as the diaphragm is divided flush with its insertion to the abdominal wall. The pleura is pushed away by blunt finger dissection. Depending on the degree of exposure needed, dissection of the 12th rib may proceed up to the vertebral column. After division of the costovertebral ligament, the 12th rib can be swung outward like a door. A rib retractor or modified Wickham ring retractor permits excellent exposure. The peritoneum is retracted medially, and the ureter is exposed in the retroperitoneal space below the lower pole of the kidney, where it already lies outside of Gerota's fascia. If the stone lies higher, Gerota's fascia is incised, and the ureter is followed upward to the stone, tilting the kidney anteriorly.

FIG. 19-2. Anterior supracostal incision. The latissimus dorsi is first incised along the upper rim of the 12th rib (A). The external, internal, and transversus abdominis muscles are divided in extension of the 12th rib (B) as far medially as needed (C). (Reprinted with permission from Marberger M, Stackl W. Surgical treatment of renal calculi. In: Schneider HJ, ed. Urolithiasis. Heidelberg: Springer, 1986;107.)

Posterior Lumbotomy The proximal third of the ureter (and renal pelvis) can be reached with minimal muscle trauma through the thoracolumbar fascia lateral to the sacrospinalis and quadratus lumborum muscles. In terms of postoperative pain and morbidity, this incision is superior to all other lumbotomies. The 12th rib above and the iliac crest below limit exposure of the kidney and midureter. Therefore, the ideal stone for this approach should be one that is firmly embedded in the upper third of the ureter or at the ureteropelvic junction. The patient is placed in the lateral recumbent position, with approximately 15-degree anterior rotation, and the table is flexed at the tip of the 12th rib. Simultaneous bilateral surgery may be performed with the patient prone. 10 The most commonly used access6 utilizes an oblique skin incision parallel and 3 cm lateral to the erector trunci, from the 12th rib down to the iliac crest ( Fig. 19-3). Fat and subcutaneous tissue are divided until the lateral fibers of the latissimus dorsi are exposed. The muscle is split to expose the subjacent 12th rib. The posterior leaf of the lumbodorsal fascia is divided in the line of the skin incision, and the lateral margin of the sacrospinalis muscle so exposed is retracted medially. The middle layer of the thoracolumbar fascia is then seen and incised somewhat lateral to the fleshy belly of the sacrospinalis muscle. The lateral border of the quadratus lumborum now comes into view and may be retracted with a hook toward the vertebral column. The deep layer of the thoracolumbar fascia is exposed and opened, care being exercised to spare the twelfth subcostal nerve and the iliohypogastric nerve coursing obliquely and laterally on its deep aspect. Gerota's fascia is incised, and the perirenal fat is divided by blunt dissection to expose the renal pelvis. In a modification, 4 the incision runs from a point three fingerbreadths lateral to the dorsal spines to the junction of the middle and anterior third of the iliac crest. The lower parts of the latissimus dorsi and the serratus posterior inferior and the costovertebral ligament are divided, and the middle and deep leaves of the thoracolumbar fascia are split. A Finochiettio rib retractor is inserted to expose the field.

FIG. 19-3. Posterior lumbotomy. (A) Positioning of patient with oblique and vertical incision. (B) Oblique incision: division of latissimus dorsi and posterior inferior serratus muscle. (C) Oblique incision: division of costovertebral ligament. (D) Oblique incision: opening of Gerota's fascia. (Reprinted with permission from Anderson EE. Ureterolithotomy. In: Glenn JF, ed. Urologic surgery, 4th ed. Philadelphia: JB Lippincott, 1993;276–286.)

Suprainguinal Approach The distal half of the ureter is best approached by a suprainguinal extraperitoneal incision. The patient is in a prone position with the ipsilateral flank supported by a cushion. Depending on the exposure needed, the skin is incised in an oblique direction along a line from the pubic tubercle upward to a point about two fingerbreadths anterior to the superior iliac crest ( Fig. 19-4). The external oblique abdominal muscles and the transversalis fascia are divided with cutting diathermy in

the same direction. After ligation and transection of the epigastric vessels, the peritoneal fold is reflected medially to expose the ureter. It can be identified without problems either where it crosses the common iliac artery or where it runs immediately below the obliterated umbilical artery. The latter structure is routinely divided and ligated.

FIG. 19-4. Suprainguinal extraperitoneal approach. (A) Incision. (B) Exposure of ureter following incision of external and internal oblique muscles and splitting of transversalis fascia. (Reprinted with permission from Anderson EE. Ureterolithotomy. In: Glenn JF, ed. Urologic surgery, 4th ed. Philadelphia: JB Lippincott, 1993;276–286.)

With stones in the distal third of the ureter, a urethral catheter should routinely be placed to keep the bladder empty during the procedure. Ureterolithotomy Once the ureter is identified, the stone is located by palpation, carefully avoiding any milking movements that could dislocate it. Without mobilizing the ureter, it is snared with vessel loops just above and below the stone. The ureteral wall is incised with a scalpel directly onto the stone in a longitudinal direction. As soon as the mucosa is opened, the incision is enlarged with angulated scissors so that the stone can be extracted with nerve hooks. The ureter is then probed in both directions with a soft ureteral catheter to ascertain complete stone removal and is irrigated copiously. With the slightest possibility of obstruction, extravasation, or difficult closure, a self-retaining stent is inserted, taking care to position the two ends properly in the bladder and renal pelvis. A standard double-J stent can usually be inserted from the ureterotomy. Because the distal segment of the ureter is, in general, more difficult to negotiate, especially after previous endoscopic maneuvers, it is intubated first. By reversing the stent, i.e., advancing the blunted, closed tip of the stent, which is otherwise advanced up to the kidney, down to the bladder, the ureterovesical junction can usually be passed. The guide wire is then removed, and the stent is advanced further down the ureter until it almost disappears in the ureterotomy. Its upper end can now be straightened and advanced up into the renal pelvis. If the proximal corner of the ureteric incision is elevated with a nerve hook during this procedure, this rarely causes problems. It is important to note the markings on the stent for correct placement. To ascertain that the distal end of the stent is in the bladder, indigo carmine can be administered through the urethral catheter; it should reflux freely into ureter and stent. Intraoperative fluoroscopy offers a more elegant alternative. If any problems are encountered in intubating the ureter, the ureter should be inspected with a thin flexible ureteroscope inserted through the ureterotomy to avoid missing an additional ureteric stone. Any additional stone is either removed through a second ureterotomy or, preferably, by endoscopic lithotripsy. A guide wire is then advanced under endoscopic control down into the bladder, and the stent is inserted over it. Problems in the ureter proximal to the incision are handled in a similar manner, but this segment of the ureter is usually dilated and therefore easier to engage. Whenever self-retaining stents are used, the patient should routinely be subjected to a flexible cystoscopy at the end of the procedure to be certain the vesical end of the stent is indeed in the bladder. The ureterotomy is closed with one to three interrupted sutures of 5-0 chromic catgut. The sutures should grasp only the superficial seromuscular layers to approximate the ureteric wall rather than achieving watertight closure. If placed too tight or too deep, they may compromise ureteric blood supply and promote leakage. Obstructing sutures have a similar effect. Whenever closure is difficult because of scarring, it is safer not to close the ureterotomy at all and to stent the ureter. The site of the ureterotomy should be covered with retroperitoneal fat or an omental flap. 7 Every ureterotomy has to be drained precisely. We routinely use a No. 21 tube drain of silicone rubber with one or two side holes, which is brought out through all layers of the abdominal wall via a separate stab incision lateral to the lower end of the incision. The tip of the drain must be in a dependent position to the ureter-otomy, but not in the immediate vicinity or in contact with the ureter. If the ureter was approached transperitoneally, it should be drained through the retroperitoneum. The wound is closed in layers with absorbable suture material. Postoperatively, patients are mobilized within 24 hours, with analgesics administered generously as needed. Antibiotics are given only with proven infection and according to appropriate sensitivity testing. The patients are well hydrated, with intravenous fluid replacement in the first two postoperative days. Especially after lumbar ureterolithotomy, bowel function may take 2 to 3 days to normalize, and an enema and even cholinergic agents may be needed.

OUTCOMES
Complications The wound drain should not be removed before the fourth or fifth postoperative day. Prolonged discharge of urine from the drain is usually caused by impaired drainage caused by a missed calculus, clot, or ureteral obstruction. Occasionally leakage results from incorrect positioning of the tip of the drain immediately adjacent to the ureterotomy. Careful retraction of the drain by 1 to 2 cm then rapidly dries up the wound. If urine leaks from the drain longer than 5 days, an indwelling ureteric stent should be inserted. Permanent urinary fistulas are extremely rare and, when present, almost always result from obstruction below the level of surgery. Urinary extravasation may cause severe problems if the wound is improperly drained because the drain either was not placed in a dependent position or was removed too early. Urinoma formation is usually heralded by fever and flank pain but may occur inconspicuously. A high degree of suspicion should therefore be directed toward this potential complication. Any unexplained fever, flank pain, or delayed healing should be investigated immediately by ultrasonography, an antegrade pyelogram (if the nephrostomy is still in place), or excretory urography with delayed films. The situation can usually be corrected by draining the kidney with an indwelling stent or a nephrostomy and by draining the urinoma percutaneously. In the pre-SWL era, the most frustrating complication of any stone operation was the retained calculus. Although the availability of SWL should still not be an excuse for a less careful attempt at complete stone removal once open surgery is decided on, retained stones can be treated in this manner highly successfully some days after the operation. Likewise, if a calculus below the level of surgery was overseen and resulted in obstruction and/or extravasation, it can be treated endoscopically or by SWL, just as any other ureteral stone in the immediate postoperative period. CHAPTER REFERENCES
1. Adams JB. Ureteral surgery. In: Badlani G, Bagley DH, Clayman RV et al. (Eds.), Smith's textbook of endourology. St. Louis: Quality Medical Publishers, 1996;962–976. 2. Anderson EE. Ureterolithotomy. In: Glenn JF, ed. Urologic surgery, 4th ed. Philadelphia: JB Lippincott, 1993;276–286. 3. Conlin MJ, Marberger M, Bagley D. Ureteroscopes and working instruments. In: Badlani G, Bagley DH, Clayman RV et al. (Eds.), Smith's textbook of endourology. St. Louis: Quality Medical Publishers, 1996;370–387. 4. Gil Vernet J. New surgical concepts in removing renal calculi. Urol Int 1965;20:255–288. 5. Hofbauer J. Electrohydraulic versus pneumatic disintegration in the treatment of ureteral stones: a randomized, prospective trial. J Urol 1995;153:623–625. 6. Lutzeyer W. Lumbodorsal exploration. In: Glenn JF, ed. Urologic surgery, 2nd ed. Philadelphia: JB Lippincott, 1975;127–133.

7. 8. 9. 10.

Marberger M, Fitzpatrick JM, Jenkins AD, Pak CLYC. Stone surgery. Edinburgh: Churchill Livingstone, 1991. Marberger M, Hofbauer J, Türk C, Höbarth K, Albrecht W. Management of ureteric stones. Eur Urol 1994;25:265–272. Marberger M, Türk C, Steinkogler I. Piezoelectric extracorporeal shock wave lithotripsy in children. J Urol 1989;142:349–352. Marberger M, Stackl W. Surgical treatment of renal calculi. In: Schneider HJ, ed. Urolithiasis. Heidelberg: Springer, 1986;107.

Chapter 20 Ureteral Reconstruction Glenn’s Urologic Surgery

Chapter 20 Ureteral Reconstruction
Bernd J. Schmitz-Dräger and Rolf Ackermann

B. J. Schmitz-Dräger and R. Ackermann: Department of Urology, Heinrich-Heine-University, D-40225 Düsseldorf, Germany.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preparation Position and Incision Pyeloileal Anastomosis or Ileoureteral Anastomosis Ileocystostomy Outcomes Complications Results Chapter References

Partial or complete ureteral replacement remains a challenge for the urologist. The maintenance of kidney function is the primary goal of this procedure. The crucial problem with ureteral replacement is that in most cases several surgical procedures and frequently radiotherapy have been performed beforehand, and kidney function is usually already impaired at the time of intervention. Therefore, standardized recommendations may not be applicable, and the approach must be tailored according to the individual situation. Several ureteral substitutes have been proposed, including blood vessels, fallopian tubes, and the appendix. Nevertheless, all these tissues or organs and the corresponding procedures are correlated with a significant morbidity, including recurrent infection, stone, or stricture formation, with a worsening of kidney function in many patients. Because larger series have rarely been reported, the results and morbidity of the various procedures differ significantly.

DIAGNOSIS
Because partial or complete ureteral replacement with small bowel is always a secondary treatment, every attempt should be made to preserve as much ureteral length as possible. Antegrade and retrograde pyelography will usually provide sufficient information regarding the length and degree of the stricture, but in selected cases ureteroscopy may provide valuable information. In patients treated for malignant disease, tumor recurrence must be excluded by CAT scan or MRI. Kidney function needs to be examined by preoperative isotope scintigraphy to ensure it is appropriate because with severely decreased function, the results of ureteral reconstruction are poor. Furthermore, preoperative isotope scintigraphy provides a baseline for follow-up examinations after surgery.

INDICATIONS FOR SURGERY
Partial or complete ureteral replacement with small bowel is indicated as a second-line treatment after failure of ureter-sparing surgery.

ALTERNATIVE THERAPY
Every attempt should be made to maintain the ureter. Depending on the length and location of the stricture, the preoperative evaluation should demonstrate whether ureter-sparing approaches are viable alternatives to ureteral replacement ( Table 20-1). Because of their relatively low morbidity, endoscopic procedures should be considered first. A success rate of only 50% to 70% with endoscopy necessitates further measures in a number of patients. 2,4,6

TABLE 20-1. “Ureter-sparing” treatment options for ureteral strictures

Autotransplantation carries several disadvantages and risks, especially after previous surgery and/or radiotherapy and impaired renal function. Although no prospective study addressing this question has been conducted, ureteral replacement with small bowel seems preferable in this patient population.

SURGICAL TECHNIQUE
Preparation Bladder outlet obstruction is a contraindication for ureteral replacement with small bowel and, therefore, must be excluded. If necessary, urodynamic examination of the lower urinary tract should be performed. In patients with obstructive prostatic hyperplasia, prostatectomy should be performed before ureteral replacement. Preoperative bowel preparation should be performed as in patients undergoing bowel surgery for other urologic procedures (see Chapter 79). A nephrostomy tube is already in place in most patients or should be inserted during surgery. Position and Incision If both sides are affected, a midline incision or a right paramedian incision is recommended. If only one ureter is to be replaced, a lateral flank position (twist position) with the table flexed and the chest positioned at about a 60-degree angle to the table is preferable. Allow the pelvis to fall back. The incision starts between the 11th and 12th ribs, continues semiobliquely nearly to the midline, and ends as a paramedian incision to the os pubis. The peritoneum is opened, and the small bowel packed away. For a right ureteral replacement, mobilize the cecum and divide the lateral attachments of the ascending colon ( Fig. 20-1). Mobilize the ascending colon as for an extensive retroperitoneal lymph node dissection (see Chapter 63). Mobilize the peritoneum from the bladder dome and the lateral aspect of the bladder. Carefully determine the length of intestine required for ureteral replacement.

FIG. 20-1. Mobilization of cecum and ascending colon. Selection of ileal segment.

Select an appropriate segment from the preterminal ileum ( Fig. 20-1). Consider adequate vascularization of the chosen segment (see ileal conduit, Chapter 77). It is mandatory to select the sites of transection to permit a dissection deep enough for the proximal end to reach the renal pelvis and the laborial or distal end to reach the bladder. The bowel is divided, and the continuity restored, as described for the ileal conduit. The mesenteric defect is then closed with 3-0 vicryl to prevent internal herniation. The excised bowel segment is irrigated with saline solution until the effluent is clear. The mesentery of the ascending colon is incised depending on the location of the mesentery of the ileal segment, and the ileum is passed into the retroperitoneal space. The ileal segment is then rotated to place the distal end near the bladder and the proximal end close to the renal pelvis or the ureter. The defect in the colonic mesentery is closed using 3-0 vicryl sutures. In this closure, it is important to avoid compression of the ileal mesenteric vessels. Pyeloileal Anastomosis or Ileoureteral Anastomosis In cases of partial ureteral replacement, the proximal opening of the ileal segment is closed with a running 3-0 chromic catgut suture. Before the anastomosis, the ureter is stented with a 6- to 9-Fr catheter held in place with a 4-0 catgut suture. The ileoureteral anastomosis of the spatulated ureter is performed end to side with a single-layer technique with either a running suture or interrupted sutures of 4-0 or 5-0 vicryl sutures. For complete ureteral replacement, the proximal opening of the ileal segment is brought to the renal pelvis. The renal pelvis is opened widely to permit end-to-end anastomosis to the ileum. In case of a small renal pelvis, it may become necessary to taper the ileum by closing the proximal opening of the ileum partially on the antimesenteric side. The pyeloileal anastomosis is performed in a single layer with either a running suture or interrupted sutures of 3-0 or 4-0 chromic catgut ( Fig. 20-2). Because a nephrostomy tube is already in place, it is not necessary to stent the ileal ureter.

FIG. 20-2. Pyeloileal anastomosis.

Ileocystostomy We prefer to perform ileocystostomy on the posterior bladder wall about 1 to 2 cm craniolaterally to the native ureteral orifice to avoid extensive angulation and possible obstruction of the ileum during bladder filling. The anastomosis is performed in a double-layered technique with a running mucosa-to-mucosa suture (4-0 vicryl) and interrupted seromuscular–detrusor muscle sutures of 3-0 vicryl ( Fig. 20-3). Bilateral ureteral replacement can be accomplished by modifying the above technique as shown in Fig. 20-4. Except for the fixation of the mutual stent, vicryl may be replaced by chromic catgut.

FIG. 20-3. Ileocystostomy.

FIG. 20-4. Complete bilateral ureteral replacement with ileum.

OUTCOMES
Most reports on ureteral replacement with ileum are case reports including only a few patients. Because no larger contemporary series are available for review, the assessment of the outcome of this procedure is difficult. Complications No data are available regarding perioperative complications of ureteral replacement with ileum. It is assumed that the complications are similar to those observed after surgery for an ileal neobladder. In patients with partial replacement of the ureter, strictures at the ureteroileal anastomosis presumably occur at a similar frequency as in patients after an ileal conduit. Ileoureteral reflux is observed in 50% to 85% of the patients depending on whether the bladder is filling or emptying. 3,6 The significance of this reflux is unknown. Hyperchloremic metabolic acidosis requiring treatment should be anticipated in approximately 50% of patients. 6 Careful follow-up examination including routine measurements of base excess, serum bicarbonate, and pH are mandatory. This group of patients is certainly prone to urinary tract infections (UTI). In 30% to 100% of the patients, UTI will occur. treatment in cases of proven UTI are required. Results The results of ureteral replacement with ileum are difficult to assess because this form of surgery is not standardized, and patient selection varies considerably between the different series. 1,2,4,6,7 Furthermore, the objective goals of the procedure are not clearly defined. The values of BUN or serum creatinine that have been reported in some series are probably not sufficient to define the outcome in patients with bilateral kidneys. Dilation of the upper urinary tract is another parameter used in the literature. However, it is difficult to discriminate between persistent dilation despite reduction of ureteral obstruction and those cases in which dilation persists as a result of obstruction and/or reflux after surgery. To date, no reports including diuretic isotope scintigraphy have been published. In the few studies with long-term results, a favorable outcome has been reported in up to 85% of the cases. This does not include patients with impaired renal function with serum creatinine levels greater than 2.0mg/dl. In this population, fewer than 50% will benefit from ureteral replacement. In general, to avoid metabolic problems, the length of the ileal segment should be as short as possible. Hinman and Oppenheimer, however, have shown in the dog that an ileal segment greater than 18 cm will block the transmission of 20 to 30 cm H 2O pressure. 5 These experiments also form the basis for the introduction of a nontubularized ileal segment in modifications of the neobladder. 8 Clinical reports on ureteral replacement with ileum apparently do not support this experimental observation, however, because cystoileal reflux and/or ileal–ureteral reflux can be observed in some patients at an intravesical pressure of only 3 to 8 cm H 2O.6 It is questionable whether a pressure of less than 20 cm H 2O can lead to damage to the upper urinary tract. So far, no clinical data including simultaneous measurement of intravesical and intrapelvic pressures are available. In addition, long-term results seen after ureteral replacement with ileum and experiences with the intestinal neobladder suggest that some protection of the upper urinary tract may be afforded by the ileal segment. We therefore prefer to use a bowel segment at least 15 cm in length. In summary, ureteral replacement with ileum is a feasible technique that carries considerable perioperative and long-term morbidity and should therefore be considered only as a second-line treatment in selected patients. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. Baum W. The clinical use of terminal ileum as a substitute ureter. J Urol 1954;72:16. Boxer RJ, Fritzsche P, Skinner DG, et al. Replacement of the ureter by small intestine: Clinical application and results of the ileal ureter in 89 patients. J Urol 1979;121:728. Fritzsche P, Skinner DG, Craven JD, Cahill P, Goodwin WE. Long-term radiographic changes of the kidney following the ileal ureter operation. J Urol 1975;114:843. Goodwin WE, Winter CC, Turner RD. Replacement of the ureter by small intestine: Clinical application and results of the ileal ureter. J Urol 1959;81:406. Hinman F Jr, Oppenheimer R. Functional characteristics of the ileum as a valve. J Urol 1958;80:448. Prout GR Jr, Stuart WT, Witus WS. Utilization of ileal segments to substitute for extensive ureteral loss. J Urol 1963;90:541. Skinner DG, Goodwin WE. Indications for the use of intestinal segments in management of nephrocalcinosis. J Urol 1975;113:436. Studer UE, Gerber E, Springer J, Zingg EJ. Bladder reconstruction with bowel after radical cystectomy. World J Urol 1992;10:I-1.
2,6

Regular urinalysis and appropriate antibiotic

Chapter 21 Ureteral Stricture Glenn’s Urologic Surgery

Chapter 21 Ureteral Stricture
Glenn S. Gerber

G. S. Gerber: Department of Surgery, The University of Chicago Hospitals, Chicago, Illinois 60637.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Ureteroureterostomy Psoas Hitch Procedure Boari Flap Procedure Outcomes Complications Results Chapter References

Ureteral strictures may result from a variety of causes, including stone passage, endoscopic urologic procedures, radiation therapy, open or laparoscopic surgery, and penetrating traumatic injuries. 3 The incidence of ureteral strictures has increased in recent years, largely as a result of the introduction and widespread use of upper urinary tract endoscopy. Several factors may contribute to the development of ureteral strictures following ureteroscopy, which may be seen in as many as 3% to 11% of patients.2,6 These include stone impaction, relative ischemia, often because of the use of larger instruments for prolonged intervals, ureteral injury with extravasation of urine, and direct mechanical or thermal trauma. 3 Ureteral injury leading to stricture, fistula formation, or obstruction may also be seen following a variety of open surgical procedures, including abdominal and vaginal hysterectomy, repair of vascular lesions, and pelvic exploration in patients with prostate, colon, and rectal malignancies. Although the need for surgical repair of a ureteral injury may be immediately evident in some cases, many patients with ureteral obstruction and/or fistulas do not present until weeks or months following surgery. In particular, ureteral strictures that result from endoscopic manipulation of the upper urinary tract may not be noted for prolonged intervals because of the slow development of ureteral fibrosis. The presentation of patients with ureteral strictures is variable and may range from acute flank pain with sepsis and pyelonephritis to the incidental finding of hydronephrosis in an asymptomatic individual. As a result, the initial evaluation and treatment of patients with suspected ureteral obstruction must be tailored to the clinical situation.

DIAGNOSIS
A variety of diagnostic procedures are available to evaluate patients suspected to have ureteral strictures. Those patients presenting with acute pain and/or upper urinary tract infection will likely require immediate decompression of the obstructed kidney by internal (ureteral stent) or external (percutaneous nephrostomy) drainage. In these cases, immediate or delayed radiographic imaging of the affected renal unit may be performed in an antegrade or retrograde fashion. Those patients not requiring immediate intervention may be evaluated by intravenous pyelography (IVP), renal scintigraphy or computerized tomographic (CT) scanning. Advantages of an IVP include the ability to assess renal function as well as the level and degree of obstruction. Nuclear studies provide limited anatomic detail but allow for a more precise quantification of renal function and drainage. CT scans provide an assessment of perirenal and periureteral structures, which may be important in some patients with ureteral strictures.

INDICATIONS FOR SURGERY
The indications for surgical management of ureteral strictures, disruptions, and fistulas are dependent on the etiology of the lesion and the clinical situation. Ureteral disruption that results from external violence will generally require immediate repair. Similarly, iatrogenic injuries to the ureter noted during open pelvic or abdominal surgery should be repaired without delay. In contrast, initial attempts at conservative or endoscopic management of patients with ureteral injuries detected several days or weeks following surgery may be appropriate in some cases. Ureteral strictures that result from upper urinary tract endoscopy may be managed successfully with endoscopic dilation and/or incision (endoureterotomy) , with open surgical repair reserved for patients with recurrent obstruction.

ALTERNATIVE THERAPY
Many patients with ureteral strictures can be successfully managed using a variety of endourologic methods. Balloon dilation via a percutaneous, antegrade, or transurethral, retrograde approach is generally simple to perform with little risk of significant morbidity. Alternatively, full-thickness endoscopic incision of the stricture (endoureterotomy) followed by stent placement may have a higher long-term success rate than dilation but entails a greater risk of vascular and/or intestinal injury. In general, endourologic techniques appear to be most successful in patients with short, nonirradiated strictures of the distal ureter. Overall, an initial attempt at endoscopic management is appropriate in most patients with ureteral strictures because the potential morbidity and the recovery period are generally less with these procedures. In addition, a failed endoscopic procedure does not appear to jeopardize the success of subsequent open surgical repairs. Finally, some patients with complex ureteral strictures and/or disruption may be best managed by nephrectomy or cutaneous ureterostomy. This approach may be particulary appropriate in those with significant comorbid disease or patients in whom a urine leak could be disastrous, such as those with vascular grafts in the area of the ureteral injury.

SURGICAL TECHNIQUE
Ureteroureterostomy This procedure is most appropriate in patients with short ureteral strictures involving the abdominal ureter (superior to the bifurcation of the iliac vessels). In addition, direct anastomosis of the ureter can be performed in patients with intraoperative injuries and those with ureteral damage secondary to external violence. If ureteroureterostomy is to be performed as a separate procedure, a flank, anterolateral, or midline approach is appropriate based on the level of the ureteral injury. In general, an incision that allows for renal mobilization is helpful in case additional ureteral length is necessary. Initially, the ureter is identified through either an extraperitoneal or intraperitoneal approach. If the ureter is difficult to locate, it can be reliably found as it crosses the bifurcation of the common iliac artery. In patients who have undergone previous surgery or those with severe ureteral distortion, the ureter may be confused with other structures such as the gonadal vein. Careful dissection and aspiration with a small needle may be helpful in such cases. Once identified, the ureter is freed by blunt and sharp dissection with abundant soft tissue left surrounding the structure. It is essential to avoid damage to the ureteral adventitia because this may lead to ischemia and poor healing of the anastomosis. In cases in which the ureter has been surgically transected or completely divided by a penetrating knife or gunshot wound, both ends of the ureter must be identified. In patients with ureteral strictures, the narrowed segment is excised after the ureter has been well mobilized both proximally and distally. It is important to remove all damaged tissue and to perform the anastomosis with healthy, well-vascularized ureter. After removal of the strictured area, stay sutures of 4-0 or 5-0 chromic catgut or absorbable synthetic suture, such as Vicryl or Dexon, are placed on each end of the ureter. It is essential that the anastomosis be performed without tension, and additional ureteral length can be obtained by mobilization of the kidney. The ureteral ends should be transected obliquely to allow for a wide-mouthed anastomosis. In addition, spatulation of each end of the ureter 180 degrees apart over a length of 2 to 4 mm may also facilitate the repair. Alternatively, a fishmouth or Z-plasty anastomosis can be performed (Fig. 21-1), although these are used less commonly. In patients with high-velocity missile injuries to the ureter, extensive debridement of the ureteral ends should be performed because the degree of tissue devitalization is often underestimated at the time of surgery.

FIG. 21-1. The techniques of reanastomosis include (A) oblique, (B) Z-plasty, and (C) fishmouth.

Once the ends of the ureter have been prepared, two sutures of 4-0 or 5-0 Vicryl or Dexon are placed through the apex of the spatulated area on one end and out the middle of the nonspatulated area on the other end. The knots should lie outside the ureteral lumen. These two initial sutures should be located 180 degrees apart and are held to facilitate the completion of the anastomosis. Running or interrupted sutures are placed approximately 2 mm apart on one side of the anastomosis, and the holding sutures are then rotated to help expose the opposite side. A double-pigtail stent may then be placed by initially passing the guide wire into the bladder and then threading the stent over the wire. The guide wire is then removed. In order to assure proper distal positioning of the stent, have an assistant fill the bladder through the urethral catheter with saline mixed with an ampule of indigo carmine. Visualization of blue-stained solution at the level of the anastomosis assures that the stent is in the bladder, except in unusual cases of preexistent vesicoureteral reflux. The guide wire is then passed proximally into the renal pelvis, and the stent is threaded over the opposite end of the wire, which is brought out through a side hole in the stent. The stent is then passed proximally until it is straightened in the ureter, and the wire is removed. The anastomosis is completed with interrupted or running sutures. If the repair is tenuous or in patients who have previously received radiation therapy, the ureteral anastomosis may be wrapped with omentum to facilitate healing. This may also be helpful in patients with associated injuries to the bowel or pancreas or those in whom a vascular graft has also been placed. A closed suction drain is placed lateral to the anastomosis and brought out through a separate stab incision before closure of the abdomen. The stent may be left in place for 2 to 4 weeks. Psoas Hitch Procedure The psoas hitch procedure is the simplest method for substitution of the lower third of the ureter. 5 With this technique, the bladder can be mobilized in a cephalad direction to a level well above the bifurcation of the iliac vessels. In most cases, adequate mobilization of the bladder can be achieved to avoid the need for a technically more difficult and less often successful Boari flap procedure (see below). The choice of incision is based on surgeon preference, and options include a lower midline, Pfannenstiel, or suprapubic V approach. The ureter is initially identified and carefully dissected at or above the crossing of the common iliac vessels. It is generally best to approach the ureter at a level above the site of obstruction and then proceed distally after encircling the uninvolved ureter with a small Penrose drain or vessel loop. Care is taken to avoid damage to the ureteral blood supply. The ureter is sharply divided at or below the pelvic inlet, based on the extent of ureteral pathology, and a stay suture is placed. The distal ureteral stump is ligated with a 2-0 chromic catgut suture. After the ureter is dissected, the bladder is mobilized to allow it to be hitched to the psoas muscle. The peritoneum is dissected from the dome of the bladder and the space between the rectum and bladder is opened. The superior and middle vesical pedicles are ligated and divided on the contralateral side, leaving the inferior vesical pedicle as the only attachment on the contralateral side of the bladder. The blood supply on the ipsilateral side rarely needs to be divided to provide sufficient upward mobilization of the bladder. The initial incision into the bladder is then made transversely across the middle of the anterior wall at the level of its maximum diameter.5 The bladder should be opened slightly more than halfway around. A common mistake is to open the bladder over too short a distance, which limits the cephalad displacement that can be achieved. The bladder is then brought to the psoas muscle at a point superior and lateral to the bifurcation of the iliac vessels by placing two fingers into the fundus of the bladder ( Fig. 21-2). The ureter can then be gently pulled toward the bladder to determine if an adequate anastomosis without tension can be achieved. If further length is needed, the kidney can be mobilized downward, or the contralateral endopelvic fascia can be opened. Once it is determined that an adequate anastomosis can be performed, the bladder is fixed to the psoas minor tendon or psoas major muscle using three to six 2-0 Vicryl sutures. These sutures should be carefully placed to avoid injury to the genitofemoral nerve. In addition, femoral neuropathy has also been reported after the psoas hitch procedure, and care should be taken not to place the tacking sutures too deeply into the muscle. 1

FIG. 21-2. Psoas hitch. The contralateral superior and midvesical arteries are ligated, and Babcock clamps are attached to stretch bladder to the point of suture fixation. The transverse bladder incision is closed vertically.

Before sutures are tied between the bladder and psoas muscle or tendon, the site of the ureteroneocystostomy should be selected. It is important to anastomose the ureter to an immobile portion of the bladder base so that intermittent obstruction of ureteral drainage does not occur as the bladder fills to varying degrees. The ureter is drawn through the bladder wall by passing a small clamp from inside the bladder and grasping a stay suture that has been placed through the distal ureter. The ureter should not be trimmed until it has been brought into the bladder, and it is clear that there is sufficient length to perform a tension-free anastomosis. If there is adequate ureteral length, a tunneled, nonrefluxing anastomosis may be performed. However, a direct, refluxing ureteroneocystostomy is also acceptable in most adults. The anastomosis is performed using interrupted 4-0 or 5-0 Vicryl sutures after the ureter has been spatulated on its anterior surface. At the distal apex of the anastomosis, one to three sutures are placed deeply into the bladder muscle and then through the tip of the ureter. The remaining sutures are taken through bladder mucosa and the ureter. The ureteral adventitia is loosely attached to the bladder wall where it exits using two or three 4-0 Vicryl sutures placed longitudinally. An 8-Fr infant feeding tube or double-pigtail stent is then placed into the renal pelvis. The feeding tube is generally preferred because it can be brought out the anterior bladder and body wall through a stab incision. This allows for direct monitoring of drainage; the tube can be irrigated if it becomes obstructed, and the need for stent removal by cystoscopy is avoided. The feeding tube is loosely tied to the bladder mucosa adjacent to the anastomosis using a 5-0 chromic suture to avoid inadvertent displacement during the perioperative period. A suprapubic tube can be placed if desired, although a urethral catheter is generally adequate. The bladder is closed vertically in watertight fashion using two layers of 2-0 and 3-0 chromic catgut. A Penrose or closed suction drain is positioned laterally in the perivesical area. A cystogram and ureterogram are performed through the externally draining stent 7 to 10 days postoperatively to assure adequate healing. Boari Flap Procedure A bladder flap operation is rarely needed in patients with distal ureteral strictures and/or fistulas. In most cases, a psoas hitch procedure is adequate to replace the lost or damaged section of ureter. If a Boari flap is necessary, the choice of incision, dissection of the ureter, and initial bladder mobilization are identical to that described above for patients undergoing a psoas hitch ureteral reimplantation. When renal and ureteral mobilization, as well as the psoas hitch, are inadequate to

allow a tension-free anastomosis to be performed, a bladder flap may be used. A psoas hitch should accompany the Boari flap to help decrease the length of flap that is needed. 4 Once the bladder has been mobilized and a psoas hitch performed, the site for the base of the flap should be identified on a fixed portion of the bladder, and the length of flap needed should be measured. A stay suture is placed at each end of the base of the flap, which should be approximately 4 cm wide. In order to assure adequate vascularity of the flap, the base should be wider if a longer flap is necessary. A stay suture is placed at each end of the apex of the flap, which should be 3 cm in width. If needed, a longer flap can be created using a spiral incision in the bladder ( Fig. 21-3). The flap should be developed by incising the bladder wall using electrocautery. The flap is then brought up to the ureter, which may be anastomosed to the apex using either a direct or tunneled technique ( Fig. 21-4). A 5- to 8-Fr infant feeding tube or double-pigtail stent should be placed. The bladder is closed, and the flap is rolled into a tube over the stent using a running 3-0 chromic catgut suture on the mucosa and interrupted 2-0 chromic catgut suture on the muscularis and adventitia. A perivesical drain is placed, and radiographic evaluation of the bladder and ureter is performed before the ureteral and urethral catheters are removed.

FIG. 21-3. (A and B) The bladder should be fixed by a psoas hitch, a spiral transverse incision made 4 cm at the base and 3 cm at the apex. (C) The bladder flap is tubed over a catheter and closed in two layers with a 3-0 running mucosal and 2-0 interrupted muscularis suture.

FIG. 21-4. The ureter either is anastomosed directly to the bladder flap using additional sutures to fix the ureter to the external bladder wall and retroperitoneum (A) or is tunneled for 3 to 5 cm to create an antirefluxing anastomosis (B).

OUTCOMES
Complications Prolonged urinary drainage is the most common problem in the early postoperative period in patients undergoing ureteroureterostomy, psoas hitch reimplantation, or the Boari flap procedure. In most cases, the leak will seal as long as adequate drainage of the upper urinary tract and bladder is assured. If there is concern regarding the site of leakage, a cystogram and/or ureterogram may be obtained. If an externally draining ureteral catheter is not present, an intravenous pyelogram may be helpful. Rarely, injury to the contralateral ureter may occur, and this possibility should be considered in patients with unusual or complicated problems. In patients with unexplained fever and/or sepsis, the presence of an undrained urine collection (urinoma) should be considered. Ultrasonography or CT scans may be useful in such cases both in establishing the presence of a fluid collection and guiding percutaneous placement of a drainage catheter. The most significant long-term risk associated with the surgical repair of ureteral strictures is recurrent obstruction. Although some patients may present with flank pain and/or infection, others will remain asymptomatic, presumably because of the slow development of a recurrent stricture. For this reason, all patients should undergo radiographic evaluation of the upper urinary tract 6 to 12 weeks following stent removal and again 6 to 12 months after surgery. Risk factors that increase the likelihood of recurrent obstruction include previous ureteral and bladder surgery, a history of pelvic or lower abdominal irradiation, devascularization of the ureter at the time of surgery, and an anastomosis performed under tension. If detected early, recurrent strictures may respond favorably to balloon dilation and/or endoureterotomy, and these procedures should be considered before resorting to open surgical revision. Results Surgical repair of ureteral strictures is associated with excellent long-term success rates. The most important factors responsible for these results include selection of the most appropriate surgical technique, based on the site and length of stricture. In addition, consideration of preoperative issues, such as a history of irradiation or prior bladder and ureteral surgery, is also essential to avoiding recurrent obstruction. In general, as long as a tension-free anastomosis can be performed using well-vascularized ureteral tissue, the long-term success of open surgical repair of ureteral strictures should be assured in the vast majority of patients. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Kowalczyk JJ, Keating MA, Ehrlich RM. Femoral nerve neuropathy after the psoas hitch procedure. Urology 1996;47:563–565. Lytton B, Weiss RM, Green DF. Complications of ureteral endoscopy. J Urol 1987;137:49–53. Motola JA, Smith AD. Complications of ureteroscopy: prevention and treatment. AUA Update Ser 1992;11:161–169. Olsson CA, Norlen LJ. Combined Boari bladder flap–psoas hitch procedure in ureteral replacement. Scand J Urol Nephrol 1986;20:279–282. Turner-Warwick R, Worth PHL. The psoas hitch procedure for the replacement of the lower third of the ureter. Br J Urol 1969;41:701–709. Weinberg JJ, Ansong K, Smith AD. Complications of ureteroscopy in relation to experience: report of survey and author experience. J Urol 1987;137:384–387.

Chapter 22 Simple and Partial Cystectomy Glenn’s Urologic Surgery

Chapter 22 Simple and Partial Cystectomy
Paul LaFontaine and John A. Petros

P. LaFontaine: Section of Urology, The Emory Clinic, Atlanta, Georgia 30322. J. A. Petros: Department of Surgery, Division of Urology, Emory University School of Medicine, Atlanta, Georgia 30322.

Simple Cystectomy Indications for Surgery Alternative Therapy Surgical Technique Partial Cystectomy Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

SIMPLE CYSTECTOMY
Simple cystectomy is defined as removal of the bladder without removal of adjacent structures or organs and is infrequently performed today. In the man, this would mean leaving behind the prostate, urethra, and seminal vesicles with the advantage that potency is conserved. However, in the impotent male in whom the urinary diversion is not expected to be reversed, removal of these structures adds little morbidity to the operation. In the woman, this means leaving behind the urethra, uterus, and anterior wall of the vagina. Simple cystectomy also implies that there is no dissection of the pelvic lymph nodes. Indications for Surgery Upper tract diversion has been a popular treatment alternative for a range of benign lower tract pathology and upper tract obstruction since the development of ureteroilieal cutaneous diversion in the 1950s. The indications for supravesical diversion are varied and include radiation cystitis after treatment of pelvic malignancies, interstitial cystitis, cyclosphosphamide cystitis, severe incontinence, neurogenic bladder, severe urethral trauma, and obstruction of the upper tracts. Initially, simple cystectomy was not routinely included during supravesical diversion because of the increased morbidity involved in simple cystectomy. However, complications from the retained bladder occur in up to 80% of patients undergoing supravesical diversion without simple cystectomy and include pyocystis, hemorrhage, sepsis, pain, vesicocutaneous fistula, colovesical fistula, feelings of incomplete emptying, and development of cancer in the retained bladder. 2 Indeed, the rate of secondary cystectomy approaches 20% in some series. Because of the high complication and reoperation rate, we recommend simple cystectomy as a part of upper tract diversion in any patient whose urinary diversion is not expected to be reversed, and especially in patients who have some component of bladder outflow obstruction. Alternative Therapy Other alternatives to simple cystectomy include conservative management, total cystoprostatectomy, radical cystectomy, and partial cystectomy. With the recent advances in laparoscopic technology and laparoscopic surgical techniques, laparoscopic simple cystectomy has become a potential treatment alternative, with several centers reporting success with the procedure as well as shortened postoperative convalescence times. 8 Surgical Technique A male patient is prepped and draped in the standard position as for radical cystectomy, in a supine position with the legs apart with gentle hyperextension. A female patient is placed in a lithotomy position for access to the perineum. This operation can be performed entirely extraperitoneally if prior upper tract urinary diversion has already taken place. This is preferred because it obviates the need for lysis of adhesions, which can be numerous in a patient who has had prior intra-abdominal surgery and/or radiation therapy. Obviously, if urinary diversion is to take place at the same time, an intraperitoneal approach is used. In the extraperitoneal approach, we use a lower midline incision extending from the pubis to immediately lateral to the umbilicus. The space of Retzius is entered by dividing the rectus abdominis in the midline. The retropubic space is developed down to the bladder, using a combination of blunt and sharp dissection to separate the parietal peritoneum from the dome and posterior wall of the bladder ( Fig. 22-1). It is important to repair with 3-0 or 4-0 chromic any tears made in the parietal peritoneum, as the parietal peritoneum provides an important boundary between the peritoneal contents and the raw surface of the pelvis after simple cystectomy. During dissection of the parietal peritoneum in the man, the vas deferens are encountered. If we are going to leave the seminal vesicles and prostate, we do not sacrifice the vas but instead dissect it posteriorly. If the seminal vesicles and prostate are to be sacrificed, we divide the vas. During the dissection of the parietal peritoneum posteriorly, the superior vesical pedicle is encountered. At this point it is clamped and divided between 2-0 silk ties.

FIG. 22-1. Bladder exposed extraperitoneally, with the space between the rectum and bladder developed by blunt dissection.

After control of the superior vesical pedicles bilaterally, the ureters are identified where they enter the bladder ( Fig. 22-2). In patients who have had prior upper tract diversion, the ureters are dissected proximally to the point where they were divided previously in order to ensure complete excision of the distal ureters. In patients who are to undergo urinary diversion at the same time, we use an intraperitoneal approach and divide the ureters close to the bladder wall.

FIG. 22-2. The vesicle pedicles are clipped and divided.

The bladder is then divided from the prostate at the prostatovesical junction using electrocautery starting anteriorly at the bladder neck and working laterally on both sides until the posterior bladder neck is reached ( Fig. 22-3). If the patient has significant prostatic hypertrophy, which will impede adequate closure of the bladder neck, then a suprapubic prostatectomy is performed with rigorous attention to hemostasis afterward because a Foley catheter can not be used to help control hemorrhage. The posterior bladder wall is then divided from within using electrocautery until the ampulla of the vasa are seen. The base of the bladder is then bluntly dissected off the seminal vesicles and ampulla of the vas ( Fig. 22-4). During this dissection, the lateral vascular pedicles are identified and divided between 2-0 silk ties. The bladder is now removed from the operative field. The prostate is then oversewn with a double layer of 0 chromic catgut.

FIG. 22-3. The bladder neck is incised anteriorly.

FIG. 22-4. The posterior bladder neck is divided, and the bladder removed. The distal bladder neck/urethra is oversewn.

In women, after development of the retropubic space, the parietal peritoneum is dissected off the dome and posterior wall of the bladder until the anterior vaginal fornix is reached. The superior vesical pedicles are divided as in a man. The ureters are handled the same way as in a man, with care taken not to injure the uterine artery during their dissection. A sponge on a stick placed in the vagina is used for cephalad traction, and the plane between the bladder and anterior vaginal wall is developed. During dissection of this plane, the lateral bladder pedicles are divided between 2-0 silk ties as they are encountered. Once the urethra is reached, the Foley catheter is removed, and the urethra is divided and oversewn with 0 chromic catgut suture. If simple cystectomy is being performed for interstitial cystitis, it is important to remove the entire urethra and external rethral meatus because failure to do so may result in persistent symptoms. If an upper tract urinary diversion is to be performed at the same time, an intraperitoneal approach is used with an incision from the symphysis pubis to a point midway between the xyphoid process and the umbilicus. Our approach for simple cystectomy in this case is the same, with the parietal peritoneum carefully preserved as a boundary between the peritoneal contents and the raw pelvic surfaces. Postoperatively, a drain is left in the pelvic space if the operation was performed in the presence of pyocystis. We rarely find it necessary to leave this drain more than 48 hours. Otherwise, we do not routinely leave a pelvic drain unless the patient will have placement of an orthotopic bladder.

PARTIAL CYSTECTOMY
Partial cystectomy has had a role in the management of bladder cancer for many years, though its exact role today is not well defined. The advent of improved means of transurethral resection of bladder tumors plus an improved understanding of the natural biology of bladder tumors has ensured that partial cystectomy today is a much less practiced procedure than in the past. There are certain advantages to partial cystectomy, such as sparing potency in men, retaining a functioning urinary reservoir, and the ability to achieve full thickness resection of bladder tumors and sample perivesical nodal tissue. This makes partial cystectomy an attractive procedure in selected patients. The major drawback in the use of partial cystectomy in the treatment of bladder cancer lies in the high tumor recurrence rates, which range from 40% to 80% in the reported series. Though this ensures that partial cystectomy is an uncommonly performed procedure, we believe it is an important part of the urologic surgeon's repetoire. Indications for Surgery Certain criteria must be met before a patient can be considered for partial cystectomy. The tumor must be a solitary, primary lesion located in a part of the bladder that allows for complete excision with adequate margins. We feel that margins of at least 3 cm are necessary for adequate resection. Other indications for partial cystectomy include patients who are not candidates for a complete transurethral resection of a bladder tumor because of a combination of patient body habitus, hypomobility of the hips secondary to osteoarthritis, or a fixed prostatic urethra. In this case, partial cystectomy may be required for complete diagnosis. It has also been recommended that tumors located in bladder diverticuli be managed with partial cystectomy. This is because bladder diverticuli have attenuated walls that may easily be perforated with transurethral resection, allowing for tumor spillage into the perivesical space. Other indications for partial cystectomy are in the management of genitourinary sarcomas in adults and children, the management of urachal carcinomas involving the dome of the bladder, involvement of the bladder by tumors in adjacent organs, and in the palliation of severe local symptoms. Nonmalignant indications for partial cystectomy include the

management of colovesical or vesicovaginal fistulas and the management of localized endometriosis of the bladder. Few other indications for partial cystectomy exist. At one time, partial cystectomy was offered to patients who were considered to be poor cardiopulmonary risks. However, improvements in surgical technique, perioperative care, and postoperative care have markedly reduced the operative mortality so that this is no longer considered an indication for partial cystectomy. Contraindications to partial cystectomy include patients with multiple lesions, recurrences, or tumors located on the trigone, where adequate excision is not possible because of the proximity of the ureteral orifices and bladder neck. In addition, patients must have biopsy-proven absence of cellular atypia or CIS in the remainder of the bladder and prostatic urethra. If there is evidence of fixation of the tumor to adjacent pelvic structures, or if segmental resection of the tumor would require removal of so much of the bladder as to necessitate augmentation cystoplasty, then a partial cystectomy should not be performed. These criteria, therefore, limit the number of patients who are candidates for partial cystectomy to 6% to 19% of patients who present with bladder cancer. The ideal candidate for partial cystectomy is a patient with a solitary primary lesion located on the dome of the bladder with no evidence of diffuse involvement of the urothelium. Alternative Therapy Other potential therapies include transurethral resection of bladder tumors, laser ablation of bladder tumors, intravesical chemotherapy, and radical cystectomy. There are several problems associated with the performance of partial cystectomy in the treatment of bladder cancer. After this operation, it can be difficult to treat tumor recurrence, and the operation cannot be repeated. There is also the real risk of tumor implantation in the wound, which is both difficult to treat and implies a poor prognosis for the patient. Several authors have advocated the use of perioperative radiotherapy to the incision to minimize the chance of tumor seeding. Perhaps the greatest contraindication to partial cystectomy lies in its questionable efficacy in the treatment of bladder cancer. Though randomized trials comparing partial cystectomy with other surgical therapies stage for stage in the treatment of bladder cancer are lacking, recurrence rates ranging from 40% to 80% have been reported. Surgical Technique The patient is placed on the operating room table in the supine position and is sterilely prepped and draped. The sterile field includes the penis in men and vulva and vagina in women. This allows for sterile insertion of a Foley catheter into the bladder after resection of the tumor and before closure of the incision. We prefer a lower midline incision to a transverse suprapubic incision because it allows for easier access to the peritoneal cavity if needed. We position the patient on the table such that the break in the table is at the anterior superior iliac spine, which allows for adequate flexion of the patient and elevation of the bladder into the wound. The standard incision extends from the pubic symphysis to the level of the umbilicus. The rectus abdominis is divided in the midline, and the space of Retzius is entered. The patient is then placed in the Trendelenburg position to elevate the abdominal contents out of the pelvis. Depending on the location of the tumor in the bladder, we proceed with either an extraperitoneal or an intraperitoneal approach. For tumors located on the dome or anterior part of the bladder, we prefer an extraperitoneal approach. For tumors located on the posterior aspect of the bladder, an intraperitoneal approach is preferred. Extraperitoneal Partial Cystectomy For our extraperitoneal approach, we expose the anterior surface of the bladder through the space of Retzius, mobilizing the peritoneum where it is readily separable from the bladder. A bilateral pelvic lymph node dissection with the boundaries from the bifurcation of the common iliac artery superiorly to Cooper's ligament inferiorly and from the external iliac artery laterally to the internal iliac artery medially. The bladder is freed laterally and posteriorly well beyond the site of the tumor. The fat over the site of the tumor is left attached to the bladder, and the superior vesicle pedicle can be divided if necessary ( Fig. 22-5).

FIG. 22-5. Partial cystectomy. The peritoneum is incised over the affected portion of the bladder after a pelvic lymph node dissection has been performed.

Several stay sutures are then placed in the bladder at a site known from cystoscopy to be distant from the tumor. The wound edges are packed away from the bladder with laparotomy pads or plastic drapes, and the bladder is entered between the stay sutures using electrocautery, taking care to minimize the amount of spillage of urine in order to minimize the risk of tumor implantation. The incision is extended for several centimeters anteriorly and posteriorly to allow for adequate visualization of the tumor and its relationship to the ureteric orifices and bladder neck. The tumor is then excised, with care taken to leave a 3-cm margin of normal-appearing bladder surrounding the tumor (Fig. 22-6). The tumor should be removed en bloc with the overlying perivesical fat and peritoneum using electrocautery or sharp dissection. If the tumor lies less than 3 cm from the ureteric orifice, sacrifice the ureteric orifice and perform a ureteral reimplantation. If enough ureter remains, a Leadbetter–Politano reimplantation is preferred, though a nonrefluxing ureteroneocystostomy or simple nipple reimplantation is acceptable. If excision of the tumor involves the bladder neck, it is possible to excise the bladder neck and the surrounding prostatic capsule after enucleation of the prostate gland. We do not recommend excising any portion of the bladder neck in women in order to avoid incontinence.

FIG. 22-6. A 3-cm margin of normal bladder is taken around the tumor.

After removal of the tumor, the bladder should be closed in two layers using a 3-0 Vicryl suture to close the urothelium and a 2-0 Vicryl to close the muscular layer (Fig. 22-7). A suprapubic cystostomy catheter is contraindicated in these patients because of the risk of tumor spillage, so it is essential that a wide-bore Foley catheter be used. We drain the perivesical space only if there is concern about the adequacy of bladder closure or a lymphadenectomy has been performed. The

abdominal wall is then closed in the standard fashion.

FIG. 22-7. The bladder is closed in two layers.

Postoperatively the urethral catheter should be left in place for 7 to 10 days. If there is any doubt as to the integrity of the repair, a gentle gravity cystogram may be performed. If perivesical drains are placed, they may be removed when drainage is minimal, usually on the third or fourth postoperative day. Intraperitoneal Partial Cystectomy For posteriorly located tumors, we take an intraperitoneal approach. After dividing the rectus abdominis muscles in the midline, we open the peritoneum in the midline. We then put the patient in the Trendelenburg position and pack the abdominal contents out of the pelvis with laparotomy pads. The peritoneum over the iliac vessels is incised, and we proceed with our bilateral pelvic lymph node dissection as described previously. We follow the obliterated hypogastric artery to the takeoff of the superior vesical artery, which we clamp and divide. The bladder is then freed posteriorly as needed, and stay sutures are then placed in the bladder, and the bladder is opened as described previously. Removal of the bladder tumor including the perivesical fat and peritoneum, reimplantation of the ureters, closure of the bladder, management of urethral catheters and perivesical drains, and wound closure are all handled as described previously.

OUTCOMES
Complications The perioperative and early post operative complications of partial cystectomy and simple cystectomy include hemorrhage and infection. In patients undergoing partial cystectomy, urinary extravasation is also a possible complication. Long-term complications of partial cystectomy include reduced bladder capacity and recurrence of the tumor in the pelvis or in the incision. This latter complication may be prevented by 2,000 cGy external beam therapy given immediately preoperatively. Results When utilized for interstitial cystitis with a substitution of bowel for reconstruction, partial (subtotal) cystectomy results in relief of pain and return of voiding ability in approximately two-thirds of patients. When utilized for superficial transitional cell carcinoma, the results are comparable to transurethral resection of superficial bladder tumors, but in invasive tumors, the results are inferior to those of radical cystectomy. Pathologic examination of the specimen will reveal the grade and stage of the tumor. If perivesical fat or pelvic lymph nodes are involved, it is recommended that the patient receive three courses of MVAC. Follow-up includes an IVP 3 months after surgery. Cystoscopies should be performed with bladder washings every 3 months for 2 years, then every 6 months for 2 years, and every year thereafter. With careful selection of the case, the reduction in bladder volume should not be so great as to cause urinary frequency. It is surprising how much of the bladder may be removed without causing urinary frequency. Even if postoperative frequency does occur, in the majority of cases it resolves spontaneously within 6 months. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Dandekar NP, Tongaonkar HB, Dalal AV, Kulkarni JN, Kamat MR. Partial cystectomy for invasive bladder cancer. J Surg Oncol 1996;60:24. Eigner EB, Freiha FS. The fate of the remaining bladder following supravesical diversion. J Urol 1990;144:31. Freiha FS. Open bladder surgery. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED, eds. Campbells urology, 6th ed. Philadelphia: WB Saunders, 1992;2765–2768. Herr HW, Scher HI. Neoadjuvant chemotherapy and partial cystectomy for invasive bladder cancer. J Clin Oncol 1994;12:975. Herr HW. Urachal carcinoma: the case for extended partial cystectomy. J Urol 1994;151:365. Hicks BA, Hensle TW, Burbige KA, Altman RP. Bladder management in children with genitourinary sarcoma. J Pediatr Surg 1993;28:1019. Kinouchi T, Hanafusa T, Kuroda M, Usami M, Kotake T. Ossified cystic metastasis of bladder tumor to abdominal wound after partial cystectomy. J Urol 1995;153:1049. Parra RO, Worischek JH, Hagood PG. Laparoscopic simple cystectomy in a man. Surg Laparosc Endosc 1995;5:161. Sweeney P, Kursh ED, Resnick MI. Partial cystectomy. Urol Clin North Am 1992;19:701.

Chapter 23 Radical Cystectomy in Men Glenn’s Urologic Surgery

Chapter 23 Radical Cystectomy in Men
Mohamed A. Ghonheim

M. A. Ghonheim: Urology and Nephrology Center, Mansoura, Egypt.

Diagnosis Indications for Radical Cystectomy in Men Alternative Therapy Surgical Technique Preparation of the Patient Anesthesia and Instrumentation Position and Initial Exposure Lymphadenectomy Cystoprostatectomy Variations on a Theme Postoperative Management Outcomes Complications Results Chapter References

One of the first detailed operative descriptions of radical cystoprostatectomy and pelvic lymphadenectomy was probably provided by Marshall and Whitmore in 1949. In the 1950s and early 1960s, the operation was attended with significant mortality and morbidity. Although more complex urinary diversions are increasingly employed, contemporary cystectomy is associated with very low mortality. Furthermore, the advent of nerve-sparing cystectomy and orthotopic bladder substitution has significantly reduced functional losses and provided many patients with good locoregional control as well as a good quality of life. The technique, herein described, is based on cumulative experience of more than 20 years during which more than 1,000 cystectomies were carried out at the Department of Urology, Mansoura University, Egypt.

2

DIAGNOSIS
The diagnosis of transitional-cell carcinoma is generally made by transurethral resection of the tumor in the bladder (see Chapter 111). Once the diagnosis has been established, it is important to know the histologic stage, particularly if the tumor invades the muscularis propria. Invasion of the muscularis mucosa is not considered as constituting a muscle-invasive tumor. The clinical staging of transitional-cell carcinoma can generally be performed by abdominal and pelvic computerized tomographic (CT) scans. Occasionally radionuclide bone scans are indicated if there is either symptomatic bone pain, abnormalities on the CT scan, or the patient has either an elevated serum calcium or alkaline phosphatase.

INDICATIONS FOR RADICAL CYSTECTOMY IN MEN
The major indication for cystectomy in men is carcinoma of the bladder. In general, the operation is carried out for: 1. Patients with superficial tumors in whom endoscopic control has failed in spite of adjuvant intravesical chemo- and/or immunotherapy. Although these measures had proved effective in the management of such cases (£T 1), an important minority fail. High tumor grade, multifocal lesions, diffuse carcinoma in situ, and involvement of the prostatic urethra were all reported as high-risk factors. 2. Infiltrating tumor without evidence of distant metastasis. These include tumors infiltrating the muscle layers (P 2, P 3a) or the perivesical fat short of the pelvic wall (P3b). Infiltration of adjacent organs (P 4) or involvement of the regional lymph nodes is not considered as a contraindication for the procedure. The extent of the radical operation in the male includes the removal of the bladder, its peritoneal covering, the perivesical fat, the lower ureters, the prostate, the seminal vesicles, and the vasa deferentia. In the standard procedure, as much as possible of the membranous urethra is also removed, and total urethrectomy is carried out only if there is involvement of the prostatic urethra. 4

ALTERNATIVE THERAPY
Alternatives to radical cystectomy include local therapy, partial cystectomy, intravenous chemotherapy, radiation therapy, or a combination of chemotherapy and radiation therapy. Local therapy in invasive disease generally results in progression of the disease and death of the patient within 5 years. Systemic chemotherapy or radiation therapy is associated with a 25% 5-year survival, though the combination of the two modalities results in significant synergy with up to 50% 5-year survival.

SURGICAL TECHNIQUE
Preparation of the Patient In view of the extent of surgery and the length of the operative time, a thorough medical evaluation and anesthetic consultation are required. Bowel preparation is necessary before surgery. If it is planned to use the small bowel, oral neomycin and a low-residue diet are all that is needed. More rigorous preparation is required if the colon is utilized. This includes soapsuds enemas until the colonic contents return clear. A neomycin sulfate enema is given on the evening before the day of operation. Intravenous fluids are also administered to maintain hydration. Patients with histories of thromboembolic disease or varicose veins should receive a prophylactic dose of heparin (5,000 units subcutaneously) the night before the operation and every 12 hours thereafter until ambulation. A parenteral broad-spectrum antibiotic is given just before induction of anesthesia and continued postoperatively for 3 days. The region extending from the midchest to the midthigh should be cleaned and prepared on the night before surgery. Anesthesia and Instrumentation Full relaxation of the abdominal muscles by an appropriate anesthetic is necessary throughout the entire procedure. Hypotensive anesthesia would provide an additional advantage and would reduce blood loss. The choice of instruments depends mainly on surgeon's preference. Standard retractors of various sizes and curves as well as long curved and angled scissors are needed. Long curved clamps should also be available. In our practice, we prefer to retract the abdominal wall on the side where the dissection is carried out using one or two ordinary retractors. A ring retractor is applied once the lymphadenectomy is completed. Position and Initial Exposure The patient is put in the supine position with a Trendelenberg tilt. Slight bending of the knees would further help in the relaxation of the abdominal muscles, facilitate retraction, and provide a wider exposure. If a total urethrectomy is planned, the patient is put in a slight lithotomy position for access to the perineum. The surgical area to be sterilized and draped extends from the lower chest down to the root of the penis. A self-retaining catheter is introduced into the bladder and

kept indwelling for its evacuation throughout the procedure. A long, vertical, right paramedian incision extending from the symphysis pubis inferiorly to a point half way between the umbilicus and xyphoid process of the sternum superiorly is generally employed. Alternatively, a midline incision encircling the umbilicus can also be utilized. For obese patients a lower abdominal muscle-cutting transverse incision is preferred. Under such circumstances it provides a wide and direct exposure of the pelvis. Initially, the abdominal and pelvic cavities are explored. The growth is palpated, its degree of mobility determined, and its relation to the adjacent structures assessed. The endopelvic and aortic lymph nodes are palpated, and frozen sections are taken if necessary. The general peritoneal cavity, omentum, intestinal tract, kidney, spleen, and liver are thoroughly examined. If the decision is to proceed with the radical operation, the intestines are packed out of the pelvis, and the retropubic space is opened by blunt dissection. Any small bleeders are coagulated. This dissection is extended inferiorly and laterally until the ventral surface of the bladder and prostate are exposed. The peritoneal incision is extended inferiorly on either side of the urachal remnant. The urachal remnant is dissected off its attachment with the umbilicus and clamped. In this manner a triangular peritoneal flap with its apex pointing superiorly is raised and will be removed later en bloc with the bladder. Lymphadenectomy The peritoneal incision, on either side, is extended posterolaterally along the lateral border of the external iliac and common iliac vessels up to the aortic bifurcation. The vas deferens is identified and ligated near the internal ring. The fascia on the iliopsoas is incised and reflected medially. The triangle of Marceille is exposed by retracting the common and external iliac arteries medially and dissecting the space between these vessels and the medial border of the psoas muscle. 3 Dissection of the fibrolymphatic tissues in this space will expose the obturator nerve as it emerges from the medial border of the psoas muscle ( Fig. 23-1). The fibrofascial sheath covering the distal half of the common iliac and the external iliac vessels is then opened and stripped medially to remove the perivascular lymphatics and lymph nodes. The vessels are gently retracted, laterally and immediately below and medial to the cleaned external iliac vein, and the obturator space is entered. By working right on the psoas and obturator muscles, one can strip all the pelvic fascia medially without difficulty. The obturator neurovascular bundle is included in the stripped mass. The obturator nerve is identified and separated from the vessels, which are divided and ligated as they leave the pelvis through the obturator foramen. Dissection is facilitated and the operating time reduced by the use of electrocoagulation to control lymphatic and small blood vessels throughout the lymphadenectomy.

FIG. 23-1. Dissection of the triangle of Marceille. The Psoas muscle is retracted laterally and the iliac vessels medially. The obturator nerve is exposed in the floor of the triangle as it emerges from the medial border of the Psoas muscle.

Cystoprostatectomy The fibrolymphatic mass is now reflected medially. The internal iliac artery is dissected free, and its anterior division is divided and ligated. The ureter is identified where it crosses the common iliac bifurcation, dissected free for 3 to 4 cm, divided, and its distal end ligated. While traction is applied on the ligated ureteric stump of the ureter, finger dissection along its posteromedial border opens the space of Denonvillier laterally. The step greatly helps in the definition of the plane between the bladder and rectum, which will be required at a later stage in the operation. The phases of the lateral dissection are illustrated in Fig. 23-2.

FIG. 23-2. Lymphadenectomy. (A) The fibroareolar tissue has been dissected from the anterior and medial aspects of the Psoas major muscle. The external and common iliac arteries are exposed and skeletonized. (B) Further dissection exposes the external vein. The obturator fossa is cleared with separation of the obturator nerve. (C) Further dissection of the internal iliac artery and its branches prior to their control. (D) The lateral dissection is completed. The anterior division of the internal iliac artery is divided with control of its parietal branches. The ureter is divided, and the ligature on its stump is used for traction.

The endopelvic fascia on either side on the prostate is then opened by the tip of a blunt pair of scissors ( Fig. 23-3). The optimal site for the creation of this opening is a white line marking the fusion of the parietal fascia lining the pelvic surface of the levator ani with the visceral fascia covering the lateral surface of the prostate. A right-angled clamp is used to lift the fascia from the underlying venous plexus, and it is further incised medially until the prostatic ligaments are reached. By blunt dissection, this plane is further developed posteriorly on either side of the prostate. Further anterior dissection is deferred to the final stages of the procedure to minimize the possibility of sudden blood losses from inadvertent injury of the prostatic venous plexus.

FIG. 23-3. The bladder and prostate are retracted medially by a Deaver's retractor. The reflection of the endopelvic fascia from the ventral surface of the levator ani to

the prostate is opened. Blunt dissection would further develop this space and expose the lateral surface of the prostate.

The specimen is now lifted ventrally by applying traction on the median umbilical ligament (urachus). The two planes developed along the posteromedial borders of the ureter on either side are easily joined together by blunt dissection. As a result, the peritoneal reflection from the anterior surface of the rectum to the back of the bladder could be stretched and safely incised by diathermy. The potential space between the rectum posteriorly and the bladder, seminal vesicles, and prostate anteriorly is opened by blunt dissection ( Fig. 23-4). As the prostatic apex is reached, this space becomes obliterated as a result of fusion of the two layers of the fascia of Denonvillier. This cul-de-sac is opened by the blunt tip of long angled scissors. Once this is completed, the tip of the surgeon's forefinger would readily feel the apex of the prostate as well as the catheter in the urethra in the midline. Alternatively, if it is directed laterally, it would appear through the previously created openings on either side of the prostate ( Fig. 23-5).

FIG. 23-4. The peritoneum of the floor of Douglas pouch is incised by diathermy. The space between the rectum posteriorly and the bladder and seminal vesicles anteriorly is dissected and opened.

FIG. 23-5. The cul-de-sac formed by the fusion of the two layers of the fascia of Denonvillier is opened by the tip of a blunt pair of long scissors. Thereafter, the tip of the surgeon's forefinger would feel the apex of the prostate and the catheter in the urethra. If the forefinger is directed laterally it would appear through the previously created openings on either side of the prostate. Thus, a thick wide fascial band is defined (the vesicoprostatopelvic fascia). It is divided piecemeal between clamps (arrow and interrupted line).

In this manner, a thick and wide fascial band is created on either side, connecting the bladder, vesicles, and prostate anteriorly with the pararectal fascia posteriorly (the vesicoprostatopelvic fascia). This is divided piecemeal between clamps, which are underrun by 2-0 polyglactin sutures. The bladder is now free laterally and posteriorly, and the mass is left to drop in the pelvis. Attention is now focused on the anterior and final phase of the procedure. The puboprostatic ligaments are identified by applying traction on the prostate in a cephalad and posterior direction. These ligaments are carefully severed at the point of their insertion in the pubic bone. The prostatic venous plexus is controlled by one or two sutures of 3-0 polygalactic acid placed near the prostatic apex. A transverse incision is made proximal to these sutures with a long scalpel and extended with sharp dissection by scissors, exposing the urethra, within which the catheter can be palpated (Fig. 23-6). The catheter is then withdrawn; the urethra is clamped and transected; the distal end is ligated; and the specimen is removed. Final hemostasis is achieved by inserting deep 2-0 polyglactin sutures between the edges of the levator ani muscles on either side ( Fig. 23-7). No attempt is made to reperitonealize the pelvis. Two tube drains are placed in the pelvic cavity and brought out through separate incisions in the abdominal wall. The wound is closed in layers with particular attention for careful closure of the anterior rectus sheath. This is closed with interrupted sutures of nylon with the knots tied to the inside.

FIG. 23-6. The puboprostatic ligaments were incised and the prostatic venous complex controlled. The membranous urethra is thus exposed.

FIG. 23-7. The specimen was removed. Final hemostasis is achieved by two or three interrupted sutures of 2-0 polyglactin between the medial borders of the levator

ani muscles.

Variations on a Theme One-Stage Cystoprostatourethrectomy Urethrectomy is indicated in a subpopulation of patients with multifocal tumors, diffuse carcinoma in situ, or tumor involving the bladder neck and/or the prostate. Following incision of the puboprostatic ligaments and control of the prostatic venous plexus described, traction is applied on the cystectomy specimen in a cephalad direction. The urethra is dissected from the urogenital diaphragm with a long pair of dissecting scissors. In this manner, 2 to 3 cm of the membranous urethra can be mobilized. The pelvis is temporarily acked with gauze, and further steps are carried out perineally without urethral transaction. A midline incision in the perineum is usually employed. The skin, subcutaneous tissue, and the bulbocavernosus muscle are incised in the midline. The Foley catheter can now be palpated in the urethra. The urethra is dissected sharply from the overlying corpora cavernosus. Further dissection is carried out in the direction of the glans penis. Traction on the urethra results in inversion on the penis, allowing dissection of the urethra as far as the coronal sulcus. The penis is then allowed to restore its normal position. The urethral meatus is circumscribed sharply, and the glans penis is incised in the midline to allow dissection of the fossa navicularis. The entire penile urethra is now free. The glans penis is reconstructed by a few sutures of interrupted 3-0 chromic catgut. Attention is now focused on dissection of the bulbar urethra. The relatively avascular tissues ventral to the bulbar urethra and beneath the symphysis pubis are dissected first. Thus, the corresponding part that had been previously dissected in the pelvis can be reached, and the pelvic and perineal exposures joined. Dissection is further developed laterally and posteriorly with control of the bulbar urethral arteries. In this manner the urethra is freed totally, and the whole specimen is removed in one block. Radical Cystoprostatectomy with Orthotopic Bladder Substitution A standard radical cystoprostatectomy is performed except that the final stages of the operation must be done with attention to detail to avoid damage to the urethra and periurethral musculature. The integrity of these structures has a central role in the functional success of orthotopic substitution. Following lymphadenectomy and control of the pedicles, the endopelvic fascia on either side of the prostate is opened by the tip of a blunt pair of scissors. A right-angled clamp is used to lift the fascia from the underlying venous plexus, and then it is further incised medially until the puboprostatic ligaments are reached. These ligaments are carefully severed at the point of their insertion in the pubic bone. The prostatic venous plexus is controlled by one or two suture ligatures of 3-0 polyglactin just distal to the vesicoprostatic junction. A transverse incision is made proximal to these sutures and extended by sharp dissection with scissors toward the apex of the prostate. The catheter is palpated in the urethra, the anterior wall of which is then incised just distal to the prostatic apex. The exposed Foley catheter is transected, clamped, and held for traction. At this point, three stay sutures of 4-0 polyglactin are placed through the urethra at the 3-,9-, and 12-o'clock positions, incorporating the mucosa as well as the periurethral musculature ( Fig. 23-8). These sutures prevent retraction of the urethra following its complete transection and are used later for the urethroileal anastomosis. The posterior urethral wall is then incised to expose the dorsal fibrous raphe formed by the fascia of Denonvillier, which is lifted from the anterior surface of the rectum by a right-angled clamp and divided. The divided fascia is then included in two posterior stay sutures at the 5- and 7-o'clock positions for its later incorporation in the urethrointestinal anastomosis.

FIG. 23-8. Preparation of the urethral stump for orthotopic substitution. (A) The anterior wall of the urethra is transacted. Three stay sutures of 4-0 polyglactin are placed through the urethra at the 3-,9-, and 12-o'clock positions. (B) Transection of the urethra is completed, and further stay sutures are applied. These will prevent retraction of the urethral stump and will be later used for the urethroileal anastomosis.

Radical Cystoprostatectomy with Nerve Sparing This procedure was initially described by Schlegel and Walsh. 5 It can be carried out in an antegrade or a retrograde manner, though in our practice we prefer the antegrade approach. During radical cystectomy there are two points where the neurovascular bundle could be injured: (a) posterolateral to the prostate and (b) behind the seminal vesicles. If the extent of the pathology allows the surgeon to avoid these areas, potency could be preserved. Bilateral lymphadenectomy with creation of the space between the rectum posteriorly and the bladder, seminal vesicles, and prostate anteriorly is carried out as previously described. By a combination of blunt and sharp dissection, the lateral surface of the seminal vesicles is freed from the medial aspect of the vesicoprostatopelvic fascia. This allows the control of these ligaments at a more ventral plane. As a result, the neurovascular pathway behind the seminal vesicles is avoided. These pedicles are controlled by a series of simple interrupted sutures of 3-0 Vicryl. The use of heavy clamps, clips, and diathermy should be avoided. The dorsal vein complex is now controlled, and the urethra isolated carefully from the adjacent fascia. The urethra is then transected, and the Foley catheter is clamped and held for traction. The prostate can thus be elevated superiorly. A right-angle clamp is used to identify branches of the neurovascular bundle to the prostate. These are ligated and divided, freeing the prostate from all its lateral attachments. Postoperative Management Intravenous alimentation and nasogastric suction are maintained until normal bowel activity is resumed. Systemic antibiotics are continued for 3 days postoperatively. Chest exercises and physiotherapy to the lower limbs should be carried out. Subcutaneous heparin should be administered if indicated. The tube drains are removed when drainage becomes less than 100 ml/day. It is advisable to estimate the creatinine content of the fluid to ensure that it is not the result of a urinary leak. Patients with an ileal conduit can be discharged on the 10th to the 12th postoperative day. Following orthotopic substitution, patients are usually kept in the hospital for 3 weeks. Before discharge, a pouchography is carried out to make sure that there are no leaks from the neobladder or from the urethroileal anastomosis.

OUTCOMES
Complications The two most serious complications that may occur during the procedure are excessive blood loss or rectal perforation. Sudden massive bleeding is usually venous in origin, arising from tributaries of the external iliac vein during the lymphadenectomy: the deep circumflex iliac vein laterally and an abnormal obturator vein medially. Since both are located near the inguinal ligament, good retraction, illumination and suction are needed. A laceration of the external iliac vein is then sutured with 5-0 Prolene.

Another source of bleeding is in relation to the dorsal vein complex. Dissection of this area has to be deferred to the final phase of the procedure. This venous complex is usually injured when the puboprostatic ligaments are incised. Compression of the bleeding area by a piece of gauze (4×8) and the tip of a long thin blade of a Deaver's retractor are necessary until the dissection of the urethra is completed. Thereafter, bleeding is controlled by one or two interrupted 3-0 sutures of polyglactin acid placed between the two medial borders of the levator ani muscles. However, the most serious source of bleeding is from the internal iliac vein or one of its tributaries. Sudden excessive bleeding occurs from the depth of a narrow deep recess. Blind attempts to control the bleeding with clamps usually fail and result in more damage. In our experience, one has to achieve an initial temporary control by packing. One or two 4×8 pieces of moist gauze are sufficient. The pack is tightly and constantly compressed for a few minutes and then is left in place. The operator should proceed with further operative steps until the specimen is removed. Now, the working space is wide enough to allow manipulations under vision. The ipsilateral external iliac and common iliac veins as well as the main stem of the internal iliac artery are controlled by bulldog clamps. The gauze pack is then removed. There will still be some back bleeding, but with the help of a little suction, the bleeding vessels are readily located and easily secured by suture ligation using 4-0 silk. The other serious intraoperative complication is rectal perforation. This usually takes place during the final phase of the operation if the space between the prostate and rectum was not adequately and completely opened. Under such circumstances, traction on the specimen will lead to tenting of the anterior wall of the rectum. Sharp dissection with scissors or application of clamps would result in an injury of the anterior wall of the rectum well below the peritoneal reflection. If this injury is recognized, the tear is meticulously repaired. The edges are trimmed and closed in two layers using 3-0 polyglactin acid: the first through and through, and the second inverting as a fascia muscular layer (Lembert technique). An omental flap is raised, brought down to the pelvis and sutured over the repair for additional security. The pelvic cavity is then thoroughly irrigated with 1% solution of kanamycin in saline. At the end of surgery, while the patient is still under anesthesia, anal dilation is carried out up to three to four fingers to establish adequate decompression. Generally, by following these principles, one can avoid the need of a temporary proximal colostomy. The postoperative mortality following contemporary cystectomy is 2% or less. 6 The most common postoperative complication is prolonged ileus. This is treated by nasogastric suction, intravenous alimentation, and hyperalimentation if necessary. Septic complications including abdominal and/or pelvic abscesses, wound sepsis, and septicemia are not uncommon. These are treated by the appropriate antibiotics and drainage of the infected collection. This is best achieved by ultrasound-guided aspiration and/or insertion of a percutaneous tube drain. Wound dehiscence should be immediately repaired by proper closure using tension sutures. Urinary collections (urinoma) are drained under ultrasound guidance. If the source of leak is the ureterointestinal anastomosis, a percutaneous nephrostomy tube is inserted until healing is achieved and checked with an antegrade study. Results Radical cystectomy has evolved as the standard therapeutic modality for muscle-invasive bladder cancer. It can be accomplished with very low mortality, and technical innovations with nerve sparing and orthotopic substitution can provide many patients with a good quality of life with minimal functional losses. All contemporary series demonstrate that radical cystectomy can result in a substantial rate of cure with overall survival ranges between 48% and 53%. 1,8 For tumors with low stage (<P 2), the survival could be as high as 75%. With further stage progression, the survival expectancy is decreased. Radical cystectomy with pelvic lymphadenectomy also provides a survival advantage for cases with nodal disease (5-year survival in the range of 20%). 7 Evidence has also been provided that adjuvant cisplatin-based polychemotherapy improves the chances of survival among patients with advanced locoregional disease. 9 CHAPTER REFERENCES
1. Frazier HA, Robertson JE, Dodge RK, Paulson DF. The value of pathologic factors in predicting cancer-specific survival among patients treated with radical cystectomy for transitional cell carcinoma of the bladder. Cancer 1993;71:3993–4001. 2. Marshall VF, Whitmore WF Jr. A technique for the extension of radical surgery in the treatment of vesical cancer. Cancer 1949;2:424–428. 3. McGregor AL. A synopsis of surgical anatomy, 9th ed. Bristol: John Wright & Sons, 1963;99. 4. Schellhammer PF, Whitmore WF Jr. Transitional cell carcinoma in men having cystectomy for bladder cancer. J Urol 1976;115:56–60. 5. Schlegel PN, Walsh PC. Neuroanatomical approach to radical cystoprostatectomy with preservation of sexual function. J Urol 1987;138:1402–1406. 6. Skinner DG, Crawford, ED, Kaufman JJ. Complications of radical cystectomy for carcinoma of the bladder. J Urol 1980;123:640–643. 7. Skinner DG. Management of invasive bladder cancer: A meticulous pelvic node dissection can make a difference. J Urol 1982;128:34–36. 8. Soloway MS, Lopez AE, Patel J, Lu Y. Results of radical cystectomy for transitional cell carcinoma of the bladder and the effect of chemotherapy. Cancer 1994;73:1926–1931. 9. Stockle M, Meyenburg W, Wallek S, et al. Adjuvant polychemotherapy of non-organ-confined bladder cancer after radical cystectomy revisited: Long-term results of a controlled prospective study and further clinical experience. J Urol 1995;153:47–52.

Chapter 24 Radical Cystectomy in Women Glenn’s Urologic Surgery

Chapter 24 Radical Cystectomy in Women
James E. Montie

J. E. Montie: Section of Urology, University of Michigan Medical Center, Ann Arbor, Michigan 48109-0330.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Perioperative Care Operative Technique Postoperative Care Outcomes Complications Chapter References

Total cystectomy is the most effective means of pelvic control of potentially lethal transitional-cell carcinoma (TCC) of the bladder. Bladder cancer is more common in men, and thus, cystectomy is more commonly performed in men than women, generally by a factor of 3:1 or 4:1.2 In some aspects, cystectomy in a woman is easier than in a man because of a larger pelvis and better exposure. However, in many ways the cystectomy is more difficult in women. Urologists have fewer opportunities for familiarity with major pelvic surgery in women than in men. Commonly, the uterus is removed with the bladder, and a hysterectomy is a procedure that urologists do not routinely perform. Bleeding from the paravaginal tissue and venous plexus around the urethra can be brisk and tedious to control. If the urethra is removed entirely, its reconstruction and reconstruction of the vagina can be done in several ways. Finally, orthotopic diversion in women is coming of age; preparation of the urethra for maximum preservation of a normal voiding pattern is different from that done in a male. The challenges of a female cystectomy are unique, and they may be accentuated for the surgeon who uncommonly performs a cystectomy.

DIAGNOSIS
The diagnosis of transitional-cell carcinoma is covered in the chapter on transurethral resection of bladder tumors ( Chapter 111).

INDICATIONS FOR SURGERY
The broad indication for a cystectomy is an invasive bladder cancer unlikely to be eliminated with transurethral resection and intravesical therapies. On occasion, recurrent, high-grade, noninvasive cancers require a cystectomy to completely eliminate a potentially lethal cancer. 1 The decision on when to proceed with a cystectomy is always difficult for the patient and physician alike. The majority of muscle-invading bladder cancers do so at presentation; occasionally cystectomy is delayed even in this setting because of the risk of the surgery and the negative consequences on urinary function if the bladder is removed. Personally, I believe cystectomy is commonly delayed by either the patient or the physician, and diminished survival after cystectomy may in part be a consequence of this delay. Recent experience suggests that earlier surgery prompted by the availability of orthotopic diversion may translate into an improved outcome. Perhaps our indication for a cystectomy (or an aggressive bladder preservation strategy using chemotherapy and radiation if this proves effective) should be the inability to reliably and completely eliminate a potentially lethal cancer rather than the traditional indication of muscle invasion. A recent review highlighted the frequency of understaging of high-grade minimally invasive or noninvasive tumors. Of 182 patients undergoing a cystectomy for high-grade clinical stage Ta and Tl tumors and CIS, 34% had a higher stage at cystectomy with a subsequent diminished survival. 1 Orthotopic diversion is an appropriate consideration in the majority of women. 9 Currently, the primary contraindication to orthotopic diversion is cancer involving the bladder neck, seen in approximately 25% of women undergoing cystectomy. Based on mapping studies, women without TCC at the bladder neck are unlikely to have urethral TCC. 8 In the absence of TCC at the bladder neck, the patient should have an orthotopic diversion provided as a choice. Older patients may have minimal alteration of their life style with an ileal conduit and may prefer this method of diversion. The experience with orthotopic diversion in women is small, currently at an estimated 100 to 200 cases worldwide. 9,10 As experience accumulates, it is apparent that urinary continence is as good in women as in men. Daytime incontinence in women is extremely rare, and nighttime incontinence may resolve more quickly than in men. A female cystectomy allows preservation of the GU diaphragm and innervation of the external sphincter. 7,10 Urinary retention rather than incontinence is seen more commonly in women and may be attributable to a variety of factors as discussed later.

ALTERNATIVE THERAPY
Alternatives to cystectomy include observation, systemic chemotherapy, radiation therapy, or a combination of chemotherapy and radiation. These modalities are generally offered to patients who are a poor surgical risk, who refuse surgery, or who are elderly. However, cystectomy is often the best option for invasive bladder cancer in the elderly who are otherwise in reasonably good health. 3 Invasive bladder cancer is not an indolent disease, and death from uncontrolled TCC is high in the first 3 to 4 years. Thus, for a healthy 75-year-old, the invasive bladder cancer is the biggest health risk. The morbidity from a cystectomy is substantial, and the risk for complications from the operation is undeniable. However, the risk of radiation therapy and chemotherapy are also substantial, and elimination of the cancer is less likely. 4 Unfortunately, a therapeutic strategy considered inadequate in younger patients may be employed in the healthy elderly, often with reduced doses of chemotherapy making the treatment even less likely to be successful. Improved perioperative care allows cystectomy to be done with a low operative mortality, supporting an expedient cystectomy as the overall safest and most effective approach. 6

SURGICAL TECHNIQUE
Perioperative Care Life-style factors that contribute to the development of bladder cancer, such as smoking, also contribute to cardiovascular and pulmonary disease. Evaluation and optimization of cardiac and pulmonary function are essential before a cystectomy. Pulmonary toilet, bronchodilators, occasionally steroids, and idealized cardiac function will lessen the risk for perioperative complications. Preoperative stoma marking by an enterostomal therapist is a necessity. Body habitus and previous surgery may require placement of the stoma in an unusual site. Improper placement of the stoma transforms an otherwise successful operation into a nightmare for the patient. Poor nutrition is an important risk factor for perioperative complications. Routine perioperative parental nutrition is not necessary in the well-nourished patient; however, if a complication does occur that delays prompt resumption of eating, there should be no hesitation to institute enteral or parenteral feeding. In our experience, approximately 10% to 15% of patients need postoperative nutritional support. Either GOLYTELY (Braintree Laboratory, Braintree, MA) or Phosphosoda (Fleet Corporation, Lynchburg, VA) solutions are used as a single-day bowel prep. Perioperative antibiotics for wound prophylaxis relies on a second-generation cephalosporin. Patients without medical problems requiring special monitoring are admitted to the hospital the same day as the surgery. Intraoperative anesthesia monitoring allows prompt recognition of hypoxemia or hypovolemia during the case. Epidural analgesia, patient-controlled analgesia (PCA), and nonsteroidal anti-inflammatory agents are useful adjuncts to provide better postoperative pain control, which translates into better pulmonary hygiene, ambulation,

and return of bowel function. 1,5 Intermittent compression stockings are currently used for deep venous thrombosis (DVT) prophylaxis. Early ambulation and a high index of suspicion for DVT are important. Establishment of a “care” or “critical” pathway is useful for cystectomy patients. 4 Compliance with the ideal postoperative course is difficult after cystectomy because of a high frequency of comorbid disease and complications, which may delay discharge even if they are not life-threatening. Nonetheless, the pathway provides a target for the patient, physicians, and nursing staff for anticipated events during the hospital course. An example of the pathway in use in 1997 at the University of Michigan Hospitals is provided in Fig. 24-1.

FIG. 24-1. Critical pathway for radical cystectomy/neobladder in 1997 at the University of Michigan Medical Center.

Operative Technique Access to the urethra and vagina is necessary during a female cystectomy. A modified lithotomy position is used with either Allen or Lloyd–Davies stirrups. Careful padding to prevent pressure points, which may cause a perineal nerve compression or anterior compartment syndrome, is important. The vagina and perineum must be well prepped. A vertical midline incision gives ideal exposure. The urachal remnant provides a convenient handle for traction on the bladder ( Fig. 24-2). The peritoneum is divided along the lateral umbilical ligaments, and the round ligament is clipped and divided ( Fig. 24-3). Usually, the fallopian tubes and ovaries are present but nonfunctional and thus are removed with the uterus and bladder after dividing the gonadal vessels above the ovaries ( Fig. 24-4). The ureters are mobilized with a large amount of periureteral adventitial tissue to preserve optimal blood supply and divided ( Fig. 24-5). Typically, a pelvic lymphadenectomy is performed with the following boundaries: the lateral limit is the genitofemoral nerve, the superior limit is the bifurcation of the common iliac artery, the inferior limit is the inguinal ligament, and the medial limit is the perivesical tissue. All tissue is removed from around the obturator nerve. Specific aspects of this surgical dissection are not covered in this chapter (see Chapter 32).

FIG. 24-2. The urachal remnant provides a convenient handle for traction on the bladder through the case and is divided between Kelly clamps and igated.

FIG. 24-3. The division of the peritoneum follows the course of the lateral umbilical ligament until the round ligament is identified, clipped, and divided.

FIG. 24-4. The fallopian tubes and ovaries, if present, are commonly removed with the uterus and bladder. The gonadal vessels are divided and ligated cephalad to the ovary.

FIG. 24-5. Each ureter is mobilized with a large amount of periureteral adventitial tissue to preserve optimal blood supply. The ureter is divided a short distance above the bladder.

The following maneuver isolates the lateral blood supply to the bladder and uterus coursing from the internal iliac artery and vein: expose the endopelvic fascia and the perirectal “fat pad”; with medial traction on the bladder and ureter, develop bluntly with the index finger a plane just medial to the well-defined superior vesicle artery, aiming obliquely toward the perirectal tissue. This will isolate the superior vesicle artery, which needs to be ligated separately, and many small arteries and veins, which are controlled with Ligaclips “marching down” the lateral pedicle under direct vision. Medical traction on the bladder with fingers above and below the pedicles enhances exposure (Fig. 24-6).

FIG. 24-6. Division of the lateral pedicle is one of the more important technical aspects of the procedure. The endopelvic fascia should be well exposed with blunt dissection. A perirectal fat pad lying adjacent to the rectum defines the lower limit of the lateral pedicle coursing from the internal iliac vessels. Medial traction on the urachal remnant and retraction of the ureter medially allow an index finger to create a plane just medial to the origin of the superior vesicle artery outlining the lateral pedicle. The caudad extent of blunt dissection is the perirectal fat tissue. The medial traction on the bladder with the surgeon's nondominant hand easily exposes the entire pedicle. The superior vesicle artery is ligated, but the remainder of the small vessels can be controlled with clips on both sides.

After division of the lateral pedicles on each side, the technique is modified depending on the amount of perivesical soft tissue to be removed with the bladder ( Table 26-1). In a classical anterior pelvic exenteration, the bladder, uterus, bilateral fallopian tubes and ovaries, anterior vaginal wall, and urethra are removed en bloc (Fig. 24-7). This is warranted for an invasive posterior bladder wall cancer in which orthotopic urinary diversion is not planned.

TABLE 24-1. Radical cystectomy in a woman: extent of dissection

FIG. 24-7. (A–C) The classical anterior pelvic exenteration includes removal of the bladder, uterus, bilateral fallopian tubes and ovaries, anterior vaginal wall, and urethra. An incision is made with cautery at the apex of the vagina. This is often facilitated with a Betadine-soaked sponge stick placed in the vagina on upward traction. The incision should be as close as possible to the posterior aspect of the cervix. The incision is then carried around laterally along the anterolateral aspect of the vagina. There is commonly a rich blood supply to the lateral aspect of the vagina, and this is most easily controlled with multiple suture ligatures in a stepwise fashion. The incision stops just before the endopelvic fascia. (D) With this dissection, the entire anterior vaginal wall, posterior bladder wall, and urethra are removed en bloc.

After division of the lateral pedicles on both sides, an incision is made in the peritoneum down to the rectal vaginal cul-de-sac. Blunt dissection in the midline mobilizes the posterior vaginal wall; this mobility will allow the posterior vaginal wall to be rolled anteriorly for vaginal reconstruction. A Betadine-soaked sponge is placed in the vagina, elevating the apex of the vagina just posterior to the cervix. Cautery is used to open the apex of the vagina in the midline; this incision is carried laterally down the anterior vaginal wall on each side ( Fig. 24-7A, Fig. 24-7B and Fig. 24-7C). Venous bleeding from the incised vaginal

wall and adjacent tissue may be heavy, and multiple suture ligatures with 2-0 Vicryl provide for hemostasis ( Fig. 24-7C). This dissection continues to near the bladder neck. The endopelvic fascia is not opened now; the dissection moves to the perineum after reasonable homeostasis has been ensured in the pelvis. The labia are retracted laterally with suture ligatures. Army–navy retractors or a self-retaining retractor provides exposure to the urethral meatus. An inverted U-shaped incision is made around the urethra and the urethra is mobilized anteriorly and laterally ( Fig. 24-8). Returning to the pelvic approach, the endopelvic fascia is incised on each side. Suture ligatures are placed in the venous plexus anterior to the urethra, analogous to control of the dorsal venous plexus in men. Then, from below, the anterior vaginal wall posterolateral to the urethra is divided to connect with the pelvic dissection, which allows removal of the entire specimen.

FIG. 24-8. Attention is now turned to the perineum, where an inverted U-shape incision is made around the urethral meatus and the anterior and lateral aspects of the urethra are mobilized. Dissection returns to the pelvic exposure, where suture ligatures are placed in the venous plexus anterior to the urethra, analogous to control of the dorsal venous complex in the man. The endopelvic fascia is then opened, and the incisions in the anterolateral vaginal wall are connected between the perineum and the pelvic dissection. This removes the entire specimen.

The vagina is reconstructed by rotating the apex of the posterior vaginal wall anteriorly to create a foreshortened vagina that maintains the previous width ( Fig. 24-9A,B). A stay suture in the apex of the posterior vaginal wall brings the vaginal wall to the perineum; this flap of vagina is sutured to the periurethral vaginal tissue anteriorly in the midline and then sequentially on each side. After two to three interrupted sutures are placed on each side from the perineum, additional sutures higher up on the vaginal wall are more easily placed from the pelvic exposure. A watertight closure provides optimal homeostasis of the paravaginal tissue. Closed-suction drains are left in the pelvis and brought out through separate stab wounds on the abdominal wall. A vaginal pack soaked in Betadine is left in the vagina for 24 hours.

FIG. 24-9. (A) Several methods of closure of the vagina are feasible. One that appears to supply strong support of the pelvis uses the posterior vaginal wall as a flap to create a neovagina. This foreshortens the vagina but does not narrow it as a side-to-side closure does and thus provides for a better return of sexual function, if appropriate. A suture is placed in the apex of the posterior vaginal wall from above; this is used to bring the posterior flap of the vagina down to the perineum. Several sutures are placed from below, incorporating a full thickness of the periurethral vaginal tissue to the posterior vaginal wall flap. Sutures closer to the apex of the vagina are more easily placed from the pelvic exposure. (B) Diagrammatic illustration of the soft tissue removed with the anterior vaginal wall, bladder, and urethra. The posterior vaginal wall is then rotated anteriorly down to the perineum.

For a tumor located high on the posterior wall, and when orthotopic diversion is anticipated, the uterus may be removed en bloc with the bladder, but the anterior vaginal wall need not be removed. For a non- or minimally invasive cancer in the bladder and planned orthotopic diversion, the uterus need not always be removed, and a plane can be established between the posterior bladder wall and the anterior vaginal wall. Care must be taken to enter the proper plane adhering to the anterior vaginal wall. Dissection too close to the bladder causes additional bleeding. If the urethra is to be used for reconstruction, the above dissection must be modified. The key points are: (a) do not open the endopelvic fascia and thus avoid disruption of the support of the external sphincter; (b) avoid dissection of the lateral wall of the vagina to prevent injury to the neurogenic innervation of the rhabdoid sphincter; and (c) remove the bladder neck entirely to minimize postoperative urinary retention. Once the bladder has been mobilized off the vagina down to the bladder neck, fine sutures are used anteriorly in the periurethral tissue as necessary for homeostasis of the venous plexus. The urethra is amputated sharply at the junction with the bladder neck, avoiding distal mobilization or dissection of the urethra ( Fig. 24-10). After the bladder has been removed, exposure is ideal for the enterourethral anastomosis, using fine absorbable sutures with small bites on the urethra; eight to ten sutures are generally necessary using 3-0 Monocryl. Mobility of the intestinal reservoir to the urethra is not a problem in the woman as it may be in the man.

FIG. 24-10. In the situation in which an orthotopic diversion is planned, neither the endopelvic fascia nor the anterior venous plexus is disturbed. An incision is made in the anterior urethra just distal to the junction with the bladder neck. The urethra is divided completely, and the dissection of the bladder off the anterior vaginal wall can be done in either an antegrade or a retrograde fashion.

In the initial experience with orthotopic diversion, the concern for stress incontinence was such that anterior urethral fixation sutures were placed to prevent

hypermobility of the urethra. This maneuver is not only not necessary (unless documented stress incontinence from hypermobility is evident preoperatively) but is probably counterproductive by contributing to increased urinary retention or “hypercontinence.” A possible additional mechanism for postoperative urinary retention may be exacerbation of a preexisting but previously insignificant cystocele. After the cystectomy, the urethra is fixed anteriorly, and the patient voids by Valsalva maneuver after relaxing the external sphincter. If a cystocele is present, the increased abdominal pressure needed for voiding across the fixed urethra and bladder neck could be blunted by the cystocele. This particular circumstance has been noted as a possible contributing factor for urinary retention in two of our first eight female patients undergoing orthotopic diversion. Confirmation of the pathophysiology, appropriate preoperative recognition, and preventative measures useful at the time of reconstruction are lacking. Postoperative Care Excellent pain control, early ambulation, judicious use of diuretics as needed to combat fluid retention, and pulmonary toilet are important. If a conduit has been used for the diversion, the conduit is kept decompressed with a catheter for 6 to 7 days. If a neobladder or continent cutaneous diversion is used, the catheter is left indwelling for 3 weeks and is removed in the outpatient area. A cystogram is no longer routinely obtained. Education of the inpatient nursing staff and the patient on frequent irrigation of the catheter to prevent plugging by mucus is crucial.

OUTCOMES
Complications A cystectomy is a difficult operation. Under the 1995 relative value (RVU) scale rating, cystectomy and continent diversion are ranked as the most difficult procedures in urology, ranked at 75.44 RVUs compared with 54.27 for radical retropubic prostatectomy and 40.80 for radical nephrectomy. A mortality rate of 1% to 2% is feasible and should be a standard to strive for. Twenty to thirty percent of patients will have a complication delaying discharge. Currently, our mean and median length of stay for 15 women undergoing cystectomy in 1995 and 1996 was 9.4 and 8.5 days, respectively (range 7 to 16 days), including six with an orthotopic diversion. Many of the specific complications after a cystectomy are a consequence of the urinary diversion and thus beyond the specific scope of this chapter. Complications from the cystectomy include bleeding and subsequent coagulation abnormalities and rectal injury. In my experience, female cystectomies tend to be associated with more blood loss than those in men, and although special blood products such as platelets and fresh frozen plasma are rarely needed, they may be lifesaving. A rectal injury should be extremely rare in women and seen only in association with previous surgery or radiation therapy. The postoperative care after cystectomy requires a diligence over and above that seen with other urologic procedures. Some complications are preventable. A regimented, reproducible “game plan” for the technique of cystectomy and diversion is enormously helpful to prevent errors during a 4 to 6 hour operation. Some complications are not preventable, but recognition early in their evolution may drastically minimize the negative consequences, and a high index of suspicion is essential. Early recognition of a complication may prevent a cascade of other successive complications, which may ultimately lead to more morbidity or the patient's death. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Freeman JA, Esrig D, Stein JP, et al. Radical cystectomy for high risk patients with superficial bladder cancer in the era of orthotopic urinary reconstruction. Cancer 1995;76(5):833–839. Harris RE, Chen-Backlund J, Wynder EL. Cancer of the urinary bladder in blacks and whites. Cancer 1990;66:2673–2680. Holmiing S, Borghede G. Early complications and survival following short term palliative radiotherapy in invasive bladder carcinoma. J Urol 1996;155:100–102. Koch MO, Smith JA Jr. Influence of patient age and comorbidity on outcome of a collaborative care pathway after radical prostatectomy and cystoprostatectomy. J Urol 1996;155:1681–1684. Leibovitch I, Avigad I, Ben-Chaim J, et al. Is it justified to avoid radical cystoprostatectomy in elderly patients with invasive transitional cell carcinoma of the bladder? Cancer 1993;71(10):3098–3101. Montie JE, Pavone-Macaluso M, Tazaki H, et al. What are the risks of cystectomy and the advances in perioperative care? Int J Urol 1995;2(Suppl 2):89–104. Montie JE. Orthotopic bladder replacement in women. In: Webster GC, Goldwasser B, eds. Urinary diversion: Scientific foundations and clinical practice. Oxford: Blackwell Scientific Publications, 1995, 193–199. Stein JP, Cote RJ, Freeman JA, et al. Indications for lower urinary tract reconstruction in women after cystectomy for bladder cancer: a pathological review of female cystectomy specimens. J Urol 1995;154:1329–1333. Stein JP, Stenzl A, Esrig D, et al. Lower urinary tract reconstruction following cystectomy in women using the kock ileal reservoir with bilateral ureteroileal urethrostomy: initial clinical experience. J Urol 1994;152:1404–1408. Stenzl A, Colleselli K, Poisel S, et al. Rationale and technique of nerve sparing radical cystectomy before an orthotopic neobladder procedure in women. J Urol 1995;154:2044–2049.

Chapter 25 Bladder Diverticulectomy Glenn’s Urologic Surgery

Chapter 25 Bladder Diverticulectomy
J. M. Gil-Vernet

J. M. Gil-Vernet: Catedra de Urologia, Facultad de Medicina, 08036 Barcelona, Spain.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Intravesical Diverticulectomy Combined Intravesical and Extravesical Diverticulectomy Outcomes Complications Results Chapter References

A bladder diverticulum is the protrusion of mucosa through the detrusor muscle fibers as a result of a structural defect (congenital or primary diverticulum) or of chronic dysfunction of bladder voiding (diverticulum secondary to obstructive pathology of the lower urinary tract). The diverticulum wall is composed of the following layers from inside out: mucosa, subepithelial connective tissue or lamina propria, isolated and thin muscle fibers, and adventitial tissue ( Fig. 25-1 and Fig. 25-2). The most frequent causes provoking an increase in bladder voiding pressure and the eventual formation of diverticulum are benign prostatic hyperplasia, urethral strictures, contracture or sclerosis of the bladder neck, urethral valves, and vesicosphincteric dyssynergy. The diverticula are located in the weakest points of the bladder, such as the ureteral hiatus (paraureteral or Hutch diverticulum) and both posterolateral walls. 1,2 and 3,5,7

FIG. 25-1. Posterior view of bladder and diverticulum. (1) Ampulla of vas deferens. (2 and 2¢) Ureters. (3) Posterior longitudinal bundle of the outer layer of the detrusor. (4) Diverticulum. (5) Circular fibers of the middle layer of the detrusor around the diverticular neck. (From Gil-Vernet S, 1948.)

FIG. 25-2. Lateral view of bladder and diverticulum. (1) Seminal vesicle. (2) Ureter. (3) Prostate. (4) Anterolateral longitudinal fibers of the outer layer of the detrusor. (5) Diverticulum. (6) Fine circularly oriented fibers around the diverticular neck. (From Gil-Vernet S, 1948.)

DIAGNOSIS
Diverticula are most commonly found on ultrasonography performed for the study of a prostatic syndrome in men or repeated urinary infections in women. Echography is highly useful for assessing whether a diverticulum is inhabited by lithiasis or tumor, although endoscopic examination should be performed if intradiverticular pathology is suspected or hematuria is present. Cystograms obtained by excretory urography or by retrograde instillation of contrast medium may provide the same information as ultrasonography with regard to the number, location, size, and urinary retention volume of the diverticulum. However, a voiding cystourethrogram with lateral and oblique projections will be indispensable when a urethral obstructive cause is suspected or in congenital cases to rule out possible vesicoureteral reflux. 8,10 Differential diagnosis should be established with the “pseudodiverticular” images observed in cystograms: bladder ears, hourglass bladder, and vesical hernias. On ultrasound, they should be differentiated from urachal cysts, prostatic utricle cysts, or Müllerian duct cysts and blind-ending bifid ureters. Other less-frequent congenital anomalies should also be considered, such as vesicourachal diverticulum, incomplete bladder duplication, and septation of the bladder, which may be mistaken in both examinations.10

INDICATIONS FOR SURGERY
The presence of intradiverticular disease (tumor or lithiasis), spontaneous diverticular rupture, or complications related to the size (³4 cm diameter) or location of the diverticulum are absolute indications for open surgery. A large diverticulum may be the cause of deficient voiding and chronic urinary infection or obstruction of the ureter and even of the posterior urethra in children, whereas paraureteral or hiatal diverticulum are usually associated with different degrees of reflux. 4,5 With the aim of improving vesical voiding, we recommend the simultaneous resection of all bladder diverticula, even those of small size (1 to 3 cm diameter) if the patient must undergo open prostatectomy, cystolithotomy, ureteroneocystostomy, or Y-V plasty of the bladder neck. Similarly, a vesical diverticulum should never be operated on without previously or simultaneously correcting the cause of obstruction, whether anatomic or functional (neurogenic bladder), that provoked it.

ALTERNATIVE THERAPY
A “wait-and-see” approach may be adopted in children with asymptomatic small-sized (rare) congenital or paraureteral diverticula and with low-grade associated reflux. Saccules and small diverticula may be treated successfully by electrocoagulation of their mucosa with the ball electrode when the primary obstructive disease is endoscopically resolved. However, we do not consider the laparoscopic approach to a diverticulum to be indicated because it will not solve the cause and will

prolong surgery. 6,9

SURGICAL TECHNIQUE
The bladder is approached via an infraumbilical midline extraperitoneal laparotomy incision. The dissection is carried into the space of Retzius with a sponge stick, and anterior bladder wall and vesical neck are identified. After reflecting the peritoneum from the bladder dome, we normally perform transverse cystotomy at this level, as it provides better exposure of bladder contents and facilitates placement of a small self-retaining retractor and additional stay sutures. The trigone, both ureteral meati, the bladder neck, and all possible diverticular orifices are clearly visualized from the bladder dome opening. In cases of intradiverticular tumor, we instill 30 mg of mitomycin by urethral catheter before the surgery and carefully protect the surgical field with moist sterile cloths to avoid possible tumor contamination during diverticulectomy. The bladder mucosa should also be thoroughly inspected to rule out papillary tumors that may have gone unnoticed on the previous endoscopic examination. Diverticulum excision has been described in three different approaches: extravesical (V. V. Czerny, 1896), intravesical (H. H. Young, 1906), and the intravesical and extravesical combination (G. Marion, 1913). The most commonly used procedures and the points of technique that we use are the following. Intravesical Diverticulectomy If the diverticulum is small (£5 cm diameter), we perform intravesicalization and eversion of its wall, grasping and tractioning its bottom gently with an Allis or Pean-type clamp inserted through its neck. If this maneuver is performed carefully, and fibrosis secondary to infection is absent, the majority of these diverticula are rapidly and easily removed. The mucosa of the everted diverticular neck is divided using electrocautery, and the defect of the bladder wall is sutured with 3-0 chromic catgut using separate submucosal and muscular sutures. In case of a saccule, a fine ligature of the neck and resection of its everted mucosa will suffice. If this maneuver is not feasible because of peridiverticular adhesions, we proceed to sharply split the mucosa around the diverticular orifice and dissect with scissors as far as the periadventitial space. In this way, the diverticular neck remains separated from the bladder wall and is pulled toward the vesical cavity with Allis-type clamps. At the same time, the adventitial adhesions that fix the diverticular sac are freed gently with a small moist gauze, and the sac is drawn into the bladder. The bladder wall is then closed as mentioned previously ( Fig. 25-3).

FIG. 25-3. Technique of intravesical diverticulectomy.

Combined Intravesical and Extravesical Diverticulectomy In a large diverticulum complicated with peridiverticulitis or in a paraureteral location, it is obligatory to place a 7- or 8-Ch ureteral catheter in the corresponding side before dissection. This will avoid an inadvertent lesion of the ureter or at least facilitate its immediate repair. These diverticula must be excised by a combined intraand extravesical approach, first identifying and dissecting the diverticular neck. For this, the maneuver of inserting the surgeon's index finger into the diverticulum and gently tractioning the upper face of its neck toward the surface is very useful. We also recommend completely filling the diverticular sac with a moist gauze to unfold its wall and delimit its margins as accurately as possible. Dissection must begin at the diverticular neck, which is sectioned extravesically with electrocautery and separated from the bladder wall, whose orifice is sutured with 3-0 chromic catgut using extramucosal separate stitches ( Fig. 25-4 and Fig. 25-5).

FIG. 25-4. Procedure of combined intravesical–extravesical diverticulectomy.

FIG. 25-5. Procedure of combined intravesical–extravesical diverticulectomy.

Tractioning the edges of the diverticular mouth toward the surface with Allis-type clamps allows the sac wall to be dissected from neighboring tissue with scissors and a small moist swab. It should always be borne in mind that the ureteral course may have been modified by the great diverticular volume, and the ureter may be closely adhered to its wall if repeated infectious processes have occurred. This dissection will be very difficult if great peridiverticulitis is present, and it is more advisable

simply to denude it of its mucosal lining with fine scissors or with the cutting current and the ball electrode from inside the diverticular cavity and then place a suction drain within it (first described by Pousson in 1901 and Geraghty in 1922). The bladder wall is closed with absorbable 3-0 interrupted sutures. We leave an aspirating drain in the Retzius space and a urethral 18-Fr Foley catheter, which may both be removed after 5 or 6 days.

OUTCOMES
Complications The most serious specific complication of excision of a bladder diverticulum is an injury to the juxtavesical or pelvic ureter during dissection of large diverticulum. With prior placement of an ipsilateral ureteral catheter, this lesion will not go unnoticed by the surgeon and can be easily sutured with absorbable 5-0 or 6-0 separate stitches if it is a partial or incomplete section. If the ureter has been severely damaged, or its section is complete and near the vesical hiatus, the distal ureter must be abandoned, and it is preferable to carry out ureteral reimplantation following the technique of Leadbetter–Politano with or without vesical lateralization to the psoas muscle (“psoas hitch”). End-to-end suture of ureteral edges must never be performed in precarious conditions because it is highly likely that it will be complicated by urinary fistula or ureteral stenosis, which will further aggravate the situation. If the ureteral lesion is more extensive and located higher, and the bladder in turn is reduced in size and of limited mobility, we prefer to perform transureteroureterostomy and ureteral suture with the aid of the surgical microscope. Less serious complications include vesical urine leakage, which may cease spontaneously if the Foley catheter is maintained for some more days, providing the obstructive pathology has been resolved. If a urinary fistula is established, we advise closing it with a flap from the bladder wall itself. Results The excision of the diverticulum is generally curative for that particular lesion, although correction of the underlying cause (e.g., outlet obstruction) is required to prevent formation of additional diverticulum. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Gil Vernet S. Bladder diverticula. Urol Nephrol 1969;1:323. Gil Vernet S. Morphology and function of vesico-prostatourethral musculature. Treviso: Canova, 1968;21. Hutch JA. Anatomy and physiology of the bladder, trigone, and urethra. New York: Appleton Century Crofts, 1972;8. Jarow JP, Brendler CB. Urinary retention caused by a large bladder diverticulum: a simple method of diverticulectomy. J Urol 1988;139:1260. Miller A. The aetiology and treatment of diverticulum of the bladder. Br J Urol 1958;30:43. Parra RO, Jones JP, Andrus CH, Hagood PG. Laparoscopic diverticulectomy: preliminary report of a new approach for the treatment of bladder diverticulum. J Urol 1992;148:869. Peterson LJ, Paulson DF, Glenn JF. The histopathology of vesical diverticula. J Urol 1973;110:62P. Quirinia A, Hoffmann AL. Bladder diverticula in patients with prostatism. Int Urol Nephrol 1993;25:243. Vitale PJ, Woodside JR. Management of bladder diverticula by transurethral resection: reevaluation of an old technique. J Urol 1979;122:744. Walker RD. Bladder and bladder neck. In: Kelalis P, King L, Belman B, eds. Clinical pediatric urology, 2nd ed. Philadelphia: WB Saunders, 1985;513–519.

Chapter 26 Bladder Augmentation Glenn’s Urologic Surgery

Chapter 26 Bladder Augmentation
R. Duane Cespedes and Edward J. McGuire

R. D. Cespedes: Department of Urology/MKCU, Wilford Hall Medical Center, Lackland AFB, Texas 78236. E. J. McGuire: Division of Urology, University of Texas, Houston, Texas 77030. The opinions contained herein are those of the authors and are not to be construed as reflecting the views of the Air Force or the Department of Defense.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Bladder augmentation is the addition of a segment of bowel to the in situ bladder to increase capacity, improve compliance, or abate uncontrollable detrusor contractility. It is frequently used in the reconstruction of neurogenic bladders that have failed medical therapy or other conservative therapies. Augmentation cystoplasty has replaced cutaneous urinary diversion in this group of patients because of decreased morbidity of the procedure, fewer postoperative complications, the widespread use of clean intermittent catheterization (CIC), and improved postoperative quality of life. Additionally, it has been shown that patients with cutaneous urinary diversions, draining continuously via any segment of bowel, have a worse long-term outcome with regard to infections and upper tract deterioration than patients with a large low-pressure reservoir that uses the urethra as the continence mechanism. In some cases, such as chronic, intractable interstitial cystitis with a small bladder capacity and severe symptoms, a supratrigonal cystectomy may be performed, and the bladder replaced by a bowel segment. Although we prefer ileum in most cases, many different bowel segments have been used, each with its own specific advantages and disadvantages. However, no bowel segment is clearly superior in all circumstances. The most important factor is detubularization of the bowel to reduce intravesical pressure from peristalsis or mass contractions. In most cases of neuropathic vesical dysfunction, the simplest low-pressure continent reservoir to construct involves the addition of bowel to augment the in situ bladder, utilizing the patient's own urethra as the continence mechanism.

DIAGNOSIS
Before consideration is given for an augmentation cystoplasty, all medical therapies and conservative treatments directed at improving detrusor compliance and increasing capacity should be exhausted. When these therapies have failed, the basic evaluation includes a cystometrogram, preferably using fluoroscopy, to evaluate bladder compliance, status of the bladder neck (open or closed at rest), and presence of vesicoureteral reflux. The bladder's functional characteristics such as capacity and compliance can sometimes dictate the type and length of bowel required. The evaluation of urethral function is problematic, particularly in patients with poor compliance or a defunctionalized bladder, where the urethra may deceptively appear worse than it truly is. The abdominal leak point pressure (ALPP) is a good method of determining urethral resistance to abdominal pressure as an expulsive force.5 If, in addition to the augmentation cystoplasty, one of the goals of the operative procedure is to achieve continence, the abdominal pressure required to cause leakage is essential. If that pressure is very low, 0 to 60 cm H 2O, then a sling procedure, an artificial urinary sphincter, or an injectable agent will be required to prevent leakage. If the resistance to abdominal pressure is high, perhaps 140 to 150 cm H 2O or more, then no procedure to improve urethral function is usually necessary. In the middle range, 60 to 140 cm H 2O, any of the above treatments and perhaps a urethral suspension in women, can be used to treat stress incontinence. It is important to recognize that creation of outlet resistance that completely resists intrareservoir pressure is inherently dangerous to the upper tracts and may contribute to rupture of the augmentation. 6 An upright cystogram, which demonstrates that continence is maintained at the bladder neck, can be a useful adjunct to the ALPP. Even if both studies are done and the urethra appears functional, leakage may still occur after augmentation cystoplasty when the bladder is very full and high intra-abdominal pressure is applied. Another important urethral function is that it must function as a compliant voiding conduit. If the patient is neurologically normal and voiding is anticipated after augmentation cystoplasty (as in selected patients with interstitial cystitis), the urethra should open normally on a voiding urodynamic study. Voiding pressures less than 30 to 40 cm H2O usually indicate that the conduit opens normally with voiding, and CIC may not be required postoperatively. Voiding pressures greater than 40 cm H2O generally indicate poor conduit function, and the patient will often require CIC after the augmentation cystoplasty. It is now clear that ureteral size (or dilation) has little to do with ureteral function, whereas, if ureteral peristalsis is present, as seen by ultrasound, fluoroscopy, or during a Whitaker perfusion test, the ureters will function adequately when placed into a low-pressure reservoir.

INDICATIONS FOR SURGERY
Bladder augmentation is a useful technique for the following indications: 1. Uncontrolled bladder contractility, most commonly for multiple sclerosis, spinal cord injury, and idiopathic urge incontinence. 2. Poor bladder compliance for neuropathic conditions after pelvic radiation therapy, intravesical chemotherapy, prolonged catheter drainage, or from untreated obstructive uropathy. 3. Prolonged bladder defunctionalization after long-term cutaneous vesicostomy drainage or bilateral cutaneous ureterostomy. 4. Patients previously diverted who are candidates for undiversion into a large, low-pressure reservoir. Patient selection remains an important issue before augmentation cystoplasty. Chronic renal failure (documented by creatinine clearance) is a relative contraindication to an augmentation because both small and large bowel resorb many urinary solutes that may deleteriously alter the metabolic status of the patient. In these patients, the use of stomach has been recommended. 1 All patients should be able, both mentally and physically, to perform CIC before a bladder augmentation, even if postoperative voiding is anticipated. This is especially important at both age extremes. If any question exists, it is better to place a cutaneous catheterizable stoma or, alternatively, a noncontinent stoma. No bowel segment is clearly superior to another in all circumstances, and the segment used is usually based on surgeon preference. In some circumstances, restrictions may exist, however. Stomach should probably not be used in patients with peptic ulcer disease, and large bowel should not be used if a history of ulcerative colitis, previous colon cancer, or diverticulitis exists. Similarly, in cases of extensive pelvic radiation, transverse colon or stomach may be preferable to small bowel. Finally, consideration for preservation of the ileocecal valve should be given in patients with myelodysplasia, as significant problems with diarrhea and fecal incontinence have been reported. 4

ALTERNATIVE THERAPY
In autoaugmentation of the bladder (also called partial detrusor myomectomy), the detrusor muscle over the dome and anterior wall is excised, which allows the bladder epithelium to distend outward, thereby improving storage capacity and detrusor compliance. 2 In our experience, the gains in bladder capacity and compliance

in patients with neuropathic voiding dysfunction are small, and we do not recommend its routine use in this group of patients. We continue to use this technique in selected patients with intractable urge incontinence, with favorable results. For patients in whom a simple, low-pressure cutaneous diversion is preferable, a noncontinent ileovesicostomy (bladder “chimney”) can be performed. Schwartz and colleagues reported excellent long-term results in 23 patients, with few complications. 8 Nonautologous tissues (Gore-Tex, Dacron, bovine dura, pericardium, etc.) have all been utilized to augment the bladder; however, complications with the anastomosis, infections, or stone formation preclude their routine use.

SURGICAL TECHNIQUE
Preparation of the patient is important, and all patients should undergo preoperative bowel preparation. In the neuropathic patient, chronic constipation is usually a problem, and 2 to 3 days of clear liquids and a full mechanical bowel preparation the day before the procedure will be necessary to insure removal of all solid stool. Nonneurogenic conditions and the use of small bowel for the augmentation may allow a less rigorous prep. If large bowel is used, oral nonabsorbable antibiotics such as neomycin and erythromycin base should be considered. Note that sometimes the desired segment of bowel is found intraoperatively to be unsuitable, and therefore, a full bowel prep is recommended for the majority of cases. A preoperative dose of IV antibiotics is also given with special consideration given to those patients with implanted prosthetic materials such as a ventriculoperitoneal shunt or orthopedic hardware. Other considerations include preoperative normalization of any metabolic or electrolyte disorders, documentation of sterile urine, and, in selected cases where colon will be used, a preoperative barium enema or colonoscopy. After preparation of the skin from xiphoid to genitalia, a urethral catheter is placed, a midline (preferably) or Pfannenstiel incision is made, and the retropubic space is dissected until the bladder is free of adhesions. In general, if a procedure to improve continence is necessary, it is performed first. Additionally, if low-pressure vesicoureteral reflux has been documented preoperatively, ureteral reimplantation should be considered. The ureters should be reimplanted into the bladder or into a colonic augmentation, as ureteral reimplantation into the ileum is tenuous and is not as favorable. We do not reimplant functional ureters that reflux with high bladder pressures because augmentation will decrease bladder pressures. A self-retaining retractor is placed, the bladder is filled with saline, and the peritoneum is dissected off the bladder to the level of the trigone ( Fig. 26-1). Using electrocautery, a U-shaped incision is made on the bladder starting 3 cm above the ureters, effectively creating an anteriorly based bladder flap ( Fig. 26-2). This technique avoids the hourglass configuration that can develop, making the augmentation little more than a poorly draining bladder diverticulum. The peritoneum is opened last to minimize third-space fluid loss and urine contamination of the peritoneal cavity. A 25- to 30-cm segment of ileum at least 15 cm away from the ileocecal valve is selected and marked with sutures. The ileum should easily reach the bladder without tension. The mesentery is cleared from both ends to create a window, and the ileum is divided using a standard stapling device ( Fig. 26-3).

FIG. 26-1. The peritoneum is dissected off the posterior bladder to the level of the trigone and 2 to 3 cm above the level of the ureters.

FIG. 26-2. A U-shaped incision is made on the bladder with the transverse portion just superior to the trigone and the limbs of the incision extending to the dome of the bladder.

FIG. 26-3. Approximately 15 cm proximal to the ileocecal valve, a 25-cm segment of bowel is selected, the mesentery cleared, and the segment removed using standard stapling techniques. The segment of ileum is oversewn at the ends to exclude the staples and then opened along the antimesenteric border.

The exact amount of ileum required varies among patients, but enough should be used to allow a minimum of 4 hours between catheterizations after the bowel is fully stretched over the ensuing months. Ileal continuity is then achieved using one of the hand-sewn or stapled techniques, and the mesenteric defect is closed. In Fig. 26-3, the ileal ends are oversewn with a running 2-0 chromic catgut to exclude the staples (to prevent stone formation), and the antimesenteric surface of the bowel is opened using electrocautery. Towels should be placed under the bowel, and the opened ileum irrigated into a kidney basin until clear. As seen in Fig. 26-4 and Fig. 26-5, the posterior wall of the ileum is folded back on itself and sutured together using running 2-0 chromic catgut. The required size of the augmentation opening is roughly measured, and the superior, anterior wall is partially closed with running 2-0 chromic catgut to match this opening ( Fig. 26-6). A large-bore suprapubic (SP) tube is placed through the bladder wall before placement of the augmentation on the bladder. The SP tube allows reliable postoperative drainage and irrigation of mucus until the suture lines are healed. The ileal segment is then sewn onto the opened bladder using running 2-0 chromic catgut with the initial suture placement

shown in Fig. 26-7 and the posterior closure seen in Fig. 26-8. The completed enterocystoplasty is seen in Fig. 26-9. A closed suction drain is placed near the suture line and brought through the skin on the side opposite the SP tube. The patient is closed in the usual manner. If a continence procedure has been performed, it is imperative that a catheter can be easily passed, otherwise the patient will be unable to catheterize postoperatively.

FIG. 26-4. The ileal segment is folded, and closure of the posterior wall is initiated using a running absorbable suture.

FIG. 26-5. The posterior wall of the augmentation is completely closed.

FIG. 26-6. The superior, anterior wall of the augmentation is partially closed. The size of the augmentation opening should roughly correspond to the size of the opened bladder.

FIG. 26-7. Initial suture placement for enterocystoplasty. Note that the bladder “flap” opens anteriorly.

FIG. 26-8. Closure of the anterior aspect of the enterocystoplasty.

FIG. 26-9. View of the completed enterocystoplasty.

Although this technique is our preferred technique, other methods of performing an augmentation exist. One such method involves splitting the bladder sagittally from just above the bladder neck and ending near the level of the ureters posteriorly to form a clam. A 25- to 30-cm segment of ileum is isolated and divided completely along the antimesenteric border 9 (Fig. 26-10). The posterior wall of the augmentation is closed with running 2-0 chromic catgut and is then either anastomosed to the bladder as a “patch” or folded again and partially closed to form a “cup.” A cup is especially useful if the patient's own bladder is very small but sometimes requires the use of up to 40 cm of bowel. In both cases, the anastomosis is started on the posterior wall until it is approximately one-third closed, and then the anterior wall is closed. The lateral walls are closed last, and any redundant bowel is closed to itself.

FIG. 26-10. Augmentation ileocystoplasty with formation of cup patch. (A) Isolation of a distal ileal segment. (B and C) Isolated ileal segment opened on antimesenteric border and double-folded to create a reservoir. (D) Cup-patch reservoir sutured to bladder incised sagittal plane.

Postoperative care is generally straightforward. Fluid and electrolyte management is important because of large third-space losses and drainage from the nasogastric tube, which remains in place until bowel function returns. The drain is removed after a few days, when drainage tapers off. The bladder is irrigated at least three times per day with 30 to 60 ml of saline to clear mucus. A cystogram is performed at 2 to 3 weeks, and the Foley is removed if no extravasation is noted. The patient begins CIC with the SP tube in place until the patient is proficient at CIC. At 3 to 4 weeks, the SP tube can usually be removed, and the patient continues CIC every 2 to 3 hours during the day and twice at night. Sometimes the augmentation takes several months to stretch, during which time frequent CIC is necessary. This may be distressing to the patient; however, liberal use of anticholinergics can help in many cases. Daily irrigation to clear mucus is essential, especially for the first few months. As capacity increases, the intervals can increase, with most patients able to go 4 to 5 hours between catheterizations during the day and once at night. Patients who are able to void must document consistently small postvoid residuals. Routine electrolytes, creatinine, BUN, and upper tract studies should be performed at regular intervals.

OUTCOMES
Complications A recent long-term study of 122 patients by Flood and colleagues reported an overall 28% early and 44% late complication rate in this difficult group of patients. 3 Most of the complications were minor and involved prolonged ileus, transient urinary extravasation, or stomal problems. Surgical interventions were necessary in only 15% of patients and were mainly stomal revisions. Small bowel obstructions occur in approximately 3% of patients. This is similar to the rate reported in urinary diversions. Bladder or kidney stones vary from study to study depending on the patient population and surgical techniques used. Stones commonly form secondary to retained mucus or exposed staples as a nidus. Routine bladder irrigation, treatment of infections, and staple exclusion at time of surgery minimize stone formation. Reservoir perforation is perhaps the most feared complication with reported rates of approximately 6%. 3,7 Fatalities are uncommon if diagnosed early. Metabolic problems such as metabolic acidosis or vitamin B 12 deficiency that are not medically treatable are uncommon if patients are properly selected and followed with appropriate labs. Carcinogenesis in all bowel segments has been reported. Although the risk to any individual patient is small, surveillance after 10 years should be considered. Voiding dysfunction is common even in nonneurogenic patients after augmentation cystoplasty. In a review by Flood and colleagues, 89% and 67% of neurologically intact men and women, respectively, required lifelong CIC. 3 These numbers reinforce the need to counsel patients on the high likelihood for lifelong CIC and preoperative demonstration of proficiency at CIC. Results A tabulation of success or failure after augmentation depends on the original reason for performance of the procedure. In patients with neuropathic bladders requiring improved compliance and capacity, an augmentation is almost uniformly successful. An augmentation is less successful in treating the symptoms of interstitial cystitis and, by itself, does not guarantee continence, especially in patients with high rates of intrinsic sphincter deficiency (ISD) such as myelomeningocele and radiation cystitis. A preoperative urethral evaluation is essential in diagnosing ISD in these patients. CHAPTER REFERENCES
1. Adams MC, Mitchell ME, Rink RC. Gastrocystoplasty: an alternative solution to the problem of urologic reconstruction in the severely compromised patient. J Urol 1988;140:1152. 2. Cartwright PC, Snow BW. Bladder auto-augmentation: early clinical experience. J Urol 1989;142:505. 3. Flood HD, Malhotra SJ, O'Connell HE, Ritchey ML, Bloom DA, McGuire EJ. Longterm results and complications using augmentation cystoplasty in reconstructive urology. Neurourol Urodyn 1995;14:297. 4. Gonzalez R, Cabral BHP. Rectal continence after enterocystoplasty. Dial Pediatr Urol 1987;10(12):1. 5. McGuire EJ, Fitzpatrick CC, Wan J, et al. Clinical assessment of urethral sphincter function. J Urol 1993;150:1452. 6. McGuire EJ, Woodside JR, Borden TA, Weiss RM. Prognostic value of urodynamic testing in myelodysplastic patients. J Urol 1981;126:205. 7. Rink RC, Woodbury PW, Mitchell ME. Bladder perforation following enterocystoplasty (abstract). J Urol 1988;139:234A. 8. Schwartz SL, Kennelly MJ, McGuire EJ, Faerber GJ. Incontinent ileo-vesicostomy urinary diversion in the treatment of lower urinary tract dysfunction. J Urol 1994;152:99.

9. Webster GD. Bladder augmentation and reconstruction. In: Glenn JF, ed. Urologic Surgery, 4th ed. Philadelphia: JB Lippincott, 1991;500–502.

Chapter 27 Vesicovaginal Fistula Glenn’s Urologic Surgery

Chapter 27 Vesicovaginal Fistula
Hubert G. W. Frohmüller

H. G. W. Frohmüller: Department of Urology, University of Würzburg School of Medicine, Würzburg D-97080, Germany.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Transvaginal Repair Suprapubic Approach Outcomes Complications Results Chapter References

Vesicovaginal fistulas (VVFs) have been recorded as a clinical entity since ancient times. The American surgeon Sims is commonly credited with having performed the first successful surgical repair of a VVF, as reported by him in 1852. 7 Vesicovaginal fistulas are extremely distressing as well as disabling to the patient. In industrialized soci-eties, VVFs occur most frequently as a result of iatrogenic injury at the time of gynecologic surgery, in particular hysterectomy. Other causes of VVFs are technically difficult surgical procedures or impaired wound healing (as a result of infection, neoplasia, previous radiation therapy, foreign bodies, or pelvic trauma) that is frequently complicated by medical conditions such as diabetes mellitus or atherosclerosis. In underdeveloped countries, especially in some parts of Africa, however, the most common cause of VVF is obstetric injury. The mechanism involved in prolonged labor is pressure necrosis of the bladder where it is caught between the obstructed head of the infant and the pubic bone. 4

DIAGNOSIS
The classic symptom of VVF is continuous leakage of urine per vaginam, with varying degrees of severity. This may develop immediately following the surgical procedure or be delayed, as in necrosis of the tissue or after radiation therapy, in the latter case months or even years after such treatment because of progressive obliterative endarteritis with resultant ischemia. 1 For diagnosis, a careful history, including the details of prior surgery, and a thorough physical examination are the usual prerequisites. An excretory urogram is advantageous in order to evaluate the upper urinary tract, particularly looking for associated ureteral injuries. The exact location and the size of the fistula and its relationship to the ureteral orifices are usually identified by cystoscopic examination. Vaginoscopy is very helpful in exactly localizing the fistula. If cystoscopy and vaginoscopy are performed at the same time, a ureteral catheter can be threaded through the fistula from the bladder into the vagina, thereby facilitating the recognition of the fistulous tract. Using both endoscopic inspections, it is important to determine the mobility of the fistulous segment of the bladder and also the degree of inflammation surrounding the fistula. In a patient with a history of prior pelvic neoplasia, a biopsy of the fistula site is mandatory to rule out recurrent tumor. Small fistulas that may escape detection by cystoscopy or vaginoscopy can be demonstrated by the following maneuver: a gauze sponge is placed into the vagina, and methylene blue or indigo carmine is instilled into the bladder. In case a fistula is present, the tampon will turn blue. Blue coloring after intravenous administration of indigo carmine is not necessarily diagnostic for the presence of a VVF because with this maneuver the leakage can also be induced by a ureterovaginal fistula.

INDICATIONS FOR SURGERY
The successful laparoscopic repair of VVF was reported recently. 5 The vast majority of VVFs, however, need repair by a surgical method, either transvaginally or by the suprapubic route. The choice of surgical approach is subject to the personal inclination of the urologist. Provided sound surgical principles are adhered to, i.e., complete excision of diseased tissue and reconstruction of the bladder using healthy, well-vascularized tissues, any surgical repair will succeed. Controversy still remains concerning the timing of fistula repair. Among patients with VVF there is quite understandably a strong desire for an expedient repair. The conventional wisdom, however, is to delay surgical intervention for 3 to 4 months. When ischemic necrosis complicated by inflammatory reaction of the tissue gives rise to the defect, delayed repair is obligatory. In clean iatrogenic injuries, e.g., in the course of hysterectomy, however, there is no disadvantage in early repair, i.e., as soon as possible after the fistula is discovered. 1

ALTERNATIVE THERAPY
A variety of treatment options exist for the closure of a VVF. In fistulas of very small caliber it may be worthwhile to attempt a conservative approach. 10 This can be done by curettage or cauterization of the fistulous tract or by application of silver nitrate as well as by simple drainage of the bladder via a transurethral catheter for a prolonged period of time. Such a conservative trial would certainly not prejudice future surgery and seems indicated in cases where delayed repair of the VVF is contemplated. There are several reports of successful closure of a VVF by introducing a fibrin clot into the fistula either cystoscopically or transvaginally or both ways.
2,6

SURGICAL TECHNIQUE
Successful treatment of VVF depends largely on careful preparation of the patient and on the ability of the urologist to vary the operative technique according to the requirements of each individual patient and to perform the technical details with meticulous precision. Transvaginal Repair Generally, the transvaginal approach to repair a VVF ( Fig. 27-1, Fig. 27-2, Fig. 27-3, Fig. 27-4 and Fig. 27-5) is simpler and less disturbing to the patient. The majority of patients with VVF can be treated by the vaginal route. Indications for using this approach are as follows:

FIG. 27-1. Vesicovaginal fistula repair utilizing an anterior vaginal wall flap for closure.

FIG. 27-2. Creation of anterior and posterior flaps.

FIG. 27-3. Excision of fistulous tract. Fig. 27-4. Transverse closure of vaginal wall.

FIG. 27-4. Transverse closure of vaginal wall.

FIG. 27-5. Mobilization of Martius flap. (Modified from Käser O, Iklé FA, Hirsch HA. Fistulas of the lower urinary tract. In: Friedman EA, ed. Atlas of gynecological surgery. New York: Thieme-Stratton, 1985.)

Fistulas less than 3 to 4 cm in diameter Tissues pliable Vaginal size normal or larger than normal No cancerous tissue involved No previous extensive radiotherapy The advantages of the transvaginal approach include: Avoidance of an abdominal incision Minimal blood loss Reduced postoperative morbidity Less postoperative discomfort to the patient Decreased length of hospitalization A disadvantage of performing the transvaginal procedure is the relative lack of familiarity with the surgical anatomy of this region by many urologists. Before repair, cystoscopy is carried out to confirm the position of the VVF and its relation to the ureteral orifices. Bilateral ureteral catheters are inserted, irrespective of the position of the fistulous tract. If possible, a ureteral catheter is introduced through the fistula into the vagina. After the removal of the cystoscope, an 18-Fr Foley catheter is inserted transurethrally, and the three previously mentioned ureteral catheters are tied to it just outside of the urethral meatus. Then the patient is placed in the hyperflexed dorsal lithotomy position. A weighted vaginal speculum is inserted, and the labia minora are sutured to the inner thighs in order to obtain good

exposure. A small Foley catheter is placed through the fistula into the bladder. In small fistulas the cystoscopically introduced ureteral catheter can be used advantageously in pulling the tip of the Foley catheter tied to its tip into the bladder. After its balloon has been inflated, traction can be applied to the catheter, which provides an additional means for exposure. The vaginal mucosa is incised circumferentially around the opening of the fistula, and the cicatricial or necrotic tissue of the fistulous tract is excised to the margin of fresh, healthy tissue after removal of the catheter. The defect is closed in three layers. The first layer, utilizing 3-0 polyglactin or polyglycolic acid sutures, ties the submucosa of the bladder thus approximating the mucosa without injuring it. The second layer, using the same suture type, ties the muscularis and the adventitia of the bladder. The third layer, utilizing 2-0 or 0 polyglactin or polyglycolic acid sutures, knits the vaginal mucosa. It is advisable to close successive layers in perpendicular directions, i.e., vertically versus transversely, in order to avoid overlapping suture lines. When difficulties arise, sutures with a 5/8; needle can be of advantage. Although no drain is used, a tampon is placed into the vagina and extracted on the second postoperative day. The ureteral catheters are left in place and connected to urine-collecting bags. They are withdrawn on the third postoperative day. The Foley catheter is removed 7 to 10 days after surgery. Alternatively, a suprapubic tube can be utilized. Uninterrupted catheter drainage during this time is of utmost importance. Antibiotics are given as long as the urethral or the suprapubic catheter is in place. The patient is instructed to abstain from intercourse for approximately 2 months. If the tissue to be closed is tenuous, or if the transvaginal repair is difficult, a vascularized fibrofatty labial segment can be utilized for interposition between bladder and vagina. This segment, called the Martius flap, 3 is easy to harvest because of its convenient location. Through a separate incision in the lateral aspect of the labia majora the underlying fat pad is mobilized and then pulled through a subcutaneous tunnel into the vaginal incision. There it is interposed between bladder and vagina with absorbable sutures 8 (Fig. 27-6 and Fig. 27-7).

FIG. 27-6. Fat pad pulled through to vaginal wound bed. (Modified from Käser O, Iklé FA, Hirsch HA. Fistulas of the lower urinary tract. In: Friedman EA, ed. Atlas of gynecological surgery. New York: Thieme-Stratton, 1985.)

FIG. 27-7. Flap sutured into position. (Modified from Käser O, Iklé FA, Hirsch HA. Fistulas of the lower urinary tract. In: Friedman EA, ed. Atlas of gynecological surgery. New York: Thieme-Stratton, 1985.)

Suprapubic Approach Indications for a suprapubic approach include: Associated pelvic pathology Cases where ureteral reimplantation may be required Limited access because of a high retracted fistula in a narrow vagina Some cases with multiple fistulous tracts Complications related to previous irradiation The confirmation of the location of the fistula by cystoscopic examination is the same as when the transvaginal approach is used. The patient is placed in a supine position, and a lower midline incision is made. The perivesical space is mobilized, and the peritoneum is retracted cephalad from the dome of the bladder. The bladder is opened by a longitudinal midline incision and then split posteriorly and downward toward the fistula. The ureters may be catheterized if desired. The fistulous tract is excised all the way into the vagina. The opening of the vagina is closed with interrupted 2-0 absorbable sutures (Vicryl or Dexon) in one or two layers. In uncomplicated cases the bladder is then closed in two layers with continuous sutures of the same material. It is important to mobilize the vagina as well as the bladder flaps widely in order to avoid any tension on the suture lines ( Fig. 27-8, Fig. 27-9, Fig. 27-10, Fig. 27-11 and Fig. 27-12).

FIG. 27-8. Longitudinal cystotomy down to fistula site.

FIG. 27-9. Incision around fistula in a racquet fashion.

FIG. 27-10. Excision of fistula tract.

FIG. 27-11. Closure of vaginal and bladder walls.

FIG. 27-12. Closure of cystotomy.

In previously irradiated tissue or in complicated cases, instead of simple closure of the vagina and the bladder, it is safer to use interposition of an omental graft order to prevent recurrent fistula formation ( Fig. 27-13, Fig. 27-14 and Fig. 27-15).

9

in

FIG. 27-13. Mobilization of omentum based on right gastroepiploic artery. (Modified from Turner-Warwick RT. Urinary fistulas in the female. In: Walsh PC, Gittes RF, Perlmutter AD, et al, eds. Campbell's urology, 5th ed. Philadelphia: WB Saunders, 1986;2718.)

FIG. 27-14. Extra length of omentum created by a pedicle, maintaining continuity of gastroepiploic vessels. Patch of omentum brought to the pelvis via a retroperitoneal course dorsal to the mobilized ascending colon.

FIG. 27-15. Interposing omentum between bladder and vagina. (Modified from Turner-Warwick RT. Urinary fistulas in the female. In: Walsh PC, Gittes RF, Perlmutter AD, et al, eds. Campbell's urology, 5th ed. Philadelphia: WB Saunders, 1986;2718.)

The blood supply of the omentum from the left gastroepiploic artery and branches of the splenic artery, and from the right gastroepiploic artery and the gastroduodenal artery, can be observed by transillumination. This dual blood supply permits mobilization of the omentum from the greater curvature of the stomach. In some cases it is possible to extend the lower margin of the omentum down to the fistula without mobilization of the omentum from above. In most cases, however, is has to be dissected from the transverse colon. Either the left or the right gastroepiploic artery is divided between 3-0 silk ligatures close to the stomach until a well-vascularized omental flap is created, long enough to be brought down to the pelvis without tension. The omental apron is transferred to the pelvis extraperitoneally, dorsal to the ascending colon. A portion of the flap is interposed between anterior vaginal wall and posterior bladder wall and tacked in position with absorbable sutures. The omentum must extend well beyond the margins of the repairs. Either a urethral catheter or a cystotomy tube can be used for bladder drainage. If ureteral catheters had been inserted, they should be left in place to keep the wound free of urinary drainage. Retrovesical drains are placed, and then the incision is closed in the usual fashion. The ureteral catheters and the drains are removed around the fifth postoperative day. The removal of the suprapubic tube or the urethral catheter takes place 2 weeks postoperatively. For interposition between bladder and vagina, a peritoneal flap can be used instead of omentum. It is usually readily available. In large fistulas, however, particularly in radiogenic ones, a pedicled omental apron is the optimal tissue. In a small, uncomplicated VVF a simple closure in layers, as described previously, will usually suffice without the necessity of resorting to the use of peritoneum.

OUTCOMES
Complications Most authors agree that the vast majority of VVFs can be successfully repaired transvaginally. It is of particular advantage that the complication rate is definitely less when this route is used rather than the suprapubic, transabdominal approach. The complication of most concern is a recurrent urine leak. One can try to manage it by reinserting a catheter in order to drain the bladder for 3 or 4 weeks. If this fails to close the fistula, a new attempt at reconstruction is inevitable. Reasons for failed repair are insufficient debridement of nonviable and scar tissue before closure, excessive tension on the suture lines, inadequate closure of dead space, postoperative bladder distension, e.g., because of a plugged catheter, abscess formation, and poor tissue healing as a result of persistent or recurrent neoplasia or radiation-induced damage. After an abdominal approach, it is not uncommon to encounter a significant period of ileus, particularly following extensive omental mobilization. Delayed healing and wound infections occur more frequently after transabdominal fistula repair than following the transvaginal procedure. Bowel obstruction secondary to adhesions is a typical, if infrequent, complication of the transabdominal procedure. It is not seen when the transvaginal method is used. Results At the author's institution, 64 vesicovaginal fistulas were treated during the 30-year period between 1966 and August 1996. The 64 VVFs were caused by abdominal hysterectomy in 42 cases, vaginal hysterectomy in 12, radiation therapy in seven, obstetric complications in two, and colporrhaphy in one patient. In 60 patients (94%), the transvaginal approach was used; in four patients the transabdominal route (6%). Fourteen of the 60 patients had had prior attempts to repair the VVF. Of these 14 patients, six had undergone a transabdominal attempt, five a transvaginal procedure, and three a combined transvaginal and transabdominal surgery. In 55 of the 60 transvaginally repaired VVFs (92%), the primary closure was successful. In five patients (8%), a secondary transvaginal procedure became necessary for a successful closure. The vast majority of vesicovaginal fistulas can be closed with a proper surgical approach and meticulous attention to detail by the urologist. In those few unfortunate patients in whom every attempt fails to successfully repair the fistula—which is usually the result of irradiation—urinary diversion can become necessary as a last resort to improve their quality of life. 11 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Blandy JP, Badenoch DF, Fowler CG, Jenkins BJ, Thomas NWM. Early repair of iatrogenic injury to the ureter or bladder after gynecological surgery. J Urol 1991;146:761. Hedelin H, Nilson AE, Teger-Nilsson AC, Thorsen G. Fibrin occlusion of fistulas postoperatively. Surg Gynecol Obstet 1982;154:366. Martius H. Die operative Wiederherstellung der vollkommen fehlenden Harnröhre und des Schliegssmuskels derselben. Zentralbl Gynäkol 1928;52:480. Naslund MJ. Vesical fistulae. In: Marshall FF, ed. Operative urology. Philadelphia: WB Saunders, 1991;151–154. Nezhat CH, Nezhat C, Rottenberg H. Laparoscopic repair of a vesicovaginal fistula: a case report. Obstet Gynecol 1994;83:899. Pettersson S, Hedelin H, Jansson I, Teger-Nilsson AC. Fibrin occlusion of a vesicovaginal fistula. Lancet 1979;i:933. Sims JM. On the treatment of vesicovaginal fistula. Am J Med Sci 1852;23:59. Steidle CP, Rowland RG. Bladder fistulae. In: Droller MJ, ed. Surgical management of urologic disease. An anatomic approach. St Louis: Mosby Year Book, 1992;619–628. Walters W. Omental flap in the transperitoneal repair of recurring vesicovaginal fistulas. Surg Gynecol Obstet 1937;64:74. Woo HH, Rosario DI, Chapple CR. The treatment of vesicovaginal fistulae. Eur Urol 1996;29:1. Zimmern PE, Hadley HR, Staskin DR, Raz S. Genitourinary fistulae. Vaginal approach for repair of vesicovaginal fistulae. Urol Clin North Am 1985;12:361.

Chapter 28 Vesicoenteric Fistula Glenn’s Urologic Surgery

Chapter 28 Vesicoenteric Fistula
Luis Gonzalez-Serva

L. Gonzalez-Serva: Jet International M-211, Miami, Florida 33102.

Diagnosis Clinical Manifestations Preoperative Assessment Indications for Surgery Alternative Therapy Surgical Technique Repair of Rectovesical Fistula (York-Mason Procedure) Outcomes Complications Results Chapter References

Although rare in daily clinical practice, the diagnosis of an acquired communication or fistula between the urinary tract and the intestines is always a challenge to the ingenuity and clinical acumen of the urologist and, later, to the good surgical judgment as to whether surgery is indicated or not, and whether to do it in a single or in several stages. Most manifestations of the condition occur toward the urinary tract, with passage of fecal matter and flatus from the gut to the bladder rather than in the opposite direction, directing these patients toward a urologist who will initiate the proper workup and participate actively in the surgical care. Vesicoenteric fistulas are usually the consequence of inflammatory processes of the bowel, mainly diverticulitis; colorectal cancer and, more rarely, carcinoma of the bladder; trauma; and iatrogenic fistulas between bladder and sigmoid colon as a result of prostatic surgery, either endoscopic or open. The specific causes of vesicoenteric fistulas are listed in Table 28-1.

TABLE 28-1. Specific causes of vesicoenteric fistulas

Diverticulitis has long been identified as the most common cause of vesicoenteric fistula, ranging in incidence from 36% to 85%, with an average of 50%, with colonic malignancy, granulomatous bowel disease, and radiation therapy accounting for the majority of the remainder in a large series. 1,5 Before 1950, one in 3,000 surgical hospitalizations and 23% of patients who required surgery for diverticulitis had vesicoenteric fistulas reflecting late diagnosis and treatment. 3 Diverticulosis occurs in at least half of patients older than 60 years old. Inflammation of these mucosal outpouches occurs 1.5 to 3.0 times more frequently in men than in women, whereas vesicocolonic fistula is up to five times as often in men as in women, indicating that the uterus may play a role in the prevention of fistulization to the bladder. Moreover, in women with diverticulitis, vaginocolonic fistulas are much more frequent than vesicolonic fistulas, 7.5:1.0, further supporting the role of the uterus as a protective anatomic barrier. 3 Colorectal adenocarcinoma is the cause of fistula in 10% to 16% of cases, being the second most frequent entity after diverticulitis. Half of these neoplastic fistulas are vesicosigmoideal. 1 Other pelvic neoplasias, such as bladder and cervical carcinoma, are less frequent causes of fistula, ranging in incidence between 5% and 10% each. 1 Enterovesical fistula is a rare complication of pelvic radiotherapy for gynecologic cancer, with recurrent neoplasia being the most common cause of fistulization, though some patients without recurrence will develop a fistula. High-risk factors for radiation morbidity include previous surgery, pelvic inflammatory disease, adjuvant hyperbaric oxygen, or locally high doses of radiotherapy caused by suboptimal geometry and technique. The range of radiation morbidity is variable, some patients had small fistulas, others had extensive fistulization and radionecrosis. The site of the radiation-induced fistula varies: colovesical fistulas, enterovesical fistulas, and some patients with fistulas involving both the small and large bowel. 4 Rarely, the cause of the vesicoenteric fistula is transitional-cell carcinoma of the bladder. Inflammatory bowel disease accounts for approximately 3.5% of cases of vesicoenteric fistulas, with Crohn's disease being the most common cause. 1,8 Generally, patients with Crohn's disease were an average of 20 years younger than the patient with cancer or diverticulitis. 3,6 Crohn's disease is more likely to produce multiple fistulas: ileovesical, ileoascending colonic, and ileosigmoidal. Other less common inflammatory processes associated with vesicoenteric fistulas include appendiceal abscess, tuberculosis, and actinomycosis.

DIAGNOSIS
Clinical Manifestations Despite the fact that most vesicoenteric fistulas are the result of primary bowel disease, patients most often have urologic symptoms, with a significant absence of localizing intestinal symptoms in up to half of patients. A wide range of symptoms exist that are of variable intensity but generally of insidious nature, making the diagnosis of a fistula elusive and delayed. Gastrointestinal symptoms of vesicoenteric fistulas include abdominal pain, diarrhea, constipation, intestinal obstruction, and acute abdomen, which are manifestations of the underlying pathology. Urologic symptoms may be lower tract such as bladder irritation, frequency, urgency, dysuria, and hematuria; those related to recurrent urinary tract infections; and upper tract with accompanying systemic symptoms, fever, and chills. The most relevant and pathognomonic manifestations, pneumaturia and fecaluria, passage of gas and feces, respectively, per urethra, occur in 40% to 83% of patients and immediately point toward the diagnosis. However, their absence does not rule out the entity; moreover, even if present, these symptoms may be overlooked or not recognized if the patient is not directly questioned or instructed to search for them carefully. Pneumaturia occurs twice as commonly as fecaluria because the passage of solid particles through the fistula is difficult. Interestingly, passage of urine per rectum is rare because of the higher colonic pressure and occurs only after a diverting colostomy or in association with severe bladder outlet obstruction. Bladder irritative symptoms associated with known diverticulosis and bouts of diverticulitis or Crohn's disease should arouse suspicion of a prodromic stage of a vesicoenteric fistula. Positive physical findings are few. An abdominal mass is felt in a third of patients, and abdominal tenderness and guarding, indicating local peritonitis, is seen in one-third to half of the patients. Rarely, one can find a cutaneous fistula and acute scrotum secondary to urinary tract infection. Urinalysis may show striated muscle

fibers (rhabdomyocytes) derived from undigested meat residue from stool, in addition to the usual findings of urinary infection. Urine cultures generally grow a single species, predominantly Escherichia coli, instead of the falsely expected mixed growth. Temporary closure of the fistula by edema may render a urine culture negative in 5% to 10% of patients. 1 Preoperative Assessment Once the diagnosis of vesicoenteric fistula is entertained, it is important to document both the presence and site of the fistula. Although it is more frequently found between the sigmoid colon and bladder, other locations (ileal, appendicular, colic, rectal, and Meckel's diverticulum) must be ruled out. Additionally, any involvement of adjacent viscera needs to be identified. Enterovesical fistula is a challenging entity, the etiology of which may be suspected from the patient's history or physical assessment. The definite diagnosis of enterovesical fistula can remain difficult despite the many methods of diagnosis, including functional, imaging, and endoscopic studies. Functional Studies 1. Activated charcoal, nonabsorbable through intestinal mucosa, administered orally or through the defunctionalized end of a colostomy, produces charcoaluria. This method is easy, inexpensive, noninvasive, and yields the highest rate of positive results. 2. Visible dyes (phenazopyridine, methylene blue, indigo carmine, Congo red) administered through enemas in undiverted patients or through urethral or suprapubic catheter, searching for stained stools. 3. Search for pneumaturia with the patient voiding in a water-filled tub. 4. Oral nonabsorbable radioisotope, e.g., 51Cr-labeled sodium chromate, producing quantitated radioactivity in urine over 2 to 3 days. Imaging Studies 1. Intravenous pyelogram is generally of little help. Rarely, one sees dye in the colon; more frequently, some nonspecific abnormalities related to the primary process or associated inflammatory process, gas in urinary tract, bladder wall irregularities, ureteral strictures, or hydronephrosis. 2. Retrograde and voiding cystogram—twice as successful as an intravenous pyelogram in diagnosing this condition, 35% to 44%—and one of the mandatory methods of evaluation. 3. Retrograde pyelogram, used only if there is any suspicion of ureteral involvement in the process. 4. Barium enema is unequivocally positive in only 20% to 42% of cases. However, it remains a very important study to delineate the process that is producing the fistula. Postevacuation films may enhance the sensitivity of the study. Also, the patient can be instructed to collect the voided urine in a plastic or glass container over a few hours in an attempt to see the barium directly or by means of a radiograph of the container. A more refined version of this concept is the Bourne test, in which a drop of sediment of centrifuged urine is placed on an x-ray cassette and compared radiologically with a control urine sample. 5. Abdominal ultrasound has been used to diagnose complications of Crohn's disease such as bowel loop wall thickening, abscess, lymph node enlargement, and vesicoenteric fistulas. Abdominal ultrasound has been found to yield the first diagnostic information on inflammatory bowel with a high correlation to surgical findings. Also, it can differentiate pelvic complications of these intestinal conditions. 6. Upper gastrointestinal (UGI) series rarely are indicated unless the colon workup is normal or there is a suspicion of a complex fistula involving the ileum. 7. Computed tomographic scan (CT) of abdomen and pelvis is a very useful study, with the following positive findings in the cases examined: air within the bladder in 83%, thickening of the intestine and bladder wall at the site of the fistula in 100%, and paravesical mass in 87%. This justifies it as a preliminary step in any suspected case of vesicoenteric fistula. 9 A CT finding consistent with the diagnosis of appendicovesical fistula is calcification of a thickened bladder wall adjacent to the cecum on noncontrast CT, which is a fecalith in the lumen of the fistula. 2 A CT scan of the abdomen and pelvis is, therefore, recommended in the evaluation of the majority of patients with suspected enterovesical fistulas. 4 8. Magnetic resonance imaging (MRI) has been used in patients with Crohn's disease to evaluate cutaneous, deep perineal, or enterovesical fistulas or abscesses, with good correlation with clinical examination under anesthesia. It is more likely that a negative MRI correlates better with other ancillary methods. Endoscopic Studies 1. Cystoscopy is, by far, the most valuable study to diagnose and localize the fistula, with a success rate of 32% to 87%. 1 Nine of ten cystoscopies show mucosal abnormalities, feces, particulate matter, barium, and the so-called Pugh's villous reaction, papillomatous epithelial growth, which sometimes can be misdiagnosed as low-grade transitional cell carcinoma. 3 The fistula or suspicious area usually is located in the high left posterolateral wall when the primary cause is diverticulitis or cancer (colonic, vesical, or uterine). The fistula is usually located in the right or anterior wall in cases of Crohn's disease or lesions of the cecum or appendix.9 It sometimes is possible to catheterize the suspect tract and obtain radiographic studies. Likewise, whether or not an obvious fistulous opening is visualized, a biopsy specimen can be taken from the suspect area to rule out an urothelial, gynecologic, or intestinal malignancy. From 10% to 20% of patients with vesicoenteric fistulas have completely normal endoscopic results. 2. Endoscopic procedures of the lower gastrointestinal tract, colonoscopy or proctosigmoidoscopy, are of limited help. Perhaps the insufflation of air during these procedures may help uncover pneumaturia. These studies are indicated to evaluate the large bowel, to assess the magnitude of the primary problem (diverticula, neoplasia) and its sequelae (obstruction, mass) and to plan an optimal operation. Also, even if the fistula is the result of inflammatory bowel disease, there can be a concurrent bowel neoplasm in 3% to 14% of cases. 3 In summary, cystoscopy (60%) and cystography (44%) seem to be the most sensitive diagnostic studies. 5 Computerized axial tomography scanning, cystoscopy, charcoaluria, and barium enema are useful in making the diagnosis. 6 The IVP and colonoscopy are generally not useful procedures for the diagnosis of VE fistulas. 6 Other imaging techniques, though less effective for diagnosis, were useful in assessing the status of the GI tract and, sometimes, determining the etiology of the fistula.

INDICATIONS FOR SURGERY
Surgical therapy aimed toward resolution of the primary process and abnormal communication is required and feasible in most cases. Ideally, the surgeon should consider a primary resection in properly selected patients, without risking the repair or the life of the patient. All possible etiologies can be treated this way. A selection of criteria regarding patient selection and characteristics of the fistula and the bowel anastomosis must be met for primary (one-stage) resection to be successful (Table 28-2).

TABLE 28-2. Criteria for patient selection for one-stage repair

ALTERNATIVE THERAPY

On occasions, especially when the vesicoenteric fistula does not meet all above-mentioned criteria, it may be wise to treat preliminarily with a diverting colostomy in order to control the inflammatory process and avoid the continuance of infection and sepsis. Generally, this is done alone, and future procedures are planned. Traditionally, when a diverting colostomy has been done as a first stage, a second operation resolves the primary bowel process and the fistula proper (bowel resection plus partial cystectomy), and the colostomy is preserved rather than closed to protect the intestinal anastomosis. The closure of the colostomy can be done 4 to 6 weeks later as a third stage, to take full advantage of it. Although it is safe, at present this extended multistaged approach does not seem cost-efficient. Ideally, once the diverted patient is fully recovered of the acute event, one can proceed with excision of the fistula and reconstruction, with closure of the colostomy in the same session if local conditions allow it, in order to avoid a third operation. During a single-stage operation, an unexpected intraoperative finding may induce a change of plans and convert a one-stage procedure to a multistaged operation. In this case, the main problem, the fistula, is excised, bowel and bladder reconstruction is accomplished, and then the anastomosis is protected by a colostomy, which will be closed as a second stage. All these alternatives should be fully discussed with the patient. In debilitated patients with reasonable and comfortable life expectancies, a palliative colostomy may be the only procedure considered. In extremely debilitated patients, especially if there is unresectable or metastatic neoplasia, an expectant and supportive medical management (intermittent antibacterial treatment) may be all that is indicated. In one series one-tenth of the patients were not candidates for operation, and one-fourth of the patients did not undergo complete operative resolution and restoration of enteric and urinary continuity. 6

SURGICAL TECHNIQUE
Regardless of the site of the vesicoenteric fistula, the bowel needs to be prepared mechanically and bacteriologically. The patient is started in a low-roughage diet several days before entering the hospital, and cathartics and enemas are administered the day before surgery. Nonabsorbable oral antibiotics are given starting 24 hours for prophylaxis against both gram-negative and anaerobic bowel flora. It is advisable to initiate a program of parenteral hydration and antibiotic prophylaxis several hours before surgery, using drugs that cover gram-negative bacteria and, in the case of a colon fistula, anaerobics. If the patient is severely malnourished, oral or parenteral hyperalimentation can be started days to weeks before surgery and maintained through the postoperative period. Surgery is started by a midline or paramedian incision in all cases, regardless of whether a prior fecal diversion has been done, to allow a transperitoneal approach with careful assessment of the inflammatory mass and intervening viscera as well as the remainder of the bowel and other intra-abdominal organs. This incision affords cephalad extension if needed. The abdomen is entered, and all adhesions are lysed; small bowel contents are removed from the pelvis. The area of the fistula is identified, and the intestine is sharply dissected from its attachment to the bladder. In patients with Crohn's disease, the indication for surgery was the fistula alone in a third of the cases and the fistula plus another complication of the disease in two-thirds of the patients. 8 These include enteroenteral, ileogenital, and enterocutaneous fistulas and intra-abdominal abscesses. The surgical team must be prepared for this eventuality. If the patient has not been diverted previously, the surgeon must decide either to proceed with a primary reconstruction without colostomy or, if the conditions are less than ideal, to protect the anastomosis with a colostomy. In case of a previous colostomy, one either takes advantage of the colostomy and plans to close it several weeks after the reconstruction or closes it immediately as mentioned. Surgical treatment of a vesicoenteric fistula ideally consists of the excision of the diseased bowel, partial resection of the involved bladder, and interposition of a vascularized tissue between the two viscera ( Fig. 28-1). Urinary diversion is ordinarily attained with a urethral catheter. The use of a suprapubic catheter is optional and depends on the confidence of the surgeon in the vesical repair, local and urethral conditions, and presumed duration of urinary leakage. In case of ureteric involvement, stenting or ureteroureterostomy may be required. Bowel anastomosis is done in the standard one- or two-layer fashion or with a bowel stapler.

FIG. 28-1. Management of vesicoenteric fistula using bowel resection and restitution with primary closure of bladder.

After both systems are reconstructed, it is advisable to fill the potential dead space between them with a vascularized tissue to support the repair and improve tissue healing, which is impaired by inflammation and infection. An omental flap is ideal to accomplish this task, either by simply fixing it between bladder and bowel with tacking sutures if the omentum is long or by creating a pedicle flap based in the right gastroepiploic artery, separating the left gastric attachments. Also, one may interpose flaps of peritoneum, muscle, or fibroadipose tissue in less serious cases. Gold foil and lyophilized human dura have been used in some cases. Colovesical fistulas associated with diverticulitis can be treated laparoscopically in a one-stage repair. It has been suggested that this operation is safe, has minimal pain, absent ileus, and a short postoperative stay. 7 Inflammatory bowel disease, especially Crohn's disease, may produce filiform tracts, which can make the fistula localization difficult. Moreover, these patients are subject to more recurrences. Therefore, although simple separation of bowel and bladder with bowel resection and oversewing of the bladder wall may suffice, it may be better to proceed with a limited partial cystectomy. The involved bladder is resected to allow a two- or three-layer closure with absorbable sutures, bringing together edges of fresh, noninflamed tissue. If the defect seems too large, one may resort to rotated bladder flaps or rearrangements of bladder tissue as the ventrodorsal repair, wherein a longitudinal cystotomy is closed in a transverse manner. Rarely, a small bowel graft may be necessary to cover an unusually large defect, and in some cases, the cystotomy has been left open, and a urethral catheter left in place for at least 1 week with healing and no late recurrences. 8 Repair of Rectovesical Fistula (York-Mason Procedure) A special situation may arise when the fistula involves the bladder and rectum instead of the sigmoid colon, as observed after injuries sustained during prostatic surgery and pelvic fractures. Repair of these fistulas using a posterior sagittal, transanal, transrectal (modified York–Mason) approach has been advocated with excellent results. 10 This repair is possible without colostomy in a patient who has undergone a complete mechanical and bacteriologic bowel prep. Before surgery, a combined cystoscopy and rectal examination are performed to localize the fistula and establish all anatomic relationships with the ureteral orifices, urinary sphincter, and anorectal anatomy. One can insert a catheter or guide wire through the fistula to facilitate dissection. The patient is placed in the prone jackknife position, with buttocks spread apart. An incision is made from the anal verge to the sacrococcygeal articulation, dividing all muscular bundles of the posterior anus and the entire thickness of the posterior rectal wall and tagging them with sutures for accurate reconstruction at the end of the procedure (Fig. 28-2). Once the fistula is located, it is excised down to its junction with the bladder or prostate, removing all inflammatory tissues to optimize the closure of both defects. Both organs are closed in two layers of absorbable sutures. The first rectal wall layer involves muscle and submucosa, and the second, rectal mucosa in an everting fashion. Sometimes a demucosalized rectal flap can be used to buttress this repair. Then, the dorsal rectal and anal mucosa are closed with a chromic running suture. Careful approximation of the dentate and pectinate lines is desirable. The anal sphincter is reconstructed with the previously placed tagged opposing sutures tied together. Drains are removed on the fifth postoperative day; however, discharge can be accomplished on the second postoperative day. Vesical drainage is effected with a Foley catheter as in any vesical reconstruction.

FIG. 28-2. Posterior sagittal, transrectal, transanal (York–Mason) approach. The patient is placed in a prone position with the buttocks taped in position laterally. After division of the posterior anus and rectum, the fistula is identified for excision and repaired.

This approach is simple, rapid, and performed through fresh tissues uninvolved in the inflammatory process and is a significant improvement over the transabdominal, transvesical, or transperineal approaches, where the depth of these pelvic organs makes these procedures difficult and lengthy. Postoperative pain is minimal, and no instances of fecal incontinence or anal strictures were reported in this series.

OUTCOMES
Complications Any intervention involving an intestinal anastomosis needs to be protected by nasogastric suction for several days until the ileus resolves, generally in 3 to 5 days, heralded by the passage of flatus or by a bowel movement. Complications include bowel anastomotic leaks with resultant peritonitis, external fecal fistula or pelvic abscess, prolonged ileus, or bowel obstruction. In the event of any of these complications, surgery may again be indicated, perhaps in the form of fecal diversion, and appropriate drainage or repeat reconstruction may be necessary. Results When the vesicoenteric fistula is caused by inflammatory disease (diverticulitis, Crohn's disease, etc.), it is likely that a one-stage bowel resection and closure of the fistula can be done (66%). In patients with a colonic malignancy, pelvic abscess, or with postradiation changes, it is more prudent to close the fistula in stages with fecal diversion and later closure. 1 Comparisons between groups of patients treated with a single-stage versus two-stage repair showed lower morbidity in the patients who underwent one-stage repair, concluding that single-stage repair can be achieved with low morbidity and mortality in many candidates. 6 In another series, there was no statistical difference in the complication rate between groups treated with single- and multistage repair. The recurrence rate of vesicoenteric fistula following surgical repair is up to 6.5% of patients, especially if the fistula is the result of inflammatory bowel disease other than diverticulitis (Crohn's disease) or of pelvic neoplasia (prostate, colon, or endometrium). 1 Surgical procedures that resect necrotic fistulized bowel and result in complete separation of the gastrointestinal and genitourinary tracts provided the best results in patients with radiation-induced enterovesical fistulas. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Carson CC, Malek RS, Remine WH. Urologic aspects of vesicoenteric fistulas. J Urol 1978;119:744–746. Fraley EE, Reinberg Y, Holt T, Sneiders A. Computerized tomography in the diagnosis of appendicovesical fistula. J Urol 1993;149(4):830–832. Gonzalez-Serva L. Vesicoenteric fistula. In: Glenn JF, ed. Urologic surgery, 4th ed. Philadelphia: JB Lippincott, 1993;484–490. Levenback C, Gershenson DM, McGehee R, Eifel PJ, Morris M, Burke TW. Enterovesical fistula following radiotherapy for gynecologic cancer. Gynecol Oncol 1994;52(3):296–300. McBeath RB, Schiff M Jr, Allen V, Bottaccini MR, Miller JI, Ehreth JT. A 12-year experience with enterovesical fistulas. Urology 1994;44(5):661–665. Pontari MA, McMillen MA, Garvey RH, Ballantyne GH. Diagnosis and treatment of enterovesical fistulas. Am Surg 1992; 58(4):258–263. Puente I, Sosa JL, Desai U, Hartmann R. Laparoscopic treatment of colovesical fistulas: technique and report of two cases. Surg Laparosc Endosc 1994;4(2):157–160. Saint-Marc O, Frileux P, Vaillant JC, Chevalher JM, Texeira A, Parc R. Enterovesical fistulas in Crohn's disease: diagnosis and treatment. Ann Chir 1995;49(5):390–395. Sarr MG, Fishman EK, Goldman SM, Siegelman SS, Cameron JL. Enterovesical fistula. Surg Gynecol Obstet 1987;164:41–48. Stephenson RA, Middleton RG. Repair of rectourinary fistulae using a posterior sagittal transanal transrectal modified York–Mason approach: an update. J Urol 1996;155(6):1989–1991.

Chapter 29 Vesical Trauma and Hemorrhage Glenn’s Urologic Surgery

Chapter 29 Vesical Trauma and Hemorrhage
Badrinath R. Konety, Michael P. Federle, and Robert R. Bahnson

B. R. Konety: Department of Urology, University of Pittsburgh, Pittsburgh, Pennsylvania 15213. M. P. Federle: Department of Radiology, University of Pittsburgh Medical Center, Presbyterian University Hospital, Pittsburgh, Pennsylvania 15213. R. R. Bahnson: Division of Urology, Ohio State University Medical Center, Columbus, Ohio 43210-1228.

Vesical Trauma Anatomy Mechanism of Injury Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Vesical Hemorrhage Treatment of Hemorrhagic Cystitis Chapter References

VESICAL TRAUMA
Vesical injury can occur as a result of blunt or penetrating trauma to the lower abdomen and pelvis. It is more commonly associated with blunt trauma such as that sustained from motor vehicle accidents, falls, blows, and during contact sports. Penetrating trauma resulting in vesical injury occurs from gunshot wounds and knife wounds. Bladder injuries can also be iatrogenic from transurethral surgery, gynecologic procedures, laparoscopy, and other intra-abdominal surgery. Bladder injury, particularly bladder rupture, is associated with pelvic fractures in 75% to 83% of patients. 5 However, only 5% to 10% of patients with pelvic fractures will have associated bladder rupture. 2,4 There is also a high incidence (>85%) of injuries to other organs in patients with bladder rupture. 4 Concomitant bladder rupture is found in 10% to 29% of patients who present with rupture of the posterior urethra, and this is the most common injury to the genitourinary tract associated with bladder rupture. 4,5 The mortality rate in patients with bladder rupture ranges from 11% to 44% and is mainly attributable to other associated organ injuries. Anatomy In children, the bladder is mainly an abdominal organ located behind the anterior abdominal wall. Growth of the bony pelvis allows the bladder to assume its position behind the pubic symphysis by the end of the sixth year of life. 2 In its extraperitoneal location, the bladder is protected by the bony ring of the pelvis. It is attached to the pelvic bones and the lateral pelvic wall by means of various ligaments. The superior surface (dome) of the bladder in women and the dome and a portion of the base of the bladder in men are covered by parietal peritoneum. A fibrous cord, the median umbilical ligament, extends from the apex of the bladder to the umbilicus and is a remnant of the urachus. The dorsolateral ligamentous attachments of the bladder contain the nerves and vascular supply to the bladder. The fascial attachments between the bladder and the pubic bones are termed the pubovesical ligaments in women and the puboprostatic ligaments in men. Ligamentous attachments also connect the bladder anteriorly and laterally to the pelvic side wall. The arterial supply to the bladder is derived from the superior, middle, and inferior vesical arteries, which are branches of the anterior division of the internal iliac (hypogastric) artery. The vesical venous plexus drains into the internal iliac veins. The sympathetic nerve supply to the bladder originates in the thoracolumbar sympathetic trunks and is via the superior hypogastric plexus to the pelvic plexus, where it joins with the parasympathetic nerves. The parasympathetic nerve supply is from the sacral parasympathetic outflow to the pelvic plexus and then to the bladder. Mechanism of Injury Bladder injury occurs as three predominant types: contusion with only intramural injury and extraperitoneal or intraperitoneal bladder rupture. The exact incidence of bladder contusion is not known because of the lack of large studies involving this type of bladder injury. It is a partial-thickness tear of the bladder mucosa with ecchymosis of the bladder wall. It is often associated with a “teardrop” bladder, which occurs as a result of the presence of compressive pelvic hematomas from pelvic fractures.17 It is usually self-limiting and rarely requires treatment. Extraperitoneal bladder rupture occurs less frequently than intraperitoneal rupture (34% versus 58% of cases). Combined intra- and extraperitoneal rupture is seen in 8% of cases. 16 It was initially believed that bladder rupture, especially extraperitoneal rupture, resulted from the traumatic dislodgement of the bladder from its points of attachment. Penetration of the bladder wall by fragments of the fractured pelvic bones was also thought to be another possible etiologic mechanism. 2 However, Carroll and McAninch noted that only 35% of bladder ruptures in their series were accompanied by ipsilateral pelvic fractures. 3 Hence, it is likely that the bladder may sustain an extraperitoneal rupture when it suffers from a bursting-type injury. In intraperitoneal rupture, the dome of the bladder, which is the weakest portion of the wall, usually gives way, resulting most often in a horizontal tear. 2 Diagnosis Patients with bladder injury usually complain of lower abdominal pain and tenderness. Such an injury should be suspected in any patient with a pelvic fracture. Most patients with bladder trauma, including those with bladder contusions, will have gross or microscopic hematuria. Patients with contusion alone are usually able to void, whereas those with a ruptured bladder are often unable to void spontaneously. Acidosis with prerenal azotemia and elevated blood urea nitrogen is sometimes noticeable when there is a delay in diagnosis. 16 Presence of blood at the urethral meatus mandates performing a retrograde urethrogram. This is performed to rule out urethral injury before catheterization or instrumentation. If the retrograde urethrogram is normal, a urethral catheter is placed, and a retrograde cystogram is obtained. This is performed by instilling at least 250 to 400 cc of water-soluble contrast (cystografin) in the bladder under gravity to ensure adequate distension and visualization of possible areas of rupture. 5,16 One of the principal reasons for false-negative cystograms is instillation of an inadequate amount of contrast in the bladder. Static anteroposterior, oblique, or lateral films are obtained with the bladder full, and a washout film is obtained after drainage of the contrast material from the bladder. These additional films are useful in evaluating patients with posterior wall ruptures, which may be obscured in the anteroposterior view by a contrast-filled bladder. The drainage film also helps detect residual extravasation. The cystogram is usually normal in the presence of a bladder contusion. Intraperitoneal rupture results in ill-defined spillage of contrast into the peritoneum ( Fig. 29-1). The extravasated contrast may outline loops of bowel or accumulate in the paracolic gutters, beneath the diaphragm or over the bladder, in an hourglass pattern. Extraperitoneal rupture is seen as streak-like extravasation of contrast confined to the pelvis on retrograde cystogram ( Fig. 29-2). Corriere and Sandler 6 further distinguished extraperitoneal ruptures as simple (confined to the perivesical space) or complex (extravasation into scrotum, retroperitoneum, abdominal wall, etc.). Displacement of the bladder by a pelvic hematoma can result in a “teardrop”-shaped bladder on cystogram. 6

FIG. 29-1. In intraperitoneal rupture of the bladder, free contrast within the peritoneal cavity oulines bowel loops.

FIG. 29-2. In extraperitoneal rupture of the bladder, contrast fills the pelvic cavity around the bladder.

Recently, examination of a contrast-filled bladder during CT scan has been used as a method of assessing injury. This is particularly applicable in patients who first undergo abdominal CT scans to rule out suspected visceral injuries. In these situations, the ability to simultaneously evaluate the bladder would obviate the need for an additional plain-film cystogram. However, during routine abdominopelvic CT scan, the bladder may not be adequately distended to allow evaluation for rupture. Mee et al.12 reported on two patients who were evaluated for bladder rupture on CT scan. Both patients received intravenous and oral contrast, and their Foley catheters were clamped to allow bladder filling. In spite of this, one of the patients had a false-negative result. The bladder rupture was subsequently visualized on plain-film cystography in both cases. The results of CT cystography are better when the bladder is filled in a retrograde fashion with large volumes of contrast (>350 cc).11 Intraperitoneal bladder rupture can be distinguished from extraperitoneal rupture on CT scan. Presence of contrast around the bladder ( Fig. 29-3A) and in the paracolic gutters on either side ( Fig. 29-3B) and around abdominal viscera such as the liver ( Fig. 29-3C) indicates intraperitoneal rupture. In the case of extraperitoneal rupture, contrast extravasation is usually seen around the bladder, in the presacral space ( Fig. 29-4A) and in the retroperitoneum anterior to the great vessels (Fig. 29-4B). Bladder contusions may be seen on CT scan as intramural hematomas. In spite of the improved accuracy of CT scans, plain-film cystography is still the diagnostic modality of choice for detecting bladder ruptures. The accuracy of CT cystography may be significantly improved if retrograde bladder filling with adequate amounts of contrast is employed. In these situations, its accuracy may even approach that of plain-film cystography. Computed tomographic cystography may be particularly useful in the select group of patients who undergo a CT scan as their initial radiologic evaluation and are unable to undergo routine cystography because of the nature of their injuries or time constraints.

FIG. 29-3. A CT cystogram showing evidence of intraperitoneal rupture of the bladder with extravasation around the bladder (A) (arrow), paracolic space (B) (solid arrows), and around the abdominal viscera (C) (solid arrows).

FIG. 29-4. Extraperitoneal rupture showing contrast in the presacral space (A) (curved arrow) and retroperitoneum (B) (solid arrow); B = bladder.

Intraoperatively, bladder rupture can be diagnosed by extravasation of saline, sterile milk, methylene blue, or indigo carmine, which is instilled in the bladder through a Foley catheter. In some situations, an intravenous pyelogram may be required to rule out other ureteral or renal injuries. Indications for Surgery 1. 2. 3. 4. Intraperitoneal bladder rupture. Bladder rupture or perforation sustained during another surgical procedure. Extraperitoneal bladder rupture in the presence of other intra-abdominal injuries requiring surgical intervention. Extraperitoneal bladder rupture with the bladder being inadequately drained by urethral catheter drainage.

Alternative Therapy Alternative treatments of bladder trauma are predominantly Foley catheter drainage, which is indicated in patients with bladder contusions and extraperitoneal extravasation. Injuries occurring during other procedures such as laparoscopic surgery may be repaired laparoscopically. Surgical Technique Intraperitoneal Bladder Rupture Intraperitoneal bladder rupture requires immediate surgical repair. The abdomen is opened through a vertical lower midline incision, which affords better exposure and is extendable in case a laparotomy is required. The rupture, which is usually placed horizontally on the dome of the bladder, is identified. In some situations, this may require instillation of saline or dye in the bladder through a previously placed urethral catheter. In cases where additional extraperitoneal ruptures are suspected, the opening in the bladder wall can be extended to allow better visualization of the interior and bladder neck. These extraperitoneal tears can be closed from inside the bladder in one or two layers using running absorbable suture (3-0 chromic or polyglycolic/polygalactic acid). The intraperitoneal rupture(s) are closed in at least two layers using running 3-0 chromic or polyglycolic/polygalactic acid suture. The mucosa, muscle, and peritoneum are all closed in separate layers. The bladder is filled with saline after completion of the closure to evaluate for leaks. If any leaks are detected, they can be closed using interrupted figure-of-eight sutures. In some situations, bony spicules that have penetrated the bladder wall may need to be removed before closure of the bladder. In cases of penetrating trauma or erosion of the bladder wall by pelvic abscess, nonviable tissue must be debrided, and the edges of the perforation freshened prior to closure. In these cases, the tissue may be extremely friable, and a single-layer closure may need to be performed. The ureteral orifices should be identified and observed to ensure normal efflux of urine. This may be done after administration of intravenous indigo carmine to facilitate visualization. If efflux of urine is not seen, proximal ureteral obstruction, especially by fractured bony fragments, should be ruled out. This can be done by performing a retrograde or intravenous pyelogram on the operating table. An 8-Fr Malecot suprapubic catheter is placed through a separate cystotomy to drain the bladder. Care must be taken not to disturb the pelvic hematoma that is invariably present. Disruption of the pelvic hematoma may give rise to significant bleeding. This can be controlled by packing the area with Gelfoam, Surgicel, or laparotomy tapes. The abdomen can be temporarily closed with the packing in place for about 24 hours, and the packing removed at the time of reexploration. In extreme cases, angiographic embolization of the pelvic vessels may be necessary. A ½-inch Penrose drain is placed adjacent to the bladder and left in place for 48 hours. In some cases, if the pelvic hematoma has not been disturbed, and the bladder closure is truly watertight, drains can be omitted altogether. The abdominal fascia and skin are closed in the usual fashion. In patients with small bladder ruptures, we have opted to drain the bladder postoperatively via a urethral catheter and have noted no significant adverse effects. The catheter is left in place for 7 to 10 days. A gravity cystogram is obtained at the end of this time to ensure absence of extravasation. The catheter is then removed if no extravasation is evident on cystogram. Iatrogenic bladder injury, if suspected to have occurred during other operative procedures, should be documented by instillation of methylene blue or indigo carmine in the bladder and noting any extravasation. The rupture or tear can be closed primarily in two or three layers using absorbable suture as in other cases of rupture. Bladder perforations sustained during laparoscopic procedures can be diagnosed by noting distention of the urethral catheter drainage bag with gas. 18 These injuries can be repaired as described previously by laparotomy or even laparoscopically. 14 Extraperitoneal Bladder Rupture Until the 1970s, extraperitoneal bladder rupture was managed as an intraperitoneal rupture. Since then several studies have demonstrated that these injuries can be managed nonoperatively.5,6 Corriere and Sandler successfully managed 41 patients with extraperitoneal bladder rupture by prolonged urethral catheterization alone. All patients healed the bladder injury spontaneously without complications. 6 Since then, other studies have duplicated these results. Isolated extraperitoneal rupture can be treated by simple urethral catheter drainage. Once urethral injury has been ruled out by means of a retrograde urethrogram, a urethral catheter is placed. The catheter is left in place for 10 to 14 days. Repeat cystograms are performed at the end of this period. If no extravasation is observed, the catheter can be removed. If any contrast extravasation is evident on the cystogram, catheter drainage is continued. Cystograms are repeated at weekly intervals until no extravasation is demonstrable. A majority of extraperitoneal ruptures treated in this manner will heal by 2 weeks, and almost all will show healing within 3 weeks. Severe bleeding with clots or sepsis should prompt surgical exploration even in cases of extraperitoneal rupture. If patients are undergoing laparotomy for other intra-abdominal injuries, it is reasonable to repair extraperitoneal ruptures surgically. Outcomes Complications Some patients may notice persistent urgency and increased frequency of micturition after repair of bladder ruptures. These symptoms are usually temporary and tend to subside with time. Vesical neck injuries increase the risk of subsequent incontinence, and attention should be paid to careful repair of these injuries. Infection of pelvic hematomas can result in abscess formation requiring prolonged drainage and antibiotic treatment. This can be prevented to some extent by taking care to avoid disrupting the hematoma intraoperatively. Unrecognized injury to adjacent structures can lead to subsequent vesicovaginal or vesicoenteric fistula formation. Otherwise, this complication is uncommon. Complications such as clot retention and pseudodiverticulum formation are seen in fewer than 10% of patients treated with catheter drainage alone for extraperitoneal rupture. 4,5 Significant sepsis, delayed healing, formation of bladder calculi, and vesicocutaneous fistula formation have been noted to occur in patients treated with urethral or suprapubic catheter drainage for extraperitoneal rupture. 9 These patients most often had poorly functioning catheters or did not receive prophylactic antibiotics. Hence, it is important to ensure that urethral catheters are functioning adequately when used in these situations. Use of larger catheters and resorting to immediate open repair if catheters remain nonfunctional after 24 to 48 hours will help avoid these complications. Prophylactic antibiotics with gram-negative coverage, when administered for the duration of catheterization, will help prevent urinary tract infections. Results Open repair with adequate closure of the rupture is almost uniformly successful in all patients treated in this manner, and 74% to 87% of patients managed with urethral catheter drainage for extraperitoneal rupture will show evidence of healing by 10 to 14 days. 5,9 The remainder will heal with an additional week to 10 days of catheter drainage.

VESICAL HEMORRHAGE
Significant bleeding from the bladder in the absence of trauma is usually associated with hemorrhagic cystitis. This can result from a variety of infectious and noninfectious etiologies. Sports hematuria or stress hematuria is a well-known cause of vesical hemorrhage seen mainly in marathon runners and other athletes. It is believed to be caused by the repeated impact of the posterior wall and base of the bladder, which results in mucosal contusions. 1 An empty bladder at the time of running facilitates this process. Maintaining a partially full bladder in which the urine acts as a hydrostatic cushion will help prevent it. Infections with various viruses such as the BK virus, adenovirus, and the influenza A virus can result in hemorrhagic cystitis. Bacterial infections, most often with E. coli, can also result in hemorrhagic cystitis. Fungal infections seen frequently after treatment with broad-spectrum antibiotics can also result in this condition. Parasitic infections with organisms such as Schistosoma hematobium are known to be associated with this form of cystitis. A list of possible etiologic factors for hemorrhagic

cystitis is listed in Table 29-1. This indicates that hemorrhagic cystitis is a symptom of an underlying condition rather than a disease in itself. Infection-related hemorrhagic cystitis is usually treatable by addressing the underlying cause.

TABLE 29-1. Etiologic agents for hemorrhagic cystitis

Radiation therapy to the prostate, bladder, or other pelvic organs can result in hemorrhagic cystitis. Initially there is mucosal edema with submucosal hemorrhage. Chronically, radiation causes obliterative endarteritis with subsequent urothelial ischemia. Various measures such as steroids, vitamin E, and trypsin have proved futile in treating radiation-induced cystitis, which can manifest many years after exposure. Coating the bladder mucosa with synthetic agents such as sodium pentosanpolysulfate has some beneficial effect. 15 Hyperbaric oxygen therapy has also proved effective. 13 Amyloidosis of the bladder, which tends to occur in patients with rheumatoid arthritis and Crohn's disease, can also result in hemorrhagic cystitis. The hemorrhage can be particularly severe after instrumentation or biopsy of the bladder. This hemorrhage may require aggressive treatment measures including angiographic vessel occlusion or cystectomy. Urothelial malignancies can also cause significant bleeding, which can be controlled by transurethral resection of tumor and fulguration with electrocautery in most cases. In patients with metastatic or unresectable bladder tumors and severe hematuria, local radiation can be used to palliate the symptoms. In some cases, cystectomy or urinary diversion by means of percutaneous nephrostomy or conduit urinary diversion may be the only viable option. A wide range of drugs and industrial toxins can also give rise to hemorrhagic cystitis. Aniline and toluidine dyes are well known to be associated with this side effect. Treatment is largely conservative, as the cystitis is self-limiting and resolves after removal of the offending agent. Antibiotics such as nitrofurantoin, ether, and accidental insertion of spermicidal contraceptives into the urethra can also result in severe hemorrhage from the bladder. Conservative treatment with adequate hydration, bladder irrigation, and discontinuing the causative agent would suffice as treatment for most cases. Chemotherapeutic agents are a major cause of hemorrhagic cystitis. Busulfan, commonly used for treatment of leukemia, can give rise to hemorrhagic cystitis. Administration of N-acetylcysteine (mucomist), used to treat some other forms of chemically induced cystitis, would only worsen the condition. Most other measures are also ineffective in this situation. Intravesical administration of Thiotepa can result in this complication, which can occur up to 6 months after cessation of therapy. Alkylating agents such as cyclophosphamide and isophosphamide, which are employed in chemotherapeutic regimens for various solid tumors and lymphoproliferative disorders, are common culprits for hemorrhagic cystitis. The incidence ranges from 2% to 40%, and significant mortality rates have also been reported.10 It is dose dependent and related to the route of administration of the chemotherapeutic agent (higher with intravenous administration). It is more severe in dehydrated patients. Acrolein, which is a liver metabolite of cyclophosphamide, is the principal inciting agent and acts by direct contact with the bladder mucosa. Histologic changes that occur in the bladder are similar to those seen with radiation and include edema, ulceration, neovascularization, hemorrhage, and necrosis. Prophylactic hydration and the use of protective agents such as mucomist or 2-mercaptoethane sulfonate (Mesna) can reduce the incidence of this complication. Systemic administration of mucomist can decrease the antineoplastic effect of cyclophosphamide. Treatment of Hemorrhagic Cystitis A practical algorithm for the management of hemorrhagic cystitis is outlined in Fig. 29-5. Mild hematuria can be managed by vigorous hydration and oral administration of agents such as aminocaproic acid or conjugated estrogens, which inhibit fibrinolysis and promote coagulation.

FIG. 29-5. Algorithm for treatment of hemorrhagic cystitis.

Moderate hematuria can be treated with continuous saline irrigation through a Foley catheter after all clots have been evacuated. In some situations such as with radiation cystitis, irrigation with cold saline for 24 to 48 hours may prove more effective. If hematuria persists, continuous bladder irrigation with 1% alum (potassium or ammonium aluminum sulfate) is helpful. The alum acts as an astringent and precipitates the surface proteins. Aluminum levels must be monitored, particularly in patients with renal insufficiency. Severe acidosis and encephalopathy can occur in such patients as a result of high aluminum levels. Periodic intravesical instillation of prostaglandins (PGE 2, PGF2a) and PGF2a analogs (Carboprost) have also proved effective. They decrease the inflammatory response and reduce the hemorrhage. They can be used prophylactically or therapeutically. Prostaglandin E 2 has been used in a dose of 0.75 mg in 200 cc of normal saline instilled for 4 hours. The effective dose of PGF 2a has been 1.4 mg in 200 cc of normal saline. Carboprost has been used in a dose of 0.8 mg/dl diluted in normal saline instilled for 1 hour at 6-hour intervals with good results in 62% of patients according to one study. 8 Instillation of silver nitrate (0.5% to 1% solution) for short periods of time followed by saline irrigation of the bladder to remove residual silver nitrate is also an effective technique. Persistent severe hemorrhage that has not subsided in spite of the above-mentioned measures can be treated with intravesical instillation of carbolic acid (phenol) or 1% formalin. This requires general anesthesia. Phenol is instilled in a dose of 30 cc of a 100% solution mixed with an equal volume of glycine for 1 minute. This is washed out with 95% ethanol (60 cc) and saline to prevent methemoglobinemia. It is necessary to rule out vesicoureteral reflux by performing a voiding cystourethrogram before using formalin, as it can cause fibrosis and scarring of the ureters and renal pelvis. If need be, the ureters can be occluded with Fogarty balloon catheters to prevent reflux while formalin is instilled. Fifty milliliters of 1% formalin (0.37% formaldehyde) diluted with saline should be instilled for 4 to 10 minutes. This should then be washed out with saline, and the saline irrigation is continued for 24 hours. The external genitalia are covered with towels or Vaseline to prevent irritation. In recalcitrant cases, use of medical antishock trousers and cryotherapy have been reported. 7 Embolization of the hypogastric arteries with autologous clot, Gelfoam,

coils, or ethanol can also be resorted to in such cases. This may result in temporary gluteal claudication. Open ligation of the hypogastric artery can also be performed. Supravesical urinary diversion by means of percutaneous nephrostomy tubes or ileal or sigmoid conduit urinary diversion with or without cystectomy remains as a final but viable option. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. Blacklock NJ. Bladder trauma in the long-distance runner. “10,000 meters hematuria.” Br J Urol 1977;49:129. Bodner DR, Selzman AA, Spirnak JP. Evaluation and treatment of bladder rupture. Semin Urol 1995;13:62. Carroll PR, McAninch JW. Major bladder trauma: mechanisms of injury and a unified method of diagnosis and repair. J Urol 1984;132:254. Cass AS. Diagnostic studies in bladder rupture. Indications and techniques. Urol Clin North Am 1989;16(2):267. Cass AS, Luxenberg M. Features of 164 bladder ruptures. J Urol 1987;138:743. Corriere JN, Sandler CM. Mechanisms of injury, patterns of extravasation and management of extraperitoneal bladder rupture due to blunt trauma. J Urol 1988;139:43. deVries CR, Freiha FS. Hemorrhagic cystitis: A review. J Urol 1990;143:1. Ippoliti C, Przepiorka D, Mehra R, et al. Intravesicular carboprost for the treatment of hemorrhagic cystitis after marrow transplantation. Urology 1995;46:811. Kotkin L, Koch MO. Morbidity associated with non-operative management of extraperitoneal bladder injuries. J Trauma 1995;38(6):895. Krane DM. Hemorrhagic cystitis. AUA Update 1992;XI Lesson 31. Lis LE, Cohen AJ. CT cystography in the evaluation of bladder trauma. J Comput Assist Tomogr 1990;14(3):386. Mae SL, McAninch JW, Federle MP. Computerized tomography in bladder rupture: diagnostic limitations. J Urol 1986;137:207. Norkool DM, Hampson NB, Gibbons RP, Weisman RM. Hyperbaric oxygen therapy for radiation induced hemorrhagic cystitis. J Urol 1993;150:332. Parra RO. Laparoscopic repair of intraperitoneal bladder perforation. J Urol 1994;151:1003–1005. Parsons CL. Successful management of radiation cystitis with sodium pentosanpolysulfate. J Urol 1986;136:813. Peters PC. Intraperitoneal rupture of the bladder. Urol Clin North Am 1989;16(2):279. Sandler CM. Bladder trauma. In: Pollack HM, ed. Clinical urography. Philadelphia: WB Saunders, 1990;1505–1521. Schanbacher PD, Rossi LJ, Salem MR, Joseph NJ. Detection of urinary bladder perforation during laparoscopy by distension of the collection bag with carbon dioxide. Anesthesiology 1994;80:680–681.

Chapter 30 Interstitial Cystitis Glenn’s Urologic Surgery

Chapter 30 Interstitial Cystitis
Gary J. Faerber

G. J. Faerber: Department of Urology, University of Michigan, Ann Arbor, Michigan 48109-0330.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Patient Selection Hydrodistention and Instillation of Vesical Agents Endoscopic Resection or Fulguration Cystolysis Urinary Diversion Augmentation Cystoplasty Partial Cystectomy and Substitution Cystoplasty Total Cystectomy with Urinary Diversion Outcomes Complications Results Chapter References

Interstitial cystitis is a symptom complex comprised of chronic irritative voiding symptoms, sterile and cytologically negative urine, and bladder pain that is exacerbated by bladder filling but relieved, in many instances, by bladder emptying. This is predominantly a woman's disease because they comprise more than 90% of all patients diagnosed with this disease. The low percentage of men reported in epidemiologic studies involving interstitial cystitis may in part reflect the difficulty in differentiating interstitial cystitis from chronic abacterial prostatitis because both conditions have similar overlapping symptoms. The exact etiology and pathogenesis of this debilitating disease are as yet unknown but may be related to an insufficient glycosaminoglycan layer, an abnormal inflammatory response to noxious urinary components, a primary sensory neuronal abnormality, and chronic infection with an as yet unidentified and/or fastidious organism, or a combination of these different etiologies.

DIAGNOSIS
The diagnosis of interstitial cystitis is based on the presence of irritative voiding symptoms, the absence of urinary organisms, and typical cystoscopic appearance. In 1988, an NIDDK-sponsored workshop was convened to propose inclusion and exclusion criteria to assist those investigating this disease. 3 These specific inclusion and exclusion criteria are listed in Table 30-1.

TABLE 30-1. Diagnosis of interstitial cystitis

Cystoscopic examination at the time of hydrodistention is mandatory in confirming the diagnosis of this disease as well as in ruling out other possible etiologies that could be responsible for bladder pain and irritative voiding symptoms. Hydrodistention is performed under regional or general anesthesia. The bladder is distended with either sterile water or normal saline irrigant at a pressure of 100 to 120 cm H 2O and then emptied after 5 minutes. The presence of glomerulations or Hunner's ulcers, which universally spare the trigone, are highly suggestive, though not pathognomonic, of interstitial cystitis ( Fig. 30-1, Fig. 30-2 and Fig. 30-3 see also Color Plate 1). Some authors prefer to categorize those patients exhibiting Hunner's ulcers separately from those patients exhibiting the much more common finding of glomerulation.

FIG. 30-1. (A) Normal appearance of the bladder urothelium before hydrodistention in a patient with symptoms consistent with interstitial cystitis. (B) Same patient following hydrodistention. The urothelium is abnormal, revealing minimal to moderate glomerulation. (C) Cystoscopic appearance of a patient with moderate glomerulations and submucosal hemorrhage. (D) Hunner's ulcer with marked hemorrhage surrounding the ulcer. This patient was successfully treated with focal Nd:YAG laser ablation therapy. (see also Color Plate 1.)

FIG. 30-2. Hydrodistention is carried out with the irrigation bag hung 100 to 120 cm above the level of the bladder. Bladder capacity is reached when irrigant flow stops. During filling, the bladder mucosa is typically normal-appearing. After 5 minutes of distention, the bladder is drained, and the bladder capacity is measured. At the termination of bladder emptying, the irrigant fluid is often blood-tinged or grossly hemorrhagic in patients with interstitial cystitis. Repeat hydrodistention in some cases may be necessary to elicit glomerulation.

FIG. 30-3. Transvaginal denervation. (A) Diagram depicting the position of the nerves innervating the bladder and their position relative to the trigone and bladder neck. The circle indicates the nerve branches that are cut during the Ingelmann–Sundberg transvaginal denervation. (B) Outline of the posterior-based inverted-U-shaped incision. The Foley balloon depicts the location of the trigone. (C) Completed sharp dissection of the vaginal epithelial flap exposing the bladder neck and trigone. After completion of the dissection, the vaginal epithelium is reapproximated with a running 2-0 or 3-0 chromic catgut suture.

Pathologic features of interstitial cystitis include nonspecific chronic inflammatory infiltrate, edema, and vasodilation of the submucosa and detrusor layers. Bladder mastocytosis is often found on pathologic examination, but it is not pathognomonic, nor does its absence exclude the diagnosis of interstitial cystitis.

INDICATIONS FOR SURGERY
The surgical treatment for interstitial cystitis is categorized in Table 30-2. Other than hydrodistention and the intravesical instillation of certain agents, the primary treatment for interstitial cystitis is usually not surgical in nature. Most would agree that surgical treatment is appropriate only for a small and select group of patients with incapacitating and debilitating symptoms resistant to conventional medical and/or behavioral therapy.

TABLE 30-2. Surgical options for the treatment of interstitial cystitis

ALTERNATIVE THERAPY
Alternatives to surgical treatment would include anticholinergics, anti-inflammatories, behavioral modification, and other methods of managing chronic pain, including TENS units and tricyclic antidepressants.

SURGICAL TECHNIQUE
Patient Selection Once the diagnosis of interstitial cystitis is established, a reasonable attempt should be made to assess the impact of the disease on daily activities, work activities, leisure activities, and interpersonal relations. It is imperative that the patient be counseled extensively regarding the realistic goals and limitations of surgical therapy for this poorly defined disease. This is especially true in patients who are considering cystectomy with an incontinent or continent diversion. Instruction on the care involved in stoma care and use of a urinary appliance are certainly a vital part of patient teaching for those contemplating this type of surgical intervention. For those considered candidates for augmentation cystoplasty or continent diversion, assessment of manual dexterity and ability to perform self-catheterization is necessary before proceeding with surgery. A dedicated nurse or stomal therapist available to advise and instruct patients preoperatively is a crucial element in the preoperative preparation process. Hydrodistention and Instillation of Vesical Agents Hydrodilation of the bladder as a treatment for interstitial cystitis was first described nearly 65 years ago and remains the most common surgical treatment for the relief of bladder symptoms. 2 Hydrodistention is an easy and relatively safe technique and is usually necessary as part of the diagnostic algorithm in patients with the symptom complex of irritative bladder symptoms in the face of sterile and cytologically negative urine. The therapeutic effect of hydrodistention appears to result from ischemia of the suburothelial nerve plexus with resultant sensory denervation secondary to bladder overdistention. 5 Before any endoscopic manipulation is done, the patient must have a sterile urine. Preoperative antibiotics are recommended. The patient then undergoes cystoscopy under general or regional anesthesia, which allows for adequate distention of the bladder. The bladder is distended with the irrigant bag (sterile water or saline)

elevated to 100 to 120 cm above the bladder, which assures that distention up to at least 100 cm H 2O occurs during filling ( Fig. 30-2). The bladder is inspected before and during filling. In the vast majority of patients, the bladder before distention is unremarkable. Bladder capacity is reached when the irrigant no longer flows in the drip chamber on the irrigant tubing. The bladder is distended for 3 to 5 minutes, drained, and the bladder volume measured. The presence of hematuria at the terminal portion of the drained fluid is very common in patients with interstitial cystitis. Reinspection of the bladder in patients with interstitial cystitis reveals diffuse glomerulations in most cases and, in some cases, Hunner's ulcers. On rare occasions the lesions may be focal. Biopsies of the affected area(s) are necessary to rule out other possible pathologic conditions such as carcinoma in situ. Additional intravesical agents such as dimethylsulfoxide (DMSO), silver nitrate, heparin sulfate, and chlorpactin WCS 9 can be instilled at the same time as the hydrodistention. Fifty milliliters of 50% aqueous solution of DMSO with or without steroids (i.e., 100 mg hydrocortisone or 50 mg methylprednisolone) are left indwelling for 20 to 30 minutes. Intravesical heparin in a dose of 10,000 IU can also be instilled and left indwelling much the same manner as DMSO. Silver nitrate in concentrations of 0.5% to 2% can also be instilled and left to dwell for 5 to 7 minutes. At the end of the dwell time, the silver nitrate solution is irrigated with copious amounts of saline solution. A white precipitate of silver chloride is formed, and saline irrigation should continue (usually 1 to 2 liters) until the irrigant is clear. It is best to avoid using silver nitrate at the same time as biopsy or in patients with vesicoureteral reflux. Chlorpactin at 0.4% is instilled in a similar manner as DMSO. However, the same precautions should be observed with chlorpactin instillation as with silver nitrate in that it should not be administered if vesicoureteral reflux is present or if bladder biopsies were performed. Endoscopic Resection or Fulguration Endoscopic resection or fulguration of lesions can also be performed in those select few patients who have Hunner's ulcers or localized disease. Resection can be carried out with the loop resectoscope in which the continuous-flow resectoscope is very helpful because it allows a constant bladder volume during the resection, minimizing the risk of inadvertent bladder rupture. Fulguration of discrete areas of glomerulation can be performed with electrocautery, using either the Bugbee or rollerball electrode, or with the neodymium:YAG laser. With the laser set at 25 watts continuous and the tip set 1 to 2 mm from the bladder wall, the entire lesion is treated including a 2- to 3-mm margin of normal mucosa. Retreatment can be repeated at 4 to 6 weeks. Cystolysis Bladder denervation procedures have been reported in the treatment of patients with intractable bladder pain and urinary frequency and urgency. Division of the posterior sacral roots, posterior rhizotomy, or division of the inferior vesical neurovascular pedicle has resulted in temporary improvement in urinary frequency, urgency, and pain. However, the return of symptoms and the development of poorly compliant bladders over the long term have resulted in the abandonment of these procedures as a viable surgical treatment for interstitial cystitis. Ingelmann-Sundberg has recently described a more selective denervation in which a transvaginal approach is used to resect the inferior hypogastric plexus, whereby both the sympathetic and parasympathetic fibers to the bladder are divided 4 (Fig. 30-3). Candidates for transvaginal denervation are selected by first performing a subtrigonal injection with bupivacaine. Patients amenable to denervation should experience a period of complete or significant relief from their irritative symptoms. To perform the denervation, place the patient in a lithotomy position and determine the bladder neck and trigone by palpation of the Foley catheter balloon. Ureteral stents should be placed before the vaginal dissection is done to avoid inadvertent injury to the ureters during the vaginal dissection. A posterior-based U incision is made in the anterior vagina, and the vaginal epithelium is sharply dissected off the underlying proximal urethra, bladder neck, and distal trigone. The vaginal epithelium is then reapproximated using a running suture of 2-0 or 3-0 chromic catgut. The ureteral stents are removed, and the Foley catheter is left indwelling for 24 hours. Urinary Diversion Urinary diversion without removal of the diseased bladder usually is not sufficient to relieve symptoms secondary to interstitial cystitis. Most patients will continue to be symptomatic, and therefore, urinary diversion alone is not an adequate or appropriate method of treatment. Augmentation Cystoplasty Simple augmentation of the bladder without excision of the diseased bladder has been described as a method of treating patients with intractable interstitial cystitis who are found to have small-capacity bladders (<400 cc under general anesthesia). Both small and large bowel segments can be used 7,9 (Fig. 30-4). Concern remains regarding the relative wisdom of leaving a significant portion of affected bladder behind in interstitial cystitis patients when performing simple augmentation. Intuitively, removal of as much diseased or affected bladder would be preferred in patients with interstitial cystitis, and therefore, partial or subtotal cystectomy with substitution cystoplasty would appear to be a better surgical choice.

FIG. 30-4. Ileal augmentation cystoplasty. (A) Posterior-based U-shaped incision is created on the anterior bladder. (B) Completion of the posterior-based bladder flap. (C) A 30-cm segment of distal ileum is isolated and divided along the antimesenteric border. (D) The ileal segment is then folded, and the posterior surface is closed completely with a running absorbable suture. The anterior segment is partially closed. (E) Completed ileal bladder anastomosis. Dotted line indicates the position of the posterior bladder flap, which ensures a widely patent anastomosis.

Partial Cystectomy and Substitution Cystoplasty Supratrigonal cystectomy with enterocystoplasty is the preferred surgical choice for patients with small-capacity bladders (<400 cc under general anesthesia) and urinary frequency and urgency related to this small bladder capacity ( Fig. 30-5). Patients who have a predominant pain component, especially if it is unrelated to bladder fullness, are not good candidates for partial cystectomy and substitution cystoplasty because they are unlikely to experience symptomatic relief. Patients undergoing partial cystectomy and cystoplasty should also be able to perform intermittent catheterization in order to achieve complete bladder emptying as well as to perform bladder irrigation of the mucus secreted by the bowel segment used.

FIG. 30-5. Technique of subtotal cystectomy and substitution cystoplasty. (A) The bladder is being bivalved with electrocautery. (B) View of the bladder with both ureteral orifices cannulated with ureteral catheters to avoid injury to ureters during bladder resection. (C) Completion of subtotal cystectomy with only a small cuff of bladder remaining, which consists of urethra, bladder neck, and trigone. (D) Completed anastomosis of bowel onto bladder cuff. A Foley catheter (not shown) and a suprapubic tube are placed to ensure adequate drainage during the immediate postoperative period.

The patient is given a bowel prep and adequate hydration before surgery and is placed in a supine position. A Foley catheter is placed sterilely in the bladder and connected to a three-way irrigation. The peritoneal cavity is entered through a vertical midline incision, and an appropriate segment of either large or small bowel with a mesentery long enough to reach down to the bladder is selected. The preferred bowel segments are the cecum, sigmoid colon, or ileum. The bladder is filled with irrigant via the three-way irrigation and is then divided in a clam-shell technique, exposing the trigone. Ureteral catheters are placed before resection of the bladder to avoid inadvertent injury to the ureters. Using electrocautery, supratrigonal cystectomy is performed, resecting all but a 1- to 2-cm cuff of bladder that includes the trigone and bladder neck. Hemostasis during the resection is controlled by placement of Allis clamps on the edges of the remaining bladder. The vesicoenteric anastomosis is completed using a two-layer running closure of 3-0 chromic on the mucosa and 2-0 chromic on the muscularis layer. In addition to the Foley, a 22-Fr Malecot suprapubic catheter is left indwelling via a separate “cystotomy” to provide adequate postoperative drainage. Starting on postoperative day 2, gentle irrigation of the suprapubic tube and Foley catheter is done to prevent obstruction secondary to mucus production from the bowel segment. Patients are usually discharged on postoperative day 7 to 10 when normal bowel function returns. Before discharge, the Foley catheter is removed, and patients are taught how to irrigate via the suprapubic tube. At 3 to 4 weeks postoperatively, a cystogram is performed, and the suprapubic tube is removed if no leakage is noted. Patients are instructed to perform intermittent catheterization to ensure adequate bladder emptying as well as to ensure irrigation of mucus. An alternative to the supratrigonal cystectomy is a total cystectomy and orthotopic urinary diversion, which has been described in both men and women after cystectomy for bladder cancer (Chapter 76, Chapter 77, Chapter 78, Chapter 79, Chapter 80, Chapter 81, Chapter 82 and Chapter 83). This type of diversion may be a viable alternative to partial cystectomy because less of the affected bladder is left behind. Total Cystectomy with Urinary Diversion Total cystectomy with urinary diversion is the treatment option for patients who have failed to respond to all previous conservative treatments or who have failed partial cystectomy and enterocystoplasty. Patients with a significant component of urethral pain are probably better candidates for complete cystectomy and urinary diversion rather than partial cystectomy. The choice of performing a continent versus an incontinent diversion is based mainly on patient preference. As continent diversions have become more popular in recent years, the majority of my patients have preferred this type of urinary diversion. To be considered a candidate for a continent diversion, patients must show a proficiency in performing intermittent catheterization and be highly motivated. The technique of cystectomy and urinary diversion, including complications, is described in detail elsewhere in Chapter 24, Chapter 25, and Chapter 78, Chapter 79 and Chapter 80.

OUTCOMES
Complications Complications of hydrodistention with or without intravesical therapy are listed in Table 30-3. The most serious complication is bladder rupture, but fortunately, this is very uncommon, occurring in fewer than 0.1% of more than 1,500 hydrodistentions performed for the treatment of interstitial cystitis at our institution over the past 8 years. No return of irrigant fluid after distention, or the sudden return of irrigant fluid at the end of bladder filling, or severe suprapubic and/or abdominal pain should alert one to the possibility of a spontaneous bladder rupture. Immediate cystogram should be performed. Prolonged Foley catheter drainage is probably all that is necessary to allow the rupture to heal spontaneously. Any bladder rupture that occurs following instillation of agents such as chlorpactin or silver nitrate probably warrants open exploration with copious irrigation of the site of extravasation because of the severe caustic properties of these agents.

TABLE 30-3. Complications of hydrodistention with or without intravesical instillation therapy

Bladder perforation is more likely to occur following fulguration or excision of interstitial cystitis lesions because the bladder wall is normally quite thin. In addition, bowel injury can occur following aggressive loop resection or injudicious use of laser energy. Early and late complications secondary to augmentation cystoplasty are listed in Table 30-4. Persistence of symptoms is probably the most common and disheartening for both patient and surgeon.

TABLE 30-4. Complications of augmentation cystoplasty or subtotal cystectomy with substitution cystoplasty a

Results Hydrodistention alone without the addition of intravesical agents relieves the symptoms in up to 30% of patients. The addition of DMSO with or without steroids has been shown to relieve symptoms in about 50% of patients. Similar clinical responses are also seen with chlorpactin and heparin. Interestingly, after intravesical therapy, symptoms may improve, but the cystoscopic appearance of the bladder, regardless of the agent used, typically remains unchanged. The choice of intravesical agent used, and in which order, is not critical to how a patient may respond clinically. It is common for patients to become resistant to one treatment and respond favorably to another treatment. The advantage to DMSO and heparin is that these agents can be administered in the office under topical anesthesia, whereas silver nitrate and chlorpactin almost universally require general or regional anesthesia. Resection or fulguration in select patients results in clinical improvement in 33% to 80% of patients, with those with Hunner's ulcers responding more favorably than those patients with focal glomerulations. The more aggressive, open surgical approaches have shown good results in selected patients. Relief of symptoms in a highly select group of patients undergoing supratrigonal cystectomy and enterocystoplasty has been reported to range between 60% and 90%. 6,8 The results of total cystectomy for the treatment of incapacitating interstitial cystitis are varied. Although a significant percentage of this highly select group of patients who undergo cystectomy will experience significant relief, there are reports of patients having persistent pelvic pain despite having undergone complete cystectomy and urethrectomy. 1 Interstitial cystitis is a difficult disease to treat in a surgical manner aside from hydrodistention and intravesical instillation therapy. Fortunately, the vast majority of patients with interstitial cystitis do not ever have symptoms so severe or incapacitating as to warrant any further invasive surgical intervention aside from hydrodistention. In those few patients who continue to have significant symptoms and who are interested in further surgical intervention, a thorough discussion of treatment options between physician and patient is critical to ensure a satisfactory outcome. Suffice it to say that our basic understanding of this poorly understood disease is still in its infancy, and therefore, proper therapy including surgery is as yet poorly delineated. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Bakins LS, Tanagho EA. Pelvic pain with pelvic organs. J Urol 1992;147:683–686. Bumpus HC. Interstitial cystitis: its treatment by overdistention of the bladder. Med Clin North Am 1930;13:1495. Gillenwater JY, Wein AJ. Summary of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases workshop on interstitial cystitis. J Urol 1988;140:203. Ingelmann-Sundberg A. Partial bladder denervation in the treatment of interstitial cystitis in women. In: Hanno PM, Staskin DR, Krane RJ, Wein AJ, eds. Interstitial cystitis. London: Springer-Verlag, 1990;189. Irwin PP, Galloway NTM. Surgical management of interstitial cystitis. Urol Clin North Am 1994;21:145. Kontturi M, Hellstrom PA, Tammela TL, et al. Colocystoplasty for the treatment of severe interstitial cystitis. Urol Int 1991;46:50. Smith RB, VanCaugh P, Skinner DG, et al. Augmentation enterocystoplasty: a critical review. J Urol 118:35,1977. von Garrelts B. Interstitial cystitis: 13 patients treated operatively with intestinal bladder substitutes. Acta Chir Scand 1966;13:436. Webster G, Maggio M. The management of chronic interstitial cystitis by substitution cystoplasty. J Urol 1989;141:287.

Chapter 31 Open Prostatectomy Glenn’s Urologic Surgery

Chapter 31 Open Prostatectomy
Ray E. Stutzman

R. E. Stutzman: Department of Urology, The Johns Hopkins University, Baltimore, Maryland 21287.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Management Suprapubic Prostatectomy Retropubic Prostatectomy Perineal Prostatectomy Outcomes Complications Results Chapter References

Open prostatectomy is the enucleation of the hyperplastic adenomatous growth of the prostate. This procedure does not involve total removal of the prostate. A tissue plane exists between the adenoma and the compressed true prostate, which is left intact. Three surgical approaches to the prostate are described in this chapter: suprapubic, retropubic, and perineal.

DIAGNOSIS
Over 90% of prostatectomies for benign prostatic hyperplasia are performed by transurethral resection of the prostate (TURP). When the obstructing tissue is estimated to weigh more than 50 g, serious consideration should be given to an open procedure. Digital examination, prostatic ultrasound, and cystourethroscopic measurement of the prostatic length may aid in the estimating of the size of the gland. Findings on cystourethroscopy may indicate an open procedure, such as sizable bladder diverticuli, which justify removal, or large bladder calculi, which are not amenable to easy fragmentation. The association of an inguinal hernia with an enlarged prostate may lead to a suprapubic or retropubic procedure because the hernia may be repaired by way of the same lower abdominal incision. 12

INDICATIONS FOR SURGERY
The indications for prostatectomy include the following symptoms or findings secondary to prostatic obstruction: acute urinary retention; recurrent or persistent urinary tract infections; recurrent gross hematuria; documented significant residual urine after voiding with or without overflow incontinence; pathophysiological changes of the kidneys, ureters, or bladder; abnormally low urinary flow rate; and normal flow rate with abnormally high intravesical voiding pressure and intractable symptoms such as nocturia, frequency, and urgency. Contraindications to an open prostatectomy include a small, fibrous gland, carcinoma of the prostate, or prior prostatectomy in which most of the prostate has previously been resected or removed and the planes are obliterated.

ALTERNATIVE THERAPY
Alternative therapies to open prostatectomy include transurethral resection of the prostate (TURP), endoscopic procedures including incision of the prostate, laser ablation, vaporization techniques, thermotherapy, and medical management. Most of these therapies are effective for moderate (medical management) to severe symptoms in prostates less than 60 g (alternative surgical techniques) and are therefore not indicated in the majority of patients who are candidates for open prostatectomy. The patient's bladder outlet symptoms could also be managed alternatively by intermittent catheterization, an indwelling catheter, or a suprapubic cystostomy. None of these are good alternatives if the patient is a reasonable surgical risk.

SURGICAL TECHNIQUE
Preoperative Management The average age of patients is about 70 years. Many of the patients have histories of cardiovascular disease, chronic obstructive pulmonary disease, diabetes, or hypertension. It is preferable to evaluate the upper urinary tract with either an intravenous pyelogram and a postvoid film if the patient's renal function is normal or an abdominal radiograph and a renal sonogram. Cystourethroscopic examination should be performed to rule out unexpected bladder pathology. This can be done just before surgery under the same anesthetic. If the patient has a documented urinary tract infection, it should be treated before planned elective surgery and may necessitate indwelling catheter drainage before the procedure. Transfusion of blood may be required in about 15% of patients undergoing open prostatectomy. It is prudent to have 2 or 3 units of blood available when contemplating the procedure. The safest transfusion is autologous blood, and individual units can be drawn a week apart while the patient is on oral iron medication. Spinal or epidural anesthesia is preferred in all prostatectomy procedures. If regional anesthesia is contraindicated, a general anesthetic with adequate relaxation may be used. Informed consent is necessary. The patient must be made aware of the risks and complications. Most patients can be evaluated as an outpatient and then admitted to the hospital on the day of surgery. This is cost effective and reduces hospitalization. Suprapubic Prostatectomy Suprapubic prostatectomy or transvesical prostatectomy is the enucleation of the hyperplastic adenomatous growth of the prostate performed through an extraperitoneal incision of the anterior bladder wall. 9 Eugene Fuller of New York is credited with performing the first complete suprapubic removal of a prostatic adenoma in 1894. This was a blind procedure with digital enucleation of the gland. Suprapubic and perineal drainage tubes were placed to wash out clots and control bleeding. Peter Freyer of London popularized the operation and subsequently published his results of over 1,600 cases with a mortality rate of just over 5%. The entire operation was usually a 15-minute procedure. A 5- to 8-cm midline suprapubic incision was made, and the bladder was opened without opening the lateral tissue spaces or entering the space of Retzius. Digital enucleation of the prostate was then performed. One or two fingers were placed in the rectum for counterpressure while the suprapubic enucleation was accomplished. The prostatic fossa was left alone because Freyer thought that the capsule and surrounding tissues at the bladder neck would contract down enough to control bleeding somewhat like a parturient uterus immediately after childbirth. He left an indwelling urethral catheter and a large suprapubic tube to evacuate clots. His low mortality and morbidity rates are remarkable considering that no blood transfusions or antibiotics were available at that time. This blind enucleation remained popular for over 50 years. The low transvesical suprapubic prostatectomy with visualization of the bladder neck and prostatic fossa and placement of hemostatic sutures has supplanted the blind procedure. 4 This operation is presented in more detail. The patient is placed in the supine position with the umbilicus positioned over the kidney rest; the table is slightly hyperextended and in a mild Trendelenburg's position. A catheter is introduced into the urinary bladder; the bladder is irrigated and then filled with 200 to 250 ml of water or saline, and the catheter is then removed. The abdomen and genitalia are prepped from nipple line to midthigh. A vertical midline suprapubic incision is made through the skin and linea alba with the incision extending from below the umbilicus to the symphysis ( Fig. 31-1). The rectus muscles are retracted laterally, and the prevesical space is developed, with the peritoneum swept superiorly. It is not necessary or desirable to expose the retropubic or lateral vesical spaces. For more adequate exposure, a self-retaining retractor

is used.

FIG. 31-1. Midline vertical incision.

Two sutures are placed in the anterior bladder wall below the peritoneal reflection. A vertical cystotomy is then made, and the incision is opened down to within 1 cm of the bladder neck, allowing visualization of the bladder neck and prostate. A medium Deaver retractor is placed into the open bladder, retracting superiorly. A narrow Deaver is then placed over the bladder neck just distal to the trigone. The curved end of the Deaver retractor provides an excellent semilunar line for incising the mucosa around the posterior bladder neck just distal to the trigone ( Fig. 31-2). By this method, the ureteral orifices are well visualized and are not compromised. Metzenbaum scissors are introduced at the 6 o'clock position, and, by gentle dissection, the plane between the adenoma, bladder neck, and the capsule of the prostate is developed.

FIG. 31-2. Incision of mucosa over the adenoma.

The remainder of the procedure is done by digital dissection, freeing the posterior lobes down to the apex of the prostate and then circumferentially sweeping anteriorly (Fig. 31-3). The urethra is firmly attached at the apex. It is preferable to use scissors to sharply incise the urethra, keeping close to the prostatic adenoma so as not to cause injury to the sphincter and subsequent incontinence. With large glands, it is often preferable to remove one lobe at a time, or, if there is a large intravesical protrusion of the middle lobe, this may be removed separately.

FIG. 31-3. Digital enucleation of adenoma.

After removal of the adenoma, the prostatic fossa is inspected, and a digital sweep is made to ascertain if there is any remaining nodular adenomatous tissue ( Fig. 31-4). There is usually minimal bleeding; however, bleeding is frequently seen in the 5- and 7-o'clock positions. The prostatic arteries enter the capsule and prostate at this level near the bladder neck. Suture ligature of these vessels is done even if there is no active bleeding ( Fig. 31-5). Figure-of-eight sutures of 2-0 chromic on a 5 /8 circle needle provide good hemostasis.

FIG. 31-4. View of empty prostatic fossa.

FIG. 31-5. Hemostatic suture ligatures at 5- and 7-o'clock positions.

A 22-Fr, 30-ml balloon, three-way Foley catheter is passed through the urethra. A 26- or 28-Fr Malecot suprapubic tube is passed through a separate stab wound in the anterior bladder wall and brought out through a stab wound in the lower abdominal wall ( Fig. 31-6). A watertight, single-layer, interrupted closure of the bladder with either 2-0 chromic catgut or Vicryl is done, just missing the mucosa but including full thickness of the muscularis and serosa. The balloon of the Foley catheter is inflated to 45 ml and placed on no traction. A 4-0 chromic catgut suture is placed as a pursestring around the suprapubic tube; this prevents any leakage and helps to hold the suprapubic tube gently in position during wound closure. A Penrose drain is placed down to the cystotomy site and brought out through a separate stab wound. The bladder is irrigated until clear and checked for significant leakage.

FIG. 31-6. Foley catheter snug at bladder neck: prevesical drain and Malecot catheter.

The wound is irrigated, and the linea alba is closed with a running #2 nylon or #1 PDS suture. The skin is approximated with skin staples. The drain and suprapubic tube are sutured to the skin with nylon sutures, and a dressing is applied. 11 Postoperatively, excessive blood loss is the most common immediate complication encountered; about 15% of patients require a blood transfusion. If excessive bleeding from the prostatic fossa is noted intraoperatively, two techniques are effective in stopping the bleeding. Malament described the placement of a #1 or #2 nylon pursestring suture around the vesical neck; the suture was brought out through the skin and tied snugly. 7 This effectively closes the bladder neck and tamponades the prostatic fossa with control of bleeding ( Fig. 31-7). Between 24 and 48 hours after placement, the suture is cut on one side and removed. O'Conor 10 described placation of the posterior capsule using 0 chromic catgut on a 5/8 curved needle. This placation narrows the fossa and results in effective hemostasis ( Fig. 31-8). Point fulguration of bleeders in the fossa may also provide hemostasis.

FIG. 31-7. Pursestring suture of bladder neck.

FIG. 31-8. Plicating sutures in posterior capsule.

Antibiotics are not indicated for elective prostatectomy in patients who have had no urinary tract infections and have sterile urine. If there has been a long-term indwelling catheter or preoperative infection, appropriate perioperative antibiotics, a cephalosporin and an aminoglycoside or a fluoroquinolone, are indicated. The patient is usually limited to intravenous fluids the day of surgery, but the following day he can usually tolerate oral nutrition, often having a full diet. A stool softener or mild laxative is given to prevent straining with bowel movements or fecal impaction. Continuous bladder irrigation by way of the three-way Foley catheter is maintained for 12 to 24 hours. The Foley catheter usually is removed after 3 days, although one can remove the suprapubic catheter first. If the Foley catheter is removed first, the suprapubic tube is clamped at 5 days to give the patient a trial at voiding. It is removed the following day if voiding is satisfactory with little residual. The drain is removed a few hours after removal of the suprapubic tube if there is no drainage. The skin staples are removed on the seventh postoperative day, and

the skin is covered with sterile strips. On discharge from the hospital, the patient is encouraged to increase his activity gradually and should be able to resume full activity 4 to 6 weeks postoperatively, with outpatient visits at 3 and 6 weeks. Retropubic Prostatectomy Simple retropubic prostatectomy is the removal of the hyperplastic prostatic adenoma by way of a prostatic capsule incision. Van Stockum is credited with performing the first retropubic prostatectomy, which he called “extravesical suprapubic prostatectomy.” 14 A longitudinal capsular incision was made on one side of the midline. Millin 7 reported his operative technique and results in 1945. His procedure gained wide acceptance, and he is credited with popularizing retropubic prostatectomy. Various modifications have subsequently been described. 1,3,13 The patient is placed in the supine position with the umbilicus positioned over the kidney rest; the table is slightly hyperextended and in a mild Trendelenburg's position. The lower abdomen and suprapubic area are shaved, and the entire operative field from nipple line to midthigh is scrubbed with surgical solution. A Pfannenstiel incision ( Fig. 31-9) may be used, but I prefer a vertical midline incision extending from below the umbilicus to the symphysis (see Fig. 31-1). The linea alba is opened, and the rectus muscles are retracted laterally.

FIG. 31-9. Modified Pfannenstiel incision.

The prevesical and retropubic space is developed, with the peritoneum and extraperitoneal fat swept superiorly. A self-retaining retractor is placed in the incision to obtain maximal exposure. There are several large veins in the loose areolar tissue and fat over the anterior capsule of the prostate. These should be suture ligated and divided; smaller vessels may be fulgurated to avoid troublesome bleeding ( Fig. 31-10).

FIG. 31-10. Ligation and division of periprostatic veins over the prostatic capsule anteriorly.

The vesical neck can be visualized and palpated. Two traction sutures are placed in the prostatic capsule above and below the planned site of the capsular incision, which is made about 1 cm distal to the bladder neck (Fig. 31-11). As the capsule is opened, one can recognize the white outer part of the adenoma. The length of the incision depends on the size of the gland and should be sufficient to dissect the adenoma.

FIG. 31-11. Transfixion sutures placed on the capsule anteriorly with an incision made transversely in the prostatic capsule.

The cleavage plane between the prostatic adenoma and the surgical capsule or true prostate is developed using Metzenbaum scissors ( Fig. 31-12 and Fig. 31-13). The dissection may be completed with scissors; in large adenomas, digital enucleation can easily be performed. The urethra is firmly attached at the apex. It is preferable to use scissors to incise the urethra sharply, keeping close to the prostatic adenoma to avoid causing injury to the sphincter and subsequent incontinence.

FIG. 31-12. Identification of the cleavage plane between the prostatic adenoma and the surgical capsule of the prostate.

FIG. 31-13. Continuing the dissection of the prostate of the adenoma from the prostatic capsule using Metzenbaum scissors.

After removal of the prostatic adenoma, the fossa is inspected for any remaining nodules of adenoma and for sites of bleeding. The main sources of bleeding are the arteries at the 5- and 7-o'clock positions, which lie just distal to the bladder neck. Figure-of-eight suture ligatures of 2-0 chromic catgut are placed to secure hemostasis (Fig. 31-14). Bleeding vessels in the prostatic fossa can be fulgurated under direct visualization. If the surgeon wears a headlight for illumination, visualization is much improved.

FIG. 31-14. Figure-of-eight sutures are applied at the vesicle neck at 5- and 7-o'clock positions to secure hemostasis.

In some patients, the posterior lip of the vesical neck is prominent and protrudes into the lumen. This can be removed by wedge resection by grasping the midline with an Allis clamp, and, with either scissors or a knife, the wedge can be excised ( Fig. 31-15). A running suture may be placed for hemostasis.

FIG. 31-15. The wedge resection of the vesical neck posteriorly.

A 22-Fr Silastic-coated three-way irrigating Foley catheter with a 30-cc retention balloon is inserted through the urethra into the bladder. The transverse incision of the prostatic capsule is then closed with a continuous suture of 2-0 chromic catgut or Vicryl, ensuring a watertight closure ( Fig. 31-16). Slight catheter traction is applied, and continuous bladder irrigation is instituted. If excessive bleeding from the prostatic fossa is noted, the source should be sought before wound closure. Suture placation of the prostatic fossa may be helpful (see Fig. 31-8). A suprapubic catheter is used only if there is significant bleeding.

FIG. 31-16. Closure of the incision of the prostate capsule anteriorly with interrupted chromic sutures.

A Penrose drain is placed into the space of Retzius and brought out through a stab wound lateral to the incision and sutured to the skin. The wound is irrigated, and the linea alba is closed with a running #2 nylon or #1 PDS suture. 11 The skin is approximated with skin staples. A wound and drain dressing is applied. Postoperatively, the Foley catheter is irrigated until it runs clear, and continuous bladder irrigation with saline is used for several hours. The catheter is usually removed on the fifth or sixth postoperative day. The Penrose drain is moved partially outward on that day and is removed the following day if no drainage occurs. The

skin clips are removed, and sterile strips are applied. Perineal Prostatectomy The first operations for relief of urinary retention from prostatic enlargement were probably done through the perineum, and early medical writings contain references to division of the bladder neck through the perineum for this purpose. Covillard, in 1639, was apparently the first to remove a hypertrophied middle lobe by tearing it away with forceps after perineal lithotomy. In 1848, Sir William Fergusson exhibited specimens of hypertrophied prostates he had enucleated through the perineum after removal of bladder calculi. Kuchler, in 1866, formulated the first systematic technique for radical perineal prostatectomy, but his operations were done only in the cadaver. In 1867, Billroth used Kuchler's method to carry out the first two intentional prostatectomies in living subjects. Apparently, however, the lobes were not entirely removed in these patients. In 1873, Gouley advocated systematic enucleation of the lateral lobes and excision of the median lobe through the perineum. Goodfellow is credited 4 as the first to perform a perineal prostatectomy successfully on a routine basis. His method involved the use of a midline vertical incision from the base of the scrotum to the anal margin, followed by incision of the membranous urethra, extension of the opening into the bladder, and complete enucleatlon of the prostatic lobes. His technique, although differing in certain respects from that used today, nevertheless forms the basis of current methods. During the next decade, a number of technical modifications were suggested by Nicoll, Alexander, Albarran, Proust, dePezzer, Legueu, and others. For the most part, those changes were concerned with improving delivery of the prostatic lobes into the perineal incision for enucleation. In 1903, Young described his operative technique developed at the Johns Hopkins Hospital; this is still the approach most widely used. In 1939, Belt and colleagues introduced an important modification in the perineal approach to the prostate, which did much to reduce the risk of rectal injury inherent in the operations of Young and earlier surgeons. Belt's method of closure also was a great improvement over earlier methods and shortened convalescence considerably. 2 Either spinal or general anesthesia can be used. Caudal block is also acceptable. With general anesthesia, tracheal intubation ensures adequate respiratory exchange. Preoperatively, the patient should self-administer an enema to clean the lower bowel and rectum and receive appropriate antibiotics for a 1-day bowel prep. The genitalia are cleansed thoroughly, after which cystoscopy is performed. The entire operative area, from costal margins to midthigh, is then prepped. Bilateral vas ligation is now rarely performed. Perfect positioning is essential for the perineal operation. The patient is placed in the exaggerated lithotomy position on any ordinary operating table ( Fig. 31-17). Sandbags are placed beneath the sacrum to position the perineum as close to horizontal as possible. The table is then elevated to bring the operative area up to the level of the operator's chest. This makes the operation a good deal easier and improves visualization considerably. The perineum can usually be positioned adequately without resort to Trendelenburg's position, but occasionally a slight Trendelenburg's position may be necessary. Under no circumstances should shoulder braces be used for fear of causing postoperative brachial palsy. All other points where pressure is likely (e.g., popliteal areas) are carefully padded.

FIG. 31-17. Perineal prostatectomy. Standard operating table is used for exaggerated lithotomy position. Classic inverted-U incision is shown.

The curved Lowsley tractor is passed through the urethra and held upright with blades unopened ( Fig. 31-18). A curved skin incision is made about 1 cm from the anal margin. The anus is excluded from the operative field by being covered with a towel secured by three Allis clamps to the posterior edge of the incision. With the index fingers, the ischiorectal fossae are developed perpendicular to the plane of tile perineum. The central tendon is gently separated from the underlying rectum and cut across distal to the external anal sphincter, with care taken not to disturb that structure ( Fig. 31-19). A bifid posterior retractor is placed in the ischiorectal fossae, and gentle traction is exerted. The lateral fossae are developed next and held with two small lateral retractors. The rectourethralis muscle is identified and cut (Fig. 31-20).

FIG. 31-18. Perineal prostatectomy. Curved Lowsley tractor is in place at the outset of the operation.

FIG. 31-19. Perineal prostatectomy. After the perineal incision, ischiorectal fossae are developed by blunt dissection. The central tendon of the perineum is isolated. The line of incision is shown.

FIG. 31-20. Perineal prostatectomy. (A) Anal sphincter is elevated, exposing the rectum. (B) Retractors in lateral fossae. The rectourethralis muscle has been developed.

By carefully incising the pararectal fascia (posterior layer of Denonvillier's fascia), the rectum can be gently peeled posteriorly off the apex of the prostate. The Lowsley tractor is passed fully into the bladder, and the blades are opened. The bifid posterior retractor is replaced by a plain posterior one (the lipped Richardson is useful here), with a moistened pad used to protect the rectum. The posterior layer of Denonvillier's fascia is progressively incised and retracted posteriorly until a window appears through which the anterior layer of Denonvillier's fascia—the “pearly gates”—can be seen clearly. At this point, the operator simultaneously depresses the handle of the Lowsley tractor toward the abdominal wall and exerts firm downward traction on the posterior Richardson retractor (Fig. 31-21). The remaining posterior fascial laver is thereby stripped away from the prostate, which comes clearly into view, covered only by the glistening anterior layer of Denonvillier's fascia. This is a most effective maneuver, but it should not be done before dissection of the posterior fascial layer has been completed at the apex.

FIG. 31-21. Perineal prostatectomy. The sagittal view shows exposure of posterior capsule of prostate.

An inverted-V or curved prostatotomy is made (Fig. 31-22), and a plane of cleavage is established with the dissecting scissors ( Fig. 31-23A). Care is taken to peel back and preserve the posterior flap for subsequent closure of the prostatotomy. The urethra is incised, the curved Lowsley tractor is removed, and the regular Young prostatic tractor is inserted gently into the bladder through the prostatotomy, using a rotary motion. The blades of the tractor are then opened, the prostate is drawn down, and enucleation is begun.

FIG. 31-22. Perineal prostatectomy. The usual inverted-V capsulotomy is used in perineal enucleation.

FIG. 31-23. Perineal prostatectomy. (A) Enucleation of an adenoma is initiated by developing a cleavage plane between capsule and adenoma. (B) An incision is made into the prostatic urethra. Young's tractor is placed into the bladder, enabling mobilization of the adenoma and amputation of the urethra at the apex.

As soon as possible, the urethra at the apex of the adenoma is cut across with the scissors, thereby facilitating enucleation distally and minimizing the danger of damage to the external urethral sphincter (see Fig. 31-23B). Enucleation is carried out essentially under direct vision, using the scissors and the finger. Enucleation can sometimes be facilitated by removing the Young tractor and grasping the lobes with forceps that are especially designed for this purpose. The lobes can then be drawn progressively into the operative field. The hypertrophied lobes are cut away sharply from the bladder neck under direct vision ( Fig. 31-24). With care, the bladder neck can be preserved intact, even after removal of a large adenoma.

FIG. 31-24. Perineal prostatectomy. Adenoma is freed from the bladder neck by blunt and sharp dissection.

After enucleation has been completed, the bladder neck is grasped with Millin T-clamps, which were originally designed for the retropubic operation. These have the advantage of being offset so that one can obtain an unimpeded view of the bladder neck ( Fig. 31-25). A careful search is made for bleeding vessels (especially at the 5- and 7-o'clock positions). Smaller ones are controlled effectively by electrocoagulation. Larger arteries require mattress sutures of 2-0 plain catgut. The interior of the bladder is explored with the finger, and any blood clots are removed. The entire prostatic fossa is inspected carefully for residual adenomatous tissue. Remaining tags of tissue are trimmed away from the bladder neck.

FIG. 31-25. Perineal prostatectomy. The vesical neck is grasped and drawn down with Millin T-clamps, enabling hemostatic mattress sutures to be placed.

A 22-Fr Foley catheter is passed through the urethra and into the bladder, where the balloon is inflated with 30 to 45 ml of water. The bladder neck, which feels like a soft cervical os dilated to about two fingerwidths, retains the balloon nicely. Wedge resection of the posterior lip is generally unnecessary. If desired, a three-way Foley, catheter may be used to permit through-and-through irrigations postoperatively. Closure is simple. The edges of the prostatotomy are approximated with interrupted 2-0 chromic catgut sutures ( Fig. 31-26). The rectum is inspected for possible injury. No effort is made to bring the levator ani fibers together. A Penrose drain is left in the retroprostatic space. Skin edges are approximated with interrupted Dexon or Vicryl sutures (Fig. 31-27). A simple dressing is applied to the wound, using a split-T binder. The lower extremities are brought down simultaneously and gradually. Too rapid depositioning may result in hypotension because of the sudden rush of blood into the legs, particularly if they have not been wrapped preoperatively.

FIG. 31-26. Perineal prostatectomy. The surgical capsule is closed with interrupted sutures.

FIG. 31-27. Perineal prostatectomy: closure of muscles and subcutaneous fat of perineum. Penrose drain is left indwelling, and skin edges are approximated with interrupted sutures.

Excessive bleeding is seldom encountered during perineal prostatectomy. If care is taken to obtain adequate exposure, bleeding vessels can usually be identified and secured without difficulty. The only other complication that may occur during the operation is laceration of the rectum, which is readily recognized from the characteristic appearance of the rectal mucosa. The injury should be completely mobilized and repaired with interrupted 4-0 chromic catgut sutures placed so that the mucosal edges are inverted. The muscularis should be closed in two additional layers, again using interrupted sutures of 4-0 chromic catgut. If the injury is recognized before the urinary tract is opened, it is best to close the perineal incision and enucleate the gland through a suprapubic incision. If the rectal injury is not appreciated until after the urinary tract has been entered, the laceration should be repaired meticulously, as just outlined. Postoperatively, the patient

should be maintained on a low-residue diet, and bowel activity should be completely suppressed with paregoric for 7 days. After removal of the hypertrophied lobes, the raw surfaces of the prostatic fossa soon reepithelialize. The compressed outer prostate (prostate proper, or surgical capsule) eventually reexpands to normal size. Scattered areas of induration usually persist indefinitely and can be detected by rectal palpation. Postoperatively, if a regular Foley catheter has been used, it is simply attached to straight bedside drainage. From time to time, gentle manual irrigation may be carried out to keep the system free of clots. The catheter is secured to the thigh, but no traction is necessary. If a three-way catheter has been used, it is attached to a through-and-through irrigating system containing sterile saline solution, which is run in just rapidly enough to keep the efflux reasonably clear. The patient is given appropriate antibiotics. Fluids may be given by mouth during the first day, and they are customarily supplemented by intravenous infusions to maintain a satisfactory intake. The perineal Penrose drain is usually removed on the first postoperative day. At this time, the patient may be placed on a soft or regular diet and allowed out of bed. Early ambulation is encouraged. Usually, the perineal wound heals benignly, but sometimes partial separation of the skin edges may occur. Healing may be promoted by removal of the dressing and exposure to a heat lamp. Warm sitz baths are also effective. The urethral catheter is removed between the seventh and tenth postoperative days. Not infrequently, urinary leakage may occur from the wound for a day or two after the catheter has been taken out. If it continues longer than this, an 18-Fr, 5-ml Foley catheter may be reinserted for a day or two. Care must be taken in passing the catheter to be certain it does not curl up in the prostatic fossa. Sometimes a stylet is helpful, with the aid of a finger in the rectum. During the immediate postoperative period, it is important that no rectal instrumentation be performed. No thermometers or rectal tubes should be inserted; this must be made clear to the nursing staff.

OUTCOMES
Complications Hemorrhage Delayed bleeding as occasionally seen after TURP is uncommon after open prostatectomy. Infections Wound infections occur in fewer than 5% of patients and are usually limited to the skin and subcutaneous tissue. Postoperative epididymo-orchitis is uncommon and may occur early or late. This complication is most commonly seen in patients who have had a long-term indwelling catheter or urinary tract infection. Incontinence Incontinence of urine is an uncommon complication of open prostatectomies and usually results from perforation and partial avulsion of the prostatic capsule or avulsion of the urethra at the apex of the prostate. With careful enucleation of the adenoma, the capsule is not perforated. With sharp excision of the urethra at the apex rather than avulsion, incontinence should not occur. Some patients may experience stress incontinence or urge incontinence, and detrusor instability may be the cause. In perineal prostatectomies, about 10% of patients experience some urinary incontinence for a few days after removal of the catheter. This disappears rapidly in the vast majority, but up to 6 months may be required for complete cessation of leakage in the occasional patient. Permanent incontinence is highly uncommon after an uneventful perineal prostatectomy. Other Urologic Complications In suprapubic and retropubic prostatectomies, urinary fistulas have been reported. Persistent perineal urinary fistula has been feared by those unfamiliar with perineal surgery; in actuality, this complication is rarely seen. Its occurrence should lead one to suspect some form of urethral obstruction, for example, a postoperative stricture. Urethral stricture and bladder neck contracture occur most commonly as complications of transurethral resection and are uncommon after suprapubic prostatectomy. A single, gentle dilation with a urethral sound usually suffices to take care of this. Erectile dysfunction after prostatectomy should not occur unless the capsule has been violated. Retrograde ejaculation is common. Other Complications Rectal injury is a rare occurrence. Osteitis pubis is rarely seen but can be disabling. The condition is usually self-limited. Analgesics and anti-inflammatory drugs provide symptomatic relief. Surgical mortality for open prostatectomy should be less than 1%; myocardial infarction, pneumonia, and pulmonary embolus are the most common causes. Early ambulation, leg movement in bed, and breathing exercises decrease morbidity. Results Enucleation of the enlarged prostatic adenoma by an open procedure is applicable in 5% to 10% of patients presenting with significant bladder outlet obstruction. The operative mortality and morbidity are minimal. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Blue GD, Campbell JM. A clinical review of one thousand consecutive cases of retropubic prostatectomy. J Urol 1958;80:257–259. Brendler H. Perineal prostatectomy. In: Glenn JF, ed. Urologic surgery, 3rd ed. Philadelphia: JB Lippincott, 1983;867–878. Dittmar H. Modification of technique for retropubic prostatectomy: report of 100 cases. J Urol 1959;81:558–561. Freyer PJ. One thousand cases of total enucleation of the prostate for radical cure of enlargement of that organ. Br Med J 2:868, 1912. Gibson TE. George E. Goodfellow (1855–1910). Invest Urol 1969;7:107–109. Harvard BM. Low transvesical suprapubic prostatectomy with primary closure. In: Campbell MF, ed. Urology. Philadelphia: WB Saunders, 1954;1965–1968. Malament M. Maximal hemostasis in suprapubic prostatectomy. Surg Gynecol Obstet 1965;120:1307–1312. Millin T. Retropubic prostatectomy: new extravesical technique: report on 20 cases. Lancet 1945;2:693–696. Nanninga JE, O'Conor VJ Jr. Suprapubic and retropubic prostatectomy. In: Walsh PC, Gittes RF, Perlmutter AD, and Stamey TA (Eds.), Campbell's urology, 5th ed. Philadelphia: WB Saunders, 1986;2739–2744. O'Conor VJ Jr. An aid for hemostasis in open prostatectomy: capsular plication. J Urol 1982;127:448. Poole GV Jr. Mechanical factors in abdominal wound closure: the prevention of fascial dehiscence. Surgery 1985;97:631–640. Schlegel PN, Walsh PC. Simultaneous preperitoneal hernia repair during radical pelvic surgery. J Urol 1987;137:1180–1183. Straffon RA. Retropubic prostatectomy. In: Glenn JF, ed. Urologic surgery, 3rd ed. Philadelphia: JB Lippincott, 1983;861–866. Weyrauch HM. Surgery of the prostate. Philadelphia: WB Saunders, 1959.

Chapter 32 Pelvic Lymphadenectomy Glenn’s Urologic Surgery

Chapter 32 Pelvic Lymphadenectomy
Ralph W. deVere White and Andrew Huang

R. W. deVere White and A. Huang: Department of Urology, University of California, Davis, School of Medicine, Sacramento, California 95817.

Diagnosis Indications for Surgery Alternative Therapy Description of Procedure Prostate Cancer Bladder Cancer Outcomes Complications Results Chapter References

The pelvic lymph nodes are the initial site of spread of prostatic, bladder, and proximal urethral malignancies. Tumors of the penis, scrotum, and distal urethra spread primarily to the inguinal lymph nodes but can involve the pelvic lymph nodes. Testicular tumors rarely involve the pelvic lymph nodes unless there is massive retroperitoneal disease or a history of orchidopexy or prior pelvic procedures. Most urologists perform pelvic lymph node dissections in patients with prostate or bladder cancer.

DIAGNOSIS
Pelvic lymphadenectomy is performed as a staging procedure in patients with prostate or bladder cancer. The diagnostic modalities for these entities are discussed in Chapter 23, Chapter 24, Chapter 33, Chapter 34, and Chapter 111.

INDICATIONS FOR SURGERY
Presence of lymph node metastases from prostate cancer has prognostic significance and is a harbinger of disease progression. 1 In a series from Memorial Sloan-Kettering Cancer Center, 86% of patients with pelvic lymph node metastases measuring over 3 cc had disease progression within 5 years. Pelvic lymph node dissection has not been shown to be of therapeutic benefit in the treatment of prostate cancer. 4 However, there are patients who have small-volume nodal disease and prolonged survival after lymphadenectomy and radical prostatectomy. It is unclear whether this prolonged survival is a consequence of tumor biology, the treatment, or a combination of the two. P>Pelvic lymph node dissection may be performed as a staging procedure before definitive radiation therapy, as a separate procedure for prostate cancer, i.e., before a radical perineal prostatectomy, or concomitant with a radical retropubic prostatectomy. Serum PSA screening has led to a stage migration in newly diagnosed cases of prostate cancer and to a lower incidence of pelvic lymph node positivity. The incidence of unsuspected pelvic lymph node disease found at exploration has decreased. In the past, nearly 25% of patients with clinically localized prostate cancer were found to have lymph node metastases at the time of surgery. 10 A review of the last 140 pelvic lymph node dissections performed at the University of California–Davis, for clinically localized prostate cancer revealed four cases of lymph node involvement. We perform pelvic lymph node dissections concomitantly with a retropubic prostatectomy but feel that, with the lower incidence of nodal disease, performing them as a separate procedure, i.e., laparoscopically, is not cost-effective. For patients undergoing a perineal prostatectomy, there are a small number who would benefit from a lymphadenectomy. For patients being treated with radiation therapy, we no longer perform a lymph node dissection. For bladder cancer, there is more agreement in favor of the benefits of pelvic lymphadenectomy. A series from Memorial Sloan-Kettering Cancer Center demonstrated that the 5-year survival rate for patients with pelvic lymph node metastases was 9%. 3 This would suggest that the pelvic lymph node dissection is mainly of prognostic rather than therapeutic value. However, Skinner reported a 35% 3-year survival rate for patients who were found to have a limited number of positive lymph nodes at the time of surgery, suggesting that the node dissection may have therapeutic benefits. 9 The presence of positive lymph nodes in bladder cancer patients has implications for treatment options. Some urologists would treat with chemotherapy and reevaluate the bladder rather than proceeding with the radical cystectomy, but others favor debulking the tumor with cystectomy followed by postoperative chemotherapy. If we find grossly enlarged lymph nodes with histologic evidence for metastatic disease, we follow the former course of therapy. However, if the lymph nodes are grossly normal, we proceed with the radical cystectomy without sending the lymph nodes for frozen section evaluation. A national study is under way assessing the value of neoadjuvant chemotherapy. If it reveals an advantage for preoperative chemotherapy, the finding of nodal disease at the time of surgery becomes more challenging. Some urologists might argue that, in this situation, surgery would be the only chance for the patient, but others would argue that, neoadjuvant chemotherapy having failed, surgery would have little role in the management of these patients.

ALTERNATIVE THERAPY
Noninvasive methods of detecting pelvic lymph node spread have not been reliable. Various radiologic tests, i.e., ultrasound, computerized tomography, magnetic resonance imaging, and pedal lymphangiography, have been used to detect pelvic lymph node disease, but these have had low sensitivity. As a marker for prostate cancer, the use of serum prostate acid phosphatase (PAP) has low sensitivity and specificity. In addition, the use of serum prostate-specific antigen (PSA) alone is not a good predictor of pathologic stage, as there can be significant overlap between the PSA and the pathologic stage. Pelvic lymph node dissection is the definitive means of evaluating lymph node status in patients undergoing extirpative procedures for urologic malignancy. With the arrival of minimally invasive techniques, laparoscopic lymphadenectomy has been performed. Another modification is to perform the dissection through a minilaparotomy incision when it is done as a separate procedure. In comparing the efficacy, morbidity, and cost effectiveness of a minilaparotomy pelvic lymph node dissection to those of a laparoscopic or standard open pelvic lymphadenectomy, the minilaparotomy approach had a similar lymph node yield to the standard open procedure and a similar shortened hospital course to the laparoscopic dissection, but at a lower cost compared to the other two. 8 Much has recently been written questioning the need for pelvic lymph node dissection before definitive treatment for prostate cancer. On preoperative evaluation, there may be patients with a low likelihood of lymph node disease who do not need a lymphadenectomy. This may not eliminate lymphadenectomies completely, but it could decrease the number performed. A similar change occurred with PSA values and bone scans for prostate cancer staging. Narayan et al. found an 11% overall incidence of positive pelvic lymph nodes. 6 In patients with PSA values less than or equal to 10 and Gleason grades less than or equal to 6, fewer than 3% had lymph node metastases. They felt that, in this group of patients, a staging pelvic lymphadenectomy would not be necessary. 6 Campbell et al. reported a lymph node positive rate of 6.5% in patients with clinically localized prostate cancer, whereas if at least one favorable characteristic (Gleason grade less than 6, PSA less than 10, or nonpalpable tumor) was present, only 2.2% had lymph node involvement. 2 Patients with a low probability of lymph node involvement, i.e., low Gleason grade and low PSA, might not require a pelvic lymph node dissection, which adds both cost and time to the definitive treatment. However, this is still somewhat controversial.

DESCRIPTION OF PROCEDURE
Prostate Cancer The boundaries of the traditional pelvic lymph node dissection were those used in combination with radical cystectomies for bladder malignancies. They included the pelvic sidewall laterally, the paravesical fascia and peritoneum medially, the genitofemoral nerve superiorly, the obturator nerve inferiorly, and the femoral canal

distally (Fig. 32-1). Proximally, the dissection was carried varying distances up the common iliac artery ( Fig. 32-2). Most urologists now feel that only the obturator nodal packet need be removed for three reasons:

FIG. 32-1. Right lateral pelvic wall. Anatomy of pelvic blood vessels and nerves encountered in a pelvic lymph node dissection is depicted.

FIG. 32-2. Shaded area shows the iliac and obturator nodes that are to be removed.

First, the obturator nodes are involved in 87% of cases when lymphatic metastases are found. 5 Second, the procedure is for staging and not therapy, so a more extensive dissection is of little benefit. Third, if radiation therapy is used for local control following surgery, patients who had an extensive lymphadenectomy have a higher incidence of scrotal or lower limb edema. Preoperatively, pneumatic compression devices (PCDs) are placed, and patients are given subcutaneous heparin as prophylaxis against deep venous thrombosis. Supine or lithotomy position may be used, although we use the low lithotomy position. The sacrum is positioned over the table break or a roll to allow for hyperextension of and better vision into the pelvis. The bladder is emptied using a Foley catheter. A midline incision is made from below the umbilicus to the symphysis pubis down through the anterior rectus sheath. The posterior rectus sheath is incised for 2 to 3 cm above the linea semilunaris to aid in lateral retraction of the wound. An extraperitoneal lymph node dissection is usually performed. If the peritoneum is entered during this incision, the defect is closed with absorbable suture. The transversalis fascia is sharply divided in the midline to allow lateral dissection superficial to the peritoneum. This helps to avoid injuring the inferior epigastric vessels. The iliac vessels are exposed by bluntly sweeping the peritoneum superomedially. The vas deferens are encountered during this maneuver and may be divided. The table is tilted toward the first side for evaluation. If the prostate cancer is confined to one lobe, the dissection is begun on that side. A self-retaining retractor is applied, with care taken not to injure the inferior epigastric vessels. We use the Buchwalter retractor without the post, as it can be more quickly applied and the post can interfere with the surgeon. Other self-retaining refractors may be used. We place the Buchwalter retractor on top of sterile towels, one on each thigh and one on the abdomen. A bladder blade and moist lap sponge are used to retract the wound laterally on the side of the dissection. A malleable retractor and moist lap are placed on the bladder and used to retract the bladder toward the contralateral side. A third retractor is placed at the apex of the wound. With these three retractors, excellent visibility can be obtained. The nodal packet is palpated to detect grossly enlarged lymph nodes. If such nodes are found, they are sent for frozen section evaluation following removal. If no enlarged nodes are palpated, we continue with the lymphadenectomy and prostatectomy and do not send the lymph nodes for frozen section. The external iliac artery is identified, and dissection of the lymph node packet is begun over its anteromedial aspect. The correct plane of dissection is easily found here, and there are no other structures in this area to be damaged ( Fig. 32-3). The dissection is brought proximally to the bifurcation of the common iliac vessels and distally to the femoral canal. The lymph node of Cloquet is the most distal aspect of the dissection. Lymphatic channels into this node and surrounding the external iliac vein are carefully clipped and divided. We place a right-angle clamp around the lymph node packet and ligate it with a 2-0 silk tie. A large right-angle clip is placed below the tie. As the nodal packet is divided and swept superiorly, an accessory obturator vein may be found and should be ligated and divided to avoid avulsion. Identification of this vein is necessary, as damage can cause extensive bleeding.

FIG. 32-3. Incision of fibroareolar tissue loosely adherent to adventitia of iliac artery and vein. This allows a portion of areolar tissue to pass lateral to iliac vessels into the obturator fossa.

With gentle lateral retraction of the external iliac vein, the lymph node packet is dissected off the pelvic sidewall laterally. Although a vein retractor is usually used for this maneuver, we use a peanut/Kittner dissector. Identification of the obturator nerve is essential as the dissection is carried into the pelvic fossa to avoid injuring it. The packet is freed from the obturator nerve and vessels. The obturator vessels are spared if they are in their usual location below the nerve. If they are above the nerve or involved with the lymph node packet, it is best to ligate and divide them before bleeding can occur. This is especially true near the femoral canal. The superior attachment of the packet is now near the internal iliac artery. We previously identified and dissected out the ureter, but, in most cases, we no longer do this.

To ensure the ureter is not damaged by a clip, the specimen is split over the obturator nerve, and a right-angle clip is placed over either limb of the split packet on each side of the nerve so that the ureter cannot accidentally be included in the clip. Additional loose attachments to the proximal hypogastric vessels are clipped and divided. The entire packet is sent to pathology as the two portions divided over the obturator nerve. The obturator fossa is irrigated. It had been our routine to leave a gauze sponge in the fossa for hemostasis; however, we now do this only for minimal oozing. The same dissection is then performed on the contralateral side to complete the lymph node dissection. We place one or two Jackson–Pratt drains in the pelvis postoperatively. Bladder Cancer The dissection is similar to the one described above for prostate cancer with some differences. The incision is carried to just above the umbilicus and down to the pubic bone. We palpate the pelvic lymph nodes while remaining extraperitoneal. If no grossly enlarged nodes are felt, the dissection becomes intraperitoneal. The peritoneum is entered in midline, and inspection is performed of the intra-abdominal organs for signs of metastatic disease. If none are found, dissection is continued by mobilizing the cecum and ascending colon. The peritoneum is incised along the white line of Toldt, and the right hemicolon is rolled medially. The right ureter is identified and freed superiorly and inferiorly. Inferiorly, this leads to the bifurcation of the iliac vessels. In freeing the peritoneum, we routinely divide the vas deferens. On the left, the peritoneum is incised lateral to the sigmoid colon, and it is reflected medially. The left ureter is identified and freed as on the right. Mobilization is aided by dividing the vas deferens. The self-retaining retractor may be placed as described above. The bowel can easily be retracted into the upper abdomen, as it has been mobilized. The node dissection begins over either common iliac artery just proximal to the bifurcation. It is carried down the hypogastric artery to the superior vesicle artery, which is identified and divided. The remainder of the lymph node dissection is similar to that for prostate cancer.

OUTCOMES
Complications Although pelvic lymph node dissection is usually a relatively short procedure with little morbidity, it has a potential for significant complications. These can be divided into intra- and postoperative complications ( Table 32-1). Paul et al. reported an 8.6% incidence of intraoperative complications, an 8.7% immediate postoperative wound complication incidence, and an additional 31.4% immediate non-wound-related complication rate. They also reviewed the complication rates reported in multiple studies. These ranged from 4% to 53% with a mean rate of 26.6%. 7 Intraoperative complications can be minimized by familiarity with the pelvic anatomy and careful dissection to identify vulnerable structures. The most common vascular injury is to the accessory obturator vein. Care should be taken not to avulse the obturator vessels as they enter the pelvic foramina, as they will retract caudally, and ligation will be difficult. If this occurs, bone wax can be used. Significant injuries to the external iliac vessels require repair, sometimes with the aid of a vascular surgeon. Transection or avulsion of the obturator nerve leads to difficulties with adduction of the ipsilateral leg and is usually irreparable. Splitting the nodal packet as described reduces the chance of inadvertent nerve injury.

TABLE 32-1. Complications of pelvic lymphadenectomy

Ureteral injuries are uncommon and require repair when encountered. A problem with ureteral injuries is that they are not always identified at the time of surgery. These are often the result of a clip inadvertently being placed across the ureter. Therefore, as we now dissect out the lymph node packet, we no longer specifically look for the ureter. However, we always place a clip on the upper end of the nodal packet after splitting it over the obturator nerve in a cranial-to-caudal direction to avoid injury to the ureter, identified or not. If there is any concern for ureteral injury, the ureter must be dissected out and fully visualized. Postoperative complications include those related and unrelated to the wound. Wound infections are uncommon, especially when prophylactic antibiotics are given. Dehiscence is similarly uncommon. Seroma and hematoma formation are more common and may require drainage and local wound care. Prolonged lymph drainage and lymphocele formation may occur in 3% to 12% of patients. Prolonged drainage is treated by instilling autologous blood or Betadine solution through the preexisting drains as sclerosing agents. If Jackson–Pratt or similar drains are used, tissue will eventually grow into the drains. This has occurred twice in our experience, and in both cases a general anesthetic was required for drain removal. We now remove all drains or treat them as described above for prolonged drainage at the end of 2 weeks. Although it has been reported that the use of subcutaneous heparin increases the incidence of prolonged lymph drainage, this has not been our experience. Our rate of prolonged lymph drainage and/or symptomatic lymphocele formation is under 3%. Symptomatic lymphoceles can often be successfully treated with percutaneous drainage under radiologic guidance. Although some lymphatic drainage is expected, careful dissection and meticulous ligation of lymphatic channels help minimize the risk of prolonged drainage. Any patient with prolonged or excessive lymph drainage must be evaluated for a urinary leak. This may be done by sending a sample of the fluid for creatinine. If a urine leak is found, this may be from either the anastomosis or an unrecognized ureteral injury. If the latter is suspected, it should be immediately investigated. Thrombophlebitis and deep venous thrombosis are recognized complications of pelvic lymph node dissection. Although the studies are conflicting, most have shown that some method of anticoagulation, low-dose heparin or pneumatic compression stockings, are beneficial in reducing the risk of these complications. We routinely administer subcutaneous heparin preoperatively and every 8 to 12 hours postoperatively as well as use pneumatic compression stockings until the patient is discharged. Chronic lymphedema of the lower extremities and external genitalia may occur, and these may be worsened by radiotherapy or an extensive dissection. The modified pelvic lymph node dissection has been a reliable way of preventing chronic lymphedema. Results Staging pelvic lymphadenectomy is an essential part in the clinical evaluation of patients with prostate and bladder cancer. Within the obturator fossa, 87% of positive nodes will be detected. In patients with low-grade prostate cancer and with a PSA < 10, the incidence of positive nodes is so low that it may become prudent not to perform a pelvic node dissection as a separate procedure. In bladder cancer, the value of a node dissection may be to identify those patients in whom preoperative chemotherapy may be of benefit, though this remains to be shown in a clinical trial. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Barzell W, Bean MA, Hilaris BS, Whitmore WF Jr. Prostatic adenocarcinoma: relationship of grade and local extent to the pattern of metastases. J Urol 1977;118:278. Campbell SC, Klein EA, Levin HS, Piedmonte MR. Open pelvic lymph node dissection for prostate cancer: a reassessment. Urology 1995;46:352. Dretler SP, Ragsdale BD, Leadbetter WF. The value of pelvic lymphadenectomy in the surgical treatment of bladder cancer. J Urol 1973;109:414. Gervasi LA, Mata J, Easley JD, et al. Prognostic significance of lymph nodal metastases in prostate cancer. J Urol 1989;142:332. McLaughlin AP, Saltzstein SL, McCullough DL, Gittes RF. Prostatic carcinoma: incidence and location of unsuspected lymphatic metastases. J Urol 1976;155:89. Narayan P, Fournier G, Gajendran V, et al. Utility of preoperative serum prostate-specific antigen concentration and biopsy Gleason score in predicting risk of pelvic lymph node metastases in

prostate cancer. Urology 1994;44:519. 7. Paul DB, Loening SA, Narayana AS, Culp DA. Morbidity from pelvic lymphadenectomy in staging carcinoma of the prostate. J Urol 1983;129:1141. 8. Perrotti M, Gentle DL, Barada JH, Wilbur HJ, Kaufman UP Jr. Mini-laparotomy pelvic lymph node dissection minimizes morbidity, hospitalization, and cost of pelvic lymph node dissection. J Urol 1996;155:986. 9. Skinner DG. Management of invasive bladder cancer: a meticulous pelvic lymph node dissection can make a difference. J Urol 1982;128:34. 10. Smith JA, Seaman JP, Gleidman JB, Middleton RG. Pelvic lymph node metastasis from prostatic cancer: influence of tumor grade and stage in 452 consecutive patients. J Urol 1983;130:290.

Chapter 33 Radical Retropubic Prostatectomy Glenn’s Urologic Surgery

Chapter 33 Radical Retropubic Prostatectomy
Joseph A. Smith, Jr.

J. A. Smith, Jr.: Department of Urologic Surgery, Vanderbilt University Medical Center, Nashville, Tennessee 37232-2765.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

The unprecedented increase in the incidence of carcinoma of the prostate that has occurred over the last decade has resulted in a corresponding increase in the rate with which radical prostatectomy is performed. Refinements in patient selection and surgical technique have diminished the morbidity of surgery. Nonetheless, many patients maintain concern about the risks and potential side effects of the operation. Knowledge of surgical anatomy and attention to operative technique can help diminish both the side effects and the overall expense of treatment.

DIAGNOSIS
Clinically localized carcinoma of the prostate is diagnosed most frequently in asymptomatic men who present for early detection or screening programs or during the course of a routine physical examination. Digital rectal examination and serum prostate-specific antigen (PSA) are accepted modalities for prostate cancer detection. Almost 20% of clinically diagnosed cancers are detected by digital rectal examination in the face of a normal serum PSA level. 7 On the other hand, stage T1c cancers, i.e., those detected on a biopsy prompted by an abnormal PSA level in the face of a normal digital rectal examination, constitute the most common category of cancers diagnosed in most clinical series. Tumor grade is one of the most important prognostic criteria, though in more recent studies, preoperative PSA may also be predictive. 4 Computerized tomography scanning of the pelvis to identify lymph node metastasis is usually not indicated. A radioisotope bone scan is usually performed if the serum PSA level is greater than 20 ng/dl but rarely is positive for metastasis if the PSA is below that level.

INDICATIONS FOR SURGERY
Radical prostatectomy is indicated in patients with carcinoma of the prostate seemingly confined within the surgical capsule of the gland, who would be expected to have a life expectancy of at least 10 years. Thus, appropriate patient selection requires accurate tumor staging as well as an assessment of patient comorbidity. Despite its well-recognized limitations, digital rectal examination (DRE) remains the standard method for assessing tumor extent within the prostate, though DRE often understages palpable tumors. 10 Large, palpable tumors frequently are found to have histologic evidence of extracapsular extension. Transrectal ultrasonography (TRUS) is the most commonly used imaging modality for prostate cancer. Typically, carcinoma of the prostate is identifiable as a hypoechoic area in the peripheral zone of the prostate. However, prospective studies have not shown the superiority of TRUS compared with DRE for staging the local extent of prostate cancer. The serum PSA level is useful in identifying patients with localized disease. In general, a serum PSA level >20 ng/ml is highly suspicious for extracapsular extension. Radical prostatectomy usually is not indicated in patients who otherwise have a life expectancy less than 10 years. Although clinical carcinoma of the prostate is a progressive disease, the rate of tumor growth is such that competing causes of death dominate in elderly patients or those with poor overall health. Both chronologic as well as physiological age must be taken into account. Life table analysis indicates that most men less than 70 years of age with good overall health can anticipate at least 10 years of additional life.

ALTERNATIVE THERAPY
The term “watchful waiting” has been applied to a nonaggressive treatment approach that consists primarily of intermittent monitoring of disease status with institution of delayed hormonal therapy on evidence of symptomatic disease progression. This is probably the preferred approach in most men with less than 10 years of life expectancy. External beam irradiation is an alternative treatment for clinically localized prostate cancer. Five- and ten-year survival figures are comparable to those obtained with radical prostatectomy, but good comparative figures beyond that time range are not available. Concerns frequently are expressed about a high rate of positive biopsies after radiation therapy and the failure of the treatment to suppress serum PSA levels to the undetectable range. Interstitial brachytherapy using radioactive seeds was abandoned in most centers during the early to mid-1980s because of inferior results. More recent series use a perineal template with ultrasound guidance for seed implantation, but there are insufficient patient numbers and follow-up to draw conclusions about the merits of this approach. Cryotherapy is being explored as a treatment approach, but only short-term results are available. Once radical prostatectomy is chosen, either a retropubic or perineal approach is most often used. The advantages of a retropubic approach are the opportunity to simultaneously inspect or sample pelvic lymph nodes, the potential for wide excision of periprostatic tissues, and a very low risk of rectal injury.

SURGICAL TECHNIQUE
Radical retropubic prostatectomy is performed at our medical center with the patient in a supine position and the table slightly flexed ( Fig. 33-1). We prefer a general anesthetic and have not observed any advantage for the use of epidural catheters. Invasive anesthetic monitoring such as central venous catheters or arterial lines are not used. A 20-Fr Foley catheter is inserted in the bladder after the field has been prepped and draped sterilely.

FIG. 33-1. Position of the patient for a radical retropubic prostatectomy with the table flexed at the hips to enhance visualization and dissection.

A midline incision is made from the umbilicus to the pubis ( Fig. 33-2). The anterior rectus fascia is incised, and then the rectus muscles are retracted laterally. Blunt and sharp dissection are used to mobilize the peritoneal envelope superiorly. Blunt finger dissection can help create a pocket directly over the psoas muscle and just

lateral to the common iliac artery, which facilitates superior retraction of the peritoneum. The vas deferens is retracted along with the peritoneum. Care should be taken to make certain that the epigastric vessels are not injured beneath the belly of the rectus muscle during this maneuver.

FIG. 33-2. (A) A midline incision is made from the symphysis pubis to the umbilicus. ( B) Division of the medial insertions of the rectus abdomini.

We use a Bookwalter self-retaining retractor. An oval ring is used, and bladder blades retract the skin and rectus muscle at the inferior and lateral aspect of the incision (Fig. 33-3). Deeper retractors at this point can result in femoral neuropathy or compression of the iliac vein. At the superior and lateral portion of the incision, a malleable blade is used to retract the peritoneum.

FIG. 33-3. A Buckwalter retractor is placed to aid in dissection.

Placement of the retractors in this manner provides excellent exposure of the operative field. Pelvic lymph node dissection, as discussed in Chapter 32, is performed if the serum PSA level is greater than 20 ng/ml or the Gleason sum of the tumor grade is 7 or greater. Otherwise, the iliac and obturator lymph nodes simply are inspected for any evidence of enlargement or induration. A sponge stick is used to displace the prostate and bladder medially ( Fig. 33-4). A Kittner dissector is used to remove some of the loose fat overlying the endopelvic fascia and to expose the junction between the lateral prostate and the levator ani muscle. Electrocautery is used to incise the endopelvic fascia overlying this space. A Kittner dissector then separates the levator muscle from the lateral margin of the prostate. If the incision is made too close to the prostate, bleeding from the overlying venous plexus can occur on the lateral margin of the prostate. Under these circumstances, it is best to place a suture of 3-0 Vicryl along the lateral prostate in order to gain hemostasis.

FIG. 33-4. The endopelvic fascia is incised and the neurovascular bundles are exposed.

After the endopelvic fascia has been incised bilaterally, a sponge stick is used to depress the bladder posteriorly. The fatty tissue overlying the anterior prostate is carefully teased away using forceps and the sucker tip. This exposes the superficial dorsal vein. We prefer to control this vein separately from the deep dorsal vein complex. If the superficial dorsal vein is small, it can simply be cauterized and then divided. A larger vein is ligated with 3-0 Vicryl. Care must be taken in applying the ligature to the distal end of the vein, as it is quite friable and can be avulsed from the underside of the pubis. After the superficial dorsal vein is divided, a Kittner dissector is used to define the puboprostatic ligaments. These usually are evident as a distinct white ligamentous structure. Often, there is some adherent levator muscular tissue just lateral to the puboprostatic ligament, and this sometimes contains a small vein. A right-angle clamp passed just lateral to the puboprostatic ligament can define this tissue well and allow it to be dissected from the lateral prostatic apex using electrocautery. It is important to define the puboprostatic ligaments precisely, as described above. This allows them to be divided with Metzenbaum scissors with little risk of bleeding (Fig. 33-5). Also, this minimizes injury to the anterior sphincter mechanism. After the puboprostatic ligaments are divided, the anterior portion of the prostatic apex falls away partially from the pubic bone.

FIG. 33-5. Lateral incision of the puboprostatic ligaments after division of the superficial dorsal vein.

The deep dorsal vein complex runs parallel to the urethra at the prostatic apex and then fans out over the anterior of the prostate ( Fig. 33-6). We feel that it is important to control these vessels preemptively rather than simply to incise them and place sutures afterward. A McDougal clamp is useful for this purpose. The dorsal vein complex can be pinched partially off from the anterior urethra with the thumb and forefinger of the left hand, and the jaws of the McDougal clamp passed just anterior to the urethra. Bleeding is rare during this maneuver if the anatomic structures are carefully dissected and defined. Spreading the jaws of the clamp should be avoided so as not to injure the anterior sphincter mechanism. A #1 Vicryl ligature is then tied around the dorsal vein complex. The McDougal clamp is then passed through the same space again and held in position while a 2-0 Vicryl suture on an SH needle is passed through the center of the dorsal vein complex just distal to the previous ligature ( Fig. 33-7). After the anterior portion of this stick suture is tied, the McDougal clamp is used to pull the end of the suture around the posterior aspect of the dorsal vein complex, which is then controlled further by tying the two suture ends. Back bleeding is prevented by placing a 0 Vicryl suture on a CT 1 needle through the veins of the dorsal vein complex, where they fan out over the anterior prostate. Grasping the lateral margins of the veins with an Allis clamp helps to bunch them in the middle and facilitates placement of this hemostatic suture ( Fig. 33-7). The McDougal clamp is once again passed through the previously defined space. Electrocautery is then used to divide the dorsal vein complex just proximal to the 0 Vicryl ligature placed earlier. Sometimes this ligature becomes displaced during this process, but the backup 2-0 Vicryl fixed suture maintains hemostasis. As the electrocautery is used, the McDougal clamp is gently lifted to displace the electrical energy away from the sphincter muscle. This entire technique usually allows total hemostasis during division of the dorsal vein complex, and no additional hemostatic sutures generally are required.

FIG. 33-6. (A) Following division of the puboprostatic ligaments, the deep dorsal venous complex is identified. ( B) A right-angled clamp is placed around the dorsal vein, and it is ligated proximally. ( C) Lateral view of the plane between the prostate and the dorsal vein showing how posterior depression of the superior prostate aids in the dissection.

FIG. 33-7. Suture ligation of the distal dorsal venous complex with an Allis clamp aiding in identification of the bundle.

At this point, the anterior sphincter muscle should be untouched and the prostatic apex and membranous urethra visible. Metzenbaum scissors are then used to incise directly over the anterior portion of the membranous urethra just at the prostatic apex ( Fig. 33-8). The Foley catheter is identified and withdrawn through the partially severed urethra after the injection port and connector end have been cut off. The Foley catheter is not used for traction, as this can injure the urinary sphincter complex or cavernous nerve. The posterior portion of the urethra is divided under direct vision using Metzenbaum scissors.

FIG. 33-8. Division of the membranous urethra.

The sequence of the next maneuver depends on whether or not a nerve-sparing approach for potency preservation is to be used. If nerve sparing is attempted, a right-angle clamp is used to lift the thin reflection of the endopelvic fascia onto the lateral margin of the prostate. This is incised with a knife up to the level of the prostatic pedicle. A Kittner dissector is then used to displace this fascial leaf posterolaterally. This carries with it the blood vessels and nerves that can be displaced

from the lateral margin of the prostate. Small bleeding vessels that may be encountered during this maneuver are usually left alone at this point to avoid injury to the neurovascular complex. After the neurovascular bundle has been satisfactorily displaced from the prostate, the layer of fascia and muscle tissue commonly termed the rectourethralis muscle is incised sharply ( Fig. 33-9). Excellent hemostasis is necessary in order to perform this maneuver under direct vision, which is highly preferable. Blunt finger dissection can then be used to develop the plane between the prostate and rectum, but it is important not to pull hard on the prostate ( Fig. 33-10). The remaining posterior attachment of the prostatic apex is divided sharply with Metzenbaum scissors immediately adjacent to the prostate.

FIG. 33-9. Division of the rectourethralis taking care to avoid the neurovascular bundles.

FIG. 33-10. The prostate is mobilized off the rectum in the midline, and the lateral pelvic fascia is incised carefully between the posteriolateral surface of the prostate and the neurovascular bundle. (Modified from Catalona WJ. Prostate cancer. Orlando, FL: Grune & Stratton, 1984;107.)

If a nerve-sparing approach is not used, the incision in the rectourethralis muscle is performed without dissecting the neurovascular bundle from the lateral prostate. The plane between the prostate and rectum is developed bluntly, and then the neurovascular bundle is excised widely at the prostatic apex by ligating it with 2-0 Vicryl sutures before incising it as widely as possible from the prostatic apex. However, even if a non-nerve-sparing approach is used, care should be taken to avoid excessive traction on the prostate, which could tear sphincter muscle tissue. At this point, the table is placed in partial Trendelenberg position to allow better visualization of the posterior prostate, the seminal vesicles, and the prostatic pedicle. An incision is made through the posterior layer of Denonvillier's fascia with electrocautery. This exposes the seminal vesicles and ampulla of the vas deferens ( Fig. 33-11). A right-angle clamp can then be placed just lateral to the seminal vesicle and around virtually the entire prostatic vascular pedicle ( Fig. 33-12). Care should be taken to avoid going too far cephalad with the right-angle clamp, as this can cause bleeding in some of the veins in the perivesical fat. A 2-0 Vicryl suture is used to ligate the prostatic pedicle. Because it is difficult to place two ties far enough apart to allow an incision between them, we usually simply use electrocautery to incise the tissue on the prostate side of a single tie. Small back bleeders from the prostate can then be controlled individually with the electrocautery.

FIG. 33-11. Denonvillier's fascia is incised between the base of the prostate and the anterior rectal wall, exposing the ampullary portions of the vasa deferentia and the medial wall of the seminal vesicles. The proper plane of dissection for the prostatic vascular pedicles is just lateral to the seminal vesicles.

FIG. 33-12. A right-angled clamp is used to dissect out the superior vascular pedicle, taking care to divide the pedicle close to the prostate and avoiding injury of the neurovascular bundles.

Complete division of the prostatic pedicle greatly facilitates visualization and dissection of the seminal vesicles ( Fig. 33-13). A right-angle clamp is passed around the ampulla of the vas deferens just medial to the seminal vesicle. The ampulla is divided with the electrocautery, and the proximal end of the vas deferens is clamped with the right angle. This can then be used for traction in dissecting the medial edge of the seminal vesicle with the electrocautery. A right-angle clamp is then passed around the seminal vesicle and used to help dissect the surrounding tissue down to the tip of the seminal vesicle. The artery to the seminal vesicle enters at the tip

and usually can be coagulated with an electrocautery after it is individually identified. We prefer not to use clips at any point in the procedure, as these have been known to migrate through the urethrovesical anastomosis.

FIG. 33-13. The prostatic vascular pedicles have been divided to a point cephalad to the seminal vesicles, exposing the ampullae and the seminal vesicles and mobilizing the prostate. (Modified from Catalona WJ. Prostate cancer. Orlando, FL: Grune & Stratton, 1984;110.)

After both seminal vesicles have been divided, they are lifted superiorly to expose the junction between the posterior prostate and the bladder neck. This is developed partially with electrocautery, and any remaining lateral attachments are also taken down with electrocautery ( Fig. 33-14). This should provide good definition of the bladder neck so that the electrocautery can then be used to incise anteriorly. If the bladder neck is well defined, identification of the ureteral orifices is not necessary before the removal of the surgical specimen is completed by dividing the remainder of the bladder neck attachments ( Fig. 33-15). However, if there is a question of tumor encroachment toward the bladder neck, a wider margin should be taken. Under these circumstances, it is best to visualize the bladder trigone before incising the posterior bladder neck. In rare circumstances, 5-Fr pediatric feeding tubes are passed up the ureteral orifices to facilitate their identification and preservation during the dissection.

FIG. 33-14. Division of the anterior bladder neck with the cautery.

FIG. 33-15. The bladder neck is transected distal to the ureteral orifices. The catheter is pulled through for use as a tractor on the prostate. (Modified from Catalona WJ. Prostate cancer. Orlando, FL: Grune & Stratton, 1984;110.)

After the surgical specimen is removed, careful hemostasis is obtained. If a potency-sparing approach is used, small bleeding vessels along the neurovascular bundle are left alone. Larger bleeding vessels near the bundle are controlled with a 4-0 chromic suture. Otherwise, virtually all other vessels are controlled with electrocautery. In order to identify bleeding vessels in the pelvis, any bleeding from around the genitourinary diaphragm is compressed with a sponge held in place with a small Deaver retractor. Often, if an anatomic dissection has been performed, bladder neck reconstruction is not required. Otherwise, 2-0 Vicryl sutures are placed in a posterior-to-anterior manner to close the bladder neck to the point where it just admits the tip of an index finger. We have found a grooved urethral sound most useful for placement of the urethral anastomotic sutures. External perineal compression has been unnecessary, as the urethral stump generally is readily visible, and suture placement is not difficult. The operating surgeon uses the left hand to manipulate the urethral sound and the right hand to place the anastomotic sutures. We use five sutures spaced proportionately around the urethral circumference. The sutures on the patient's left side are most easily placed outside to in and incorporate only the urethra. We use 2-0 Vicryl for the anastomosis and have a double-armed needle that allows inside-out placement of the sutures on the patient's right ( Fig. 33-16). The sutures are then placed in a corresponding position in the bladder neck. We do not place mucosal eversion sutures before forming the anastomosis, but an Allis clamp placed anteriorly on the bladder neck can nicely evert the mucosa ( Fig. 33-17). A 20-Fr Foley catheter with a 5-cc balloon is passed through the anastomosis before placement of the most anterior sutures to prevent any tangling of the sutures that could result in catheter entrapment. The same style of clamps are always used to tag these sutures and avoid any confusion. For example, straight hemostats are used for the two most posterior sutures, curved hemostats for the two paired anterolateral sutures, and a Kocher clamp for the most anterior suture.

FIG. 33-16. Initial sutures placed in the urethra.

FIG. 33-17. Anastomosis of urethra and bladder.

The Foley catheter balloon is inflated with 8 cc of water, and the table flexion is released. A sponge stick is used to displace the bladder medially so that the anastomotic sutures can be tied. After all of the sutures have been tied, the anastomosis is inspected both visually and manually to make certain that all of the ties have gone down completely and that there is no portion of the catheter either visible or palpable. Rarely, the anastomosis is not felt to be completely secure. Under these circumstances, we take down the anastomosis and redo it. Although this is painstaking, a secure anastomosis can help avoid long-term complications after the procedure. We place a single Jackson–Pratt drain directly over the anastomosis and bring it out through a stab wound just lateral to the incision. The incision is closed with a running 1 Vicryl suture, and the skin is closed with a subcuticular 4-0 Vicryl suture.

OUTCOMES
Complications When the aforementioned technique is used, intraoperative bleeding complications are unusual. Our median blood loss is less than 500 cc. 6 We do not ask patients to donate autologous blood preoperatively, as our transfusion rate is less than 2%. Excessive drainage is extremely uncommon, and the Jackson–Pratt drain is removed on the second or third postoperative day. Thromboembolic problems remain the most frequent cause of serious morbidity and occasional mortality after radical retropubic prostatectomy. Clinically recognized deep venous thrombosis occurs in 3% to 5%, and pulmonary embolus in 1% to 3% of patients. 3 Routine use of anticoagulants in the perioperative period diminishes the rate of thromboembolic complications but does introduce a risk of wound or operative site bleeding or hematoma. We use heparin prophylaxis in patients at a high risk for thromboembolism such as those with obesity or a history of deep venous thrombosis. The most concerning long-term complications of radical prostatectomy are the possibilities of impotence or incontinence. The risk of impotence is directly related to patient age, potency status before surgery, and whether a nerve-sparing approach is used unilaterally or bilaterally. 9 Over half of younger men undergoing a bilateral nerve-sparing approach can anticipate the return of erections suitable for intercourse postoperatively. Incontinence requiring further treatment occurs in around 2% of patients, but up to 15% to 20% of patients have at least occasional episodes of stress incontinence. 1,11 Results We have designed and implemented a collaborative care pathway for patients undergoing radical retropubic prostatectomy. The target date for hospital discharge is the second or third postoperative day. Ninety-four percent of our patients have accomplished this goal with no apparent adverse effect on morbidity. This is, however, but one factor in overall hospital charges. Operating room charges constitute the major component of overall costs in patients undergoing radical retropubic prostatectomy. By addressing all aspects of perioperative care, we have diminished hospital charges and costs by over 40% through a collaborative care pathway. 5 After surgical removal of the entire prostate, the serum prostate-specific antigen level should be undetectable. The success with which this occurs depends on pathologic tumor stage.8 If the tumor is completely confined within the prostatic capsule and of low to intermediate grade, nearly 90% of patients have an undetectable PSA, which apparently translates into long-term cure. 2 As with any surgical procedure, the success of radical retropubic prostatectomy depends on appropriate patient selection and careful attention to anatomic detail and surgical technique. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Geary ES, Dendinger TE, Freiha FS, Stamey TA. Incontinence and vesical neck strictures following radical retropubic prostatectomy. Urology 1995;45(6):1000–1006. Gerber GS, Thisted RA, Scardino PT, et al. Results of radical prostatectomy in men with clinically localized prostate cancer. JAMA 1996;276(8):615–619. Heinzer H, Graefen M, Noldus J, Hammerer P, Huland H. Early complication of anatomical radical retropubic prostatectomy: lessons from a single-center experience. Urol Int 1997;59(1):30–33. Hoenig DM, Chi S, Porter C, et al. Risk of nodal metastases at laparoscopic pelvic lymphadenectomy using PSA, Gleason score, and clinical stage in men with localized prostate cancer. J Endourol 1997;11(4):263–265. Koch MO, Smith JA Jr. Influence of patient age and co-morbidity on outcome of a collaborative care pathway after radical prostatectomy and cystoprostatectomy. J Urol 1996;155(5):1681–1684. Koch MO, Smith JA Jr. Blood loss during radical retropubic prostatectomy: Is preoperative autologous blood donation indicated? J Urol 1996;156(3):1077–1079. Partin AW, Criley SR, Subong EN, Zincke H, Walsh PC, Oesterling JE. Standard versus age-specific prostate specific antigen reference ranges among men with clinically localized prostate cancer: A pathological analysis. J Urol 1996;155(4):1336–1339. Pound CR, Partin AW, Epstein JI, Walsh PC. Prostate-specific antigen after anatomic radical retropubic prostatectomy. Patterns of recurrence and cancer control. Urol Clin North Am 1997;24(2):395–406. Quinlan DM, Epstein JI, Carter BS, Walsh PC. Sexual function following radical prostatectomy: influence of preservation of neurovascular bundles. J Urol 1991;145(5):998–1002. Smith JA Jr, Scardino PT, Resnick MI, Hernandez AD, Rose SC, Egger MJ. Transrectal ultrasound versus digital rectal examination for the staging of carcinoma of the prostate: results of a prospective, multi-institutional trial. J Urol 1997;157(3):902–906. Steiner MS, Morton RA, Walsh PC. Impact of anatomical radical prostatectomy on urinary continence. J Urol 1991;145(3):512–514.

Chapter 34 Radical Perineal Prostatectomy Glenn’s Urologic Surgery

Chapter 34 Radical Perineal Prostatectomy
Sam D. Graham, Jr.

S. D. Graham, Jr.: Department of Urology, Emory University School of Medicine, The Emory Clinic, Atlanta, Georgia 30322.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Radical perineal prostatectomy, first performed in 1869 by Buchler and popularized in the United States by Young in 1905, remained the primary surgical approach to carcinoma of the prostate until the mid-1970s. 11 With the recognition of the importance of assessing pelvic lymph nodes preoperatively and the advantage that retropubic prostatectomy offered with concomitant pelvic node dissection, perineal prostatectomy declined in popularity for the treatment of prostate cancer. The perineal approach, however, has seen a resurgence in the 1990s for several reasons: (a) the trend toward minimally invasive surgery with a focus on reducing the morbidity and therefore the hospital stay of patients, (b) the advent of laparoscopic surgery for lymph node assessment, (c) the introduction of PSA for screening for prostate cancer with reduction in the numbers of patients with node-positive disease, and (d) algorithms that may predict patients at high risk for positive lymph nodes. The procedure is also associated with reduced blood loss, low morbidity, and can be modified to incorporate the neurovascular sparing techniques for preservation of potency.

DIAGNOSIS
All patients who are potential candidates for radical perineal prostatectomy should undergo preoperative staging to assure that the patients are operable candidates. Methods of differentiating local from advanced disease include digital rectal examination, transrectal ultrasonography, radionuclide bone scan, assessment of pelvic lymph nodes, as well as pathologic indicators of progression such as Gleason sum and other markers. Since the late 1980s, PSA has made a significant impact on the preoperative stage of patients with prostate cancer. 2 Patients presenting for surgery are generally younger, healthier, and more likely to have organ- confined prostate cancer than the population treated only a decade earlier, and this in many ways contributes to the large increase in the number of radical prostatectomies done in the United States in the past decade. Other contributing factors are the modifications in technique that have reduced morbidity, including the nerve-sparing technique described by Walsh and alternative methods of restoring potency. 10 Digital rectal examination has a limited role in the clinical staging of prostate cancer. Its primary capability is to crudely estimate the volume of the cancer. Transrectal ultrasonography is another modality that also has limitations in assessing local disease but, combined with digital rectal examination, at least gives some gross assessment of likelihood of extracapsular disease. Other modalities such as transrectal MRI, CT scan, and pelvic MRI have been shown to have limited usefulness. Radionuclide bone scans are useful in assessing advanced bony disease but generally are not positive in patients with PSA below 20 ng/ml and no other sign of advanced disease. For the past 11 years we have prospectively applied an algorithm to the preoperative assessment of patients with prostate cancer based on the evaluation of over 400 patients who had undergone pelvic node dissection at our institution. The current algorithm includes patients with a Gleason 7 or less, with low-volume cancer (T 1b–c, T2a), normal acid phosphatase, and PSA of less than 20. Patients meeting all of these criteria have a less than 5% chance of positive lymph nodes, and, therefore, we do not perform pelvic lymph node dissections. Patients exceeding any one of the above criteria are considered to be in the high-risk group and have undergone pelvic lymph node dissections. 4 With this method of assessment, our PSA recurrence rate from 1988 through 1994 was 27%, which compares favorably to our own and other series of retropubic prostatectomies, which have shown PSA recurrence rates between 24% and 29%. 4,7,9 Because 75% of the patients who did not develop PSA recurrences had PSA <10 ng/ml, we have now modified our algorithm to place patients in the low-risk group if the serum PSA is <10 ng/ml. 4

INDICATIONS FOR SURGERY
Patients who are candidates for radical prostatectomy must have clinically organ-confined prostate cancer (T 1–2). In addition to having organ-confined disease, other factors that need to be taken into consideration are the patient's life expectancy, other comorbidities, or any other factors that may affect the patient's choice. We generally would not offer a radical prostatectomy to patients who have a life expectancy less than 10 years. Over the age of 70, we would offer a radical prostatectomy only in selected cases in which we feel that the benefits that can be obtained from radical prostatectomy outweigh the potential risks, particularly when compared to alternative therapies.

ALTERNATIVE THERAPY
Alternatives to radical prostatectomy include observation, hormonal deprivation, and radiation therapy. We do not consider either observation or hormonal deprivation to be curative, and the patients for whom this is a good option are those patients with less than 5 years of life expectancy, patients who are over 70 years old with a well-differentiated cancer, or patients who are a high risk for surgery and refuse radiation. Overall, observation is associated with a 75% mortality over 10 years. 5 Radiation therapy, however, may be definitive and has an equivalent 5- and 10-year survival. The recurrence rates with radiation therapy are bimodal, with initial recurrences within 1 to 2 years of treatment and a delayed peak at 5 to 7 years after treatment. 8 In our institution, 359 patients who received brachytherapy from 1972 to 1984 were compared to a contemporaneous series of 161 patients undergoing radical prostatectomy. The 5-year recurrence rates and survival were similar in both groups, which were equivalent in preoperative stage, age, Gleason sum, and other demographics. By 7 years, however, the recurrence rate among the brachytherapy group was 48% (9% local recurrence, 39% distant recurrence) versus 27% in the radical prostatectomy patients (8% local and 19% distant recurrence). If the patient is young and has a 15-year or longer outlook, we feel that our results would favor radical prostatectomy.

SURGICAL TECHNIQUE
Before the patient is put in position, the legs are wrapped with Ace bandages. The patient is placed in an exaggerated lithotomy position ( Fig. 34-1). It is important that the patient's perineum be parallel to the floor in that this directly affects exposure. We use a standard operating room table with folded sheets under the patient's sacrum supporting the patient's entire weight. We do not use shoulder braces, and if a patient tends to slide off the sheets, we will place the table in a slight reverse Trendelenburg position. The patient's legs are stabilized using candy cane stirrups, again taking precautions to prevent stretching the hamstring or causing pressure on the legs.

FIG. 34-1. Exaggerated lithotomy position is achieved by placing the patient's sacrum on a roll of sheets. Care is taken not to stretch the hamstrings or place pressure on the shoulders.

Four instruments are significant in assisting the surgeon for this operation. These include the Lowsley curved tractor, the Young straight prostatic tractor, a halogen headlamp, and an Omnitract miniwishbone retractor system. The curved Lowsley tractor is used to bring the prostate up into the perineum to allow the dissection against the prostate while separating the rectum from the prostate. The straight Young tractor is used to manipulate the prostate laterally as well as cephalad and caudad after the membranous urethra has been divided. The halogen headlamp is important in that it allows the surgeon to aim a strong light into the operative field, which may be deep and narrow, preventing standard operating lights from adequately illuminating the structures. The Omnitract miniwishbone allows virtually unlimited retraction in any direction. It should be noted that, in manipulation of the prostate from the perineum, the pelvis should be viewed as a cone with its apex at the site of the incision ( Fig. 34-2). To get better visualization at times, it may be necessary to actually push the prostate further into the pelvis. Also note that traction is not placed directly on the bulb or membranous urethra, as this will decrease the likelihood of restoration of potency and potentially affect the patient's continence postoperatively.

FIG. 34-2. The pelvis is an inverted cone with the apex at the site of the incision. To gain better visualization during perineal procedures, it is helpful to push the prostate further into the wound.

The incision is made from the ischial tuberosity, crossing the midline at the juncture between the squamous epithelium and mucocutaneous border of the rectum ( Fig. 34-3). The incision extends posteriorly to a line equal to the posterior portion of the anus. By use of sharp dissection and electrocautery, the ischiorectal fossae are entered, and the central perineal tendon is identified by blunt dissection and transected with an electrocautery. At this point, we employ the Belt approach and dissect down to the white fascia of the rectum and proceed subsphincterically ( Fig. 34-4). Predominantly by blunt dissection with an index finger in the rectum, the rectal sphincter and levator ani can be dissected free of the rectum with minimal bleeding ( Fig. 34-5), and the blades from the miniwishbone retractor are then used to retract these muscles anteriorly and laterally. With tension on these muscles and tension on the rectum, the rectourethralis is identified and divided, which allows the surgeon to dissect the rectum free of the apex of the prostate ( Fig. 34-6).

FIG. 34-3. The incision for a perineal prostatectomy begins with an incision extending from ischial tuberosity to ischial tuberosity approximately 1.5 cm from the anus. Posteriorly, the incision is in a plane with the posterior anus. The ischiorectal fossa is entered, and the central tendon is divided, allowing dissection onto the rectal fascia.

FIG. 34-4. The Belt technique (subsphincteric approach) allows the surgeon to stay on the rectum and reduces the likelihood of rectal injury. Passive stretching of the sphincter does not contribute to rectal incontinence.

FIG. 34-5. The levator ani are dissected from the rectum using blunt dissection and retracted anterolaterally to allow access to the pelvis.

FIG. 34-6. (A) The rectourethralis is divided to allow the rectum to be retracted posteriorly. (B) The surgeon's finger in the rectum during the dissection aids in identifying the rectal anatomy.

If this is to be a nerve-sparing technique, the dissection is carried down to approximately 1.5 to 2 cm from the apex, at which point the posterior layer of Denonvillier's fascia is divided, and dissection is carried between the two layers of the Denonvillier's fascia ( Fig. 34-7). Care is taken not to damage the neurovascular bundles that course along the lateral posterior prostate on either side ( Fig. 34-8).

FIG. 34-7. Development of the plane between the prostate and rectum. The most common approach is to incise the posterior layer of Denonvillier's fascia (“The Pearly Gates”) and dissect between the two layers. Alternatively, the dissection may be along the anterior rectum, resecting both layers of Denonvillier's fascia with the specimen.

FIG. 34-8. The neurovascular anatomy as it relates to potency can easily be visualized from the perineal approach. The neurovascular bundle is seen lateral and posterior to the prostate at the base and crosses the apical portion of the prostate to enter the pelvic floor posterior to the membranous urethra. The superior prostatic pedicle is a constant structure, but the inferior pedicle is more variable.

The distal portion of Denonvillier's fascia is then incised in the midline with scissors, and the tag is then used to facilitate dissection of the neurovascular bundle from the prostate; the inferior pedicle, if present, is ligated and divided ( Fig. 34-9). If the dissection is in any way impaired by fibrosis such that there is a potential for prostatic tissue to be left behind, the neurovascular bundle is sacrificed on that side. It should be noted during this dissection that the neurovascular bundle actually courses across the posterior surface of the prostate at the apex and enters the urogenital diaphragm just posterior to the membranous urethra. This proximal relationship is important in that the vesicourethral anastomosis may incorporate the neurovascular bundle if the sutures are placed too deeply during the anastomosis. Once the neurovascular bundle has been dissected, a Vessi-loop is placed around this to aid in lateral traction.

FIG. 34-9. (A) The apical portion of the posterior layer of Denonvillier's fascia is incised to the membranous urethra. (B) The inferior pedicle is identified and ligated. (C) The inferior pedicle is divided, and the neurovascular bundle is dissected from the prostate.

The retraction of the neurovascular bundle on either side thereby exposes the proximal membranous urethra, allowing a right-angled clamp to be placed around the membranous urethra. There should be little resistance anterior to the urethra to the passage of the tip of the right-angled clamp anterior to the urethra if one stays posterior to the endopelvic fascia ( Fig. 34-10). The Lowsley tractor is removed, and the membranous urethra is divided. The Young tractor is then placed into the bladder via the severed prostatic urethra, and the endopelvic fascia is dissected free of the anterior prostate. In most cases, there is insignificant bleeding from the dorsal venous complex, but if there should be communicating veins, they should be ligated using 3-0 Vicryl.

FIG. 34-10. (A) The division of the membranous urethra is performed by placing a right-angled clamp around the membranous urethra just distal to the apex of the prostate. (B) The endopelvic fascia is dissected from the anterior prostate to allow access to the bladder neck. Note Young tractor in place.

The groove between the prostate and bladder is identified, and, with either sharp or blunt dissection, the prostate and bladder can be separated. If there is any resistance to blunt dissection, the bladder neck should be sharply divided, and biopsies taken of the bladder neck to assure that the resistance is not secondary to bladder neck invasion with the cancer. Patients who have had prior transurethral resections may have an obliteration of the plane between the prostate and bladder, and the blades of the Young tractor can be used to aid in this identification. The prostate is dissected from the bladder anteriorly to the 5-o'clock and 7-o'clock positions, respectively, on the patient's left and right; the bladder neck is divided over the Young tractor, and the Young tractor is removed ( Fig. 34-11). The bladder is evacuated of any urine, and the posterior bladder neck is divided at its juncture with the prostatic urethra. The prostate is then dissected free from the posterior bladder, allowing identification of the superior pedicles of the prostate as well as the seminal structures (Fig. 34-12). The superior pedicles are isolated, divided, and ligated. The seminal vesicles are dissected to their tips with blunt dissection, and the artery from the seminal vesicle is either cauterized or ligated; the vasa deferentia are generally cauterized. Finally, the Denonvillier's fascia overlying the seminal structures is divided, allowing removal of the prostate.

FIG. 34-11. Division of the bladder neck. (A) In most cases, a plane can be developed between the prostate and the circular fibers of the bladder neck. (B) If such a plane is not possible, the bladder neck is resected using the Young tractor as a guide.

FIG. 34-12. Mobilization of the retrovesical structures. (A) After the bladder neck is divided, the bladder is bluntly dissected from the posterior structures. The superior prostatic pedicles are divided, and the seminal structures are dissected. (B) The vasal ampullae are divided, and the seminal vesicles are dissected from the surrounding tissue. Care should be taken to control the artery to the seminal vesicle that is found at the apex.

After complete hemostasis has been ensured, the bladder neck is reconstructed using 3-0 Vicryl on an SH needle, beginning posteriorly to anteriorly ( Fig. 34-13) as described by Dees. 3 This direction of the closure, beginning in the posterior bladder, is done to facilitate the closure without injury to the ureters and also to take

advantage of the anatomic relationship between the bladder neck and the membranous urethra with the shorter distance being anteriorly. The anastomosis is performed using 3-0 Vicryl simple sutures and an RB-1 controlled-release needle ( Fig. 34-14) around a 22-Fr 5-cc Foley catheter. Generally, seven or eight interrupted sutures are used for this anastomosis, though, alternatively, the anastomosis can be performed with a running suture. Care should be taken that small portions of the membranous urethra are incorporated in the anastomosis so that the continence mechanism is left undisturbed and the neurovascular bundles that contribute to potency are avoided ( Fig. 34-15).

FIG. 34-13. The vesical neck is reconstructed from posterior to anterior.

FIG. 34-14. The anastomosis is performed with interrupted sutures over a 22-Fr Foley catheter.

FIG. 34-15. The anastomosis should avoid injury to the neurovascular bundles posteriorly by incorporating only the urethra.

The rectum is then inspected; a Foley balloon is inflated, and a Penrose drain is placed through the left ischiorectal fossa and a separate stab incision ( Fig. 34-16). The incision is closed with a 3-0 chromic gut closure. One suture is placed to reapproximate the central tendon, and the remainder of the sutures are used to close the skin in a horizontal mattress.

FIG. 34-16. A drain is placed near the anastomosis, and the incision is closed.

Postoperatively, the patients have very low requirement for pain medication. Most patients either do not require parenteral pain medication or are off the parenteral medications within 12 to 24 hours. Average time to discharge is approximately 48 hours from the time of surgery. The patient's catheter is removed on the 12th day. The Penrose drain is removed before discharge.

OUTCOMES
Complications Perioperative complications include hemorrhage, wound infection, cardiovascular complications, and rectal injury. The incidence of rectal injury is less than 2%, and with current techniques, preoperative bowel preparation (Golightly), and antibiotics, these are closed primarily in two layers without the need to perform a diverting colostomy. Wound infection rates are less than 1%, and cardiovascular complications are approximately 1%. The average blood loss in these patients is approximately 450 cc, and our transfusion rate is less than 5%. Long-term complications include incontinence and impotence. Incontinence requiring intervention such as pads, clamps, or inflatable devices occurs in 2.8% of our patients. We have found that incontinence generally occurs in patients who are older, obese, or have had prior radiation therapy. Potency following nerve-sparing

perineal prostatectomy is dependent on the patient's age and preoperative status. Patients under the age of 60 who are fully potent and have both neurovascular bundles spared have approximately a 50% potency rate. Patients who are over the age of 60 have a reduced rate of potency, and we have not yet had a patient over the age of 70 who has spontaneously regained his potency. Patients who are having difficulties with potency before surgery and patients in whom the neurovascular bundles could not be spared will likely be impotent. We generally advocate early use of pharmacologic or other means of assistance in these patients to help them regain their potency. Results Following radical prostatectomy, recurrence can be measured using PSA, which is exquisitely sensitive. Any patient who undergoes a radical prostatectomy can expect his PSA to fall below detectable levels. Its failure to do so generally means that the patient has significant residual disease, either locally or distantly. Another group of patients will have an initial drop in their PSA to undetectable levels and then a return to measurable levels. These patients may have local and/or distant recurrence of their disease or possibly residual malignancy. Patients who develop recurrence based on PSA will generally manifest a clinical progression within 18 months. Additionally, if the PSA is going to rise, it will do so within 2 years in 90% of patients and within 4 years in virtually every patient. Based upon surrogate markers such as PSA, we can now predict disease recurrence earlier, allowing assessment of the outcome of radical prostatectomy within 5 years as opposed to the older data, which required a 7- to 10-year follow-up. The primary predictor of recurrent disease after radical prostatectomy is the presence of positive margins. In SEER data, radical prostatectomy has been shown to have a long-term disease-free rate approaching 90% at 10 years with organ-confined disease. 6 Positive margins may reflect capsular penetration, invasion of the periprostatic tissue, or may reflect a pathologic discrepancy. Patients with negative margins will have a 10% risk of recurrence, and the risk of recurrence with positive margins will be 30% to 50%. Before PSA screening, the incidence of positive margins in patients undergoing radical prostatectomy in our institution was 48% with an overall PSA recurrence rate of 24%. Whether the current reports of 20% positive margin rates with serial PSA screenings has a proportional drop in the PSA recurrence rates remains to be seen. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Belt E, Ebert CE, Surber AC Jr. A new anatomic approach in perineal prostatectomy J Urol 1939;41:482. Catalona WJ, Smith DS, Ratliff TL, Basler JW. Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening. JAMA 1993;270(8):948–954. Dees JE. Vesicourethral anastomosis in radical perineal prostatectomy. Urol Dig 1969;8:16–19. El-Galley Rizk ES, Keane TE, Petros JA, et al. Evaluation of staging lymphadenectomy in prostate cancer. Urology (in press). Gronberg H, Damber L, Jonson H, Damber JE. Prostate cancer mortality in northern Sweden, with special reference to tumor grade and patient age. Urology 1997;49(3):374–378. Krongrad A, Lai H, Lai S. Survival after radical prostatectomy. JAMA 1997;278(1):44–46. Oesterling JE, Chan DW, Epstein JI, et al. Prostate specific antigen in the preoperative and postoperative evaluation of localized prostatic cancer treated with radical prostatectomy. J Urol 1988; 139(4):766–772. Partin AW, Kattan MW, Subong EN, et al. Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update. JAMA 1997;277(18):1445–1451. Scardino PT, Frankel JM, Wheeler TM, et al. The prognostic significance of post-irradiation biopsy results in patients with prostatic cancer. J Urol 1986;135(3):510–516. Trapasso JG, deKernion JB, Smith RB, Dorey F. The incidence and significance of detectable levels of serum prostate specific antigen after radical prostatectomy. J Urol 1994;152(5 Pt 2):1821–1825. Walsh PC, Mostwin JL. Radical prostatectomy and cystoprostatectomy with preservation of potency: results utilizing a new nerve sparing technique. Br J Urol 1984;56:694–797. Young HH. The early diagnosis and radical cure of carcinoma of the prostate: being a study of 40 cases and presentation of radical operation which was carried out in 4 cases. Bull Johns Hopkins Hosp 1905;16:315–321.

Chapter 35 Brachytherapy for Localized Prostate Cancer Glenn’s Urologic Surgery

Chapter 35 Brachytherapy for Localized Prostate Cancer
Haakon Ragde

H. Ragde: Department of Prostate Brachytherapy, Urologic Services, Pacific Northwest Cancer Foundation, Northwest Hospital, Seattle, Washington 98133.

Indications for Surgery Alternative Therapy Surgical Technique Preimplant Planning Operative Implant Postimplant Evaluation and Management Outcomes Complications Results Chapter References

The Crookes' tube from which the Roentgen rays are emitted is, of course, too bulky to be admitted into the middle of a mass of cancer. But there is no reason why a tiny fragment of Radium sealed in a fine glass tube should not be inserted into the very heart of the cancer, thus acting directly upon the diseased material. Would it not be worthwhile making experiments along this line? Excerpt from a letter written by Dr. Alexander Graham Bell to the editor of Archives of Roentgen Ray in 1903 “Brachy”(therapy), meaning “short” in Greek, describes treatment with radioactive sources or materials placed into, or at a short distance from, the tissue to be radiated. Brachytherapy stands in contrast to “tele”(therapy), Greek for “long,” which refers to external radiation delivered at a distance from the patient and the tumor. Although early attempts at delivering this form of spatially controlled radiation to the prostate date back to the early part of the century, significant clinical investigation of this treatment method did not get under way until the 1960s. At that time, Carlton and co-workers began using radioactive gold combined with external beam irradiation, and Whitmore and associates implanted radioactive iodine. Both procedures consisted of retropubic exposure, bilateral pelvic lymph node dissection, and free-hand insertion of the implant needles for placement of the seeds ( Fig. 35-1).

FIG. 35-1. Schematic illustration of the now obsolete freehand technique of prostate seed implantation.

Initially, this innovative treatment for prostate cancer excited interest. A highly confined radiation dose was delivered to the prostate, sparing adjacent, uninvolved tissue. The complication rates, notably incontinence and impotence, were lower than those reported after surgery and external beam therapy of the time period. The free-hand placement technique, however, all too often resulted in poor radiation dose distributions and unsatisfactory local control rates 6,7 (Fig. 35-2). Additionally, in an era of digital tumor detection, the local failure rates were further escalated by implanting bulky prostates at advanced stages of disease. The unfavorable local control rate, coupled with ongoing improvements in competing treatment modalities, soon led to dwindling interest in prostate brachytherapy.

FIG. 35-2. X-ray of surgically removed prostate implanted by the freehand technique. Note large cold spot (arrows).

Significant new developments occurred in the 1980s that served to remedy the shortcomings of the open implant method and rekindle interest in the procedure. These improvements included high-tech imaging and computer software that permitted precise measurement of prostate volume and shape and optimal dose determination. Further, transperineal insertion of the implant needles under real-time transrectal ultrasound monitoring allowed for accurate and reproducible seed placement and more uniform source distribution 1 (Fig. 35-3). Using low-energy, short-range radioisotopes that favored protection of the adjacent uninvolved tissue permitted delivery of higher radiation doses than could safely be administered by external beam techniques and had the potential for improving local control.

FIG. 35-3. Schematic illustration of modern transrectal ultrasound-guided prostate seed implantation.

Today, modern prostate brachytherapy can be performed on a cost-effective outpatient basis. It does not require a surgical incision, early results are encouraging, morbidity is minimal, and the patient can usually resume his normal activities in a day or two after treatment. Dictated by isotope selection, transperineal prostate seed implantation can be divided into temporary and permanent implants. Temporary implants, always combined with external beam irradiation, utilize high-energy sources, such as iridium-192, that are left in the patient for a specific time period and removed. Permanent implants, on the other hand, are left in the patient to decay to an inert state over a specific time period. Although a few centers continue to use gold-198 for implantation, the majority of present-day permanent implants use iodine-125 and palladium-103, both low-energy, short-range sources. This chapter is limited to permanent implantation with these two radionuclides. Information available to patients to help evaluate and select a treatment program for prostate cancer has significantly increased over the last several years. Prostate brachytherapy, therefore, incorporating up-to-date dosimetry and seed placement technology, should provide the urologist with a timely and practical treatment alternative for patients who either are medically unfit for surgery or do not wish to have it. Essential to successful application of brachytherapy techniques are proper planning, technical expertise, and meticulous execution. As practiced today, it is a team effort requiring urologic surgical skills, radiotherapeutic planning expertise, and medical physics. The combined efforts and continuous participation of these specialties are important for the success of the procedure. The technique is a three-step process: preimplant planning, operative implant, and postimplant quality evaluation.

INDICATIONS FOR SURGERY
Indications for brachytherapy are any patient with localized adenocarcinoma of the prostate who otherwise has a life expectancy greater than 5 years.

ALTERNATIVE THERAPY
Alternatives to brachytherapy include observation, hormonal deprivation, external beam radiation (teletherapy), and radical prostatectomy.

SURGICAL TECHNIQUE
Preimplant Planning The main purpose of implant planning is to assure a systematic approach to the individual patient. The custom-tailored plan permits delivery of a high dose of radiation to the prostate with maximal sparing of juxtaposed healthy tissue. The “preplan” has several components. Patient Selection With the effective radiation encompassing only a 5-mm margin beyond the prostate, patients selected for interstitial radiation with iodine-125 and palladium-103 as monotherapy should have a high probability of organ-confined disease. No generally accepted and applied criteria for the selection of organ-confined lesions exist. However, increasing data indicate that the combination of clinical stage, number and locations of positive biopsies, presenting PSA level, and biopsy Gleason pattern scores are helpful in singling out patients with the highest probability of having lesions confined to the prostate, i.e., having all the prostate cancer included in the effective radiation field. 3,4,8 For patients suspected of having periprostatic extension of the disease, an implant alone is not adequate therapy. The addition of external beam irradiation to a dose of 45 Gy may be indicated in order to sterilize periprostatic tumor extension. Comorbidity status as it relates to anesthetic risks must be evaluated, though restrictions for a 45-minute bloodless seed implantation need not be as stringent as for surgery. Additionally, it is important to identify risk factors that may give rise to urinary and rectal complications after implantation. Significant urinary obstructive symptoms, for example, may lead to urinary retention because of swelling from the radiation. If the lateral lobes are the cause of obstruction, a 3-month course of total androgen ablation before implantation will usually rectify the condition. Median bar obstruction, in our experience, responds poorly to hormonal manipulation and is best corrected before the implant with a transurethral incision of the prostate (TUIP). A previous transurethral prostate resection (TURP), performed even years before, carries increased risk for stress incontinence. 2 In such patients, differential loading of the sources away from the prostatic urethra may help prevent this complication. Selecting large glands for implantation carries two types of risks. First, the bony pubic arch may overlay the anterolateral part of the prostate and prevent transperineal needle insertion ( Fig. 35-4). Second, large glands require more seeds. Consequently, there is an increase in total dose, which may adversely affect adjacent tissue, such as the rectum and bladder. For these reasons, current knowledge dictates caution in implanting glands over 50 cc. The majority of large glands may be reduced to acceptable volumes by total androgen ablation for 3 months before brachytherapy ( Fig. 35-5).

FIG. 35-4. Correlated transverse CT image of the prostate demonstrating significant pubic arch obstruction that would interfere with needle insertion.

FIG. 35-5. Correlated transverse CT scans of the prostate. (A) Before treatment with an LH-RH agonist and an androgen receptor blocker, showing significant pubic arch obstruction. (B) After 3 months of treatment, bony interference to needle insertion is no longer a problem.

Prostate Volume Specification An accurate prostate volume description with delineation of adjacent rectum, urethra, and bladder is fundamental to precise source positioning and conformal dosimetry. Although volume determination and treatment planning can be performed using computed tomography, most centers prefer to use transrectal ultrasound step-sectional planimetry. Advantages include its low cost, excellent cross-sectional anatomy, and high correlation with real-time monitoring during the actual implant. The circumference of 5-mm spaced transverse prostate images from apex to base of the gland, overlaid with a template configuration corresponding to the template needle puncture channels, are demarcated with a lightpen ( Fig. 35-6). Computer software calculates the volume.

FIG. 35-6. TRUS volume specification, depicting 5-mm transverse sections of the prostate, apex to base, with overlaid template coordinates and lightpen-demarcated glandular margins.

Prostate–Pubic Arch Relationship Pubic arch interference, where the pubic rami may prevent the transperineal insertion of implant needles, most commonly occurs in glands over 50 cc but may also be encountered with smaller glands. It is important to recognize this condition in the planning stage rather than in the operating room. A simple method of detecting pubic arch obstruction entails the superimposition of the ultrasound image of the pubic arch over the widest transverse image of the prostate ( Fig. 35-7). In the face of pubic arch obstruction, shrinkage of the gland through total androgen blockade for 2 to 3 months may be necessary to make an implant possible.

FIG. 35-7. Transrectal ultrasound image of the pubic arch superimposed over the widest transverse image of the prostate. No bony obstruction to interfere with needle insertion is evident.

Seed-Loading Method Selection Once the prostate volume and its spatial geometry have been determined, the seed-loading pattern is selected. The Quimby pattern is characterized by a uniform source distribution across the target volume ( Fig. 35-8). The Patterson–Parker motif uses a peripherally weighted source allocation, implanting 60% to 70% of the total activity into the periphery of the gland ( Fig. 35-9). The Quimby method results in a radiation distribution that is characterized by a high central dose at the midpoint of the prostate and by a larger number of lower-strength seeds. Implants by this method are technically easier because small seed movements are less apt to result in underdosage or overdosage.

FIG. 35-8. Schematic illustration of the uniform (Quimby) technique of implantation.

FIG. 35-9. Schematic illustration of the peripherally weighted (Patterson–Parker) technique of implantation.

The high central dose may have the further advantage of greater tumor destruction but must be applied with caution in patients who have had a prior transurethral resection and consequent damage to the urethral blood supply. In these patients, the peripherally weighted Patterson–Parker method, with its lower central dose and lower likelihood of urethral damage, may be preferable. However, peripheral loading represents a more difficult implant and demands greater precision in source placement for a homogeneous dose distribution. It has the further disadvantage of delivering lower central doses to the tumor and involves a higher risk of rectal injury because high-activity seeds are placed close to the rectum. Target Volume Specification The target volume differs from the prostate volume in that it encompasses a 2- to 5-mm margin beyond the prostate periphery on the 5-mm-spaced transverse images, with slightly more generous margins at the apex, base, and biopsy-positive tumor sites ( Fig. 35-10).

FIG. 35-10. Transverse TRUS image of prostate. The inner white line comprises the glandular margins, and the outer white line the target delineation for this particular image.

Isotope Selection (Iodine-125/Palladium-103) The two sources are both low-energy emitters. Both are physically similar, enclosed in miniaturized biocompatible titanium cylinders. With their low tissue penetration ability, they pose little or no hazard to medical personnel. The isotopes differ primarily in their half-lives and, therefore, in the rate at which they deliver radiation. Iodine-125, with a half-life of 60 days, emits energy at 8 to 10 cGy per hour, whereas palladium-103, with a half-life of 17 days, delivers at a rate of 20 to 24 cGy per hour. With lower dose rates, recovery of sublethally damaged cells may, at least conceptually, lessen the ultimate tumoricidal effect of the radiation. Some investigators therefore prefer to use the higher-dose-rate palladium-103 when treating poorly differentiated tumors. Because of the increased biological effect of the higher dose rate of palladium-103, some reduction in the total target dose is necessary. Conformal, Computer-Based Dosimetry Most institutions performing brachytherapy use the matched peripheral dose (MPD) convention developed at the Memorial Sloan-Kettering Cancer Center to describe the dose of radiation delivered to the tumor over the entire period of decay for the radioisotope used. The aim here is to deliver a dose to the prostate margins approximating that achieved with external beam therapy, which has been determined to be 160 Gy for iodine and 115 Gy for palladium. When brachytherapy is combined with a preliminary course of 45 Gy of external beam supplement, the iodine dose is lowered to 120 Gy, and palladium to 90 Gy. The spatial seed configuration is determined by entering each of the serial target volume images into a dosimetry computer to determine source spacing and strength for optimum dose homogeneity and distribution that will deliver the prescribed MPD to the periphery of the prostate while limiting radiation to adjacent structures ( Fig. 35-11). Isodose curves are then calculated for each transverse ultrasound image to determine if the proposed dose adequately covers the entire gland ( Fig. 35-12). If not, the seed location, number of seeds, or seed strength is adjusted to assure optimal coverage.

FIG. 35-11. Computer-generated three-dimensional model of the prostate showing the proposed seed distribution.

FIG. 35-12. Ultrasound-based computer image of transverse section of midprostate showing planned seed positions and resultant isodose coverage.

The Implant Worksheet The computer-derived isodose plan with digitized target volume contours and resultant seed configuration is tabulated onto a worksheet ( Fig. 35-13). This reference facilitates needle loading and guides the urologist and radiation oncologist during the operative implant.

FIG. 35-13. Computer-generated worksheet detailing an individually prepared implant itinerary used at Northwest Hospital. The plan depicts individual needles to be implanted, the number of seeds per needle, and the depth of needle insertion.

Operative Implant The implant, as described by Holm and associates in 1983, consists of implanting 18-gauge needles preloaded with seeds and spacers. The treatment plan designates specific template coordinates and prostate location. In order to ensure accurate placement of the needles within the prostate, needles are guided to their specific designations in the prostate by transrectal ultrasound. The procedure is done under spinal anesthesia with the patient in the dorsal lithotomy position and the urethra injected with an ultrasound enhancement agent (emulsion of air and K-Y jelly) for visualization. The scrotum is displaced toward the abdominal wall with a plastic/adhesive drape, and the perineum is prepped. Brackets fastened to the operating table support a stepping unit to hold a biplanar, multifrequency endorectal transducer (Bruel & Kjaer model 8551, Marlborough, MA) that is attached to an ultrasound system (Bruel & Kjaer model 3535). A multichannel needle-steering device, which corresponds to the electronic grid matrix superimposed on the transverse ultrasound prostate images of the volume specification, is attached to the rectal probe ( Fig. 35-14).

FIG. 35-14. TRUS probe with transducer and template anchored in a stepping unit (Bruel & Kjaer model UA 1084).

The probe with the transducer is inserted into the rectum. While scanning through the gland with the template coordinate grid activated, the probe is adjusted until the sequential images on the TV monitor correlate with the planning scan images. At that time the support brackets are locked in position. The prostate is very mobile and may need to be stabilized before needle insertion ( Fig. 37-15 and Fig. 37-16). The implantation begins anteriorly and proceeds posteriorly to prevent target shadowing of seeds already placed. Each needle is guided to its preplanned position in the gland under direct transverse and/or sagittal ultrasound observation ( Fig. 37-17 and Fig. 37-18).

FIG. 35-15. Drawing of the Morgenstern stabilizing needle (MD Tech, Gainesville, FL). (A) Closed (insert) position. (B) Open (anchor) position.

FIG. 35-16. Schematic of coronal section of prostate stabilized with a pair of Morgenstern needles.

FIG. 35-17. Transverse ultrasound image of the prostate base showing the bright echo of an implant needle tip at template coordinates C 3.5.

FIG. 35-18. Implant needle echo as seen in the sagittal scanning mode.

Real-time monitoring of the needle insertion process is of critical importance. Even though the prostate has been stabilized, needle insertion may distort and move the gland. Any deviation and internal distortion should be recognized and adjusted for. When a needle is correctly positioned, the obturator is held stationary by an assistant, and the needle is withdrawn. In this way, rows of alternating seeds and spacers are deposited into the preplanned positions in the gland. The position of the base and prostatorectal interface are monitored throughout the procedure. The operator must regularly observe the transverse and sagittal images. The ultrasound transducer is adjusted in 5-mm increments in the caudal direction for those template coordinates that do not call for seeds at the most cephalad portion (base) of the gland. At the completion of the needle insertion, an AP fluoroscopy is performed to assess the uniformity of seed distribution ( Fig. 35-19). Extra seeds may be implanted wherever there appears to be a spatial deficiency. Also, fluoroscopy will portray stray seeds in the bladder, which may be removed cystoscopically and reinserted.

FIG. 35-19. Postimplant AP fluoroscopy depicting seed distribution.

Postimplant Evaluation and Management Evaluation of implant quality is performed on every patient using three-dimensional CT-based calculations. The evaluation consists of dose computation and dose analysis for target and surrounding structures based on the actual implant. Five-millimeter slice thicknesses are scanned using soft-tissue-density images for prostate volume and bone density windows for seed delineation ( Fig. 35-20). Isodose curves are generated for each CT image, yielding a detailed analysis of radiation distribution relative to the CT-determined target volume ( Fig. 35-21).

FIG. 35-20. Postimplant transverse CT images of the prostate showing uniform seed distribution.

FIG. 35-21. Postimplant transverse CT image of the prostate showing isodose coverage obtained.

Patients are routinely examined at 3-month intervals for the first 2 years, every 6 months for the next 3 years, and yearly thereafter. All follow-up visits consist of clinical evaluation and serum PSA measurement. Four-quadrant TRUS-guided postimplant needle biopsies are recommended for all patients. Treatment results are determined based on clinical freedom from disease, freedom from biochemical (PSA) failure, and repeat biopsy results. There is now increasing evidence that, of these, the PSA assay is the most sensitive index of tumor biological activity. 5,10 Unlike surgery, where the presence of PSA activity after removal of the prostate is a reliable indicator of treatment failure, irradiated prostate epithelial cells continue to secrete measurable amounts of the antigen, albeit in small quantities. What the exact level should be is as yet uncertain, but several investigators believe the posttreatment level should decrease and remain at 1.0 ng/ml or less to validate biological cure. 9

OUTCOMES
Complications Early Complications Adverse effects up to 12 months after the implant include obstructive and irritative symptoms. They occur to some extent in most patients, with the severity of symptoms often related to the degree to which such symptoms were experienced before treatment. The symptoms subside over a few weeks to a few months. Should obstructive symptoms persist, it is best to manage the patient by intermittent catheterization for at least three half-lives of the isotope—6 months for iodine and 2 months for palladium—before intervening surgically. A prostate incision at the 6-o'clock position (TUIP) is often all that is required. Transurethral resection is associated with an increased risk of incontinence in patients who received a uniform-distribution (Quimby) implant and is best avoided. Peripherally weighted (Patterson–Parker) implants minimize the urethral dose and do not appear, at least in a short-term follow-up study, to be associated with any increase in complications in patients with transurethral resections. Late Complications In one series, 320 patients were implanted by the uniform source distribution method (Quimby) and followed for 7 years. Two of the 320 (0.6%) required urinary diversion for severe urethral stricture and incontinence, and another 25 (8%) required minor office procedures such as cystoscopy, urethral dilation, and sigmoidoscopy. Erectile dysfunction did not present itself in patients below 60 years of age but occurred in 20% of patients between the ages of 60 and 70 years. Another series reported on 71 patients implanted with the peripherally weighted (Patterson–Parker) implant technique and followed for a mean of only 2 years. Investigators noted only mild radiation proctitis in 4.2% and urinary retention in 5.6%. During this time period, only 6% of the patients experienced erectile difficulties. Results Three hundred twenty T1/T2 patients were treated with iodine-125 or palladium-103 as monotherapy at Northwest Hospital between January 1987 and June 1993. Median presenting PSA was 6.7 ng/ml (range 0.2 to 74.6), and the Gleason pattern scores at diagnosis, available in 313 patients, were: 2 to 4 (130), 5 and 6 (161), and 7 to 10 (22). Median follow-up was 50 months (range 24 to 97 months). No patients were pathologically staged, and none was treated with hormones. The 7-year actuarial local control rate was 97%, the distal disease-free control rate 95%, and the PSA progression-free status 83%. In patients with no clinical/pathologic evidence of prostate cancer, PSA progression is defined as two consecutive increases in serum PSA from a minimum inflection (nadir) regardless of absolute PSA levels. In this series of patients, PSA progression—however minimal—preceded all local and distal clinical failures. Posttreatment biopsy was obtained in 192 of 320 (60%). Biopsy results and corresponding PSA levels are: negative, 83% (median PSA 0.3); indeterminate, 13% (median PSA 0.6); positive, 4% (median PSA 3.0). Biopsy specimens were interpreted as indeterminate based on the presence of residual neoplastic cells exhibiting severe radiation effect. The viability of such cells is uncertain; however, experience with this pathologic category suggests that the majority of these biopsies convert to negative with time. A subset of 112 patients in this cohort has been followed for a minimum of 5 years. The PSA progression-free rate (non-Kaplan–Meier) is 105 of 112 (94%); the PSA £ 1.0 ng/ml rate is 101 of 112 (90%); and the PSA £ 0.5 ng/ml rate is 95 of 112 (85%). Eighteen patients agreed to repeat needle biopsy 5 years or more beyond treatment, three of whom were biopsied because of PSA progression. Negative biopsy was found in 16 of 18 (89%), one patient (5.5%) was graded indeterminate, and one patient (5.5%) failed locally with a positive biopsy. These results are encouraging, but randomized trials to support the therapeutic advantages of prostate brachytherapy are lacking. Thus, the definitive curative potential of brachytherapy must await longer-term follow-up. Brachytherapy does, however, offer some distinct advantages over the common treatment modalities. It is a one-time, low-morbidity, and cost-effective outpatient procedure that is less socially disruptive than either surgery or external beam treatment. The utilization of transrectal ultrasound and template guidance systems has greatly improved radioisotope implantation of the prostate. These improvements allow for high radiation doses to be safely delivered to the target tissue in an accurate and reproducible manner. The curative results with brachytherapy, at least at intermediate durations of follow-up, are comparable to those achieved with radical surgery and external beam irradiation. Current success and the prospects for further improvement argue that, at least for the immediate future, incorporating brachytherapy in the treatment armamentarium will allow the urologist to play a wider

9

role in providing care for patients with localized prostate cancer. CHAPTER REFERENCES
1. Blasko JC, Ragde H, Grimm PD. Transperineal ultrasound guided implantation of the prostate: Morbidity and complication. Scand J Urol Nephrol 1993;137:113–118. 2. Blasko JC, Ragde H, Schumacher D. Transperineal percutaneous iodine-125 implantation for prostate carcinoma using transrectal ultrasound and template guidance. Endocuriether Hypertherm Oncol 1987;3:131–139. 3. Epstein JI, Carmichael MJ, Pizov G, Walsh PC. Influence of capsular penetration on progression following radical prostatectomy: A study of 196 cases with long-term follow-up. J Urol 1993;150:135–141. 4. Epstein JI, Walsh PC, Carmichael M, Brendler CB. Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA 1994;271:368–374. 5. Frazier HA, Robertson JE, Humphrey PA, Paulson DF. Is prostate specific antigen of clinical importance in evaluating outcome after radical prostatectomy? J Urol 1993;1499:516–518. 6. Kuban DA, el-Mahdi AM, Schellhammer PF. I-125 interstitial implantation for prostate cancer. What we have learned 10 years later? Cancer 1989;63:2415–2420. 7. Morton JD, Peschel RE. Iodine-125 implants versus external beam therapy for stages A2, B, and C prostate cancer. Int J Radiat Oncol Biol Phys 1988;14(6):1153. 8. Partin AW, Yoo J, Carter HB, Pearson JD, Epstein JI, Walsh PC. The use of prostate specific antigen, clinical stage, and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol 1993;150:110–114. 9. Stone NN, Stock RG. Brachytherapy for prostate cancer: Real-time three-dimensional interactive seed implantation. Tech Urol 1995;12:72–80. 10. Zietman AL, Coen JJ, Shipley WU, et al. Radical radiation therapy in the management of prostatic adenocarcinoma: the initial prostate specific antigen value as a predictor of treatment outcome. J Urol 1994;151:640–645.

Chapter 36 Prostatic Ultrasound and Needle Biopsy Glenn’s Urologic Surgery

Chapter 36 Prostatic Ultrasound and Needle Biopsy
Johan Braeckman and Louis J. Denis

J. Braeckman: Department of Urology, Vrije Universiteit Brussels, 1090 Brussels, Belgium. L. J. Denis: Department of Urology, Oncology Centre Antwerp, 2020 Antwerp, Belgium.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Precautions Preparations Biopsy Technique Outcomes Complications Results Chapter References

Prostate cancer is now the most common malignancy in men, both in the United States and in Europe. Approximately 97% of all prostate cancers are adenocarcinoma. Other histologic types include neuroendocrine tumors, sarcoma, carcinoid tumors, melanoma, and metastases to the prostate. Both the morbidity and mortality from prostate cancer are increasing. Whether this is a true finding or just a result of a longer life expectancy and improved technology for early diagnosis is still debated.

DIAGNOSIS
Until the development of transrectal ultrasound (TRUS) of the prostate and of detection methods for prostate-specific antigen (PSA), the digital rectal examination (DRE) was the primary tool in the clinical diagnosis of prostate cancer. Likewise, the histologic diagnosis, indispensable for further management of the disease, has evolved from “blind” finger-guided punctures with a Tru-Cut needle to a procedure of refined echo-guided tissue sampling. The almost simultaneous introduction of a spring-driven “biopsy gun” (Biopty) was an invaluable adjunct to this technique.

INDICATIONS FOR SURGERY
Obtaining a histologic diagnosis of prostate cancer should be the goal whenever this opens therapeutic prospects for patients, but only then. In other words, the suspicion of prostatic malignancy in a man with a limited life expectancy for any reason other then prostate cancer should not automatically lead to any prostatic biopsy procedure. Assuming a reasonable life expectancy, however, the indications for TRUS are either an elevated PSA (actual or age adjusted) or abnormal DRE with any level of PSA.2 Transrectal ultrasound-guided biopsy is also appropriate in an older man with a limited life expectancy who has obstructive or irritative voiding symptoms. Such a patient may benefit from palliative treatment of a prostate cancer. In addition to securing the diagnosis, TRUS can be used to assist in the collection of directed biopsies from selected areas in and around the prostate, which will assist in staging the disease.

ALTERNATIVE THERAPY
Among the available and generally accepted methods to collect prostate tissue for histologic diagnosis, the transrectal ultrasound-guided biopsy with the “biopsy gun” has become the most popular one over the last years. Blind, digitally guided prostatic biopsies can be done with the same biopsy gun, but they are less accurate, especially for the diagnosis of small, nonpalpable lesions. 5 Transperineal echo-guided biopsies are too laborious and thus reserved for patients in whom the rectal access is impossible. In selected patients the diagnosis is made, sometimes unexpectedly, on histologic examination of tissue material obtained after transurethral prostatic resection.

SURGICAL TECHNIQUE
Although transrectal ultrasound-guided biopsy of the prostate is an outpatient procedure, some precautionary measures and preparations are recommended in order to optimize its accuracy and to minimize complications. Precautions Whenever a urinary tract infection might be present, a urinary sample for cytobacteriologic examination should be collected, and eventual infections should be treated adequately. Patients with acquired or iatrogenic coagulation disorders should be, when judged reasonable, medically accommodated with styptics or by temporarily interrupting their anticoagulation treatment. Though infrequently necessary, it is occasionally appropriate to provide some tranquilizing or analgesic drug to very anxious patients. Preparations Preoperative cleansing enemas the evening before and the morning of the procedure empty the rectal ampulla of gas and feces, optimizing the visualization. Antibiotic prophylaxis is given empirically, generally a fluoroquinolone the morning of the procedure, followed by a full dose of the fluoroquinolone for 3 days. Biopsy Technique The patient is placed on a comfortable examination table, in left lateral decubitus position, knees flexed toward the chest. The examiner takes his or her place on a stool on the right side of the patient and explores the rectal ampulla and the posterior aspect of the prostate with an index finger. The gloved and lubricated finger palpates for gross abnormalities and prepares the patient's anus for the introduction of the ultrasound probe ( Fig. 36-1).

FIG. 36-1. Positioning of patient.

The ultrasound equipment should generate high-frequency (7 to 10 MHz) multiplane images of the prostate. If the probe works with an external needle guide, the guide should be fixed to the probe before its introduction. Internal needle guides can be added with the probe in position. Some probes have an optional small balloon on top, to be filled with water, for better visualization. Even though this balloon will probably be pierced by the biopsy needle, its use is recommended for optimal initial examination of the prostate and seminal vesicles. A condom, filled with some jelly inside, is pulled over the distal part of the probe. After lubrication of the top of this condom, the probe is ready for introduction. Depending on the other clinical findings (DRE, PSA, previous TRUS), the prostate is first scrutinized for abnormalities. Taking into account possible rebiopsies in case of a negative result, it is good policy to note the DRE and TRUS findings at this time. Especially of interest are the volume of the prostate, the presence of hypoechoic lesions, asymmetries, and discontinuities of the prostate boundary echo. When the biopsies return with a positive result, these data will certainly be helpful in the further management of the disease (Fig. 36-2).

FIG. 36-2. Diagram of prostate and data form: transverse and saggital section of prostate with a suspicious lesion on the right side.

An 18-gauge Tru-Cut biopsy needle is placed in the mechanical actuating device (Biopty or alternative system). Before this device is introduced through the hollow needle guide, the ultrarapid discharge of the gun is checked to ensure the functioning of the system and to accustom the patient to this surprising sound ( Fig. 36-3).

FIG. 36-3. The spring-driven biopsy gun can be introduced through an internal or external puncture guide.

The targets in the prostate are now localized by moving the probe in the rectum and switching between the axial and sagittal planes. A little trick to prevent downsliding of the needle along the capsule of the prostate at the time of the biopsy itself is to turn the oblique face of the top of the needle away from the patient ( Fig. 36-4).

FIG. 36-4. Orientation of needle.

If the only goal is to know whether an ultrasonically hypoechoic lesion is malignant or not, two or three directed tissue samples of this lesion should be enough. The lesion is brought into view in the sagittal plane, meaning that it is crossed by the electronic puncture line, and the needle is advanced in the guide until its hyperechoic tip is clearly seen at the edge of the lesion. A gentle warning is addressed to the patient, and the needle is fired off into the lesion. One can easily judge the accuracy of the biopsy, as it is done in real time, and the moving needle creates a very strong reflection along its biopsy course. Even after an apparently perfectly directed biopsy, it is always recommended to take at least one extra core. In a patient who might be a candidate for curative treatment, it is also necessary to collect at least three tissue cores from the contralateral side (see sextant biopsies) (Fig. 36-5).

FIG. 36-5. Directed biopsies.

If an indurated prostatic nodule is palpated but not recognized on TRUS, geographic biopsies are indicated. The site of the suspicious lesion, i.e., left or right side, apex to base, is noted, and the biopsies are guided in that region of the prostate. Three additional biopsies from the contralateral side are again the standard ( Fig. 36-6).

FIG. 36-6. Geographic biopsies.

In patients with suspicion of prostate cancer (elevated PSA) but without palpable or echogenic abnormalities, we recommend the performance of “sextant biopsies” (Fig. 36-7).3 These biopsies are directed to the lateral margins of the peripheral zone of the prostate, where the majority of cancers originate. On each side three biopsies are taken: one at the apex, one at the base, and one in between. Here, the puncture needle should be advanced until its tip lies just in front of the prostatic capsule. That way one is sure to obtain representative tissue cores of 15 mm. If the “sextant biopsies” return negative from the pathology department and cancer suspicion persists, one can either repeat the procedure or extend it into the transectional zone of the prostate. We do not perform these biopsies routinely as they are far more invasive and painful.

FIG. 36-7. Sextant biopsies.

OUTCOMES
Complications In most cases the postbiopsy period is free of complications, but the patient is always told of possible adverse events. There might be some rectal bleeding, which is usually taken care of with a clean absorbing cloth. Exceptionally an internal rolled lubricated bandage is placed and can be safely removed at home after a couple of hours. Hematuria resulting from a urethral injury can occur, even several days after the procedure. Hemospermia, a benign and predictable event after a prostate biopsy, can frighten the patient, even weeks later. The patient is always warned of the possibility of high fever or chills and is asked to seek immediate medical help in this event. Again, this is an exceptional complication, certainly after antibiotic prophylaxis. Results In Cooner's 1990 study of the role of transrectal ultrasound, his overall detection rate was 14% for all patients undergoing biopsy. 1 He performed a biopsy only on those patients with a visible hypoechoic lesion. The detection rate increased with increasing PSA: 4.5% of patients with a PSA < 4.0 ng/ml had a hypoechoic lesion with a positive biopsy; 17% of patients with a PSA > 4.0 ng/ml but < 10.0 ng/ml had a positive biopsy; 53% of patients with a PSA > 10 ng/ml had a positive biopsy. Cooner's detection rate for transrectal ultrasound-guided biopsy in patients with a PSA > 4.0 ng/ml was 33%. Vallancien reported a detection rate of 26% with systematic sextant biopsies in men with a PSA > 4.0 ng/ml and a normal DRE.6 Smith reported a detection rate of 29% at the initial evaluation of men with PSA > 4.0 ng/ml and an overall detection rate of 45% when these same men were followed longitudinally and evaluated with repeat biopsies. 4 CHAPTER REFERENCES
1. Cooner WH, Mosley BR, Rutherford CL Jr, et al. Prostate cancer detection in a clinical urological practice by ultrasonography, digital rectal examination and prostate specific antigen. J Urol 1990;143:1146. 2. Lee F. Transrectal ultrasound: diagnosis and staging of prostatic carcinoma. Urology 1989;33:5–10. 3. Peller P, Young C, Marrnaduke D, Marsh W, Badalament R. Sextant prostate biopsies. Cancer 1995;75:530–538. 4. Smith DS, Catalona WJ, Herschman JD. Longitudinal screening for prostate cancer with prostate-specific antigen. JAMA 1996;276:1309. 5. Torp-Pedersen ST, Lee F. Transrectal biopsy of the prostate guided by transrectal ultrasound. Urol Clin North Am 1989;16:703–712. 6. Vallancien G, Prapotnich D, Veillon B, Brisset JM, Andre-Bougaran J. Systematic prostatic biopsies in 100 men with no suspicion of prostate cancer on digital rectal examination. J Urol 1991;146:1308.

Chapter 37 Stamey and Gittes Bladder Neck Suspension Glenn’s Urologic Surgery

Chapter 37 Stamey and Gittes Bladder Neck Suspension
David A. Ginsberg, Eric S. Rovner, and Shlomo Raz

D. A. Ginsberg: Department of Urology, University of Southern California/Norris Cancer Center, Los Angeles, California 90033. E. S. Rovner: Division of Urology, Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania 19104. S. Raz: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90024.

Stamey Bladder Neck Suspension Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Gittes No-Incision Urethropexy Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Chapter References

STAMEY BLADDER NECK SUSPENSION
Suspension of the bladder neck via a vaginal approach was initially described by Peyrera in 1959. Contemporary techniques of transvaginal bladder neck suspension have arisen as modifications of Peyrera's description. The endoscopic needle suspension of Stamey, described in 1973, contributed several concepts to the surgical technique of bladder neck suspension. This procedure was the first to utilize the cystoscope to precisely place sutures at the bladder neck and visualize closure of the bladder neck with elevation of the suspension sutures. In addition, the procedure incorporates a knitted dacron graft as a bolster to buttress either side of the urethra and aid in the prevention of suture pull-out. Diagnosis The complete evaluation and work-up for urinary incontinence is described in Chapter 38. Indications for Surgery The technique described by Stamey is indicated for correction of stress incontinence in the absence of a significant cystocele. We currently have abandoned simple bladder neck suspensions and perform vaginal wall slings for patients with stress incontinence and no significant cystocele. The choice between these techniques depends on the surgeon's training and experience with the different procedures. Alternative Therapy Alternatives to needle suspensions of the bladder for stress urinary incontinence include transabdominal suspensions, vaginal wall slings, fascial slings, periurethral injections of collagen or Teflon, and conservative measures such as pessaries, pelvic floor stimulation, behavior modification, biofeedback, a-agonist therapy, and urinary collection devices including pads or diapers. Surgical Technique Positioning and Retraction The patient is placed in the dorsal lithotomy position and prepped and draped in the standard fashion. A posterior weighted vaginal speculum and silk labial retraction sutures are placed to aid in exposure. A Foley catheter is placed, and the bladder is drained. Exposure of Bladder Neck A T-shaped incision is made in the anterior vaginal wall. The dissection is carried down to the glistening periurethral fascia and continues laterally until the surgeon is able to palpate the balloon of the catheter. This identifies the bladder neck and allows adequate exposure for later placements of the dacron pledgets. Needle Passage Two suprapubic stab wound incisions are made on each side of the lower abdomen, and the anterior rectus fascia is exposed. The single-pronged Stamey needle is then inserted into the medial edge of one of the suprapubic wounds and advanced, under fingertip control, into the vaginal incision ( Fig. 37-1). The needle passes through the rectus fascia, adjacent to the periosteum, alongside the bladder neck, and through the periurethral fascia as it traverses from the abdomen to the vagina. The Foley catheter is removed, and cystoscopy is performed to confirm correct positioning of the needle. An appropriately positioned needle, when moved medially, will indent the ipsilateral bladder neck ( Fig. 37-2). If the needle penetrates the bladder, it should be removed and repassed.

FIG. 37-1. The needle is guided along the posterior surface of the pubic symphysis under fingertip control to avoid injury to the bladder. (From Shortliffe LMD, Stamey TA. In: McDougal WS, ed. Operative urology. Kent, England: Butterworth, 1985.)

FIG. 37-2. The cystoscope is placed distal to the urethrovesical junction to evaluate needle positioning. An appropriately positioned needle will indent the ipsilateral bladder neck when moved medially (upper inset). If the needle penetrates the bladder ( lower inset), the needle is removed and repassed. (From Shortliffe LMD, Stamey TA. In: McDougal WS, ed. Operative urology. Kent, England: Butterworth, 1985.)

Suture Transfer and Dacron Graft One end of a #2 nylon suture is threaded through the needle and transferred suprapubically. The Stamey needle is passed a second time, 1 cm lateral to the first pass, and its position is again cystoscopically confirmed. The vaginal end of the nylon suture is threaded through a 10- by 5-mm dacron arterial graft, and the free vaginal end of the nylon suture is then placed in the needle holder and transferred suprapubically ( Fig. 37-3). During the transfer of this nylon, an Allis clamp may be used to visually maneuver the dacron graft into appropriate position at the urethrovesical junction as the Stamey needle is pulled suprapubically. The periurethral tissues are now suspended on one side of the bladder neck, and the procedure is repeated on the contralateral side.

FIG. 37-3. Stamey needle is passed a second time after the suture is passed through the dacron graft. (From Shortliffe LMD, Stamey TA. In: McDougal WS, ed. Operative urology. Kent, England: Butterworth, 1985.)

Cystoscopy and Closure Cystoscopy is performed to evaluate the placement of the needle sutures and to confirm adequate functional closure of the bladder neck with minimal tension placed on the nylon sutures. The vaginal incision and graft material are irrigated with an antibiotic solution and closed with a running, locking 2-0 polyglycolic acid suture. An antibiotic-impregnated vaginal pack is placed, and the suprapubic nylon sutures are tied, without tension, such that the knots rest against the rectus fascia. A suprapubic catheter is placed, and the suprapubic wounds are closed with a 4-0 polyglycolic acid suture following antibiotic irrigation. The vaginal packing may be removed 2 hours after surgery, and the patient may be discharged as early as 6 hours postoperatively. The suprapubic catheter is removed no earlier than 1 week after surgery once the postvoid residuals are less than 60 ml. Several important technical points have been outlined by Stamey. 9 The dacron graft should be positioned below the suture line to prevent graft erosion through the vaginal incision. Copious irrigation with an aminoglycoside solution should be performed before closing the vaginal incision to decrease the risk of dacron graft infection. The appropriate Stamey needle (0-, 15-, or 30-degree angle at the distal end containing the needle) should be used depending on the patient's anatomy.

OUTCOMES
Complications Complications particular to the Stamey needle suspension include erosion of suture and bolster material into the urinary tract, which can occur up to 7 years following the procedure. Stamey reports a 0.3% incidence of dacron buttress erosion into the bladder as well as a 0.3% incidence of failure of the vaginal incision to completely heal, resulting in an exposed piece of dacron. In both circumstances the exposed tube and suture were removed (endoscopically if the tube eroded into the bladder), and continence was maintained with the single remaining suture on the contralateral side. 9 Sutures are removed in 1% to 2% of patients for pain or infection, and long-term retention may be resolved by loosening of the nylon loop under local anesthesia. Results Evaluation of the literature to determine the success rates of this operation is difficult. The majority of the studies with an adequate number of patients obtained their data in a retrospective manner without anonymous questionnaires to the patients (thus possibly introducing bias to their results), the follow-up in most studies was short (mean follow-up often 24 months or less), and the definition of success was different from author to author (completely dry versus improved). These provisos should be remembered as the literature is reviewed. Review of the English literature in the past 5 years reports cure rates that range from 53% to 80%. 1,3 Walker and Texler evaluated patients with a mail-in questionnaire and found 82% of 192 respondents improved and 65% of patients willing to undergo the procedure again. 10 Early success rates with the Stamey bladder neck suspension may not be durable. O'Sullivan et al. reported a dry rate of 70% immediately after surgery in 67 patients, which decreased to 31% dry at 1 year (58 patients) and further decreased to 18% dry at 5 years (22 patients). 8 Mills et al. found the cure rate in 30 patients decreased from an initial 67% to 33% over a 10-year period of time.7 Factors that may place patients at increased risk for postoperative failure include obesity, respiratory disease, number of pads used per day, prior Marshall–Marchetti–Krantz procedure and concomitant abdominal hysterectomy. 8,10

GITTES NO-INCISION URETHROPEXY
The technique of Gittes and Loughlin was described in 1987. 2 Their simplified modification of the Peyrera needle suspension obviates the need for vaginal incisions (Fig. 37-4). This technique is based on the concept that as a monofilament suture pulls through the vaginal wall, it heals as an autologous pledget, creating an internal bolster that tethers the anterior vaginal wall and prevents rotational descent with Valsalva ( Fig. 37-5).

FIG. 37-4. Diagrammatic representation of just-tied no-incision urethropexy as monofilament sutures lift the vaginal wall under tension. (From Raz S, Little NA, Juma S. Female urology. In: Walsh PL, Retik AB, Stamey TA, Vaughan ED, Jr., eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992; 2782–2828.)

FIG. 37-5. Diagrammatic representation of no-incision urethropexy approximately 1 month after surgery. As the suture slowly cuts through the vaginal wall and fascia, a curtain of scar tissue is left and provides support to the vesical neck. (From Gittes RF. No-incision urethropexy. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996;328–332.)

Diagnosis The complete evaluation and workup for urinary incontinence is described in Chapter 41. Indications for Surgery The technique described by Gittes is indicated for correction of stress incontinence in the absence of a significant cystocele. We currently have abandoned simple bladder neck suspension and perform vaginal wall slings ( Chapter 39) for patients with stress incontinence and no significant cystocele. The choice between these techniques depends on the surgeon's training and experience with the different procedures. Alternative Therapy Alternatives to needle suspensions of the bladder for stress urinary incontinence include transabdominal suspensions, vaginal wall slings, periurethral injections of collagen or Teflon, artificial urethral sphincters, and conservative measures such as pessaries, pelvic floor stimulaion, and urinary collection devices including pads. Surgical Technique Positioning and Retraction The patient is placed in the dorsal lithotomy position and prepped and draped in the standard fashion. A posterior weighted vaginal speculum and silk labial retraction sutures are placed to aid in exposure. A Foley catheter is placed, and the bladder is drained. Needle Passage Two suprapubic stab wound incisions, approximately 5 cm lateral to the midline, are made on each side of the lower abdomen at the upper border of the symphysis pubis, and the anterior rectus fascia is exposed. The single-pronged Stamey needle is then inserted into the medial edge of one of the suprapubic wounds such that the tip of the needle scrapes the posterior aspect of the pubic bone. The anterior vaginal wall is identified just lateral to the Foley catheter balloon and simultaneously elevated with the surgeon's second hand. The needle is then directed, from above, toward the intravaginal fingertip. Once the needle tip is palpable by the intravaginal fingertip, the needle is advanced through the anterior vaginal wall and out through the introitus. The Foley catheter is removed, and cystoscopy is performed to confirm correct positioning of the needle. An appropriately positioned needle, when moved medially, will indent the ipsilateral bladder neck. If the needle penetrates the bladder, it should be removed and repassed. Suture Transfer One end of a #2 Proline suture is threaded through the needle, transferred suprapubically, and secured with a hemostat. The Stamey needle is passed a second time, 1 to 2 cm lateral to the first pass, to provide a base of strong fascial support for the suspension. The second pass of the needle should perforate the vaginal tissue approximately 1 cm lateral to the initial pass to avoid tenting up a large amount of vaginal tissue at the completion of the procedure. The position of the Stamey needle is again confirmed with cystoscopy. The free end of the Proline is threaded through a Mayo needle, and two or three helical bites of vaginal tissue are taken between the first and second vaginal perforation. The Mayo needle is then unthreaded, and the free end of the Proline suture is then advanced through the eye of the previously positioned Stamey needle. The needle is withdrawn, and the two ends of the suspension suture are secured with a hemostat for later tying. The periurethral tissue is now suspended on one side of the bladder neck. Needle passage and suture transfer are then repeated on the contralateral side. Cystoscopy and Closure Cystoscopy is performed to evaluate the placement of the needle and sutures and to confirm adequate functional closure of the bladder neck with minimal tension placed on the Proline sutures. An antibiotic-impregnated vaginal pack is placed, and the suprapubic Proline sutures are tied, without tension, such that the knots rest against the rectus fascia. A suprapubic catheter is placed, and the suprapubic wounds are closed with a 4-0 polyglycolic acid suture following antibiotic irrigation. The vaginal packing may be removed 2 hours after surgery, and the patient may be discharged as early as 6 hours postoperatively. The suprapubic catheter is removed no earlier than 1 week after surgery once the postvoid residuals are less than 60 ml. Outcomes Complications

An overall complication rate of 9.8% for the Gittes no-incision urethropexy has been reported. 6 Potential complications include prolonged urinary retention (2% to 7%), suprapubic pain or cellulitis, genitofemoral or ilioinguinal nerve entrapment, vaginitis, and suture infection with abscess formation, which could require the removal of a suspension suture (and possibly lead to recurrent stress urinary incontinence). 4,5 and 6 Results Reported success rates in the English literature for cure of stress incontinence with the Gittes no-incision urethropexy vary between 81% and 94% depending on the length of follow-up and definition of cure. 4,5 and 6 There are no adequately done studies that have evaluated the long-term efficacy of the Gittes no-incision urethropexy. Kursh evaluated factors influencing the outcome of this procedure and found a significantly decreased cure rate in postmenopausal women and in patients with a greater degree of incontinence preoperatively. 4 As expected, women with type 1 stress incontinence had better outcomes than patients with type 3 stress incontinence (97% versus 45% cure rate, respectively). 4 CHAPTER REFERENCES
1. Ashken MH. Follow-up results with the Stamey operation for stress urinary incontinence of urine. Br J Urol 1990;65:168. 2. Gittes RF, Loughlin KR. No incison pubovaginal suspension for stress incontinence. J Urol 1987;138:568. 3. Hilton P, Mayne CJ. The Stamey endoscopic bladder neck suspension: a clinical and urodynamic investigation, including actuarial follow-up over four years. Br J Obstet Gynaecol 1991;98:1141. 4. Kursh ED. Factors influencing the outcome of a no incision endoscopic urethropexy. Surg Gynecol Obstet 1992;175:254. 5. Kursh ED, Angeli AH, Resnick MI. Evolution of endoscopic urethropexy: Seven-year experience with various techniques. Urology 1991;37:428. 6. Loughlin KR, Whitmore WF III, Gittes RF, Richie JP. Review of an 8-year experience with modifications of endoscopic suspension of the bladder neck for female stress urinary incontinence. J Urol 1990;143:44. 7. Mills R, Persad R, Ahsken MH. Long-term follow-up results with the Stamey operation for stress incontinence of urine. Br J Urol 1996;77:86. 8. O'Sullivan DC, Chilton CP, Munson KW. Should Stamey colposuspension be our primary surgery for stress incontinence? Br J Urol 1995;75:457. 9. Stamey TA. Urinary Incontinence in the Female: The Stamey Endoscopic Suspension of the Vesical Neck for Stress Urinary Incontinence. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;2829–2850. 10. Walker GT, Texler JH Jr. Success and patient satisfaction following Stamey procedure for stress urinary incontinence. J Urol 1992;147:1521.

Chapter 38 Abdominal Approaches to Surgery for Female Incontinence Glenn’s Urologic Surgery

Chapter 38 Abdominal Approaches to Surgery for Female Incontinence
E. P. Arnold and Peter Gilling

E. P. Arnold: Department of Urology, Christchurch Hospital, Christchurch, New Zealand. P. Gilling: Promed House, Tauranga, New Zealand.

Diagnosis Indications for Surgery Principles of Surgery Alternative Therapy Surgical Technique: Retropubic Cystourethropexies Burch Colposuspension Marshall-Marchetti-Krantz Procedure Suburethral Slings Laparoscopic Burch Colposuspension Outcomes Complications Results Chapter References

Urinary incontinence in women is common. Its incidence rises after vaginal delivery, apparently as a result of nerve stretch, which can also occur after prolonged straining from constipation. Denervation can lead to incompetence of urethral and anal sphincters and to prolapse of the pelvic floor. Incontinence and prolapse are common and often, but not always, coexist. It is clear that for incontinence to occur, the urethral sphincter must be deficient whether or not a prolapse is present. In some patients with prolapse and bladder base descent, however, rotational compression of the urethra from the prolapse stops any leakage. Following repair of such prolapse, these patients might then begin leaking when the underlying intrinsic sphincter deficiency is unmasked. The second major cause of urinary incontinence is detrusor instability, a diagnosis based on urodynamic studies and referring to the occurrence of detrusor contractions during the filling phase or provoked by coughing, posture change, etc. It is often associated with symptoms of frequency, nocturia, urgency, and urge incontinence. The presence of detrusor instability preoperatively has been demonstrated by some authors to have a negative prognostic influence on the outcome of surgery. The incidence of detrusor instability detected by continuous ambulatory monitoring is higher than that found on standard cystometry to the extent that some have considered it to be almost normal. In any case, where there are high-pressure overactive bladder contractions, these patients may well require a reduction in their bladder pressures by anticholinergic medication or, in severe cases, by augmentation cystoplasty to ensure adequate bladder capacity, particularly if there is a neurologic cause.

DIAGNOSIS
The diagnosis of urinary incontinence can be made by clinical history and examination of the patient with a full bladder, observing the effects of coughing and straining. The severity of urinary incontinence can be graded ( Table 38-1).

TABLE 38-1. Severity of urinary incontinence

Urinary leakage can be quantified by standardized pad-weighing tests or by a bladder diary and can be further assessed by the performance of pressure-flow studies, preferably with video screening facilities to demonstrate bladder base and urethral hypermobility or rotational descent during coughing and straining. This will also outline any fistula or urethral diverticulum. The type of urinary incontinence can be classified on the basis of radiologic findings and the degree of bladder base descent (anatomic incontinence) or whether the urethra appeared wide open as a “stove pipe” (intrinsic sphincter deficiency, ISD) in an attempt to rationalize the type of procedure to be advised. Measurements of urethral pressures during coughing and straining and measurements of the pressure transmission ratio or assessment of the abdominal leak-point pressure during a Valsalva maneuver have been found useful by some clinicians. Low preoperative maximum urethral closure pressure (MUCP) is more likely to result in a postoperative failure to stop the leakage. 1 The abdominal leak-point pressure (ALPP) correlates with the degree of incontinence from the history and from pad-weighing tests; however, the pelvic floor is a complex mechanism, and precisely what happens to it during straining and coughing requires further study. The presence of detrusor instability is not easy to predict from the history of frequency, nocturia, urgency, and urge incontinence or from examination of the patient, and one must rely on urodynamic studies.

INDICATIONS FOR SURGERY
Essentially, if conservative measures have not succeeded, patients who have genuine stress incontinence and who are bothered by the symptom and who are fit for surgery should be offered it. 3 The significance of detrusor instability detected preoperatively in predicting the outcome of surgery is controversial. Many have shown that preoperatively detected detrusor instability compromises the results of surgery. 10 Long-term ambulatory monitoring studies show a higher incidence of detrusor instability than conventional cystometry, and the detrusor instability correlated with an increased surgical failure. 2 Many patients with detrusor instability lose their urgency and frequency symptoms postoperatively, though preoperative detrusor instability may persist in 15% to 89% of cases (mean approximately 50%). 10 Principles of Surgery Continuous Compression of Urethra

Artificial urethral sphincters and injectable bulking agents produce a continuous compression of the urethra to stop leakage. Submucosal injections of bovine collagen, fat, silicone macroparticles, and other bulking agents have been used to produce coaptation of urethral walls and to increase urethral resistance to leakage; they seem to work best for ISD where there is minimal rotation of the bladder base. Intermittent Compression Exactly how other procedures work remains somewhat conjectural but is probably on the same principle of having a backboard behind the urethra that will prevent mobility during coughing and improve the pressure transmission ratio (PTR) of abdominal pressure to the urethra. 9 There is no intrinsic change in the maximum urethral closure pressure postoperatively; hence, nothing is done to improve the efficiency of the intrinsic sphincter mechanism. These operations produce a suburethral “sling”: Of vaginal wall and periurethral tissues and endopelvic fascia in situ, as in the Stamey, Gittes, Burch, Marshall–Marchetti–Krantz (MMK), and Raz–Peryera operations Of the vaginal wall as a buried strip as in the Raz vaginal sling Using a strip of rectus fascia or fascia lata Of synthetic materials The principal difference among these procedures is whether the “sling” is sutured to a fixed point, as in the iliopectineal ligament of the Burch procedure and as in the symphysis pubis in the MMK procedure, or to the mobile rectus sheath, as in the pubovaginal sling, Stamey, and Raz operations. The theoretical advantage of being fixed to the mobile rectus sheath is that it will allow for a loose sling at rest that tightens when the abdominal pressure rises and hence resists leakage when it is needed without obstructing voiding.

ALTERNATIVE THERAPY
Conservative therapies including pelvic floor exercises, vaginal cones, and pelvic floor electrical stimulation achieve a success rate that varies between 30% and 70%.3 This improvement may avoid the need for surgery in some patients. It is usually worth trying as a first line of treatment. Success is less likely in the patient who has had previous surgery or radiotherapy and in those with grade 2 or 3 severity of leakage. Hormone replacement therapy may help frequency, urgency, and burning, but its use for incontinence has been disappointing.

SURGICAL TECHNIQUE: RETROPUBIC CYSTOURETHROPEXIES
Burch Colposuspension Informed consent should follow only after a full explanation of the procedure ( Fig. 38-1), length of time of hospitalization and recovery, expected outcomes, and possible complications, as discussed below. Prophylactic antibiotics are administered. My preference is to use a cephalosporin with a spectrum against anaerobic organisms.

FIG. 38-1. Burch colposuspension. (A) Modified Lloyd–Davies position with hips abducted and minimally flexed and the patient slightly head-down. Pfannensteil incision. (B) Initial exposure of the retropubic space, with bladder swept off the back of the symphysis pubis and superior pubic rami, demarcating the iliopectineal ligament. (C) The vagina is tented up by the operator's second and third fingers placed vaginally, and using diathermy forceps the endopelvic fascia is buttonholed alongside the bladder neck and bladder base. (D) Sutures are placed on one side and then tied before being placed on the other. No attempt is made to approximate the vaginal wall to Cooper's ligament.

The fully anesthetized patient is placed in the modified Lloyd Davies position with hips abducted and minimally flexed. This allows access for preliminary cystoscopy if indicated and for bimanual manipulation intraoperatively. Skin preparation of the abdomen and vagina is performed using Betadine. A 16-Fr urethral catheter with 7 ml instilled into the balloon is placed on free drainage. Through a Pfannensteil incision 1 cm above the symphysis pubis, the rectus fascia is incised transversely, and the recti are separated. The bladder is swept off the back of the symphysis pubis and the iliopectineal ligament, exposing it and the obturator internus and levator ani muscles. Care is required to avoid damage to the abnormal obturator veins, if present. This dissection is easy in a first approach but can be difficult if adhesions are present following previous surgery. Care is required to avoid perforating a thin-walled bladder adherent to the bones. Keeping close to the bone helps to avoid this. Some surgeons advocate intravesical installation of methylene blue to detect any transgression into the lumen. With the two fingers of the operator's left hand in the vagina, the lateral fornix is tented up, and the endopelvic fascia overlying the vaginal fornix, well lateral to the bladder, is buttonholed; this allows the vagina and its plexus of veins to be mobilized medially. The dissection should start well laterally to avoid the ureter and the bladder. Blunt dissection with the Riches diathermy forceps develops the appropriate plane lateral to the bladder neck as defined by palpation of the Foley catheter balloon, and this exposes the pale white tissue of the vagina and allows any venous bleeding to be cauterized without changing instruments. Mobilization is done on both sides. A row of usually three nonabsorbable sutures (size 1 nylon) are placed full thickness of vagina up through the iliopectineal ligament. These are then tied before sutures are placed on the opposite side to enable correct positioning. It is important not to make the suspension too tight, and it is not necessary to approximate the vagina to the iliopectineal ligament. Often the sutures “bow-string,” particularly after previous vaginal repair or vaginal hysterectomy, as of course they always do in any of the needle suspension procedures. Where an enterocele is present, this can be surgically managed with a Muskowitz procedure in which the pouch of Douglas is obliterated intraperitoneally by encircling sutures at serial levels. In the prevention of this complication of the Burch procedure, some have advised plication of the round ligament or, if a hysterectomy has been previously undertaken, of tightening the uterosacral ligaments. A suction drain is left in the retropubic space, and a suprapubic catheter is inserted to enable measurements of postvoid residuals postoperatively ( Fig. 38-2). After balloon inflation, the catheter should be withdrawn against the vault of the bladder to avoid its tip irritating the trigone and causing postoperative urgency. 12 The rectus muscles are approximated with interrupted 2-0 synthetic absorbable sutures without tension, and the rectus fascia is closed transversely with continuous 0 suture.

FIG. 38-2. Placing suprapubic catheter. (A) With Allis tissue forceps, grasp both walls of the bladder onto the balloon of the urethral catheter to ensure avoiding the peritoneum. (B) Puncture both walls of the bladder with mosquito forceps. (C) Grasp the Foley catheter tip (14 Ch). (D) Pull catheter into the bladder and direct it downward. (E) Withdraw the forceps and close the puncture wound; inflate the catheter balloon. Pull back the catheter so that the balloon abuts on the inner vault of the bladder. 12

Postoperatively the suction drain is removed, usually by day 2; the suprapubic catheter is clamped, and a trial of voiding is commenced. It is removed when the residuals are less than 100 ml or less than one-third of the bladder capacity. Patients are discharged by day 5, and by then most are voiding satisfactorily; if there is any doubt, the patient is taught intermittent self-catheterization until voiding efficiency is restored. Marshall-Marchetti-Krantz Procedure This procedure (Fig. 38-3), described in 1949, involves retropubic dissection through a Pfannensteil incision as for the Burch procedure, and the edges of the urethra are sutured to the fibrocartilage of the symphysis pubis. Serially, the row of sutures is continued onto the anterior surface of the bladder in the perception that this would alleviate prolapse and descent of the bladder base. There is some risk of damage to the urethra as the sutures pass through its lateral wall, and most surgeons have modified the original procedure to take in the periurethral tissue rather than the edges of the urethra. There are also technical difficulties in some patients because of the very thin periosteum through which to place the sutures.

FIG. 38-3. Marshall–Marchetti–Krantz procedure. (A) Endopelvic fascia incised alongside the upper urethra and bladder neck. Space opened by blunt dissection using the diathermy forceps and developing the plane by counterpressure against the operator's second and third fingers placed within the vagina. Three sutures are placed on each side of the urethra through the endopelvic fascia and vaginal wall, but not including the urethral wall, and the sutures are then passed through the cartilage of the symphysis pubis. If the sutures are placed in the cartilage too medially, then this can compress and obstruct the urethra. (B) Lateral parasagittal view of symphysis showing approximation of the urethra and anterior bladder wall to the back of the symphysis pubis.

Suburethral Slings Patients with rotational descent and adequate vaginal capacity are suitable for either a Burch colposuspension or a pubovaginal sling, but if the vaginal capacity is reduced, a pubovaginal sling may be more appropriate. These retropubic procedures use a fascial strip from the rectus fascia as a suburethral sling with each end sutured to the rectus sheath (Fig. 38-4). This technique has been largely superseded by a vaginal approach to avoid the retropubic dissection and is described in Chapter 39 and Chapter 40. In the abdominal approach for this procedure, a strip of rectus fascia 1.5 cm wide is taken from the upper margin of the Pfannensteil incision. The bladder is mobilized, and a tunnel is developed between the vagina and bladder neck and upper urethra, using the balloon of the catheter as a guide. The sling is passed through this tunnel, each end of the strip is sutured with 1 nylon, and the sutures at each end are brought out and tied anterior to the rectus, 1 to 2 cm lateral to the midline, and low down, 1 cm above the symphysis. The length of the strip is probably not critical to the procedure, but it would seem reasonable to use a strip approximately 5 to 10 cm in length. The nylon sutures can be passed up to the appropriate site using the Stamey needle. It is important to ensure that the sling is not too tight, to avoid urethral compression.

FIG. 38-4. Suburethral sling: abdominal approach. (A) A 1.5-cm strip of anterior rectus sheath harvested from the line of the suprapubic Pfannenstiel incision, 5 to 10 cm in length. (B) Endopelvic fascia incised lateral to the proximal urethra and bladder neck. (C) Sutures at one end of the fascial strip grasped and pulled through the suburethral tunnel. (D) Stamey needle puncture of the anterior rectus sheath and one suture drawn through. A second pass with the Stamey needle is made 1 cm from the first in a vertical plane to draw through the second suture. The two sutures are then tied, fixing the one end. The process is repeated on the opposite side, ensuring minimal tension on the sling.

Laparoscopic Burch Colposuspension The patient is admitted in the morning of surgery, having had nothing by mouth from midnight of the night before. Below-knee compression stockings are fitted, and a shave is performed to just below the upper border of the pubic symphysis (midpubic shave). No bowel preparation is necessary for this procedure. A suitable premedication is given, and prophylactic antibiotics are given with the induction of anesthesia as a single dose.

A general anesthetic is preferred for complete relaxation. Pneumatic calf stimulators are placed once the patient is asleep. The low lithotomy position is employed with slight hip flexion and abduction. The lower abdomen and vagina are scrubbed and prepared in the usual manner, and a Foley catheter is placed on free drainage. Surgical drapes are placed to leave the abdomen below the umbilicus and perineum exposed. The surgeon stands on the patient's left, and the assistant and a scrub nurse on the patient's right. A Mayo table is placed between the patient's legs, and this contains the trocars and instruments necessary during the procedure. Two video monitors are generally employed, one each for the surgeon and assistant, at the foot of the bed. The instrumentation required is listed in Table 38-2.

TABLE 38-2. Instrumentation

The procedure (Fig. 38-5) commences with a 12-mm incision, which is placed midway between the umbilicus and the pubic symphysis. It is extended down to the fascia, which is cleared away by finger dissection. The fascia is penetrated with a fine arterial forceps, and the surgeon's index finger is used to clear the extraperitoneal space so that the pubic symphysis can be clearly felt. It is important to be careful not to puncture the peritoneum during this maneuver. Two stay-sutures are placed to the fascial opening; then the retroperitoneal dilating balloon on a catheter introducer is passed into the space created, and 600 ml of saline is placed in the balloon. This remains inflated for several minutes while the remainder of the laparoscopic instrumentation is set up, including the video camera. The balloon is then deflated, and the Hasson cannula placed and secured with stay-sutures. The space is then inspected with the laparoscope. The room is then darkened, and the light source turned up to the maximum. The remaining two trocars are placed in the left lower quadrant. It is important to avoid the vessels that are transilluminated during this maneuver. The 5-mm trocar is placed 2 cm above the symphysis at the lateral border of the left rectus abdominis muscle. The 10-mm trocar is placed between the two previous trocars avoiding obvious blood vessels.

FIG. 38-5. Laparoscopic Burch colposuspension. (A) Site of placement of port. (B) The percutaneous suture passer is placed via side punctures into the retropubic space and thence under laparoscopic vision through the iliopectineal ligament. One end of the suture is then passed through its eye, the passer is withdrawn out of the ligament, and its end is then delivered into the retropubic space and tied. (C) Laparoscopic procedure at completion. 4

With the assistant's finger in the vagina elevating the tissue adjacent to the bladder neck, which can easily be identified using the balloon of the Foley catheter as a guide, the endopelvic fascia is cleared of its fatty tissue. The grasping forceps are in the surgeon's left hand, and the endosurgical scissors in his or her right. Once this is completed, the needle holder is used to introduce the long length of 2-0 Ticron suture down the 10-mm port. The needle is grasped at an appropriate angle, and a large bite of this tissue adjacent to the bladder neck is obtained. The suture needle is then cut off. The percutaneous suture passer is then passed under direct vision into the extraperitoneal space just above Cooper's ligament. It is then placed through the ligament. The end of the suture is passed through the eye of this needle and drawn up through the ligament. Then, the suture material is freed from the suture passer, and this is withdrawn. The free end of the suture, which now passes through both the vaginal tissue and Cooper's ligament, is then grasped with the needle holder and drawn out through the 10-mm port. An extraperitoneal knot is then tied, and this is passed down to be placed snugly on Cooper's ligament. The suture material is then cut, and two further intracorporeal knots are tied to secure this extracorporeal knot. The assistant's finger elevates the tissues throughout this maneuver. Further fixation can be obtained with a 15-cm length of 2-0 Ticron suture on a needle that is passed through both Cooper's ligament and the vaginal tissues with the knots being tied intracorporeally. The procedure is then repeated on the patient's left side in an identical manner. An alternative to this technique involves a 15-cm length of suture, which is prepared by tying a small loop at the free end. The suture is then passed through the 10-mm port with a needle holder and then through the tissue at the level of the bladder neck. The needle is then passed through this loop at the end of the suture, and it is snugged down onto the tissues. A further bite of paravaginal tissue can then be taken to form a helical suture, and the needle is then passed through Cooper's ligament. A Lapra-Ty clip can then be placed onto the suture just as it exits the ligament. A second pass through Cooper's ligament and a second Lapra-Ty clip can also be placed for extra security. The patient is able to eat and drink once she recovers from the effects of the anesthetic. Analgesic and antiemetics are given as required, though opiates are usually unnecessary. The Foley catheter remains in the bladder on free drainage until 6 o'clock the next morning, at which time it is removed, and if the patient is able to void comfortably on several occasions following this, she is discharged from the hospital. Rarely, intermittent self-catheterization will be required for the first few days if voiding is not satisfactory. Residual urine volumes are not routinely measured. No further antibiotic prophylaxis is employed.

OUTCOMES
Complications Urethral obstruction after successful surgery occurs in about 10% of cases. 5 To prevent it, care should be taken to avoid elevating the bladder neck too high or tying sutures with undue tension. Intermittent self-catheterization is best used in these cases, and generally, with time, voiding dysfunction usually resolves within about a week. It may persist longer. If dysfunction persists for more than 3 to 6 months, the sutures can be taken down, and, following cystolysis, a vagino-obturator shelf can be considered.11 Postoperative voiding difficulties may be predicted by preoperative voiding problems. However, some acontractile bladders at preoperative study can be shown to contract postoperatively. Those with low preoperative voiding pressures (less than 15 cm of water) were more likely to have problems. 7 Measurement of isometric pressure at the clinical stop test was not found to be helpful in predicting postoperative voiding dysfunction. De novo appearance of detrusor instability postoperatively has been observed in 10% to 20% of cases. Why this should occur is unclear, but ambulatory continuous monitoring studies show a higher preoperative incidence of detrusor instability, and perhaps the surgery unmasks this by changing afferent inputs from the pelvic

floor. In the patient with urgency and urge incontinence postoperatively, one should exclude infection and obstruction. Urodynamic studies should be performed to determine the presence of instability. Cystoscopy should be undertaken where the problem persists, as occasionally a stitch will erode into the bladder. In management, if instability is the only problem, then anticholinergics are often helpful. In the rare cases where this does not suffice and instability remains a problem, then a clam cystoplasty can be considered. Dyspareunia following pelvic floor surgery is not uncommon and is either downplayed or not included in the subjective symptomtology in most series. Where the vagina was narrow or shortened before the incontinence surgery, dyspareunia may occur as a result of the posterior ridge that mirrors the anterior suspension of the anterior vaginal wall. Usually this will settle in time. The question of whether vaginal delivery should be allowed after a successful incontinence surgery is often stated, but the evidence is lacking. This has led to most patients who have had a successful outcome from surgery being advised by their obstetrician to consider a cesarean section rather than risk breaking it down. The magnitude of this risk has never been properly substantiated. After the Burch procedure and needle suspension operations, pain occurs in around 10% and is usually caused by the tension of the sutures. Mostly it decreases with time, but occasionally it persists, and the sutures have to be removed. Another rarer cause of pain is osteitis pubis. Results There is a discrepancy between the results as told us by the patients and those obtained from objective tests of clinical examination, pad-weighing tests, or video pressure-flow studies postoperatively. Often it is not clear in the literature how the results are in fact assessed. In a collected series only 4,815 of 20,481 (23.5%) had had their results of surgery objectively assessed. 5 Results are probably best assessed by a health professional independent of the surgeon, as patients often unwittingly confound the results as stated. They often want to please the surgeon, the one whom they chose and whose advice they accepted. They might feel guilty about hurting the surgeon's feelings or fearful that if they complain of continuing problems, the surgeon might want to do all that again, or even something worse. Patients often enhance the results, and although they may indeed be dry, that might be at the expense of stopping jogging, aerobics, or golf, etc. These factors make comparison of outcomes from different procedures quite difficult. There is not a perfect operation for female urinary incontinence, but the results in achieving continence are quite good at 5 years. The results tend to diminish in time, although the Burch operation and pubovaginal slings tend to hold up as well as most others ( Table 38-3).5,7

TABLE 38-3. Success of urinary incontinence procedures

Patients with persistent or recurrent urinary incontinence should be thoroughly reassessed. This includes history, examination, bladder diary, pad-weighing tests, and urodynamic studies to detect any instability, fistula, etc. If genuine stress incontinence is the reason for failure, then the operation can be repeated. Or, if the urethra is “stove-pipe” and fixed, then there are the options of injecting a suburethral bulking agent such as collagen or of repeating the colposuspension or vaginal sling. In gross cases, some have advocated a reduction vesicourethroplasty. In some cases the artificial urethral sphincter can be considered. Causes of failure include incorrect placement of sutures too high or too low in relation to the bladder neck, sutures cutting out, atrophy of the periurethral and vaginal tissues contributing to the sling, or sphincter damage as might occur in the Marshall–Marchetti–Krantz procedure, where the sutures might have been placed in the wall of the urethra. The urethra may be held open by scar tissue following the first procedure, and a full urethrolysis needs to be considered at the second procedure to avoid this. A history of previous hysterectomy; obesity, but excluding morbid obesity; parity; or age had no apparent influence on the outcome, but results were slightly better in those who had no past history of prior incontinence surgery. 6 CHAPTER REFERENCES
1. Bown LW, Sand PK, Ostergard DR, Franti CE. Unsuccessful Burch retropubic urethropexy: A case-controlled urodynamic study. Am J Obstet Gynecol 1989;160:452–458. 2. Eckford SD, Bailey RA, Jackson SR, Shepherd AM, Abrams P. Occult pre-operative detrusor instability: an adverse prognostic feature in genuine stress incontinence surgery. Neurourol Urodyn 1995;14:487–488. 3. Fantl JA, Newman DK, Colling J, et al. Urinary incontinence in adults: acute and chronic management. Clinical practice guidelines no. 2, AHCPR publication no. 96-0682. Rockville, MD: Agency for Health Care Policy and Research, 1996. 4. Gilling PG, Fraundorfer MR, Sealey C, et al. Laparoscopic extraperitoneal approaches to female urinary incontinence: the colposuspension and pubovaginal sling. J Urol 1994;153:344A. 5. Jarvis GJ. Surgery for genuine stress incontinence: Review. Br J Obstet Gynaecol 1994;101:371–374. 6. Kiilholma P, Makinen J, Chancellor MB, Pitkanen Y, Hirvonen T. Modified Burch colposuspension for stress urinary incontinence in females. Surg Gynecol Obstet 1993;176:111–115. 7. Lose G, Jorgenson L, Mortenson SO, Molsted-Pedersen L, Kristensen JK. Voiding diffilculties after colposuspension. Obstet Gynecol 1978;69:33–37. 8. McDougall E. Correction of stress urinary incontinence: retropubic approach. J Endourol 1996;10:247. 9. Theofrastous JP, Bump RC, Elser DM, Wyman JF, McClish DK. Correlation of urodynamic measures of urethral resistance with clinical measures of incontinence severity in women with pure genuine stress incontinence. Am J Obstet Gynecol 1995;173:407–414. 10. Vierhout ME, Mulder AFP. Persistent detrusor instabilty after surgery for concomitant stress incontinence and detrusor instability: a review. Int Urogynecol J 1993;4:237–239. 11. Webster GD, Kreder KJ. Voiding dysfunction following cystourethropexy: its evaluation and measurement. J Urol 1990;144:670–673. 12. Woo HH, Rosario DJ, Chapple CR. A simple technique for the intra-operation placement of a suprapubic catheter. Br J Urol 1996;77:153–154.

Chapter 39 Anterior Vaginal Wall Sling Glenn’s Urologic Surgery

Chapter 39 Anterior Vaginal Wall Sling
Lynn Stothers

L. Stothers: Department of Surgery, St. Paul's Hospital, Vancouver, British Columbia V52 4E3, Canada.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preparation Suprapubic Cystostomy Anterior Vaginal Wall Sling Outcomes Complications Results Chapter References

The Raz anterior vaginal wall sling is a relatively new surgical procedure developed since 1992 at the University of California Los Angeles Medical Center. It creates a sling for the treatment of intrinsic sphincter dysfunction (ISD) or anatomic (hypermobility-related) incontinence without burying vaginal epithelium or using autologous fascial strips. Permanent sutures are placed into the periurethral supporting tissues to create a hammock of support from the naturally occurring anatomic structures adjacent to the bladder neck and urethra. To achieve improvement in continence status, two main surgical goals must be met: (a) provide support and increased coaptation to the urethra and (b) create a strong hammock of vaginal wall and underlying tissues to provide a backboard of support to the midurethra and bladder neck during times of increased intra-abdominal pressure.

DIAGNOSIS
The presence of stress urinary incontinence should be confirmed objectively by the surgeon with physical examination, cystoscopy, and urodynamics. Epidemiologic studies have shown that 1% of patients who complain objectively of stress urinary incontinence actually have an alternative diagnosis such as profuse vaginal discharge or endocervical cysts, which may mimic incontinence. If incontinence can not be demonstrated objectively by the routine evaluation, then a pyridium pad test may be helpful in confirming the diagnosis. Before surgery, administration of a self-directed incontinence-specific quality-of-life score such as the Raz Quality-of-Life Score, the Incontinence Impact Questionnaire, or the Urogenital Distress Inventory will help to quantify the degree of clinical significance urinary incontinence is having on an individual patient. 2,5 Once incontinence is demonstrated, it is classified as intrinsic sphincteric dysfunction (ISD) or anatomic incontinence (AI). Because no pathognomonic test exists for ISD, the diagnosis is made by a combination of historical, physical, and urodynamic parameters. Factors typically associated with ISD include multiple prior surgeries, prior radiation exposure, direct urethral trauma, or neurologic dysfunction. The abdominal leak-point pressure is typically low. In contrast, patients with AI have hypermobility of the bladder neck and urethra, and the abdominal leak-point pressure is found to be higher than that in those with ISD. In addition to those patients who fit classically into either the ISD or AI categories, there exists a group of patients who may exhibit characteristics of both etiologies with an abdominal leak-point pressure in the gray zone between these two categories. Although patients with a broad range of intra-abdominal leak-point pressures have been treated with the anterior vaginal wall sling, we still continue to classify patients as ISD or AI types to aid in preoperative counseling. Patients with ISD may take longer to resolve postoperative urinary retention and have a lower incidence of resolution of preoperative instability symptoms than do patients with AI. 4

INDICATIONS FOR SURGERY
This procedure is indicated for clinically significant female stress urinary incontinence secondary to bladder neck hypermobility or intrinsic sphincter dysfunction with little (grade 1) or no cystocele. Like any vaginal surgery, it is contraindicated in patients who can not adequately be placed in lithotomy position because of physical restrictions such as limited hip abduction. Modifications of this procedure can be made to treat incontinence associated with grade 2 or 3 cystocele (six-corner suspension). Grade 4 cystocele requires a more extensive procedure in which both central and lateral defects are corrected as described in Chapter 42, Chapter 43, Chapter 44 and Chapter 55.

ALTERNATIVE THERAPY
In contrast to the Raz vaginal wall sling, which uses permanent sutures placed in the periurethral supporting tissues, a sling can be created from a variety of alternative materials. These can be classified as endogenous sources (fascia lata, rectus fascia, or harvested strips of anterior vaginal wall) or exogenous sources. The latter are either synthetics such as Marlex mesh or natural sources such as banked human dura. In addition to sling procedures, ISD may be treated surgically with injectable agents (collagen, fat, Teflon, silicon) or the artificial urinary sphincter. Alternative surgical treatments for anatomic or hypermobility-related incontinence are classified by the surgical approach—either abdominal, such as the Burch suspension, or vaginal, such as the Gittes or Stamey procedure. The Raz bladder neck suspension, which was previously used to treat anatomic incontinence, has now been replaced by the anterior vaginal wall sling at our institution.

SURGICAL TECHNIQUE
Preparation In preparation for surgery, the patient should be given an oral stool softener to begin on the day before surgery. Broad-spectrum intravenous antibiotics such as an aminoglycoside plus a cephalosporin or a broad-spectrum DNA gyrase inhibitor are administered preoperatively. Following general or spinal anesthesia, the patient is placed in the lithotomy position with the buttocks just overhanging the edge of the operating table. This will allow the weighted vaginal speculum to hang freely without contact against the operating table. The feet should be adequately padded in protective boots and placed in leg-supporting stirrups. If the patient is obese, slight Trendelenburg positioning of the table will help to expose the suprapubic region. The skin is then painted with an iodine-based solution from the level of the umbilicus inferiorly to include the whole perineum. The vagina is also painted with an iodine-based solution. A speculum or long forceps should be used to aid in the skin preparation of the vagina to ensure that the entire vagina is adequately prepared to avoid bacterial contamination of the suture material. The drapes are placed to expose the suprapubic region and perineum, with the anus carefully excluded from the field. Several 3-0 silk sutures are used to anchor the drape covering the anus to ensure that it does not fall away during the procedure, thereby contaminating the field. To obtain maximum exposure, the 30-degree weighted vaginal speculum is placed in the vagina, following which single 3-0 silk labial retraction sutures are placed into the labia minora. Appropriate placement of these two retraction sutures will greatly aid in the exposure. Be sure they anchor the labia both laterally and superiorly to exposure the urethra and bladder neck region of the anterior vaginal wall. Always apply retraction sutures after the vaginal speculum is in place to avoid unnecessary tension or suture pull-out. Suprapubic Cystostomy A 14-Fr Foley catheter is used for suprapubic drainage. To place the suprapubic catheter, the closed curved Lowsley forcep is placed into the bladder by the surgeon and pressed against the anterior vaginal wall 2 cm above the symphysis pubis in the midline. The assistant feels for the tip of the forceps and makes a puncture wound with the scalpel blade cutting the skin and rectus fascia. The operator pushes the tip of the forceps through the wound, and the assistant positions the 14-Fr

Foley catheter into the jaws of the retractor. Do not lubricate the tip of the catheter or curved Lowsley retractor to ensure that the catheter does not slip out of the jaws of the retractor. Withdrawal of the retractor by the surgeon delivers the tip of the catheter out the urethra. A small forceps is used to hold the tip of the catheter inside the bladder while the assistant inflates the balloon with 10 cc of water and irrigates with 50 cc of normal saline to ensure correct positioning within the bladder. The suprapubic catheter is placed on traction, and the bladder is emptied with the suction and clamped off. A second 14-Fr Foley catheter with 10 cc of water in the balloon is placed per urethra and clamped off. Palpation of the balloon against the bladder neck is helpful in identifying this landmark vaginally. The assistant places three Allis clamps at the level of the midurethra (midway between the bladder neck and external meatus) on the anterior vaginal wall and retracts upward, exposing the anterior vaginal wall for the surgeon. Anterior Vaginal Wall Sling Before the vaginal incisions are made, 10 cc of saline is injected just beneath the vaginal wall along the anticipated suture lines to facilitate dissection. Two oblique incisions are made in the anterior vaginal wall, extending from the level of the midurethra to 2 cm below the bladder neck ( Fig. 39-1). Dissection is carried out laterally using the Metzenbaum scissors to expose the vaginal side of the urethropelvic ligament bilaterally. This dissection should be superficial. Deep dissection with perforation of the ligament can result in excess bleeding. The attachment of the urethropelvic ligament to the tendinous arc can be felt by the operator by placing a finger into the incision pointing toward the ipsilateral shoulder of the patient ( Fig. 39-2A). With gentle pressure, the curved Mayo scissors are placed into each wound against the tendinous arc and advanced until the retropubic space is entered. Opening the blades of the scissors helps to detach the urethropelvic ligament from the tendinous arc. The operator can now place a finger into the wound and feel the open retropubic space. Blunt finger dissection is used to detach any adhesions within the retropubic space from both sides. The space should feel freely open with the finger, and one should be able to palpate the urethra easily in the midline ( Fig. 39-2B).

FIG. 39-1. The patient is in the lithotomy position with the weighted vaginal speculum in place along with labial retraction sutures. The positions of the two oblique incisions in the anterior vaginal wall are shown by the dotted lines. Note that the incisions do not cross or meet in the midline.

FIG. 39-2. (A) Dissection is carried out over the glistening periurethral fascia. The curved Mayo scissors are shown entering the retropubic space. The scissors are pointed toward the shoulder of the patient. Misdirecting the scissors too far medially could result in bladder perforation. (B) The surgeon's finger is inserted into the open retropubic space through each vaginal incision, and adhesions are taken down. The inside of the pubic ramus is easily palpated. In patients with prior surgery, a Deaver retractor may be placed in the retropubic space, and sharp dissection under vision can be used to safely incise any dense urethral adhesions from the pubic bone.

Two pairs of #1 Proline suture on a half-circle tapered MO-5 needle are used to complete the sling. As each suture is passed to the surgeon, its free end is held with a small mosquito forceps, which can rest on the patient's abdomen while the surgeon completes suture placement. This keeps the free end of the suture well within the sterile field and prevents potential contamination. Begin with placement of the proximal pair of Proline sutures, which are similar to those used in the traditional Raz bladder neck suspension. A long forceps is placed into the retropubic space, and the urethra and bladder are retracted medially. A #1 Proline suture is placed in a helical fashion into the urethropelvic ligament, taking several passes. Then, with the needle kept parallel to the plane of the vagina, the suture is passed in the vaginal wall (excluding the epithelium) to incorporate a large surface area of the underlying vesicopelvic fascia. A similar procedure is carried out on the contralateral side ( Fig. 39-3).

FIG. 39-3. The first pair of Proline sutures are placed at the level of the bladder neck. On the patient's left, correct positioning of the left proximal Proline suture is shown incorporating several helical passes of the urethropelvic ligament and the underlying vesicopelvic fascia. This suture is placed in an identical fashion to the Raz bladder neck suspension.

To place the second, more distal pair of Proline sutures, the long forceps is placed into the open retropubic space. Opening the jaws of the forceps parallel to the floor and retracting inferiorly will create an open triangle in the retropubic space. At the apex of this triangle is the levator muscle as it inserts into the pubic symphysis and the midurethral segment. The urethropelvic ligament in the medial vaginal wall forms the lateral border of the triangle. The floor of the triangle is parallel to the cardinal ligaments.

Using a #1 Proline suture, incorporate several passes of the levator muscle and the edge of the urethropelvic ligament. In order to obtain an adequate amount of levator tissue, the needle must be placed deep into the retropubic space. The levator should be visualized on the arc of the needle. Reposition the forceps to put downward traction on the anterior vaginal wall in the area of the midurethra and incorporate several helical bites of the underlying periurethral fascia incorporating tissue up to but not crossing the midline. As in placing sutures into the vesicopelvic fascia it is important to keep the needle parallel with the vaginal wall to prevent suture material from entering the spongy tissue of the urethra itself. After all four Proline sutures are in place, one can visualize a rectangle of support for the bladder neck and midurethra (Fig. 39-4).

FIG. 39-4. The second more distal pair of Proline sutures is placed at the level of the midurethra. These sutures include the levator muscle, the edge of the urethropelvic ligament, and the underlying periurethral fascia. Once the four Proline sutures are in place, one can visualize the rectangle of support that will be given to the underlying urethra and bladder neck.

A clean blade is used to make a puncture wound the width of the double-pronged needle carrier in the midline, just above the superior margin of the pubic bone. If the incision is made too high, the sutures will be transferred over a mobile area of the anterior abdominal fascia, which can result in pain or incomplete support. The incision is carried down to, but not through, the rectus fascia. With a finger in the retropubic space serving as a guide, the double-pronged needle carrier is advanced through the suprapubic incision, the retropubic space, and out through the vaginal incision ( Fig. 39-5). As the needle passes into the retropubic space, it should hug the symphysis in the midline to ensure that the needle passes into the thick and less mobile area of the rectus fascia. The freed ends of one of the ipsilateral Proline sutures is placed through the needle holes in the double-pronged ligature carrier. Retraction of the needle carrier delivers both ends of the suture out through the suprapubic incision. A total of four passes are made, each suture being transferred individually. Do not attempt to transfer more than one suture at a time—this can result in tangling or knotting of the Proline sutures.

FIG. 39-5. Each Proline suture is transferred individually using the double-pronged needle carrier through a midline suprapubic stab wound the width of the double-pronged needle. Note that passage of the needle is done under fingertip control, passing the needle from the suprapubic region to the vaginal area as close to the pubic symphysis as possible. This maneuver is repeated for a total of four passes, transferring one Proline suture at a time to the suprapubic region.

Indigo carmine is injected intravenously, and cystoscopy is performed with 30- and 70-degree lenses. This ensures that (a) the suprapubic tube is in good position, (b) blue efflux is noted from both ureteral orifices, (c) no Proline suture material has entered the bladder, and (d) upward retraction on the suprapubic Proline sutures provides support to the bladder neck and midurethra. The urethral catheter is replaced, and the assistant provides upward retraction with the three Allis clamps, once again exposing the anterior vaginal wall. The two oblique incisions are closed with a running, locking absorbable suture of 2-0 polyglycolic acid on a tapered UR-5 needle. The shape of this needle allows better placement of sutures high in the vagina. A vaginal pack laden with antibiotic cream is inserted into the vagina, following which the weighted vaginal speculum is removed. Last, the Proline sutures are tied independently to their ipsilateral mates over the rectus fascia, creating the hammock of support ( Fig. 39-6). Excessive tension in the suspending sutures may lead to prolonged pain and is not necessary to achieve support. The skin edges of the suprapubic wound are freed, and the Proline knots are buried. Failure to adequately free the skin edges can result in dimpling of the skin over the Proline knots, which can cause patient discomfort. The suprapubic skin wound is closed with intradermic 4-0 absorbable sutures and Steri-strips.

FIG. 39-6. Lateral view of the pelvis demonstrating all four Proline sutures in place and tied in the midline; a hammock is created to provide support to the midurethra and bladder neck. A vaginal pack is shown in the vagina, and the suprapubic catheter is exiting the suprapubic region in the midline 2 cm superior to the Proline sutures.

Within 24 hours the vaginal pack and urethral catheter are removed. The suprapubic catheter is plugged, and the patient begins to record her voided volumes and the postvoid residual. The patient is discharged with an oral stool softener, an oral antibiotic, and analgesics. When the residual urine is consistently low, the suprapubic catheter is removed in the office.

OUTCOMES
Complications A list of the potential complications related to the vaginal wall sling are listed in Table 39-1. The majority are preventable by (a) proper positioning of the patient, (b) careful dissection and identification of the important anatomic landmarks, and (c) routine performance of intraoperative cystoscopy for early identification and correction of potential problems. Patients at increased risk for complications include those with a history of prior bladder or pelvic surgery, endometriosis, pelvic infection, pelvic fracture, or significant pelvic prolapse. These factors may alter the typical pelvic anatomy. Although a detailed review of female pelvic anatomy is beyond the scope of this chapter, several excellent resources are available. 1,3

TABLE 39-1. Classification of the potential complications related to the anterior vaginal wall sling

Intraoperatively, minor bleeding most commonly results from dissection in the wrong fascial plane, such as perforation of the urethropelvic ligament rather than dissection over its glistening surface during exposure. Bleeding may also occur following opening of the retropubic space if entrance is made too close to the urethra with subsequent injury to the periurethral vessels. Temporary placement of a pack into the retropubic space will facilitate exposure and provide hemostasis until a suture ligature can be applied. Temporary packing is always preferable to the excessive use of electrocautery on the delicate tissues of the urethra and bladder. Rarely, arterial vessels are found within the vaginal wall and are encountered shortly after the vaginal wall incisions are made. Should this occur, an Allis clamp placed over the edge of the vaginal wall will secure hemostasis until a suture ligature is applied. Postoperatively, vaginal spotting of blood may be noticed by the patient within the first two postoperative weeks. If vaginal bleeding continues or is increasing, perform a vaginal examination and place a temporary vaginal pack. Misplaced Proline sutures occur when the anatomic landmarks are not clearly exposed and identified. This can potentially result in ureteral or bladder perforation or injury. Both of these complications may be diagnosed with the use of intraoperative indigo carmine and careful cystoscopy. If suture material is identified within the bladder, immediately remove the offending suture and ensure that the ureters are intact, performing retrograde pyelograms if necessary. Ureteral injuries may require stenting for minor injuries or reimplantation for complete ligations. Perforation of the bladder during dissection is exceedingly rare and should be repaired immediately intraoperatively by a multiple layered vaginal closure adhering to the principles of vesicovaginal fistula repair. P>Postoperative suprapubic pain may be related to suspension sutures and is often activity related, subsiding with several weeks of decreased physical activity. It is often idiopathic in nature but may also be caused by cellulitis, subcutaneous abscess formation, muscle entrapment, vigorous overtying of sutures, or placement of sutures through a mobile portion of the anterior abdominal wall. Transferring of sutures in such a site may create tension over the rectus muscle during activity. Passing the double-pronged needle in the midline, as close to the pubic bone as possible, and not tying sutures under tension will minimize this complication. Vaginal stenosis or shortening may result from excessive plication of the vaginal epithelium during closure or secondary scarring. When the vaginal wall is closed, the running suture is locked to aid in the prevention of excessive plication of the tissue. A history of new onset of dyspareunia or pelvic or vaginal pain and the finding of foreshortening on physical examination confirm the diagnosis. Mild shortening or stenosis may be treated with longitudinal relaxing incisions in the lateral vaginal wall with transverse closure. Voiding dysfunction in the early postoperative period is common following surgery for stress urinary incontinence. Before surgery, the patient should be informed of common potential changes in her voiding pattern, including temporary retention and mild bladder irritability. The majority of patients with the new onset of voiding dysfunction have resolution of their symptoms within a short period of time; however, persistent voiding dysfunction should be reevaluated with physical examination, cystoscopy, and videourodynamics. Urinary retention in the immediate postoperative period resolves in the majority of patients with the use of a suprapubic catheter or intermittent self-catheterization. Permanent retention has not been reported as a complication of this procedure. Persistent or recurrent stress incontinence requires complete urodynamic evaluation and usually reoperation if it is of a severity to affect the patient's quality of life. In this case, reoperation with proper suture placement should correct the problem. When the cause is recurrent intrinsic sphincteric dysfunction, alternative corrective measures including injection of urethral bulking agents or artificial urinary sphincters may be considered. Pain, redness, and swelling in the suprapubic area should alert the surgeon to a potential infection at the site of suspension suture knots, which may require drainage should antibiotic therapy be unsuccessful. Lower urinary tract infections are common in the first month following any vaginal surgery and generally respond to a short course of oral antibiotics. Results In 1995 the clinical outcome of the sling procedure was reported in 163 women ranging in age from 31 to 81 years. 4 The cohort was followed prospectively with a median follow-up of 17 months (range 6 to 32 months). Three patients were lost to follow-up. Of the 160, 95 had intrinsic sphincter dysfunction (ISD), and 65 anatomic incontinence (AI). One hundred fifty-two patients were considered cured of stress urinary incontinence at last follow-up. 4 Eleven of 163 patients were considered failures and had recurrent incontinence that was unrelated to bladder instability and required further therapy. Time to recurrent stress incontinence comparing ISD and AI patients, as modeled using Kaplan–Meier survival curves and the log-rank test, showed no significant difference between patients with preoperative anatomic incontinence and those with intrinsic sphincter dysfunction ( p > 0.05). Conditional logistic regression covariates revealed no significant predictive factors for postoperative failures. Seven percent of patients developed de novo instability. Pre- and postoperative within-patient changes of quality-of-life scores were found to be statistically significantly improved for both AI and ISD patients. 6 These initial results indicate that excellent continence was achieved in patients with both ISD and AI using the anterior vaginal wall sling. Advantages of this technique include the absence of a laparotomy incision, short hospital stay, lateral placement of permanent nonreactive sutures, and the ability to correct mild cystocele. The procedure has been shown to be safe and effective and allows for outpatient surgical management of bothintrinsic sphincter dysfunction and anatomic incontinence. CHAPTER REFERENCES
1. Raz S. The anatomy of pelvic support and stress incontinence. In: Raz S, ed. Atlas of transvaginal surgery. Philadelphia: WB Saunders, 1992;1–21. 2. Raz S, Erickson DR. SEAPI QMM incontinence classification system. Neurourol Urodyn 1992;1:187–199. 3. Raz S, Stothers L, Chopra A. Vaginal reconstructive surgery for incontinence and prolapse. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology, 7th ed. Philadelphia: WB Saunders, 1998;1059–1094. 4. Raz S, Stothers L, Young G, et al. Vaginal wall sling for anatomic incontinence and intrinsic sphincter damage-efficacy and outcome analysis. J Urol 1996;156:166–170. 5. Schumaker SA, Wyman SF, Uebersax JS, et al. Health related quality of life measures for women with urinary incontinence; the Incontinence Impact Questionnaire and the Urogenital Distress

Inventory. Quality Life Res 1994;3:291–306. 6. Stothers L, Raz S, Chopra A. Anterior vaginal wall sling. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996;395–398.

Chapter 40 Pubovaginal Fascial Slings Glenn’s Urologic Surgery

Chapter 40 Pubovaginal Fascial Slings
R. Duane Cespedes and Edward J. McGuire

R. D. Cespedes: Department of Urology/MKCU, Wilford Hall Medical Center, Lackland AFB, Texas 78236. E. J. McGuire: Division of Urology, University of Texas, Houston, Texas 77030. The opinions contained herein are those of the authors and are not to be construed as reflecting the views of the Air Force or the Department of Defense.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

The first urethral sling procedure was described by Von Giordano in 1907. 9 However, even after numerous technical improvements and application of many different materials, the pubovaginal sling (PVS) was rarely used until repopularized by McGuire and Lytton in 1978. 2 The pubovaginal sling has traditionally been used only when other incontinence procedures such as a bladder neck suspension or retropubic urethropexy have failed. In this regard, patients with type 3 stress urinary incontinence, also called intrinsic sphincter deficiency (ISD), have often been diagnosed by default. More recently, the preoperative diagnosis of ISD has been facilitated by use of the Valsalva or abdominal leak point pressure (ALPP) during incontinence evaluations. 3 Accordingly, the diagnosis of ISD can be made before surgery and a PVS performed as the primary incontinence procedure. Stress urinary incontinence in females is classified by the presence and degree of urethral mobility and functional status of the urethra. In types I and II stress urinary incontinence, the urethral sphincter functions normally; however, abdominal pressure can drive the sphincter to a position where it doesn't function normally. Stress incontinence due to urethral hypermobility can be successfully treated by a procedure that immobilizes it, such as a retropubic urethropexy or needle suspension procedure. Type III stress urinary tract incontinence, or ISD, is usually characterized by a minimally mobile urethra and incompetence of the urethral sphincter during increases in abdominal pressure. A few patients have incontinence due to coexisting ISD and urethral hypermobility. All patients with ISD are effectively treated with a PVS.

DIAGNOSIS
The preoperative evaluation is directed to identifying ISD. The history can be helpful because patients with ISD usually have severe leakage with minimal activity or have a history of irradiation to the pelvis, a prior incontinence procedure, or are elderly (especially new onset in patients over 70 years old). The incidence of ISD increases after each failed incontinence procedure: 9% if no previous surgery, 25% after one failed procedure, and 75% after two failed procedures. 4 The physical exam is directed to demonstrating leakage, urethral hypermobility, and pelvic prolapse. Urinary leakage without significant hypermobility constitutes presumptive evidence of ISD. A careful evaluation for associated cystocele, rectocele, enterocele, and uterine prolapse is important for ALPP interpretation and in planning the appropriate operative procedures. Failure to repair associated pelvic prolapse conditions will put undue stress on any incontinence procedure, including a pubovaginal sling, which increases the failure rate. After the postvoid residual is determined, a cystometrogram is performed to exclude poor detrusor compliance and overt detrusor instability. To diagnose ISD, an ALPP is indispensable. The bladder is filled to a standard volume of 200 ml (children to one-half functional bladder capacity) and a slow Valsalva maneuver is performed with the patient in the upright position until leakage is noted. Performing this several times and determining an average improves accuracy. If a well-performed Valsalva maneuver fails to induce leakage, vigorous coughing may be required. If the ALPP is below 60 cm H 2O, then ISD is present. If the ALPP is greater than 90 cm H2O and minimal pelvic prolapse exists, pure urethral hypermobility is usually present. Patients with a significant pelvic prolapse condition may have a falsely elevated ALPP and reduction with a vaginal pack is helpful. ALPP values between 60 to 90 cm H 2O form a gray area in which hypermobility and ISD usually coexist. 3

INDICATIONS FOR SURGERY
The most common indications for a PVS are urodynamically documented ISD with or without urethral hypermobility and a prior failed incontinence procedure. Additionally, because of the long-term success and durability of a pubovaginal sling, certain patients with stress urinary incontinence due to urethral hypermobility may be better served with a sling procedure. These include females who engage in vigorous athletic activities, are significantly obese, or who cough frequently due to pulmonary disease.

ALTERNATIVE THERAPY
In selected female patients with ISD and minimal urethral hypermobility, collagen can be injected at the bladder neck with a success rate of 63% using a mean of 9.1 ml and 1.5 treatments.6 The vaginal wall sling introduced by Raz uses the in situ vaginal wall as the sling with a reported 93% short-term cure rate in patients with ISD.8

SURGICAL TECHNIQUE
Patients with atrophic vaginitis should be treated with topical estrogens for 2 weeks before the procedure. It is helpful to teach the patient clean intermittent catheterization before the procedure since incomplete emptying is common for a few days postoperatively. One dose of intravenous antibiotics should be given preoperatively. General or regional anesthesia may be used without particular advantage to either technique. The procedure is performed in the low lithotomy position using Allen stirrups with feet squarely in the stirrups to avoid pressure on the calf areas. The legs should only be moderately flexed at the hips to allow simultaneous exposure to the vagina and the lower abdomen. A 16-Fr Foley catheter is placed and the balloon inflated with 5 mls to allow palpation of the bladder neck and urethra. A weighted vaginal speculum is placed. The labia may be sewn laterally if the view is obstructed. The rectus fascia is usually harvested first to minimize vaginal bleeding. In adults, an 8- to 10-cm Pfannenstiel incision is made approximately 2 to 3 cm above the pubis (Fig. 40-1). The subcutaneous tissue is cleared from the rectus fascia and a relatively scar-free area is selected. Even the most scarred and thickened rectus fascia is usually suitable as a sling. Incising parallel to the fibers, obtain a fascial sling approximately 8 to 10 cm in length with a center portion 1.5 to 2.0 cm wide, tapering the ends to 1 cm wide (Fig. 40-1 inset). Free the upper and lower fascial leaf from the rectus muscles superiorly and inferiorly for approximately 4 to 5 cm to allow a tension-free fascia closure. The sling sutures may be placed before or after transection. The size and type of suture used is a matter of personal preference but we currently use 1-0 polyglactin absorbable suture, which decreases postoperative suture pain and does not compromise durability. The sutures are placed perpendicular to the direction of the fibers approximately 0.5 cm from the ends incorporating all of the fibers in the bites.

FIG. 40-1. Harvesting of the rectus abdominis fascial sling. The sling is 8 to 10 cm long and 1.5 to 2.0 cm wide in the center, narrowing to 1 cm at the ends. The sling ends are sutured with 1-0 absorbable suture incorporating all of the fibers.

The vaginal procedure begins by placing an Allis clamp midway between the bladder neck and the urethral meatus with traction placed superiorly. It is very important to maintain this traction throughout the vaginal procedure. Injectable saline is infiltrated beneath the vaginal epithelium over the proximal urethra to facilitate the dissection. A 3-cm midline incision is made over the proximal urethra and the initial vaginal dissection is performed with a scalpel or Church scissors, which allows one to quickly find the proper plane superficial to the white periurethral fascia. Damage to the underlying urethra and bladder is minimized when dissection proceeds in this plane. The dissection is facilitated by maintaining outward traction (toward the operator) on the developing vaginal flap and by maintaining the tips of the scissors on this flap at all times. Carry the dissection laterally and enter the retropubic space inferior to the ischium, at the level of the bladder neck, by perforating the endopelvic fascia using curved Metzenbaum scissors with tips pointed laterally and slightly superiorly ( Fig. 40-2). Blunt finger dissection should not be used to perforate the endopelvic fascia as bladder injury may occur. Once the endopelvic fascia is entered, gently advance the closed scissors laterally and slightly upward for 1 to 2 cm before opening widely. Gentle blunt finger dissection of the retropubic space superiorly to the rectus muscle is performed ( Fig. 40-3). Through the abdominal incision, the lateral border of the rectus muscle is retracted medially to expose a defect just lateral to where the rectus muscle inserts onto the symphysis (Fig. 40-4). Gentle dissection in this area allows safe and easy access into the retropubic space. If finger dissection of the retropubic space is difficult, as is sometimes the case after prior procedures, place the tips of Metzenbaum scissors directly on the posterior pubis and slowly advance them with constant pressure against the pubic periosteum. After this is completed, no tissue should be palpable between fingers inserted into the retropubic space from above and the vaginal incision below. If some intervening tissue is found at the level of the pelvic floor, penetration of that tissue is safe. If the tissue is higher than the pelvic floor, it is often the bladder attached to the posterior pubis. The bladder can be carefully dissected off the pubis by keeping the scissors on the back of the pubis at all times. A similar procedure is performed on the other side. Extensive retropubic space dissection is unnecessary and may lead to excessive bleeding or bladder injury. A Sarot or Crawford clamp is placed in the retropubic space from above and directed into the vaginal incision using manual guidance ( Fig. 40-5). The tip of the clamp should remain in contact with the pubic periosteum and under the vaginal operator's finger at all times. After clamps have been passed bilaterally, cystoscopy is performed to ensure there has been no damage to the urethra or bladder. Each sling suture is pulled into the abdominal incision placing the sling under the urethra. Proper function and longevity of the sling does not depend on the sutures to hold tension indefinitely (since the sutures are absorbable) and thus it is critical that a good portion of the sling extend into the retropubic space to allow good fixation. One or two 3-0 absorbable sutures are placed through the edge of the sling and superficially through the periurethral fascia to secure it in place ( Fig. 40-6). The sling sutures are passed through the rectus fascia, directly above the retropubic “tunnel,” using a right angle clamp before the rectus fascia is closed. If a suprapubic tube needs to be placed (we do not recommend this), it is done under direct vision at this time. The vagina is closed with a running, locking 2-0 absorbable suture. The weighted speculum and all other instruments should be removed from the vagina. The sling sutures are gently pulled up to remove any slack and tied over the rectus ( Fig. 40-7). A shodded clamp can be used to hold tension on the untied sutures until the appropriate tension is obtained. The appropriate tension is the minimum amount required to stop urethral motion, which is tested by pulling on the urethral catheter. Also, one or two fingers should easily slide under the suture knot. If in doubt it is better to err on the side of too little tension. We do not place a vaginal pack before tying the sutures. We have not found it useful to judge how tight to pull the sling by visual assessment during cystoscopy nor by tightening the sling until leakage cannot be produced by compressing the bladder. In the situation where the patient does not void and permanent urinary retention is desired, increased tension can be applied. The skin is closed and a vaginal pack placed. When the abdominal and vaginal components are performed synchronously, the average operating time is 40 minutes with 50 ml average blood loss.

FIG. 40-2. The vaginal dissection is performed superficially to the white periurethral fascia. With the scissors parallel to the plane of the perineum and tips pointing superiorly and laterally, the retropubic space is entered and subsequently enlarged by further advancing the scissors 1 to 2 cm and then opening them.

FIG. 40-3. Blunt finger dissection creates a tunnel to the rectus muscles above. Wide dissection is unnecessary and may cause significant bleeding or bladder injury.

FIG. 40-4. The approach to the retropubic space from above is located below the rectus fascia and lateral to where the rectus muscles attach to the pubic symphysis. Minimal dissection in this area allows safe and easy access to the retropubic space previously dissected by the vaginal operator.

FIG. 40-5. Using manual guidance, a Crawford clamp is passed from above toward the vaginal incision with the tip of the clamp in contact with the pubic periosteum and the vaginal operator's finger. After the passage of clamps bilaterally, cystoscopy is performed to ensure that no injury to the bladder has occurred.

FIG. 40-6. The sling ends are pulled well into the retropubic space to allow good fixation. The sling is seated at the proximal urethra and sutured to the periurethral fascia using 3-0 absorbable suture.

FIG. 40-7. The sling sutures are passed through the rectus fascia before the fascia is closed. The vaginal mucosa is closed, the weighted speculum removed, and the sling sutures are tied down over the rectus fascia under minimal tension.

On postoperative day 1, the vaginal pack is removed; if the patient is ambulating well, the Foley catheter is removed. The patient performs clean intermittent catheterization after each void, or a minimum of every 4 hours if unable to void, until the postvoid residual is consistently under 60 ml. Patients are regularly discharged within 48 hours. Oral antibiotics are not routinely prescribed postoperatively. Patients should refrain from vigorous activities and sexual intercourse for 4 to 6 weeks to allow proper fixation of the sling.

OUTCOMES
Complications When rectus fascia is used for the urethral sling, the most common complications include detrusor instability and urinary retention. Approximately 15% to 25% of patients will have residual urgency symptoms, with less than half demonstrating occasional urge incontinence. 1,2 Less than 10% will develop new onset detrusor instability. In a recent report by O'Connell and colleagues, 26% of patients had residual urgency symptoms and less than half of this group had mild urge incontinence.7 In most cases, these symptoms are responsive to anticholinergic medications and will subside over a period of 3 to 6 months. Persistent postoperative urinary retention, although believed to be a common complication, is not statistically more common after pubovaginal slings than after suspension procedures. In a recent series of 54 patients, no patient who could void preoperatively was in persistent retention postoperatively. 7 McGuire and colleagues reported a 3% incidence of prolonged retention in one series. 5 Superficial wound infections occur in approximately 4% of patients and significant blood loss occurs in 1% to 2%. 7 Wound infections have not resulted in sling failure. Although synthetic sling materials may be used, relatively high rates of infection and urethral erosion have been reported. Persistent postoperative pain is rare when absorbable suture is used. O'Connell and colleagues reported that no patient had to take analgesics chronically and no patient had a procedure to relieve pain. 7

Results In a recent series, 82% of patients were totally dry and another 11% had rare incontinence (once a week or less) for an overall 93% cured or significantly improved. Other long-term series have documented a greater than 80% cure and over 90% significantly improved rate. 1,2,5 Residual stress incontinence usually responds very well to injectable agents such as collagen. The pubovaginal fascial sling is the procedure of choice for treatment of females with urinary incontinence due to ISD. Even patients who had prior surgical failures can obtain excellent results with minimal morbidity, but such results are contingent on an accurate preoperative evaluation and careful placement of the sling at the proximal urethra without undue tension. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Blaivas JG, Jacobs BZ. Pubovaginal sling: long term results and prognostic indicators. J Urol 1993;149:292A, Abstract 313. McGuire EJ, Lytton B. The pubovaginal sling for stress urinary incontinence. J Urol 1978;119:82. McGuire EJ, Fitzpatrick CC, Wan J, et al. Clinical assessment of urethral sphincter function. J Urol 1993;150:1452. McGuire EJ. Urodynamic findings in patients after failure of stress incontinence operations. In: Zinner NR, Sterling AM, eds. Female incontinence. New York: Alan R. Liss, 1981;351–360. McGuire EJ, Bennett CJ, Konnak JA, Sonda LP, Savastano JA. Experience with pubovaginal slings for urinary incontinence at the University of Michigan. J Urol 1987;138:525. O'Connell HE, McGuire EJ, Aboseif S. Transurethral collagen injection therapy in women. J Urol 1995;154:1463. O'Connell HE, McGuire EJ, Usui A, Gudziak M. Pubovaginal slings in 1994. J Urol 1995;153:525A, abstr. 1186. Raz S, Stothers L, Young GPH, et al. Vaginal wall sling for anatomical incontinence and intrinsic sphincter dysfunction. Efficacy and outcome analysis. J Urol 1996;156:166. Ridley JH. The Goebell-Stoeckel sling operation. In: Mattingly RF, Thompson JD, eds. TeLinde's operative gynecology. 4th ed. Philadelphia: JB Lippincott, 1985;923–935.
7

Chapter 41 Injections for Incontinence in Women and Men Glenn’s Urologic Surgery

Chapter 41 Injections for Incontinence in Women and Men
Rodney A. Appell and Randy A. Fralick

R. A. Appell: Section of Voiding Dysfunction and Female Urology, Department of Urology, The Cleveland Clinic Foundation, Cleveland, Ohio 44195. R. A. Fralick: Section of Voiding Dysfunction and Female Urology, Department of Urology, The Cleveland Clinic Foundation, Cleveland, Ohio, 44195, and Holy Family Memorial Medical Center, Manitowoc, Wisconsin 54220.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Technique of Injection in the Male Patient Technique of Injection in the Female Patient Outcomes Complications Results Chapter References

In evaluating patients for the use of intraurethral injections as a treatment of urinary incontinence, it is essential to identify the cause(s) of incontinence in order to recommend appropriate therapy. Intraurethral injections benefit patients with incontinence occurring at the level of the bladder outlet. Incontinence occurring at this level may be caused by anatomic displacement of a normally functioning urethra (anatomic genuine stress urinary incontinence) in women or intrinsic incompetence of the urethral closure mechanism (intrinsic sphincteric dysfunction) in women or men. Patients with intrinsic sphincteric dysfunction (ISD) commonly have had a previous surgical procedure on or near the urethra, a sympathetic neurologic injury, or myelodysplasia. Female patients with genuine stress urinary incontinence have normal urethral function but hypermobility of the bladder neck and proximal urethra resulting from a deficiency in pelvic support. These patients benefit from bladder neck elevation and stabilization. Patients with ISD have poor urethral function and require procedures to increase outflow resistance. Patients with a fixed, well-supported urethra in association with ISD are excellent candidates for periurethral injection. In men this is most commonly encountered following radical prostatectomy, whereas in women the primary cause of ISD is a residual effect of multiple surgical resuspension procedures for genuine stress urinary incontinence. 1

DIAGNOSIS
Patients with ISD urodynamically display an open bladder outlet at rest in the absence of a detrusor contraction. However, standardization of a methodology to determine ISD has not yet been accepted. Because the incontinence of ISD is nonresistant (passive) urinary leakage, the goal of treatment should be only to coapt the urethra by passive occlusion. With respect to outlet function, maximum urethral closure pressure (UCP max) obtained during urethral pressure profilometry has been the test used to determine the presence of ISD in women. Those women with UCP max £ 20 cm H2O were said to have a “low-pressure urethra” or ISD, and these patients failed standard bladder neck suspension procedures. 9 One can infer that if urethral urinary loss can be induced by abdominal pressure, something must be wrong with the outlet. The abdominal leak-point pressure (LPP abd) is determined by direct measurement of the abdominal pressure required to overcome urethral resistance. This determination may be considered an indirect method of measuring the closure forces of the urethra during straining maneuvers. It is used primarily in women with stress urinary incontinence (SUI) to differentiate between anatomic displacement of a normal-functioning urethra (SUI caused by hypermobility) from poor outlet function (ISD). The leakage may be documented visually or fluoroscopically. At the point at which leakage occurs, LPP abd is recorded. Patients with ISD demonstrate minimal urethral resistance to straining, and therefore the urethral opening pressure is very low, whereas patients with an anatomic displacement of a normally functioning urethra have high urethral opening pressures, and therefore the LPP abd will be higher. However, no standardization of the technological methods to obtain LPP abd has been accepted. In summary, in evaluating leak-point pressures, until a universally accepted technique is established, a single mode with which the physician is comfortable should be used in the same manner on every patient whether for a preoperative evaluation or for the evaluation of a patient with an unsatisfactory result.

INDICATIONS FOR SURGERY
During the multicenter investigation of collagen in the treatment of ISD, the patients selected with anatomic (type II vesicourethral hypermobility) incontinence did not fare well.7 Therefore, the recommendation currently is to perform intraurethral or periurethral injections on patients with a poorly functioning urethra (ISD) and good anatomic support. However, recent data suggest that injectables may be used for selected female patients with anatomic stress urinary incontinence. 6

ALTERNATIVE THERAPY
The treatment of urinary incontinence related to the incompetent urethra has been a challenging problem and frequently involves surgical augmentation of intraurethral pressures by the use of slings made of autologous or synthetic materials, implantation of an artificial sphincter, or periurethral injection of bulk-enhancing agents.

SURGICAL TECHNIQUE
The technique of injection of material is not difficult; however, it is essential to perform precise placement of the material in order to ensure an optimal result. The equipment required for injection depends on the bulk-enhancing agent injected. The injection can be performed either suburothelially through a needle placed directly through a cystoscope (transurethral injection) or periurethrally with a needle inserted percutaneously and positioned in the urethral tissues in the suburothelial space, with the manipulation observed by cystourethroscopy. 1 Men are injected predominantly by the transurethral approach, and women are injected by either technique. There is certainly a learning curve with any technique chosen, which ultimately results in using less injectable material to attain the desired result of continence. Injection techniques using glutaraldehyde-cross-linked collagen (Contigen) are presented, as this is currently the only injectable approved for incontinence. Technique of Injection in the Male Patient The patient is positioned in the semilithotomy position and prepped and draped in the usual fashion, and 10 ml of a 2% lidocaine jelly is placed intraurethrally and left in place 10 minutes before instrumentation. Cystourethroscopy with a zero-degree lens is employed. The injectable material is then delivered suburothelially by way of a transcystoscopic injection needle under direct vision. The needle is advanced under the mucosa with the beveled portion of the needle facing the lumen of the urethra. This is performed in a circumferential matter, employing four quadrant needle placements ( Fig. 41-1). The material is injected until a mucosal bleb is created in each quadrant. Gradually, after the circumferential injections, the urethral mucosa meets in the midline, although additional needle placements may be required for completion.

FIG. 41-1. Transurethral injection in a male patient.

In cases of ISD following a radical prostatectomy, a short segment of urethra remains above the external sphincter. If visualization of this segment of the urethra is difficult, the needle may be placed at the level of the external sphincter and advanced to ensure deposition of the material proximal to the external sphincter. To be effective, any injectable material must be injected in the urethra superior to the external sphincter, even if this means injecting into the actual bladder neck on the proximal side of the anastomosis. It is important to note that the material should not be injected directly into the external sphincter, as this can cause pudendal nerve irritation with resultant sphincter spasm and discomfort. The depth of injection is also critical. The materials must deform the urethral mucosa so that it closes the urethral lumen. Too deep an injection site is a waste of the material and is not effective. Injection is more difficult in patients with post-radical-prostatectomy incontinence resulting from the short segment of urethra above the level of the external sphincter and extensive scarring, which usually occurs in this area following surgery. This problem can be circumvented by using an antegrade approach. The technique is performed by passing a cystoscope with a 5-Fr working port through a small suprapubic cystotomy tract. The vesical neck and proximal urethra are then visualized, and subepithelial injections are performed until the bladder outlet is coapted ( Fig. 41-2). Frequently there is less scar tissue in this location, which results in better tissue coaptation. In early clinical trials, this technique seems to facilitate more precise injection of material, generating improved results with the use of less material. 4 In the authors' opinion this technique represents an exciting new method of implantation in male patients and should be considered in any postprostatectomy man not achieving adequate success by way of the standard transurethral approach. Post-radical-prostatectomy urothelium covers scar, and there is migration of any injectable substance distally along the urethra. Once this stops, there is a “wall” to abut the freshly injected material at the bladder neck. Therefore, this additional technique is not recommended as the primary method for an initial injection.

FIG. 41-2. (A) Beginning antegrade injection in a male patient. (B) Completion of antegrade injection.

A small subset of patients continue to have some degree of incontinence after the placement of a bulbous urethral artificial urinary sphincter. To date the only options to address this problem have been to place a more distal second (tandem) cuff around the bulbous urethra or to place a higher-pressure regulating balloon. Injectable agents have generally been avoided in this setting because of fear of damaging the sphincter cuff. The antegrade approach can be used for this situation without fear of damaging the cuff, although it remains important to know the location of the pressure-regulating balloon before performing the punch cystotomy. In cases of ISD following prostatic resection, a short segment of urethra remains below the veru montanum and yet is still proximal to the external sphincter. The injections should be made in this position circumferentially until urethral coaptation is visible. Extrusion of material into the urethral lumen from the needle holes may occur but can be minimized by not traversing the injected area with the distal end of the cystoscope once the material has been injected. In other words, do not enter the bladder. Technique of Injection in the Female Patient Women may be injected by way of a transcystoscopic technique, as described for the male patient, or by a periurethral approach. 1 The patient is placed in the lithotomy position and prepped and draped in the usual fashion. The introitus and vestibule are anesthetized with 20% topical benzocaine, and the urethra is anesthetized with 10 ml of 2% lidocaine jelly. Following this, a local injection of 1% plain lidocaine is performed periurethrally at the 3- and 9-o'clock positions using 2 to 4 ml on each side. Panendoscopy is performed with a 0- or 30-degree lens, and the needle is positioned periurethrally at the 4- or 8-o'clock position with the bevel of the needle directed toward the lumen (Fig. 41-3). The needle is then advanced into the urethral muscle into the lamina propria in an entirely suburothelial plane. Once the needle is positioned in the lamina propria, it usually advances with very little force. The needle may also be introduced between the urethral fascia and vaginal epithelium at the 6-o'clock position ( Fig. 41-4), and, again, needle placement is fully observed endoscopically. Bulging of the tip of the needle against the mucosa of the urethra is observed during advancement of the needle to ensure its proper placement. When the needle tip is properly positioned 0.5 cm below the bladder neck, the material is injected until swelling is visible on each side, creating the appearance of occlusion of the urethral lumen ( Fig. 41-3). Once the urethra is approximately 50% occluded, the needle is removed and reinserted on the opposite side, and additional material is injected until the urethral mucosa coapts in the midline, creating the endoscopic appearance of two lateral prostatic lobes.

FIG. 41-3. Technique of periurethral injection in a female patient. (A) Appearance of the urethra before treatment. (B) Needle positioned in proximal urethra below the bladder neck. (Note the submucosal location within the lamina propria.) (C) Urethra after completion of injection.

FIG. 41-4. Transvaginal injection of collagen.

Although urologists and urogynecologists are more familiar with transurethral than periurethral techniques, we prefer the periurethral approach, as this minimizes intraurethral bleeding and extravasation of the injectable substance. A useful “trick” described by Neal et al. is to add methylene blue to the injectable lidocaine to aid in the location of the needle tip before injecting the bulking agent. 8 Once the needle is located at the bladder neck position, the syringe of anesthetic/methylene blue is removed, and the syringe containing the bulking agent is engaged. When the desired appearance of the coapted mucosa is attained, have the patient stand and cough to see if there is any leakage, and, if there is, reposition the patient and reinject. Perioperative antibiotic coverage is continued for up to 3 days following the procedure. Most patients are able to void without difficulty following the procedure; however, if retention does develop, clean intermittent catheterization is begun with a 10- to 14-Fr catheter. An indwelling urethral catheter is to be avoided in patients, as this promotes molding of the material around the catheter. Although it is usually unnecessary, if longer-term catheterization is needed, suprapubic cystotomy should be performed in these patients. Patients are contacted 2 weeks postprocedure in order to determine their continence status. Repeat injections are scheduled as necessary and at a time interval appropriate for the injectable substance.

OUTCOMES
Complications The perioperative complications associated with periurethral injections are uncommon ( Table 41-1). In the multicenter clinical trial using Contigen injections, transient retention developed in approximately 15% of patients, but only 1% of patients experienced irritative voiding symptoms, and 5% developed a urinary tract infection. 5 Hypersensitivity responses with Contigen are not a problem, as the possibility is assessed by skin testing (wheal and flare) with the more immunogenic and sensitizing non-cross-linked collagen prior to treatment. Those with a positive skin test are excluded from treatment. Regardless of the material, the use of periurethral injections has proven to be safe, eliciting only minor complications. All complications resolve rapidly, and a serious long-term complication from the use of periurethral injections has yet to be reported.

TABLE 41-1. Adverse events reported during multicenter study of 382 patients treated with GAX-collagen a

Results There are no controlled, long-term reports available on any injectable. In fact, it is difficult to glean information in any group reported as to etiology of the incontinence. For example, in women results of injectables are reported without differentiating among patients with hypermobility, those with ISD, and those with both; and men with prostatic resection for benign disease are not separated from those having had a radical prostatectomy. Thus, results have been a combination of anecdotal reporting mixed with conjecture, speculation, and the hope that the truth is involved. Having stated this, it appears that injectables are helpful for some incontinent patients, especially selected women. There are two major disadvantages to the use of injectables: (a) the inability to determine the quantity of material needed for an individual patient and (b) the safety of nonautologous products for injection with respect to migration, foreign body reaction, and immunologic effects. At this point in time only Contigen has been approved as an injectable for incontinence in the United States, and results presented are confined to this approved substance. Results of the North American Contigen Study Group of 134 postprostatectomy patients (17 postresection; 117 post-radical-prostatectomy) and 17 postradiation incontinent men demonstrated that only 22 men (16.5%) regained continence following injections of collagen, but 78.7% were dry or significantly improved at 1 year of follow-up, and 67% at 2 years following injections. 2,7 Use of the antegrade injection technique in men failing the standard retrograde, cystoscopic approach increased the “cured” rate at 1 year by another 37.5%. 4 Results of the North American Contigen Study Group of 127 women demonstrated 46% dry and 34% socially continent (patients requiring a single minipad/day) with 77% remaining dry once continence had been attained. 3 Worldwide independent studies have supported these findings. Patients with no anatomic hypermobility and ISD appear to be the most satisfactory candidates for intraurethral injections. In selected elderly and less mobile female patients with anatomic incontinence, recent data suggest that collagen may also be useful in this patient population. 6 Injectables are still in the developmental stages and their roles in the management of incontinence still need to be defined more precisely. Because the methods are less invasive and generally performed on an outpatient basis, medical costs should be reduced, and there should also be a more rapid return to the patient's normal activities. The ideal material is still sought and should combine ease of administration with minimal tissue reaction, lack of migration, and persistence over time. The physician considering injectables for his or her patient should consider that there is a learning curve in patient selection as well as method of delivery of the bulking agents to attain optimal results. CHAPTER REFERENCES
1. Appell RA. Collagen injection therapy for urinary incontinence. Urol Clin North Am 1994;21:177. 2. Appell RA, McGuire EJ, DeRidder PA, et al. Summary of effectiveness and safety in the prospective, open, multicenter investigation of contigen implant for incontinence due to intrinsic sphincteric deficiency in males. J Urol 1994;151(2):271A. 3. Appell RA, McGuire EJ, DeRidder PA, et al. Summary of effectiveness and safety in the prospective, open, multicenter investigation of contigen implant for incontinence due to intrinsic

4. 5. 6. 7. 8. 9.

sphincteric deficiency in females. J Urol 1994;151(2):418A. Appell RA, Vasavada SP, Rackley RR, et al. Percutaneous antegrade collagen injection therapy for urinary incontinence following radical prostatectomy. Urology 1996;48:769. Bard CR. PMAA submission to US Food and Drug Administration for IDE G850010, 1990. Faerber GJ. Endoscopic collagen injection therapy for elderly women with type I stress urinary incontinence. J Urol 1995;155(2):527A. McGuire EJ, Appell RA. Transurethral collagen injection for urinary incontinence. Urology 1994;43:413–415. Neal D Jr, Lahaye K, Lowe D. Improved needle placement technique in periurethral collagen injection. Urology 1995;45:865–866. Sand PK, Bowen LW, Panganiban R, et al. The low-pressure urethra as a factor in failed retropubic urethropexy. Obstet Gynecol 1987;69:399.

Chapter 42 Pelvic Floor Relaxation Glenn’s Urologic Surgery

Chapter 42 Pelvic Floor Relaxation
David A. Ginsberg, Eric S. Rovner, and Shlomo Raz

D. A. Ginsberg: Department of Urology, University of Southern California School of Medicine/Norris Cancer Center, Los Angeles, California 90033. E. S. Rovner: Division of Urology, Department of Surgery, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104. S. Raz: Division of Urology, Department of Surgery, University of California, Los Angeles, Los Angeles, California 90024.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Positioning and Retraction Exposure of Perineal Body Exposure of Distal Vaginal Defect Exposure of Proximal Vaginal Defect Plication of Prerectal and Pararectal Fascia Repair of the Levator Hiatus Repair of the Perineal Body Outcomes Complications Results Chapter References

A rectocele is secondary to a defect in the supporting fascia of the rectum that results in a herniation of the anterior rectal and posterior vaginal wall into the lumen of the vagina. The true incidence of rectoceles is unknown. Wells et al. reported a 12% incidence of rectoceles on physical examination when evaluating patients complaining of urinary incontinence. 9 Concomitant rectocele or enterocele repair was performed in 35% of patients undergoing a Raz bladder neck suspension 7; however, 65% of patients who underwent repair of a grade IV cystocele required rectocele repair. 6 To understand the concepts underlying repair of pelvic floor relaxation, the anatomy of the normal pelvic floor support system should be briefly reviewed. The pelvic diaphragm is the superior shelf of the pelvic floor and consists of the levator ani and the coccygeus muscles ( Fig. 42-1). The urogenital diaphragm forms the second layer of the pelvic floor and consists of the bulbocavernosus, transverse perinei, and external anal sphincter muscles. These muscles join together with the anterior fibers of the levator ani to form the central tendon of the perineum ( Fig. 42-2).

FIG. 42-1. The pelvic diaphragm consisting of the levator ani and the coccygeus muscles and their investing fascia. The levator hiatus is an area of relative weakness. Prerectal fibers of the levator ani between the rectum and the vagina help to maintain a narrow hiatus that resists prolapse of pelvic organs. (From Raz S, Little NA, Juma S. Female urology. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;2782–2828.)

FIG. 42-2. The urogenital diaphragm, consisting of the bulbocavernosus, superficial and deep transverse perinei, and external anal sphincter muscles, and their investing fascia, provides the second layer of pelvic support and is anchored in the middle by the central perineal tendon. (From Raz S, Little NA, Juma S. Female urology. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;2782–2828.)

The fascial support of the rectum consists of the prerectal fascia and the pararectal fascia. The prerectal fascia runs anterior to the rectum from the pouch of Douglas to the central tendon and prevents protrusion of the rectum into the vagina. A virtual space exists between the posterior vaginal wall and the prerectal fascia, which offers a convenient plane of dissection during rectocele repair. The pararectal fascia originates from the lateral pelvic sidewall and sweeps posteromedially to the rectum, splitting into anterior and posterior sheets and forming a fibrous envelope around the rectum ( Fig. 42-3).

FIG. 42-3. The prerectal and pararectal fascia provide support to the posterior pelvic floor. (From Babiarz JW, Raz S. Pelvic floor relaxation. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996;442–456.)

The normal vaginal axis that is seen in the well-supported pelvic floor conveniently protects against rectocele formation and further pelvic prolapse. Two distinct areas of the vagina are seen if a normal vaginal axis is maintained. The proximal vagina lies at a 110- to 120-degree angle to the horizontal. The distal vagina, with the sling-like support provided by the levators, forms an angle of 45 degrees from the vertical. This results in a midvaginal angle of 110 to 130 degrees ( Fig. 42-4). In women with significant pelvic floor prolapse, levator plate laxity and widening of the levator hiatus result in a disappearance of the normal curvature of the vagina and a near-vertical vaginal axis, which facilitates rectocele formation ( Fig. 42-5).

FIG. 42-4. Lateral view of the pelvis. The distal vagina forms an angle of 45 degrees from the vertical. The proximal half of the vagina lies on top of the levator plate, forming an angle of 110 degrees from the diatal vagina. (From Raz S, Little NA, Juma S. Female urology. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;2782–2828.)

FIG. 42-5. Levator plate laxity and forward pressure from a rectocele result in a near-vertical vaginal axis, which facilitates the downward prolapse of the pelvic organs. (From Babiarz JW, Raz S. Pelvic floor relaxation. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996;442–456.)

The high incidence of concomitant rectocele and cystocele relates to the pathophysiology of pelvic floor weakness and subsequent rectocele formation. Childbirth results in several events that weaken the pelvic floor support system: (a) passage of the child's head through the vagina stretches the prerectal and pararectal fascia and detaches the prerectal fascia from the perineal body; (b) the levator musculature and its fascia are weakened, which allows the levator hiatus to widen; (c) the normal narrowing of the vaginal opening is rendered ineffective secondary to widening of the anogenital hiatus and damage to the UG diaphragm. The changes wrought by childbirth are further enhanced by aging, loss of estrogen stimulation, obesity, smoking, strenuous work/physical activity, and chronic abdominal straining, which is often seen in patients with chronic respiratory diseases and cough, constipation, and bladder outlet obstruction. Furthermore, loss of the normal vaginal axis, which is seen with pelvic floor relaxation (and may be accentuated after cystocele repair and/or anti-incontinence surgery), results in ineffective transmission of intra-abdominal pressures. This may lead to a worsening of preexisting pelvic prolapse and an increased risk of stress incontinence. Defects of the perineal body are often a result of injuries sustained during vaginal delivery or episiotomy.

DIAGNOSIS
The majority of rectoceles are asymptomatic. If symptomatic, rectocele-related complaints are often related to bowel dysfunction and include constipation, the need to digitalize the vagina to facilitate stool passage, a feeling of blockage at the outlet, and a sensation of stool pocketing. Interestingly, although problems with constipation are often correlated with a rectocele, many patients report continued difficulties with constipation after rectocele repair. 1 Patients may also complain of dyspareunia and symptoms attributable to prolapse such as the feeling of a bulge or sitting on a ball. Defects of the perineal body are usually asymptomatic, but patients may complain of incontinence of liquid stool or flatus or loss of sensation during sexual intercourse secondary to a widened introitus. The diagnosis of a posterior wall defect is made on physical examination. Examination of the posterior compartment is best accomplished using a Sims retractor or half of the vaginal speculum to displace the anterior vaginal wall anteriorly. Perineal body defects are associated with a widened introitus and a decreased distance between the anus and the posterior aspect of the vagina and are graded as follows: I, a tear in the hymenal ring; II, a tear involving the perineal body but not the anal sphincter; III, a tear involving the anal sphincter; IV, a tear extending into the anal mucosa. A rectocele will manifest as a bulge extending from the posterior wall of the vagina and is graded as follows: I, protrusion of the posterior vaginal wall at the level of the hymenal ring; II, protrusion at the level of the hiatus; III, protrusion beyond the introitus. Rectoceles may further be classified according to their position in the vagina as low, medium, or high ( Table 42-1). Rectovaginal examination will reveal attenuation of the fascia and helps rule out coincidental enterocele, which should be suspected in the patient with a high rectocele. With posterior wall defects, loss of the normal banana-like axis of the lower and upper vagina is seen, as the vagina will assume a straight orientation. Finally, defecography and dynamic rectal radiologic examinations are used by some authors in the diagnosis and classification of posterior vaginal vault defects. 8,10

TABLE 42-1. Classification of rectoceles by position

INDICATIONS FOR SURGERY
Patients with symptomatic posterior vaginal wall defects should undergo surgical correction. The repair of asymptomatic defects coincident with other vaginal surgery is controversial. Arguments against repair of an asymptomatic rectocele include postoperative coital dysfunction and rectal injury. Jeffcoate described a 30% rate of discontinued coitus or dyspareunia after anterior and posterior repair 3; however, recent reviews evaluating outcomes using present-day techniques describe a 0% to 9% incidence of coital dysfunction. 2,4,5 Rectal injury has not been a concern with current surgical techniques. Arguments favoring repair of asymptomatic pelvic floor relaxation during concomitant vaginal surgery include the risk of larger and symptomatic pelvic prolapse (i.e., rectocele, enterocele, uterine prolapse) if repair is not accomplished and the possibility that results of simultaneous anti-incontinence surgery are improved if repair is done. Anti-incontinence procedures orient the vagina in a vertical axis; however, pelvic floor relaxation repair helps restore the normal near-horizontal axis of the vagina. Restoration of this axis decreases the incidence of postoperative prolapse, results in more effective transmission of intraabdominal pressure to the pelvis, and should improve the results of anti-incontinence surgery by helping to provide a strong backboard against which the bladder neck and urethra (which are secondarily supported by the pelvic floor) can be compressed. These arguments, combined with the ability to accomplish this surgery without introducing significant perioperative morbidity, leads us to strongly favor simultaneous repair of even asymptomatic moderate pelvic floor weakness at the time of concurrent vaginal procedures.

ALTERNATIVE THERAPY
Alternatives to repair of pelvic floor relaxation include observation and intravaginal pessaries.

SURGICAL TECHNIQUE
The essential goals of rectocele repair include (a) plication of the prerectal and pararectal fascia, (a) narrowing of the levator hiatus by reapproximating the prerectal levator fibers; 3) repair of the perineal body. Two days before surgery, the patient begins a clear liquid diet and begins oral laxatives. Broad-spectrum intravenous antibiotics to cover anaerobes, gram-negative bacilli, and group D enterococcus are administered preoperatively. Positioning and Retraction The patient is placed in the dorsal lithotomy position, and a Betadine-soaked rectal packing is placed to aid in identification of the rectum and to avoid rectal injury. The patient is draped (the rectal packing is isolated from the operative field with double draping), and a Foley catheter is placed. Anti-incontinence surgery, cystocele repair, enterocele repair, and vaginal hysterectomy, if indicated, are accomplished first. A ring retractor with hooks, applied to the perineum, aids in lateral exposure of the vaginal vault. The anterior vaginal wall is retracted upward with a Haney or right-angle retractor to improve visualization and help prevent excessive narrowing of the vagina. Exposure of Perineal Body The rectocele repair begins with the placement of two Allis clamps to the posterior margin of the introitus at the 5- and 7-o'clock positions. A V-shaped incision is made, and a triangular segment of perineal skin with the base of the triangle at the mucocutaneous junction is excised between the Allis clamps, exposing the attenuated perineal body ( Fig. 42-6).

FIG. 42-6. (A) Overlying triangle represents site of incision for this step. (B) Pelvic floor repair begins with excision of triangle of skin at the mucocutaneous junction of posterior vaginal wall and perineum.

Exposure of Distal Vaginal Defect The Allis clamps are then placed in the midline of the posterior vaginal wall, grasping and elevating the rectocele at its midpoint. Saline is injected along the posterior vaginal wall to facilitate dissection. With the use of a scalpel, a second triangular incision is made in the posterior vaginal wall with the base of the triangle at the site of the previous incision and the apex of the triangle above the levator plate 2 to 3 inches inside the hymenal ring ( Fig. 42-7). This is a superficial incision through the vaginal wall only; a deeper dissection at this point risks injury to the rectum. Metzenbaum scissors are then used to sharply develop a plane from the lateral margins of the triangle, dissecting between the herniated rectal wall and the vaginal wall. Staying as close as possible to the vaginal wall to avoid injury to the rectum, the dissection extends laterally, exposing the attenuated prerectal fascia distally. The triangular island of posterior vaginal wall that was created by the inverted V-shaped incision is sharply excised off the prerectal levator fascia and fibers. This redundant skin is not discarded until the rectocele is entirely repaired; if the repair is accidentally too tight and/or excessively narrows the vagina, the excised piece of vaginal wall may be used as a free graft.

FIG. 42-7. (A) Overlying triangle represents site of incision along posterior vaginal wall. (B) With lateral dissection and excision of the vaginal wall, the underlying

pararectal and prerectal fascia is exposed.

Exposure of Proximal Vaginal Defect The prerectal fascia is exposed by sliding the Metzenbaum scissors under the posterior vaginal wall from the apex of the previous triangular incision to the cuff of the vagina. The posterior vaginal wall is then incised along the midline. This incision is made from the apex of the previous triangular incision to the vaginal cuff. An appropriately sized rectangular strip of posterior vaginal wall is excised (a greater severity of prolapse necessitates a wider resection of posterior vaginal wall), exposing the attenuated pararectal and prerectal fascia proximally ( Fig. 42-8). Use of a Haney or right-angle retractor on the anterior vaginal wall at this point helps prevent resection of an excessive amount of posterior vaginal wall, thus decreasing the risk of vaginal stenosis postoperatively. Inadequate resection of sufficient vaginal wall risks a weak repair and the formation of painful ridges during reconstruction.

FIG. 42-8. (A) Overlying vertical line represents site of incision along posterior vaginal wall, extending from previous triangular apex to vaginal cuff. (B) A rectangular strip of posterior vaginal wall is excised, exposing the attenuated pararectal and prerectal fascia proximally.

Plication of Prerectal and Pararectal Fascia At this point attention is turned toward repair of the rectocele. The anterior vaginal wall is retracted upward, and the distal rectum is retracted downward with a Haney or right-angle retractor. This protects the rectum, reduces the rectocele, and facilitates reapproximation of the pararectal and prerectal fascia. Reconstruction begins at the apex of the rectocele and is carried out to the level of the levator hiatus with a running, locking 2-0 polyglycolic acid suture. Each needle passage incorporates the edge of the vaginal wall and generous bites of the prerectal fascia and the pararectal fascia bilaterally ( Fig. 42-9). We attempt to reapproximate the sacrouterine/cardinal ligament complex with the initial bite of this portion of the repair to decrease the risk of subsequent enterocele formation.

FIG. 42-9. (A) A running, locking, absorbable suture is used to reapproximate the vaginal wall and prerectal and pararectal fascia up to the level of the levators. (B) Schematic diagram demonstrates layers incorporated with repair.

Repair of the Levator Hiatus Two or three interrupted figure-of-eight 2-0 polyglycolic acid sutures are placed, closing the distal posterior vaginal wall to the level of the perineum ( Fig. 42-10). This suture incorporates the same layers as previously described. As the reconstruction continues, each side of the vaginal wall should proportionally come together such that the most distal aspect of the repair, at the mucocutaneous junction, is reapproximated evenly. Reapproximation of the prerectal levator fascia at this level restores the normal axis of the vagina. Therefore, examination of the repair at this point should reveal a well-supported posterior vaginal wall with a concavity (corresponding to the normal midvaginal axis of 110 degrees) to the repair proximally. Finally, a smooth contour without ridges should be noted along the suture line.

FIG. 42-10. The prerectal levator fibers are reapproximated with interrupted absorbable sutures placed in a figure-of-eight fashion.

Repair of the Perineal Body Several vertical mattress sutures of 2-0 polyglycolic acid are used to approximate the bulbocavernosus, transverse perineal, and external anal sphincter muscles ( Fig. 42-11). This brings together the muscles of the UG diaphragm, reconstructing and providing support to the central tendon. The perineal skin is closed with a running 4-0 polyglycolic acid suture ( Fig. 42-12), and an antibiotic-impregnated vaginal packing is placed.

FIG. 42-11. Vertical mattress sutures are placed to reconstruct the perineum.

FIG. 42-12. The repair is completed with reapproximation of the perineal skin with a running 4-0 absorbable suture. (From Raz S, Little NA, Juma S. Female urology. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;2782–2828.)

This procedure is performed as an outpatient surgery. The Foley catheter and vaginal packing are removed several hours after surgery, and patients are prepared for discharge within 6 to 20 hours postoperatively. Patients are sent home with oral antibiotics and are maintained on stool softeners for 1 month. Finally, patients are encouraged to resume early postoperative coitus to ensure normal resumption of sexual function.

OUTCOMES
Complications Urinary retention is the most frequent complication of rectocele repair and occurs in 12.5% of patients. 1 Retention in these patients is temporary and rarely lasts more than several days. Rectovaginal fistula was not seen in our series but has been reported in up to 5% of patients undergoing pelvic floor repair. 5 Dyspareunia can be averted by not excessively narrowing the vagina, avoiding suture placement directly into the levators, and by not leaving uneven, painful ridges along the repair. Other complications of vaginal surgery include infection, bleeding, vaginal shortening, vaginal wall inclusion cyst formation, and fistula. Results Recurrent rectocele is very uncommon and has not occurred in any of the 95 patients we recently reviewed. However, recurrent pelvic prolapse can be expected in as many as 7.5% of patients postoperatively. Constipation is not resolved in up to 50% of patients undergoing rectocele repair for this complaint; this is likely a result of the multifactorial etiology of constipation in many patients. 1 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Arnold MW, Stewart WR, Aguilar PS. Rectocele repair: Four years experience. Dis Col Rect 1990;33:684. Haase P, Skibsted L. Influence of operations for stress incontinence and/or genital descensus on sexual life. Acta Obstet Gynecol Scand 1988;67:659. Jeffcoate TN. Posterior colpoperineorrhaphy. Am J Obstet Gynecol 1958;77:490 Nichols DH. Posterior colporrhaphy and perineorrhaphy: Separate and distinct operations. Am J Obstet Gynecol 1991;164:714. Pratt JH. Surgical repair of rectocele and perineal lacerations. Clin Obstet Gynecol 1972;15:1160. Raz S, Little NA, Juma S, Sussman EM. Repair of severe anterior vaginal wall prolapse (grade IV cystourethrocele). J Urol 1991;146:988 Raz S, Sussman EM, Erickson DB, Bregg KJ, Nitti VW. The Raz bladder neck suspension: Results in 206 patients. J Urol 1992;148:845. Sentovich SM, Rivela LJ, Christensen MA, Blatchford GJ. Simultaneous dynamic proctography and peritoneography for pelvic floor disorders. Dis Col Rect 1995;38:912. Wells TJ, Brink CA, Diokno AC. Urinary incontinence in elderly women: Clinical findings. J Am Geriatr Soc 1987;35:933. Wiersma TG, Mulder CJ, Reeders JW, Tytget GN, Van Waes PF. Dynamic rectal examination (defecography). Ballieres Clin Gastroenterol 1994;8:729.

Chapter 43 Rectus Muscle Sling Procedure for Severe Stress Urinary Incontinence Glenn’s Urologic Surgery

Chapter 43 Rectus Muscle Sling Procedure for Severe Stress Urinary Incontinence
Niall T. M. Galloway

N. T. M. Galloway: Section of Urology, Department of Surgery, Emory University School of Medicine, The Emory Clinic, Atlanta, Georgia 30322.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complication Results Chapter References

Type III stress urinary incontinence results from intrinsic dysfunction of the urethra and bladder neck incompetence. Effective repair must restore closure of the deficient urethra. Current surgical techniques include the use of fascial slings, vaginal island slings, artificial urinary sphincter, or periurethral injections. A variety of natural materials have been used for sling procedures, the most popular being fascia lata or rectus fascia. Synthetic materials are convenient but are more prone to problems of erosion or infection. The narrow dimensions of a traditional sling make it important that the surgeon position the sling accurately at the proximal urethra. A more distal location can produce outflow obstruction or problems with recurrent infection or voiding difficulty. The rectus muscle provides a broad platform of support for the bladder neck and urethra, and accurate placement seems to be less of a problem. 1

DIAGNOSIS
There is a clinical pattern of sacral neurogenic deficit that is characterized by flat feet and loss of intrinsic muscle function of the toes (inability to abduct the toes), and the lateral toes may be hypoplastic. On perineal examination, there is loss of two-point discrimination (4 cm) in the postanal (S5) or perianal (S4) dermatomes, and anal examination reveals loss of anal tone and anal grip that is weak and not sustained. The severity of urinary leakage will give a clue to intrinsic urethral weakness. If the patient leaks with a flood in the supine position on the first or second cough, one should suspect type III stress urinary incontinence. Correction of bladder neck displacement with the examining finger will usually fail to correct the leakage. It is often difficult to assess urethral function in the presence of severe vaginal vault prolapse or procidentia because the prolapsing bladder base may obstruct the urethra. Surgical correction of the prolapse may reveal moderate or severe stress incontinence. Objective urodynamic findings are essential to distinguish the patient who will require a sling procedure. Selection criteria for rectus muscle sling procedure were Valsalva leak-point pressures of less than 60 cm H 2O and/or maximum urethral pressure (Brown and Wickam) of less than 20 cm H 2O and/or a urethral length of less than 1.5 cm.

INDICATIONS FOR SURGERY
Traditional indications would reserve sling procedures for those who have failed a primary surgical repair. In contemporary practice, the sling is also used as a primary procedure for patients with severe stress urinary incontinence. Clinical features would include leakage with a flood that occurs instantly with the first cough in a supine position, in a patient with a comfortably full bladder, or leaks while standing without provocation. Cystoscopic features include open bladder neck and short urethral length (<1.5 cm). Patients who present with total vault prolapse with eversion of the vagina will also require a sling. It is my practice to identify and correct all of the pelvic support defects at the time of surgery.

ALTERNATIVE THERAPY
A variety of natural materials have been used for sling procedures, the most popular being fascia lata or rectus fascia. Synthetic materials are convenient but are more prone to problems of infection and erosion. The narrow dimensions of a traditional sling make it important that the surgeon position the sling accurately at the proximal urethra. A more distal location can produce outflow obstruction, problems with recurrent infection, or voiding difficulty. The artificial urinary sphincter or injection of periurethral bulking agents may be considered for the treatment of type III stress urinary incontinence.

SURGICAL TECHNIQUE
We have used a combined abdominal and vaginal approach. Venous compression pneumatic hose's are applied and activated. The patient is placed in a modified dorsal lithotomy position with the lower limbs supported in the Sams modification of Allen stirrups. Independent grounding pads are placed for the abdominal and vaginal fields. Careful examination of the vagina and perineum confirms the presence or absence of fascial defects for repair. The most common defects are paravaginal, but anterior (pubocervical) or posterior (rectovaginal) fascial defects may also be present. A midline incision is preferred, extending from the left of the umbilicus to the pubic crest. Alternatively, a transverse incision may be made, but this will require that the anterior rectus sheath be incised with a deep U incision and reflected upward to the level of the umbilicus. The rectus abdominis muscle is mobilized from the posterior aspect of the anterior rectus sheath, with care taken to preserve the epimysium, the outer fascial envelope that encloses the muscle bundles. This dissection progresses easily if the assistant elevates the fascial edge with Allis forceps and then Richardson retractors, and the surgeon works from the pubis toward the umbilicus. There are two small perforating vessels that pass anteriorly from the deep inferior epigastrics through the rectus abdominis muscle and the anterior rectus sheath to anastomose with the superficial epigastric vessels in the subcutaneous tissues. These perforators are coagulated and divided. At the lateral border of the rectus muscle, the segmental neurovascular bundles penetrate between the posterior and anterior layers of the rectus fascia; these bundles are coagulated, but the larger bundles may be ligated and divided. Dissection to the lateral border of the muscle in the distal third allows the thin posterior fascia to be opened and the deep inferior epigastric pedicle to be palpated. If the pedicle has been ligated during an earlier procedure, the rectus muscle will be thin and not suitable, but the contralateral muscle can be used. If both pedicles have been ligated, the rectus muscle procedure would be abandoned, and a fascial sling substituted. As dissection proceeds to the level of the umbilicus, a tendinous inscription will be noted where the anterior rectus fascia is adherent to the muscle. Above this level the direction of the fibers will change, and the muscle is mobilized for a further 2 cm cephalad to the tendinous inscription. The surgeon's finger can now be passed from lateral to medial between the rectus muscle and the posterior rectus sheath. A large right-angle forceps is passed, and a heavy Vicryl ligature is carried around the muscle belly. As the ligature is tied, the muscle is divided, and the vascular pedicle is ligated ( Fig. 43-1). A second heavy ligature is tied, and the vessels are ligated and divided. Three holding sutures are placed through the tendon to facilitate the dissection and to be used later.

FIG. 43-1. Division of rectus muscle.

Dissection is continued until the muscle is free from all posterior attachments, and care is taken to clean all of the loose areolar tissue off the posterior rectus sheath to travel with the muscle; this will ensure that the vascular pedicle will be elevated with the muscle and preserved. The pubic insertion and the deep inferior epigastric vessels are not disturbed, but the muscle must be completely free from the overlying anterior rectus sheath and pyramidalis. The muscle is held up by the holding sutures in the tendinous intersection and then folded along its longitudinal axis, by tying the lateral and medial holding sutures, to enclose and protect the vascular pedicle ( Fig. 43-2). The borders are approximated with two absorbable sutures in the middle third of the muscle.

FIG. 43-2. Folding of rectus muscle to protect vascular pedicle is completed.

The vaginal dissection is similar for other slings. The vaginal mucosa is elevated with injectable saline. We favor an inverted-U incision, but if access is narrow, a vertical incision is used. Dissection is carried out laterally to the pubic rami and forward under the bladder neck. The vaginal surgeon will elevate the endopelvic fascia, lateral and close to the pelvic side wall. The abdominal surgeon will incise the endopelvic fascia and create an opening that will admit the passage of first one, and then two, fingers on the left side and then the right. The abdominal surgeon is usually able to guide the vaginal surgeon to avoid opening vaginal vessels, but on occasion the veins must be oversewn to control bleeding. Use of a spreading forceps (Knight Surgical Instrument Co.) allows the passage first of the holding sutures and then of the muscle belly through the ipsilateral defect to the vaginal surgeon. It is drawn downward to deliver the full length and then passed back into the pelvis through the other opening, while the spreading forceps within the pelvis hold open the defect in the endopelvic fascia. The tendinous inscription of the muscle is used to anchor the muscle sling to Cooper's ligament with three or four nonabsorbable sutures ( Fig. 43-3). On occasion, the length of muscle will not reach up to Cooper's ligament, and instead it will be secured to the internal obturator fascia.

FIG. 43-3. Completion of vaginal sling with end of divided rectus muscle sutured to Cooper's ligament.

The muscle belly fills the suburethral space and lies naturally without tension beneath the bladder neck and urethra. The suburethral muscle provides a broad support to the bladder neck and coaptation of the urethra. The vaginal wound is closed with interrupted absorbable sutures. If there are other support defects, the vaginal wound is closed first, before the colpopexy, and the paravaginal sutures from above (or the sacrocolpopexy fixation) are tied down. Cystoscopy is done after intravenous injection of 5 ml of indigo carmine to confirm free efflux from the ureters. Cystoscopy is not used to adjust the sling but only to confirm the normal axis of the urethra and the closed appearance of the bladder neck. The bladder is drained with a urethral catheter. If there is a defect in the rectovaginal fascia or perineal body, this would be repaired after cystoscopy. The abdominal wound is closed with interupted nylon sutures. Particular care is taken to close the fascia at the pubic crest. The patient is mobilized on the first day. The catheter is drained to a bedside bag. It is removed when the patient has begun to pass flatus or had a bowel movement, usually on the third postoperative day. Voiding trials should begin at 3 hours, and after voiding efforts, straight catheterization should be done for residual volume. It is also necessary to drain the bladder one time in the course of the night in the first days. Most patients will require an interval of self-catheterization in the first 2 weeks, and some will use it for as long as 6 weeks. Preoperative teaching is done to encourage the transition through the interval of self-catheterization. When the catheterized volumes decrease to less than 60 ml, it can be tapered off. It is usually helpful to continue with the self-catheterization twice daily for a few days to be sure that emptying has been achieved.

OUTCOMES
Complication

Complications from this operation are few. Of 100 patients treated at our center since 1992, there was no mortality, and ten early complications included superficial wound infection (6), pelvic abscess (2), deep venous thrombosis (1), and fascial dehiscence (1). These complications all occurred early in the series, and some of these were related to the use of a suprapubic catheter. Special attention has been given to the surgical technique, including abandoning the use of suprapubic catheters, the introduction of antibiotic wound irrigation, and also the use of nonabsorbable sutures for the closure. There have been no wound problems in the last 26 cases. There were 11 late complications including incisional hernia in nine. One patient had had multiple periurethral injections of Teflon, and the anterior vaginal wall was a solid indurated mass. After excision, the omentum was used to cover the muscle sling and repair the vaginal defect. This patient did well but developed prolapse of redundant omentum, which had to be trimmed after 6 months. Results One hundred patients have been treated for type III stress urinary incontinence with a rectus muscle sling since 1992. The medical record, patient interviews, and an independent questionnaire were used to evaluate the course and outcome of treatment. The interviews and questionnaires were done independently by a physician who was not a member of the surgical team. The mean age of the patients was 62 years, with a range of 24 to 83. The mean follow-up interval was 22 months, with a range of 3 to 47. Factors predisposing to incontinence included hysterectomy (79), lumbar stenosis (7), pelvic fractures (2), and traumatic paraplegia (1). Forty-nine patients had failed 86 previous surgical repairs, including MMK or bladder neck suspension (61), anterior repair (17), and periurethral collagen (8). All patients had moderate or severe stress urinary incontinence, and all patients had preoperative videourodynamics. Selection criteria for rectus muscle sling procedure were Valsalva leak-point pressures of less than 60 cm H 2O and/or maximum urethral pressure (Brown and Wickam) of less than 20 cm H 2O and/or a urethral length of less than 1.5 cm. Other anatomic defects were present in these patients, and other procedures were combined with the rectus muscle sling as needed. These procedures included sacrocolpopexy (34), paravaginal repair (31), and repair of urethrovaginal fistula (3). The mean hospital stay was 6 days. On leaving hospital, 6% were already voiding to completion. Sixty-three were voiding but needed clean catheterization to empty, and 31% used a suprapubic catheter. After 2 months, 90% were voiding to completion, no patient had a suprapubic catheter, and only 10% used clean intermittent catheterization. The pattern of voiding after a traditional sling may be slow or interrupted. It has been noticeable that voiding is quite normal for many of these patients after rectus muscle sling, and there are few complaints of irritative symptoms. Of the 84 patients who responded to the independent questionnaire, 47% were dry and used no pads, and 33% were satisfied but were still using one or two pads a day. Twenty percent were not satisfied and still used 3 or more pads a day. Eighty-one percent of patients described themselves as improved or much improved after surgery. This report represents a more complex group of patients than many series, because the majority had failed previous surgery, and no effort was made to exclude patients with risk factors of neuropathic disease, fistulas, or diabetes. The rectus muscle flap is not difficult to develop, and the generous vascular pedicle is easy to protect during mobilization and transport of the flap. The muscle flap brings its own blood supply with it, providing excellent oxygenation for the healing tissues. This is an important consideration when dealing with complex problems of incontinence in patients who have had multiple surgical procedures or radiation therapy. This versatile flap may find many other applications in pelvic surgery. CHAPTER REFERENCES
1. Wall LL, Copas P, Galloway NTM. Use of a pedicled rectus abdominis muscle flap sling in the treatment of complicated stress urinary incontinence. Am J Obstet Gynecol 1996;175:140–146.

Chapter 44 Cystocele Glenn’s Urologic Surgery

Chapter 44 Cystocele
Eric S. Rovner, David A. Ginsberg, and Shlomo Raz

E. S. Rovner: Division of Urology, Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania 19104. D. A. Ginsberg: Department of Urology, University of Southern California/Norris Cancer Center, Los Angeles, California 90033. S. Raz: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90024.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Combined Repair of Lateral and Central Defects with Mesh and Vaginal Wall Sling Repair of Lateral Defect (Six-Corner Bladder Suspension) Repair of a Central Defect Outcomes Complications Chapter References

A cystocele represents herniation of the urinary bladder through the weakened supportive fascia of the anterior vaginal compartment. Cystoceles range from involvement of only a small portion of the bladder base with a well-supported urethra to involving virtually the entire bladder and urethra 6 (Table 44-1). Cystocele is one of the manifestations of pelvic floor relaxation and prolapse. Within the context of pelvic floor relaxation, cystocele is commonly associated with other defects in the support of the superior and posterior compartments of the vagina as well. Loss of superior support (uterine prolapse, vault prolapse, and enterocele) and loss of posterior support (rectocele and perineal laxity) may coexist with defects of anterior compartment prolapse (cystocele and urethral hypermobility) and will require a coordinated approach and simultaneous repair.

TABLE 44-1. Classification of cystocelea

Anatomically, a cystocele is the result of loss of pelvic floor support because of weakness of the levator fascia. The levator fascia has the principal role in the support of the anterior vaginal wall, urethra, and bladder. 2 This fascial sheet covering the levator musculature of the pelvic floor inserts on the tendinous arch of the obturator muscle laterally and has a vaginal side and an abdominal side. The abdominal side is referred to as the endopelvic fascia. The vaginal side is called the periurethral fascia at the level of the urethra and the perivesical fascia at the level of the bladder. Together, the periurethral and perivesical fascia comprise the pubocervical fascia. The abdominal and vaginal sides fuse laterally as they insert onto the tendinous arch of the obturator. This fascia has several important condensations that provide lateral support to the bladder and urethra ( Fig. 44-1). The pubourethral ligaments support the midurethra to the inferior margin of the pubic bone. The urethropelvic ligaments suspend the urethra to the lateral pelvic sidewall from the bladder neck to the external meatus. The vesicopelvic ligament extends laterally to the pelvic sidewall, supporting the bladder, and finally, the cardinal ligaments extend from the cervix and upper vagina supporting these structures to the pelvic sidewall.

FIG. 44-1. Diagrammatic view of ligamentous and fascial support of the anterior vaginal wall as seen from the lithotomy position.

Anterior compartment weakness may involve the supporting structures of the urethra, the bladder, or both. Isolated defects in support of the urethra alone result in urethral hypermobility without cystocele. This may result in stress urinary incontinence and is discussed elsewhere in this text. Defects in the anatomic support of the bladder with or without coexisting involvement of the urethra result in cystocele. The cystocele defect may involve either the central or lateral support of the bladder and urethra, or it may be a result of weakening of both ( Fig. 44-2). Isolated central cystoceles are found when the fascia spanning the levator hiatus on the vaginal side (perivesical fascia) becomes attenuated without compromise of the lateral support (urethropelvic and vesicopelvic ligaments). Separation or attenuation of the cardinal ligaments in the midline usually contributes to the anatomic defect in central cystoceles, and reapproximation of these structures is critical to effect repair and prevent the onset of enterocele postoperatively. Isolated central cystoceles are a rare condition and comprise fewer than 10% of cystoceles. Isolated lateral cystoceles are more common and result from weakness or disruption of the lateral attachments of the vesicopelvic or anterior cardinal ligaments to the pelvic sidewall without significant weakness of the central support. Urethral hypermobility is commonly associated with lateral fascial defects. Combinations of central and lateral fascial defects are the most common presentation of cystocele and may result in severe degrees of prolapse.

FIG. 44-2. (A) Coronal diagrammatic view of the normal support of the anterior vaginal wall. Note the M shape of the anterior vaginal wall. In patients without pelvic relaxation, the middle portion of the M is shallow, indicating good central support, and the upper portions of the M are adjacent to the pelvic sidewall (tendinous arch of the obturator), indicating good lateral support. (B) In isolated central defects, the middle portion of the M sags prominently in the midline, but the lateral support is well maintained. (C) In isolated lateral defects, the upper portions of the M become displaced inferiorly away from the pelvic sidewall, but the central support is unchanged. (D) When both central and lateral defects are present, the entire M sags inferiorly.

DIAGNOSIS
Patients with cystoceles may be asymptomatic, may have associated stress urinary incontinence, or may complain of an introital bulge or a sensation of a mass in the vagina. Large cystoceles may cause kinking of the bladder neck and urethra, resulting in obstructive urinary symptoms, incomplete emptying, and, less commonly, frank urinary retention. Severe cystoceles may result in obstructive hydronephrosis and renal failure from urethral and ureteral obstruction. Rarely a cystocele may present as dyspareunia or incontinence during sexual intercourse. On physical examination, cystoceles appear as a midline mass in the vagina anterior to the cervix or vaginal cuff (if a hysterectomy has been performed). If significant pelvic relaxation is present, a cystocele may be present outside the vaginal introitus at rest or may bulge outside the introitus with Valsalva maneuver. It is important to differentiate cystocele from the other manifestations of pelvic prolapse that also may present with masses protruding from the vagina, including enterocele and rectocele. Examination with a half speculum will locate the vaginal cuff. Significant enteroceles and rectoceles will usually present posterior to the vaginal cuff or cervix. If there is doubt, cystography can be of assistance. Lateral fascial defects causing cystoceles are identified in those patients with significant urethral and anterior vaginal wall hypermobility on stress maneuvers during physical examination. On examination with a half speculum directed posteriorly reducing any coexisting enterocele or rectocele, these patients have hypermobility of the entire anterior vaginal wall on stress with laxity of support for the anterolateral vaginal wall. Isolated central fascial defects are rare and result from attenuation of the perivesical fascia in the midline and separation of the cardinal ligaments without evidence of lateral fascial weakness. Clinically these patients have a distinct anterior midline vaginal bulge without coexisting loss of support for the urethra and anterolateral vaginal wall. Typically there is no urethral hypermobility, as the lateral support of the urethra is unaffected. Combined lateral and central defects are common and can result in urinary obstruction with or without incontinence when severe. It is important to ascertain the level of sphincteric competence preoperatively in any patient undergoing cystocele repair. Clearly, if the patient is incontinent, some type of procedure to increase outlet resistance is needed in addition to cystocele repair. 2 However, many patients presenting with cystoceles may not have associated symptoms of stress incontinence. In these patients sphincteric incompetence may be masked by the valvular effect of the cystocele. If the cystocele is repaired in isolation, the protective valvular effect of the cystocele on the urethra and outlet will be lost. In this case, despite adequate repair of the cystocele, the unsuspended urethra will remain in a low-lying unprotected position, and the patient will have a substantial risk of postoperative incontinence. Therefore, any degree of urethral hypermobility in the presence of a cystocele, with or without urinary incontinence preoperatively, should be repaired at the time of surgery by simultaneous suspension or sling. Finally, a cystocele with a well-supported, nonmobile urethra from a previous suspension in combination with poor emptying ability may signal urethral obstruction. This may be caused by urethral obstruction (from the previous suspension or the cystocele or both) or detrusor hypocontractility. Patients with urethral obstruction from previous surgery may require urethrolysis and resuspension in addition to cystocele repair in order to avoid postoperative urinary retention. Patients with poor detrusor contractility should be alerted to the high risk of long-term intermittent catheterization postoperatively. Careful preoperative urodynamics with the cystocele reduced will help sort out these situations.

INDICATIONS FOR SURGERY
The repair of cystocele is based on several factors: the presence or absence of urinary incontinence, the grade of the cystocele, the inherent pathophysiological fascial weakness (central or lateral), emptying ability, and the associated vaginal or abdominal pathology to be repaired (uterine prolapse, enterocele, rectocele, etc.). Asymptomatic, small cystoceles with no evidence of stress urinary incontinence (SUI), urinary obstruction, and absence of other manifestations of pelvic prolapse do not require surgical repair. Small grade I and II cystoceles resulting from lateral fascial defects associated with SUI are usually adequately repaired using the Raz vaginal wall sling described in this text and elsewhere. 5 Cystoceles resulting from an isolated central fascial defect without concomitant SUI, urethral hypermobility, or demonstrated sphincteric incompetence with the cystocele reduced can undergo a central fascial defect repair alone. It should be noted that this is an uncommon presentation, and this procedure is rarely performed in isolation at our center. In patients with a moderate cystocele resulting from a lateral fascial defect, we utilize a six-corner bladder suspension. This group of patients has, by definition, associated urethral hypermobility as a result of the associated anatomic fascial defect. The urethral hypermobility as well as the cystocele will be corrected by the six-corner suspension. This operation is a modification of our previously described four-corner suspension. 3 This evolution came about from our improved understanding of the importance of the midurethral complex in the maintenance of continence in many women. 7 In patients with a severe cystocele (grade IV) and both a lateral and central fascial defect, a combined lateral and central fascial defect repair with mesh and vaginal wall sling will be performed. This is also a modification of a previously described procedure now redesigned to incorporate the midurethral complex into the repair. 4

ALTERNATIVE THERAPY
With rigorous physical therapy and intensive pelvic floor rehabilitation, some small cystoceles can be eliminated by strengthening the pelvic floor musculature. The addition of oral or topical estrogens may augment the response to nonsurgical therapy in those patients who are poorly estrogenized. Larger cystoceles without significant urinary obstruction but large enough to be bothersome to the patient can be reduced and treated with a pessary. This is also effective in those patients whose coexisting medical illness precludes surgery. Commonly used surgical alternatives to the transvaginal procedures described below include various transabdominal procedures including the Richardson para-vaginal repair, the Burch colposuspension, and the Marshall–Marchetti–Krantz (MMK) repair. Transabdominal retropubic procedures such as the paravaginal repair and the colposuspension are indicated when the presence of other intra-abdominal pathology (large uterine leiomyomas requiring abdominal hysterectomy, ovarian pathology, etc.) requires concomitant surgical exploration. It should be noted that these procedures do not address central fascial defects and are useful only for repairing isolated lateral fascial defects. The use of retropubic procedures to repair cystoceles resulting from central defects may actually aggravate the condition, as the lateral tension placed on the suspending sutures may create increased midline fascial separation through additional shearing forces on the already weakened central fascia. The MMK should never be utilized for the repair of cystocele, as the sutures are placed too medially over the urethra to have a significant impact on the

intrinsic fascial defect causing the cystocele. Simultaneous abdominal and vaginal approaches may be necessary in some cases with combined pathology. Transvaginal approach to cystocele repair has several distinct advantages over the abdominal approach. Vaginal incision is associated with less postoperative pain and discomfort and a faster return to regular activities. Coexisting vaginal pathology such as rectocele, enterocele, and vault prolapse are easily repaired through the same or a slightly extended incision. Finally, both the lateral and central fascial defects resulting in the appearance of the cystocele are completely isolated and repaired under direct vision with a vaginal approach.

SURGICAL TECHNIQUE
Combined Repair of Lateral and Central Defects with Mesh and Vaginal Wall Sling This procedure will repair the lateral and central fascial defects as well as the associated urethral hyper-mobility. The lateral fascial defect is repaired by non-absorbable sutures placed through the ligamentous supports of the bladder and urethra and then suspended to the anterior rectus fascia. The vaginal wall sling is accomplished by these same suspension sutures, thus repairing the urethral hypermobility. Finally, the central defect is repaired by reapproximating the cardinal ligaments in the midline and then placing several interrupted sutures plicating the perivesical fascia from the bladder neck to the level of the cardinal ligaments. Preparation Most vaginal surgery at our institution is performed on an outpatient basis. Antibiotics are administered parenterally 1 hour before incision. General anesthesia is preferentially used in all our vaginal surgery unless medically contraindicated. The patient is brought to the operating room and placed in the dorsal lithotomy position with candy-cane stirrups. The buttocks are placed just off the end of the operating room table. All pressure points are padded, and care is taken to ensure that no lower extremity joint is flexed more than 90 degrees. The lower abdomen and perineum are shaved. A povidone/iodine vaginal scrub and painting are performed. The anus is draped out of the field with a self-adherent clear plastic drape. The remaining drapes are secured with silk suture across the perineum to ensure separation of the fecal and urinary streams. A weighted vaginal speculum is placed, and labial retraction sutures of 3-0 silk are used for maximal exposure. A Lowsley retractor is placed per urethra and supported at the meatus. The tip of the Lowsley is directed anteriorly, and a suprapubic incision is performed approximately two fingerbreadths cephalad to the superior margin of the symphysis pubis in the midline. The incision is carried sharply down onto the tip of the Lowsley, and the Lowsley is extruded through the anterior abdominal wall. A 16-Fr Foley catheter is grasped and brought into the bladder and confirmed in good position by irrigation with a Toomey syringe. An additional 16-Fr Foley is placed per urethra. If indicated, a vaginal hysterectomy is now performed. Goalpost Incision A ring retractor is placed, and the hooks are used to expose the introitus. The anterior vaginal wall overlying the cystocele is grasped and everted through the introitus. Infiltration with injectable saline is performed along the anterior vaginal wall in the line of a goalpost-shaped incision ( Fig. 44-3A). The limbs of the goalpost are slightly obliqued and are located on the anterior vaginal wall 1 cm from the reflection of the lateral walls of the vagina. The obliqued limbs of the goalpost extend from the midurethra to just beyond the bladder neck. The proximal extent of the paired oblique incisions are connected across the midline under the bladder neck, and then a single incision is carried to the level of the vaginal apex in the midline. The preserved island of tissue beneath the proximal urethra and bladder neck will be used for the placement of the sling sutures.

FIG. 44-3. (A) The incision for the combined lateral and central fascial defect repair is shown. The distal limbs extend to the midurethra and are connected across the midline proximal to the bladder neck. The midline incision is carried to the level of the vaginal cuff. (Adapted from Raz S, Stothers L, Chopra A. Raz techniques for anterior vaginal wall repair. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996.) (B) Anatomy of the anterior vaginal wall after lateral and posterior dissection is completed. The area of the cardinal ligaments should be reapproximated with a suture but not yet tied. An island of vaginal wall is preserved beneath the proximal urethra to be used as the anchoring tissue for the distal sling sutures. (Adapted from Raz S, Stothers L, Chopra A. Raz techniques for anterior vaginal wall repair. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996.) (C) Locations of sutures for repair of the lateral defect and vaginal wall sling. (Adapted from Raz S, Stothers L, Chopra A. Raz techniques for anterior vaginal wall repair. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996.) (D) Location of sutures for repair of the central defect. The mesh is not depicted. The suspension sutures have already been passed and are not seen here. (Adapted from Raz S, Stothers L, Chopra A. Raz techniques for anterior vaginal wall repair. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996.) (E) After cystoscopy, the sutures closing the central defect are tied.

Development of the Vesicovaginal Space The cut edges of the vaginal wall in the midline are grasped with Allis clamps to provide countertraction for the ensuing initial dissection. Sharp, shallow dissection is then carried out laterally, using Metzenbaum scissors, to develop the vesicovaginal space. This plane is avascular, and the vaginal wall should easily separate from the underlying periurethral and perivesical fascia. In reoperative cases, the vesicovaginal space may be difficult to define, and very shallow sharp dissection on the vaginal wall is imperative to avoid inadvertent entry into the bladder. Lateral and anterior dissection toward the limbs of the goalpost incision is carried out first. Posterior dissection toward the vaginal cuff in patients who have had previous hysterectomy can be difficult, and extreme care should be taken to avoid inadvertent bladder injury. Initial mobilization of the anterior and lateral vaginal walls will make the difficult posterior dissection toward the vaginal cuff considerably less dangerous because the tissues can be reflected over the surgeon's finger in order to facilitate dissection in the correct plane. Careful attention should be paid to the possible presence of a coexisting enterocele in the region of the vaginal cuff. All dissection should be done under direct vision. The hooks of the ring retractor can be replaced onto the developing vaginal flaps to assist in exposure. Excessive bleeding during the early portion of the dissection indicates entry into an incorrect plane and may signal imminent bladder perforation. Sudden brisk bleeding in the posterior dissection toward the vaginal cuff may indicate severing of a branch of the uterine artery. This can be readily controlled under direct vision with forceps and cautery. However, cautery should be kept to a minimum to avoid tissue devitalization and subsequent development of vesicovaginal fistula. The entire bladder base with attached fascia should be dissected free of the vaginal wall. Anteriorly, the periurethral fascia should be exposed toward the inferior pubic ramus and its attachment to the tendinous arc of the obturator muscle. Posteriorly, the dissection is carried to the level of the vaginal apex in the region of the cardinal ligaments. Once the entire cystocele has been dissected from the anterior vaginal wall, a 2-0 synthetic absorbable suture (SAS) is placed through the area of the cardinal ligaments (Fig. 44-3B) so that it will reapproximate both cardinal ligaments to the midline when tied. This stitch is not tied at this time, as it marks the proximal extent

of the central defect repair. Repair of the Lateral Fascial Defect Attention is now turned to the suspension. The retropubic space is entered with Mayo scissors pointed at the ipsilateral shoulder, perforating the urethropelvic ligament at its insertion onto the tendinous arc of the obturator muscle. The urethropelvic ligament is released from its attachment to the tendinous arc. The adhesions in the retropubic space are lysed bilaterally, leaving the urethra freely mobile. The four suspending sutures (two proximal and two distal sutures) of #1 polypropylene (Prolene) are now placed. The proximal suture individually incorporates three structures: the proximal edge of the released urethropelvic ligament at the bladder neck, the perivesical fascia at the reflection of the dissected vaginal wall midway between the bladder neck and the cardinal ligaments, and, finally, the area of the cardinal ligaments at the level of the vaginal cuff ( Fig. 44-3C). Exposure of the proximal portion of the urethropelvic ligament is aided by placing blunt forceps into the retropubic space and retracting the urethra medially. The suture is then passed through all these structures again. The distal polypropylene suture is passed with helical bites through the midurethral complex, urethropelvic ligament, and the periurethral fascia (anterior vaginal wall excluding epithelium). Exposure of the midurethral complex is facilitated by opening a forceps horizontally within the retropubic space and placing downward traction with the open forceps. This suture incorporates the midurethral complex, including the pubourethral ligament and levator entrance into the urethra, and the distal aspect of the freed urethropelvic ligament and finally is passed parallel to the anterior vaginal wall to include the periurethral fascia but exclude the vaginal epithelium. The identical two sutures are placed on the opposite side. Transfer of Sutures A 1-cm skin incision is performed at the upper margin of the symphysis pubis on the lower abdominal wall. Blunt dissection is carried down to the anterior abdominal wall fascia. A double-pronged ligature carrier is now passed from the suprapubic incision to the vaginal incision under fingertip guidance. The ligature carrier should scrape the posterior surface of the symphysis as it is passed to avoid inadvertent bladder or urethral injury. Piercing the abdominal wall fascia too cephalad off the superior margin of the pubis will result in mobility of the sutures and considerable postoperative pain and discomfort. The ligature carrier is used to transfer the four sutures individually from the vagina to the abdominal incision. They are not tied at this time. These four sutures represent the repair of the lateral defect of the cystocele as well as the vaginal wall sling. Repair of the Central Fascial Defect Absorbable mesh (Dexon or Vicryl) is used to pack and reduce the cystocele cephalad in the midline. This is left in situ but is not sutured in place. Interrupted horizontal mattress sutures of 2-0 SAS are placed in the lateral edges of the perivesical fascia sequentially from the bladder neck to the area of the cardinal ligaments (Fig. 44-3D). These imbricating sutures are placed just medially to the suspending polypropylene sutures into the perivesical fascia but are not tied. Usually four or five sutures are required to close the defect. Cystoscopy The urethral Foley is removed, and cystoscopy is performed to ensure that a polypropylene suture has not been inadvertently placed through the bladder or urethra as well as to confirm efflux from both ureteric orifices. The suprapubic tube location should be confirmed as well. Gentle upward traction on the suspending sutures should elevate the bladder neck and proximal urethra as viewed through the cystoscope. Up to this point the operation is completely reversible, as no sutures have been tied. Thus, if a misplaced suture or incidental cystotomy is detected on cystoscopy, it is easily remedied. Closure of the Vaginal Wall The cardinal ligament suture is now tied, thus reapproximating these structures toward the midline at the level of the vaginal apex or cuff. The sutures repairing the central defect are now tied (Fig. 44-3E). The limbs of the goalpost incision are closed with a running interlocking 2-0 SAS. The excess vaginal wall is trimmed, and the remaining vaginal wall is closed with 2-0 SAS in a running interlocking fashion incorporating the underlying central defect repair, thus closing potential dead space. The vagina is packed with an antibiotic-impregnated gauze. Completing the Suspension The suprapubic sutures are tied with the knot laid down onto the anterior abdominal wall fascia under no tension. A cystoscope sheath should be placed per urethra at a 30-degree incline while the suspending sutures are tied down. In our experience, if the cystoscope sheath maintains elastic mobility after all the sutures are tied, then there is no undue tension on the suspending sutures. The suprapubic incision is irrigated with antibacterial solution and closed with a subcuticular 4-0 SAS. Postoperative Care The SP tube is placed on slight traction and left to gravity drainage. The vaginal pack is removed in 2 to 3 hours. The patient is allowed to void immediately postoperatively. Postvoid residuals are checked every 2 to 3 hours. The patient is discharged from the same-day surgery unit when ambulatory and able to tolerate a regular diet. She is taught suprapubic tube care and how to measure her own postvoid residuals. When the residual is less than 30 to 60 cc, the suprapubic tube is removed. Alternatively, when the postvoid residuals remain high, the patient is taught clean intermittent catheterization techniques, and the tube is removed at the end of 4 weeks. The patient may resume all regular activities immediately postoperatively except heavy lifting, running, and sexual intercourse. These limitations are removed at 4 weeks at the time of the first postoperative office visit. Repair of Lateral Defect (Six-Corner Bladder Suspension) This procedure is best suited for those patients with moderate cystoceles (grade II or III) and primarily lateral fascial defects. Preparation The positioning, preparation, and placement of suprapubic tube are identical to those for the repair described above. Incision The anterior vaginal wall overlying the distal urethra is grasped with an Allis clamp and stretched cephalad. Two oblique vaginal incisions are performed 1 cm medial to the reflection of the lateral vaginal wall onto the anterior vaginal wall, from the midurethra to the region of the vaginal cuff (if a hysterectomy has been performed) or to the paracervical region if the uterus is to be preserved ( Fig. 44-4A).

FIG. 44-4. (A) Oblique vaginal incisions used for six corner bladder suspension. The incision is carried from the mid-urethra to the apex of the vagina. (Adapted from Raz S, Stothers L, Chopra A. Raz techniques for anterior vaginal wall repair. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996.) (B) Location of the polypropylene sutures used in the six corner suspension prior to transfer. (Adapted from Raz S, Stothers L, Chopra A. Raz techniques for anterior vaginal wall repair. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996.)

Exposure of Ligamentous Supports Sharp dissection is carried laterally from both incisions using the Metzenbaum scissors. Proximally, dissection is performed to expose the area of the cardinal ligaments. Distally, lateral dissection is carried over the glistening periurethral fascia, exposing the insertion of the urethropelvic ligament onto the tendinous arch of the obturator muscle. The retropubic space is entered with Mayo scissors pointed at the ipsilateral shoulder and perforating the urethropelvic ligament at its insertion onto the tendinous arc of the obturator muscle. The urethropelvic ligament is released from its attachment to the tendinous arch bluntly. The adhesions in the retropubic space are lysed bilaterally, leaving the urethra freely mobile. Placement of Sutures to Repair the Lateral Fascial Defect Three sutures of #1 polypropylene (Proline) are placed on each side. Each suture incorporates multiple passes through the tissue ( Fig. 44-4B). The proximal suture is passed through the perivesical fascia parallel to the anterior vaginal wall, incorporating the area of the cardinal ligaments at the apex of the vagina but excluding the vaginal epithelium. If passed correctly, this suture should be anchored into very strong, supportive tissue. The middle suture is first placed at the level of the bladder neck and is passed with helical bites through the perivesical fascia parallel to the anterior vaginal wall excluding the vaginal epithelium. The bladder neck is then held medially and the suture is placed through the freed proximal edge of the urethropelvic ligament in a helical fashion at the level of the bladder neck. The third suture is placed at the level of the midurethral complex. Exposure of the midurethral complex is facilitated by opening a forceps horizontally within the retropubic space and placing downward traction with the open forceps. This suture incorporates the midurethral complex, including the pubourethral ligament and levator entrance into the urethra, and the distal aspect of the freed urethropelvic ligament and finally is passed parallel to the anterior vaginal wall including the periurethral fascia but excluding the vaginal epithelium. Transfer of Sutures A 1-cm skin incision is performed at the upper margin of the symphysis pubis on the lower abdominal wall. The ligature carrier is used to transfer the six sutures individually from the vagina to the abdominal incision in the same manner as described for the combined repair. They are not tied at this time. Cystoscopy The urethral Foley catheter is removed, and cystoscopy is carried out examining for intravesical or intraurethral suture as well as proper location of the suprapubic tube at the dome. Urinary efflux should be confirmed from both ureteral orifices. Closure of the Vaginal Wall The oblique incisions in the anterior vaginal wall are now closed with a running interlocking 2-0 SAS with care taken not to trap the polypropylene suspension sutures in the closure. Completing the Suspension The suprapubic sutures are tied under no tension as described previously, and the abdominal incision is irrigated and closed. Postoperative Care The vaginal packing is removed in 2 to 3 hours, and the patient is discharged home when ambulatory and able to tolerate a regular diet. The remaining postoperative care is identical to that previously described for the combined repair. Repair of a Central Defect This procedure is specifically indicated only for the repair of isolated central fascial defects with a well-supported, competent, nonobstructed sphincteric mechanism. This is an uncommonly performed procedure, as most central defects are accompanied by other manifestations of pelvic prolapse. Preparation The positioning, preparation, and placement of a suprapubic tube are identical to those for the repairs described above. Incision and Dissection of the Anterior Vaginal Wall The anterior vaginal wall is infiltrated with injectable saline in the midline from the bladder neck to the apex of the vagina ( Fig. 44-5A). The anterior vaginal wall overlying the cystocele is then incised sharply with the knife. The cut edges of the vaginal wall are grasped with Allis clamps to provide countertraction for the ensuing initial dissection. Sharp dissection is then carried out laterally, using Metzenbaum scissors to expose the vesicovaginal space and the perivesical fascia. This plane is avascular, and the vaginal wall should easily separate from the underlying periurethral and perivesical fascia. Lateral flaps of vaginal wall are developed from the midurethra to the vaginal apex. A ring retractor can be placed to assist in the retraction of the developing flaps.

FIG. 44-5. (A) Incision for the repair of an isolated central defect. (B) Anatomy of the anterior vaginal wall after completion of lateral and posterior dissection of the vaginal wall. (Adapted from Raz S, Stothers L, Chopra A. Raz techniques for anterior vaginal wall repair. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996.) (C) Locations of the sutures for the repair of the central defect including the reapproximation of the area of the cardinal ligaments. (D) The sutures closing the central defect are tied.

Lateral dissection is carried out until the weakened perivesical fascia found in the midline—the cause of the anatomic defect allowing for the formation of the cystocele—is no longer attenuated ( Fig. 44-5B). This strong lateral aspect of the perivesical fascia will constitute the tissue for the subsequent central defect repair. Proximal dissection toward the vaginal cuff in patients who have had previous hysterectomy can be difficult, and extreme care should be taken to avoid inadvertent bladder injury. Dissection should be carried proximally to the area of the cardinal ligaments. Repair of the Central Defect Repair commences with the reapproximation of the area of the cardinal ligaments in the midline with a 2-0 synthetic absorbable suture. Attention is now turned to the perivesical fascia lateral to the bladder neck on either side. A horizontal mattress stitch of 2-0 synthetic absorbable suture is used to plicate this fascia toward the midline. When tied, this stitch will draw the perivesical fascia beneath the bladder neck. The cystocele is reduced manually or, as we prefer, using absorbable Vicryl or Dexon (polyglycolic acid) mesh. The cystocele is thus packed cephalad, and the mesh is left in situ. The remaining perivesical fascia is then plicated from the level of the bladder neck to the cardinal ligaments with interrupted closely approximated horizontal mattress sutures of 2-0 synthetic absorbable suture ( Fig. 44-5C). The sutures are not tied at this time. Cystoscopy The urethral Foley is removed and cystoscopy is performed to ensure that a suture has not been inadvertently placed through bladder or urethra as well as ensure efflux from both ureteric orifices. Closure The plicating sutures are now tied ( Fig. 44-5D). The excess vaginal wall is trimmed, and the vaginal wall is reapproximated with a running 2-0 absorbable suture. An antibiotic impregnated vaginal packing is placed. In 4 to 6 hours the vaginal packing and Foley are removed.

OUTCOMES
Complications Complications associated with cystocele repair can be avoided by careful attention to detail during dissection of the cystocele and passage of the ligature carrier. Incidental cystotomy occurs rarely and should be repaired intraoperatively. A multiple-layer closure with nonopposing suture lines and maximal urinary drainage should be performed to prevent late-onset vesicovaginal fistula. Careful cystoscopic evaluation intraoperatively should alert the surgeon to many other potential complications. Ureteral obstruction is diagnosed by noting the lack of urinary efflux from the ureteral orifice. Removal and replacement of the offending stitch should suffice for repair. Internal ureteral stenting is considered only if there was extensive trauma to the ureter. Inadvertent intravesical or intraurethral placement of nonabsorbable suture will result in recurrent infections and stone formation. This should be recognized intraoperatively during careful cystoscope examination. Removal and replacement of the suture is easily performed if it has not yet been tied as described above. Postoperative bladder instability is a well-documented complication of cystocele repair. Detrusor instability may result from three sources: continuation of preoperative instability, de novo bladder instability temporally related to cystocele repair, and finally, urethral obstruction as a result of tying the suspension sutures under tension. Preoperative detrusor instability is expected to resolve in over 70% of patients postoperatively. 1 The remaining patients may be treated pharmacologically with anticholinergic agents. De novo instability is treated pharmacologically as well. Nonresolution of de novo instability in the presence of incomplete emptying may indicate urethral obstruction. Tying the sutures with a cystoscope sheath in the urethra under no tension as described is the best way to avoid this complication. Urethral obstruction may require formal urethrolysis or complete takedown of the suspension for resolution. In this situation, resolution of instability can be expected in over 90% of cases.8 Incomplete emptying and urinary retention may also result from poor detrusor contractility unrecognized preoperatively. Long-term clean intermittent catheterization is preferable to indwelling Foley catheterization in these unfortunate cases. Persistent pain, infection, bleeding, recurrent incontinence, vaginal stenosis and/or shortening, vesicovaginal fistula, ureterovaginal fistula, and dyspareunia are also potential complications of cystocele repair. Finally, enterocele may result months to years later from alteration of the pelvic axis and insufficient anatomic reapproximation of the cardinal ligaments to the midline during repair. Plication of the perivesical fascia without reapproximation of the cardinal ligaments during repair of central defects leaves a considerable anatomic defect in the region of the vaginal cuff, allowing for the formation of an enterocele. Results Because these procedures represent relatively new modifications of previous technique, we are still compiling data on the results. Previously we had reported on the four-corner bladder suspension for moderate cystocele, which is the forerunner of the six-corner bladder suspension described above. 3 This procedure did not incorporate the midurethral complex in the form of the vaginal wall sling as it does now. Nonetheless, with the previous technique, 105 of 107 patients with moderate cystocele were successfully treated at a mean follow-up of 2 years. We have also reported on a previous modification of the combined repair of central and lateral defects. 4 Likewise, this procedure did not incorporate the midurethral complex in the repair. In this preliminary study, we reported a 96% success rate for grade 4 cystocele at a follow-up of 34 months. The current modifications were designed to address the midurethral complex and improve postoperative continence in this complex group of patients. CHAPTER REFERENCES
1. McGuire EJ. Abdominal procedures for stress incontinence. Urol Clin North Am 1985;12:395-402.

2. 3. 4. 5. 6. 7. 8.

Raz S. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996;338–357. Raz S, Klutke CG, Golomb J. Four corner bladder and urethral suspension for moderate cystocele. J Urol 1989;142:712–715. Raz S, Little N, Juma S, Sussman E. Repair of severe anterior vaginal wall prolapse (grade IV cystourethrocele). J Urol 1991;146:988–992. Raz S, Stothers L, Young GP, et al. Vaginal wall sling for anatomical incontinence and intrinsic sphincter dysfunction: Efficacy and outcome analysis. J Urol 1996;156:166–170. Stothers L, Chopra A, Raz S. Vaginal reconstructive surgery and anterior vaginal-wall prolapse. Urol Clin North Am 1995;22(3):641–655. Stothers L, Raz S. The anatomy of female continence. Paper read at the meeting of American Urological Association Western Section, Scottsdale, AZ, November 5–9, 1995. Webster GD, Kreder KJ. Voiding dysfunction following cystourethropexy: Its evaluation and management. J Urol 1990;144:670–673.

Chapter 45 Transvaginal Enterocele Repair Glenn’s Urologic Surgery

Chapter 45 Transvaginal Enterocele Repair
Victor W. Nitti

V. W. Nitti: Department of Urology, New York University Medical Center, New York, New York 10016.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Presurgical Preparations Surgical Approach Transvaginal Enterocele Repair Enterocele Repair with Vault Suspension Sacrospinous Ligament Fixation Outcomes Complications Results Chapter References

An enterocele is a hernia of the peritoneal pouch of Douglas extending caudally between the vagina and rectum. It usually contains small bowel with or without omentum. Nichols described four types of enterocele 2: Congenital enterocele, the least common, occurs when the anterior and posterior peritoneal folds fail to fuse during fetal development. Pulsion enterocele is caused by a chronic increase in intra-abdominal pressure. Traction enterocele occurs when a pelvic organ, for example the uterus, bladder, or rectum, prolapses and causes traction on the vaginal vault and peritoneum. Iatrogenic enterocele occurs after hysterectomy, when excess peritoneum remains or the pouch of Douglas is not adequately closed. It may also occur after any procedure that alters the vaginal axis. The two types of enterocele that urologists most often encounter are traction enterocele associated with anterior vaginal wall prolapse such as cystocele and iatrogenic enterocele that follows surgery on the anterior vaginal wall. Traction enteroceles often present with a concomitant cystocele and stress incontinence in any patient who has previously had a hysterectomy. It may be part of a total vaginal eversion. Iatrogenic enterocele is not uncommon after surgery for stress incontinence and has been reported in 3% to 17% of cases. 8,10 In order to understand the pathophysiology of an enterocele, one must first consider normal pelvic anatomy. The levator plate provides the primary support for the pelvic organs and directly supports the rectum and vagina ( Fig. 45-1). The distal vagina forms an approximately 45-degree angle with the vertical line while the proximal vagina forms a 110-degree angle and sits almost horizontally over the levator plate. The distal vagina is supported primarily by perivaginal or pubocervical fascia (which is a portion of the levator fascia) and its attachment to the tendinous arch. This proximal portion of the vagina is also anchored along with the cervix over the levator plate by the cardinal and uterosacral ligarnents, which are attached to the tendinous arch and sacrum, respectively. When there is an alteration in this support, commonly seen after hysterectomy, enterocele may occur. Figure 45-2 demonstrates how a change in the vaginal axis after bladder neck suspension may predispose an already weakened pelvic floor to the development of an enterocele.

FIG. 45-1. Normal pelvic anatomy and support after hysterectomy. The horizontal levator plate supports the rectum and vagina proximally and posteriorly before opening at the levator hiatus. The bladder and urethra are located anteriorly. Note the axis of the proximal vagina, which is approximately 110 degrees with the vertical. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

FIG. 45-2. The effect of anterior vaginal wall surgery (e.g., bladder neck suspension, culposuspension) in the face of uncorrected posterior pelvic floor relaxation. The anterior vaginal wall is now well supported while the lax levator plate sags downward. The result is an altered vaginal axis with the posterior vaginal wall and vault no longer supported by the levator plate. These areas are now subject to increases in abdominal pressure and are predisposed to prolapse (i.e., enterocele and rectocele). (From Nitti VW. Transvaginal enterocele repair with variations, Contemp Urol 1994;6:50–64, with permission.)

DIAGNOSIS
An enterocele usually appears as a mass bulging from the vagina. If the uterus is in place, the enterocele will appear posterior to the cervix. However, more commonly it is seen as a bulge from the apex of the vagina after hysterectomy. Larger enteroceles can be seen bulging out of the vagina introitus. A systematic examination of the vagina and pelvis is the first step in the proper diagnosis of an enterocele. The major differential diagnosis is cystocele and high rectocele. An enterocele may exist independently or in combination with these other forms of pelvic prolapse. It is also important to ascertain if the vaginal vault is prolapsed, as this will affect the type of enterocele repair performed.

The extent of prolapse is first evaluated with the patient in the lithotomy position. The posterior blade of a vaginal speculum may be used to retract the posterior vaginal wall to view the anterior vaginal wall. At this time, the presence of urethral hypermobility, stress incontinence, and cystocele may be assessed. The patient should be instructed to cough and strain during these maneuvers. Next, the blade of the vaginal speculum is rotated to retract the anterior vaginal wall. The posterior wall and vaginal vault are inspected. Prolapse of the posterior vaginal wall can be observed readily with the bladder retracted out of the way. If the patient has not undergone a hysterectomy, the uterus can be evaluated for prolapse and movement. In posthysterectomy patients, support of the vaginal vault must be assessed. A rather large enterocele can occur posterior to the well-supported vault, or the entire vagina may be everted. Although large enteroceles are often obvious, smaller ones associated with rectocele may appear as a high continuation of the rectocele bulge in the posterior vaginal wall. An enterocele may distinguished from a high rectocele by bimanual examination. A finger may be placed in the patient's rectum, and she is instructed to cough or bear down. The impulse of the enterocele may be felt against the fingertip as it would during the examination an inguinal hemia. With the index finger in the rectum and the thumb in the vagina, an increased thickness in the rectal vaginal septum may be felt as the enterocele is trapped between the two fingers. This maneuver can be repeated in the standing position (with one foot elevated on a stool) if there is any doubt about the diagnosis. Standing the patient provides a true impression of the degree of prolapse experienced during daily activity. When there is difficulty in distinguishing between cystocele and enterocele, a cystogram with anterior–posterior, oblique, and lateral resting and straining views can be done to help delineate the bladder. Sometimes it is difficult to determine the degree of cystocele and enterocele strictly by physical examination, and this may not be sorted out until the time of surgery.

INDICATIONS FOR SURGERY
Generally, the degree of an enterocele and the amount of discomfort it causes are the indications for surgery. Small enteroceles are often asymptomatic and need not be treated. However, there is a tendency for enteroceles to increase in size over time if left untreated. Larger enteroceles that prolapse outside the vaginal introitus are usually quite uncomfortable. Generally, treatment is driven by patient's symptoms of discomfort, incontinence, obstructive voiding, and constipation. In cases of severe vaginal prolapse, erosion and ulceration of the vaginal wall may occur, causing a great deal of discomfort. When other pelvic surgery is being performed, e.g., stress incontinence surgery, an enterocele of any size should be repaired at the same time, as they are likely to worsen postoperatively. When surgery is contemplated, it is important to consider all of the anatomic abnormalities including vaginal vault prolapse, cystocele, urethral hypermobility with or without stress incontinence, rectocele, and the presence of a uterus. Also, the patient's degree of sexual activity will play a role in the type of surgical procedure performed. Age may also influence the type of procedure being performed. Finally, if the patient might undergo a laparotomy for other pathology, this may influence the type of surgery performed, as a transabdoniinal procedure may be elected.

ALTERNATIVE THERAPY
A nonsurgical alternative for treatment of enterocele and pelvic prolapse is the use of a pessary. Pessaries come in a variety of shapes and sizes and are fit on a trial-and-error basis. Some patients with severe pelvic prolapse are unable to hold the pessary. In others, pessaries are found to be uncomfortable or to cause vaginal infections. A pessary can be used as a temporizing measure until surgery can be performed or as a chronic management of enterocele and pelvic prolapse in patients who do not wish surgical intervention. The type of pessary used is based on the degree of enterocele, presence of a uterus, and other coexisting pathology. Another surgical alternative for enterocele treatment is colpocleisis, in which the entire vaginal canal is closed. This may be elected in the very elderly or in patients who have failed multiple attempts at repair.

SURGICAL TECHNIQUE
Presurgical Preparations Once the patient has elected to have surgery, preoperative preparation is simple and consists of a modified bowel prep, which can be performed at home. The day before surgery the patient can begin on a clear liquid diet and also should take an oral laxative such as citrate of magnesia. The evening before surgery the patient should take a self-administered enema. As in all vaginal procedures, the patient receives broad-spectrum antibiotic prophylaxis perioperatively. We prefer to use gentamicin and ampicillin, or vancomycin in patients who are penicillin allergic. Doses of antibiotics are given just before the procedure and are continued for 24 hours after the procedure. At this time, the patient is switched to a broad-spectrum oral antibiotic for 10 days. Surgical Approach The choice of the specific type of enterocele repair will depend on several factors. We always prefer a transvaginal repair when possible, as this will reduce morbidity and recovery time. In cases in which laparotomy is being performed for another reason, an abdominal approach is preferred. In selecting the type of transvaginal repair, it is important to note the type and extent of the enterocele. If the vaginal vault is prolapsed, this will require a suspension or fixation of the vaginal vault in addition to the repair of the enterocele. We follow the algorithm set forth in Fig. 45-3 in choosing a procedure for enterocele repair. 4,7 This takes into consideration whether vault prolapse exists and also the degree of anterior vaginal wall prolapse or cystocele. If there is no vault prolapse and no cystocele, a simple repair can be performed. In cases in which there is vault prolapse and cystocele, a vaginal vault suspension is chosen. In cases of vault prolapse with no significant cystocele, a sacrospinous ligament fixation is performed. The sacrospinous ligament fixation can also be used in cases of vault prolapse with cystocele; however, we have found the vault suspension technique to be easier and to yield equal or better results in properly selected patients.

FIG. 45-3. Algorithm for transvaginal repair of enterocele based on the presence of vault prolapse and cystocele.

Transvaginal Enterocele Repair All four variations start with the technique of simple enterocele repair, and other procedures may be performed after this if necessary. The patient is placed in the dorsal lithotomy position and prepped, with attention to adequately scrub the inside of the vagina in preparation for surgery. We usually place an iodoform-soaked pediatric laparotomy pad into the rectum so that it can be easily identified by palpation of the posterior vaginal wall. This is especially helpful when a concomitant rectocele repair is to be performed. The labia are retracted with silk sutures. If a cystocele repair is to be performed, we usually place a suprapubic tube at the beginning of surgery either by the Lowsley tractor technique or percutaneously. A Scott ring retractor (Lone Star Medical Corporation) is very useful in helping to expose the operative field. The first step is to isolate, repair, and remove the enterocele sac. This is begun by grasping the enterocele with two Allis clamps and bringing it outside of the vaginal introitus. The vaginal wall is then infiltrated with normal saline to facilitate dissection and separation of tissue planes. A longitudinal incision is made in the vaginal wall

along the entire length of the enterocele ( Fig. 45-4). The vaginal wall is then carefully dissected away from the underlying pubocervical fascia and enterocele sac. In the initial dissection, care must be taken to stay very superficial and develop the proper plane. This is best accomplished by placing the curve of the Metzenbaum scissors against the vaginal wall. A finger can be placed on the outside of the vaginal wall to stabilize the initial dissection. Once the proper plane is entered, it is usually quite easy to dissect the vaginal wall away from the underlying enterocele sac. Care taken here will prevent early entry into the peritoneal cavity. The dissection of the enterocele is continued all the way to the neck of the enterocele sac ( Fig. 45-5). After the enterocele has been completely isolated, the sac is opened, and the peritoneal cavity is entered. At this time, one may see small bowel, omentum, or ovary and fallopian tube in cases where previous hysterectomy without oophorectomy has been performed (Fig. 45-6).

FIG. 45-4. The vaginal wall is grasped with two Allis clamps and is brought outside the vaginal introitus. A midline incision is made. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

FIG. 45-5. Enterocele sac completely dissected to its neck. (From Nitti VW. Transvaginal enterocele repair with variations, Contemp Urol 1994;6:50–64, with permission.)

FIG. 45-6. The enterocele sac is opened, exposing intraabdominal contents. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

The next step is closure of the enterocele defect or pouch of Douglas. Retraction of the peritoneal contents is best performed using a moist pediatric lap pad and a narrow Deaver retractor. This is assisted by placing the patient in Trendelenburg position so that abdominal organs fall slightly cephalad. The enterocele repair begins posteriorly while the abdominal contents are retracted anteriorly using the Deaver. A #1 PGA suture is first placed through the peritoneum and into the prerectal fascia that overlies the rectum ( Fig. 45-7). A circumferential closure of the defect is then performed by placing the pursestring suture laterally in the right in the uterosacral–cardinal ligament complex, anteriorly in the peritoneum, overlying the base of the bladder, laterally on the left in the uterosacral–cardinal ligament complex, and finally again posteriorly in the prerectal fascia ( Fig. 45-8). After this pursestring suture has been placed, a second one is placed in the identical structures in close proximity to the first. Care should be taken to place these sutures deep enough to ensure that adequate vaginal depth can be achieved. After the second pursestring suture has been placed, a third #1 PGA suture is placed from the right to the left uterosacral–cardinal ligament complex. This suture helps to reinforce the repair and also will be left tagged to help identify this complex later should it be necessary. After all sutures are placed, the assistant cinches down and places tension on one of the pursestrings while the surgeon ties the other. After this has been tied, the second pursestring is tied in a similar manner, followed by the uterosacral–cardinal ligament suture. The two pursestring sutures may now be cut while the third is left tagged. The excess enterocele sac may be excised, and the ends oversewn with a 2-0 PGA suture (Fig. 45-9). If only a simple enterocele repair is performed, the tagged suture may now be cut. Excess vaginal wall is then excised, and the vaginal wall is closed with a running 2-0 PGA suture incorporating deep tissue to obliterate any dead space. An antibiotic-impregnated vaginal packing is placed for a period of 24 hours.

FIG. 45-7. A Deaver retractor is used to retract abdominal contents so that pursestring sutures can be placed. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

FIG. 45-8. Placement of pursestring sutures. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

FIG. 45-9. Completed closure of enterocele with excision of excess sac. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

Enterocele Repair with Vault Suspension The vault suspension procedure may be used when the vaginal vault has prolapsed and a cystocele is present. The size of the cystocele will determine the choice of vault suspension. For moderate cystocele (grades 2 and 3) with primarily lateral defects, the four-corner vault suspension and cystocele repair is the procedure of choice. When severe anterior vaginal wall prolapse is present with a grade 4 cystocele and a large central defect of the bladder through the pubocervical fascia, vault suspension with repair of grade 4 cystocele is used. These procedures incorporate techniques of two previously described repairs for moderate and severe cystocele. 5,6 Four-Corner Vault Suspension After simple enterocele repair has been completed, the vaginal wall is left open, and the anterior vaginal wall is further infiltrated with normal saline. An inverted-U incision with the apex halfway between the bladder neck and the urethral meatus is made. The sides of the U are extended proximally to the level of the vaginal cuff and the uterosacral–cardinal ligament complex where the previous enterocele repair had been performed. The vaginal wall is then dissected laterally off the glistening surface of the periurethral fascia and out to the pubic bone in a manner identical to that performed for the Raz bladder neck suspension. Once the pubic bone is reached, the retropubic space is entered with sharp dissection, detaching the urethropelvic ligament from the tendinous arch using a curved Mayo scissors. A finger can then be placed in the retropubic space, and any adhesions bluntly lysed. Two #1 polypropylene suspension sutures are placed on each side. The distal sutures are identical to those for a Raz needle bladder neck suspension. They include two or three helical bites of full-thickness vaginal wall without its epithelium, pubocervical fascia, and urethropelvic ligament at the level of the bladder neck. The proximal sutures incorporate two to three helical bites of full-thickness vaginal wall without epithelium at the level of the vaginal cuff, pubocervical fascia, and the uterosacral–cardinal ligament complex. This complex can be identified by placing tension on the previously placed #1 PGA suture, which had been left tagged ( Fig. 45-10). After these sutures are placed, tension should be placed on them individually to make sure they are in strong tissue. The patient should be able to be moved on the table by pulling on each suture. The procedure is then repeated on the opposite side. Once these sutures have been placed, a stab incision is made in the anterior abdominal wall at the superior border of the symphasis pubis in the midline. A Pereyra–Raz double-pronged ligature carrier (Cook Urological) is placed in this incision and brought through the retropubic space under direct finger guidance. Each of the four suspension sutures is transferred to the anterior abdominal wall individually, as in other needle suspension procedures. Once these sutures have been transferred, gentle tension is placed to make certain that there is reduction of the cystocele.

FIG. 45-10. Four-corner vault suspension. After completion of simple enterocele, an inverted-U incision is made in the anterior vaginal wall, and the retropubic space is entered bilaterally. Two #1 polypropylene suspension sutures are placed on each side, one proximally and one distally, as described in the text. Transferring these sutures to the anterior abdominal wall suspends the bladder neck and base and vaginal vault. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

Cystoscopy is performed to document that there has been no injury to the bladder and that the bladder neck and proximal urethra elevate and coapt nicely. Ureteral injury is ruled out by having the anesthesiologist give intravenous indigo carmine and observing for efflux of blue from each of the ureteral orifices. The vaginal wall is closed with a running interlocking 2-0 PGA suture. After closure of the vaginal wall, the previously placed suspension sutures are tied with minimal tension. Antibiotic-soaked vaginal packing is placed, and the small suprapubic incision is closed with a subcuticular 4-0 PGA suture. Vault Suspension with Grade 4 Cystocele Repair A vertical midline incision is made in the entire anterior vaginal wall after it has been infiltrated with normal saline. This incision will extend from the midurethra all the way through the vaginal cuff and sometimes into the posterior vaginal wall. As described above, the vaginal wall is then dissected from the underlying cystocele, enterocele, and pubocervical fascia. In these cases, it is usually the bladder that is first encountered. The large cystocele is dissected out almost in its entirety. Once

most of the posterior portion of the cystocele has been dissected, the enterocele sac is usually seen. After the enterocele sac is identified, its posterior margin can be dissected off of the vaginal wall. It is extremely important to completely separate the anterior margin of the enterocele sac from the bladder. If this plane is not obvious on initial dissection, it can be seen more easily after the enterocele has been opened. Adhesions of the enterocele sac to the bladder can safely be dissected with a finger inside the enterocele sac. Once the cystocele and enterocele have been completely separated, the retropubic space is entered by perforating the endopelvic fascia as described above for the four-corner vault suspension. The bladder may be reduced and packed up in its normal position with a gauze sponge so that enterocele repair can be performed first. The enterocele sac is opened, and the peritoneal cavity entered. Once the enterocele sac has been mobilized all the way to its neck and the cystocele is adequately reduced, enterocele repair can be performed (as described above for simple enterocele repair). After completion of the enterocele repair, attention is turned to the cystocele. Anterior vaginal wall and vault suspension sutures of #1 polypropylene are placed. The distal sutures include the urethropelvic ligament, pubocervical fascia, and full thickness of vaginal wall without the epithelium at the level of the bladder neck. The proximal sutures are placed in the pubocervical fascia, uterosacral–cardinal ligament complex (again identified by the previously placed tagged suture), and the full thickness of the vaginal wall without the epithelium ( Fig. 45-11). These sutures are transferred to the anterior abdominal wall through a stab incision in the identical manner described for the four-corner vault suspension. Next, the central defect of the cystocele is closed, approximating the attenuated pubocervical fascia in the midline, using a 2-0 PGA suture. During the repair of the central defect, the bladder is kept reduced with either a gauze sponge, which is removed, or PGA mesh, which can be left in place. After completion of the cystocele repair, cystoscopy is performed as above. The excess anterior vaginal wall is closed with a running interlocking 2-0 PGA suture incorporating deep tissue to avoid any dead space. If present, the rectocele is repaired at this time. After closure of the vaginal wall, the suspension sutures are tied with minimal tension. Antibiotic-soaked vaginal packing is placed, and the suprapubic incision is closed with a 4-0 subcuticular PGA suture.

FIG. 45-11. Vault suspension with grade 4 cystocele repair. The enterocele has been repaired through the single midline incision after its sac was completely separated from the cystocele. The attenuated pubocervical fascia is seen laterally with the bladder bulging medially (central defect). Two #1 polypropylene suspension sutures have been placed on each side, as described in the text. These are then transferred to the anterior abdominal wall. Not shown: The edges of the attenuated pubocervical fascia are then brought together in the midline with interrupted 2-0 PGA sutures. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

Sacrospinous Ligament Fixation This technique is used to repair enterocele and vault prolapse in cases in which the anterior vaginal wall is well supported. This commonly occurs after bladder neck suspension or colposuspension. Certainly, one would not want to jeopardize the previous anterior vaginal wall repair by performing a vault suspension. In this case, vaginal depth and axis are restored by posterior fixation of the vaginal vault to the sacrospinous ligaments. The sacrospinous ligament stretches from the ischial spine to the sacrum and is covered by the coccygeus muscle ( Fig. 45-12). This procedure may also be used with cystocele repair, but we prefer the vault suspension in this instance.

FIG. 45-12. Sacrospinous ligament extending from the ischial spine to the sacrum.

After simple enterocele repair is completed, the posterior vaginal wall must be opened far enough distally to facilitate dissection to the sacrospinous ligament. When a simultaneous rectocele repair is to be performed, the entire posterior vaginal wall is opened through the perineum. After the posterior vaginal wall is incised in the midline, it is gently dissected laterally from the underlying prerectal fascia for a short distance. Next, the sacrospinous ligament must be identified. This is done by penetrating the right or left rectal pillar (pararectal fascia) sharply and entering the pararectal space ( Fig. 45-13). Blunt dissection of the pararectal space can be performed with a combination of finger dissection and the use of deep Breisky–Navratil retractors. This dissection is performed until the sacrospinous ligament is palpated and overlying coccygeus muscle is seen. The Breisky–Navratil refractors will help to expose the ligament. Once the ligament is identified, a #1 PGA suture is placed through the ligament and coccygeus muscle complex 2 cm medial to the ischial spine, which is also identified by palpation. It is important to place the suture in this position to avoid injury to the pudendal nerve and vessels, which run just below the ischial spine. It is also important to include the strong ligament in addition to the overlying coccygeus muscle. Tension should be placed on this suture to make certain that it is in the ligament. A second suture should be placed adjacent to the first. Each of these sutures is then placed through the full thickness of the vaginal wall at the level of the dome, approximately 1 cm apart, and left untied. If a rectocele is present, it is repaired at this time. The dome of the vagina can be directed under finger guidance to the deepest possible portion, where it will be fixed. The vaginal wall is then closed with a running interlocking 2-0 PGA suture, and then the previously placed sacrospinous ligament fixation sutures are individually tied. Antibiotic-impregnated vaginal packing is then placed.

FIG. 45-13. Dissection for sacrospinous ligament fixation. The rectal pillars are sharply penetrated, and the pararectal space entered. The space is widened with blunt

dissection to expose the superior surface of the pelvic diaphragm. The sacrospinous ligament can then be palpated, and the coccygeus muscle overlying it can be seen. (From Nitti VW. Transvaginal enterocele repair with variations. Contemp Urol 1994;6:50–64, with permission.)

An intra-abdominal approach to enterocele may also be performed. We usually reserve this for when laparotomy is being performed for other reasons. The abdominal approach described by Moschcowitz is similar to the simple enterocele repair that we use except that the approach is from above. 1 In cases of vault prolapse, a colposacropexy can be performed using autologous rectus fascia or a synthetic mesh. After completion of the surgical procedure, the antibiotic-soaked vaginal packing is left in the vagina until the next morning. Patients receive two to three postoperative doses of intravenous antibiotics before they are switched to broad-spectrum oral antibiotic. In the case of simple enterocele repair or sacrospinous fixation, when no suprapubic tube is used, the oral antibiotic is continued for 7 to 10 days. In cases where a suprapubic tube is left indwelling, antibiotics are usually continued until normal voiding resumes and all tubes are removed. Patients are usually hospitalized for 24 to 48 hours. They may resume light activity on discharge and only restrain from heavy lifting, strenuous exercise, and intercourse for 6 weeks.

OUTCOMES
Complications When the algorithm described above was used for 83 patients, 14% experienced complications. 5 Operative complications included one bladder and one ureteral injury. Most of the delayed complications were minor and included suprapubic wound infection (2.5%), cystocele (1.2%), rectocele (1.2%), flap of excess vaginal tissue requiring excision (1.2%), and chronic suprapubic pain (1.2%). In one patient in whom hysterectomy was not performed with enterocele, uterine prolapse developed, and vaginal hysterectomy was done. Other possible complications that did not occur in our series include small bowel or rectal injury, vaginal shortening limiting the ability to have intercourse, prolonged urinary retention, de novo stress or urge incontinence, and pelvic pain from pudendal nerve entrapment following sacrospinous ligament fixation. Results We used the above algorithm on 83 consecutive patients undergoing enterocele repair. Forty-nine (60%) underwent simple repair, 25 (31%) had vault suspension with enterocele repair (eight had four-corner and 17 had grade 4 cystocele repair), and seven (9%) had sacrospinous ligament fixation. Overall success (no recurrence) was 86%: 82% for simple repair, 96% for vault suspension, and 86% for sacrospinous ligament fixation. A total of 11 patients suffered recurrence at a mean of 11 months (range 4 to 32 months). Two of these occurred after further vaginal surgery, and one after pelvic trauma. Success for sacrospinous ligament fixation has previously been reported to be 62% to 97%. 3,9 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Moschcowitz AV. The pathogenesis, anatomy, and cure of prolapse of the rectum. Surg Gynecol Obstet 1912;15:7. Nichols DH. Types of enterocele and principles underlying choice of operation for repair. Obstet Gynecol 1972;40:257. Nichols DH. Sacrospinous fixation for massive eversion of the vagina. Am J Obstet Gynecol 1982;142:901. Nitti VW. Transvaginal enterocele repair, with variations. Contemp Urol 1994;6:50–64. Raz S, Nitti VW, Bregg KJ. Transvaginal repair of enterocele. J Urol 1993;149:724. Raz S, Klutke CG, Golomb J. Four-corner bladder and urethral suspension for moderate cystocele. J Urol 1989;142:712. Raz S, Little NA, Juma S. Repair of severe anterior vaginal wall prolapse (grade IV cystourethrocele). J Urol 1991;146:988. Raz S, Sussman EM, Erikson DE, Bregg KJ, Nitti VW. The Raz bladder neck suspension: results in 206 patients. J Urol 1992;148:845. Richter K. Massive eversion of the vagina: pathogenesis, diagnosis, and therapy of the true prolapse of the vaginal stump. Clin Obstet Gynecol 1982;25:897. Wiskind AK, Creighton SM, Stanton SL. The incidence of genital prolapse following the Burch culposuspension operation. Neurourol Urodyn 1991;10:453.

Chapter 46 Vaginal Hysterectomy Glenn’s Urologic Surgery

Chapter 46 Vaginal Hysterectomy
Eric S. Rovner, David A. Ginsberg, and Shlomo Raz

E. S. Rovner: Division of Urology, Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania 19104. D. A. Ginsberg: Department of Urology, University of Southern California/Norris Cancer Center, Los Angeles, California 90033. S. Raz: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90024.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preparation Incision Anterior Dissection Posterior Dissection Opening the Peritoneum Division of the Cardinal and Sacrouterine Ligaments Division of the Uterine Vascular Pedicle Incision of the Anterior Peritoneum Division of the Broad Ligament Vaginal Vault Fixation, Closure of theCul-de-Sac, and Peritoneal Closure Closure Outcomes Complications Chapter References

In the United States hysterectomy is the second most commonly performed operation on women following cesarean section. It may be indicated for a variety of gynecologic conditions including symptomatic uterine leiomyomas, endometriosis, carcinoma of the female genital tract, endometrial hyperplasia, and uterine prolapse. Many of these conditions have confounding diagnostic and therapeutic implications outside the realm of urology; thus, in this chapter the discussion is limited to vaginal hysterectomy only as it pertains to the surgical treatment of uterine prolapse. Uterine prolapse is particularly well suited to vaginal hysterectomy as the laxity of the ligamentous support of the uterus resulting in the prolapse allows excellent operative exposure transvaginally. Uterine prolapse may be classified anatomically into four stages based on the position of the cervix relative to the vaginal outlet ( Table 46-1). It is rarely an isolated condition and is more commonly associated with the other manifestations of generalized pelvic relaxation, which may include cystocele, rectocele, enterocele, perineal laxity, and/or urethral hypermobility.

TABLE 46-1. Classification of uterine prolapse

Multiple etiologic factors contribute to uterine prolapse, including congenital, neurologic, racial, social, and coexisting medical factors; however, the most important factor is parity. 1 Stretching of the various paravaginal, parauterine, and paracervical supports during parturition causes significant trauma. Most patients recover from this initial insult; however, with advancing age and loss of estrogens postmenopausally, the effects of the multiparity may manifest as uterine prolapse. Normal uterine support is provided primarily by the sacrouterine and cardinal ligaments ( Fig. 46-1). The cardinal ligaments extend from the upper vagina and cervix laterally to the pelvic sidewall in the base of the broad ligament. The sacrouterine ligaments extend from the posterolateral aspect of the cervix and extend below the peritoneal folds in the pouch of Douglas on either side of the rectum to insert onto the periosteum of the sacrum. These ligaments tether the cervix posteriorly such that the corpus of the uterus lies over the levator plate in its normal anteverted position. Increased intra-abdominal pressure is thus exerted on the posterior surface of the uterus, further assisting in the maintenance of uterine anteversion as the uterus is compressed into the levator plate. The round ligaments provide little support to the uterus but do assist in maintaining the uterus in its anteverted position. Other structures also assisting in support of the uterus include the bony pelvis and the urogenital diaphragm including the central tendon of the perineum.

FIG. 46-1. Diagram of uterine support. (From Raz S. Atlas of transvaginal surgery. Philadelphia: WB Saunders, 1992;4.)

Uterine prolapse results from weakening and laxity of its anatomic supports, including the sacrouterine and cardinal ligaments. With weakness of the sacrouterine ligaments the cervix may shift anteriorly over the levator plate. This may result in a change in the uterine axis as the corpus of the uterus swings backward. In addition, the predominant intra-abdominal forces may now be exerted on the anterior surface of the uterus, resulting in a tendency toward retroversion and further

prolapse. Weakness of the cardinal ligaments allows further loss of support over the levator plate, and uterine prolapse results.

DIAGNOSIS
Uterine prolapse may present as an isolated mass or a bulge in the vagina. It also may be found when the patient is assessed for other urologic complaints such as irritative or obstructive voiding symptoms, urinary incontinence, or retention. A history of pelvic pain, back pain, dyspareunia, or recurrent urinary tract infections may be present. Classically, back pain caused by uterine prolapse is aggravated by standing as the uterus prolapses further through the introitus and is relieved by resuming a recumbent position. The diagnosis is confirmed by physical examination, which reveals significant uterine descent. During pelvic examination, the prolapsed uterus should be assessed for mobility, size, and ligamentous laxity. Associated manifestations of pelvic relaxation, including cystocele, rectocele, and enterocele, should be noted and included in a comprehensive pelvic reconstruction plan when discussed with the patient. Sometimes significant uterine prolapse is not detected by physical examination, and it is not until examination under anesthesia before repair for other associated pelvic prolapse that the degree of uterine prolapse is fully appreciated. For example, in some patients presenting for repair of grade IV cystocele, the full extent of uterine prolapse may not be appreciated in the preoperative office evaluation. It is only when the patient is under anesthesia and fully relaxed that the degree of uterine prolapse becomes evident. Thus, patients who present for repair of grade IV cystocele should be advised of the strong association with uterine prolapse and the potential need for vaginal hysterectomy. With complete uterine prolapse (procidentia), the upper tracts should be assessed for hydronephrosis, as ureteral obstruction can be insidious in onset and yet quite severe. Patients may present with recurrent pyelonephritis, upper tract calculi, or renal failure. This is usually reversible if the hydronephrosis has not progressed to complete renal parenchymal destruction. 3 Ultrasonography of the pelvis may be helpful in patients suspected of having large uterine leiomyomas, as their size may preclude vaginal hysterectomy because of size disproportion.

INDICATIONS FOR SURGERY
The decision to perform vaginal hysterectomy is based on the patient's symptoms, the degree of uterine prolapse, the associated manifestations of pelvic relaxation requiring repair, and the patient's desire to remain fertile. Vaginal hysterectomy is indicated in symptomatic patients with significant discomfort and/or dyspareunia as a result of the prolapsed uterus as well as asymptomatic patients with severe prolapse and urinary obstruction. Patients with moderate or severe prolapse and significant associated cystocele or enterocele should undergo simultaneous hysterectomy and repair of the associated pelvic pathology. Vaginal hysterectomy should not be performed in the presence of significant size disproportion (i.e., large uterus or leiomyomata with stenotic vagina), adnexal or uterine malignant tumor, acute or subacute pelvic inflammatory disease, extensive endometriosis, and/or known obliteration of the cul-de-sac. Confounding factors such as uterine or cervical dysplasia or dysfunctional bleeding should be addressed before vaginal hysterectomy is done, as these situations may mitigate against a vaginal approach. Finally, in patients who desire continued fertility, uterus-preserving procedures should be considered. Vaginal hysterectomy is not indicated in patients with stress incontinence and adequate uterine and pelvic support. Isolated removal of the uterus in these patients will have no impact on continence.

ALTERNATIVE THERAPY
Asymptomatic or mildly symptomatic patients with uterine prolapse may not require any therapy. Minimal degrees of uterine prolapse may respond to Kegel exercises or hormonal therapy. The primary nonsurgical alternative to hysterectomy for significant uterine prolapse involves the use of a pessary, which may be used in patients who are unable or unwilling to undergo surgery. This device is inserted transvaginally and requires suitable perineal support for efficacy. Unfortunately, many of the patients with uterine prolapse have significant perineal laxity, and thus, a pessary may not be effective. Many uterus-sparing procedures have been described as alternatives to vaginal hysterectomy. These include both transabdominal and transvaginal procedures. The uterus may be fixed to the sacrospinous ligament either transvaginally (sacrospinous fixation) or abdominally (abdominal sacral colpopexy). Many other operations have been devised in the past to be expeditious and low risk in order to avoid hysterectomy and its historically high morbidity in the elderly population. These include the Manchester–Fothergill operation and the LeFort procedure. These operations are rarely employed now, as the technique of vaginal hysterectomy has evolved into a simpler operation with low associated morbidity. The Manchester–Fothergill operation involved amputation of the cervix, anterior and posterior colporrhaphy, combined with plication and suturing of the cardinal ligaments to the anterior surface of amputated cervix for uterine support. The LeFort procedure essentially involved excision of rectangular strips of vaginal wall on the anterior and posterior vaginal walls with reapproximation of the denuded areas over the cervix. This procedure was quite expeditious but left the patient with a very shallow vagina that was usually too short for coitus.

SURGICAL TECHNIQUE
Many techniques of vaginal hysterectomy exist. This chapter focuses on the technique that we have utilized with success for many years. Preparation Antibiotics are administered parenterally 1 hour before incision. General anesthesia is preferentially used in all our vaginal surgery unless medically contraindicated. The patient is brought to the operating room and placed in the dorsal lithotomy position with candy-cane stirrups. The buttocks are placed just off the end of the operating room table. All pressure points are padded, and care is taken to ensure that no lower extremity joint is flexed more than 90 degrees. The lower abdomen and perineum are shaved. A povidone/iodine vaginal scrub and painting are performed. A povidone/iodine-soaked pediatric lap sponge is packed into the rectum, and the anus is draped out of the field with a self-adherent clear plastic drape. The remaining drapes are secured with silk suture across the perineum to ensure separation of the fecal and urinary streams. A weighted vaginal speculum is placed, and labial retraction sutures of 3-0 silk are used for maximal exposure. A ring retractor with hooks can be utilized to assist in retraction. If a cystocele repair or bladder neck suspension is planned, a suprapubic tube is placed at this time. A Foley catheter is placed per urethra to empty the bladder. Incision A tenaculum is used to grasp the cervix and evert it through the vaginal introitus. Mobility of the uterus should be confirmed by this maneuver. Normal saline is injected circumferentially around the cervix in order to facilitate dissection. A circumferential incision is performed 1 cm proximal to the cervix, and the anterior dissection is begun. Anterior Dissection Sharp dissection with Metzenbaum scissors is performed from the cervical incision anteriorly in the midline, developing a plane beneath the vaginal wall. Dissection is continued over the cervix, separating it from the posterior bladder wall and perivesical fascia. Care is taken not to dissect laterally. The point of the scissors should remain angled toward the uterus to avoid entry into the bladder. This dissection is aided by gently retracting the anterior vaginal wall and bladder cephalad with a Heaney retractor, especially when a significant cystocele is present. Dissection continues until the vesicouterine peritoneal fold (anterior cul-de-sac) is reached. Posterior Dissection Similar dissection is now carried out posteriorly from the cervical incision in the midline, separating the vaginal wall from the posterior fascia of the uterine cervix ( Fig.
1,4

46-2). Again, a Heaney retractor may assist in visualization of the critical anatomic structures by retracting the vaginal wall and rectum downward. Dissection is continued until the posterior peritoneal fold (posterior cul-de-sac or pouch of Douglas) is identified. If difficulty is encountered in locating the posterior peritoneum, the hysterectomy is begun in an extraperitoneal fashion. The cardinal and uterosacral ligaments can be divided first, thus enabling more mobility of the uterus and easier identification of the posterior peritoneal fold.

FIG. 46-2. Posterior dissection toward the pouch of Douglas. (From Raz S. Atlas of transvaginal surgery. Philadelphia: WB Saunders, 1992, p. 125.)

Opening the Peritoneum The posterior cul-de-sac is entered sharply through a small peritoneotomy. The posterior peritoneum is gently explored digitally for adhesions and masses. A retractor is placed into the peritoneal cavity elevating the cervix and uterus anteriorly. Adhesions in the posterior cul-de-sac are sharply divided. The peritoneum may be tagged at this point for later identification during closure. Division of the Cardinal and Sacrouterine Ligaments The cervix is retracted through the vaginal introitus and slightly laterally to one side. The tip of a large right-angle clamp is placed into the cul-de-sac with the tips directed anteriorly 1 to 2 cm from the cervix. The ligaments are bluntly dissected out and isolated at their point of attachment to the cervix as the right-angle tip is brought from posterior to anterior against the uterus ( Fig. 46-3). The ligaments are individually clamped, divided, and ligated 1 to 2 cm lateral to the cervix using a figure-of-eight suture ligature. The suture ends are left long and are anchored laterally to one of the grooves of the ring retractor. The cervix is now retracted slightly to the other side, and the opposite ligaments are taken in the same fashion.

FIG. 46-3. Retraction on the cervix with division of the cardinal and sacrouterine ligaments. (From Raz S. Atlas of transvaginal surgery. Philadelphia: WB Saunders, 1992, p. 126.)

Division of the Uterine Vascular Pedicle The right angle is then passed again in the same direction slightly higher along the uterus isolating the uterine vascular pedicle on one side. The vascular pedicle is isolated, clamped, divided, and ligated in a similar manner as the ligaments ( Fig. 46-4). The sutures are left long and anchored to the ring retractor. The opposite vascular pedicle is taken in the same fashion. It should be noted that any traction on the cardinal ligaments implies traction on the uterine vessels. This traction will bring the ureters closer to the operative field.

FIG. 46-4. Division of uterine vascular pedicle (From Raz S. Atlas of transvaginal surgery. Philadelphia: WB Saunders, 1992, p. 126.)

At this point the uterus should be markedly mobile. If the uterus is somewhat fixed, confounding factors should be considered such as ventral fixation, endometriosis, adhesions, carcinoma, and size disproportion. Incision of the Anterior Peritoneum The fundus of the uterus is now rotated and everted through the posterior peritoneotomy and out through the vaginal introitus. A finger is passed over the fundus of the uterus to identify the anterior peritoneal reflection. The anterior peritoneum is then entered safely and sharply by incising the peritoneum tented up by the fingertip. Thin peritoneal attachments to the fundus of the uterus are identified by gentle upward retraction on the bladder and divided. Division of the Broad Ligament A Heaney retractor is placed into the anterior peritoneal space and is used to retract the bladder cephalad. At this point, only the broad ligament and its enclosed

structures hold the uterus in place. The uterus is retracted laterally to better expose the broad ligament on one side. A large right-angle clamp is placed across the entire broad ligament next to its insertion into the uterus. Within this clamp lie the utero-ovarian ligament, the fallopian tube, and the round ligament in succession. The broad ligament and its enclosed structures are then divided and suture ligated ( Fig. 46-5). The opposite broad ligament is divided similarly. The sutures are again left long. The uterus is now removed.

FIG. 46-5. Eversion of uterus. (From Raz S. Atlas of transvaginal surgery. Philadelphia: WB Saunders, 1992, p. 127.)

Three pedicles are identified bilaterally: the anterior pedicle is the divided broad ligament; the middle pedicle is the uterine vessels; and the posterior pedicle is the sacrouterine and cardinal ligaments. Vaginal Vault Fixation, Closure of the Cul-de-Sac, and Peritoneal Closure Modified McCall culdoplasty sutures are used to support the vaginal vault and close the cul-de-sac. A #1 synthetic absorbable suture (SAS) is placed through the vaginal wall starting from within the vagina as high as possible on the lateral fornix. The suture is then passed successively through the area of the sacrouterine and cardinal ligament pedicle on the same side and then the prerectal fascia and the sacrouterine and cardinal ligament pedicle on the opposite side. The suture is brought back to the original side, traversing the same structures in reverse order, and finally exiting the vaginal wall 1 cm from the site of entry. An identical suture is placed in the other direction from the opposite fornix. These sutures are not tied down at this time. The peritoneal cavity is closed with two pursestring sutures of #1 SAS incorporating the prerectal fascia, the prevesical fascia, the posterior peritoneal surface of the bladder, and the sacrouterine cardinal pedicle. High placement of the McCall and peritoneal closure sutures ensures adequate vaginal depth on closure ( Fig. 46-6).

FIG. 46-6. Modified McCall culdoplasty and peritoneal pursestring closure sutures with the three pedicles from the hysterectomy.

Closure The broad ligament pedicles are tied to each other across the midline. The sutures on the uterine pedicles are trimmed. Next, the peritoneal pursestring sutures are tied down snugly. If no other vaginal surgery is planned (i.e., cystocele repair, vaginal wall sling, etc.), then the McCall sutures are now cinched down and tied. The vaginal mucosa is trimmed and closed with a running interlocking 2-0 SAS. The vagina is packed with an antibiotic-impregnated gauze. Patients are admitted for 24 to 48 hours postoperatively or until they are able to ambulate and are tolerating a regular diet.

OUTCOMES
Complications Potential complications at the time of surgery include ureteral or bladder injury, bleeding, and rectal or other bowel injury. Injury to the urinary tract may be heralded by the sudden appearance of hematuria in the urinary drainage bag or a sudden gush of clear fluid into the wound. If an injury to the urinary tract is suspected intraoperatively cystoscopy may be performed or, alternatively, the bladder may be filled retrograde through the Foley catheter with indigo carmine and saline until the laceration or injury is seen. If these maneuvers fail to demonstrate the injury, then intravenous indigo carmine is given, and the ureteric orifices are observed cystoscopically for blue efflux. Ureteral injury is much less common after vaginal hysterectomy than after abdominal hysterectomy. This may be because of superior retraction of the anterior vesicoperitoneal fold and thus a higher displacement of the ureters during dissection. 2 Nonetheless, the ureters are most likely to be injured during the case at two distinct junctures: while the uterine vessels are being clamped and divided (the ureters lie just below and lateral to the vascular pedicle) or during pursestring closure of the peritoneal cavity. To avoid ureteral injury during clamping and division of the uterine pedicle, these vessels should be taken very close to the cervix and as distal as possible. During closure of the peritoneum, the ureters lie anterolaterally at the 2- and 10-o'clock positions; thus, the pursestring suture should not incorporate any tissue in the anterolateral peritoneal folds and should be placed rather shallow on the posterior peritoneal surface of the bladder in the midline. Many times, ureteral injury may not become evident until the postoperative period when anuria, fever, flank pain, and/or tenderness may manifest. A high degree of suspicion for ureteral injury should be maintained in this setting with the early use of adjunctive diagnostic studies and operative repair when necessary. Bladder injury may result from retractor injury or from dissection misadventure during the development of the anterior peritoneal fold. This should be repaired at the time of hysterectomy with a multiple layer closure and maximal urinary drainage to prevent the formation of a vesicovaginal fistula. Significant bleeding may result from a laceration of a branch of the uterine artery. This should be controlled carefully with pinpoint accuracy as indiscriminate cautery may result in ureteral injury, tissue devitalization, and potential fistula. Bowel or rectal injury is rare if the uterus is freely mobile and there are minimal intraperitoneal adhesions in the pelvis. Careful, controlled dissection and peritoneal entry as described is the best way to avoid this unfortunate complication. Late postoperative complications include urinary fistulas, vaginal stenosis or shortening, and the appearance of vaginal vault prolapse. A small, asymptomatic vault prolapse is common after hysterectomy and requires no further therapy, as this is usually self-limited. However, significant vault prolapse may indicate that an enterocele was missed at the time of hysterectomy and requires repair.

CHAPTER REFERENCES
1. 2. 3. 4. Chopra A, Stothers L, Raz S. Uterine prolapse. In: Raz S (ed), Female urology, 2nd ed. Philadelphia: WB Saunders, 1996;457–463. Hofmeister FJ, Wolfgram RL. Methods of demonstrating measurement relationships between vaginal hysterectomy ligatures and the ureters. Am J Obstet Gynecol 1962;83:938–948. Mattingly RF, Thompson JD. In: Mattingly RF and Thompson JD (eds.), Te Linde's operative gynecology, 6th ed. Philadelphia: JB Lippincott, 1985;541–567. Raz S. Atlas of transvaginal surgery. Philadelphia: WB Saunders, 1992;121–129.

Chapter 47 Vaginal Repair of Vesicovaginal Fistula Glenn’s Urologic Surgery

Chapter 47 Vaginal Repair of Vesicovaginal Fistula
David A. Ginsberg, Eric S. Rovner, and Shlomo Raz

D. A. Ginsberg: Department of Urology, University of Southern California/Norris Cancer Center, Los Angeles, California 90033. E. S. Rovner: Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania 19104. S. Raz: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90024.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Vaginal Approach (Basic Technique) Adjuvant Techniques for Interposition of Tissue Outcomes Complications Results Chapter References

There are numerous causes for the formation of a fistula tract between the bladder and the vagina ( Table 47-1). In developing countries the primary etiology is prolonged and obstructed labor, but in more developed countries the cause of approximately 90% of vesicovaginal fistula (VVF) is surgical trauma following gynecologic procedures. Total abdominal hysterectomy for benign disease accounts for the majority offistulae secondary to gynecologic surgery. 2 Common nonsurgical causes include advanced local carcinoma (cervical, vaginal, endometrial) and radiation therapy. Risk factors for VVF formation include prior uterine surgery (Cesarean section), endometriosis, infection, diabetes, arteriosclerosis, pelvic inflammatory disease, and prior radiation therapy. 5

TABLE 47-1. Etiology of vesicovaginal fistulae

As the primary cause of vesicovaginal fistulae is iatrogenic, prevention should always be the surgeon's chief goal. Bladder injuries during hysterectomy are 3 times more common with an abdominal versus a vaginal approach. 8 Such injuries can often be avoided with sharp dissection in the appropriate plane and the use of an indwelling catheter during dissection. All attempts should be made to diagnose and repair suspected damage to the bladder at the time of surgical injury. The bladder may be filled to check for leakage or methylene blue or indigo carmine can be administered intravenously to identify a potential site of fistula formation. All identified sites of injury should be appropriately repaired after adequate mobilization of tissues. An important maxim with these injuries is that the first operation to repair a vesicovaginal fistula has the best chance of success.

DIAGNOSIS
The classic presentation of a VVF is continuous daytime and nighttime urinary leakage following a pelvic operation. A patient with a small fistula should still normally void a significant quantity of her urine; larger fistulae do not permit adequate collection of urine in the bladder to permit voiding. Approximately two-thirds of VVF secondary to pelvic surgery are clinically evident within 10 days following the initial injury; radiation-induced fistulae may occur as late as 20 years following therapy. Physical examination, with the aid of a speculum, may determine the source of leakage and can help differentiate a urinary fistula from urinary incontinence of other causes. A fistula site may be identified if pelvic exam is negative by placing a Foley catheter, introducing a methylene blue–tinted solution into the bladder and inspecting the vagina for leakage. If blue-tinged leakage is not apparent and the diagnosis of VVF is in doubt, the sensitivity of this test is increased by placing a vaginal packing and ambulating the patient for a short period. If the vaginal packing remains dye-free with this maneuver, then a ureterovaginal fistula should be ruled out with the use of a clean vaginal packing and intravenous indigo carmine. A modified double dye has also been described to distinguish various vaginal fistulas. 6 Phenazopyridine is systemically administered, blue dye is placed intravesically, and a tampon is placed. After 5 minutes the tampon is examined; an orange stain at the top is consistent with a ureterovaginal fistula, blue in the midportion suggests VVF, and blue at the tip is indicative of urethral leakage, likely due to stress incontinence. Cystoscopy, voiding cystourethrography (VCUG), and an upper tract study should be performed in patients evaluated for a urinary fistula. During cystoscopy the fistula size, the presence of collateral fistulas, and the bladder capacity are assessed; the location of the ureteral orifices in relation to the fistula are noted; and a biopsy of the fistula is done if there is a history of previous pelvic malignancy. VCUG may demonstrate the extent of the fistula as well as associated pelvic prolapse, stress incontinence, or vesicoureteral reflux that may require concomitant repair. Finally, upper tract evaluation with intravenous pyelography or computerized tomographic urography can rule out concomitant ureteral obstruction, suggestive of a ureterovaginal fistula.

INDICATIONS FOR SURGERY
Surgical repair of a VVF is indicated when conservative measures fail. Approximately 10% of posthysterectomy fistula will close with bladder drainage and antibiotics. Greatest success with this treatment option has come with fistulas only a few millimeters in diameter. 1,4 Fistulous tracts that remain open 3 weeks after adequate Foley drainage are unlikely to resolve without surgical intervention.

ALTERNATIVE THERAPY
Cystoscopy with superficial bladder fulguration of the fistula is an option in patients with small, solitary, uncomplicated fistulous tracts. 3,7 If the tract is not free of infection or the vesicovaginal septum is too thin at the time of fulguration, the surgeon risks increasing the diameter of the fistula with this procedure. An alternative to the vaginal repair of VVF is an abdominal approach. Advantages of the vaginal repair of VVF include no abdominal incision, decreased morbidity, a quicker recovery, and avoidance of bivalving of the bladder. We use the abdominal approach when there is associated intraabdominal pathology such as ureteral fistula or obstruction or the need for concomitant augmentation cystoplasty (often seen in patients with radiation cystitis). Abdominal repair of VVF is described in Chapter 27.

SURGICAL TECHNIQUE
There are numerous techniques describing repair of a VVF in the literature. This section discusses the technical aspects of our vaginal approach to a single, uncomplicated VVF as well as modifications for a complicated or radiation-induced fistula. Several techniques for interposing tissue and flaps that can be incorporated into the repair are also described. The timing of VVF repair is somewhat controversial. The classic teaching is to perform fistula repair 3 to 6 months after the initial injury to allow maximum healing and resolution of inflammatory reaction. This is especially important if the repair is to be done through an abdominal approach. However, we routinely perform our repair through a vaginal approach 2 to 3 weeks after the initial injury if conservative therapy fails (i.e., the patient remains wet with a Foley catheter in place and with adequate drainage of the bladder provided) and the patient is in good general health. Retrospective review of the results have not shown a difference when comparing early transvaginal repair with delayed abdominal and vaginal repair. 9 Early repair is contraindicated in patients with infection of the vaginal cuff or pelvis, and prolonged antibiotic therapy is required before reconstruction can be attempted in these patients. Estrogen replacement is begun immediately after a VVF is diagnosed and continued up to the surgery date. Broad-spectrum intravenous antibiotics to cover anaerobes, gram-negative bacilli, and group D enterococcus are administered preoperatively. Vaginal Approach (Basic Technique) Positioning/Preparation/Retraction The patient is placed in the dorsal lithotomy position, a rectal packing is placed (to aid in identification of the rectum if peritoneal flap is to be done in conjunction with the fistula repair), and the lower abdomen and perineum are prepped and draped in the usual standard fashion. Any concomitant anti-incontinence or other vaginal surgery that is to be done simultaneous with VVF repair should be done prior to reconstruction so as not to disturb the repair once completed. A suprapubic tube is placed with the use of the Lowsley retractor through a puncture wound and ureteral catheters are cystoscopically placed if the fistula tract is close to the ureteral orifices. Single-J stents can be used for difficult repairs in an attempt to keep the bladder (and the repair) dry during the early postoperative period. Appropriate exposure is maintained with use of a posterior vaginal weighted speculum, silk labial retraction sutures, and a ring retractor with hooks. Incision The fistula tract is initially dilated with metal sounds until a small catheter can be inserted which can be used for retraction later during the dissection ( Fig. 47-1). Saline is then injected into the anterior vaginal wall surrounding the fistulous tract. An inverted J-shaped incision that circumscribes the fistula tract is made with the long end of the J extending to the apex of the vagina ( Figure 47-2). The asymmetric nature of this incision allows for creation of a vaginal wall flap that can be advanced and rotated over the fistula repair. This helps avoid vaginal shortening as well as overlapping suture lines during reconstruction. If the fistula is high in the vaginal cuff the incision should be inverted, placing the base of the flap distally, facing the urethral meatus.

FIG. 47-1. A urethral Foley catheter is inserted, a weighted vaginal speculum is placed, and an Allis clamp is used to elevate the anterior vaginal wall. The fistulous tract is dilated with metal sounds until a small Foley catheter may be inserted into the tract to aid in dissection. (Adapted from Stothers L, Chopra A, Raz S. Vesicovaginal fistula. In: Raz S, ed. Female urology. 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

FIG. 47-2. Diagram showing the J-shaped incision that circumscribes the fistulous tract for vesicovaginal fistula repair. (Reprinted with permission from Raz S, Little NA, Juma S. Female urology. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology. 6th ed. Philadelphia: WB Saunders, 1992;2782–2828.)

Creation of Vaginal Wall Flaps The vaginal wall flaps are created by dissecting in a proximal, distal, and lateral direction away from the incision ( Fig. 47-3). Each flap is mobilized 2 to 4 cm from the fistulous tract, exposing the perivesical fascia. The ring of vaginal tissue where the initial incision circumscribed the fistula opening is left intact; thus, flap creation is done in healthy tissue, avoiding dissection of the actual fistulous tract ( Fig. 47-4). This technique facilitates dissection in proper tissue planes, avoids bleeding edges at the resected fistula tract (which may require fulguration and the possible conversion of a small fistula defect into a larger defect with use of the electrocautery), ensures that closure of the fistula is done with healthy tissue (vaginal wall flaps), and decreases the risk of potential bladder perforation. In addition, adequate mobilization of the bladder allows for easier construction of a tension-free closure.

FIG. 47-3. Schematic diagram showing the incision in the anterior vaginal wall and dissection of the vaginal wall flaps in an anterior and posterior direction from the fistulous tract. (Reprinted with permission from Stothers L, Chopra A, Raz S. Vesicovaginal fistula. In: Raz S, ed. Female urology. 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

FIG. 47-4. Creation of two flaps on either side of the fistula tract; the ring of the fistula is left intact. (Reprinted with permission from Raz S, Little NA, Juma S. Female urology. In: Walsh PC, Retik AB, Stamey TA, Vaughn ED Jr, eds. Campbell's urology. 6th ed. Philadelphia: WB Saunders, 1992;2782–2828.)

Fistula Closure Closure of the fistulous opening is now done. The intrafistula catheter is removed and the first layer of the repair is placed with closure of the fistula tract with interrupted 2-0 polyglycolic acid sutures placed in a transverse fashion. These sutures incorporate bladder wall and the fistulous tract itself, starting in healthy tissue approximately 2 to 3 mm away from the margin of the fistula (Fig. 47-5). Inclusion of the fistulous tract in the repair (and not resecting the fistula) provides a strong anchor of supporting tissue for the first layer of the repair. The second layer of the repair is placed with interrupted 2-0 polyglycolic acid sutures. These sutures are placed to invert the previous layer by imbricating the perivesical fascia and the deep musculature of the bladder over the first layer/fistula tract ( Fig. 47-6). The sutures should be applied at least 3 to 5 mm from the prior suture line, free of tension, and at a right angle from the first suture line to minimize overlapping of the two lines of repair. The integrity of the repair is confirmed by filling the bladder with indigo carmine.

FIG. 47-5. Schematic diagram showing the first layer of the fistula closure with incorporation of the fistulous tract and the bladder wall in a transverse fashion. (Reprinted with permission from Stothers L, Chopra A, Raz S. Vesicovaginal fistula. In: Raz S, ed. Female urology. 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

FIG. 47-6. Schematic diagram showing the second layer of the fistula closure with incorporation of the perivesical fascia and deep musculature of the bladder. These sutures should be tension-free and placed at a right angle to the first layer as the first layer of closure is inverted. (Reprinted with permission from Stothers L, Chopra A, Raz S. Vesicovaginal fistula. In: Raz S, ed. Female urology. 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

Advancement and Closure of Vaginal Wall Flap The final and third layer of closure is done with the vaginal wall flaps that were previously created. The redundant, excess anterior (distal) vaginal flap is excised and the posterior (proximal) vaginal flap is advanced beyond the fistula closure. This covers the fistula site with fresh, healthy vaginal tissue, which helps avoid overlapping of suture lines ( Fig. 47-7). The flap is advanced at least 3 cm beyond the fistula closure ( Fig. 47-8) and the vaginal wall is closed with a running, locking 2-0 polyglycolic acid suture ( Fig. 47-9).

FIG. 47-7. Schematic diagram showing the first two layers of closure of a vesicovaginal fistula. The outermost layer consisting of the vaginal flap has not yet been closed. (Reprinted with permission from Stothers L, Chopra A, Raz S. Vesicovaginal fistula. In: Raz S, ed. Female urology. 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

FIG. 47-8. Intraoperative photograph demonstrating advancement of the vaginal flap. (Reprinted with permission from Stothers L, Chopra A, Raz S. Vesicovaginal fistula. In: Raz S, ed. Female urology. 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

FIG. 47-9. Advancement and rotation of the posterior vaginal flap covers the underlying two layers of repair with fresh, healthy tissue. (Reprinted with permission from Stothers L, Chopra A, Raz S. Vesicovaginal Fistula. In: Raz S, ed. Female urology. 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

An antibiotic-impregnated vaginal packing is placed for 24 hours postoperatively. The urethral Foley and suprapubic catheters are left to drain for 10 to 14 days. Anticholinergics are given to decrease bladder spasms and oral antibiotics are continued until the catheters are removed. A cystogram is done prior to catheter removal to document the integrity of the repair. Sexual intercourse is avoided for 3 months postoperatively. Adjuvant Techniques for Interposition of Tissue The interposition of healthy tissue during reconstruction is advised when repairing fistulas that are recurrent, located high in the vaginal vault, related to previous radiotherapy, ischemic (obstetric), large, associated with a difficult or doubtful closure, and when there is poor tissue quality secondary to a lack of estrogens or atrophic vaginitis. A flap of peritoneum or a Martius graft can be used when repairing a complex VVF via a vaginal approach; these two techniques are described below. The peritoneal flap is the method we prefer and is what we have primarily used for graft interposition over the past 5 years. Other procedures include gluteal skin flaps and myocutaneous gracilis muscle flaps, which are particularly useful for postirradiation fistulas, in the presence of vaginal atrophy and when no other viable skin source is available. Any one of these techniques enhances the quality of the repair by providing an additional layer of healthy tissue during closure of a complex fistula. Martius Graft The Martius graft, or fibrofatty labial flap, consists of adipose tissue and is the preferential graft for fistulas involving the trigone, bladder neck, and urethra. The blood supply to the graft is provided inferiorly by the posterior labial vessels (off the internal pudendal), superiorly by the external pudendal artery, and laterally by the obturator artery (Fig. 47-10). The lateral blood supply is sacrificed during mobilization of the graft; the graft may be divided at either its most superior or most inferior margin (basing the blood supply on the inferior or superior vascular pedicle, respectively), depending on where the graft will be transferred.

FIG. 47-10. The blood supply to the fibrofatty labial (Martius) graft is supplied by the inferior labial artery inferiorly, external pudendal artery branches superiorly, and obturator artery branches laterally. (Reprinted with permission from Stothers L, Chopra A, Raz S. Vesicovaginal fistula. In: Raz S, ed. Female urology, 2nd ed. Philadelphia: WB Saunders, 1996;490–506.)

Following the closure of the vesical portion of the fistula, the previously placed labial retraction suture is removed on the side where the flap is to be harvested and the ring retractor repositioned to eliminate tension from the labium. A vertical incision is made over the labia majora. The borders of dissection include the labiocrural fold laterally, the labia minora, and the bulbocavernosus muscle medially and Colles' fascia covering the urogenital diaphragm posteriorly. Graft harvest is accomplished in a lateral-to-medial fashion. Dissecting down to the adductor muscles laterally before coming around the width of the Martius graft facilitates the harvest of a thick, fatty segment for graft placement. The entire thickness of the fibrofatty flap is included in a small Penrose drain and gentle downward traction is applied to aid in dissection superiorly. The main vascular supply to the graft is located at the base of the labia majora. The anterior segment is clamped and transected interior to the pubic symphysis. The free segment of the graft is dissected from the underlying structure down to the posterior based vascular pedicle ( Fig. 7-11).

FIG. 47-11. Diagram demonstrating exposure of the fibrofatty tissue of the Martius flap with dissection of the superior and the lateral vascular pedicle. The blood supply off of the inferior pedicle remains intact. (Reprinted with permission from Raz S. Vesicovaginal fistulas. In: Raz S, ed. Atlas of transvaginal surgery. 1st ed. Philadelphia: WB Saunders, 1992;141–165.)

Dissection is carried out from the site of the repair, between the vaginal wall and the perivaginal tissue, to create a tunnel to transfer the graft to the area of the fistula. A hemostat is used to transfer the fibrofatty pad from the harvest site through the tunnel to the vaginal area ( Fig. 47-12). The graft is placed over the fistula repair and secured with interrupted absorbable sutures in a tension-free manner.

FIG. 47-12. The fibrofatty fat pad is transferred from the labia to the vagina underneath the vaginal tunnel. (Reprinted with permission from Raz S. Vesicovaginal fistulas. In: Raz S, ed. Atlas of transvaginal surgery. 1st ed. Philadelphia: WB Saunders, 1992;141–165.)

The vaginal wall flap is advanced over the Martius graft and closed as previously described. A small Jackson-Pratt may be left in the labial incision if the operative field is not completely dry. The labial incision is closed and a pressure dressing may be applied to the labial skin incision. Peritoneal Flap The use of a peritoneal flap during repair of a complex VVF is a simple procedure that does not require extravaginal harvesting of the graft as does a Martius flap. This technique is primarily used in conjunction with repair of a high-lying VVF. Following formation of the vaginal flaps, dissection of the posterior (proximal) vaginal flap is continued into the cul-de-sac. The peritoneum and preperitoneal fat is identified, isolated, and mobilized using sharp dissection ( Fig. 47-13).

FIG. 47-13. Closure of high vesicovaginal fistula with a peritoneal flap. (A) Fistula location with relationships of bladder, vagina, and peritoneum. (B) Closure of fistula with peritoneal flaps.

The first two layers of the fistula are closed as described above. At this point the peritoneal flap is advanced over the fistula repair and secured with interrupted absorbable sutures in a tension-free manner. If a peritoneotomy is made the defect can be closed as the flap is secured to the perivesical fascia over the fistula repair. The vaginal flap is then advanced and closed as previously described, providing a fourth layer of closure. Radiation-Induced Fistula The pathologic changes that lead to formation of a VVF after radiation also result in different strategies when a radiation-induced fistula is repaired. Radiation damage leads to obliterative endarteritis and results in poorly vascularized tissue along the fistula site and the surrounding tissue. Therefore, spontaneous healing of this kind of defect is unlikely. Radiation-induced VVFs are typically found in the trigone. This area of the bladder is fixed and more susceptible to the effects of radiation. Primary repair of a radiation-induced fistula is difficult because the surrounding tissue is often fixed, easy to slough, and nonpliable. Extensive dissection of the fistulous tract combined with augmentation cystoplasty is a successful option, especially in the bladder that is contracted from previous radiation. We prefer to use ileum for augmentation; however, virtually all segments of bowel have been used by various authors for bladder augmentation. Whatever segment is used, the bowel should be nonirradiated, minimizing any compromise to the blood supply that could lead to suture line breakdown and failure of the reconstruction. Finally,

inter-position of healthy tissue (such as omentum or a Martius graft) between the repair is performed, eliminating the possibility of an overlapping suture line and decreasing the risk of recurrent fistula.

OUTCOMES
Complications Early postoperative complications include vaginal infection, bladder spasms, and bleeding. Bladder spasms should be treated with anticholinergics and vaginal bleeding is treated with bed rest and vaginal packing. Late complications include vaginal shortening or stenosis, unrecognized ureteral injury, and recurrence of the fistula. Excessive resection of the vaginal wall causes vaginal shortening or stenosis, and is treated with vaginoplasty. Ureteral injuries are initially managed with a percutaneous nephrostomy drainage; retrograde pyelography and ureteroscopy should be avoided in the early postoperative period as they may result in disruption of the fistula repair. Recurrent fistula may be repaired again through a vaginal approach. This should be done no sooner than 3 months after the previous repair to allow for resolution of postoperative inflammation and is often done in conjunction with adjuvant measures such as a Martius or peritoneal flap. Results With a follow-up of 6 months to 12 years (median 5 years), 93% of our patients were cured with this technique. The majority of these patients (60%) had a previous failed repair of the fistulous tract. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Davits RJ, Miranda SI. Conservative treatment of vesicovaginal fistulas by bladder drainage alone. Br J Urol 1991;68:155. Everett HS, Mattingly RF. Urinary tract injuries resulting from pelvic surgery. Am J Obstet Gynecol 1957;71:234. Geertsen UA, Nielsen HV. Vesicovaginal fistulas with special references to therapeutic possibilities and time. Ugeskr Laeger 1993;155:2046. Kliment J, Berats T. Treatment of vesicovaginal fistulas. Cesk Gynekol 1992;57:267. Kursh ED, Morse RM, Resnick MI, Persky L. Prevention of the development of a vesicovaginal fistula. Surg Gynecol Obstet 1988;166:409. O'Brien WM, Lynch JH. Simplification of double-dye test to diagnose various types of vaginal fistulas. Urology 1990;36:456. Stovsky MD, Ignatoff JM, Blum MD, Nanninga JB, O'Conor VJ, Kursh ED. Use of electrocoagulation in the treatment of vesicovaginal fistulas. J Urol 1994;152:1443. From Vesicovaginal fistulas. In: Mattingly RF, Thompson JD, eds. Te Linde's operative gynecology. 6th ed. Philadelphia: JB Lippincott, 1985;637–667. Zimmern PE, Hadley HR, Staskin DR, Raz S. Genitourinary fistulae: vaginal approach for repair of vesicovaginal fistula. Urol Clin North Am 1985;12:361.

Chapter 48 Female Urethral Diverticula Glenn’s Urologic Surgery

Chapter 48 Female Urethral Diverticula
Kumaresan Ganabathi and Gary E. Leach

K. Ganabathi: Brookville, Punxsutawney, and Clarion Hospitals, Brookville, Pennsylvania 15825. G. E. Leach: Tower Urology Institute for Continence, Cedar Sinai Medical Center, and Department of Urology, University of Southern California, Los Angeles, California 90048.

Diagnosis Indications for Surgery Alternative Therapy Endoscopic Procedures Marsupialization Miscellaneous Techniques Surgical Technique Preparation Diverticulectomy Bladder Neck Suspension (Optional) Postoperative Care Outcomes Complications Results Chapter References

More female urethral diverticula are now being diagnosed than ever before because of a higher index of clinical suspicion and improved diagnostic techniques such as voiding cystourethrogram (VCUG) and transvaginal ultrasound. Because of the complexity and variability of diverticula, thorough evaluation is required to completely assess important pretreatment facors and plan appropriate management. With proper pretreatment evaluation, surgical treatment is generally associated with an excellent outcome. The true incidence of urethral diverticulum is unknown. The reported incidence varies from 1.4% to 5% depending on the population studied. Although diverticula are reported in all age groups, they most commonly present in the third through fifth decades, and 15% to 20% of women with diverticula are nulliparous. Though they are reported to be more common in the black population, the authors have not found any racial predilection. 3 Female urethral diverticulum arises from the wall of urethra and consists mainly of fibrous tissue lined with epithelium. In many cases the epithelial lining may be absent because of chronic inflammation, and the diverticulum may be adherent to the neighboring structures including the periurethral fascia and anterior vaginal wall. Although the exact mechanism of diverticular formation is unknown, the most commonly accepted theory implicates the periurethral glands. Obstruction of the periurethral gland duct is associated with infection in the occluded gland, which results in abscess formation. The abscess subsequently ruptures into the urethral lumen, either as the result of trauma or progression of the infection, forming the diverticulum. The complications of female urethral diverticulum include infection (either acute or chronic), stone formation, and malignancy (adenocarcinoma in 61%, transitional cell carcinoma in 27%, and squamous cell carcinoma in 12% of reported cases). 8

DIAGNOSIS
The presenting symptoms of urethral diverticulum vary considerably and include the three Ds (dysuria, postvoid dribbling, and dyspareunia), urinary frequency/urgency, recurrent infection, urinary incontinence (stress and urge), hematuria, anterior vaginal pain, and swelling (particularly after voiding). The symptoms of urethral diverticulum may mimic those of simple or chronic bacterial cystitis, interstitial cystitis, pelvic inflammatory disease, endometriosis, nonspecific or gonococcal urethritis, carcinoma in situ of the bladder, detrusor instability, and bladder outlet obstruction. The symptoms are related to neither the size nor the number of diverticula, but more likely, the symptoms are dictated by the size and patency of the diverticular opening and the recurrent urinary tract infections associated with the diverticula. Two percent to 11% of urethral diverticula are asymptomatic and found incidentally on routine pelvic examination, radiography such as a postvoid view of an intravenous urography (IVU) for hematuria, or VCUG performed for vesicoureteric reflux, cystocele, or recurrent urinary tract infection. A high index of suspicion for the possibility of urethral diverticulum is essential in making the diagnosis. One should always consider the possibility of a urethral diverticulum in a woman with persistent or recurrent lower urinary tract symptoms that fail to respond to routine treatment. Physical examination of the urethra should be routinely performed by compressing the anterior vaginal wall beneath the urethra and looking for tenderness, mass, or external urethral meatus discharge. In the authors' experience, all urethral diverticula of significant size (>1 cm×1 cm) have been suspected by demonstrating a periurethral mass on pelvic examination. However, all periurethral masses do not represent urethral diverticula; thus, the differential diagnosis of an anterior vaginal wall mass must be considered before the physician assumes thepresence of urethral diverticulum. 2 The differential diagnosis of a periurethral mass includes Skene's gland abscess (located lateral to the urethral meatus), Gartner's duct cyst (located in the anterior lateral vaginal wall), ectopic ureterocele (located beneath the distal urethra and filled with clear fluid), vaginal wall inclusion cyst (spontaneous or after vaginal surgery), urethral carcinoma, periurethral or vaginal fibroma or myoma, hemangioma, urethral varices, endometriosis of the urethra, sarcoma botryoides, and vaginal wall metastasis. Compression of the mass (which is usually tender) may result in pus, blood, or urine extruding from the external urethral meatus. Anterior vaginal wall tenderness may be noticed on palpation (even without obvious swelling), and point tenderness along the urethra may indicate the presence of urethral diverticulum. Induration or hardness in the area of diverticulum suggests the possibility of malignancy or stone. During pelvic examination, other findings such as urethral hypermobility and vaginal prolapse are also considered in order to plan appropriate investigation and treatment. A properly performed VCUG is the best radiographic test to confirm the presence, extent, size, number, and configuration of a urethral diverticulum ( Fig. 48-1).10 In the authors' experience, 60 of 63 diverticula were adequately demonstrated on VCUG. The VCUG should be performed under fluoroscopic control in the standing position to clearly document all the characteristics of the diverticulum. Sometimes, an “air–fluid level” may be seen within the diverticulum because of partial filling, suggesting that the diverticulum is much larger than the portion of the diverticulum seen on the x-ray. Filling defects seen within a diverticulum may suggest the possibility of stone, tumor, or inflammatory mass. The lateral straining views also provide information regarding urethral support and hypermobility, and the presence of stress incontinence with loss of contrast across the bladder neck during coughing or straining.

FIG. 48-1. Voiding cystourethrogram showing the details of a urethral diverticulum.

Occasionally, the diverticulum may be demonstrated on the postvoid film of a standard IVU. However, the area below the pubic symphysis is frequently excluded on the postvoid film. The authors perform an IVU before diverticulectomy to identify an ectopic ureterocele presenting as a periurethral mass because excision of such an unrecognized lesion would result in total incontinence. Retrograde positive-pressure urethrography is used in the investigation of urethral diverticulum in selected cases, when there is a strong suspicion of diverticulum not demonstrated by other methods. Practically, retrograde urethrography is a difficult, time-consuming, and unsatisfactory procedure. Ultrasound examination (vaginal, perineal, translabial, transrectal, suprapubic, or urethral endoluminal 1) has been increasingly used in the investigation of urethral diverticulum. Ultrasound examination not only identifies the contents of the diverticulum but may also show multiloculation within a diverticulum ( Fig. 48-2) or the presence of a second diverticulum that might otherwise have been missed at operation. The ultrasound also helps to differentiate urethral diverticulum from other periurethral masses, and commonly the diverticular orifice can also be visualized ( Fig. 48-2). Given the quality of the ultrasound images, this study should replace positive-pressure urethrography in cases with a normal VCUG when a urethral diverticulum is suspected. Ultrasound is also commonly used to confirm or augment the findings of VCUG. Although they are interesting from a research perspective, the authors do not find CT or MRI scans of the urethra to be either clinically useful or necessary in the investigation of urethral diverticulum.

FIG. 48-2. Multiloculated proximal urethral diverticulum and its opening in through the proximal urethra just below the bladder neck (between + marks) as seen on transvaginal ultrasound (B, Foley catheter balloon; BL, bladder; C, diverticular communication; CU, Foley catheter in the urethra; D, multiloculated diverticulum).

In addition to radiographic and ultrasound studies, complete cystourethroscopy with a 20-Fr female “short beaked” urethroscope is performed with a 0- and 30-degree lens. The regular cystoscope will not expand the urethral wall properly because of its long beak, making diagnosis more difficult. Palpation of the suburethral mass over the instrument allows a better appreciation of the location, size, and consistency of the diverticulum. Also, endoscopic observation of the urethral lumen while the mass is compressed will frequently demonstrate the site of communication as purulent material extrudes into the urethra. Even without the ability to extrude the diverticular contents into the urethra, the communication site usually can be identified with careful observation. It is important to identify the communication site before surgery to completely excise the diverticulum and to prevent recurrence. Another essential aspect of preoperative urethroscopy is to evaluate the competence of the bladder neck and the degree of urethral hypermobility with stress. This endoscopic information is extremely useful to determine whether a suspension procedure should be combined with diverticulectomy. Urodynamic evaluation should be considered before diverticulectomy if the patient has a history of stress or urge urinary incontinence, symptoms of bladder dysfunction such as urinary urgency or frequency, or urethral hypermobility ( Fig. 48-3). The presence of stress incontinence requires a simultaneous bladder neck suspension along with diverticulectomy. The presence of detrusor instability may necessitate prolonged postoperative anticholinergic therapy to prevent “breakdown” of the urethral reconstruction site secondary to high intravesical pressures and to avoid persistent irritative symptoms such as frequency, urgency, and urgency incontinence in the postoperative period. Fluoroscopic examination, if available, should be used along with urodynamic evaluation (videourodynamic studies) to differentiate “paradoxic or spurious stress incontinence” (leakage of contrast from the diverticulum with coughing or straining without any leakage across the bladder neck) from genuine stress urinary incontinence (GSUI). When videourodynamic studies are not available, one must rely on the VCUG to confirm contrast loss across the bladder neck with stress. The urethral pressure profile (UPP) is of only historical interest and has no role in the diagnosis of urethral diverticulum and stress urinary incontinence.

FIG. 48-3. Urodynamic findings in 58 women with urethral diverticulum (DI, detrusor instability; GSUI, genuine stress urinary incontinence).

The authors use the L/NS/C3 (location/number/size/configuration, communication, continence) classification system for an accurate description of the diverticula to document all the critical pretreatment factors before diverticulectomy ( Table 48-1).6 This classification also allows the authors to compare different patients before and after operations and to compare the results in different series as well.

TABLE 48-1. Preoperative classification of female urethral diverticulum (L/N/S/C3) in the authors' experience of 63 women

INDICATIONS FOR SURGERY
The authors recommend urethral diverticulectomy if the patient is symptomatic and/or the diverticulum is of significant size (more than 0.5 cm in diameter). The authors prefer transvaginal excision of diverticula with a vaginal flap technique, which provides excellent exposure and complete excision of the diverticulum with minimal risk of recurrence. A three-layer closure is performed, and overlapping sutures are avoided to minimize the risk of urethrovaginal fistula or recurrent diverticulum. A Martius labial fat pad graft interposition is also used between the urethra and vaginal wall during wound closure after diverticulectomy in selected cases. 5 The indications for the Martius graft include fibrotic and scarred tissues, history of radiation treatment, absent or tenuous periurethral fascia for the second layer of closure, and recurrent diverticula. Along with the vaginal flap technique, simultaneous bladder neck suspension can be performed ( Fig. 48-4) without fear of infection spreading into the retropubic space. To prevent infection, preoperative antibiotics are given, the transvaginal needle suspension is performed before the diverticulum is manipulated, and manual compression of the diverticulum is avoided during surgery. The indications for concomitant bladder neck suspension are documented GSUI, urethral hypermobility, and large proximal urethral diverticulum (which may make the placement of the bladder neck suspension sutures in the anterior vaginal wall difficult without entering the diverticulum).

FIG. 48-4. Inverted-U incision and placement of suspension sutures in the anterior vaginal wall at the level of bladder neck prior to manipulation of the diverticulum.

Pubovaginal sling procedures have also been performed along with diverticulectomy when the patients had either type II or type III stress urinary incontinence. 9 The authors believe that it is risky to use a fascial sling over a delicate urethral closure site for fear of urethral erosion, especially in the absence of type III stress urinary incontinence or intrinsic sphincteric deficiency (ISD). In addition, the sling procedure is associated with postoperative urinary retention, and the authors are concerned regarding the potential need for self-catheterization through the “reconstructed urethra.” Occasionally, a female urethral diverticulum presents as a large periurethral abscess that may not respond to antibiotics. The abscess may be initially drained with incision, and diverticulectomy is performed later, after satisfactory treatment of infection.

ALTERNATIVE THERAPY
Endoscopic Procedures Lapides transurethrally “saucerized” the diverticula by opening the diverticula into the urethral lumen with a knife electrode, especially in women who had previous surgical intervention with multiple recurrent diverticula. The authors have successfully performed the saucerization technique in two distal recurrent diverticula. Other techniques include using a 10-Fr pediatric resectoscope and Collins knife or Pott scissors. Endoscopic procedures are useful mainly in diverticula situated in the distal urethra, creating a wide-mouthed diverticulum that is expected to drain freely. When these are used for mid- or proximal diverticula, the risk of urinary incontinence is greater. Endoscopic procedures do not address the need for concomitant treatment of stress urinary incontinence when a combined problem exists. Marsupialization Spence and Duckett described the technique of marsupialization of distal diverticula. It is basically an incision of the urethral floor through the diverticulum to the diverticular orifice. The diverticulum should be inspected thoroughly, and a biopsy should be taken from any suspicious area to exclude malignancy. The epithelium of the vagina and urethra are then co-opted by running an absorbable suture and keeping vaginal packing for 24 hours and a Foley catheter for 2 or 3 days. Though the reported incidence of stress incontinence after marsupialization of a distal diverticulum in experienced hands is as low as 0.3%, this technique should not be considered with proximal or midurethral diverticula. Other complications of this procedure include recurrent diverticula, vaginal voiding, and spraying of urine with micturition. Miscellaneous Techniques Ellick treated diverticulum with incision and packing of the diverticular cavity with oxidized cellulose (oxycel) or Gelfoam. Problems were encountered with a multiloculated diverticulum because the entire diverticulum was not obliterated. Tancer et al. described a partial ablation technique by transvaginally opening the diverticulum and using the sac as a second layer after closing the orifice. Periurethral injection of Polytef paste adjacent to the diverticulum was employed by Mizraki and Bitterman to collapse the diverticulum. Most diverticula contain infected material, and so, the risk of infection and abscess formation with the synthetic material is considerable. There are several other techniques of diverticulectomy, including (a) a vertical vaginal incision, excision of the diverticulum, and closure of the periurethral fascia in vest-over-pants fashion; (b) a two-layer vaginal flap technique; (c) closure of the urethral defect, retaining a portion of the sac and marsupializing it to the vaginal mucosa to prevent extensive subtrigonal dissection; and (d) excision of the urethral floor from the external urinary meatus to the distal diverticulum and closing of the urethra in layers. The authors prefer a three-layer closure with the closures oriented in different directions to avoid overlapping sutures, which could result in urethrovaginal fistula formation. Several techniques have also been described in the literature to define the diverticulum during surgery, including sounds, gauze packing, Foley or Fogarty balloon catheters, ureteral catheters, injecting methylene blue, coagulated cryoprecipitate, or a silicone and rubber mixture, and urethral endoluminal ultrasound. 1,4 Kohom and Glickman have introduced a 7-Fr Foley catheter (with its tip cut off) through the defect of a diverticulum inadvertently opened during diverticulectomy and then inflated its balloon with saline to distend the diverticulum and facilitate continuation of dissection. In the experience of the authors, these diverticulum-defining procedures are rarely required during operation, as a thorough preoperative evaluation almost always reveals satisfactorily the diverticulum location, extent, and communication site.

SURGICAL TECHNIQUE

Preparation At the time of preoperative consent, the patient is informed about the procedure and postoperative management in detail. Possible complications of diverticulectomy such as infection, bleeding, recurrent diverticulum, urethrovaginal fistula, and urinary incontinence are discussed. Vaginal douches and lower abdominal scrubbing are performed by the patient the night before and the morning of operation. Because most diverticula are filled with infected purulent material, perioperative parenteral antibiotics are given the morning of surgery, usually preceded by 1 week of oral suppressive antibiotic therapy. Patients are admitted the morning of surgery. Diverticulectomy After induction of general or spinal anesthesia, the patient is placed in the modified dorsolithotomy position after application of intermittent pneumatic calf compression. The vagina and lower abdomen are prepared and draped with isolation of the rectum from the operative field. The bladder is filled, and a 22-Fr suprapubic Foley catheter is placed as a “safety valve” for bladder drainage using a modified curved Lowsley tractor. Cystourethroscopy is performed to check the position of the suprapubic catheter and to reconfirm the site of urethral communication of the diverticulum. A 14-Fr Foley catheter is inserted transurethrally. Saline is infiltrated in the anterior vaginal wall along the site of incision, which is made in a U-shaped manner with the apex distal to the diverticulum. If a bladder neck suspension is being performed concomitantly, the suspension sutures are placed at this point (see below). The anterior vaginal flap is reflected toward the bladder neck with sharp, spreading dissection using scissors in the correct plane on the shiny white surface of the vaginal wall (Fig. 48-5A). Dissection in the wrong tissue plane (usually too deeply) results in entry into either the periurethral fascia or the diverticulum itself. Premature entry into either structure makes the remainder of the dissection more difficult. Preservation of the periurethral fascia is important to provide a second layer of closure between the urethra and the vaginal wall.

FIG. 48-5. Technique of diverticulectomy. (A) Inverted-U incision and dissection of anterior vaginal wall flap. (B and C) After transverse incision of the periurethral fascia, anterior and posterior flaps are mobilized to expose the underlying diverticulum. (D) Excision of the diverticulum, creating a large urethral defect. (E) Completed closure of the vaginal flap, with avoidance of overlapping suture lines.

The urethral diverticulum is usually quite obvious once the vaginal flap is dissected inferiorly. Next, the periurethral fascia is incised transversely to allow subsequent exposure of the diverticulum beneath this fascial layer ( Fig. 48-5B). The plane between the periurethral fascia and the diverticulum is defined using sharp dissection, away from the midline, with care taken not to dissect too deeply to enter into the diverticulum at this stage. Once the dissection is completed, the periurethral fascia can be opened like “leaves of a book” to completely expose the underlying diverticulum ( Fig. 48-5C). The diverticulum is carefully dissected around until its communication with the urethra can be defined. Rarely, when there is any difficulty identifying the diverticulum or its communication site to the urethra during the operation, urethroscopy is performed, and a probe or curved pediatric sound is passed under visual control from the urethral lumen into the diverticulum for vaginal palpation. The diverticulum is then excised in its entirety with its communication with the urethra and the adjacent urethral wall ( Fig. 48-5D), thus creating a large urethral defect. When the diverticulum is multiloculated, the diverticulum should be opened and inspected to ensure that all intercommunicating pockets are identified and removed. It is very important that all abnormal, weak, and attenuated tissue at the urethral communication site be excised to reduce the risk of recurrent diverticular formation. On the other hand, care must be exercised not to remove an excessive amount of urethral wall, or urethral closure over a 14-Fr catheter without tension may be difficult. The urethral defect is closed vertically without tension, using a running locking 4-0 Vicryl suture starting at the proximal margin. Care is taken to incorporate both the muscular and mucosal layers of the urethral wall into the closure. A watertight closure is essential to reduce the risk of postoperative extravasation. Next, meticulous hemostasis is obtained to prevent hematoma formation and disruption of the suture lines. The periurethral fascia is closed transversely with a running 3-0 Vicryl suture. Care is taken to space the sutures to obliterate any “dead space” beneath the periurethral fascia. When indicated, a Martius labial fat pad graft is harvested and placed between the periurethral fascia and the anterior vaginal wall closure. 5 The anterior vaginal wall is closed with a running 2-0 Vicryl suture. The wound closure is thus completed in three layers: the urethral wall vertically, the periurethral fascia horizontally, and the overlying vaginal wall flap, which covers the underlying suture lines ( Fig. 48-5E). The bladder neck suspension sutures, if placed, are then tied with minimal tension before the suprapubic incision is closed. At the end of the procedure, an antibiotic-soaked vaginal packing is placed, and both the suprapubic and urethral catheters are placed on gravity drainage. Bladder Neck Suspension (Optional) When indicated, a bladder neck suspension is performed after the initial vaginal wall incision. The bladder neck is identified by palpating the urethral Foley catheter balloon under traction. The plane between the vaginal wall and endopelvic fascia is developed laterally at the level of the bladder neck toward the pubic bone. The dissection must be in the correct plane between the vaginal wall and the medial reflection of the endopelvic fascia to avoid excessive bleeding or bladder injury. After the bladder is emptied, the endopelvic fascia is sharply or bluntly perforated laterally, entering the retropubic space. Blunt dissection is carried out with a wiping motion on the posterior aspect of the pubic bone and continued inferiorly to the level of the ischial tuberosity on each side. Blunt dissection is also continued anteriorly on each side to facilitate passage of the ligature carrier. Number-one polypropylene sutures (Ethicon D-6731) are placed in a helical fashion incorporating the vaginal wall as the anchoring tissue at the level of bladder neck (Fig. 48-4). Care is taken not to enter into the diverticulum at this point to avoid spillage of infected material from the diverticulum into the retropubic space. When a very large proximal diverticulum is present, placement of these helical sutures may be impossible without entering the diverticulum. In this situation, bladder neck suspension is deferred. A 3- to 4-cm transverse suprapubic incision is made just above the symphysis pubis, and the anterior rectus fascia is exposed. After the bladder has been emptied, the modified Pereyra ligature carrier is transferred under finger guidance from the suprapubic area, through the retropubic space, into the vagina. Each suspension suture is threaded into the eye of the ligature carrier and pulled to the suprapubic position. Cystourethroscopy is performed to confirm the absence of bladder perforation or suture material inside the bladder, satisfactory ureteral efflux of previously injected intravenous indigo carmine, and adequate elevation of bladder neck with minimal traction on the suprapubic suspension sutures. When the bone fixation technique is used to anchor the bladder neck suspension sutures, the sutures should be passed through the pubic tubercle before the diverticulectomy is performed in order to reduce the risk of infection. In order to facilitate diverticulectomy and closure of the vaginal flap, the suspension sutures are not tied until the end of the operation. Postoperative Care Perioperative parenteral antibiotics are continued for 24 hours, followed by oral antibiotics until the catheters are removed. Belladonna and opium suppositories are given postoperatively until the patient can tolerate oral anticholinergics (oxybutynin and imipramine hydrochloride) to prevent bladder spasm. The vaginal packing is removed on the first postoperative day. Seven to 10 days postoperatively, after a vaginal examination has demonstrated an intact vaginal flap and suture lines, and

following discontinuation of anticholinergics for 24 hours, a VCUG is performed. During the VCUG the urethral catheter is removed, and the bladder is filled with contrast through the suprapubic tube. The urethra is carefully observed fluoroscopically during voiding. Should any extravasation occur (as noted in approximately 50% of patients), the patient is asked to stop voiding, and the bladder is left on drainage via the suprapubic catheter. The urethral catheter is not replaced, and anticholinergics are restarted. The VCUG is repeated 7 to 10 days later. When there is no extravasation on the initial VCUG, the patient is allowed to empty her bladder, and if the postvoid residual urine volume is less than 100 ml, the suprapubic catheter is removed. The suprapubic catheter is removed only after satisfactory bladder emptying is established (postvoided residual urine consistently less than 100 ml). Intermittent self-catheterization is avoided for fear of disrupting the urethral reconstruction site.

OUTCOMES
Complications The urethral diverticulum may recur as a result of incomplete excision of the diverticulum and its urethral communication site. Preoperative evaluation should identify the presence of multiple diverticula and/or communication sites for complete excision during operation. All pockets of a multiloculated diverticulum should also be removed. Care should also be taken to completely close the urethral defect after excision of the diverticulum to prevent formation of a pseudodiverticulum between the layers of closure. The VCUG should be performed postoperatively to rule out extravasation from the urethral closure site before the patient resumes voiding. Recurrence of a diverticulum is generally suggested by the presence of persistent or recurrent symptoms, periurethral mass, tenderness along the urethra, or recurrent urinary tract infection. Care should be taken in evaluating a patient with VCUG after diverticulectomy, as ballooning or irregularity of the urethra may be seen at the site of diverticulectomy and should not be mistaken for recurrence. Recurrent diverticulum may be managed with a second excision procedure with Martius fat interposition. A small distal recurrent diverticulum may also be satisfactorily treated with transurethral “saucerization” or Spence marsupialization procedure. Urethrovaginal fistula is also a well-described complication of urethral diverticulectomy. In the authors' experience the fistula is usually seen when a vertical vaginal incision is used for diverticulectomy with overlapping suture lines. The use of three-layer technique that avoids overlapping closures greatly minimizes the chance of fistula formation. Also, the urethral defect should be closed watertight, with care taken to avoid any tension on the suture line. Use of Martius fat interposition in selected cases (as described before) further reduces the risk of fistula. Urethrovaginal fistula can be repaired transvaginally without excising the fistula, using the three-layer anterior vaginal wall flap technique and interposition of Martius labial fat pad graft 5 (Fig. 48-6).

FIG. 48-6. Labial fat pad (Martius graft) after mobilization passed through medial tunnel from labial area to the fistula site and fixed in position with absorbable sutures.

Bladder injury can also occur during dissection of a large proximal urethral diverticulum extending beneath the trigone, resulting in the formation of vesicovaginal fistula. Instillation of indigo carmine into the bladder and/or cystoscopy should be performed in such cases to rule out bladder injury. Urethral diverticulectomy may be followed by urinary incontinence as the result of persisting GSUI (if simultaneous stress incontinence surgery is not performed), de novo GSUI, recurrent diverticulum with paradoxical loss with stress, urethrovaginal fistula, and/or postoperative de novo detrusor instability. Genuine SUI may develop after diverticulectomy from periurethral dis-section, further compromising the urethral support. Genuine SUI is also common after excision of a large proximal diverticulum extending beneath the proximal urethra and bladder neck area because of the extensive dissection required to remove the diverticulum. Preoperative evaluation is essential to identify individuals who may develop postoperative stress incontinence, which may be prevented with a concomitant bladder neck suspension when diverticulectomy is performed (see indications). Postoperative stress incontinence is evaluated and appropriately treated. 7 Irritative symptoms may be caused by urinary tract infection or bladder dysfunction. Urinary tract infection is treated with appropriate antibiotics. Preoperative evaluations including urodynamic studies are essential to identify candidates likely to develop postoperative irritative symptoms. In our recently published series, 12 women had preoperative detrusor instability during urodynamic studies, and it persisted in four women (33%) after diverticulectomy. 3 Persistent detrusor instability is treated with anticholinergic medications, and postoperative de novo bladder dysfunction requires further evaluation. Narrowing of the urethra is avoided by not excising too much of the urethral wall during excision of diverticular communication site. Sufficient urethral wall is preserved to close over a 14-Fr Foley catheter. Urethral stricture after diverticulectomy may require reconstruction of urethra. The complications of concomitant bladder neck suspensions include bladder injury, ureteral injury, bleeding, nerve injury (secondary to patient positioning or nerve entrapment from lateral fixation of suspension sutures), acute and chronic urinary retention, immediate and late urinary incontinence, detrusor instability, and infection. Results The results of urethral diverticulectomy from published series are summarized in Table 48-2. In our series, urethral diverticulum was diagnosed in 63 women, and urethral diverticulectomy was performed in 56 women (88.9%), including concomitant bladder neck suspension in 27 women (48.2%). 3 With a mean follow-up of 70 months (range 6 to 136 months), 48 women (85.7%) had relief of their presenting symptoms. Two recurrent diverticula (noted at the distal urethral closure site) were documented by periurethral mass and/or tenderness on follow-up vaginal examination. Both diverticula were managed satisfactorily with transurethral “saucerization.” Suprapubic tenderness occurred in one woman in relation to the bladder neck suspension sutures secured suprapubically. This pain resolved with conservative treatment within 6 months of the procedure. Early urinary tract infections were identified and treated satisfactorily in six women, and none of these patients developed recurrent urinary tract infections. Four women who had persistent detrusor instability after surgery noted significant urgency symptoms. No patient developed a retropubic or postoperative wound infection or urethral stricture.

TABLE 48-2. Complications of diverticulectomy from published series compared to authors' series

In this series, the authors did not find any patient with intrinsic sphincteric deficiency (ISD) and did not use a sling procedure with diverticulectomy. 3 The overall continence status in the treatment group of 56 women with a mean follow-up of 70 months ( Table 48-3) may be summarized as follows: totally continent or minimal incontinence (dry or no pad for protection), 45 (80.4%) women; moderate incontinence (1 or 2 minipads/day for protection), ten (17.9%) women, and severe incontinence (several pads/day for protection) in one woman (1.8%) because of detrusor hyperreflexia secondary to cerebellar degeneration.

TABLE 48-3. Continence status of 56 women after urethral diverticulectomy with a mean follow-up of 70 months

Three of 27 patients who had diverticulectomy alone developed postoperative GSUI. Two of them had no GSUI before diverticulectomy, and the other one, as noted before, had preoperative GSUI, but simultaneous bladder neck suspension could not be performed because of the extremely large size of the diverticulum. All three women subsequently underwent bladder neck suspension with a satisfactory outcome. Six women developed recurrent GSUI after diverticulectomy and bladder neck suspension, but none had ISD. One woman in this group had a subsequent Burch colposuspension during hysterectomy, and the other five did not desire further treatment because their incontinence was significantly improved. Female urethral diverticula are more common than previously thought, and the suspicion should always be high in the clinician's mind. It is most important to be aware of this clinical entity in women with recurrent lower urinary tract symptoms such as urinary frequency, urgency, dysuria, postvoid dribbling, hematuria or dyspareunia, and/or urinary tract infections, with or without periurethral mass. An appropriate history coupled with vaginal examination, VCUG, endoscopic examination, and urodynamic studies will facilitate the diagnosis. With the use of the L/N/S/C3 classification all preoperative factors are easily addressed. The three layer vaginal flap technique of diverticulectomy is associated with an excellent success rate and minimal complications. Thus, the authors recommend this procedure as the treatment of choice for female urethral diverticula. In appropriate patients, diverticulectomy can be safely combined with bladder neck suspension without any significant additional risk and with a satisfactory outcome. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Chancellor MB, Lin JB, Rivas DA, Karasick S, Bagley DH, Goldberg BB. Intraoperative endo-luminal ultrasound evaluation of urethral diverticulum. J Urol 1995;153:72. Dmochowski RR, Ganabathi K, Zimmern PE, Leach GE. Benign female periurethral masses. J Urol 1994;152:1943. Ganabathi K, Leach GE, Zimmern PE, Dmochowski RR. Experience with the management of urethral diverticulum in 63 women. J Urol 1994;152:1445. Ganabathi K, Sirls L, Zimmern PE, Leach GE. Operative management of female urethral diverticula. In: Lytton B, Catalona WJ, Lipshultz L, McGuire EJ, eds. Advances in urology, vol 7. St Louis: Mosby Yearbook, 1994;199–228. Leach GE. Urethrovaginal fistula repair with Martius labial fat pad graft. Urol Clin North Am 1991;18:365. Leach GE, Sirls LT, Ganabathi K, Zimmern PE. L/N/S/C3: a proposed classification system for female urethral diverticula. Neurourol Urodyn 1993;12:523. Leach GE, Sirls L. Transvaginal needle suspension. Atlas Urol Clin North Am 1994;2:13. Rajan N, Tucci P, Mallough C, Choudhury M. Carcinoma in female urethral diverticulum: case reports and review of management. J Urol 1993;150:1911. Swierewski SJ III, McGuire EJ. Pubovaginal sling for treatment of female stress urinary incontinence complicated by urethral diverticulum. J Urol 1993;149:1012. Zimmern PE. The role of voiding cystourethroscopy in the evaluation of the female lower urinary tract. In: Paulson DF, ed. Problems in urology, vol 5. Philadelphia: JBLippincott, 1991;23–41.

Chapter 49 Closure of Bladder Neck in the Male and Female Glenn’s Urologic Surgery

Chapter 49 Closure of Bladder Neck in the Male and Female
Scott E. Litwiller and Philippe E. Zimmern

S. E. Litwiller and P. E. Zimmern: Department of Urology, University of Texas Southwestern Medical Center, Dallas, Texas 75235-9110.

Diagnosis Indications for Surgery Female Male Alternative Therapy Surgical Technique Vaginal Approach (Female) Abdominal Approach Female Approach in the Male Perineal Access Abdominal Closure in the Male Vascularized Interposition Postoperative Care Outcomes Complications Results Chapter References

Bladder neck closure (BNC) is an uncommon procedure that has traditionally been reserved as a final alternative for the management of the female patient with neurogenically induced intractable incontinence arising from long-term Foley catheter drainage. 3,7,10 It has also been used in the treatment of nonneuropathic conditions such as traumatic urethral destruction or recalcitrant fistula. BNC in the male is usually reserved for patients with neurogenic bladder or a history of incontinence secondary to trauma or urethrocutaneous fistula failing multiple prior attempts at surgical correction or artificial sphincter placement. With the many other options that exist for the treatment of these complex conditions, there is a limited but distinct role for BNC. Though initially fraught with a high failure rate, 7 patient selection and technical refinements have allowed some authors to achieve a success rate of nearly 100%. 10 This chapter will focus on the technique of both abdominal and vaginal bladder neck closure in the female and abdominal bladder neck closure in the male with emphasis on the principles necessary to achieve both successful and durable results.

DIAGNOSIS
Preoperative evaluation and patient selection is extremely important to the success of BNC. A careful history should include special attention to prior abdominal or pelvic surgeries including prior reconstructive flaps or grafts. During the physical exam it is important to carefully assess the presence of lower extremity contractures that may limit access to the vagina, perineal skin integrity or presence of decubiti, and the potential for intermittent catheterization to be carried out successfully. In the patient with adequate manual dexterity or a reliable caregiver, a catheterizable efferent limb from the bladder may be chosen for postoperative drainage. When intermittent catheterization is not feasible, options for postoperative bladder drainage primarily consist of suprapubic tube or incontinent ileovesicostomy. 6 Study of the upper urinary tract by either ultrasonography or intravenous pyelography is important to exclude hydronephrosis or ureteral obstruction as often the same processes responsible for the patient's incontinence may promote upper tract deterioration. When upper tract deterioration is noted, strong consideration must be given to supravesical diversion or preserving the bladder and lowering intravesical pressures by augmentation cystoplasty. A static or voiding cystogram assists in detecting bladder diverticula, vesicoureteral reflux, and calculi. In the case of urethral fistula or stricture, a retrograde urethrogram or fistulogram can document the nature and extent of the patient's underlying disease. Cystoscopy with biopsy to exclude malignancy is essential for the patient who has been managed for an extended period with an indwelling catheter. The extent of urodynamic evaluation is tailored to the choice of postoperative bladder management. In the patient who desires continent, catheterizable access to the bladder, preoperative urodynamic evaluation of bladder storage parameters such as compliance and detrusor instability (hyperreflexia) must be documented to determine the need for concomitant augmentation cystoplasty. When the bladder outlet is patulous, occlusion of the outlet during urodynamic evaluation can be readily accomplished using gentle traction on an inflated Foley catheter. A sterile urine culture should be documented preoperatively. When it is impossible to completely sterilize the urine, culture-specific preoperative parenteral antibiotics must be administered to ensure adequate tissue levels at the time of surgery.

INDICATIONS FOR SURGERY
Female The indications for bladder neck closure in the nonneurogenic patient are urethral destruction, severe intrinsic sphincteric deficiency that is not amenable to conventional treatment, and urethrovaginal fistula failing prior attempts at repair. Patients suffering from neurogenic incontinence often have intractable leakage from urethral destruction due to the long-term effects of an indwelling urethral catheter. A common indication is the patient with advanced multiple sclerosis with urethral destruction from chronic Foley catheter drainage who is not a candidate for aggressive reconstruction. Though control of incontinence has been achieved by some using a pubovaginal sling, 1,4 many patients with urethral destruction and reduced urethral length are not suitable candidates for this procedure. Likewise, for the female patient who has failed attempts at urethrovaginal fistula closure, BNC with a continent catheterizable efferent channel, incontinent vesicostomy, or suprapubic tube may represent a viable option for management. The vaginal approach is favored in the patient without history of prior radiation who desires suprapubic tube drainage. An abdominal approach is desirable for the patient with a history of radiation in whom vaginal tissues may be poorly vascularized and in whom omental interposition between the bladder neck and vagina is desirable. It is also the approach of choice in the patient who elects for a continent catheterizable efferent limb (made of bowel or appendix), incontinent ileovesicostomy, or has failed a prior attempt at vaginal closure of the bladder neck ( Table 49-1).3,6

TABLE 49-1. Comparison of approaches to bladder neck closure

Male The role of BNC in the male resides in the management of refractory urethrocutaneous or urethrorectal fistulas, and in cases of severe neurogenic or postoperative incontinence (with low outlet resistance) when the artificial sphincter is not an option. It may also be used in the treatment of recalcitrant urethral strictures when reconstruction is impossible or undesired.

ALTERNATIVE THERAPY
Options for local reconstruction in females with severe incontinence or fistula are limited. Though urethral reconstruction with vaginal wall or bowel is an available option, maintaining a urethral outlet that is both patent and continent can prove extremely challenging. Continence following these reconstructive procedures may be provided by autologous or synthetic sling materials, 1,4 injectable bulking agents (collagen or fat), artificial urinary sphincter, or bladder neck reconstruction (Young-Dees). In the male with refractory incontinence or fistula, the artificial urinary sphincter and formal fistula closure are other viable alternatives. Historically, supravesical diversion and ureterosigmoidostomy (nonneurogenic patients) have been advocated for treatment of patients with this severity of incontinence. However, it is our opinion that BNC should be considered before embarking on these more extensive surgical options. BNC not only preserves the bladder but preserves the integrity of the ureterovesical junction, thereby protecting the upper tracts.

SURGICAL TECHNIQUE
The goals of the procedure are the same for both male and female patients regardless of the approach utilized, i.e., wide mobilization of the bladder neck to allow for tension-free closure, multilayer closure of the outlet without overlapping suture lines thereby reducing the opportunity for fistula, interposition of vascularized tissue between the vesical outlet and urethral stump or vagina, and adequate postoperative bladder drainage with a large-bore catheter. Vaginal Approach (Female) The vaginal approach is preferred in the female who desires suprapubic tube drainage, has no history of prior radiation, and is not undergoing a concomitant abdominal procedure. Preoperative preparation includes antibiotics, vaginal douching, enema, and deep venous thrombosis prophylaxis. The patient is placed in high lithotomy position with careful attention to padding of all pressure points and extremities. A Lonestar ring retractor (Houston, TX) is suggested along with a weighted vaginal speculum and headlight to provide maximal vaginal exposure. In the case of a small contracted bladder, the curved Lowsley retractor is employed to place a suprapubic tube.9 The patient is placed in deep Trendelenburg position to displace bowel contents and the curved retractor is introduced through the urethra and directed to the anterior abdominal wall 1 to 2 cm above the symphysis pubis. A small suprapubic incision is made over the tip of the Lowsley, which can be palpated beneath the fascia. The tip of the retractor is then pushed out through the skin incision and a 20-Fr Foley catheter is grasped between the open jaws and delivered back into the bladder. Its intravesical position can be confirmed with cystoscopy or irrigation with normal saline. A waterproof surgical ink pen marks the proposed inverted U-shaped vaginal wall incision. A dilute solution of vasopressin 60 U/100 cm 3 is injected into the periurethral tissues and anterior vaginal wall to facilitate dissection of both the urethra and anterior vaginal flap and reduce local bleeding ( Fig. 49-1).

FIG. 49-1. Proposed incision for closure of the bladder neck.

After a circumscribing incision has been made around the urethral meatus, the broad-based vaginal flap is elevated using sharp scissors dissection ( Fig. 49-2). This flap not only aids in the exposure of the remainder of the urethral dissection but serves as an advancement flap to close over the amputated vesical outlet and interposed labial fat pad. When dissecting in the proper plane, the vaginal wall exhibits a distinctly recognizable white glistening surface. Significant venous bleeding may be encountered when the dissection is carried out too deep and into the venous sinuses of the bladder wall. Sharp dissection is then used to free the urethra from its lateral and anterior fascial attachments. To achieve a tensionfree closure, the bladder neck must be completely detached from its surrounding (mostly anterior) attachments (Fig. 49-3). The pubourethral ligaments are sharply transected and the endopelvic fascia is perforated bilaterally using blunt or sharp dissection. After entering the retropubic space, blunt dissection is used to free the lateral and anterior aspects of the bladder neck. Indigo carmine is given intravenously to aid in visualizing the ureteral orifices and the urethral edges may be trimmed to expose fresh healthy tissues for formal closure.

FIG. 49-2. Elevation of vaginal flap.

FIG. 49-3. Detachment of bladder neck.

The bladder neck is first closed in a vertical fashion with absorbable 3-0 polyglycolic acid (PGA) suture ( Fig. 49-4). The integrity of the closure is checked by filling the bladder by gravity through the suprapubic tube. A second horizontal layer of interrupted 2-0 PGA serves to both imbricate the first layer and transfer the closed bladder outlet to a position high behind the symphysis pubis, thus rotating it into a nondependent position ( Fig. 49-5). This technique not only avoids a dependent closure but also directs the force of bladder spasms away from the vagina, thereby reducing the likelihood of secondary vesicovaginal fistula.

FIG. 49-4. Primary closure with tension-free anastomosis.

FIG. 49-5. (A) Photograph of closure of bladder neck. (B) Lateral view of bladder neck closure.

The use of a Martius flap is recommended to reinforce the bladder neck closure and reduce the risk of fistula. The technique relies on a well-vascularized fibrofatty labial pad (from the labia majora) that is based posteriorly on a labial branch of the internal pudendal artery. 8 The Martius flap is tunneled beneath the vaginal wall and fixed in place over the bladder neck closure with 3-0 PGA sutures ( Fig. 49-6). The vertical labial incision can be closed with absorbable suture over a small suction drain. The vaginal flap is trimmed and advanced to close the vaginal incision with running 3-0 PGA suture ( Fig. 49-7). The suprapubic tube is then irrigated to ensure patency and the vagina is packed for 24 hours with an antibiotic-soaked pack.

FIG. 49-6. (A) Martius flap tunneled beneath labia minora. (B) Intraoperative photograph of Martius flap to close over bladder neck.

FIG. 49-7. Closure of vaginal flap.

Abdominal Approach Female The patient is placed in low lithotomy position to provide continuous access to the vagina. Alternatively, if lower extremity contractures prohibit lithotomy position, supine position may be appropriate. A urethral Foley catheter is placed and an infra-umbilical midline incision made. This incision not only provides excellent exposure but also can be extended for omental harvest or use of bowel for an efferent catheterizable limb. A Pfannenstiel incision may be considered if a chronic suprapubic tube is chosen for postoperative bladder management. The rectus muscles are retracted laterally and the prevesical space (Retzius) is developed bluntly. The peritoneum is retracted superiorly, and a self-retaining retractor (Balfour or Bookwalter) provides exposure of the retropubic space. With the aid of the Foley catheter and its balloon, the bladder neck and urethra are identified. A 2-0 PGA figure-of-8 suture is placed through the distal most aspect of the deep dorsal vein of the clitoris and the proximal urethra. Using electrocautery or sharp dissection, the anterior bladder neck is amputated from the pelvic inlet over the most distal aspect of the Foley catheter. The anterior bladder neck is grasped with traction sutures or Allis clamps and the Foley catheter is identified and delivered into the surgical field. After intravenous administration of indigo carmine, open-ended ureteral catheters may be placed to assist in a safe dissection of the posterior bladder neck (Fig. 49-8). A hand in the vagina can help to identify and maintain the appropriate plane between the posterior bladder neck and the vaginal wall. Using electrocautery or sharp dissection, the posterior bladder neck is freed from the anterior vaginal wall. This division continues until the bladder neck is rolled up and out of a dependent position. The edges of the bladder neck are trimmed to allow approximation of healthy tissues and the ureteral stents are removed. When an incontinent vesicostomy or catheterizable efferent limb is selected for postoperative bladder drainage they may be fashioned at this time. Otherwise, a large-bore (24-Fr) Malecot or Foley catheter is placed through a stab wound in the bladder dome. The bladder neck is then closed in two layers as described for vaginal closure.

FIG. 49-8. Opened bladder with mobilization of bladder neck.

Approach in the Male The technique of bladder neck closure in the male differs from the female in several distinct ways: (a) lack of direct perineal access to the bladder neck; (b) options for vascularized tissue are more limited; and (c) the prostatic anatomy poses a challenge to both intraoperative closure and postoperative care. Perineal Access Perineal approach to BNC, though conceptually and technically feasible, is not considered to be the procedure of choice in the male. Perineal access to the bladder neck necessitates either a concomitant prostatectomy with its own inherent morbidity or closure of the infraprostatic urethra, a procedure associated with a high rate of spontaneous fistulization. Infraprostatic closure of the urethra, though easily performed, is not desirable: 1. The surgical closure continues to remain in a dependent position. 2. With the exception of a gracilis or gluteal flap, there is little opportunity for interposition of a large healthy segment of vascularized tissue. 3. Prostate secretions can only drain in a retrograde fashion into the bladder or, in dyssynergic patients, remained trapped in the prostatic fossa. This results in a high rate of fistulization. 4. Perineal closure does not preserve antegrade ejaculation and compromises future fertility. Abdominal Closure in the Male The abdominal approach to bladder neck closure has two distinct advantages over perineal closure: (a) the bladder neck can be rotated anteriorly and out of a dependent position and (b) the choices for vascularized interposition are abundant (omentum, rectus flap, and peritoneal flap). Two techniques have historically been employed for abdominal closure of the bladder neck: supraprostatic and infraprostatic closure. Supraprostatic bladder neck closure has been our choice as it offers several distinct advantages over infra-prostatic closure. It is technically easier and does not involve deep pelvic dissection or transection of the dorsal venous complex. It allows for a better mobilization of the bladder neck, resulting in a tension-free closure. Lastly, it provides opportunity for future fertility as an antegrade flow of ejaculate is preserved. After supine or low lithotomy positioning, the patient is prepped and a catheter placed sterilely. An infraumbilical vertical midline incision is performed and the retropubic space is accessed as described earlier. The bladder neck is identified and absorbable suture is used to ligate the superficial dorsal venous complex at the prostatovesical junction. The prostate and vesical neck are grasped and electrocautery or sharp dissection is then used to amputate the anterior vesical neck from the prostate. Once the bladder neck mucosa is entered, the Foley balloon may be deflated and removed to permit visualization of the posterior vesical neck. Indigo carmine and ureteral catheters are used as previously described. The posterior bladder wall is transected and the plane between the bladder and the rectum identified. Mobilization of the posterior bladder neck from Denonvillier's fascia and rectum should continue until the vesical outlet has reached an anterior, nondependent position. Excessive mobilization should be avoided to prevent injury to the ureters or vascular pedicles of the bladder. A large-bore (22- to 24-Fr Malecot or Foley) suprapubic tube is then placed through a separate stab incision. If an alternative bladder drainage method is desired (incontinent vesicostomy, catheterizable efferent limb), it may be constructed at this time. Depending on its size, bladder neck closure can be performed by one of two methods. In the patient with a small bladder neck, a series of two absorbable purse-string sutures of 3-0 PGA may be used to invert the outlet similar to the inversion of an appendiceal stump. For a larger bladder neck, or where closure is more difficult, the outlet is closed in two layers as described above. Placement of a well-vascularized flap of omentum, rectus muscle, or peritoneum in the fossa between the bladder neck closure and prostate is performed to not only facilitate healing but to help prevent fistulization ( Fig. 49-9). Concomitant prostatectomy is generally not indicated unless a strictured urethra or prostatorectal fistula poses a problem to postoperative prostatic drainage.

FIG. 49-9. Omental interposition in male.

Vascularized Interposition Following the BNC it is highly advisable to interpose vascularized tissue between the bladder neck and the pelvic outlet to reduce the risk of secondary fistula. Choices for interposition include omentum, a flap of adjacent peritoneum, or a rectus flap. We prefer omentum because of its size, reliable blood supply, and abundant lymphatic drainage. In patients with a generous omentum, a tongue may be easily mobilized with only limited dissection. 8 If, however, the patient is extremely thin, or has had radiation or prior intraabdominal surgery, a more extended incision may be needed and the omentum may be mobilized on a pedicle supplied by the right gastroepiploic artery. The right side is preferred due to its more dependent position in the abdomen and its more generous blood supply. The omentum is positioned between the BNC and the pelvic outlet, and sutured in place with absorbable sutures. When a rectus flap is selected, it may be mobilized and based on an inferior epigastric vascular pedicle with careful attention to tie all lateral vascular collaterals. The mobilized rectus flap is then rotated downward and positioned as described above for omentum. Alternatively, a paravesical peritoneal flap may be interposed; however, its vascular supply may not be as reliable as omentum or rectus flap.8 A suction drain is left in the pelvis and brought out through a separate stab wound along with the suprapubic catheter. Postoperative Care Postoperative antibiotics are used for 3 to 5 days after which patients are placed on daily oral antibiotic suppression. The suction drain is usually left for 1 to 2 days. In our experience a nasogastric tube is not usually necessary. The suprapubic tube is carefully secured to avoid kinking or dislodgement. Patients are kept on either oral or rectal anticholinergic medication (oxybutynin with or without imipramine or belladonna and opium suppositories) to prevent bladder spasms. A cystogram is obtained at 2 to 3 weeks to document the integrity of bladder neck closure. If there is no evidence of leak or fistula, the suprapubic tube may be changed or removed if a catheterizable stoma was chosen for bladder drainage.

OUTCOMES
Complications The primary complication of bladder neck closure is postoperative fistula. Such a fistula may occur as early as 1 week postoperatively or as late as 1 year. Prevention of fistula formation is accomplished by careful debridement of the bladder neck edges, use of two nonoverlapping suture lines, nondependent positioning of the bladder neck, interposition of well-vascularized tissue, interposition over the closure, and avoidance of postoperative bladder spasms. When a fistula is suspected, the patient should undergo a cystogram with a mixture of 30% iodinated contrast and methylene blue dye. The site of leakage (vagina or perineum) should then be assessed both visually and radiographically. If a small fistula is encountered early in the postoperative period, bilateral percutaneous nephrostomies may be used to divert the urine away from the fistulous site. Reoperation is a more complex but reliable method of dealing with postoperative fistula. When the initial procedure was performed from a vaginal or perineal approach, reoperation should be performed suprapubically to allow extensive bladder mobilization and allow for interposition of a large, well-vascularized omental flap. Supravesical diversion is reserved for patients in whom all other attempts at repair have failed. Loss of access to the bladder may also represent a source of postoperative morbidity. Loss of a suprapubic tube and closure of its tract is an underreported but not uncommon complication. Access may be reestablished by using a flexible cystoscope or ureteroscope and may require fluoroscopy to negotiate the tract and pass a flexible wire down to the bladder. If this procedure fails, the patient may be given a fluid bolus and the bladder may be percutaneously accessed under sonographic guidance. Once access has been established, the tract may be dilated and a council catheter passed over the wire. Inability to catheterize a continent efferent limb may be treated similarly and endoscopic negotiation of the conduit usually suffices to reestablish access. Results Though a number of authors have reported their results with BNC, most series have been small, retrospective, and with a great deal of variability in technique. 2,3,6,10 Consequently, long-term outcomes and overall success rates are difficult to judge. In series where the bladder neck is anteriorly mobilized and appropriate vascularized interposition is utilized, long-term continence rates range from 86% to 100% with a 7% to 8% reoperation rate. 2,3,10 In series where these principles have not been employed, fistula formation and reoperation rate approach 30% and 25%, respectively. 7 Upper tract deterioration has been noted in a single series (11%) and has been causally related to use of a continent catheterizable efferent channel in patients with persistent bladder dysfunction. 3 Though closure of the bladder neck is not the procedure of choice for control of fistula or severe incontinence due to urethral loss, it nonetheless represents a viable treatment option in selected patients with advanced neurologic diseases or comorbidities precluding a more aggressive option. It may also be a useful alternative to supravesical urinary diversion in patients who have failed prior reconstructive lower urinary tract procedures. Decisions regarding surgical approach and postoperative bladder management should be individualized to meet each patient's needs. Essential techniques for closure include wide mobilization of the bladder neck out of a dependent position; multilayer, nonoverlapping, tension-free closure; and interposition of a well-vascularized flap between the closed bladder neck and the pelvic outlet. Following these principles should yield a high level of success with acceptable risks and postoperative morbidity. CHAPTER REFERENCES
1. Chancellor MB, Erhard MB, Kiilholma PJ, Karasick S, Rivas DA. Functional urethral closure with pubovaginal sling for destroyed female urethra after long-term catheterization. Urology 1994;43(4):499. 2. Hensle TW, Kirsch AJ, Kennedy WA, Reiley EA. Bladder closure in association with continent urinary diversion. J Urol 1995;154:883. 3. Jayanthi VR, Churchill BM, McLorie GA, Koury AE. Concomitant bladder neck closure and Mitrofanoff diversion for the management of intractable urinary incontinence. J Urol 1995;154:886. 4. Kakizaki H, Shibata T, Shinno Y, Kobayashi S, Matsuma K, Koyanagi T. Fascial sling closure for the management of incontinence due to sphincter incompetence. J Urol 1995;153:644. 5. McGuire EJ, Savastano J. Comparative urological outcome in women with spinal cord injury. J Urol 1986;135:730. 6. Schwartz SL, Kennelly MJ, McGuire EJ, Faerber GJ. Incontinent ileo-vesicostomy urinary diversion in the treatment of lower urinary tract dysfunction. J Urol 1994;152:99. 7. Stower MJ, Massey JA, Feneley RC, Urethral closure in management of urinary incontinence. Urology 1989;34(5):246. 8. Windle B, Kursh E. Use of interposition pedicle material for vesicovaginal fistula repair. In: McGuire EJ, ed. Female urology. Philadelphia: JB Lippincott, 1994;393–402. 9. Zeidman EJ, Chiang H, Alarcon A, Raz S. Suprapubic cystostomy using Lowsley retractor. Urology 1988;32(1):54. 10. Zimmern PE, Hadley HR, Leach GE, Raz S. Transvaginal closure of the bladder neck and placement of a suprapubic catheter for destroyed urethra after long term indwelling catheterization. J Urol 1985;134:554.

Chapter 50 Reconstruction of the Severely Damaged Female Urethra Glenn’s Urologic Surgery

Chapter 50 Reconstruction of the Severely Damaged Female Urethra
Jerry G. Blaivas

J. G. Blaivas: Department of Urology, New York Hospital–Cornell Medical Center, New York, New York 10022.

Diagnosis Indications for Surgery Alternative Therapy Description of Procedure General Principles of the Surgical Technique Timing of Surgery and Preoperative Management Technique Bladder Flap Techniques Outcomes Complications Results Conclusion Chapter References

The severely damaged female urethra is a rare occurrence that has two main causes—obstetric injury and surgical trauma. Obstetric injuries are exceedingly uncommon in industrial countries but not so in the Third World. Damage to the trigone, vesical neck, and urethra during delivery is thought to be the result of prolonged and neglected labor, most often associated with maternal–fetal disproportion wherein the fetal head compresses these structures against the pubis, causing pressure necrosis. 7 Surgical damage may occur during any of the Peyrera-type bladder neck suspension procedures, anterior colporrhaphy, urethral diverticulectomy, and, much less commonly, vaginal hysterectomy. In our experience, urethral diverticulectomy is the most common cause of extensive urethral damage.2 This most likely results from failure to obtain a tension-free closure of the urethral defect that results from excision of the diverticulum. During bladder neck suspension, inadvertent injury to the bladder or urethra may occur, or an errant suture may result in fistula formation or tissue necrosis. We have also seen several patients who sustained extensive tissue loss after a seemingly simple Kelly plication. It is postulated that the plication sutures were tied too tightly around a urethral catheter, resulting in pressure necrosis. Rarely, indwelling urethral catheters may cause pressure necrosis of the urethra, and even more rarely, trauma to the pelvis may result in fracture or separation of the symphysis pubis, which lacerates the urethra and/or vesical neck. Finally, there may be local invasion of these tissues from carcinoma of the cervix or damage from radiation treatment. Regardless of the cause of urethral damage, the diagnostic and therapeutic challenges to the surgeon are considerable. The goals of surgical correction are to create a continent urethra that permits the painless, unobstructed passage of urine and is of sufficient length to ensure that the patient does not void into the vagina. It is our belief that these goals can almost always be accomplished with a single operative procedure; only rarely is a staged procedure necessary.

DIAGNOSIS
Most patients with extensive damage to the urethra have overt incontinence; in them, the diagnosis is obvious upon examination of the vagina. Sometimes, though, when the vesical neck remains intact, there may be loss of the entire remaining portion of the urethra without any symptoms. These are generally discovered incidentally on physical examination and need no treatment. With all of these injuries, one must have a high index of suspicion that there are concomitant abnormalities such as vesicovaginal or ureterovaginal fistula, ureteral obstruction, low bladder compliance, vesicoureteral reflux, and intrinsic sphincteric deficiency. A careful evaluation to exclude each of these potential conditions should be undertaken before surgery. In addition, detrusor function may be compromised in the form of impaired detrusor contractility or detrusor instability, but these conditions generally do not require evaluation unless they persist postoperatively. We recommend that intravenous pyelography be performed in all patients except those in whom it is contraindicated. If there is any index of suspicion for ureteric injury, retrograde pyelography should be performed even in women with normal-appearing intravenous pyelograms. Cystoscopy and pelvic examination are essential in order to evaluate (a) the extent of the anatomic defect, (b) the possibility of unrecognized secondary fistulas, (c) the pliability of local tissue, (d) the need for securing bulk-ensuring tissue pedicle flaps, (e) the need for concomitant pelvic reconstructive surgery, and (f) the timing of surgery. Secondary vesicovaginal fistulas are usually apparent at cystoscopic examination. If a vesicovaginal fistula is suspected but not seen at the time of cystoscopy, the bladder should be filled with fluid to which a dye such as methylene blue has been added. The vagina should then be inspected for signs of urinary leakage with the urethra occluded with a Foley balloon catheter or with the surgeon's examining finger occluding the urethra in order to prevent urethral leakage. Urinary incontinence is not always caused by what appears to be the most overt lesion. For example, neither a urethrovaginal fistula nor a destroyed distal urethra should cause urinary incontinence unless the proximal urethra and vesical neck are also damaged. A careful stepwise evaluation should be carried out in all patients to delineate the pathophysiology underlying incontinence. Other causes of urinary incontinence commonly seen in these patients include (a) a previously undiagnosed vesicovaginal or ureterovaginal fistula, (b) detrusor instability, (c) low bladder compliance, and (d) sphincteric abnormalities.

INDICATIONS FOR SURGERY
The mere presence of extensive urethral damage is not an indication for surgery. The two main indications for reconstruction are sphincteric incontinence and urethral obstruction, but neither is absolute. Of course, if there is an associated condition such as a vesicovaginal fistula, it should be repaired at the same time. Urethral reconstruction is technically demanding and requires a considerable degree of experience and skill. In inexperienced hands, the risks may be prohibitive, and in some instances, when there is insufficient local tissue for reconstruction, it may be more prudent to consider urinary diversion than urethral reconstruction. When sphincteric incontinence is present preoperatively, we believe that it should be surgically corrected at the time of urethral reconstruction. In general, we prefer to construct a fascial pubovaginal sling 3 with an interposed free graft of labial fat pa 2,3 and 4,8 between the sling and the reconstructed vesical neck, but other authors have recommended the modified Peyrera technique in patients with less extensive anatomic damage and incontinence caused by urethral hypermobility. 5 Although it is important to document the presence of detrusor instability or low bladder compliance preoperatively, it has been our experience that after surgery these conditions abate in the vast majority of women. Accordingly, we do not recommend concomitant surgical interventions to treat these conditions at the time of urethral reconstruction. There are three generic approaches to urethral reconstruction: (a) anterior bladder flaps, 5,12 (b) posterior bladder flaps, 9 and (c) vaginal wall flaps. 2,6,7 and 8,11 These techniques appear to be comparable with respect to creation of a neourethra. However, whenever the vesical neck and proximal urethra are involved, which is usually the case, postoperative incontinence rates of about 50% are to be expected unless a concomitant anti-incontinence procedure is performed. 5,9,12 We believe that vaginal reconstruction is considerably easier and faster, much more amenable to concomitant anti-incontinence surgery, and associated with much less morbidity than the bladder flap operations.

ALTERNATIVE THERAPY

Alternatives to urethral reconstruction include various forms of urinary diversion and chronic catheter drainage. Chronic catheter drainage is indicated only in patients in whom there is no alternative, who are generally too ill to undergo any other procedure, and is associated with chronic infections, stones, and vesical spasms.

DESCRIPTION OF PROCEDURE
General Principles of the Surgical Technique In these women, vaginal tissues are often scarred, fibrotic, and ischemic. Before surgery careful examination of the vagina is necessary to determine the actual extent of urethral tissue loss and to assess the availability of local tissue for use in the reconstruction. In most instances there is sufficient tissue in the anterior vaginal wall that can be mobilized and rolled into a tube or patch graft for urethral reconstruction. 2,3,5,6,7 and 8,11 Occasionally, it may be necessary to use an adjacent labial flap. Alternatively, an anterior bladder flap can be used. 4 After reconstruction of the urethra or fistula repair, it is usually advisable to interpose a well-vascularized pedicle flap over the site of the repair. These include labial, 3,5,11 rectus abdominis, 6 and gracilis. 2 The most important surgical principles include (a) clear visualization and exposure of the operative site, (b) creation of a tension-free, multiple-layered closure, and (c) assurance of an adequate blood supply and (d) adequate bladder drainage. Operative exposure often requires two or more assistants and the use of self-retaining retractors. A tension-free closure can usually be accomplished by wide mobilization of surrounding tissue but sometimes requires the use of local pedicle flaps or relaxing incisions in the anterior vaginal wall. Bladder drainage is best accomplished with a large suprapubic catheter, which should be placed at the beginning of the procedure to avoid having to distend the bladder after the reconstruction. Timing of Surgery and Preoperative Management In the past, much controversy surrounded the timing of surgical repair. For decades it had been taught that surgery should be delayed for 3 to 6 months or longer to allow adequate time for tissue inflammation and edema to subside. In our experience, surgery can be safely performed as soon as the vaginal wound is free of infection and inflammation and the tissues are reasonably pliable. It is usually possible to perform the surgery within 3 to 6 weeks after the original surgery, but we have performed one reconstruction 7 days after birth trauma. Management of incontinence while waiting for healing of the vaginal tissue is often a difficult problem. In most patients Foley catheter drainage is insufficient. If significant leakage occurs with a Foley catheter in place, we generally recommend that the catheter be discontinued and the patient be managed with superabsorbent pads, which are changed frequently throughout the day. Technique The patient is placed in the dorsal lithotomy position, and cystourethroscopy is performed to assess the relationship of the ureteral orifices to the damaged urethra. If the ureteral orifices are in close proximity to the fistula, single-J ureteral stents are left indwelling. A 16-Fr Foley catheter is inserted into the bladder, and the balloon is inflated with enough fluid to hold it securely at the vesical neck. A percutaneous suprapubic cystotomy tube (at least 14 Fr) is placed and sewn to the anterior abdominal wall unless a pubovaginal sling or modified Pereyra is planned, in which case this is deferred until the end of the procedure. The choice of incision depends on the local anatomy of the tissue loss and whether or not a pubovaginal sling or other anti-incontinence procedure is planned. If a pubovaginal sling is planned, the dissection for the sling is completed first, and the sling is passed around the site of the vesical neck, but the sutures are not tied, and the sling is retracted so that it does not interfere with the subsequent dissection to reconstruct the urethra. A Pfannenstiel incision is made and carried down to the rectus fascia. The surface of the rectus fascia is dissected free of subcutaneous tissue, and a suitable site is selected for excision of the fascial strip that will be used as a free graft for creation of the sling. Two parallel horizontal incisions 2 to 3 cm apart are made near the midline in the rectus fascia ( Fig. 50-1A). The incisions are extended superolaterally for the entire width of the wound, following the direction of the fascial fibers. The undersurface of the fascia is freed from muscle and scar. Before the strip is excised, each end of the fascia is secured with a long 2-0 monofilament nonabsorbable suture using a running horizontal mattress, which is placed at right angles to the direction of the fascial fibers ( Fig. 50-1B, Fig. 50-1C). No attempt is made to mobilize the bladder or vesical neck from above.

FIG. 50-1. (A) A 2- to 3-cm-wide graft is outlined, with the incision kept parallel to the direction of the rectal fascial fibers. The incision is extended laterally to the point where the fascia divides and passes to the internal and external oblique muscles. (B) A 2-0 nonabsorbable running horizontal mattress suture is placed across the most lateral portion of the graft, and the ends are left long. (C) Each end of the fascial graft is transected approximately 1 cm lateral to the mattress suture. (D) Dissection is begun with Metzenbaum scissors in the avascular plane just beneath the vaginal epithelium. The tips of the scissors are directed toward the patient's ipsilateral shoulder. (E) The endopelvic fascia is perforated with the index finger, and the retropubic space is entered. (F) A long DeBakey clamp is passed from the abdominal to the vaginal wound lateral to the urethra. (G) The fascial graft is passed around the urethra and brought to the abdominal wound on either side. ( A-G from Blaivas JG. Pubovaginal sling procedure. In: Whitehead ED, ed. Current operative urology. Philadelphia: JB Lippincott, 1990;93–101.) (H) The long ends of the sling are tied together in the midline with no tension. The labial fat pad is positioned between the sling and the vesical neck.

The lateral edges of the vaginal wound are grasped with Allis clamps and retracted laterally. The dissection continues just beneath the vaginal epithelium with Metzenbaum scissors pointed in the direction of the patient's ipsilateral shoulder until the periosteum of the pubis or ischium is palpated with the tip of the scissor ( Fig. 50-1D). During this part of the dissection, it is important to stay as far lateral as possible. This is best accomplished by dissecting with the concavity of the scissors pointing laterally and by exerting constant lateral pressure with the tips of the scissors against the undersurface of the vaginal epithelium. Once the periosteum is reached, the endopelvic fascia is perforated, and the retropubic space is entered. In most instances this is easily accomplished by blunt dissection with the surgeon's index finger (Fig. 50-1E). The tip of the finger, opposite the nail, palpates the periosteum. With the back edge of the fingertip, the bladder and urethra are mobilized medially as the finger advances and perforates the fascia. This completely mobilizes the vesical neck and proximal urethra, freeing these structures from their vaginal attachments. In some instances this dissection must be performed sharply with Metzenbaum scissors. The surgeon's left index finger is reinserted in the vaginal wound, retracting the vesical neck and bladder medially. The tip of the finger is palpated by the right index finger, which is placed just under the inferior leaf of the rectus fascia. A long sharp curved clamp (DeBakey) is inserted into the incision and directed to the undersurface of the pubis ( Fig. 50-1F). The tip of the clamp is pressed against the periosteum and directed toward the index finger, which is retracting the vesical neck and bladder medially. In this fashion, the clamp is guided into the vaginal wound. When the tip of the clamp is visible, one end of the long suture, which is attached to the fascial graft, is grasped and pulled into the abdominal wound ( Fig. 50-1G). The procedure is repeated on the other side. A Kocher clamp is placed on the inferior edge of the rectus fascia, and the fascia is retracted vertically. An incision is made in the rectus fascia just above the pubis and lateral to the midline, just

large enough to accept the sling. The long ends of the suture attached to the sling are grasped and pulled through the fascial incisions on either side. The rectus fascia is closed with a running suture of 2-0 PDS. The fascial sling is now positioned from the abdominal wall on one side around the undersurface of the vesical neck and back to the abdominal wall on the other side. The long sutures attached to either end of the sling are marked with clamps, and the sling is retracted out of the surgical field while the urethral reconstruction is being performed. At the conclusion of the operation, the long ends of the sling are tied together in the midline over the rectus fascia without any tension at all ( Fig. 50-1H). There are three basic techniques for urethral reconstruction: 1. Primary closure. If the anterior portion of the urethra is intact, it is usually possible to close the urethra primarily. The defect is circumscribed ( Fig. 50-2A), and lateral vaginal wall flaps are elevated ( Fig. 50-2B). If a pubovaginal sling is to be done, the dissection is completed at this time. The lateral urethral walls are mobilized, and the urethra is closed with interrupted sutures of 3-0 chromic catgut ( Fig. 50-2C). We prefer chromic catgut to longer-acting synthetic absorbable sutures for urethral closure because, in our experience, the latter often make subsequent voiding and/or urethral instrumentation painful. If a Martius labial fat pad graft is needed, it is prepared at this time (see below) and placed over the reconstructed urethra and vesical neck. The pubovaginal sling is positioned over the Martius flap at the level of the bladder neck, and the vaginal incision is closed primarily with 2-0 chromic catgut ( Fig. 50-2D).

FIG. 50-2. (A) The fistula is circumscribed. (B) Lateral vaginal wall flaps are elevated, and the lateral urethral walls are mobilized. (C) The urethra is closed primarily with interrupted sutures of 3-0 chromic catgut. (D) The vaginal wall is closed. (Modified from Mattingly RF, Thompson JD. In: Telinde's operative gynecology, 6th ed. Philadelphia: JB Lippincott, 1985;662.)

2. Advancement flap. If there is insufficient tissue on the anterior vaginal wall at the site of the urethra to mobilize lateral flaps, it may be possible to repair the urethra and cover the repair with an advancement flap from the anterior vaginal wall cranial to the damaged urethra. A U incision is made in the anterior vaginal wall long enough to be advanced and rotated to form the posterior and lateral walls of a neourethra ( Fig. 50-3A). The flap is mobilized with Metzenbaum scissors, reflected caudally, and sutured in place over the urethral catheter ( Fig. 50-3B). When necessary, the dissection for a pubovaginal sling is done at this time before the vaginal wall flap is sutured. When indicated, a Martius flap is prepared and sutured in place between the reconstructed urethra and the pubovaginal sling. At the end of the procedure, the vaginal wall is closed primarily with 2-0 chromic catgut ( Fig. 50-3C,D). If there is insufficient tissue for primary closure, it may be possible to use an inverted-U advancement flap or, rarely, a labial pedicle graft (see below).

FIG. 50-3. (A) A U-shaped incision is made with the arms of the U extending caudally as far as the planned urethral meatus. (B) The flap is elevated and rotated 180 degrees. The flap is sutured to the edges of the parallel distal incisions over the catheter to form the new urethra. (C,D) The vaginal wall is approximated in the midline to cover the neourethra. (Modified from Mattingly RF, Thompson JD. In: Telinde's operative gynecology 6th ed. Philadelphia: JB Lippincott, 1985;660–661.)

3. Tube graft. If there is circumferential loss of the urethra and sufficient vaginal wall tissue on the anterior vaginal wall, parallel incisions are made over the site of the neourethra, and the resulting strip is rolled into a tube over the urethral catheter. If there is a urethral fistula present, we prefer to retain the remaining bridge of urethral tissue and close the fistula to preserve local blood supply. The fistula is circumscribed, and an inverted-U-shaped incision is made ( Fig. 50-4A). The flap is reflected caudally, and the dissection for the pubovaginal sling is completed ( Fig. 50-4B). The fistula is closed by approximating its edges with interrupted sutures of 4-0 chromic catgut (Fig. 50-4C). Two parallel incisions are made on either side of the urethral catheter for construction of the neourethra ( Fig. 50-D). The flaps are mobilized, rolled into a tube over the urethral catheter, and sutured in the midline with interrupted sutures of 4-0 chromic catgut. A (Martius) labial fat pad graft is prepared to provide a second layer of tissue over the neourethra. The inverted-U flap is advanced over the entire reconstructed urethra. The vaginal and labial wounds are closed, and a ¼-inch Penrose drain is left in the labial wound ( Fig. 50-4G). If it is not possible to close the vaginal incision primarily, a labia minora (hair-free) pedicle graft should suffice. Since we began using this technique, we have never found it necessary to use a gracilis myocutaneous graft.

FIG. 50-4. (A) An inverted-U-shaped incision is made in the anterior vaginal wall with the apex of the U at the vesical neck just proximal to the urethral fistula. The fistula is circumscribed. (B) A plane is created in the avascular plane just underneath the vaginal epithelium, and the vaginal wall flap is reflected posteriorly. If a pubovaginal sling or other anti-incontinence procedure is to be performed, the dissection into the retropubic space is completed at this time. (C) The urethrovaginal fistula is closed with interrupted sutures of 3-0 or 4-0 chromic catgut. (D) Two parallel incisions are made alongside the Foley catheter, and medially based flaps are elevated. (E) The vaginal and labial wounds are closed. (From Blaivas JG. Vaginal flap urethral reconstruction. An alternative to the bladder flap neurourethra. J Urol 1989;41:542–545.

The technique for creation of a Martius flap is as follows. A vertical incision is made over the labia majora, and the fat pad is mobilized ( Fig. 50-5A). The fat pad is tunneled underneath the vaginal epithelium and sewn in place over the suture lines of the neourethra ( Fig. 50-5B,C). If a single labial fat pad graft does not provide adequate coverage, a second graft may be obtained from the other side, or a gracilis, perineal, or rectus pedicle graft may be harvested. The pedicle graft is placed between the sling and the reconstructed urethra.

FIG. 50-5. Technique of obtaining Martius fibrofatty pedicle graft. (Modified from Mattingly RF, Thompson JD. In: Telinde's operative gynecology, 6th ed. Philadelphia: JB Lippincott, 1985;663.)

The technique for performing a labial graft is depicted in Fig. 50-6. At the conclusion of the procedure, the Foley catheter is sutured to the anterior abdominal wall with a gentle loop to insure that undue tension is not placed on the urethra. Failure to maintain the correct position of the catheter may result in necrosis of the urethra.

FIG. 50-6. Technique of obtaining full-thickness labial fibrofatty pedicle graft to cover the reconstructed urethra when there is insufficient local vaginal wall tissue. (Modified from Mattingly RF, Thompson JD. In: Telinde's operative gynecology, 6th ed. Philadelphia: JB Lippincott, 1985;665.)

The Penrose drain is removed on the first postoperative day. The urethral wound is checked daily, and the catheter is removed as soon as feasible, usually within the first 2 to 5 days. A voiding cystourethrogram is performed though the suprapubic catheter at about day 14. If the patient voids satisfactorily and there is no extravasation, the suprapubic tube is removed. If not, the tube is left in place, and another voiding trial is undertaken 2 to 4 weeks postoperatively. Bladder Flap Techniques We believe that bladder flap reconstructions are almost never necessary in these patients, and the single patient in whom we performed this procedure was a failure. The basic technique is depicted in Fig. 50-7.10 In the most recent and extensive series, Elkins et al. 4 recently reported their experience with a Tanagho-like procedure in 20 West African women with extensive urethral damage subsequent to obstructed labor. These patients all had large vesicovaginal fistulas and, because of extensive scarring, were not suitable for vaginal flap techniques. The procedure was performed entirely through the vaginal approach.

FIG. 50-7. Technique of performing anterior bladder flap urethroplasty. (From Tanagho EA. Bladder neck reconstruction for total urinary incontinence: 10 years of experience. J Urol 1981;125:321.)

The anterior and lateral fistula edges are dissected sharply away from the pubic bone beneath the arch of the pubic ramus, thus entering the retropubic space. The anterior bladder wall is dissected free of surrounding tissues to the level of the peritoneal re-flection. The anterior bladder is mobilized, and a 3 × 3 cm flap is raised and rolled into a tube over a 16-Fr catheter. The new urethra is sutured either to the remaining distal urethra or at the site of the new meatus. The posterior edges of the vesicovaginal fistula are approximated, and fixation sutures are placed through the top portion of the neourethra to reattach the urethra to the base of the pubic periosteum. In the last three patients a modified Peyrera procedure was performed instead. A Martius fat pad graft was then placed beneath the suture lines.

OUTCOMES
Complications Early in our series, we did not routinely perform concomitant pubovaginal slings, and 50% of the women who underwent a modified Pereyra procedure had persistent sphincteric incontinence; all were subsequently cured by a pubovaginal sling. Only one of the remaining women developed sphincteric incontinence, which was associated with necrosis of the reconstructed urethra. Three patients had flap necrosis, one a previously unrecognized vesicovaginal fistula, and one had urethral

obstruction caused by the pubovaginal sling. All but one of the failures were subsequently reoperated on, and all had a successful outcome. This is consistent with the reported complications from large international series, where stress incontinence is a relatively frequent sequela of reconstruction and requires a second procedure (see Results). In our series of patients who have undergone bladder flap reconstruction, 18 of the 20 women had a satisfactory anatomic repair of the fistula, but four of the 18 had persistent stress incontinence that required further surgery. Two others had refractory detrusor instability or low bladder compliance. Results To date there has been a paucity of studies concerning reconstruction of the severely damaged urethra. Among all series we could find in the English language literature, there were fewer than 400 patients ( Table 50-1). Overall, successful anatomic reconstructions were reported in 67% to 100% of women. Most authors emphasized the need for well-vascularized pedicle flaps to insure a successful outcome. Continence was achieved in only 55% to 92% after a single operation, and postoperative urethral obstruction was reported in 2% to 17%. In the great majority of studies, the criteria for incontinence and urethral obstruction were not specified, and especially in view of lack of follow-up, the results cited above should be considered optimistic ones. It does seem evident, however, that it is important to perform an anti-incontinence procedure at the same time as the urethral reconstruction. Failure to do so resulted in incontinence rates varying from 50% to 84%. Most series indicated that secondary procedures to correct incontinence are successful in the majority of patients.

TABLE 50-1. Summary of cases reported in English language literature

The largest series was reported by Hamlin in 1969 from West Africa. All of the women suffered childbirth injuries, and all were quite extensive. An excellent anatomic repair was achieved in 49 of the 50 women, but eight (16%) had severe incontinence, and many more had lesser degrees of incontinence. They stated that the incontinence was usually cured after a second procedure. Elkins et al. 3 reviewed the results of 36 vesicovaginal and/or urethrovaginal fistula repairs performed by American visiting professors in West Africa. All of the vesical neck and urethral fistulas resulted from obstructed labor. In this series, two of 13 proximal urethral fistulas were complicated by severe stress urinary incontinence postoperatively. To date we have operated on 51 women with extensive anatomic vesical neck and urethral defects caused by complications from urethral diverticulectomy (24), modified Pereyra bladder neck suspensions (18), anterior colporrhaphy (eight), and childbirth injury (one). All but one patient underwent a vaginal reconstruction (one patient had a Tanagho anterior bladder flap, and that failed). We used a Martius flap in all of the remaining patients except one who had a gracilis flap. We have previously published the results of 49 of these women. 1 With the modifications described above, no patient required intermittent catheterization, and at least 1 year after their last surgery all were dry (except for the one woman who refused reoperation).

CONCLUSION
Reconstruction of the severely damaged urethra is a technically challenging undertaking and requires considerable surgical expertise and decision making. The vast majority of women with traumatic injuries have sufficient vaginal tissue for a vaginal flap reconstruction, and we believe that the vaginal approach offers the best chance for a successful outcome. However, in patients with extensive vaginal scarring from childbirth injuries, such as those reported by Elkins, bladder flap techniques may be useful. The most important principles to keep in mind are: (a) clear visualization and exposure of the operative site, (b) creation of a tension-free, supple, multiple-layered closure, (c) assurance of an adequate blood supply and soft tissue base with a Martius flap, (d) concomitant pubovaginal sling when anti-incontinence surgery is indicated, and (e) adequate bladder drainage. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Blaivas JG. Vaginal flap reconstruction of urethra and vesical neck in women: Report of 49 cases. J Urol 1996;155:1014. Blaivas JG. Vaginal flap urethral reconstruction: an alternative to the bladder flap neourethra. J Urol 1989;41:542–545. Blaivas JG. Female urethral reconstruction. In: Webster G, ed. Reconstructive urology. Boston: Blackwell Scientific, 1993;873. Elkins TE, DeLancey JO, McGuire EJ. The use of modified Martius graft as an adjunctive technique in vesicovaginal and rectovaginal fistula repair. Obstet Gynecol 1990;75:727–733. Elkins TE, Ghosh TS, Tagoe GA, Stocker R. Transvaginal mobilization and utilization of the anterior bladder wall to repair vesicovaginal fistulas involving the urethra. Obstet Gynecol 1992;79:455–460. Ellis LR, Hodges CV. Experience with female urethral reconstruction. J Urol 1969;102:214. Gray LA. Urethrovaginal fistulas. Am J Obstet Gynecol 1968;101:28. Hamlin RHJ, Nicholson EC. Reconstruction of urethra totally destroyed in labor. Br Med J 1969;1:147. Leadbetter GW Jr. Surgical correction of total urinary incontinence. J Urol 1964;91:261. Mattingly RF, Thompson JD: Telinde's operative gynecology, 6th ed. Philadelphia: J.D. Lippincott, 1985; 622. Symmonds RE. Loss of the urethral floor with total urinary incontinence: a technique for urethral reconstruction. Am J Obstet Gynecol 1968;103:665–678. Tanagho EA. Bladder neck reconstruction for total urinary incontinence: 10 years of experience. J Urol 1981;125:321–326.

Chapter 51 Urethral Stricture and Disruption Glenn’s Urologic Surgery

Chapter 51 Urethral Stricture and Disruption
E. James Wright and George D. Webster

E. J. Wright: Department of Surgery, Division of Urology, University of Kentucky, Lexington, Kentucky 40536. G. D. Webster: Department of Surgery, Division of Urology, Duke University Medical Center, Durham, North Carolina 27710.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Preparation Strictures of the Glanular Urethra Strictures of the Pendulous Urethra Bulbar Urethral Stricture Repair Panurethral Stricture or Failed Urethroplasty Posterior Urethroplasty Perineal Anastomotic Repair Outcomes Complications Results Chapter References

Urethral strictures result from scarring induced by local tissue injury. Trauma is currently the most common cause of urethral stricture, and this can be of accidental or iatrogenic origin. Blunt trauma in the form of a straddle injury often results in bulbar stricture as does iatrogenic injury from endoscopic instrumentation. With modern day proliferation of firearms, penetrating missile injuries involving any portion of the urethra are increasing in frequency. Inflammatory processes, including gonococcal urethritis and balanitis xerotica obliterans, can precipitate stricture formation. An uncommon cause of urethral stricture is malignancy, usually of urethral, penile, or lymphoid origin. The pathophysiology of urethral stricture involves progressive fibrosis of the epithelium with subsequent involvement of the underlying spongiosal tissue. Narrowing of the urethral lumen results, culminating in near-obliteration and, occasionally, fistula and abscess formation.

DIAGNOSIS
Urethral stricture most often presents with obstructive voiding symptoms although irritative symptoms, including frequency, urgency, and, occasionally, dysuria, may also occur. Microscopic hematuria is not uncommon and pooling of urine in the dilated urethra proximal to a stricture may lead to prolonged postvoid dribble. Recurrent urinary tract infection can accompany these symptoms, especially if the etiology is inflammatory. In cases of traumatic stricture, the history will include prior urethral instrumentation or an episode of external injury. Difficult catheter placement is sometimes the initial presentation of urethral stricture disease. Radiologic studies provide the best means for evaluation of urethral strictures. Retrograde urethrography will define the location and severity of luminal narrowing. When the tightness of the stricture precludes good visualization of the urethra proximal to the stricture, voiding cystourethrography may be used. Simultaneous voiding cystourethrography and retrograde filling is used to evaluate the obliterated urethra such as occurs following pelvic fracture with urethral injury. Ultrasound and magnetic resonance imaging have also been used in the evaluation of stricture disease.
4

Physical examination should include urethral palpation to determine the depth and severity of spongiofibrosis. Assessment of penile shaft skin and circumcision status should also be included in planning surgical repair. Laboratory evaluation includes measurement of serum creatinine and urine culture. Finally, cystourethroscopy should be carried out to visually assess the urethra and strictured segment for signs of malignancy and correlation with imaging studies.

INDICATIONS FOR SURGERY
The occurrence of urinary retention, recurrent urinary infection, hydronephrosis, and stone formation are indications for the treatment of stricture. Treatment is often initially by dilation or visual urethrotomy; however, such intervention is not frequently curative and surgical urethroplasty is then indicated. 8 More specifically, patients in whom urethral stricture recurs rapidly following dilation or urethrotomy; those in whom dilation or endoscopic management is complicated because of local adverse features such as false passages, diverticula, and fistula; and those in whom dilation is often accompanied by bacteremic episodes or excessive bleeding are candidates for surgical intervention. Additionally, strictures in children justify earlier consideration for surgery. Many urethroplasty techniques have been described and it is true to say there is no single technique appropriate for the management of all strictures. Techniques may be broadly grouped into anastomotic and substitution repairs. In substitution repairs the urethral lumen is augmented using either flap or graft techniques, with the grafts customarily obtained from penile skin although buccal mucosa is a current alternative. Some techniques involve a combination of anastomotic and substitution approaches and may be performed in single or multiple stages. The selection of appropriate repair is governed to a large degree by the location of the stricture, its length, and the presence of local adverse features such as fistulas, inflammation, or scarring. The goal of surgical repair should not only be to achieve urethral patency, but to avoid compromising normal sexual function and to avoid altering penile cosmetics. Repair of pendulous urethral strictures can result in penile chordee if urethral shortening results, so that stricture excision and anastomotic repairs are relatively contraindicated, and indeed, even skin graft repairs in the penile urethra may prove inelastic. Hence, flap repairs are optimal in this location. In the bulbar urethra, strictures may result from straddle injury in which case they are short and may be repaired by excision and reanastomosis. Inflammatory strictures tend to be longer and are treated by onlay graft or flap techniques. Pelvic fracture urethral injury often causes distraction defects of the membranous urethra; these are invariably managed by a perineal anastomotic repair. In performing urethroplasty care must be taken to extend the repair proximally and distally into healthy urethra. Failure to accomplish this is likely the commonest cause for recurrent stricture following surgical repair.

ALTERNATIVE THERAPY
There are a number of therapeutic alternatives for the treatment of urethral stricture. These include chronic dilation, preferably with filiform and followers or by balloon catheter; visual urethrotomy using cold knife, cautery, or laser; and surgical urethroplasty. Other endoscopic techniques that have been described include transurethral placement of a tubularized skin graft and prosthetic stents.

SURGICAL TECHNIQUE
Preoperative Preparation The urine culture should be negative prior to surgery and on call parenteral antibiotics are given. If the patient has a suprapubic catheter, then broad-spectrum antibiotics are given for 24 hours prior to surgery. The patient takes an antiseptic sitz bath the night before surgery and genital shaving is deferred to the operating

room. General endotracheal anesthesia is preferable if it poses no increased risks to the patient. Loss of vascular tone with regional anesthesia techniques can exacerbate venous congestion and engorgement of the urethral tissues. In addition, inadvertent movement of the patient as often occurs during spinal or epidural anesthesia can compromise surgical exposure and success of the repair. Most urethroplasties are performed in the lithotomy position using a variety of leg support systems. Care must be taken in positioning the patient to pad all pressure points, particularly the lateral aspect of the lower leg so as to avoid injury to the peroneal nerve. Strictures of the Glanular Urethra Strictures in the glanular urethra are most commonly due to inflammatory conditions, instrumentation, or improper circumcision. Surgical repair of these strictures may be accomplished by simple meatotomy, a glans-based Y-V advancement flap for very distal stenosis, or a ventral shaft skin onlay flap in the case of stricture involving the entire fossa navicularis. For Y-V advancement a V-shaped incision is made on the dorsal aspect of the glans with the apex terminating at the urethral meatus ( Fig. 51-1). A dorsal midline extension of this incision is made into the urethra to widen the lumen to accept a 22-Fr sound. The glans flap can be elevated using skin hooks and sharp dissection to allow its mobilization into the most proximal limit of the dorsal urethral incision where it is anchored with a 4-0 polyglycolic acid (PGA) suture. Care must be taken to leave this glans flap with a wide base to avoid vascular compromise. A 14-Fr Foley catheter is left in place for approximately 5 days.

FIG. 51-1. Glans flap meatoplasty.

More extensive strictures of the glanular and distal penile urethra may result from balanitis xerotica obliterans (BXO). These are best repaired with a flap of skin obtained from the penile shaft. The flap may be fashioned as an onlay or as a tube, depending on the extent of the disease. There are a number of variations of this technique but one commonly used is depicted in Figure 51-2. In performing this procedure it is important that the repair be continued into healthy urethra proximal to the stricture, and in cases where BXO is the cause, strictured urethral excision and tubed skin flap replacement may be optimal. An excellent cosmetic result can be achieved by mobilization of glans wings and their reapproximation around the neourethra. A 14-Fr urethral Foley catheter remains in place for 7 days following the repair.

FIG. 51-2. Repair of meatal and distal penile urethral stricture using a pedicled island of distal penile skin. This procedure allows for reapproximation of glans tissue around the neourethra giving an excellent cosmetic and functional result.

Strictures of the Pendulous Urethra Pendulous urethral strictures are most often repaired using an onlay flap of penile skin. Although many variations of this technique have been described, the laterally based pedicled island flap originally described by Orandi gives excellent results and is technically forgiving. 6 It is generally suitable for repairs measuring up to 8 cm (depending on penile length), although longer repairs can be performed by harvesting a penile flap that has a vertical and a distal circumferential component. A flap width up to 25 mm can be used without compromising penile skin closure. The patient is placed in either lithotomy or a supine frog-legged position. Methylene blue is injected retrogradely into the urethra to accentuate the stricture margins. An incision is made along the ventral penile raphe through Buck's fascia down to the strictured segment of urethra. Using skin hooks to elevate the flap and fine scissors for dissection, one skin margin is mobilized from the corpus cavernosum for 10 to 15 mm staying deep to Buck's fascia ( Fig. 51-3). With the urethra adequately exposed, a suitable catheter or filiform is placed in the urethral lumen as a guide and the urethra incised ventrally along the stricture. This incision is then extended using scissors to expose at least 1 cm of healthy urethra proximal and distal to the stricture. Significant blood loss can be controlled by oversewing the urethral edge with 5-0 PGA suture, although this is seldom necessary. An onlay flap of suitable width to achieve a 24-Fr lumen (generally 18 to 25 mm) is then outlined on the mobilized lateral penile shaft skin. The flap length corresponds to the length of the urethrotomy and is tapered at the proximal and distal ends. Care must be taken not to incorporate a significant segment of hair-bearing shaft skin in the repair.

FIG. 51-3. Onlay urethroplasty of a pendulous urethral stricture (Orandi repair). The stricture is approached through a ventral penile incision. The outline of the skin

island to be used for onlay is marked once the stricture has been opened and its length and caliber determined. The skin island is raised on a subcutaneous vascular pedicle and the skin island is rotated inward and sutured as an onlay to augment urethral caliber. Skin closure is in at least two layers to avoid fistula formation.

Once the flap has been outlined, the projected new skin edges are approximated in the midline to ensure that wound closure will be possible with a minimum of tension after the flap has been raised. The lateral aspect of the flap can then be incised superficially through the subcutaneous connective tissue, leaving the underlying dartos and deeper Bucks fascia intact. With exposure of the correct tissue plane, the lateral penile shaft skin will retract with minimal additional dissection. The medial border of the island flap is then anchored to the lateral edge of the open urethra with 5-0 PGA stay sutures. Beginning at the distal margin, the flap is sutured to the urethra using a running 5-0 PGA suture, incorporating the urethral epithelium in the closure. When the lateral suture line is complete, the free edge of the island flap can be rolled over and secured with 5-0 PGA stay sutures to the other urethral margin, recreating and augmenting the urethral lumen. A similar running suture line is performed with 5-0 PGA to complete the onlay repair. A 14- to 16-Fr silastic Foley catheter is placed for urine diversion and stenting. Local wound drainage may be accomplished with a 7-Fr suction drain placed underneath the vascular pedicle along the length of the flap if necessary. This guards against hematoma or urinoma formation and potential flap compromise. Wound closure is performed in two layers using 4-0 interrupted PGA suture. Care must be taken when approximating the subdermal connective tissue of the skin margins to not injure the vascular pedicle of the flap. This first layer of closure serves to cover the exposed urethral suture line and minimize the risk of fistula formation. Final skin closure is performed with a running subcuticular 4-0 PGA suture. The wound is supported with adhesive strips and covered with a gauze and loosely applied Coband® dressing to reduce edema formation. The Foley catheter is secured to the abdominal wall and remains in place for up to 3 weeks. The suction drain is usually removed within 24 hours and the supportive dressing taken down on the third postoperative day. The patient is allowed to ambulate on the first day following surgery and he may be discharged within 24 to 48 hours. Upon return visit at 3 weeks a pericatheter retrograde urethrogram is performed. If there is no extravasation at the repair site, the catheter is removed and the patient resumes normal voiding. Further follow-up is at 3 months and then annually. Bulbar Urethral Stricture Repair Anastomotic Urethroplasty Strictures in the bulbar urethra are most often of traumatic origin from straddle injury, endoscopic instrumentation, or catheterization. When less than 1 cm in length, these are best repaired with primary anastomosis following excision of the stricture. The patient is placed in the lithotomy position with the buttocks at the edge of the table. We have not found any exaggeration of this position to be necessary. Following retrograde injection of methylene blue into the urethra, a lambda incision is made 2 cm above the anus to the base of the scrotum. Subcutaneous tissue is divided using cautery to expose the underlying bulbospongiosus muscle. The bulbospongiosus is incised sharply in the midline and separated from the underlying bulbar urethra. Circumferential mobilization of the bulbar urethra is facilitated by sharply dividing the reflection of Buck's fascia lateral to the urethra where it drapes over the corporal bodies just proximal to the crus. A space deep to Buck's fascia and between the separating corporal bodies is entered on each side, and a right angle clamp can then be passed over the roof of the urethra to commence the dorsal dissection. This dorsal dissection, separating the urethra from the corporal bodies, is then continued sharply until the entire strictured urethra has been mobilized. Further mobility of the proximal bulbar urethra is obtained by dividing the ventral attachments to the perineal body. A 16-Fr catheter is advanced to the strictured segment and the urethra transected at this point. The strictured segment is excised proximally to expose healthy urethral tissue. It should be emphasized once again that complete removal of fibrotic tissue with exposure of healthy proximal and distal urethra is essential for successful repair, but no more than 1 cm of urethra should be excised. If the urethral ends appose in a tension-free fashion they are spatulated sharply for 1 cm. The proximal urethral segment is spatulated on its dorsal aspect and the distal end on the ventral aspect ( Fig. 51-4). Using 4-0 interrupted PGA suture the dorsal aspect of the distal urethra is spread-fixed to the underside of the tunica albuginea of the corporal bodies to anchor and stabilize the anastomosis. Circumferential full-thickness anastomotic sutures of 4-0 PGA complete the repair by approximating the proximal spatulated urethra to the spread-fixed distal segment. A 16-Fr silastic catheter, fenestrated in the bulbar portion, is used to stent the repair ( Fig. 51-5).

FIG. 51-4. Anastomotic repair of bulbar urethral stricture that may follow straddle injury. (A) The urethra is exposed through a midline perineal incision and the stricture site identified and excised. (B) Following stricture excision apposing spatulations of 1 cm are accomplished. (C) The distal urethral opening is spread-fixed to the underlying corporal body and spatulated anastomosis performed with interrupted sutures. (D) The anastomosis is completed and additional tension-relieving sutures between urethral adventitia and adjacent corporal bodies are placed to avoid anastomotic tension during erection.

FIG. 51-5. The fenestrated urethral catheter stent is optimal following urethroplasty. Either a balloon or straight catheter may be used, and fenestrations are made in the catheter shaft in the region of repair. The fenestrations allow debris and exudate to be washed away from the site of repair. Bladder irrigation cannot be performed unless the catheter is inserted all the way into the bladder so that the most distal fenestration is above the bladder neck.

A no. 7 Jackson-Pratt suction drain is placed in the perianastomotic space and the soft tissue closed using interrupted 3-0 PGA suture. The bulbospongiosus and subcutaneous connective tissue are approximated in the midline in two layers with interrupted 3-0 PGA suture. Final skin closure is made with a running subcuticular

4-0 PGA suture. The perineum is dressed and bolstered with fluff gauze. A 14- to 16-Fr Foley catheter is placed to stent the repair and is secured to the abdominal wall to minimize traction and pressure on the urethra. It may be removed between 7 and 14 days postoperatively and ideally its removal is preceded by a pericatheter urethrogram to ensure an absence of extravasation. Augmented Anastomotic Urethroplasty Bulbar urethral strictures of more than 1 cm in length cannot be repaired by the above anastomotic procedure without the risk of causing some penile chordee. Excision of 1 cm of urethra and a subsequent 1-cm spatulated repair results in a total of 2 cm of urethral shortening, an amount that can be easily accommodated by the elasticity of the mobilized urethra. Note that the distal urethra should only be mobilized to the suspensory ligament of the penis. If a 2-cm stricture is excised and a 1-cm spatulated anastomosis performed, the total shortening would be 3 cm, and this is excessive. Hence, a variety of other options may be used for strictures that are 1 cm or more in length. For strictures of up to 2 cm in length, stricture excision and augmented anastomotic repair is a good option. This procedure is similar to the above technique (anastomotic urethroplasty), the 2-cm segment being excised, but the urethra is then spatulated proximally and distally for 1 cm into healthy urethra on the same side, i.e., at the 6 o'clock position (ventral spatulation). The unspatulated dorsal urethral ends are then approximated end-to-end in a spread-fixed fashion, securing the anastomosed urethral roof strip to the overlying corporal body. The anastomosis is not completed circumferentially, resulting in a diamond-shaped ventral urethral defect into which is laid either a graft of penile skin or buccal mucosa or, alternatively, a diamond-shaped flap of penile skin mobilized on a pedicle through the scrotum. This repair is called an augmented roof strip anastomotic procedure, and it enjoys the advantages of anastomosis while augmenting the anastomotic site with a graft or flap (Fig. 51-6).9 Total penile shortening is limited to 2 cm, which is acceptable. This repair may also be performed as a floor strip anastomosis, placing the augmenting graft on the dorsal aspect where it is spread-fixed to the overlying corporal body. 1 This may be a superior technique in that the spread fixation of the augmenting graft reduces the risk of shrinkage and enhances graft take because of the apposition to the corporal body. These repairs are stented with a 16-Fr silastic Foley catheter and the perineal area is drained with a suction drain. Closure is routine and postoperative care is similar to that of the previous technique. If a graft is used, catheter removal is deferred for 3 weeks following surgery and is preceded by urethrography.

FIG. 51-6. Augmented roof strip anastomotic repair. (A-D) A healthy roof strip of urethra is created by stricture excision and roof strip reanastomosis. (E-G) An appropriately sized and shaped island of ventral penile skin is mobilized on its vascular pedicle, tucked through the scrotum, and sutured over the anastomosis augmenting its caliber. (Closure is in layers.) (H-K) This series illustrates the augmented roof strip anastomotic repair using a fenestrated full-thickness skin graft from the ventral penile shaft rather than a pedicled island of penile skin.

Onlay Bulbar Urethroplasty In the case of more extensive bulbar strictures which are usually of inflammatory etiology, it is rarely possible to complete any type of anastomotic repair without causing significant penile chordee and tension on the anastomosis. In these situations we have found success using a free skin graft onlay applied to the dorsal aspect of the strictured portion of the bulbar urethra ( Fig. 51-7).1 The urethral bulb is exposed circumferentially in the fashion previously described. A 16-Fr catheter is advanced to the site of narrowing and the urethra rotated 180 degrees to expose its dorsal aspect. Stay sutures of 4-0 silk are placed along the exposed dorsal aspect and the urethra is then incised along the strictured segment in the 12 o'clock (dorsal) position, extending the incision into healthy urethra proximal and distal to the stricture. Confirmation of proximal and distal patency is confirmed with a 26-Fr bougie à boule. A suitable skin donor site is selected on the ventral penile shaft or prepuce and a graft harvested. The graft is spread-fixed on a paraffin block and defatted using scissors. It is then fenestrated with a scalpel, and sized and shaped to fit the defect created by urethral incision. This graft is anchored in a spread-fixed fashion to the corporal bodies opposing the dorsally incised strictured urethra using 4-0 interrupted PGA suture. The urethra is then rotated back to its normal anatomic position and the margins of the urethral incision sutured to the fixed graft edge and corporal body using interrupted 4-0 PGA suture. In this fashion, the dorsal graft becomes the new urethral roof, augmenting the urethral caliber at the stricture site. A 16-Fr Foley catheter with fenestrations at the graft site is placed and the wound closed in three layers as described. This dorsal approach mitigates blood loss from a ventral bulbar incision and provides excellent graft stabilization.

FIG. 51-7. The dorsal onlay graft bulbar urethroplasty. (A) Exposure of the urethra. (B, B1) Fenestrated graft sutured to under surface of corporal body. Strictured portion of urethra either excised or rotated. (C) Suture of opened urethra to dorsal onlay graft. (D) Completed repair.

Panurethral Stricture or Failed Urethroplasty The management of extensive stricture of the anterior urethra involving both the pendulous and bulbar portions can be challenging. These are usually of inflammatory origin and can be up to 20 cm in length. Urethral repair in these circumstances is undertaken either by combination repairs using grafts and flaps or by multistaged repairs that may also use perineal inlays of full- or split-thickness skin graft. One-Stage Combination Urethroplasty One-stage combination procedures make use of both a flap repair, as described previously for the pendulous portion of the urethra, and a skin graft repair, as described for the bulbar area, with the two procedures being performed in continuity. The patient is placed in the lithotomy position for access to the perineum as well as the penile shaft. In the uncircumcised male the prepuce can be used as a skin graft donor site, leaving the remaining shaft skin for island pedicle flap construction. If the prepuce is absent, a careful assessment of stricture length must be made in order to deploy penile shaft skin most effectively for the repair. In the event there is insufficient penile skin an alternate donor site such as buccal mucosa is used for the graft repair of the contiguous scrotal/bulbar urethra.

A ventral incision is made over the pendulous urethra and this portion of the stricture is repaired using a ventral onlay repair in the fashion of Orandi, as described earlier. The more proximal scrotal and bulbar portion of the stricture is then repaired in continuity by a dorsal or ventral onlay graft of penile skin or buccal mucosa. This portion of the repair is approached through a separate incision in the perineum for exposure of the bulbar urethra. This onlay may be performed either dorsally or ventrally, although our preference is for the former because of improved stabilization of the graft. If a dorsal onlay is used, the bulbar urethra is circumferentially mobilized as far distally as the proximal limit of the penile flap repair and the urethra is then incised dorsally through the stricture, with the distal limit being the visible ventral onlay. A dorsal onlay graft is completed as described earlier using 4-0 PGA suture. Hence the long stricture is repaired by an onlay flap for the pendulous portion and an onlay graft for the more proximal portion, with the procedures being completed nose to tail. Wound drainage is established with a fenestrated suction drain. A 16-Fr silastic Foley catheter fenestrated in the bulbar region is left in place for 3 weeks. Staged Repair Extensive anterior urethral stricture disease, particularly full-length stricture, stricture complicated by fistula, inflammation, etc., long recurrent stricture following prior failed repairs and in the absence of a good donor site for urethral substitution, are best managed by staged procedures. Historically, such repairs were performed as scrotal inlay procedures; however, the resulting neourethra was then formed from scrotal skin, which has proven to give suboptimal results. Variations on this theme now inlay fenestrated full-thickness preputial or split-thickness thigh skin alongside the marsupialized urethra in the first stage so that the urethra is ultimately made from this tissue in the second stage. 2,7 Full-thickness skin is superior for this purpose but is not often available in communities in which circumcision is common. Split-thickness skin is generally harvested from the inner thigh, which is easily accessible with a patient in the lithotomy position. The strictured anterior urethra is exposed through a ventral midline perineal incision that may even bifurcate the scrotum and it is incised along the length of the stricture to healthy spongiosal tissue proximally and distally. This is checked by calibrating the urethra with a 28- and 24-Fr bougie à boule, which should pass easily through the proximal and distal ostia, respectively, at the limit of the urethrotomy. Using a dermatome set to 20 thousandths of an inch a thick-split thickness graft is harvested from the thigh and meshed to a ratio of 1:1.5. A strip of meshed graft measuring 3 to 5 cm is then inlayed around the marsupialized strictured urethra ( Fig. 51-8). Medially it is sutured to the incised urethral margin and laterally to the incised scrotal and perineal skin edges using 4-0 PGA. This width of graft allows for the up to 50% shrinkage that may occur without leaving too little skin for second-stage urethral tubularization. A suprapubic tube is placed for urinary diversion in the postoperative period and the graft is dressed with an Adaptic gauze and sterile cotton fluffs soaked in Bunnell's solution. The penis is kept on the abdominal wall and erection prophylaxis is administered. The donor site is covered with xeroform gauze and heat lamp application begun on the second postoperative day. The patient is mobilized with care the day after surgery.

FIG. 51-8. Staged urethroplasty for full-length stricture disease using meshed split-thickness skin graft. The meshed graft is laid alongside the marsupialized urethra so that the neourethra will be tubularized from the graft rather than from hair-bearing scrotal skin. Tubularization of the neourethra is delayed for 2 to 6 months.

Completion of the repair is delayed for at least 12 weeks to allow thorough vascularization of the skin graft. The proximal and distal urethral openings should again freely calibrate to 28 and 24 Fr, respectively, prior to second-stage closure. If the ostia have narrowed, revision rather than dilation should be undertaken, usually using a Y-V advancement technique. Second-stage repair is begun by incising the graft circumferentially to a width that will allow neourethral tubularization to approximately 28-Fr. The tubularization is performed by midline anastomosis using 4-0 PGA running suture, interlocking every third pass. A fenestrated 16-Fr silastic Foley is placed and the wound closed in layers using 4-0 PGA interrupted suture. Posterior Urethroplasty Posterior urethroplasty is most commonly performed for the repair of pelvic fracture urethral distraction defects (PFUDDs). The topic of urethral injury is presented elsewhere in this text, and here we will only discuss the later surgical repair of the urethral defect. The surgical options include endoscopic cut for the light procedures, perineal anastomotic repairs, abdominoperineal transpubic repairs, and staged substitution inlay urethroplasty. Short distraction defects may be appropriate for endoscopic management and a variety of techniques have been described to facilitate the procedure including cutting for the light, radiographic guidance, and use of an endoscopically guided stiletto. These techniques are ideally performed by two experienced endoscopists. Staged inlay urethroplasty is rarely required for the management of PFUDDs and is generally only performed in those patients with associated anterior urethral stricture disease. Abdominoperineal transpubic repairs (Waterhouse procedures) are also rarely required, only being indicated in some cases with associated rectourethral fistulas, bladder base fistulas, associated epithelialized pelvic floor cavities, and in some patients with associated bladder or bladder neck pathology requiring simultaneous management. Perineal Anastomotic Repair The vast majority of PFUDDs can be resolved using a perineal anastomotic repair that is generally performed 3 months or later following the initial pelvic fracture injury, with the patient managed by suprapubic catheter drainage in the interim. The length of the defect does not significantly influence the procedure, which is elaborated as it progresses to accommodate the individual needs. The patient is positioned in standard lithotomy and prepped and draped to give access to the perineum, genitalia and suprapubic region. A midline perineal incision is made, bifurcated posteriorly to improve perineal access, and dissection is taken down to the bulbospongiosus muscles, which are reflected from the bulbar urethra. The bulbar urethra is circumferentially dissected, as described earlier, proximally as far as the obliteration (identified by the passage of a 20-Fr catheter) and distally no further than the suspensory ligament of the penis ( Fig. 51-9). The urethra is transected at the obliteration, creating a distally based urethral flap. A 24-Fr van Buren sound is passed through the suprapubic tract and its tip negotiated through the bladder neck into the proximal urethra. Providing the distraction defect is not too long, the tip can be palpated from the perineal side, through the scar. In this event, a vertical incision is made through the scar onto the tip of the sound and the tip of the sound delivered into the perineal field. A long-bladed nasal speculum can now be negotiated retrogradely through the scar into the prostatic urethra, guided by the sound as it is removed. With the speculum in place incision is then performed at the 6 o'clock position, enlarging the opening in the pelvic floor scar, ultimately allowing for identification of the healthy prostatic urethral mucosa and the verumontanum. Little pelvic floor scar is excised because injudicious dissection, particularly in the anterolateral region, may damage the cavernous nerves. The posterior incision in the scar is deepened posteriorly until the hiatus is wide enough to easily accept the index finger, and generally speaking the proximal limit of the incision is at the verumontanum. In some cases where the defect is long and the tip of the descending van Buren sound cannot be palpated through the scar, one must boldly make a vertical incision in the perineal scar just behind the symphysis until the tip is felt or exposed. The bulbar urethra is spatulated on the opposing (dorsal or 12 o'clock) side and is then approximated to the prostatic urethra. If approximation is easy and tension-free the anastomosis can now be accomplished using radial 4-0 Vicryl sutures. Suture placement can be difficult, particularly if the prostate is inaccessible because it has been displaced cephalad. Sutures are placed clockwise, each suture being tagged with a hemostat, the hemostats being stacked in order. Once all sutures have been placed they are then tied in the order in which they were inserted. Suture placement is best performed by straightening the SH needle into a J shape ( Fig. 51-10). The hub of the needle is then grasped end on with the needle driver and the needle tip pushed through the proximal urethral wall from the outside in. Once the needle tip appears inside the prostatic urethral lumen it is grasped with the needle driver and advanced until the needle hub clears the urethral wall and the needle is then withdrawn. The needle is then passed through the corresponding portion of the bulbar urethra. Generally, eight sutures are placed to create a water-tight anastomosis.

FIG. 51-9. Perineal repair of pelvic fracture urethral distraction defects. (A) A midline perineal incision is bifurcated posteriorly. (B) After dissecting the bulbospongiosus muscle from the bulbar urethra, it is circumferentially mobilized proximally to the obliterative defect. Incision of the posterior urethral attachments (scissors) facilitates mobilization. (C) Once transected posteriorly the urethra is mobilized distally as far as the suspensory ligament of the penis, if necessary. (D) Penile corporal bodies are separated, right from left, from the crus distally for 5 to 7 cm. The dorsal penile vessels dorsal to the corpora lie beneath the inferior ramus of the pubic above. (E) A channel is excised from the inferior ramus of the exposed pubis between the separated corporal bodies using bone osteotome and/or bone rongeur. (F) The corporal body is circumferentially dissected and the mobilized urethra rerouted around it and through the resected bony defect. The mobilized bulbar urethra is spatulated dorsally for anastomosis to the posteriorly spatulated prostatomembranous urethra.

FIG. 51-10. (A) Bulboprostatic anastomosis is facilitated using a standard suture needle bent into a J shape. The needle is advanced through prostatic urethral edge and the needle tip retrieved in the prostatic lumen. The needle is advanced through bladder neck to clear needle and then withdrawn. (B) Suture placement is commenced with the 12 o'clock suture in the prostatic urethra. (C) Sequential sutures are then placed in a clockwise direction around the prostatic urethral opening but are not tied. Hemostats placed on the end of each suture are stacked sequentially on an Allis clamp. After approximately eight sutures have been placed they are then individually tied, commencing with the 12 o'clock suture proceeding clockwise. A catheter is inserted following suture placement and tying.

In a series of 113 cases reported by this author, the anastomosis could be accomplished using the above technique alone in only 8%. Other cases required elaboration by tension-relieving techniques that commence with corporal body separation. The corporal bodies can be separated in an avascular midline plane from the crus distally for about 5 to 10 cm, following which the mobilized urethra will lay in the groove created, shortening the distance to the anastomosis. This technique allows a further 44% of cases to be completed. The next tension-relieving maneuver, if needed, is inferior pubectomy. If the separated corporal bodies are retracted with a Weitlanner retractor the underlying dorsal venous structures can be easily dissected and mobilized laterally from the inferior surface of the pubis that lies immediately above the corporal body. Once this inferior surface has been cleared a wedge is excised using osteotome and bone rongeurs, the wedge being large enough to allow the urethra to lie within the groove. This maneuver allows for more direct routing of the urethra to the dissected prostatic apex and may require further judicious pelvic floor scar excision. This maneuver allowed a further 28% of cases to be anastomosed. The remaining 23% required the final tension-relieving maneuver, which is supracorporal rerouting of the urethra. For this maneuver one corporal body is circumferentially mobilized at or just proximal to the suspensory ligament. This dissection is not close to the corporal body, so that there is no risk of injury to neurovascular structures. A small groove of bone is removed from the symphysis just beneath the corporal body so that the urethra can course around the corporal body, lie in the bony groove, and then head toward the prostatic urethra through the previously excised pubic channel (see Fig. 51-9, inset). This markedly reduces the distance and allows all cases to be anastomosed tension-free. This maneuver does not cause significant penile torsion or chordee. Regardless of which of the tension-relieving maneuvers are used the anastomotic technique is the same, i.e., stenting is with a 14- to 16-Fr fenestrated silastic Foley catheter, and the suprapubic tube is replaced. The wound area is drained with a suction drain and closure is routine. The patient is mobilized from the bed the day after surgery and generally discharged by postoperative day 2 or 3. Catheter removal is at 3 weeks following surgery and is preceded by a retrograde urethrogram around the catheter. In the event that healing is adequate a voiding trial is undertaken and the suprapubic catheter is removed. This technique addresses the vast majority of pelvic fracture distraction defects of the urethra and carries an approximately 95% success rate. The procedure has also been used for salvage urethroplasty in previously failed situations. 11 The operation has not caused erectile dysfunction in those in whom potency has been preserved following their injury, and continence is not compromised for it resides at the bladder neck level. Only a small percentage of cases will require an abdominoperineal repair or a substitution scrotal inlay operation, as noted earlier.

OUTCOMES
Complications Hematoma/Hemorrhage Hematoma or hemorrhage is generally rare if attention is paid to hemostasis during repair. Occasionally a bulbar or cavernosal artery can be divided with subsequent retraction and spasm leading to delayed bleeding. Placement of silastic fenestrated suction drains can mitigate hematoma formation postoperatively. Infection Wound infection is an uncommon complication that generally presents with erythema and induration of incision lines. Antibiotic coverage for coliforms and skin pathogens will usually resolve this complication. As progression of wound infection can lead to graft failure and flap loss, every attempt should be made to sterilize the urine prior to reconstructive surgery to avoid tissue infection. Flap Necrosis Flap necrosis is most often due to technical errors in preparation of skin flaps or poor selection of suitable flaps. It may also be contributed to by prior surgery, infection, tissue ischemia, poor nutrition, and smoking. Graft Failure Contraction of 15% to 25% is anticipated for full-thickness skin grafts and must be allowed for during preparation. Removal of subcutaneous fat and fascia must be achieved for adequate exposure of the subdermal vascular plexus but should not be so excessive as to convert the graft to a split-thickness variety. Fenestration of free grafts will promote drainage of the graft bed, and a well-vascularized graft bed must be assumed, sometimes requiring redeployment of spongy tissue.

Split-thickness grafts are not advocated for one-stage urethroplasty because their shrinkage is unpredictable and may be excessive (up to 50%). Fistula Fistula is generally uncommon in adult urethral surgery but is occasionally encountered in the setting of underlying infection or vascular compromise. It most commonly occurs following repair of pendulous urethral strictures. Suprapubic urinary diversion may allow for spontaneous healing of fistulas encountered within the first few weeks following urethroplasty. Fistulas that fail to close with proximal diversion generally need reoperation after a minimum of 6 months observation. Chordee Penile curvature may result because of inappropriate procedure selection in attempting anastomotic repair or long-graft onlay repair to the pendulous urethra distal to the suspensory ligament. It may also follow excessive urethral excision (more than 2 cm total shortening) when performing anastomotic bulbar urethroplasty. Although considerably longer defects are bridged in the anastomotic repair used in PFUDDs, the tension-relieving maneuvers and the elasticity of the healthy bulbar urethra generally avoid this. Diverticulum Formation With use of an oversized or poorly supported ventral onlay flap or graft, particularly in the bulbar urethra, there is occasionally formation of a redundant urethral segment in which urine may pool, leading to poor urethral emptying and recurrent infection. Proper measurement of flaps and grafts to provide a 26- to 28-Fr lumen, as well as adequate spread fixation, may alleviate this outcome. Stone Formation Urethral stones can form as a result of retained hair on flaps and grafts. Preoperative epilation and proper selection of donor sites can alleviate this occurrence. Results As understanding of urethral anatomy and tissue transfer techniques has grown, so too has the effectiveness of urethral reconstruction for stricture disease. Patency rates in excess of 90% are reported for anastomotic repairs in both posterior urethral distraction injury and bulbar strictures 3,10 With regard to substitution repairs, tube grafts fare less well than onlay patches, both of which are best applied to the more proximal portion of the anterior urethra. A steady annual attrition rate of 5% per year is reported for substitution urethroplasty at follow-up of 10 years. 5 For repair of pendulous urethral strictures, pedicled flaps perform best in terms of patency and avoidance of penile curvature, but are subject to the same rate of attrition. Later attrition may prove to be less common using buccal mucosal grafts. Certainly the commonest cause for stricture recurrence appears to be ongoing fibrosis in the adjacent unmanaged urethra, particularly that proximal to the repair. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Barbagli G, Selli C, Tosto A, Palminteri E. Dorsal free graft urethroplasty. J Urol 1996;155(1):123–126. Carr LK, MacDiarmid SA, Webster GD. Treatment of complex anterior urethral stricture disease with mesh graft urethroplasty. J Urol 1997;157:104. Jakse G, Marberger H. Excisional repair of urethral stricture: follow-up of 90 patients. Urology 1986;27:233. McCallum RW. The adult male urethra: normal anatomy, pathology and method of urethrography. Radiol Clin North Am 1979;17:227. Mundy AR. The long-term results of skin inlay urethroplasty. Br J Urol 1995;75(1):59–61. Orandi A. One stage urethroplasty. Br J Urol 1968;40:717. Schreiter F, Noll F. Mesh graft urethroplasty using split thickness skin graft or foreskin. J Urol 1989;142:1223. Steenkamp JW, Heyns CF, deKock MLS. Internal urethrotomy vs dilation as treatment for male urethral strictures: prospective, randomized comparison. J Urol 1997;157:98. Warwick, RT. Principles of urethral reconstruction. In: Webster GD, Kirby R, King LR, Goldwasser B, eds. Reconstructive urology. Oxford: Blackwell Scientific, 1993;609. Webster GD, Ramon J. Repair of pelvic fracture posterior urethral defects using elaborated perineal approach: experience with 74 cases. J Urol 1991;145:744–748. Webster GD, Ramon J, Kreder KJ. Salvage posterior urethroplasty after failed initial repair of pelvic fracture membranous urethral defects. J Urol 1990;144:1379.

Chapter 52 Surgery for Urethral Trauma Glenn’s Urologic Surgery

Chapter 52 Surgery for Urethral Trauma
David M. Nudell, Allen F. Morey, and Jack W. McAninch

D. M. Nudell: Department of Urology, University of California, San Francisco, California 94110. A. F. Morey: Department of Surgery (Urology Service), Uniformed Services University of the Health Sciences, Brooke Army Medical Center, Fort Sam Houston, Texas 78258. J. W. McAninch: Department of Urology, University of California, and San Francisco General Hospital, San Francisco, California 94110. The opinions contained herein are those of the authors and are not to be construed as reflecting the viewpoints of the Armed Forces or the Department of Defense.

Anterior Urethral Injury Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Posterior Urethral Injury Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Chapter References

ANTERIOR URETHRAL INJURY
The anterior urethra extends from the membranous urethra to the external meatus. It is surrounded at all points by the corpus spongiosum and Buck's fascia. Anterior urethral injuries are rarely associated with pelvic fractures. Blunt injury is often caused by motor vehicle accidents, direct blows to the perineum, or “straddle”-type injuries wherein force directed on the relatively immobile bulbous urethra may crush it against the pubis. Penetrating trauma usually occurs from low-velocity gunshot wounds with equal involvement of the pendulous and bulbar urethra. Iatrogenic anterior urethral injuries occur from instrumentation, endoscopy, and urethral dilation. The urethra may be injured in approximately 20% of cases of penile fracture. 9 Diagnosis The diagnosis of urethral injury should be suspected from the history. A direct blow or fall on the perineum suggests anterior urethral injury. Any penetrating penile injury can cause urethral disruption. Symptoms include pain, inability to void, and hematuria. Urine or blood may extravasate only to the penile base if Buck's fascia remains intact but may spread in a “butterfly” distribution over the perineum and lower abdomen if Buck's fascia is disrupted. Blood at the external meatus is seen in approximately 70% of anterior urethral injuries. 7 All patients should undergo dynamic retrograde urethrography (RUG) before any attempt at urethral catheterization is made. The patient is positioned obliquely at a 45-degree angle with the bottom leg flexed 90 degrees at the knee and the top leg kept straight ( Fig. 52-1). A scout film is taken to insure proper positioning and radiographic technique. A 12- or 14-Fr Foley catheter is inserted into the fossa navicularis without lubrication, and the balloon is filled with 2 to 3 ml of water to prevent inadvertent dislodgment. Approximately 20 to 30 ml of undiluted water-soluble contrast agent is injected. A film is taken during injection with the penis held on slight stretch to prevent telescoping of the urethra. Films are taken again after injection and after emptying.

FIG. 52-1. Proper positioning for retrograde urethrography. The patient is placed obliquely at a 45-degree angle with the bottom leg flexed and the top leg straight. Slight traction on the penis during injection of contrast helps avoid telescoping of the urethra and allows more accurate characterization of injury.

Extravasation of contrast is diagnostic of urethral injury. A complete transection reveals extravasation without contrast seen in the bladder, whereas a partial disruption will show extravasation, but some contrast usually enters the bladder. The absence of contrast material in the bladder alone without extravasation is nondiagnostic and usually related to technique. If a catheter has already been placed into the bladder at the time of urologic consultation, it should not be removed. Instead, a plastic 16-gauge angiocatheter is placed alongside the catheter to instill contrast material for a “pericatheter” RUG when the patient is stabilized. Indications for Surgery The acute management of anterior urethral injuries depends on the extent and method of the injury. Blunt injuries causing urethral transection are generally managed with suprapubic cystostomy diversion until the associated spongiosal and soft tissue inflammatory response subsides. Primary surgical repair is avoided except in cases of concomitant penile fracture or penetrating injuries. Trochar suprapubic cystostomy is preferred in patients with penetrating urethral trauma in which there is hemodynamic instability, multiple associated injuries, or a known large urethral defect. Percutaneous tube placement may be facilitated by intravenous hydration and portable ultrasonography. Broad-spectrum antibiotics are mandatory in all patients with significant extravasation of urine or blood. Urinary diversion is maintained for 3 weeks when VCUG is performed. In cases of partial disruption, we believe the judicious urologist may gently attempt passage of a 16-Fr Coudé urethral catheter without exacerbation of the injury ( Fig. 52-2). If this is unsuccessful, no further attempt at realignment is performed, and trochar suprapubic cystostomy is done.

FIG. 52-2. Partial urethral disruption. This patient sustained partial bulbar urethral disruption after a straddle-type injury. A 16-Fr Coudé catheter was gently passed acutely and retained for 2 weeks, and the patient healed uneventfully. If the catheter does not pass easily in such instances, suprapubic cystostomy tube placement is performed.

Alternative Therapy There are no significant alternative therapies to the surgical treatment of urethral injuries. The question is whether the patient should be managed initially with conservative (i.e., establish urinary drainage) or aggressive urethral repair. Surgical Technique For immediate repair, the patient is placed supine and draped widely, as penetrating injuries often require exploration for associated injuries to the penis, testes, and perineum. Clean wounds need only minimal debridement followed by urethral closure with 6-0 Maxon sutures over a catheter. Contaminated wounds need to be thoroughly irrigated and debrided of devitalized tissue. Urethral defects of up to 25 mm in the bulbar urethra can be repaired with end-to-end anastomosis. 1 For bulbar repair, the patient is placed in high lithotomy position, and a vertical perineal incision is made. The urethra is exposed after division of the bulbocavernosus muscles. The corpus spongiosum is mobilized at the injury site, and the necessary urethral length is obtained by liberally dissecting the urethra distally. The urethral ends are spatulated, and the anastomosis is carried out using 6-0 Maxon sutures to create a tension-free, watertight closure ( Fig. 52-3). The dorsal surface is closed in one layer incorporating mucosa and spongiosum while the ventral closure is done in two layers ( Fig. 52-4).1 For pendulous repair, exposure is obtained through a subcoronal, circumferential incision. Both repairs are done over a 16-Fr silicone urethral catheter, which is removed 2 weeks postoperatively.

FIG. 52-3. Repair of bulbar urethral injuries. The urethra is mobilized and spatulated in preparation for tension-free end-to-end anastomosis.

FIG. 52-4. Repair of bulbar urethral injuries. Anastomosis is carried out with a single-layer dorsal closure and a two-layer ventral closure using interrupted 5-0 or 6-0 Maxon suture.

Outcomes Complications The complications of anterior urethral injury include infection and stricture formation. Although many strictures are of minimal clinical significance, some will require formal urethral reconstruction at a later date. Extravasated blood and urine can lead to periurethral fibrosis and abscess formation. Potential complications of infection include necrotizing gangrene, urethrocutaneous fistulas, and periurethral diverticula. 10 Unlike posterior urethral injuries, incontinence and impotence are not common.

POSTERIOR URETHRAL INJURY
Blunt injury accounts for more than 90% of prostatomembranous urethral injuries. In contrast to anterior urethral disruptions, posterior urethral injuries almost always occur in conjunction with pelvic fractures. Injury is caused by opposing forces between the prostatic urethra, which is attached to the pubis via the puboprostatic ligaments, and the membranous urethra, which is attached to the urogenital diaphragm. These injuries are often complex and have great potential for future urologic complications. Diagnosis The diagnosis of posterior urethral injury is not difficult as long as suspicion is high. Retrograde urethrography is inexpensive, readily available, and accurate and should be performed as described in the previous section ( Fig. 52-5). Almost all patients will have blood at the external meatus and a palpably distended bladder ( Fig. 52-6).7 Rectal exam may reveal the classic “high-riding” prostate in the setting of pelvic trauma; in such cases, Foley catheterization should not be attempted. It is often difficult to assess prostate position in this setting because of pelvic hematoma. Although CT scanning for associated injuries is commonly done in this setting, it is a poor modality to assess acute prostatomembranous urethral injuries. 3 If RUG is normal, the catheter should be carefully advanced into the bladder, and a formal

cystogram obtained to assess for concomitant bladder perforation, which occurs in 15% of prostatomembranous injuries. 2

FIG. 52-5. Retrograde urethrogram demonstrates copious extravasation secondary to prostatomembranous disruption.

FIG. 52-6. Physical findings suggestive of prostatomembranous disruption include a distended bladder and blood at the urethral meatus.

Indications for Surgery We believe open suprapubic cystostomy tube placement should be performed in all cases of posterior urethral disruptions because of the association with concomitant bladder ruptures. 7 If the patient is unable to undergo surgical placement of a cystostomy tube, percutaneous placement may be guided by CT or portable ultrasonography. Immediate urethral repair or reapproximation is indicated if the patient requires pelvic exploration for associated rectal or vascular injury, there is a suspected severe bladder neck laceration, or the bladder is fixed high in the pelvis by its attachment to comminuted bone fragments. 3 This last condition, known as a “pie-in-the-sky” bladder, leads to very long strictures that may be difficult to repair in a delayed fashion. Many procedures have been described to realign the ruptured urethra, including passage of interlocking or magnetic sounds and simultaneous passage of antegrade and retrograde flexible cystoscopes with guide wire and subsequent Council tip catheter placement. 2,3,4 and 5,10 These manipulations must be done gently with minimal disturbance of periprostatic tissues and pelvic hematoma. If catheter placement is not accomplished easily with the above methods, attempts at alignment are aborted, and open suprapubic cystostomy is performed. 5 Urethral injuries in women are exceedingly rare and usually occur when a bladder laceration extends through the bladder neck into the urethra and often the vaginal wall. These injuries require immediate repair to restore bladder neck function and help prevent subsequent vesicovaginal fistulas. 2 Alternative Therapy As in anterior urethral injury, the question is whether to divert the urine initially or to proceed with immediate repair. Surgical Technique For open cystostomy tube placement, the patient is placed supine, and the genitals are prepped into the field. A 6-cm infraumbilical incision is used to gain access to the prevesical space anteriorly near the dome of the bladder. The lateral perivesical spaces and space of Retzius are not entered to avoid severe bleeding that can ensue if the pelvic hematoma is disturbed. After 2-0 silk stay sutures have been placed to provide traction, the bladder is opened vertically with electrocautery. The cystotomy should be long enough to inspect the bladder neck for lacerations. Because prostatomembranous urethral injury may threaten the integrity of the external urethral sphincter, vesicle neck lacerations should be repaired at this initial procedure to avoid loss of the internal sphincter mechanism. Bladder lacerations, if encountered, are repaired in two layers using a running 3-0 Maxon suture in the muscle and 4-0 chromic suture in the mucosa. The cystotomy is then closed using a running 2-0 Maxon to obtain a watertight closure. A 24-Fr silicone Foley catheter is placed in the bladder dome and is brought out through the upper part of the midline incision ( Fig. 52-7). This position facilitates bladder neck identification during subsequent reconstruction.

FIG. 52-7. A large-caliber Foley catheter is placed in the bladder dome after inspection for concomitant bladder laceration. With the catheter placed through the superior end of the midline incision, bladder neck identification is facilitated at the time of subsequent reconstruction.

Outcomes Complications Complications of the treatment of posterior urethral injuries are directly related to the severity of the injury. The potential initial complications include infection of the

pelvic hematoma and blood loss if the pelvis is explored. Delayed complications include impotence or incontinence as well as stricture formation. Results Delayed reconstruction is preferred in most cases of posterior urethral injury at our institution. This can usually be accomplished at 3 to 6 months following the injury. Although multiple reconstructive techniques have been described, we believe that long-term impotence and incontinence are related more to the severity of pelvic injury then to the method of urethral management. Overall, impotence occurs in 50% to 80% of patients. 4,5 and 6,8 Delayed return of potency, however, is noted in up to 63% of impotent patients after definitive urethroplasty. 6 Nearly all patients remain continent if their bladder neck continence mechanism is undamaged, regardless of treatment method. Endoscopic treatment, either initial or delayed, is safe and well tolerated, but multiple subsequent internal urethrotomies are required in up to 90% of cases.4,8 CHAPTER REFERENCES
1. Armenakas NA, McAninch JW. Acute anterior urethral injuries: Diagnosis and initial management. In: McAninch JW, ed. Traumatic and reconstructive urology. Philadelphia: WB Saunders, 1996;543–550. 2. Corriere JN Jr. Trauma to the lower urinary tract. In: Gillenwater JY, Grayhack JT, Howards SS, Duckett JW, eds. Adult and pediatric urology, 3rd ed. St Louis: Mosby Year Book, 1996;563–585. 3. Dixon CM. Diagnosis and acute management of posterior urethral disruptions. In: McAninch JW, ed. Traumatic and reconstructive urology. Philadelphia: WB Saunders, 1996;347–355. 4. El-Abd SA. Endoscopic treatment of posttraumatic urethral obliteration: Experience in 396 patients. J Urol 1995;153:67. 5. Follis HW, Koch MO, McDougal S. Immediate management of prostatomembranous urethral disruptions. J Urol 1992;147:1259. 6. Koraitim MM. The lessons of 145 posttraumatic posterior urethral strictures treated in 17 years. J Urol 1995;153:63. 7. McAninch JW. Traumatic injuries to the urethra. J Trauma 1981;21(4):291. 8. Morey AF, McAninch JW. Reconstruction of posterior urethral injuries: Outcome analysis in 82 patients. J Urol 1997;157(2):506. 9. Nicolaisen GS, Melamud A, Williams RD, McAninch JW. Rupture of the corpus cavernosum: surgical management. J Urol 1983;130:917. 10. Oesterling JE, Kletscher BA. Endoscopic management of traumatic urethral injuries and bulbar urethral strictures. In: McAninch JB, ed. Problems in urology. Philadelphia: JB Lippincott, 1994;8:302–314.

Chapter 53 Artificial Genitourinary Sphincter Implantation Glenn’s Urologic Surgery

Chapter 53 Artificial Genitourinary Sphincter Implantation
James A. Dugan and David M. Barrett

J. A. Dugan and D. M. Barrett: Department of Urology, Mayo Clinic, Rochester, Minnesota 55905.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Management Patient Positioning General Principles Male Bulbous Urethral Cuff Implantation Male and Female Bladder Neck Cuff Placement Female Bladder Neck Cuff Transvaginal Approach Bladder Neck Cuff Placement in Children Implantation of Agus at the Time of Enterocystoplasty Postoperative Care Outcomes Complications Results Chapter References

The fundamental function of the urinary tract is to transport, store, and evacuate urine. The human species is endowed with the unique ability to achieve social continence. Social dryness depends on (a) accommodation of increasing volumes of urine at a low intravesical pressure accompanied by appropriate sensation of filling, (b) a bladder neck and voluntary sphincter that is closed at rest and that remains so during increases in intraabdominal pressure, and (c) absence of involuntary detrusor contractions. This ability can fail because of illness of several organ systems. It is estimated that millions of men, women, and children in this country have difficulty with urinary control. A subset of patients with urinary incontinence leak urine because of an isolated incompetent or absent sphincter mechanism. Urine loss in these patients is often quite severe and adversely affects quality of life. In the presence of normal bladder compliance, the artificial genitourinary sphincter (AGUS), an implantable prosthetic device, can restore urinary control in patients with sphincteric incontinence, The first reliable artificial sphincter (AS 721) was implanted by Scott in 1972. The history of prosthetic devices for incontinence therefore spans about 25 years. The current AS 800, introduced in 1982, represents the evolution of this science ( Fig. 53-1). The components of the system are made of Dacron-reinforced silicone elastomer and stainless steel and include a reservoir, control assembly, inflatable urethral compression cuff, and connecting tubing. The pressure-regulating reservoir controls the amount of pressure transmitted to the cuff surrounding the urethra or bladder neck. Different pressure balloons are manufactured to deliver a range of preset pressures (51 to 60, 61 to 70, 71 to 80, and 81 to 90 cm H 2O) that can be selected, based on the clinical situation and site of cuff placement. The control assembly contains a manual pump, a deactivation button, and a delayed refill resistor. Once it is activated, the patient squeezes the control assembly deflation pump, which transfers fluid out of the occlusive cuff into the pressurized reservoir. At this point, the delayed refill resistor slowly allows the cuff to automatically refill over 3 to 5 minutes to allow time for voiding.

FIG. 53-1. Components and cycling of the AS 800 AGUS. (A) Squeezing of the scrotal/labial pump transfers fluid out of the compression cuff and into the reservoir initiating voiding. (B) A delayed refill resistor slowly allows the cuff to refill automatically over 3 to 5 minutes.

The device can be deactivated (prevents fluid in the reservoir from refilling the cuff) by pushing the deactivation button after transferring fluid from the cuff to the pressurized reservoir. The surgeon deactivates the AGUS at implantation to allow the tissues to heal without urethral compression. Patients will often deactivate the AGUS at night to allow urethral tissues to have maximal blood flow, thereby minimizing the risk of cuff erosion. This deactivation button therefore allows externally controlled nonsurgical delayed activation and complete control by physician and patient over periods of cuff compression. The cuff is implanted around the bladder neck or urethra, and its size is selected on the basis of the anatomy. Cuffs are currently available in 4- to 10-cm sizes. In 1987, the narrow-backed cuff design was introduced to improve the transmission of cuff pressure to the underlying tissues and to decrease the incidence of tissue pressure atrophy and cuff erosion. A quick-connect system has been developed to supplant the hand-tied connector system. Finally, color-coded kink-resistant tubing has further improved ease of implantation and device reliability.

DIAGNOSIS
Only those patients with moderate to severe urinary incontinence, who have failed more conservative therapies, and who have documented sphincteric absence or incompetence as the principal etiology of their incontinence are candidates for the AGUS. Incontinence can result from detrusor dysfunction, sphincteric dysfunction, neurologic disease, structural changes in the bladder or outlet from trauma, surgery, or radiation, or any combination of these abnormalities. The diagnosis of sphincteric incontinence is suggested by clinical history, surgical history, physical examination, neurologic examination, urinary bacteriologic studies, renal function assessment, excretory urography, voiding cystourthrography (VCUG), retrograde urethrogram (RUG), cystourethroscopy, and urodynamic and videourodynamic evaluation. This workup must be tailored to the individual patient. History, physical examination, laboratory evaluation, radiology, urodynamics, and cystoscopy comprise the most used diagnostic tools. Excretory urography and VCUG will identify upper tract urolithiasis and vesicoureteral reflux. Upper tract calculus disease should be completely treated before AGUS implantation. Vesicoureteral reflux that was not clinically significant in a patient with total incontinence may lead to upper tract deterioration after placement of an AGUS. Reflux of grade 2 or higher should be corrected either before or at the time of sphincter implantation. Filling cystometry is the preferred urodynamics study. Complex cases may necessitate pressure–flow studies or videourodynamics. Sphincteric incontinence is characterized by normal bladder storage, sensation of filling, low urethral pressure profile, and a low Valsalva leak-point pressure. Involuntary detrusor activity can overcome sphincter cuff resistance and lead to persistent incontinence after AGUS implantation. Similarly, low detrusor compliance can cause persistent incontinence and increased intravesical pressure leading to upper tract deterioration. Either medical therapy (anticholinergics) or surgical therapy (augmentation cystoplasty) must result in a stable, low-pressure bladder before AGUS implantation. Urodynamic diagnosis of an atonic large-capacity bladder may influence the decision as to cuff site because these patients will likely have to perform clean intermittent catheterization (CIC) after surgery. If significant detrusor–external sphincter dyssynergia is

present, a decision must be made either to perform sphincterotomy before device placement or to institute CIC postoperatively. Cystourethroscopy is recommended to identify urethral and bladder defects such as stricture, false passage, diverticulum, vesical neck contracture, anatomic anomaly, and foreign body. These abnormalities should be definitively managed and considered stable before AGUS implantation. Bladder neck contracture after prostatectomy that has been incised should easily accept passage of at least a 14-Fr catheter and should remain open for at least 3 months before AGUS surgery. Retrograde urethrography may be helpful in patients with a history of recurrent urethral strictures, pelvic trauma, and/or urethral reconstruction. Midbulbous or distal urethral stricture disease or previous urethral reconstruction may prevent implantation of the cuff at this level, and an alternative site should be chosen. Visual inspection of the urethral mucosa can also give some indication of the vascular integrity of the tissues, which is particularly important in patients with a history of trauma or radiation therapy.

INDICATIONS FOR SURGERY
Urinary incontinence is an infrequent but important complication of radical perineal/retropubic prostatectomy and rarely occurs after transurethral prostatectomy as well. Patients should have had an adequate trial of anticholinergic and/or a-sympathomimetic medication along with pelvic floor exercises before AGUS implantation. A minimum of 6 months should be allowed between the time of prostatectomy and AGUS placement, for incontinence can dramatically improve during this period. These patients are excellent candidates for AGUS implantation, and the bulbous urethra is the most common site for cuff placement. In addition, patients rendered incontinent after urethral reconstructive surgery are also candidates for bladder neck AGUS placement. A number of congenital disorders including sacral agenesis and exstrophy–epispadias complex can result in loss of sphincteric continence. However, myelomeningocele is the most common cause. If these patients have suitable bladder storage and normal upper tracts, then they too may benefit from AGUS implantation. The bladder neck is often the optimal site for cuff placement in this patient population because CIC is often required postoperatively. The bladder neck is the only site for cuff placement in children. Patients with sphincteric incontinence from neurologic disease or injury, including pelvic fracture and spinal cord injury, are candidates for an AGUS, assuming that their urodynamics have stabilized and associated detrusor compliance and contractility abnormalities can be satisfactorily managed. Absolute contraindications to AGUS implantation include patients with low-volume detrusor hyperreflexia or instability, decreased compliance and a small functional bladder capacity, active stone disease, or recurrent bladder tumors (frequent instrumentation increases the chance of cuff erosion). Unstable recurrent urethral stricture disease or a urethral diverticulum at the level of the potential cuff site, active urinary, genital, perineal, or systemic infection, or insufficient dexterity or intelligence to understand the workings of the device are also contraindications. Through a combination of pharmacologic management, neurologic blockade, or additional surgical procedures such as augmentation cystoplasty, many of these patients may become suitable candidates for an AGUS. Women with hypermobility-related stress urinary incontinence are better served by a bladder neck suspension procedure. Intrinsic sphincter deficiency (ISD) in women as a result of birth trauma, aging, or prior pelvic or vaginal surgery or radiation is also best managed primarily with a vaginal reconstructive procedure (pubovaginal sling). Women with ISD from neurogenic deficits who respond poorly to conventional surgery may then be considered for AGUS implantation. The bladder neck is the only suitable site of cuff placement in women. All female patients should be instructed in CIC before AGUS implantation. Those of childbearing age must be forewarned that pregnancy is acceptable, but cesarean section may be indicated to minimize the risk of damage to the bladder neck and its surrounding cuff. Deactivation of the AGUS is recommended during the last trimester of pregnancy to reduce the risk of erosion.

ALTERNATIVE THERAPY
Alternative treatment for patients with sphincteric urinary incontinence includes clamps or urethral inserts, external collection devices, indwelling catheters or suprapubic tubes, medical or surgical therapy to improve or restore sphincteric function, urethral wall injectable bulking agents, or urinary diversion. Realistic goals and outcomes of device implantation must be discussed with patients. Patients are told that they will have significant improvement in continence, although they may not become 100% dry. They also must understand the mechanical nature of the device and accept the occasional need for reoperation to repair device components. In the event of a prosthetic infection, the device may need to be completely explanted, and the patient needs to accept the risk of this eventuality. If the patient has realistic goals and understands all risks, the operation can be undertaken, and the outcome is usually very pleasing to the patients. These can be some of your most grateful patients, because they can return to the hobbies and activities they most enjoy.

SURGICAL TECHNIQUE
Preoperative Management Because infection is a potentially devastating complication of prosthetic surgery, all patients must have a documented negative urine culture before AGUS implantation. An antiseptic shower is prescribed for the night before surgery. Bulbous urethral cuff placement requires no bowel preparation. Male patients having bladder neck cuff placement, women, and children are instructed to perform an outpatient limited lower bowel preparation the night before surgery. All female patients perform a standard vaginal douche preparation on the morning of surgery. Morning enemas can be given if the patient reports poor bowel cleansing. Patients are admitted to the hospital on the morning of surgery. Broad-spectrum intravenous vancomycin and gentamicin are given on call to the operating room and are continued throughout the inpatient hospital stay. Operating room precautions are taken to minimize the risk of bacterial seeding of the prosthesis. Hair removal from the lower abdomen, genitals, and perineum is done immediately preoperatively. Operating room skin preparation is performed for a full 10 minutes using an iodophore skin scrub. Traffic in and out of the operating suite is kept to a minimum, and all members of the surgical team wear protective hoods. Surgeons perform a standard 10-minute hand scrub. Antibiotic irrigation solution is used liberally throughout the operation. Patient Positioning The dorsal lithotomy position is used in male bulbous urethral implantation and in female patients for bladder neck implantation. This position gives excellent access to the perineum for placement of a bulbous urethral cuff in men and vaginal access for bimanual capabilities in bladder neck cuff placement in females. The supine position with the legs slightly abducted is used for men and boys undergoing placement of a bladder neck cuff. A 12-Fr Foley catheter is inserted in all male patients for bulbous urethral cuff placement, 16-Fr in all female patients and male bladder neck cuff patients, and appropriately smaller sizes in children. For bladder neck cuff placement, a rectal tube is placed to facilitate identification of the rectum in men, and an iodoform-impregnated gauze is used to pack the vagina to aid the identification of the plane between the vagina and the bladder neck in women. It is important then to isolate the rectum from the operating field with an adhesive drape, penetrating towel clamps, or sutured towel drapes. Fecal soiling of the wound during implantation can be catastrophic. General Principles The principles outlined in this section may be applied to all subsequently described operative approaches. The device should be handled delicately and as infrequently as possible. It should not be placed in prolonged contact with surgical sponges, which may shed fibers. The tubing lumen should be kept clear of blood and tissue that might enter and cause later malfunction. Only rubber-shod clamps should be used to clamp the tubing. These clamps should be closed only to the first click to avoid crush injury to the tubing that could result in a tubing leak and device malfunction. Throughout the procedure no air should be allowed into the system, for this could result in an “airlock” and device malfunction. Absolute attention must be exercised with all sharps, and nontoothed forceps are recommended once the device is in the surgical field. Although saline can be used, the authors recommend using iso-osmotic contrast material to fill the device. Device function and component location are much easier to demonstrate radiographically when contrast medium fills the AGUS. Male Bulbous Urethral Cuff Implantation The bulbous urethra is the most common site of cuff placement in the adult male patient. The 12-Fr Foley catheter should be placed using sterile technique. As the assistant holds the scrotum cephalad, and once the catheter has been palpated as it dives from the deep bulbous urethra toward the membranous portion, a 5 to 6 cm

vertical midline incision is made in the perineum over the bulbous urethra ( Fig. 53-2A). An incision that is too low will miss the catheter altogether whereas an incision that is too high will allow the scrotum and its contents to hamper exposure. The deeper fibrofatty layers of the perineum are carefully divided with sharp and blunt dissection until the bulbospongiosus muscle is identified. A Young retractor is used to gain exposure superiorly, and a Gillepie retractor provides lateral exposure ( Fig. 53-2B). Optimally, dissection should take place in the plane around the bulbospongiosus muscle and not directly on the urethra itself. This leaves another layer of tissue between urethra and cuff to reduce the risk of erosion.

FIG. 53-2. (A) Bulbous urethral perineal raphe incision for placement of bulbous urethral cuff. (B) Sharp dissection and circumferential isolation of 2-cm length of bulbous urethra.

The urethra is circumferentially dissected off the tunica albuginea of the corpora cavernosa for a length of 2 cm to accommodate the width of the cuff. This is best accomplished by grasping the bulbospongiosus muscle and urethra with the thumb and forefinger of the nondominant hand and pulling it out of the wound while spreading with the scissors tips. This gives the surgeon the tactile sense of the catheter and urethral location throughout the dissection. There are frequently penetrating bulbous urethral vessel branches at this lateral extent of the dissection. These branches are spared if possible and can also be a source of significant bleeding. A right-angle clamp is passed behind the urethra by carefully penetrating the septum of the corporal bodies at the 12-o'clock position. This is the site where most urethral injuries occur. To minimize the chance of urethral injury, again grasp the catheter with the nondominant thumb and forefinger and protect the urethra by passing the right-angle point to the forefinger tip. The thumb guards the urethra and catheter as the septum is pierced, after which a gentle spreading motion will open up the plane. A Penrose drain is passed behind the urethra, tagged, and used for traction as the 2-cm length of urethra is dissected free. If the urethral lumen is violated, it may be possible to close the defect primarily with 4-0 or 5-0 absorbable suture, and the cuff may then be placed at a more distal site during the same procedure. If a large injury occurs and the repair is questionable, an indwelling catheter should be left in place and the procedure aborted. A suspected urethral injury can be confirmed by removing the catheter and injecting antibiotic solution down the urethra using a bulb syringe or injecting methylene blue alongside the catheter with a 20-gauge angiocatheter sheath. At this point the surgeon must decide on which side the control assembly and reservoir will be implanted. The cuff tubing should be routed toward that side. Although a 4.0-cm cuff size has recently been introduced, the authors routinely implant a 4.5-cm narrow-backed cuff around the bulbous urethra. The tag is removed from the Penrose drain, the side to which the tubing will be routed is regrasped with the true right-angle clamp, and the opposite side of the Penrose is pulled. This provides gentle traction to again bring the right angle behind the urethra. The Penrose drain is released, and the end tab of the 4.5-cm narrow-backed cuff is grasped by the right angle. The empty cuff is passed tab-first behind the urethra ( Fig. 53-3), the tubing is passed through the cuff opening, and the cuff is snapped into place by gentle opposing traction applied to the tab and tubing ( Fig. 53-4). The lumen of the tubing is always shod-clamped near its end throughout this maneuver.

FIG. 53-3. The cuff is pulled behind the urethra with a right-angle clamp.

FIG. 53-4. The cuff tubing is passed through the hole in the cuff tab and then snapped into place.

The reservoir is usually placed on the side of patient hand dominance. A 5- to 7-cm transverse incision is then made in the lower abdomen over the rectus muscle on the side where the pump and reservoir will be located. Electrocautery is used to carry the incision down to anterior rectus fascia, which is then opened vertically approximately 2 to 3 cm. The fingertip is then used to split the rectus belly and then is swept circumferentially beneath the muscle. Care must be taken to avoid injury to the inferior epigastric vessels. A preperitoneal pocket is thus created for reservoir placement. A 61- to 70-cm H 2O pressure balloon reservoir is used in most patients. In high-risk patients with evidence of tissue ischemia from prior surgery or radiation, the 51- to 60-cm H 2O pressure reservoir is used to minimize tissue pressure ischemia. An angled clamp is passed down through a separate stab incision in the anterior rectus fascia, and the reservoir tubing is brought out. The reservoir is then tucked into the pocket and then test filled with 22 ml of fluid ( Fig. 53-5). The fingertip is used to ensure adequate pocket size and spherical inflation of the reservoir. The reservoir is then drained, and the fascia is carefully closed with a running 0 Vicryl suture. The reservoir is then filled with 22 ml of iso-osmotic contrast, and a shod clamp is placed near the end of the tubing.

FIG. 53-5. The cuff tubing is brought up to the abdominal incision. The reservoir balloon is then placed behind the rectus muscles, and the tubing exits the anterior fascia via a separate stab incision.

A long curved clamp is passed down over the pubis to the perineal incision in the plane inferior to Scarpa's fascia but superior to the rectus fascia. The cuff tubing is grasped and directed up into the lower abdominal wound. From the abdominal wound a fingertip or a 10-Fr Hegar dilator is passed superior to Scarpa's fascia in a lateral then inferomedial direction into the ipsilateral anterolateral hemiscrotum, thus creating a subcutaneous pouch. This pouch is then sequentially dilated to 15 Fr. As long as the surgeon stays above Scarpa's fascia, no harm will come to the spermatic cord structures. This process is made easier by wetting the dilator with antibiotic solution, driving the dilator to your opposite hand fingertips protecting the scrotal skin from a perforation, and making the pocket as subcutaneous and dependent as possible. This will allow easy patient manipulation after the postoperative edema resolves. The pump is then placed in the pouch with the deactivation button facing laterally and easily palpable through the scrotal skin. Once the pump is in a dependent position, a Babcock clamp is clamped to one click around the tubing and scrotal skin just cephalad to the pump to prevent proximal migration during tubing connection. All tubing is then cropped to the appropriate length to avoid a large amount of redundancy, and appropriate tubing connections are then made. Color-coded tubing has simplified this step. The cuff tubing is connected to the control assembly tubing with a right-angle connector while the reservoir tubing is connected with a straight connector (Fig. 53-6). Connectors are either tied in place with 2-0 Proline suture or are assembled with the quick-connect set. Gentle traction on the tubing connections will ensure the integrity of the join. All clamps are then taken off the tubing, and the cuff is allowed to pressurize. The authors do not pressurize the cuff as a separate step after it is placed around the urethra. Once the device has been tested and subsequently allowed to cycle, it is deactivated with the cuff left in the open (deflated) position. This is accomplished by completely emptying the cuff by squeezing the pump until it is flat then letting 30 seconds pass. At this point press the deactivation button. The incisions are then closed in layers with 3-0 chromic for the subcutaneous tissues then 4-0 undyed Vicryl subcuticular suture for the skin. Sterile occlusive mild-pressure dressings are applied.

FIG. 53-6. The tubing from the cuff is brought to the lower quadrant incision beneath Scarpa's fascia using a long curved clamp. Inset: After placement of the control assembly in the ipsilateral hemiscrotum, appropriate color-coded tubing connections are made: right-angle connector for the cuff tubing and a straight connector for the reservoir tubing.

Male and Female Bladder Neck Cuff Placement In adult male patients a midline lower abdominal incision is made and carried down through the anterior rectus and the transversalis fascia. In women a Pfannenstiel incision may be preferred. The retropubic space is defatted, and the endopelvic fascia is bluntly developed. The peritoneal cavity need not be violated. In adult men, a plane is then bluntly created around the bladder neck cephalad to the endopelvic fascia and prostate and caudad to the ureterovesical junction ( Fig. 53-7). With the Foley catheter in place, the demarcation between the bladder neck and the prostatic urethra can be palpated. With a combination of sharp and blunt dissection, a plane is then established posterior to the bladder neck and anterior to the rectum. Palpation of the rectal tube may facilitate identification of this plane. This portion of the dissection should be performed carefully to avoid injury to thin rectum, bladder, ureters, or urethra ( Fig. 53-8). In women, the vesicovaginal plane is then established at a point cephalad to the endopelvic fascia and caudad to the ureterovesical junction. Palpation of the Foley catheter and the vaginal pack aids in identifying this plane. Scarring from previous retropubic or bladder neck surgery may make this dissection very difficult. In these cases, it may be necessary to open the bladder to facilitate the dissection. The incision must be made in the anterior midline, well away from the bladder neck, to avoid the risk of future cuff erosion through the cystotomy closure suture line. Palpation of the anterior vaginal wall with a finger in the vagina may also help to better define the vesicovaginal plane. In men and women the width of the plane should span 2 cm to allow unrestricted placement of the cuff. Using a long right-angle clamp, pass an umbilical tape around the bladder neck and tag it. The bladder is filled with a mixture of antibiotic solution and methylene blue to identify any small tears that have occurred during dissection. Bladder or vaginal wall tears can be closed in two layers with 3-0 or 4-0 absorbable suture. However, if a rectal injury occurs, this should be closed primarily, and the procedure aborted. The bladder neck is then measured circumferentially with the measuring tape ( Fig. 53-9). If a 12-Fr Foley catheter is used, it does not need to be removed before measurement, but a 16-Fr Foley should be removed. A size 8- to 14-cm cuff is usually required for adult men, 6 to 8 cm in most adult women. Sizing is critically important because a cuff that is too small can lead to prolonged retention, whereas an oversized cuff does not correct the incontinence. A measurement of greater than 10 cm is uncommon and may represent dissection in a wide, soft tissue plane around the bladder neck. The cuff is then passed around the bladder neck and snapped into place ( Fig. 53-10, Fig. 53-11 and Fig. 53-12). The cuff tubing is passed through the rectus muscle as well as the anterior rectus fascia 4 to 5 cm above the pubis on the side of the patient where the pump will be placed. A shod clamp is placed at the end of the tubing and closed to one click.

FIG. 53-7. Sharp dissection of the vesical neck for placement of a bladder neck cuff.

FIG. 53-8. Passage of a right-angle clamp in the appropriate plane anterior to the rectum, behind the vesical neck, and measurement for cuff size.

FIG. 53-9. Measuring for the proper bladder neck cuff length.

FIG. 53-10. The bladder neck cuff is passed behind the vesical neck with a right-angle clamp.

FIG. 53-11. Final position of the bladder neck cuff in the adult man.

FIG. 53-12. The bladder neck cuff placed via the abdominal approach in the woman.

The pressure balloon reservoir is placed in the prevesical space. Its tubing is brought out through the rectus muscle near the cuff tubing and is clamped. A 61- to 71-cm H2O pressure reservoir is routinely used in most patients; however, if a larger cuff size is needed, the 71- to 80-cm H 2O pressure balloon may be used. The anterior abdominal wall fascia is then closed in the midline with absorbable suture. A Jackson Pratt drain may be left in the retropubic space if necessary. The balloon is then filled with 22 ml of iso-osmotic contrast. Male pump placement is accomplished with Hegar dilators in a similar fashion to that described above. The pump is placed in an anterolateral subcutaneous pocket in the dependent portion of the ipsilateral hemiscrotum with the deactivation button laterally positioned for easy palpation through the skin ( Fig. 53-13). In women, the pump is implanted into the ipsilateral labia majora ( Fig. 53-14). The pump is placed in a subcutaneous pocket in the dependent portion of the labia, where it can be easily palpated and manipulated. This pocket is developed by sequential dilation with Hegar dilators. Appropriate tubing connections are then made between the

AGUS components using straight connectors that are either tied in place with 2-0 Proline suture or joined using the quick-connect system. The device is tested and cycled and then deactivated with the cuff in the open (deflated) position. The incision is closed in layers with absorbable suture.

FIG. 53-13. Final position of the components of a bladder neck AGUS after appropriate tubing connections in the man (view from above).

FIG. 53-14. Anterior view of bladder neck AGUS components in adult woman.

Female Bladder Neck Cuff Transvaginal Approach After appropriate scrub, vaginal preparation, and draping of the anus out of the surgical field, the labia minora are sutured to the skin laterally, and a posterior weighted speculum is placed in the vagina for exposure. An inverted U-shaped incision is made in the anterior vaginal wall with the apex midway between the bladder neck and the urethral meatus (Fig. 53-15). The vaginal wall is dissected off the urethra back to the level of the bladder neck. With light traction on the Foley catheter, palpation of the balloon can be helpful for locating the bladder neck–urethral junction. The retropubic space is entered with curved blunt-tip scissors on either side of the bladder neck at the underside of the pubic bone. This is accomplished with the scissors tips pointing laterally and with pressure in the direction of the patient's ipsilateral shoulder. After this fascia has been pierced, a gentle spread of the scissors tips will open this space. The fingertip is then used to sweep the endopelvic fascia laterally off the pubic bone. This dissection should be done sharply if scar is encountered from prior bladder neck surgery. After the retropubic space is entered and the endopelvic fascia is reflected, the bladder neck and proximal urethra are bluntly mobilized off the underside of the pubis. This portion of the dissection is not done under direct vision, and care must be taken not to inadvertently enter the bladder neck, the urethra, or the dorsal vein complex anteriorly. Prior retropubic suspension procedures or bladder neck surgery can cause dense scarring of the anterior urethra, making the dissection quite difficult. In these cases, one author describes making a second crescent-shaped incision above the external urethral meatus and then sharply dissecting a midline plane between the urethra and the underside of the pubic bone. Very dense scar or significant bleeding requires conversion to suprapubic approach. Once the bladder neck and urethra have been completely mobilized, the Foley catheter is removed, and the bladder neck is measured with the measuring strap. The appropriate-size cuff is then snapped into place (Fig. 53-16). Cystoscopy, with or without intravenous methylene blue, can then be performed to confirm appropriate cuff location and ureteral integrity. A transverse suprapubic incision is made and carried down to the anterior rectus fascia. This is opened in the midline, and a 61- to 70-cm H 2O reservoir is placed in the prevesical space, test inflated, drained, and the fascia closed. The reservoir tubing exits the pelvis via a separate hole in the fascia made with a right-angled clamp. The pump is placed into the ipsilateral labia majora in the same fashion as described above. Under fingertip guidance, the cuff tubing is passed up into the abdominal incision using a special tubing trocar in a fashion similar to the needle passage during a bladder neck suspension procedure. After the balloon is filled with 22 ml of iso-osmotic contrast material, appropriate tubing connections are made. The device is cycled, and the cuff is left in the open (deflated), deactivated position. The abdominal and vaginal incisions are closed in layers using absorbable suture. An antibiotic-soaked vaginal pack is placed, and a 12-Fr Foley catheter is reintroduced into the bladder.

FIG. 53-15. Incision for transvaginal approach for bladder neck cuff placement.

FIG. 53-16. (A) After circumferential dissection, the cuff is passed behind the vesical neck with a right-angled clamp. (B) Final configuration of a bladder neck cuff placed transvaginally.

Bladder Neck Cuff Placement in Children The bladder neck is the only site of cuff implantation in both male and female children. The most common patients are myelodysplastics and spinal cord injury patients and will require clean intermittent catheterization postoperatively. The child must have the emotional maturity and be able to demonstrate the manual dexterity to perform CIC before AGUS implantation. The usual minimal age for sphincter implantation is 6 years for boys and 8 to 9 years for girls. Artificial sphincter implantation in children follows the same preoperative preparation and intraoperative technique as in adults with a bladder neck cuff. The child is placed in the supine position with the knees bent and legs slightly abducted. A lower midline or Pfannenstiel incision provides excellent exposure for dissection of the bladder neck. Dissection is carried out in an extraperitoneal plane. Prepubertal girls have more fragile urethral and vaginal tissues and require meticulous technique. A special cutter clamp may be necessary to develop the plane between the bladder neck anteriorly and the vaginal wall or rectum posteriorly. Occasionally, a cystotomy is required to facilitate identification of the proper posterior plane. The usual cuff size is 6 to 8 cm in children. A low-pressure, 61- to 70-cm H 2O pressure balloon is used because higher pressures in children may lead to substantial tissue atrophy and cuff erosion over time. The reservoir is positioned in the prevesical space and the pump is placed in either the scrotum or within the labia majora. A small drain may be left in the retropubic space. Lengthening of the pump tubing into the scrotum or labia is the only usual revision needed with somatic growth. The implantation of an AGUS with subsequent alteration of voiding mechanics need not be associated with change in renal growth or function. Long-term follow-up with urodynamics studies and VCUG is recommended. Implantation of AGUS at the Time of Enterocystoplasty Sphincteric incontinence can exist simultaneously with detrusor hyperreflexia and/or decreased compliance. Enterocystoplasty can increase bladder capacity and decrease intravesical pressure but does not always result in acceptable continence. Combining augmentation cystoplasty with AGUS implantation can result in urinary continence with decreased risk of upper tract damage in patients with neurogenic bladders. The augmentation is usually completed before AGUS implantation. Sterile technique with the open viscus portion of the case is critical. Copious irrigation of the pelvis and regowning and gloving should precede the AGUS portion of the case. In general, the authors bivalve the bladder high and away from the bladder neck for augmentation. Care must be taken when anastomosing the bowel segment to the bladder to allow space in the area of the bladder neck for the cuff to be placed without impinging on any suture line ( Fig. 53-17). The cuff is placed around the bladder neck using the standard technique described earlier. A suprapubic drainage tube is left in place for 14 to 21 days. A 14- to 16-Fr Foley catheter is also left in place, but this should be removed within 48 hours to prevent tissue ischemia in the area of the recently dissected bladder neck.

FIG. 53-17. AGUS placed contemporaneously with an augmentation enterocystoplasty and ureteroneocystostomy.

Postoperative Care Hospital stay for patients with a bulbous urethral cuff is 2 days; with bladder neck placement, it is 4 to 5 days. Ice packs applied to the labia or scrotum for the first 24 to 48 hours will reduce edema at the pump implantation site. Traction on the Foley catheter should be avoided, particularly in patients with a bladder neck cuff. Pain is usually minimal and can be controlled with oral agents. The urethral catheter is usually removed on the morning after implantation. After catheter removal, many patients report immediate improvement in continence, a result of increased urethral resistance from edema at the cuff site. This is usually transient, and patients are incontinent again in a few days. Most patients are able to void without difficulty. If a patient is unable to urinate, CIC with a well-lubricated small-caliber catheter is warranted until spontaneous voiding returns. Prolonged indwelling catheterization through a recently implanted cuff significantly increases the risk of erosion. If a drain was placed during surgery, it should be removed as soon as is reasonable. Preoperative intravenous broad-spectrum antibiotics are continued throughout the inpatient convalescence phase. Early ambulation is encouraged, and a general diet may be initiated after the patient has fully recovered from anesthesia. While in the hospital, patients are instructed to gently pull the pump down into the most dependent portion of the scrotum or labia once a day to avoid upward migration of the pump as capsule formation occurs. Patients are continued on oral antibiotics consisting of a cephalosporin for 14 days after discharge. They may shower daily and can use a bath 5 days after surgery. Bulbous urethral cuff patients are restricted from heavy lifting for 2 weeks, bladder neck cuff patients for 6 weeks. Patients are restricted from driving for 2 weeks. Patients are advised to abstain from sexual activity until the device is activated 6 to 8 weeks after implantation. Patients with a bulbous urethral cuff are cautioned to avoid undue compression of the perineum from prolonged sitting. Permanent restrictions include standard saddle seat bicycle or exercise bike riding and horseback riding. With appropriate seat or saddle modifications, these can be relative restrictions. Incontinence is expected to persist during the initial deactivation period, and penile compression clamps should be avoided. The patient is instructed to use either protective pads or a condom catheter. Patients are made aware of the signs and symptoms of wound infection and are instructed to contact their surgeon immediately if these are experienced. A Medic-Alert bracelet that notes their AGUS should be worn by all patients. This ensures that the treating emergency physician will know to deactivate the device before passing a urethral catheter. Patients are usually ready for AGUS activation in 6 to 8 weeks, regardless of cuff implantation site. By this time the abdominal and perineal incisions are well healed and scrotal or labial tenderness has generally subsided. Activation of the device is performed in the office by firmly squeezing the pump to compress it. The deactivation pin will “pop” into the activated position, allowing fluid to circulate from the reservoir, through the pump, to fill the cuff. If contrast was used to fill the system at the time of implantation, activation can be confirmed by inflate/deflate roentograms, which illustrate filling of the cuff after activation. Patient education is initiated at the time of the first consultation during which AGUS implantation is considered. The patient is encouraged to cycle an office demonstration device at initial visit and on return for AGUS activation. After activation, patients are given a direct demonstration of the means by which to cycle their device. Patients must demonstrate, in front of the surgeon or a member of his or her team, the independent ability to compress the pump to open the cuff and void. Patients learn that after the cuff is deflated, recompression takes 3 to 5 minutes. Although bladder emptying may occur more quickly, full continence returns only after the cuff is completely refilled. Patients with reduced flow rate, as seen with large bladders or bladder outlet obstruction, may require a second pumping of the cuff to completely empty their bladder. Urethral compression can lead to ischemia, necrosis, and erosion. This is particularly true in patients with a history of pelvic radiation. Most patients are dry when they are supine in bed at night even without urethral cuff compression. It is therefore reasonable to teach patients to deactivate their device at night. Deactivation reduces the risk of ischemia to the underlying urethra or bladder neck and helps delay tissue atrophy. Patients who require CIC to completely empty are instructed to deflate the sphincter cuff before inserting the catheter into the bladder.

OUTCOMES
Complications The most common minor complication of AGUS implantation is hematoma. These usually collect in the dependent labia or scrotum around the pump. If very large, these can displace the pump into an unfavorable high location and cause significant pain. Small hematomas can be managed expectantly, but larger collections must be opened and drained. Urinary retention after AGUS implantation usually occurs in the immediate postoperative period as a result of edema at the cuff site. The surgeon must ensure that the cuff is indeed deactivated in the open position before proceeding to any other treatment. Most patients, regardless of cuff location, begin to void after a short course of CIC with a 10- or 12-Fr catheter. Perineal ice packs and the tincture of time will usually result in spontaneous voiding and the return of incontinence in 1 to 3 days. Urinary retention after cuff activation may be the result of recurrent obstruction from bladder neck contracture or anterior urethral stricture disease. This should be evaluated with retrograde urethrography and/or cystoscopy. Treatment can be performed endoscopically with the cuff deactivated. If the need arises for prolonged bladder drainage, a suprapubic tube should be placed under fluoroscopic guidance to safely avoid the sphincter components. Indeed, in AGUS patients, any subsequent lower abdominal incision can jeopardize the safety of the device components, and fluoroscopy can be helpful to avoid the reservoir and tubing. Urinary retention in patients with neurogenic voiding dysfunction may be the first manifestation of changing bladder dynamics. Urodynamic evaluation is necessary before instituting treatment. Cuff erosion infrequently presents as urinary retention. Periprosthetic infection can present early in the postoperative period or months or years later. The overall infection rate for primary AGUS implants is approximately 1% to 3%. When reoperation is necessary, the infection rate significantly increases. Immediate postoperative infections are most often caused by skin contaminants or airborne bacterial contamination at the time of implantation. Delayed infections usually follow indolent colonization with less virulent organisms or gram-negative bacteria of genitourinary tract origin. AGUS patients should take antibiotic prophylaxis before dental or surgical procedures to avoid hematogenous bacterial seeding of the device. Indolent infection can present with subtle signs and symptoms such as pain, mild swelling, or erythema of the pump or cuff site. Patients with purulent infections, however, can present with fever and bacteremia. There may be drainage from the abdominal incision site, scrotal or perineal abscess formation, pump erosion through the scrotal skin, or cuff erosion. Treatment usually requires complete exploration of the device. Rarely, drainage of the periprosthetic infection with instillation of antibiotic solution through the drain can salvage a device. Reimplantation should not be attempted for at least 6 months, and the patient should be warned that the chance of recurrent infection is much higher than with the primary implant. Mild superficial skin infections can be safely treated with oral antibiotics and usually do not progress to deep tissue infections. Mechanical Malfunction The incidence of mechanical malfunction has dramatically decreased with each successive generation of AGUS. Specific design modifications include kink-proof tubing, reinforced joints between the tubing and device components, control pump valve improvements, and color-coded tubing, the quick-connect connecting system. Due to the continuous improvements in the design there is no contemporary study to evaluate the precise incidence of mechanical failure with the current AS 800 device. Although the exact longevity of the current AS 800 device cannot be predicted, it appears that the 5 year reliability rate is greater than 90%. The ability of the cuff to provide satisfactory continence hinges on its ability to achieve adequate urethral occlusion pressure. There is a delicate balance, however, between adequate pressure and tissue ischemia, which can lead to tissue necrosis with subsequent cuff erosion. Meticulous urethral dissection, proper cuff size, proper balloon pressure, and deactivation during healing have decreased the incidence of cuff erosion. The introduction of the narrow-backed cuff design has also contributed to the current low rate of cuff erosion. Cuff erosion can also occur as a result of periprosthetic infection. Cuff erosion can occur at any time but is most common within the first 4 postoperative months. Cuff erosion in the immediate postoperative period is certainly caused by an unrecognized iatrogenic urethral injury. Signs and symptoms of cuff erosion include recurrent incontinence, bloody urethral discharge, urinary tract infection, pain and swelling in the perineum or scrotum, or pain referred to the tip of the penis. Cystourethroscopy confirms the clinical suspicion. Erosion mandates cuff exploration. If no infection is present, a stainless steel plug can be used to cap the tubing to the cuff. New cuff implantation may be considered after a 3-month healing period. Purulent infection requires exploration of the entire AGUS device. A silicone catheter should be left in place for 4 to 6 weeks to allow the urethra to heal completely. An 18-Fr catheter is used for bladder neck erosions, whereas a 12- to 14-Fr catheter is appropriate for bulbous urethral erosions. Urethral strictures are surprisingly rare after cuff erosion. A systematic diagnostic approach is required to discover the etiology of persistent incontinence after device activation, or recurrent incontinence after successful AGUS implantation (Fig. 53-18). The three most common etiologies are mechanical failure, tissue atrophy, and cuff erosion.

FIG. 53-18. Algorithm for evaluation of persistent or recurrent urinary incontinence after AGUS implantation.

Inadvertent cuff deactivation is common and is easily treated by reactivation. Mechanical failure secondary to fluid loss can be evaluated first with an inflate/deflate abdominal roentgenogram. If no opaque contrast is seen within the device, then surgical exploration is required to discover the site of fluid loss. In the authors' experience, the most common site for leakage is the lower surface of the cuff. This is exposed first via a perineal incision, and if the cuff is not the source of the problem, then the other components are examined in a systematic fashion. The tubing connectors are the next most common site of leakage, followed by the balloon reservoir. The pump is the least frequent site of a leak. Once the site has been discovered, the appropriate components can be individually replaced. If fluid loss is confirmed preoperatively and no distinct site is found during surgical exploration, the authors recommend replacing the entire device. Tissue pressure atrophy is the natural result of cuff compression over time. Although the narrow-backed cuff design has compensated for this problem to some degree, urethral thinning is still a significant cause of recurrent incontinence. A history of slowly progressive decrease in dryness as well as an increased number of pumps needed to empty the cuff are clues to the presence of tissue atrophy. Urodynamic leak-point pressure measurement, as well as cystoscopy revealing poor cuff compression, confirms the diagnosis. Treatment can be achieved by one of three methods. Balloon reservoir pressure can be increased to the next higher category. This should be the first choice in patients with bladder neck cuffs because the reservoir is much more accessible than the cuff. Reduction in cuff size is a second option, and the cuff should be reduced in size initially by 0.5 cm. With a bulbous urethral cuff, a third option is placement of a tandem cuff. This is connected to the first device via a perineal or abdominal Y connector and is located at a site just distal to the primary cuff. Usually, no additional fluid is required. Urodynamic evaluation may also reveal poor bladder compliance, which may require surgery to correct, or involuntary detrusor contractions, which can be managed

with pharmacologic therapy. Female patients who present with continuous incontinence after AGUS implantation should be further evaluated to rule out a vesicovaginal fistula, which may have resulted from an unrecognized iatrogenic injury. Results Many reports have presented data on the success of the AS 800 AGUS in treating sphincteric incontinence. Goldwasser et al. 5 reviewed 109 AGUS patients and reported that satisfactory continence was achieved in 83% with 21% of patients requiring reoperation. The study by Fishman et al. 3 reported 90% satisfactory continence in 148 patients. Since the introduction of the narrow-backed cuff design in 1987, two large series have reviewed results. Light and Reynold 9 reported that 95% of 126 AGUS patients had obtained satisfactory continence. No patient experienced a cuff erosion, but 19% of patients were reoperated for tissue atrophy. Leo and Barrett 7 reviewed results in 144 patients. Satisfactory continence was obtained in 92% of patients. The rate of cuff erosion was 2%, and no patient required reoperation for incontinence from tissue atrophy. The most common indication for AGUS implantation is incontinence following prostate surgery. Gundian and co-workers 6 reviewed the Mayo Clinic results with AGUS implantation in 117 patients with incontinence after radical retropubic prostatectomy. Of these patients, 83% were dry (two pads or less), whereas 6% were improved but required three pads. Ninety percent of all patients believed that the AGUS had markedly improved their continence. The infection rate in this series was 2.5%. Litwiller and co-workers found a 90% satisfaction rate, and 96% of their patients said they would recommend the AGUS to a friend. 10 Fourteen percent of these patients reported improved sexual activity. Bladder neck cuff placement of the AGUS has been successfully used to treat both primary and secondary sphincteric incontinence in women. Fishman, 4 in a series of 239 patients, reported that 86% were socially continent, and with a 9-year follow-up, only 12 patients required surgery to revise the device, and 7% had an exploration because of infection. Appell reported on 34 patients implanted via the transvaginal approach, and all of these patients were completely dry. 1 Three patients required reoperation, two for tissue atrophy and one for connector leak. No patient has experienced cuff erosion or infection. Levesque and coinvestigators reviewed AGUS results in 39 children with a 10-year follow-up. 8 They reported that 82% were dry, and six patients developed renal failure, thus emphasizing the need for close follow-up when these devices are placed in patients with neurogenic bladders. Barrett and co-workers reported on 61 AGUS implantations in 59 children and young adults, 46 of whom had incontinence resulting from myelodysplasia. 2 At a mean follow-up of 43 months, 80% had a good rate of return to continence, whereas continence was considered to be fair in another 14%. Augmentation cystoplasty was necessary in 20 patients to overcome problems of increased detrusor contractility, decreased compliance, or both. The AS 800 is a reliable, safe device with proven efficacy for the treatment of sphincteric incontinence in men, women, and children. Constant design modification and manufacturing improvement have made the current sphincter the state-of-the-art prosthetic implant for the management of incontinence. Future improvements should focus on more even distribution of pressure to the bladder neck or urethra to further decrease the incidence of tissue atrophy. There will be an expanding role for the AGUS implantation in conjunction with continent urinary diversion as well as primary therapy for postprostatectomy incontinence. CHAPTER REFERENCES
1. Appell RA. Techniques and results in the implantation of the artificial urinary sphincter in women with Type IR stress urinary incontinence by a vaginal approach. Neurourol Urodyn 1988;7:613–619. 2. Barrett DM, Parulku BO, Kramer SA. Experience with AS 800 artificial sphincter in pediatric and young adult patients. Urology 1993;42:431–435. 3. Fishman IJ, Shabsigh R, Scott FB. Experience with the artificial urinary sphincter model AS 800 in 148 patients. J Urol 1989;141:307–310. 4. Fishman IJ. Female incontinence and the artificial urinary sphincter. In: Seidmon EJ, Hanno PM, eds. Current urologic therapy 3. Philadelphia: WB Saunders, 1994;312–315. 5. Goldwasser B, Furlow WL, Barrett DM. The model AS 800 artificial urinary sphincter: Mayo Clinic experience. J Urol 1987;137:668–671. 6. Gundian JC, Barrett DM, Parulkar BG. Mayo Clinic experience with the use of the AMS 800 artificial urinary sphincter for urinary incontinence following radical prostatectomy. J Urol 1989;142:1459–1461. 7. Leo ME, Barrett DM. Success of the narrow-backed cuff design of the AMS 800 artificial urinary sphincter: analysis of 144 patients. J Urol 1993;150:1412–1414. 8. Levesque PE, Bauer SB, Tala A, et al. Ten year experience with the artificial urinary sphincter in children. J Urol 1996;156:625–628. 9. Light J, Reynolds JC. Impact of the new cuff design on reliability of the AS 800 artificial urinary sphincter. J Urol 1992;147:609–611. 10. Litwiller SE, Kim KB, Fone PD, White D, Stone AR. Post-prostatectomy incontinence and the artificial urinary sphincter. A long-term study of patient satisfaction and criteria for success. J Urol 1996;156:1975–1980.

Chapter 54 Urethral Cancer in Women Glenn’s Urologic Surgery

Chapter 54 Urethral Cancer in Women
John Naitoh, William J. Aronson, and Jean B. DeKernion

J. Naitoh, W. J. Aronson, and J. B. DeKernion: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90095-1738.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Treatment of Distal Urethral Tumors Treatment of Larger Distal Lesions Treatment of Proximal Urethral Tumors or More Locally Advanced Lesions Outcomes Complications Results Chapter References

Primary urethral carcinoma is an extremely rare entity, accounting for fewer than 1% of all adult malignancies. It is one of the few urologic malignancies for which the incidence in women is greater than that in men, with a 4:1 female-to-male ratio. Most cases of female urethral carcinoma occur in patients who are postmenopausal. The mean age at diagnosis is 55 to 60 years. 6,9 The etiology of urethral cancer is unknown, and the risk factors for its occurrence are undefined in the majority of patients. Associations between urethral cancer and chronic irritation and infection have been made, including a correlation between the occurrence of cancer and a prior history of urethral caruncle and urethral diverticuli. The most common type of urethral cancer is squamous cell carcinoma (55%), followed by transitional cell carcinoma (17%) and adenocarcinoma (16%). However, in cases of urethral cancer occurring in diverticuli, the most common histologic type is adenocarcinoma. 9 Rarer types of urethral cancer include clear-cell carcinoma, undifferentiated carcinoma, and melanoma. 7,8 The disease tends to spread first by local invasion into the muscularis of the urethra and then into the periurethral connective tissue, the anterior vaginal wall and vulva, the bladder neck, and the pubic arch. Lesions in the distal one-third of the urethra spread to the superficial and deep inguinal nodes, whereas proximal lesions spread via the deep pelvic lymphatics to the periaortic chain. 8 In patients with urethral cancer, clinically evident adenopathy is likely to represent metastatic disease, with 80% to 90% of grossly enlarged lymph nodes harboring metastatic deposits. 3 Although distant metastases are uncommon at presentation, distant disease can eventually occur via hematogenous spread to the liver, lung, bone, and brain. Patients tend to fail both locally and at distant sites, with prognosis more dependent on the clinical stage at presentation than on the histologic type or grade of the primary tumor. 1,8

DIAGNOSIS
Because the symptoms of urethral cancer are subtle, the diagnosis is usually delayed, and patients commonly present with locally advanced disease at the time of diagnosis. The most prevalent symptom at presentation is urethral bleeding, spotting, or hematuria. Other symptoms include dysuria, urgency, frequency, urethral discharge, or dyspareunia. A palpable anterior vaginal wall mass can be found on bimanual examination, or an ulcerated lesion may be found on examination of the urethral meatus. Women with urethral carcinoma are often misdiagnosed with other entities such as urethral diverticuli, polyps, hemangiomas, fistulas, urethral stenosis, caruncles, or atrophic vaginitis on original evaluation but continue to have symptoms despite medical therapy. Thus, any patient with persistent symptoms that do not respond to standard therapies should be evaluated for carcinoma with urethroscopy and transurethral biopsy of any suspicious lesion. 7,9 A TNM staging system exists for urethral carcinoma, although the Grabstald system is the one that is in most common use (see Table 54-1).7,8 Careful examination of the external genitalia and bimanual examination under anesthesia at the time of tumor biopsy will help to determine if the tumor is resectable. A helpful technique is to palpate the tumor while the cystoscope sheath is in the urethra to determine the local extent of the tumor. Evaluation of the bladder neck and proximal urethra via endoscopy and biopsy should also be performed if bladder preservation approaches are to be considered. The inguinal nodes should be carefully examined because palpable lymphadenopathy usually represents malignancy. In the absence of inguinal adenopathy, there is no need for inguinal node dissection because there is no clear benefit for prophylactic lymphadenectomy. 5

TABLE 54-1. Grabstald staging of urethral cancera

Chest radiograph and CT scanning of the abdomen and pelvis are performed to look for distant metastases or for significant adenopathy. 9 In the presence of enlarged lymph nodes in the pelvis or periaortic region, CT-guided lymph node biopsy or laparoscopic node dissection can be performed; the presence of pelvic nodal metastasis portends a poor prognosis and is a contraindication to exenterative surgery.

INDICATIONS FOR SURGERY
Without treatment, most urethral cancers will progress with a mean survival of only 12 months. 6 If the patient has clinically localized disease and has a reasonable life expectancy, then curative treatment is indicated. The presence of metastatic disease or pelvic lymph node involvement is a contraindication to performing radical surgery. For superficial, noninvasive urethral lesions, topical tumor ablative treatments can be used. 8 For distal urethral lesions, a less invasive distal urethrectomy can be performed in selected cases (see below). 8 However, the presence of a histologically confirmed invasive urethral carcinoma is an indication that radical excision of the tumor and the surrounding tissues is needed. Patients may be considered for bladder preservation surgery with continent cutaneous urinary diversion if they are compliant, have adequate hand–eye coordination to perform intermittent self-catheterization, and have a creatnine less than 2.0 to avoid electrolyte disorders. If the patient cannot manage the complexities of a continent diversion, then noncontinent forms of urinary diversion, such as a suprapubic tube, vesicostomy diversion, or ileal conduit, are preferable.

ALTERNATIVE THERAPY

For superficial lesions of the urethra, endoscopic fulguration of the lesion can be performed. However, because no pathologic material is obtained during fulguration of the lesion, adequate biopsies of the lesion must be obtained before tumor destruction to rule out the presence of any invasive disease. The neodynium:YAG laser can be used safely to destroy the lesion through an endoscope, with the advantage that the laser energy penetrates up to 3 mm (at a power setting of 60 watts) into the urethral tissue and can provide for a therapeutic margin. Use of the neodynium:YAG laser has the potential risk of postoperative urethral stricture from injury to the underlying spongiosum of the urethra. For treating urethral lesions, the noncontact laser beam is applied to the tumor site for 0.5 to 1 second at an energy level of 25 to 30 watts.8 Radiation therapy (either interstitial or external beam) can also be used to cure low-stage urethral cancer. However, when it has been used for high-stage disease, the response rates to radiation therapy were low, and the complication rates were high. Side effects such as incontinence, stricture, fistula, or radiation cystitis occurred in almost 50% of patients.2 There is evidence that radiation therapy may confer a survival advantage when it is used as an adjuvant to surgery, either in cases where the tumor is locally excised or where a larger exenteration has to be performed because of the presence of high-stage disease. 8

SURGICAL TECHNIQUE
With the exception of transitional cell carcinoma, urethral cancers represent lesions that originate in the squamous and columnar epithelium of the urethra. As such, urethral cancer does not involve a “field defect” change in the entire urothelium, and the bladder transitional cell epithelium should be spared from involvement if the tumor has not grossly invaded the bladder by direct extension. 4 In patients with a squamous cell carcinoma or adenocarcinoma of the urethra, a bladder-preserving procedure can decrease patient morbidity without compromising the outcome of the cancer surgery. If a transitional cell carcinoma is found in the urethra, then the entire lower urinary tract urothelium must be considered at risk, and the patient must be treated accordingly. Treatment of Distal Urethral Tumors The technique of distal urethrectomy can be used for patients with urethral tumors that are clinically localized to the distal one-third of the urethra. Ideally, the lesion should be restricted to the meatus because resection of more than 1 cm of the urethra can result in incontinence. The patient is placed in lithotomy position on an operating table where the footrest can be either lowered or removed. The patient is positioned so that the perineum rests at the edge of the table, allowing for the placement of a weighted vaginal speculum. Number 1 silk sutures are placed into the labia and tacked to the medial thigh to separate the labia and improve exposure to the vaginal introitus. A Scott ring retractor is also placed to provide vaginal exposure. A Foley catheter is placed to facilitate palpation of the urethra and the tumor during dissection. A circumscribing incision is made around the urethral meatus as shown in Fig. 54-1. After sharp dissection has been performed through the epithelium of the vagina, sharp dissection with Metzenbaum scissors and Bovie electrocautery are used to dissect along the urethra in a circumferential manner, using palpation of the Foley as a guide to the dissection. If the distal urethra is firmly adherent to the anterior vaginal wall, it sometimes is easier to resect the anterior vaginal wall with the urethra instead of trying to preserve the anterior vaginal wall that is attached to the urethra.

FIG. 54-1. Technique of distal urethrectomy.

A figure-of-eight, 0 silk stay suture is placed at the meatus, and gentle traction of this suture allows exposure of the urethral wall during dissection. A small right-angle clamp is used to facilitate dissection along the anterior and lateral aspects of the urethra, and bleeding sites along these planes are identified and fulgurated. Dissection is extended to a point 0.5 to 1 cm beyond the palpable tumor. An anterior incision is made in the urethra beyond the tumor to the level of the Foley, and 3-0 catgut stay sutures are placed in the urethra to prevent retraction of the urethra after the distal urethra is transected. The remaining circumference of the urethra is then divided, and the specimen is removed. Frozen sections of the margins are sent to confirm that there is no residual tumor in the urethral stump. If local involvement of the anterior vaginal wall is suspected, a portion of the anterior vaginal wall can be excised in continuity with the specimen. This is done by incising the vaginal wall epithelium 0.5 to 1 cm around the periphery of the urethral mass. Pinpoint electrocautery is used to achieve hemostasis. The urethral meatus is then reconstructed by suturing the edges of the urethral stump to the edges of the vaginal incision using 3-0 chromic catgut suture. After the urethral meatus is reconstructed, the remaining defect in the vaginal wall is approximated with interrupted 3-0 polyglactic acid suture or chromic suture. A vaginal pack and urethral catheter are placed. The vaginal pack is removed on the first postoperative day, and the Foley catheter is removed 7 days after surgery. Treatment of Larger Distal Lesions For patients with squamous cell carcinoma or adenocarcinoma of the urethra not involving the bladder neck, a bladder-preserving approach can be used by closing the bladder neck and creating a continent cutaneous bladder augmentation. However, if the primary tumor is transitional cell carcinoma of the urethra, then the entire bladder urothelium is also at risk, and bladder-preserving surgery should not be performed. The patient is placed in a low lithotomy position with the legs extended in Allyn stirrups to allow for perineal access without interference of the abdominal exploration. A weighted vaginal speculum is placed, and #1 silk sutures are used to tack the labia laterally to improve exposure to the vaginal introitus. A Scott ring retractor is placed to provide further vaginal exposure. A Foley catheter is inserted to facilitate palpation of the urethra, bladder neck, and the tumor during dissection. After examination under anesthesia and careful palpation of the inguinal nodes, a midline transperitoneal lower abdominal incision is performed. The incision runs from the pubis up to a point that is approximately 6 cm above the umbilicus. Bilateral pelvic node dissections are completed and sent for frozen section. If the lymph nodes are positive, then only a pallitive procedure is indicated. If there is no evidence of nodal involvement, the bladder neck is mobilized from the lateral attachments to the pelvis by incising the endopelvic fascia. The attachments of the urethra to the urogenital diaphragm are exposed and divided lateral to the urethra with electrocautery. The uterus and ovaries are preserved. After the anterior urethra and bladder neck have been mobilized through the retropubic space, dissection is then initiated through a transvaginal approach to excise the urethra and distal bladder neck along with a margin of anterior vaginal wall. This is done by extending the circumscribing perimeatal incision into the anterior vaginal wall to a point that is 0.5 to 1 cm from the periphery of the palpable tumor. The pubourethral ligaments that connect the distal urethra to the inferior surface of the symphysis are identified using blunt dissection with a right-angle clamp; the ligaments are then divided with electrocautery. The dissection is extended under the pubic rami to the base of the bladder. Through the abdominal incision, the bladder neck is then opened anteriorly with electrocautery. The ureteral orifices are then identified, and the ureters are cannulated with 5-Fr whistle-tip catheters. The trigone is transected distal to the orifices with the incision extended through the vaginal wall to complete the resection

(Fig. 54-2). Frozen sections of the margins are obtained to confirm that all disease has been excised. The bladder neck is closed in two layers with synthetic absorbable suture (please see Chapter 49 for a more detailed description). Just before completion of bladder neck closure, the ureteral catheters are removed. Ureteral stenting is not required if the ureters were not injured during the dissection.

FIG. 54-2. Technique of urethrectomy and bladder preservation by excision of tumor at level of bladder neck, with subsequent bladder neck closure.

If there is adequate vaginal tissue remaining, the vagina can be closed primarily with 2-0 absorbable suture. However, if there is inadequate vaginal tissue, or if the patient wants to preserve sexual function, then we reconstruct the vagina by mobilizing pedicled skin flaps from the gluteal region ( Fig. 54-3). For creation of a gluteal flap, a U-shaped incision is created, running from the defect into the vagina and toward the gluteus along the posterior-medial thigh. Sharp dissection is performed along the incisions going deep into the subcutaneous tissue to create an advancement flap that has a thickness of 2 cm. Hemostasis is achieved with pinpoint cautery. This flap is then advanced and rotated to fill in the vaginal wall defect. The vaginal epithelium is anastomosed to the advancement flap with 3-0 synthetic absorbable sutures, and the external skin on the medial thigh is then approximated with a subcuticular closure.

FIG. 54-3. Creation of gluteal flap. An inverted-U incision is created on the posterior medial thigh. A 2-cm-thick flap is created and rotated into the defect in the anterior vaginal wall. A running closure is then used to secure the flap to the remaining vaginal epithelium, completing the reconstruction.

The continent cutaneous diversion is then performed, using an ileocolic segment for augmentation (see illustrations in Chapter 55). Postoperatively, the patient is managed with a 14-Fr catheter running through the ileal stoma, a 22-Fr suprapubic tube, a vaginal pack, and a pelvic drain. In general, we do not use a nasogastric tube for intestinal decompression after performing an ileocolic anastomosis, although the patient is kept NPO until the full return of bowel activity. The vaginal pack is removed 24 to 48 hours after surgery. The pouch is irrigated with normal saline beginning on the first postoperative day, and the patient is taught how to irrigate the pouch before discharge to home. The patient is maintained on antibiotic suppression until all of the drains and catheters are removed. Three weeks postoperatively, a cystogram is performed, and all tubes are removed if no leak is detected. The patient is then managed with clean intermittent catheterization. Treatment of Proximal Urethral Tumors or More Locally Advanced Lesions For patients who have large tumors or tumors involving the bladder base, bladder-preserving approaches are not feasible, and anterior pelvic exenteration is indicated. A stoma site on the abdomen is selected preoperatively for urinary diversion, and the patient should meet with an enterostomal therapist preoperatively to help prepare for surgery. The patient undergoes a full mechanical bowel prep preoperatively. Radical cystectomy, pelvic lymph node dissection, and urinary diversion are then performed in the standard fashion (see Chapter 26, Chapter 79, and Chapter 82). For larger lesions, wide excision of the vulva, pelvic floor, and pubic rami may also be indicated. In such circumstances, closure of the perineal defect with a gracilis myocutaneous flap will be needed after removal of the tumor.

OUTCOMES
Complications Treatment of urethral tumors in the distal one-third of the urethra by partial urethrectomy is well tolerated and is relatively free of complications. Incontinence has not been reported as a complication of this procedure since the internal sphincter of the bladder is located at the proximal one-third of the urethra near the bladder neck. Meatal stenosis is another potential complication, which is easily managed with intermittent dilation or meatoplasty. Complications of continent cutaneous diversion include stomal stenosis, incontinence, electrolyte disorders, vitamin B 12 deficiency, upper tract deterioration, pyelonephritis, and pouch rupture. 10 These patients require ongoing education to properly manage their cutaneous diversion to avoid complications. Furthermore, careful long-term monitoring is required to detect and treat the complications that are described above. Results Long-term outcomes following treatment for advanced urethral cancer have been poor. Overall outcome data have been difficult to categorize because of the scarcity of cases, but prognosis seems to depend more on the stage of the tumor at the time of presentation than on the histology of the primary tumor and the type of treatment utilized.1 Only 30% to 40% of patients who undergo anterior exenteration for locally advanced tumors remained disease-free after 5 years of follow-up. 1,6 Of the recurrences that were seen, 57% were localized to the pelvis, and 43% were associated with distant disease. After recurrence occurred, the mean survival overall was only 8 months, although sporadic cases have been cured with resection of the local recurrence alone. 6 In contrast, local recurrence of disease has not been a major problem in the patient who undergoes a partial or total urethrectomy for tumor treatment. This may reflect the better prognosis that is associated with smaller, distal urethral lesions as much as it represents the success of the surgical method. Five-year survival using local tumor excision in carefully selected patients to treat distal urethral carcinomas approaches 100%. 6,8 Similar excellent results have been seen in patients selected for bladder preservation approaches. In a recent series of five well-selected patients who underwent bladder-preservation surgery for urethral carcinoma, no patient had local recurrence with a mean follow-up of 39 months. 4 One patient in that series died of a concurrent ovarian cancer, and the remaining four remained free of disease at last follow-up.

CHAPTER REFERENCES
Bracken RB, Johnson DE, Miller LS, et al. Primary carcinoma of the female urethra. J Urol 1976;116:188. Garden AS, Zagars GK, Delclos L. Primary carcinoma of the female urethra. Results of radiation therapy. Cancer 1993;71:3102. Grabstald H. Tumors of the urethra in men and women. Cancer 1973;32:1236. Hedden BJ, Husseinzadeh N, Bracken RB. Bladder sparing surgery for locally advanced female urethral cancer. J Urol 1993;150:1135. Levine RL. Urethral cancer. Cancer 1980;45:1965. Mayer R, Fowler JE, Clayton M. Localized urethral cancer in women. Cancer 1987;60:1548. Mostofi FK, Davis CJ, Sesterhenn IA. Carcinoma of the male and female urethra. Urol Clin North Am 1992;19:347. Narayan P, Konety B. Surgical treatment of female urethral carcinoma. Urol Clin North Am 1992;19:373. Poore RE, McCullough DL. Urethral carcinoma. In: Gillenwater JY, Grayhack JT, Howards SS, Duckett JW, eds. Adult and pediatric urology, 3rd ed. St Louis: Mosby Year Book, 1996;1837–1852. 10. Trapasso JG, deKernion JB. Urinary diversion and continent reservoir. In: Gillenwater JY, Grayhack JT, Howards SS, Duckett JW, eds. Adult and pediatric urology, 3rd ed. St Louis: Mosby Year Book, 1996;1465–1500. 1. 2. 3. 4. 5. 6. 7. 8. 9.

Chapter 55 Carcinoma of the Male Urethra Glenn’s Urologic Surgery

Chapter 55 Carcinoma of the Male Urethra
William J. Aronson, John Naitoh, and Jean B. deKernion

W. J. Aronson, J. Naitoh, and J. B. DeKernion: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90095-1738.

Diagnosis Indications for Surgery Management of Nodes Alternative Treatments Surgical Technique Preoperative Considerations: Carcinoma of the Penile Urethra Penile Sparing Urethrectomy and Perineal Urethrostomy Preoperative Considerations: Carcinoma of the Bulbomembranous Urethra Total Penectomy, Total Urethrectomy, Radical Prostatectomy, and Continent Cutaneous Bladder Augmentation Outcomes Complications Results Chapter References

Urethral carcinoma in men is rare, with fewer than 600 cases reported through 1980. The age incidence of urethral cancer in men ranges from 40 to 70 years of age. Causal factors may include chronic irritation, urethritis, venereal disease, and urethral stricture disease. Two studies have suggested a causal role for human papilloma virus 16 (HPV-16) in squamous cell carcinoma of the urethra. Weiner et al. reported HPV-16 in four of 14 (29%) men with squamous cell carcinoma of the urethra and found HPV-16 in metastatic sites, thus suggesting a causal effect between HPV-16 and urethral cancer. 8 Cupp et al. also found HPV-16 in six of six cases of primary squamous cell carcinoma of the pendulous urethra using polymerase chain reaction DNA amplification on archival tissue. 2 To best understand the pathophysiology, treatment, and prognosis of urethral cancer in men, we can divide the urethra into three anatomic units: (a) the penile urethra, (b) the bulbar and membranous urethra, and (c) the prostatic urethra. The most frequent cell type is squamous cell carcinoma, occurring in 78% of patients, transitional cell carcinoma in 15%, and adenocarcinoma occurring in 6%. 5 Squamous cell carcinoma occurs most commonly in the penile and bulbomembranous urethra, and transitional cell carcinoma occurs most commonly in the prostatic urethra ( Fig. 55-1).

FIG. 55-1. Changes in the histology of the epithelial lining within the divisions of the male urethra tend to dictate the pathology of the tumor most likely to occur. (From Webster GD. The urethra. In: Paulsen DF, ed. Genitourinary surgery. New York: Churchill Livingstone, 1983;567.)

Squamous cell carcinoma of the urethra usually spreads by direct extension to adjacent structures or to regional lymph nodes. The lymphatics of the penile urethra drain to the superficial and deep inguinal nodes, which then drain to the external iliac nodes. Palpable inguinal nodes in a patient with carcinoma of the penile urethra usually represent metastases, unlike in penile carcinoma, when palpable nodes frequently represent infection. The lymphatics of the bulbomembranous and prostatic urethra drain to the external iliac, obturator, and presacral lymph nodes. The presence of positive inguinal lymph nodes in a patient with urethral carcinoma of the bulbomembranous urethra usually represents retrograde spread from the pelvic lymph nodes. Whereas squamous cell carcinoma of the urethra usually spreads by direct extension to adjacent structures and to regional lymph nodes, primary transitional cell carcinoma of the prostatic urethra has a propensity for hematogenous dissemination, and patients usually succumb to their disease from distant metastases.

DIAGNOSIS
Because of the extremely low incidence of urethral carcinoma and thus the low index of suspicion of practitioners, there is usually a long time delay between the time of presentation of symptoms and establishment of a diagnosis. A diagnosis of urethral cancer should be suspected in men with recurring urethral stricture disease that is progressing and refractory to treatment. Urethroscopy combined with a retrograde urethrogram should be performed in all patients with recurring or progressing urethral stricture disease to define the extent of any benign stricture and also to rule out urethral carcinoma. Urethral cancer must be suspected when urethral stricture disease is associated with complicating factors such as a bloody urethral discharge, a urethral or penile mass, phlegmon, or urethrocutaneous fistula. The diagnosis of urethral cancer is confirmed by percutaneous or transurethral biopsy of the lesion or mass. Biopsies may need to be repeated if initial biopsies are negative and the disease process persists. If local therapy or penile or urethral sparing surgery is being considered, then biopsies of the more proximal urethra should be performed to ensure a negative margin. A careful physical exam should be performed to attempt to establish the degree of local invasion. The entire extent of the urethra, corpora cavernosa, and perineum should be palpated to evaluate for local invasion into the corpus spongiosum, corpora cavernosa, and adjacent structures. Digital rectal exam may detect invasion into the membranous and prostatic urethra and rectum, and a bimanual exam should be performed to rule out a fixed pelvic mass. The inguinal nodes should be carefully palpated because clinically suspicious nodes in men with urethral cancer usually represent metastases. Magnetic resonance imaging gives excellent resolution of the primary lesion and can identify tumor invasion into the tunica albuginea and the septum between the corpora cavernosa.6 This is especially important if penile sparing surgery is being considered. The MRI may also identify the presence of local invasion of bulbomembranous urethral cancer into the pubic bone, which may help to determine preoperatively if bone resection will be required. An MRI or a CT scan should be performed to evaluate the inguinal, pelvic, and para-aortic nodes, and a chest x-ray should be obtained to evaluate the lungs. Other screening studies include intravenous urograms, bone scan, abdominal CT scan, and liver function tests. The most useful staging system for urethral carcinoma was proposed by Ray and associates and is shown in Table 55-1.4 This staging system defines the degree of local invasion and regional and distant disease. A TNM staging classification has been proposed but is not in widespread use.

TABLE 55-1. Staging of male urethral carcinoma

INDICATIONS FOR SURGERY
Urethral carcinoma is a highly aggressive and lethal disease. The median survival with no treatment or with only palliative treatment is 3 months, and the longest survival with no treatment is 15 months. If the metastatic workup demonstrates distant disease, then only palliative treatments should be performed. If pelvic CT or MRI reveals suspicious pelvic nodes, then CT-guided needle biopsy should be performed, and if the biopsy is positive, then only palliative treatments are indicated. In patients with nonmetastatic disease, however, the treatment is dependent on both the location of the tumor within the urethra and the local stage of the malignancy. For patients with biopsy-proven squamous cell carcinoma of the penile urethra, cure depends on adequate local control. For patients with localized stage 0 or stage A tumors, local resection may suffice. This can be accomplished with transurethral resection or open resection of the involved urethra with an end-to-end anastomosis. These tumors can also be fulgurated with the Nd:YAG or the CO 2 laser. For tumors involving the penile or distal bulbar urethra but confined to the corpus spongiosum, we consider performing penile sparing surgery consisting of a urethrectomy with sparing of the corpora cavernosa and creation of a perineal urethrostomy. This approach can be performed only if negative margins can be obtained of the corpus spongiosum and a 2-cm margin of tumor-free urethra can be obtained. With these types of conservative surgery, careful examination of frozen and permanent sections is required as well as careful postoperative follow-up. In general, patients with invasive carcinoma of the penile urethra should be treated with partial penectomy if a 2-cm tumor-free margin can be obtained and the patient is still able to direct his stream with the residual penile stump; otherwise total penectomy is indicated. In the case of total penectomy, if a 2-cm tumor-free margin of urethra cannot be obtained, then more extensive surgery is indicated as described below. Squamous cell carcinoma of the urethra tends to recur locally, and death in these patients usually results from complications related to local disease. If the initial surgery performed fails to obtain adequate tumor-free margins, then the surgeon should not hesitate to reoperate if adequate local resection is feasible. Whereas patients with carcinoma of the penile urethra usually respond well to surgical therapy with a high cure rate, patients with carcinoma involving the bulbomembranous urethra have a worse prognosis and usually present in advanced stages with direct tumor extension into adjacent tissues or regional node involvement. If the metastatic workup is negative and local resection is feasible, then first perform bilateral pelvic lymph node dissection. If the pelvic lymph nodes are negative, then a variety of options exist for resecting the primary tumor. Usually these patients require total penectomy, total urethrectomy, prostatectomy, and partial pubectomy if the tumor abuts or invades the pubic bone. If biopsies of the bladder neck are negative and adequate tumor-free margins can be obtained, then we perform bladder neck closure combined with a continent cutaneous bladder augmentation. If negative margins cannot be obtained at the bladder neck, then we proceed with radical cystectomy and urinary diversion. We will consider performing the above procedures with sparing of the corpora cavernosa if the primary tumor is confined to the corpus spongiosum and the margins are negative on frozen section. But in general, total penectomy is required for these patients. Primary transitional cell carcinomas of the prostatic urethra tend to metastasize by hematogenous routes, and a careful metastatic survey is required including a CT scan of the abdomen and pelvis to rule out distant disease. If the disease is clinically localized, then bilateral pelvic lymph node dissection with radical cystoprostatectomy, en bloc urethrectomy, and urinary diversion is indicated. Management of Nodes In patients with invasive carcinoma of the penile urethra, if the inguinal nodes are palpably negative, then we observe these patients with serial exams of the inguinal nodes. In patients with suspicious inguinal nodes, we first perform a bilateral pelvic lymph node dissection, remaining extraperitoneal and extending proximally to the bifurcation of the common iliac artery, distally to Cooper's ligament, laterally and superiorly to the genitofemoral nerve, and inferiorly to the obturator nerve. If frozen section reveals the pelvic nodes to be negative, then we proceed with resection of the superficial and deep inguinal nodes on the involved side and transpose the sartorius muscle over the exposed femoral vessels. On the contralateral side, if the inguinal nodes are not clinically suspicious for malignancy, then we perform a modified inguinal node dissection as described by Catalona. If the contralateral nodes are palpable, then we perform a complete nodes dissection with resection of the superficial and deep inguinal nodes and transposition of the sartorius muscle. The management of the inguinal nodes in patients with carcinoma of the bulbomembranous urethra is the same as described above for carcinoma of the penile urethra. Clinically negative inguinal nodes are observed with serial exams. Clinically positive nodes usually represent spread from pelvic nodes, and node dissection of clinically positive nodes is performed only if the pelvic lymph nodes are negative.

ALTERNATIVE TREATMENTS
Radiation therapy has had limited success in treating carcinoma of the anterior urethra. Although some success has been reported, the complications are significant and include stricture and woody penile edema as well as local recurrence resulting in death. 1,3,7 Chemotherapy has had limited success for treating urethral cancer and, like radiation therapy, is usually reserved for palliative treatment of advanced disease. Preoperative radiation therapy and chemotherapy has been proposed for bulky disease, but this is of unproven benefit. If, on the basis of clinical staging, cure seems feasible, then surgery should be performed.

SURGICAL TECHNIQUE
Preoperative Considerations: Carcinoma of the Penile Urethra Preoperatively an enema is given the morning of surgery. Ampicillin and gentamicin are given 1 hour before surgery as prophylaxis for enterococcus and the coliform bacteria. For invasive carcinoma of the penile urethra, either partial penectomy or total penectomy should be performed, and these procedures are described in Chapter 66. When performing these procedures, it is imperative to obtain a 2-cm margin of normal urethra proximal to the tumor. The margins of the urethra and surgical specimen should be examined by frozen section at the time of surgery. Penile Sparing Urethrectomy and Perineal Urethrostomy Penile sparing surgery can be performed in patients with carcinoma of the penile urethra or carcinoma of the distal bulbar urethra if the disease is limited to the corpus spongiosum. Usually this is not the case, and patients require either partial or total penectomy. To perform penile sparing urethrectomy, the patient is placed in the dorsal lithotomy position, and an inverted-U incision is made medial to the ischial tuberosities and below the base of the scrotum ( Fig. 55-2). The superficial fascia is incised vertically, and the bulbospongiosus muscle is identified, incised, and resected. The urethra is dissected off of the corpora cavernosa and surrounded with a vessel loop. While traction is placed on the urethra with the vessel loop, the remainder of the distal penile urethra is dissected off of the corpora cavernosa using pinpoint electrocautery. Bleeding from the corpora cavernosa is oversewn with 3-0 polyglycolic acid (PGA) sutures. The urethral dissection proceeds distally to the fossa navicularis, resulting in inversion of the penis. The penis is everted, the urethral meatus and fossa navicularis are excised with the electrosurgical unit using cutting current, and the urethra is drawn into the incision in continuity. The meatus is oversewn with two interrupted 3-0 PGA sutures. The urethra proximal to the

lesion is dissected and divided to ensure a 2-cm margin of normal urethra. A ¼-inch Penrose drain is placed and brought out through a skin incision.

FIG. 55-2. (A,B) Incision used for penile sparing urethrectomy. (C) The urethra is dissected off the corpora cavernosa using electrocautery. (D) Distal dissection of the urethra results in inversion of the penis. (E) The spatulated urethra is sewn to the skin edges with interrupted 3-0 chromic sutures, and the drain is brought out through a separate stab incision.

A small circle of skin is excised several centimeters ventral to the anus, and the urethra is spatulated 1 cm and sewn to the skin edges with interrupted 3-0 chromic sutures. Carefully inspect the urethra to make sure that it is not twisted or kinked and pass a catheter to rule out obstruction. The superficial perineal fascia is closed with running 3-0 PGA suture, and the skin is closed with running 4-0 PGA subcuticular suture. The drain is removed 24 to 48 hours postoperatively, and the catheter is removed 48 hours postoperatively. Preoperative Considerations: Carcinoma of the Bulbomembranous Urethra There is usually a significant delay in diagnosis of these tumors, and they usually present in advanced stages. Rarely, radical surgery can be performed with sparing of the corpora cavernosa, but in general, total penectomy combined with prostatectomy and urinary diversion is required for these patients. If a bladder-sparing procedure is being considered, obtain biopsies of the bladder neck before the procedure. This may not be possible if extensive malignancy prevents preoperative cystoscopy, and intraoperative pathologic examination of the bladder neck margin may be required. Obtain an upper tract study to ensure that both kidneys function well and to ensure normal ureteral anatomy. For a continent urinary diversion, the patient's creatinine must be less than 2.2 mg/dl to avoid electrolyte disorders. Mark out a stoma site to be used for cutaneous urinary diversion. If the primary tumor is large, contact a plastic surgeon to assist with creation of a muscle flap to fill the perineal defect resulting from the radical surgery. Total Penectomy, Total Urethrectomy, Radical Prostatectomy, and Continent Cutaneous Bladder Augmentation For home bowel preparation, begin the patient on clear liquids 2 days before the procedure, and the morning before the procedure have the patient drink 4 liters of Go-Lytely to be completed by noon. Contact the patient in the afternoon to make sure the bowel is adequately prepared and that the patient maintains high fluid intake to avoid dehydration. Oral erythromycin and neomycin are begun the evening before surgery, and the patient is given a Fleet enema the morning of the procedure. Ampicillin, gentamicin, and metronidazole are given intravenously 1 hour before the procedure. The patient is positioned in the low lithotomy position to allow simultaneous exposure to the abdomen and perineum with the table flexed above the iliac crests to allow better exposure to the pelvis and all pressure points are well padded. A midline incision is made from 3 cm above the umbilicus down to the pubis, the peritoneum is entered, and the abdomen is explored to rule out metastatic disease. A bilateral pelvic lymph node dissection is performed with the limits being the bifurcation of the common iliac artery proximally, the genitofemoral nerve laterally and superiorly, Cooper's ligament distally, and the obturator nerve inferiorly. The nodes are sent for frozen section examination, and if they are positive, the radical procedure is terminated, and a palliative procedure is indicated. If the nodes are negative, the operation is continued through a perineal approach. A circumferential incision is made around the base of the penis extending down the median raphe of the scrotum, and the incision extended down the perineum (Fig. 55-3). The dartos fascia is divided in the midline with electrocautery without entering the tunica vaginalis, as is the superficial perineal fascia. The suspensory ligament of the penis and the neurovascular bundle of the penis are divided between 0 silk ties. The corpora cavernosa are then divided between clamps at the ischial tuberosities, and the ends are oversewn with running 2-0 PGA suture. Resect the bulbar urethral mass with great care to avoid leaving positive tumor margins. If the tumor invades the scrotum, then total scrotectomy with or without orchiectomy may be required. Part or all of the urogenital diaphragm may need to be resected. If the tumor is invading the superficial periosteum of the pubic arch, then resect the periosteum with the electrocautery. If there appears to be more extensive invasion of the pubic arch, the pubic arch is scored with the electrocautery with the cutting current to obtain a 1-cm negative margin, and the inferior portion of the pubic arch is resected with an orthopedic hammer and chisel.

FIG. 55-3. (A,B) Incision used to gain access to the perineum for total penectomy, total urethrectomy, radical prostatectomy, and continent cutaneous bladder augmentation. (C) Electrocautery and suture ligatures are used to resect the urethral mass. (D) If pubic bone invasion is present, the pubic arch is first scored with electrocautery; then hammer and chisel are used to resect the inferior pubic arch to ensure a negative margin.

The surgeon then proceeds with resection of the tumor mass up to the apex of the prostate. Through the abdominal incision the prostatectomy is performed in an antegrade fashion by starting at the bladder neck and working distally. First, the endopelvic fascia is incised bilaterally, and the prostate is dissected from the bladder neck with electrocautery along with the seminal vesicles and vas deferens. The lateral prostatic pedicles are divided between clips or ties. The puboprostatic ligaments are sharply divided, and the dorsal venous complex is identified and divided between 0 PGA ties. The entire surgical specimen is submitted en bloc for frozen sections to confirm that the tumor margins at the bladder neck and throughout the specimen are negative. If any margins are positive, then a deeper resection is indicated where possible. If the bladder neck margins are negative, the continent cutaneous bladder augmentation is begun. Intravenous indigo carmine is administered or pediatric feeding tubes are passed up the ureters to verify patency. The bladder neck is closed by opposing the mucosa and inner muscle with interrupted 2-0 PGA sutures, and the outer muscle with running 0 PGA suture. The perineal defect is packed with laparotomy pads and the urinary diversion is performed through the abdominal incision. A 15-cm segment of cecum and a 10-cm segment of adjacent ileum is isolated on its vascular pedicle ( Fig. 55-4). After bowel continuity has been restored, the cecum is opened along the antimesenteric border, and the appendix is removed. The ileum is tapered around a 14-Fr catheter using the GIA stapler. The serosa is incised

along the anterior tinea of the cecum, and the tapered ileal segment within this groove is secured with interrupted 3-0 PGA sutures. A cruciate incision is made in the bladder dome with the electrocautery, and the cecal patch is attached to the bladder with running locking 2-0 PGA sutures and interrupted 2-0 PGA sutures. The distal ileum is brought out to the skin at a preselected stoma site, any redundant ileum is excised, and it is verified that the 14-Fr catheter passes easily through the ileal stoma into the bladder. Alternatively, instead of tunneling the efferent limb, continence can be obtained by reenforcing the ileocecal valve with interrupted 3-0 silk sutures.

FIG. 55-4. (A) Mobilize a 15-cm segment of cecum and adjacent 10 cm of ileum. (B) Excise the staple lines and detubularize the cecum by incising along the tinea as shown. (C) Taper the ileum over a 14-Fr catheter using a GIA stapler. (D) Suture the detubularized cecum to the bladder. (E) Create flaps of the posterior tinea. (F) Secure the tinea flaps to the lateral wall of the ileal efferent limb.

The pouch is drained with a 14-Fr catheter running through the efferent limb of ileum and a 24-Fr suprapubic tube. We close the pelvic defect by sewing Marlex mesh to the urogenital diaphragm with interrupted 0 Proline sutures. Alternatively, a gracilis muscle flap or rectus abdominis muscle flap can be dissected by a plastic surgeon and used to close the pelvic defect. The superficial perineal fascia is closed with running 3-0 PGA suture, and the skin is closed with running 4-0 chromic suture. If there is insufficient skin to bring the skin edges together, then a pedicled gluteal flap can be created to close the skin defect (see Chapter 54). A Jackson Pratt drain is placed between the mesh and the superficial perineal fascia and is brought out through the perineum. Large Penrose drains are also placed through the abdominal wall into the pelvis before the abdominal wall fascia is closed. A cystogram is performed 3 weeks postoperatively, and if there are no leaks, the drains are removed, and the patient is taught self-catheterization. Antibiotics are continued until all drains are removed. Note: To avoid the severe disfigurement of total penectomy in patients with early invasion through the corpus spongiosum, the penile skin and glans penis can be left intact. To accomplish this, the spongiosum and cavernosum are resected completely, leaving the glans to receive blood supply through the neurovascular bundle and penile skin. This is a less mutilating procedure, and the glans penis and penile skin may be used for reconstruction at a later date, if desired.

OUTCOMES
Complications Complications of transurethral resection or laser treatment of superficial urethral carcinoma include urethral stricture and urethrocutaneous fistula. These patients need to be followed closely postoperatively and, at the first sign of urethral stricture disease, treated with daily self-dilation. If this is unsuccessful, then more extensive urethroplasty procedures may be required. The main complication resulting from partial and total penectomy is stenosis of the meatus or perineal urethrostomy. This can be managed either with self-dilation or meatoplasty. Local recurrence of tumor can cause outlet obstruction at the meatus or anywhere along the urethra, and suspicious lesions must be biopsied and treated accordingly. The main complications of radical resection of bulbar urethral tumors include pelvic abscess and perineal wound breakdown. This can subsequently lead to severe intra-abdominal complications and sepsis. These are extremely morbid complications and can only be prevented by careful attention to all technical aspects of the original operation including placement of large pelvic drains. Postoperative management must include ongoing irrigation of mucus plugs from the continent cutaneous diversion, and ongoing monitoring of the pelvic drains with careful attention to assure that the drains are not prematurely removed. Delayed postoperative fever requires appropriate imaging and prompt drainage of any fluid collections to prevent a large abscess from developing. Resection of the pubic bone may lead to osteitis pubis and may require orthopedic consultation and long-term antibiotics. Results In well-selected patients with superficial squamous cell carcinoma or transitional cell carcinoma of the urethra, transurethral resection, laser fulguration, or local excision offers excellent cure rates approaching 100%. 9 Recurrence in these patients is not uncommon, and close follow-up is mandatory. In patients with squamous cell carcinoma of the penile urethra who undergo partial penectomy, in whom a 2-cm tumor-free margin is obtained, the results are also excellent and approach 100%.9 Results of radical penectomy for squamous cell carcinoma of the urethra are not as good, and the disease-free rate reported is roughly 40%. This probably reflects the fact that these patients initially present with more advanced disease. In patients with carcinoma of the bulbar urethra, serious consideration should be given to performing more radical surgery to obtain good tumor-free margins. The overall survival for patients presenting with squamous cell carcinoma of the proximal bulbar and membranous urethra is roughly 10% because of the advanced nature of this disease. Of patients with disease amenable to surgery who undergo radical excision including radical penectomy and prostatectomy with or without cystectomy, 38% are reported to be free of disease. In the last 55 patients at UCLA in whom we performed the tunneled continence mechanism, the daytime continence rate has been roughly 96% with 1 year of follow-up. Of the two failures, one patient had a small contracted bladder and required additional augmentation, and the other patient required revision of the continence mechanism. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Bracken RB, Henry R, Ordonez N. Primary carcinoma of the male urethra. South Med J 1980;73:1003. Cupp MR, Reza MS, Goellner JR, Espy MJ, Sniith TF. Detection of human papillomavirus DNA in primary squamous cell carcinoma of the male urethra. Urology 1996;48:551. Kaplan GW, Buckley GJ, Grayhack JT. Carcinoma of the male urethra. J Urol 1967;98:365. Ray B, Canto AR, Whitmore WF Jr. Experience with primary carcinoma of the male urethra. J Urol 1977;117:591. Sullivan J, Grabstald H. Management of carcinoma of the urethra. In: Skinner DG, deKernion JB, eds. Genitourinary cancer. Philadelphia: WB Saunders, 1978;419. Vapnek JM, Hricak H, Carroll PR. Recent advances in imaging studies for staging of penile and urethral carcinoma. Urol Clin North Am 1992;19:257. Webster GD. The urethra. In: Paulson DF, ed. Genitourinary surgery. New York: Churchill Livingstone, 1984;399. Wiener JS, Liu ET, Walther PJ. Oncogenic human papillomavirus type 16 is associated with squamous cell cancer of the male urethra. Cancer Res 1992;52:5018. Zeidman EJ, Desmond P, Thompson IM. Surgical treatment of carcinoma of the male urethra. Urol Clin North Am 1992;19:359.

Chapter 56 Seminal Vesicle and Ejaculatory Duct Surgery Glenn’s Urologic Surgery

Chapter 56 Seminal Vesicle and Ejaculatory Duct Surgery
Paul J. Turek

P. J. Turek: Department of Urology, University of California, San Francisco, California 94143–0738.

Seminal Vesicle Surgery Diagnosis Indications for Surgery Alternatives to Surgery Surgical Technique Outcomes Ejaculatory Duct Surgery Diagnosis Indications for Surgery Alternatives to Surgery Surgical Technique Outcomes Chapter References

SEMINAL VESICLE SURGERY
The seminal vesicle is a uniquely male organ, a swelling derived from the mesonephric duct beginning at 13 fetal weeks. The normal adult seminal vesicle is 5 to 8 cm in length and 1.5 cm in width, with a volume of approximately 10 ml. Histologically, the organ wall is composed mainly (80%) of smooth muscle and is lined by columnar epithelium. The blood supply is derived from the deferential artery or, occasionally, from branches of the inferior vesical artery. The seminal vesicles receive primary innervation from adrenergic fibers supplied from the hypogastric nerve. Primary pathologic states of the organ are rare. Secondary lesions constitute the major reason for organ removal. Congenital lesions of the seminal vesicles include ureteral ectopy, seminal vesicle cysts, and aplasia. The vast majority of men with cystic fibrosis, and 1% of infertile men, present clinically with absence of the vas deferens and seminal vesicles. Infections of the seminal vesicles are uncommon, but tuberculosis and schistosomiasis are not infrequent causes of masses, abscesses, and calcifications in developing countries. Chronic bacterial seminal vesiculitis, a rare and difficult diagnosis, is presumably a consequence of bacterial prostatitis. Benign tumors of the seminal vesicles include papillary adenoma, cystadenoma, fibroma, and leiomyoma. Malignant neoplasms are extremely rare (100 reported cases) and include papillary adenocarcinoma and sarcoma. 1,9 Radical excision of the organ is the accepted treatment of seminal vesicle malignancies. Far more common than primary malignancies is secondary involvement from carcinoma of the bladder, adenocarcinoma of the prostate or rectum, and lymphoma. Diagnosis The normal seminal vesicle is usually not palpable on digital rectal examination. The area superior to the prostate can be enlarged and compressible in the presence of seminal vesicle dilation or firmly indurated if the organ contains tumor. These findings constitute indications for further evaluation. Transrectal ultrasound (TRUS) is the diagnostic modality of choice to evaluate seminal vesicle size and pathology. In addition to high-resolution delineation of anatomy, TRUS facilitates needle aspiration for cytology or core biopsy for tissue diagnosis. Even greater anatomic detail can be obtained with CT scan or MRI, although their use should be limited to the staging of solid lesions within the pelvis and to confirm the hemorrhagic nature of suspicious masses. In infertile patients who present with azoospermia, or in patients with hematospermia or coital pain, TRUS is generally the initial imaging study. Contrast vasography performed transurethrally or through the scrotal vas deferens accurately images the vas deferens and ampullary–seminal vesicle junction but is less reliable than TRUS for seminal vesicle pathology. Recently, there has been renewed interest in TRUS-guided direct puncture seminal vesiculography for both imaging and aspiration. Indications for Surgery Most procedures performed on the seminal vesicles are related to radical surgery for the treatment of urethral, prostate, bladder, or rectal cancer. Treatments specific to the seminal vesicle include: Transrectal aspiration of cysts or abcesses Transurethral unroofing of abcesses, obstructing cysts, or allowing stone passage Open resection to treat refractory infections, to excise an ectopic ureter, or to remove benign or malignant masses. Cysts are treated in the presence of symptoms referable to a mass effect. Surgical drainage of abcesses usually follows the failure of antibiotic therapy. The excision of solid lesions of the seminal vesicles is warranted after aspiration or biopsy confirmation of pathology. This section emphasizes the open surgical approaches to the seminal vesicle. Alternatives to Surgery Few established alternatives to surgery exist for the treatment of tumors of the seminal vesicles. Infections can often be appropriately treated with antibiotics. Many asymptomatic congenital anomalies can be observed. Surgical Technique Although a variety of open surgical approaches to the seminal vesicles has been described, the most useful are the transvesical, transperineal, paravesical and retrovesical, and transcoccygeal methods, as illustrated in Fig. 56-1. The approach that is chosen depends, in part, on nature of the lesion to be excised and, more importantly, on the experience of the surgeon. For most of these approaches, one or two units of autologous blood should be available for emergency use.

FIG. 56-1. The four different surgical approaches to seminal vesicle surgery.

Transvesical Approach The patient is positioned supine, and an infraumbilical extraperitoneal incision is made. 6 The incision can be midline or Pfannenstiel. The rectus muscles are separated, and the space of Retzius is entered. A Balfour retractor is used to expose the anterior bladder wall. The bladder wall is opened with a vertical (7- to 10-cm) incision that extends to no closer than 2 to 3 cm from the anterior bladder neck. Full-thickness bladder wall stay sutures of 2.0 Vicryl are placed at either end of the incision for traction and also to prevent tearing of the incision with retraction. The lateral blades of the Balfour retractor are repositioned within the bladder to expose the trigone and posterior wall. Several moist 4 × 8 sponges are packed in the bladder dome, and a Deaver blade is placed superiorly to stretch the bladder interior and flatten the posterior wall. It is advisable to intubate the ureters with #8 feeding tubes or ureteral stents to define the course of the intramural ureters. With a cutting Bovie stylet, a 5-cm vertical incision is made through the trigone ( Fig. 56-2A). Beyond the thick, deep muscle of the bladder wall near the bladder neck, the ampullae of the vas can be recognized (Fig. 56-2B). With sharp dissection, the seminal vesicles can be identified lateral to the ampullae at the prostate base. In the absence of prior inflammatory disease, the plane around the seminal vesicles is easily entered. They can be dissected free, ligated at the prostate base with 2.0 chromic suture, and then divided. A metal clip placed across the cut end of the seminal vesicle minimizes spillage of organ contents in the field. The distal vascular pedicle should be identified and controlled with small metal clips or ties, and the organ removed.

FIG. 56-2. The transvesical approach to the seminal vesicle. (A) Vertical incision in the trigone to expose the retrovesical seminal vesicle and, (B) Dissection of the vas ampullae and seminal vesicles.

Prior inflammation can make the dissection quite difficult. In this situation, ureteral catheters are useful to help avoid ureteral injury. It is also important to avoid too deep a dissection and risk violating Denonvillier's fascia and entering the rectum. The posterior bladder incision is closed in two layers: a running 2-0 absorbable suture for the muscle layer and a running 4-0 absorbable suture in the mucosal layer. After removal of the sponge packs and stents, a urethral catheter is placed, and the anterior bladder wall is closed in a manner similar to the posterior wall. Suprapubic drainage of the bladder is optional. A suction drain is left for 2 to 3 days and removed if drainage is less than 60 ml/day. The urethral catheter is removed in 7 to 10 days. Early postoperative ambulation is the rule. Patients may be discharged with or without the drain when pain is controllable with oral medication, generally within 3 to 4 days of surgery. Transperineal Approach The patient should undergo preoperative outpatient bowel preparation with citrate of magnesia or Go-Lytely and receive broad-spectrum antibiotics systemically before the procedure. Intermittent compression stockings are recommended to prevent thromboembolic events. The transperineal approach to seminal vesiculectomy is virtually identical to that of the radical perineal prostatectomy described elsewhere. An exaggerated dorsal lithotomy position is used to elevate the perineum so that it is parallel to the floor. An inverted-U incision is made in the perineum, and the central tendon is divided (Fig. 56-3A). An anterior retractor stretches the rectal sphincter superiorly, allowing visualization of the glistening anterior rectal fascia fibers. The rectourethralis is divided near the prostatic apex, and a weighted speculum is placed, dropping the tented rectum. To adequately expose the seminal vesicle, the rectum should be dissected off the posterior surface of the prostate to a point higher than that needed for perineal prostatectomy. Denonvillier's fascia is then incised transversely or in the midline (if nerve-sparing) on the prostate near the base of the seminal vesicles.

FIG. 56-3. The transperineal approach. (A) U-shaped incision in the perineum with takedown of the central tendon. (B) Exposure and ligation of the seminal vesicles after incising Denonvillier's fascia.

Seminal vesicle dissection is facilitated by posterior traction on the prostate provided by placement of a Lowsley retractor in the bladder. Medially, the ampullae and seminal vesicles are apparent after Denonvillier's fascia is incised ( Fig. 56-3B). The ampullae can be spared for the excision of a simple seminal vesicle cyst or small tumor but may need resection if the surgery is for cancer or infection. After dissection of the seminal vesicle at the prostatic base, an absorbable tie of 2.0 suture is used to ligate the gland ( Fig. 56-3B). Before division of the seminal vesicle, a clip is placed on the cut end of the organ to minimize spillage of gland secretions. An Allis or Babcock clamp is then placed on the freed base of the seminal vesicle to ease the apical dissection. I have found that dividing the organ at the base of the prostate first and dissecting distally along the seminal vesicle is easier than tip dissection followed by base ligation. The vascular pedicle at the apex of the gland is usually observed within 1 cm of the tip. This is ligated with small metal clips, and the gland is removed. The wound is closed in layers as outlined for a perineal prostatectomy. A Penrose drain is left in the area of dissection for 24 hours or until no drainage is noted. This approach is well tolerated, and patients are usually discharged within 24 to 48 hours of surgery. Paravesical and Retrovesical Approaches The paravesical approach is suitable for large, unilateral seminal vesicle cysts and for the correction of congenital anomalies in children. A vertical infraumbilical midline or Pfannenstiel incision is used to expose the space of Retzius; the bladder is bluntly dissected away from the pelvic sidewall. The vas deferens can be tracked medially toward the bladder base to help locate the seminal vesicle. The plane between the bladder and seminal vesicle is developed from laterally to medially. This dissection is facilitated by emptying the bladder with a Foley catheter. As the seminal vesicle is dissected, be aware that the vas deferens crosses over the ureter and that the ureter can easily be injured. As the plane between the bladder and seminal vesicle develops, roll the bladder medially for better exposure. At

the level of the prostate, do not dissect lateral to the seminal vesicles, as this may damage the neurovascular bundle. The neck of the seminal vesicle can be defined at the prostate base and the organ ligated with 2.0 absorbable suture. A drain may be left in place in the bed of the dissection and brought out through a separate stab incision. Usually the drain can be removed 24 hours later and the Foley catheter removed within 48 hours. The recovery time is rapid, with discharge approximately 48 hours after surgery. The retrovesical approach is appropriate in cases of bilateral seminal vesicle excision for small cysts or tumors. 2 Generally, a midline, infraumbilical, intraperitoneal incision is made to gain access to the dome of the bladder and the cul-de-sac between bladder and rectum (Fig. 56-4A). With intra-abdominal exposure, the small and large bowel are gently packed superiorly, and the peritoneal reflection near the posterior bladder is incised transversely over the rectum ( Fig. 56-4B). It is important not make the incision too deep and risk injury to the rectum. With careful sharp dissection, the bladder is peeled forward off of the rectum until the ampullae and the seminal vesicle apices are visualized ( Fig. 56-4C). In a manner similar to the paravesical approach, the seminal vesicles are dissected down to the prostatic base, the neck of the gland is encircled, and the organ is ligated with 2.0 absorbable suture. A suction drain can be left in the bed of dissection and secured as described above. Postoperative care is also similar to that of the paravesical approach except that the return of bowel function may be delayed 24 hours with the intra-abdominal dissection.

FIG. 56-4. The retrovesical approach. (A) Midline infraumbilical incision. (B) Incision of the posterior peritoneum over the rectum in the cul-de-sac. (C) Exposure of the vas ampulla and seminal vesicle behind the bladder.

Transcoccygeal Approach Urologists are generally not as familiar with transcoccygeal surgery as they are with the previously described methods. 4 This dissection is appropriate for individuals in whom the perineum or lower abdomen is not accessible because of limitations of patient positioning or because of multiple prior surgeries in these anatomic regions. Because it is an uncommon surgical approach, it will be discussed only briefly. The patient is placed in the prone, jackknife position. An incision is made along the coccyx and angled into a gluteal cleft ( Fig. 56-5A). The coccyx is removed, and the gluteus maximus muscle layers are retracted laterally to expose the rectosigmoid colon. After Denonvillier's fascia has been incised medial and deep to the rectum, it is dissected off the prostate with exposure of the seminal vesicles ( Fig. 56-5B). Injury to the neurovascular bundle is more likely with this approach because the bundle is superficial to the prostate and therefore directly in the path of dissection. After the seminal vesicles are removed, the rectum is carefully inspected for injury. The wound is closed in anatomic layers, and a drain is placed in the field of dissection. Postoperative recovery is usually rapid.

FIG. 56-5. The transcoccygeal approach. (A) The incision is made over the coccyx and curved along the gluteal cleft. (B) Denonvillier's fascia is incised deep to the rectum to expose the prostate and seminal vesicles.

Outcomes Complications Each approach to seminal vesicle surgery is associated with unique complications. The relative complication rates with these approaches are outlined in Table 56-1. Limited extraperitoneal rectal injuries can be handled with formal two-layer closure. Rectal injury with the retrovesical approach is intra-abdominal, however, and may require the placement of omentum over the repair or even temporary colostomy. Most bladder injuries can be closed primarily. The most important point about ureteral injuries is their recognition. Almost all of these injuries can be treated adequately with ureteral stents for 7 to 10 days.

TABLE 56-1. Relative complication rates of the different surgical approaches to the seminal vesicle

Results Resection of the seminal vesicles for cystic and inflammatory diseases is generally successful, although large case series are rare. Seminal vesicle extirpation for

cancer has also been successful, but the rarity of this indication prohibits an accurate assessment of survival and cure rates. Strict categorization of the approximately 100 reported cases of primary seminal vesicle cancer has revealed that fewer than 50% are truly primary to the seminal vesicle. 1,9 In general, reconstructive surgery in children using these approaches has resulted in satisfactory outcomes.

EJACULATORY DUCT SURGERY
The ejaculatory ducts are paired, mainly collagenous, tubular structures that commence at the junction of the vas deferens and seminal vesicle, course through the prostate, and empty into the prostatic urethra at the level of the verumontanum. As illustrated in Fig. 56-6, there are three distinct anatomic regions to the ejaculatory duct:

FIG. 56-6. Illustration of the three anatomic regions of the ejaculatory duct from the seminal vesicle to the prostatic urethra. Inset shows how longitudinal muscle layer becomes attenuated in middle segment. (Used with permission from Nguyen HT, Etzell J, Turek PJ. Normal human ejaculatory duct anatomy: a study of cadaveric and surgical specimens. J Urol 1996;155:1639–1642.)

The proximal and largely extraprostatic portion The middle intraprostatic segment A distal segment that is incorporated into the lateral aspect of the verumontanum in the prostatic urethra

5

The wall of the proximal portion of the duct, as a direct extension of the seminal vesicle itself, consists of three histologic layers: An outer muscular layer A collagenous middle layer An inner mucosal layer 5 The muscular layer becomes markedly attenuated in the middle or prostatic portion of the duct and is wholly absent in the distal segment. Diagnosis Ejaculatory duct obstruction is the cause of infertility in 5% of azoospermia men. Duct obstruction can result from seminal vesicle calculi, the presence of Müllerian duct (utricular) or Wolffian duct (diverticular) cysts, postsurgical or postinflammatory scar tissue, calcification near the verumontanum, or congenital atresia. 7 Classically, this condition presents as hematospermia, painful ejaculation, or infertility with azoospermia. Associated risk factors from the patient history include evidence of prior urinary tract infection or trauma and perineal pain or discomfort. It is important to discontinue the use of medications that may impair ejaculation. An abbreviated list of these medications is found in Table 56-2. Ejaculatory duct obstruction is suggested by finding enlarged, palpable seminal vesicles on rectal examination. The diagnosis is confirmed by a combination of findings:

TABLE 56-2. Medications associated with impaired ejaculation

An ejaculate with a volume less than 2.0 ml and a pH < 7.2 that contains no sperm or fructose A transrectal ultrasound (TRUS) demonstration of dilated seminal vesicles (>1.5 cm width) or dilated ejaculatory ducts (>2.3 mm) in association with a cyst, calcification, or stones along the duct. Recently, high-resolution TRUS has virtually replaced the more invasive vasography in the diagnosis of ejaculatory duct obstruction. With TRUS, it is also possible to determine precisely the level of obstruction in the ejaculatory duct in order to more accurately assess the feasibility of transurethral surgery. To complete the evaluation in patients presenting with infertility, it is important that the serum FSH and testosterone be normal and that a testis biopsy confirms ongoing sperm production. Indications for Surgery Patients with ejaculatory duct obstruction sufficient to cause coital discomfort, recurrent hematospermia, or infertility should be considered as candidates for transurethral treatment. Infertility with duct obstruction can present with low ejaculate volume and azoospermia or low ejaculate volume with decreased sperm density (<20 × 106 sperm/ml) or impaired sperm motility (<30% motility). Anatomic findings suggestive of obstruction and lesions located within approximately 1 to 1.5 cm of the verumontanum are usually amenable to transurethral management. Alternatives to Surgery The discontinuation of the medications listed in Table 56-2 may improve states of ejaculatory dysfunction. Anatomic lesions are most appropriately treated with surgery. Surgical Technique

Transurethral resection of the ejaculatory ducts (TURED) is performed in the outpatient setting. Following the administration of light general or regional anesthesia and a single dose of a broad-spectrum antibiotic, the patient is placed in dorsal lithotomy position with a rectal drape (O'Connor). Formal cystourethroscopy is performed. Careful examination is made of the areas lateral to the verumontanum within the prostatic urethra to visualize either ejaculatory duct orifice. A small resectoscope (24 Fr) and electrocautery loop are inserted and the verumontanum resected in the midline ( Fig. 56-7). The resection is performed with pure cutting current to minimize cauterization of the delicate ejaculatory ducts. Often several passes of the cutting loop are required to visualize the ejaculatory duct openings within the prostatic tissue. This can mean relatively deep dissection in a small prostate gland, a situation that can make even an experienced transurethral surgeon feel uneasy. At the correct level of resection, cloudy, milky fluid can usually be seen refluxing from the opened ducts. After resection, large bleeding blood vessels are lightly cauterized, with care taken to avoid fulguration of the duct openings. Because the area of resection is at the prostatic apex, near to both the external urethral sphincter and the rectum, careful and constant positioning of the resectoscope is essential. A finger placed in the rectum can help avoid rectal injuries and assist in keeping the resectoscope tip proximal to the external sphincter. A small Foley catheter is placed for 24 to 48 hours and removed on an outpatient basis. Oral antibiotics are given while the catheter is in place. After such treatment for infertility, intercourse is resumed after 7 days, and a formal semen analysis is checked first at 2 weeks and then at regular intervals thereafter, until semen quality stabilizes.

FIG. 56-7. Transurethral resection for ejaculatory duct obstruction. Midline resection of the verumontanum is shown. Lateral and deeper resection may be necessary depending on the site and reason for duct obstruction. (Conceptualized by Paul Stempen, MA.)

There are several useful aids to ensure that the resection is performed safely and completely. With an endoscopic needle, the milky ejaculatory duct fluid can be sampled transurethrally during the procedure and inspected with microscopy for sperm. The use of simultaneous, real-time TRUS during the resection has recently become a valuable addition to this procedure. The exact location of the lesion to be resected can be determined by TRUS, and the depth of resection continously assessed during the resection. Similarly, TRUS can be used to guide the instillation of indigo carmine or methylene blue into the seminal vesicles with a long, 20-gauge Chiba needle before the resection. The dye is subsequently visualized through the endoscope on relief of obstruction. Outcomes Complications The expected complication rate from TURED surgery is approximately 20%. 8 Most common among them are self-limited hematospermia, hematuria requiring recatheterization, and urinary tract infection. More concerning, but less frequent, are epididymitis and a “watery” ejaculate. High-volume watery ejaculate is presumed secondary to the reflux of urine retrograde through the ejaculatory ducts and into the seminal vesicles, as suggested by the finding of creatinine in the ejaculates of TURED patients. In addition to the social implications of this complication, the exposure of sperm to urine may significantly impair fertility potential. Several potentially major but rarely reported complications include retrograde ejaculation, rectal perforation, urinary incontinence, and recurrent seminal vesicle infection. Results Long-term relief of postcoital and perineal pain after TURED can be expected in 60% of patients. 3 Hematospermia has also been effectively treated with TURED, but the literature on this indication remains ancedotal. There is convincing evidence from several large series of patients treated for infertility that a 20% to 30% pregnancy rate can be expected from TURED.7,8 In one series, men treated for either low-volume azoospermia and low-volume oligoasthenospermia were equally likely (65% to 70%) to show improvements in semen quality after TURED 8 (Fig. 56-8).

FIG. 56-8. Expected outcome of TURED for the indications of low-volume azoospermia and low-volume oligoasthenospermia (TMC, total motile count rise of >50% after surgery). (Adapted with permission from Turek PJ, Magana JO, Lipshultz LI. Semen parameters before and after transurethral surgery for ejaculatory duct obstruction. J Urol 1996;155:1291–1293.)

Several caveats of TURED surgery should be recognized and emphasized to the patient preoperatively. We have found that 13% of men treated by TURED for low-volume azoospermia will convert to normal-volume azoospermia. Among these patients, we have found evidence of secondary obstruction at the level of the epididymis that required epididymovasostomy. The phenomenon of epididymal obstruction in these patients may reflect the effects of time and blockage on other portions of the delicate male ductal system. Notably, 4% of patients treated for low-volume oligoasthenospermia may become azoospermic after TURED, presumably from scar tissue formation. We now recommend preoperative sperm cryopreservation if TURED is planned for this indication. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Benson RC Jr, Clark WR, Farrow GM. Carcinoma of the seminal vesicle. J Urol 1984;132:483. DeAsis JS. Seminal vesiculectomy. J Urol 1952;68:747. Farley S, Barnes R. Stenosis of ejaculatory ducts treated by endoscopic resection. J Urol 1973;109:664. Kreager JA, Jordan WP. Transcoccygeal approach to the seminal vesicles. Am Surg 1965;31:126. Nguyen HT, Etzell J, Turek PJ. Normal human ejaculatory duct anatomy: a study of cadaveric and surgical specimens. J Urol 1996;155:1639–1642. Politano VA, Lankford RW, Susaeta R. A transvesical approach to total seminal vesiculectomy: a case report. J Urol 1975;113:385. Pryor JP, Hendry WF. Ejaculatory duct obstruction in subfertile males: analysis of 87 patients. Fertil Steril 1991;56:725. Turek PJ, Magana JO, Lipshultz LI. Semen parameters before and after transurethral surgery for ejaculatory duct obstruction. J Urol 1996;155:1291–1293. Williams RD. Surgery of the seminal vesicles. In: Walsh PC, Retik AB, Stamey TA, Vaughan ED, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;2942–2955.

Chapter 57 Vasectomy Glenn’s Urologic Surgery

Chapter 57 Vasectomy
Jon L. Pryor and Douglas A. Schow

J. L. Pryor: Departments of Urologic Surgery, Cell Biology and Neuroanatomy, and Obstetrics and Gynecology, University of Minnesota, Minneapolis, Minnesota 55455, and Center for Men's Health and Infertility, Reproductive Health Associates, St. Paul, Minnesota 55102. D. A. Schow: Department of Urologic Surgery, University of Minnesota, Minneapolis, Minnesota 55455, and Center for Men's Health and Infertility, Reproductive Health Associates, St. Paul, Minnesota 55102.

Indications for Surgery Preoperative Evaluation Alternative Therapy Surgical Technique Preparation Local Anesthesia Delivery and Excision of the Vas Outcomes Complications Results Chapter References

Elective vasectomy is a popular method of permanent sterilization for men. Approximately 500,000 men undergo this procedure in the United States each year. Despite being considered a minor office procedure, vasectomy has been noted to be the most common procedure involved in malpractice claims against urologists. The urologist needs to be thoroughly versed in the management of men seeking surgical sterilization. This chapter reviews the preoperative evaluation, surgical techniques, complications, and failure rates of vasectomy.

INDICATIONS FOR SURGERY
Preoperative Evaluation No strict criteria exist for determining which patient should be offered vasectomy. Patients who are unmarried, have no children, or have no clear indication for the procedure should be thoroughly counseled, as vasectomy may not be appropriate for these men. Involvement of the spouse in the decision making and in witnessing the consent, though not required in most states, is recommended. A brief review of the past medical history, specifically for bleeding diatheses, anesthetic allergies, and prior scrotal surgery (i.e., orchiopexy) is obtained. A brief physical examination, in a warm room, of the genitalia for any anatomic abnormalities is performed. Large varicoceles, hydroceles, or a vas that is difficult to palpate because of a contracted (tight) scrotum may warrant doing the procedure in the operating room. Penile or scrotal infections are to be treated before the vasectomy is done. Penile or scrotal condyloma can be treated before or after the vasectomy. The most important aspect of the preoperative evaluation is to review the procedure and associated complications in clear, easily understood language. It is also good practice to document that the patient understands the possible complications and the necessary follow-up. The essential points to review with the patient should include: That this procedure results in permanent sterility. Although vasectomy reversal is possible, there are no guarantees of success, and most insurance companies do not cover the expenses of vasectomy reversal. That semen can be cryopreserved before vasectomy. The patient must provide semen for analysis until two consecutive samples show azoospermia and that contraception is required until sperm are cleared. That there is a slight risk of failure (i.e., recanalization). Postoperatively, some patients may develop scrotal pain that usually, but not always, resolves. Finally, the surgeon may need to address concerns regarding the myths of vasectomy, specifically about changes in testicular function, sexual function, libido, quality of climax, and characteristics of semen: no studies have shown any significant affect on testicular or sexual function after vasectomy. 8 Once an informed consent is obtained and all questions have been answered, the patient is scheduled for the vasectomy.

ALTERNATIVE THERAPY
Vasectomy is the only form of male birth control that is currently available. All other forms of birth control are directed at the female partner, including oral contraceptives, tubal ligation, and other forms of contraception.

SURGICAL TECHNIQUE
The introduction of the no-scalpel vasectomy (NSV) in the United States by Li in 1986 has played a major role in refining the procedure. 7 In our opinion, major contributions of this variation of vasectomy have been threefold: the use of a single midline approach at the median raphe of the scrotum, the use of the external spermatic sheath injection for anesthetizing the vas, and the three-finger method for fixation of the vas. Details of the NSV technique, which requires much practice to master, can be found in a review of the procedure by Li and associates. 7 We describe our vasectomy technique, which relies on some aspects of the NSV but employs the scalpel to incise the skin. Preparation Shaving of the scrotum, though not mandatory, is preferred. Pubic hair in the wound or along the vasectomy site increases the risk of infection and can hinder visibility. The patient is placed supine in a comfortable position, arms resting at his sides. The operating field is prepped with a warm antiseptic solution. A warm room and warm solution keep the scrotum relaxed, which facilitates isolating the vas deferens. Surgical drapes are placed to isolate the scrotum. Local Anesthesia The method that we feel works best for achieving anesthesia of the vas and scrotal skin is the external spermatic sheath injection described by Li and others. 6 The vas deferens is held using the three-finger technique 7 (Fig. 57-1). The middle finger of the nondominant hand is used to manipulate the right vas deferens to the skin at the median raphe of the scrotum. The vas is then tensed over the middle finger, with the thumb and index finger used to stretch the skin over the vas while positioning the vas as superficial to the skin as possible. A 25- or 27-gauge 1½-inch needle is used for injection. One to two percent plain lidocaine (i.e., without epinephrine) is injected in a transverse fashion across the median raphe between the thumb and index finger to raise a superficial skin wheal 1 to 1.5 cm in diameter. Apply pressure on the skin wheal using a gauze sponge for a few seconds to enhance palpation of the vas deferens. Once the superficial wheal is made, the needle is advanced its full length immediately adjacent and parallel to the vas ( Fig. 57-2) following a path that is in the general direction of the external inguinal ring. Two to five milliliters of lidocaine is injected within the external spermatic sheath as the needle is withdrawn. After injection, the right vas deferens is released, and the left vas is manipulated to the median raphe. The left vas is anesthetized using the same technique as on the right vas. If this is done properly, the referred abdominal discomfort that occurs with manipulation of the vas during vasectomy is usually eliminated using this anesthetic technique.

FIG. 57-1. The three-finger technique for isolating the right (A) and left (B) vas defrens.

FIG. 57-2. The right vas deferens is manipulated to the midline raphe and anesthetized.

Delivery and Excision of the Vas After local anesthesia has been accomplished, the right vas is again fixed under the skin wheal with the nondominant hand. The vas should be held as superficial to the skin as possible. The scalpel is used to make a 1.0-cm transverse incision over the vas deferens ( Fig. 57-3A). The incision is carried down until the vas is identified. Grab the vas or its sheath with a towel clip to ensure that the vas does not slip away. The vas sheath is then incised vertically until the bare vas wall is clearly seen. An Allis clamp or forceps is used to grasp only the vas, with care taken to avoid including the vasal sheath in the clamp or forceps ( Fig. 57-3B). The vas should be easily pulled out of its sheath; if not, the surgeon most likely did not completely incise the sheath. At this point the vas should be covered only on the posterior aspect by the vas sheath. A mosquito clamp is passed along the undersurface of the vas and gently spread to strip the sheath and vasal vessels away from the vas (Fig. 57-4A). This maneuver should result in a 1.5- to 2.0-cm bare segment of vas.

FIG. 57-3. (A) A 1-cm transverse incision is made over the vas deferens. (B) The vasal sheath is incised vertically and the vas defrens is grasped.

FIG. 57-4. (A) The vasal sheath and vessels are stripped away from the vas deferens. (B) Transection of the vas deferens and cauterization of the lumen of the abdominal end.

The testicular end of the isolated vas segment is ligated with a 2-0 Vicryl or silk suture, leaving a long free end of suture to hold the vas with a mosquito clamp. The vas is then transected 1 cm from the ligature with iris scissors or a scalpel, exposing the lumen of the vas ( Fig. 57-4B). A battery-operated hot wire is passed 0.5 cm into the lumen of the abdominal end of the vas. The needle is slowly withdrawn as current is applied to the lumen. The surgeon should avoid a full-thickness burn of the lumen, as this can cause necrosis of the vas and increases the risk of recanalization. The idea is to create enough injury to cause fibrosis of the vasal lumen, which results in permanent occlusion of the vas. The vas is then released, allowing the abdominal end to retract. The vasal sheath is then grasped with forceps and pulled over the abdominal end of the vas. A single interrupted 3-0 chromic suture is used to close the fascial edges of the sheath over the abdominal end of the vas (Fig. 57-5). This maneuver isolates the abdominal vas from the testicular end of the vas, which decreases the risk of recanalization. The testicular end of the vas is then transected next to the suture so as to allow a 1-cm segment of vas to be removed. This segment of vas can be sent to pathology for formal confirmation of vasectomy or saved for future pathologic analysis, if necessary. The free end of suture is cut, and the testicular end of the vas is allowed to retract into the scrotum.

FIG. 57-5. The vasal sheath is closed over the vas deferens stump, and a segment of vas is transected.

The left vas deferens is then manipulated to the same incision using the same three-finger technique, and the procedure is repeated as described for the right vas. Once the left side has been completed, the incision is observed for bleeding. Hemostasis is meticulously obtained with the electrocautery, as any bleeding can cause a large scrotal hematoma. A single interrupted 3-0 chromic suture can be used to approximate the skin edges, but this is often not required, as the incision usually contracts and approximates itself. Antibiotic ointment is applied to the wound. Gauze is applied and held in place with a scrotal support. A jockstrap is recommended, but brief underwear can also be used.

OUTCOMES
Complications Scrotal hematoma is the most worrisome complication of vasectomy and occurs in up to 3% of patients. Not surprisingly, one study found that the rate of scrotal hematoma was higher for physicians who performed fewer than 50 vasectomies a year 5: the hematoma rate was 4.6% for physicians with fewer than ten vasectomies per year versus 1.6% for those who performed more than 50 per year. Most scrotal hematomas are small and do not require drainage. Warm compresses, nonsteroidal anti-inflammatory medications, and restricted activity result in resolution of the majority of these hematomas. Occasionally the hematoma expands underneath the pliable skin of the scrotum and becomes quite large. These hematomas require surgical drainage, which results in relief of discomfort and a quicker recovery. Incisional infections occur in 1% to 3% of patients. 5,12 Most infections are superficial and resolve with a course of oral antibiotics. Rarely does an abscess form, which would then require surgical incision and debridement. Chronic scrotal pain after vasectomy is not uncommon and has been reported to occur in up to 33% of patients. 9 Of the patients with chronic scrotal pain, one-half stated that the pain was troublesome; however, only 9/172 (5%) sought medical attention for the symptoms. The general belief is that the scrotal pain after vasectomy is secondary to congestion of the epididymal tubule and is sometimes associated with epididymal blowout causing an epididymal granuloma. The symptoms, which can be exacerbated by sexual activity and ejaculation, are generally self-limiting. Nonsteroidal anti-inflammatory medications are recommended to treat the inflammatory process, which usually leads to resolution of the symptoms. Patients with persistent pain despite conservative therapy can be treated with epididymectomy. Vasovasostomy has also been recommended but has the drawback of fertility. Sperm granuloma at the vasectomy site occurs in 1% to 10% of patients. 2,12 These granulomas form as a result of blowout of the testicular end of the vasal occlusion. Most granulomas are asymptomatic. The majority of patients only need reassurance that the lump in the scrotum is not cancer or something bad. Patients who experience discomfort from the granuloma should be treated initially with nonsteroidal anti-inflammatory medications. If the symptoms do not resolve, excision of the granuloma is recommended.12 Recanalization/Failure There have been numerous methods described for closure of the vasal ends matched only by the number of techniques described for isolating the vas. Reviews of the various techniques have been done both by Denniston and by Esho and Cass. 2A,2B Multiple authors have reported no recanalization with the use of luminal fulguration of both vasal ends and interposition of the fascial sheath over the abdominal end, as described previously. The luminal burn causes a fibrotic reaction that seals the vasal lumen, and the interposed fascial sheath enhances the closure. This method is easy to master, leaves minimal foreign material in the scrotum, and does not jeopardize the chances of future vasectomy reversal. The authors recommend this technique as the method for closing the vasal ends; however, we do feel that a single ligature on the testicular stump prevents leakage of sperm long enough to allow fibrosis of the abdominal end to occur. Recanalization rates of less than 1% can be expected with this technique. Vasectomy and Prostate Cancer In February 1993, Giovannucci and others published two reports showing an increased relative risk of prostate cancer in vasectomized men. 3,4 As a result of the controversy created by these reports, the NIH formed a panel of experts to assess these and other reports showing an association between vasectomy and prostate cancer. The consensus of the NIH panel was that there was insufficient evidence of a real association between vasectomy and prostate cancer. 10 The issue of the possible association of vasectomy and prostate cancer should be reviewed with the patient at the time of consent, but the surgeon should emphasize that at this time there is no conclusive evidence showing a causal relationship between vasectomy and prostate cancer. Results The patient should be instructed to lie supine as much as possible during the first 12 to 24 hours. Most vasectomies are done in the afternoon so that by the next morning the patient can ambulate. Intermittent application of an ice pack to the scrotum during the first 8 hours is highly recommended. The cooler temperature causes contraction of the scrotum and compression of any bleeding sites, decreasing the risk of hematoma formation. Patients may return to work the next day, but strenuous activity should be avoided for at least 1 week. Patients should abstain from sexual activity for 1 week to allow fibrosis of the vas to occur. Patients need to be instructed on the importance of using contraception until they have two azoospermic semen analyses. Studies have shown that 90% of patients will be azoospermic within 82 days, and 80% to 90% of patients will be azoospermic after 12 to 15 ejaculations. 11 The patient should return in approximately 2 months for the first semen analysis. Most recanalizations occur within 12 weeks, 11 so a second semen analysis is obtained 3 months postvasectomy to assure sterility. If the patient has two azoospermic semen analyses, he can discontinue contraception. Ten percent of patients will have nonmotile sperm on semen analysis. Proper handling of these patients has been reviewed extensively in the literature. 1 The risk of pregnancy could not be determined, but the few reports in the literature suggested that the risk is less than 1%. 1 They conclude that patients with nonmotile sperm can be given special clearance to have unprotected intercourse provided that the patient is aware of the extremely low risk of pregnancy. Patients with motile sperm in the semen analysis at 3 months should be advised to undergo repeat vasectomy. Vasectomy is a safe and relatively easy procedure to perform and should continue to be offered to men who desire a permanent form of contraception. The no-scalpel technique has introduced modifications that have improved the efficiency of the procedure. Surgeons performing this procedure should attempt to do at least 50 vasectomies per year, as proficiency in this technique has been shown to decrease complication rates. The evidence to date indicates that vasectomy is not associated with any long-term adverse sequelae. Further study is required to assess the validity of previous reports on the relationship of vasectomy to prostate cancer.

CHAPTER REFERENCES
1. Benger JR, Swami SK, Gingell JC. Persistent spermatazoa after vasectomy: a survey of British urologists. Br J Urol 1995;76:376. 2. Chen TF, Ball RY. Epididymectomy for post-vasectomy pain: histological review. Br J Urol 1991;68:407. 2A. Deniston GG. Vasectomy by electrocautery: Outcomes in a series of 2,500 patients. J Family Practice 1985;21:35. 2B. Esho JO, Cass AS. Recanalization rate following methods of vasectomy using interposition of fascial sheath of vas defrens. J Urol 1978;120:178. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Giovannucci E, Ascherio A, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. A prospective cohort study of vasectomy and prostate cancer in US men. JAMA 1993;269(7):873. Giovannucci E, Tosteson TD, Speizer FE, Ascheria A, Vessey MP, Colditz GA. A retrospective cohort study of vasectomy and prostate cancer in US men. JAMA 1993;269(7):878. Kendrick JS, Gonzales B, Huber DH, Grubb GS, Rubin GL. Complications of vasectomy in the United States. J Family Pract 1987;25(3):245. Li PS, Li S, Schlegel PN, Goldstein M. External spermatic sheath injection for vasal nerve block. Urology 1992;39(2):173. Li S, Goldstein M, Zhu J, Huber D. The no-scalpel vasectomy. J Urol 1991;145:341. McCormack M, LaPointe S. Physiologic consequences and complications of vasectomy. Can Med Assoc J 1988;138:223. McMahon AJ, Buckley J, Taylor A, Lloyd SN, Deane RF, Kirk D. Chronic testicular pain following vasectomy. Br J Urol 1992;69:188. Pollack AE. Vasectomy and prostate cancer. Adv Contracept 1993;9:181. Schlegel PN, Goldstein M. Vasectomy. In: Goldstein M, ed. Surgery of male infertility. Philadelphia: WB Saunders, 1995;34–45. Schmidt SS. Vasectomy by section, luminal fulgaration and fascial interposition: results of 6248 cases. Br J Urol 1995;76:373.

Chapter 58 Vasoepididymostomy Glenn’s Urologic Surgery

Chapter 58 Vasoepididymostomy
Anthony J. Thomas, Jr.

A. J. Thomas, Jr.: Section of Male Infertility, Department of Urology, The Cleveland Clinic Foundation, Cleveland, Ohio 44195.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Position of Patient and Anesthetic Considerations Preparation for Anastomosis End-to-Side Anastomosis End-to-End Anastomosis Postoperative Care Outcomes Complications Results Chapter References

Obstructions of the epididymal tubule are the result of prior infection, congenital dysjunction, vasectomy with subsequent “high-pressure” extravasation, iatrogenic injury, trauma, or other, unknown causes. Successful surgery to bypass the obstruction depends on both the skill of the surgeon and the level of obstruction. The use of the operating microscope has allowed urologists a greater measure of success that was simply not possible when the surgery was performed with the unaided eye or only optical-loupe-magnified vision. Vasoepididymostomy to the caput is significantly less successful in terms of patency and naturally occurring pregnancy than those procedures done at the level of the corpus or cauda epididymis. Although the incidence of obstructive azoospermia is less than 10% of those patients presenting with infertility, it is important to recognize because it is potentially correctable.

DIAGNOSIS
A patient with complete bilateral epididymal obstruction should have a normal semen volume, palpably normal vasa deferentia, epididymides that are full, somewhat tender, or indurated, and testicles larger than 20 cm 3. Serum follicle-stimulating hormone, luteinizing hormone, and testosterone levels will be normal. Absolute diagnosis can only be made at surgery, when active spermatogenesis and patency of the vasa are confirmed and sperm are found somewhere within the epididymal tubule.

INDICATIONS FOR SURGERY
If the patient has the findings just listed, if his health status is such that he would be a suitable surgical candidate, and if he wishes to attempt to correct the obstruction, surgery may be carried out. Vasoepididymostomy may be needed for some patients requesting vasectomy reversal. This is because of apparent high-pressure blowouts in the epididymis, which can occur after vasectomy. Although it may occur at any time interval from vasectomy to reversal, it is rarely found at intervals of less than 5 years, uncommon but not unexpected at 5 to 8 years. After 8 or 9 years, I have found it necessary to perform vasoepididymostomy on either one or both sides in about a quarter of the patients undergoing reversal surgery. The indications are not always as well defined as one would like, and there is a certain amount of decision making based on operative findings and experience. The quality of vasal fluid, the obstructive interval, and the appearance of the epididymis must be taken into account when deciding on the appropriate procedure. Vasoepididymostomy should be considered if there has been a long obstructive interval and no fluid or sperm are identified from the proximal (testicular end) vas deferens, even after gentle irrigation of the lumen with Ringer's lactate. If there is thick, paste-like azoospermic fluid oozing from the proximal vas lumen, I milk out as much fluid as possible by gently compressing the tail of the epididymis and convoluted vas. If it remains the same, I explore the epididymis and do a vasoepididymostomy at a level where normal-appearing motile or nonmotile sperm are found. Other indications are less obvious. Long obstructive intervals of more than 10 years and thick, though not pasty, fluid containing only sperm heads prompts me to expose the epididymis and decide whether to do a vasoepididymostomy or vasovasostomy based on its appearance, degree of dilation, and induration.

ALTERNATIVE THERAPY
The only other options that would allow the patient to have his own biological children would be percutaneous or open epididymal or testicular sperm aspiration combined with the assisted reproductive technique of in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI).

SURGICAL TECHNIQUE
Position of Patient and Anesthetic Considerations Because these procedures can sometimes take between 2 and 4 hours, particularly if they involve bilateral correction of obstruction, the patient is supine on the operating table and appropriately padded at all pressure points, particularly his head, buttocks, and heels, using egg-crate foam. The patient's arms are either positioned at his side or angled no more than 45 degrees from his body to minimize any chance of pressure injury to the brachial plexus. Although this operation can be done under local anesthesia and intravenous sedation, it is not recommended except for those surgeons with extensive experience in both anesthetic blocks and microsurgical reconstruction. For patient and surgeon comfort as well as for safety, either general or epidural anesthesia would be advisable. An intravenous cephalosporin or another antibiotic with similar coverage is administered 30 minutes before surgery begins. Preparation for Anastomosis The presence of active spermatogenesis is an obvious prerequisite to this surgical procedure. If a testis biopsy has not already been carried out before this surgery, it can be done in a standard fashion, and the tissue examined by frozen section or touch preparation. 6 Alternatively, a small piece of tissue can be minced on a glass slide, covered with a drop of saline, crushed with a coverslip, and examined under the microscope (400×) for the presence of sperm, some of which may be motile. It is necessary to prove that the vasa are patent. This should be done at the time of the definitive, corrective surgery, not as a separate procedure and not in conjunction with a testis biopsy, as it may lead to excess scarring and make later vasoepididymostomy more difficult to do. This author has previously described a simple method of proving the vasa patent. 6 A straight portion of the scrotal vas is isolated with great care so as not to injure the closely adherent blood supply. A 4-inch piece of plastic tubing cut from a butterfly needle is passed beneath the vas as a nontraumatic holding device. A 30-gauge lymphangiogram needle (#6657, Becton-Dickinson Co., Franklin, NJ) is passed in a spot just distal to the tubing almost at a parallel angle through the muscular wall and into the lumen of the vas deferens, pointing toward the abdominal end, not the testicle ( Fig. 58-1). Confirmation of proper position is made by the easy flow of normal saline through the needle using a 3-ml syringe. Visual proof of patency can then be obtained by injecting dilute contrast (one-to-one mixture of saline and Renografin-60®rm) and taking a radiograph of the pelvis and scrotum. Alternatively, if one is able to inject 10 to 12 ml of saline without meeting resistance, patency can be assumed. At this point, exploration of the epididymis and correction of the obstruction can take place using either the end-to-side or end-to-end technique.

FIG. 58-1. A 30-gauge needle is inserted into the lumen of the vas deferens, pointing away from the testicle.

End-to-Side Anastomosis This is the most common method used to perform a vasoepididymal anastomosis. It does not involve resection of the epididymis but merely identification of a patent portion of the epididymal tubule to which the vas deferens can be anastomosed. 1,7,9 The procedure begins by first freeing up the portion of the vas deferens that will be used for the anastomosis. If the patient has had a vasectomy, and it is determined that a vasoepididymostomy is needed, the abdominal end of the vas needs to be freed up along with its attendant blood supply. If there is insufficient length of vas to reach the epididymis without tension, the scrotal incision may need to be extended toward the groin or a separate incision made at the level of the external inguinal ring. The vas is then able to be freed from its surrounding loose tissue and other cord structures to gain length. If the vas has not been previously cut, it is preferable to use the convoluted portion of the vas, tapering the angle of the convolutions to fit flush against the opened epididymal tubule. Once the vas has been dissected and cut, the epididymis is examined under the operating microscope. Sometimes the point of obstruction is obvious by a marked narrowing of the epididymal diameter or by a blue-brown discoloration caused by a breakdown of extravasated sperm into the peritubular tissue. It is most prudent to begin to identify the unobstructed portion of the tubule starting at the level of the cauda. The epididymis is held between thumb and forefinger and, using a small round-tipped Beaver blade knife, an incision is made just into the epididymal tunic ( Fig. 58-2). With a round-tipped microscissors, a small window in the tunic is created, and a loop of epididymal tubule below is isolated from its surrounding loose connective tissue. A 2-mm-tip microknife is used to incise the tubule along its long axis (Fig. 58-3). The exuding fluid is immediately placed on a glass slide and examined by light microscopy for the presence of long-tailed motile or nonmotile sperm. A single 10-0 nylon suture (TE-70, Davis and Geck) is passed outside in through the opened tubule at the midpoint of the incision. If no sperm are identified in the fluid, the same procedure is carried out 0.5 to 1 cm more proximal, and the procedure is repeated until sperm are found. After the first identification suture is placed, two more are passed outside in through the opened tubule 120 degrees apart from one another ( Fig. 58-4).

FIG. 58-2. Incision into epididymal tunic.

FIG. 58-3. Longitudinal incision into epididymal tubule.

FIG. 58-4. Placement of 10-0 nylon identification sutures 120 degrees apart.

The vas deferens is brought into close approximation to the epididymis by creating a small opening in the parietal portion of the tunica vaginalis 1 cm from the opened patent loop of epididymal tubule. A fine curved hemostat is passed through the opening beneath the vascular cord structures, grasping the suture attached to the end of the vas and pulling the vas through the opening to lie next to the epididymis, ready for anastomosis ( Fig. 58-5).

FIG. 58-5. Approximation of vas deferens to epididymal tubule for anastomosis.

When it has been determined that the vas lies straight and is not twisted, one or two 9-0 nylon sutures (Ethicon, VAS-100) are passed through the cut edge of the window of the epididymal tunic, outside in and inside out through the edge of the vas muscularis and adventitia, securing the vas to the point of anastomosis ( Fig. 58-6). The first previously placed identification suture (so-named 0 degrees position) is passed inside out through the edge of the vas lumen and tied ( Fig. 58-7). Another 10-0 suture is then passed halfway (60 degrees) between this first tied suture and the one passed at 120 degrees from the first. It is passed outside in through the edge of the epididymal lumen and into the vas lumen but not tied at this point. A similar suture is passed on the opposite side of the first suture. The two 120-degree sutures are then passed through the edge of the vas lumen. The 60-degree sutures are tied in turn, and before the 120-degree sutures are tied, a determination is made as to whether there is room for a sixth suture directly opposite the first at 180 degrees. If so, it is passed before the other two are tied, and each is then tied in turn (Fig. 58-8). A 9-0 nylon suture is then passed at the 180-degree position outside in through the vas adventitia and muscularis into the edge of the window of the epididymal tunic (Fig. 58-9) and tied. With the vas pulled gently to one side, four or five equally spaced 9-0 nylon sutures are placed to secure first one and then the other side of the anastomosis, with sutures placed into the epididymal tunic and the opposing vas adventitia and muscularis ( Fig. 58-10). Further “bolstering sutures” can be placed along the long axis of the vas, securing it to the visceral tunic ( Fig. 58-10). These last sutures prevent any pulling on the anastomosis as the testis is manipulated and placed back into the tunica vaginalis and scrotum.

FIG. 58-6. Vas deferens is secured to epididymal tunic with 9-0 nylon.

FIG. 58-7. Placement of identification suture in the vas deferens.

FIG. 58-8. The sutures are tied.

FIG. 58-9. Second stabilization suture of 9-0 nylon from vas adventitia to epididymal tunic.

FIG. 58-10. Circumferential sutures to reinforce the anastomosis from vas to epididymal tunic. Longitudinal sutures to hold vas along the epididymal tunic.

End-to-End Anastomosis Although I have tended to use the end-to-side method of vasoepididymostomy almost exclusively, the end-to-end technique, which was originally described by Silber,4,5 is a useful procedure, particularly in the instance of a short abdominal vas when the tail and body of the epididymis can be dissected away from the testis to gain length and create a tension-free anastomosis. 2 The end-to-end anastomosis is more easily performed at the level of the cauda, where there are fewer convolutions and the wall of the tubule is more muscular. Isolation of the vas deferens and its blood supply is similar to that described above. The epididymis is examined under the operating microscope, and the distal cauda cut across in a transverse manner, exposing multiple cut ends of the epididymal tubule. Bleeding from the tunic and other small vessels is controlled with gentle finger compression and judicious use of a micro-tipped bipolar cautery. The end of the epididymis is irrigated with Ringer's lactate, and if the patent tubule is at that level, fluid containing normal-appearing motile or nonmotile sperm will be seen when it is placed on a glass slide and examined by light microscopy. If no sperm are found, the epididymis is transected 0.5 to 1 cm proximal, and the same procedure is carried out until sperm are identified. As soon as the patent end of the tubule is isolated, a single 10-0 nylon suture is passed outside in at the 6-o'clock position to act as the identification suture. The open end of the vas adventitia and muscularis is then secured to the inferior portion of the cut end of the epididymis with one or two 9-0 nylon sutures ( Fig. 58-11). Three more 10-0 nylon sutures are passed at 90 degrees from the first, and then the most inferior (0 degrees) suture is placed through the edge of the vas lumen ( Fig. 58-12). The 90-degree suture is then passed inside out through the vas, and the first suture is tied. The 270-degree suture is placed in the vas next, and finally the 180-degree suture is placed. They are tied in the same sequence. The edge of the tunic is then secured to the vas adventitia and muscularis with 9-0 nylon, first at 180 degrees from the first to prevent pulling on the anastomosis and then in a circumferential fashion ( Fig. 58-13).

FIG. 58-11. The first lumen-to-lumen 10-0 nylon suture is placed at the 6-o'clock position.

FIG. 58-12. Three more sutures are placed in the lumen of the epididymal tubule and vas deferens equidistant from one another.

FIG. 58-13. The adventitia and muscularis of the vas deferens are approximated to the tunic of the epididymis with equally spaced 9-0 nylon sutures.

Postoperative Care Regardless of which technique of anastomosis is used, postoperative instructions to the patient are the same. An ice bag is applied to the scrotal area from the time he is in the recovery or outpatient suite until he goes to sleep that night. Analgesia is limited to a combination drug of either acetaminophen with codeine or

oxycodone. This is generally required for only a day or two. Patients are restricted from strenuous physical exercise and heavy lifting for 3 weeks. Sexual activity is prohibited for the same time. A semen analysis is performed at 1 month and every 3 months thereafter for the first year. It is not uncommon for the patient to be azoospermic or severely oligospermic for the first two analyses. It may, in some instances, take as long as a year to find sperm in the ejaculate. 3

OUTCOMES
Complications Serious complications during or following vasoepididymostomy are rare. The men undergoing these procedures are, in general, healthy, and the surgery is elective, allowing time for adequate preoperative assessment of any factors that might increase morbidity. A history of prior anesthetic problems, either personal or familial, possible allergy to latex, untreated hypertension, and clotting dysfunction are a few examples of potential serious risk factors that should be evaluated prior to surgery. The most significant genital complication from this procedure is testicular atrophy resulting from inadvertent injury to the arterial blood supply of the testis. The chance of this occurring can be minimized by scrupulously avoiding the main cord structures when dissecting the vas and not separating the upper portion of the epididymis from the testis where injury to the testicular artery may occur. Scrotal hematoma following surgery can, to say the least, be very disturbing for both patient and surgeon. Meticulous hemostasis with bipolar and ophthalmologic cautery can prevent most problems. Avoid placing large suture ligatures around clusters of vessels where there is bleeding, as it may compromise the arterial blood supply. Any bleeding that does occur during the procedure must be precisely identified and controlled. I have not found it necessary to place drains in the scrotum after vasoepididymostomy. Infections, either within the scrotum or postoperative epididymitis, are rare, perhaps in part because of the antibiotic prophylaxis given before and for a few days after the procedure. When it does occur, aggressive management with antibiotic therapy and drainage if necessary will minimize morbidity. Results Success with this operation may be measured by patency (sperm in the ejaculate) or by the ultimate pregnancy rates, this latter being the most critical with regard to functional success. Time of follow-up is important because, as previously stated, it may take up to 12 months to have sperm in the ejaculate and perhaps another 12 months or more to establish a pregnancy for some couples. Therefore, any analysis of results must define the time frame in which the patients have been followed in order to place the results in proper perspective. In a review of my first 161 consecutive end-to-side procedures performed on 153 men, the overall patency rate was 79% (110 of 139 patients followed more than 6 months), and the pregnancy rate was 42% (47 of 111 patients followed for at least 1 year or reporting a pregnancy before 1 year). The best results were from anastomoses to the corpus or cauda epididymides, which had an 85% (76/85) patency and a 51% (36/71) pregnancy rate. Caput anastomosis had the poorest results, with a patency rate of only 44% (14/32) and a pregnancy rate of 13% (3/23). If the procedure was able to be bilateral, 79% had sperm postoperatively, and 51% established pregnancies. In those having a unilateral vasoepididymostomy, 71% demonstrated patency but only 27% caused a pregnancy. 8 There are a number of reasons for vasoepididymostomy to fail. The lumens are small, and any inaccuracy in approximating them will cause either sperm leak or scarring, occluding the anastomosis. It is sometimes difficult to see the opened epididymal tubule. A few drops of indigo carmine dripped over the area will stain the edge of the cut lumen, allowing for clearer identification. Inadequate vasal length causing tension at the anastomosis is another cause for failure. Taking the time to free up sufficient length of vas with its blood supply will make the operation easier and assure a tension-free anastomosis. Vasoepididymostomy is a procedure that takes practice and patience to learn. If it is not being done with some degree of regularity, the results may more than likely be disappointing to both patient and surgeon. The manual dexterity needed to perform this procedure well is best obtained in the microsurgical laboratory, and the skill kept up by performing microsurgical vasovasostomy in the operating room. The principles put forth are the same as those for any surgical anastomosis: absence of tension, adequate blood supply, and accurate alignment of the ends being brought together. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. Dewire DM, Thomas AJ. Microsurgical end-to-side vasoepididymostomy. In: Goldstein M, ed. Surgery of male infertility. Philadelphia: WB Saunders, 1995;128–134. Goldstein M. Microsurgical vasoepididymostomy: end-to-end anastomosis. In: Goldstein M, ed. Surgery of male infertility. Philadelphia: WB Saunders, 1995;120–127. Schlegel PN, Goldstein M. Microsurgical epididymovasostomy: refinements and results. J Urol 1993;150:1165–1168. Silber SJ. Microscopic vasoepididymostomy: specific microanastomosis to the epididymal tubule. Fertil Steril 1978;30:565–571. Silber SJ. Microsurgery for vasectomy reversal and vasoepididymostomy. Urology 1984;23(5):505–524. Thomas AJ, Nagler HM. Testicular biopsy and vasography for evaluation of male infertility. Urol Clin North Am 1987;14:167–176. Thomas AJ. Vasoepididymostomy. Urol Clin North Am 1987;14:527–538. Thomas AJ. Microsurgical end-to-side vasoepididymostomy: analysis and outcome of 161 procedures. Paper presented at the 88th annual meeting of the American Urological Association, San Antonio, TX, May, 1993. 9. Thomas AJ. Vasoepididymostomy. In: Thomas AJ, Nagler HM, eds. Atlas of surgical management of male infertility. New York: Igaku-Shoin Medical Publishing, 1995;62–70.

Chapter 59 Vasovasostomy Glenn’s Urologic Surgery

Chapter 59 Vasovasostomy
William Forbes Hendry

W. F. Hendry: St. Bartholemew's Hospital and Royal Marsden Hospital, London W1N 2DE, United Kingdom.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Vasectomy reversal may be requested for a variety of reasons, including remarriage, death of children, change of heart, or altered financial circumstances. Howard 7A asked the question “Who asks for vasectomy reversal and why?” and found that over half were divorced or separated from their wives and felt disadvantaged in courtship or remarriage by being infertile: among those who had remarried, many wives were said to be desperate for a pregnancy. Most requests for reversal came from men who had been under 35 at the time of vasectomy, and many had decided to have the vasectomy done at a time of emotional crisis. There is no fundamental reason why fertility should not be restored by reconnection of the testes after vasectomy because sperm production is well maintained. Normal spermatogenesis has been observed in testicular biopsies from men taken up to 10 years following vasectomy. 1A In fact, technical difficulties with the vas anastomosis, secondary changes in the epididymis, and in some cases the immunologic reaction provoked by the spermatozoa make the chances of successful restoration of fertility by vasectomy reversal only slightly better than 50%. Vasal obstruction may also occur following infection, such as gonorrhea, when it may coexist with a caudal epididymal obstruction, either on the same or on the other side. The most common sites affected are at the neck of the scrotum and at the internal inguinal ring, where the vas changes direction sharply. Totally impenetrable obstructions are occasionally encountered, which generally turn out to be tuberculous. The vas may also be obstructed following groin surgery such as hernia repair in infancy or childhood. 9,11

DIAGNOSIS
The level of the obstruction is usually obvious following vasectomy but may need to be defined by vasography in patients with postinflammatory or posttraumatic obstructions, in whom patency of the vas beyond the obstruction should also be confirmed. Antisperm antibodies develop in the serum of 60% to 80% of men following vasectomy, but their effects on fertility following vasectomy reversal have been poorly understood. In a prospective study, these antibodies were found in the sera of 79% of 130 men presenting for vasectomy reversal. However, only 9.5% had antibody in the seminal plasma before, rising to 29.5% after the reversal procedure. Eventually, 44.6% of these men impregnated their wives, among whom there was no statistically significant reduction in fertility until very high serum titers (512 or more) were reached, and even at that level, 25% were apparently fertile. 11B There is, therefore, no reason to refuse vasectomy reversal to a man because he has such antibodies, although it is useful to have this information which may allow more accurate prediction of the eventual outcome of the procedure.

INDICATIONS FOR SURGERY
Any organization that includes vasectomy in its program of family planning should include advice and referral for vasectomy reversal, in exactly the same way that the pill may be stopped or a coil can be removed. Fertility can be successfully restored by vasovasostomies in about half of the men who wish to have their vasectomies reversed, often because of a change in their circumstances beyond their direct control. During surgical correction of obstructive azoospermia, vasography can be expected to show that the vas is obstructed in about 10% of men; 6 in such cases, the vasovasostomy may be done straightaway whenever it is technically possible. Unilateral testicular obstruction is not uncommon and may be discovered during investigation of oligospermia or in association with antisperm antibodies in a subfertile man. Among 125 subfertile men with unilateral obstructions defined by surgical exploration, epididymal obstruction following previous infection was the most common cause: however, 50 patients were found with obstructions caused by vasal or epididymal damage following groin or genital surgery in childhood. 7 Vasal reconstruction may be required in such cases, although the choice of optimum treatment will depend on the nature and extent of the obstruction, the immune response to the obstruction, the quality of the opposite testis, and the sperm output in the ejaculate.

ALTERNATIVE THERAPY
Modern techniques of in vitro fertilization (IVF), particularly when combined with intracytoplasmic sperm injection (ICSI), now allow production of pregnancy using spermatozoa obtained by microscopic epididymal sperm aspiration (MESA) 12 or even by percutaneous aspiration (PESA) from patients with testicular obstruction. 4 Although this possibility should be made known to couples where there is good evidence that the man may have obstructive azoospermia, the first approach should still be scrotal exploration not only to define the nature and extent of the obstruction but also because surgical correction offers the most cost-effective and permanent solution to the problems presented by testicular obstruction.

SURGICAL TECHNIQUE
Surgery for male infertility should be done as a day case in good operating facilities, usually with general anesthesia. The basic conditions that may be adequate for vasectomy are generally not good enough for reconstructive surgery on the male genitalia. Special facilities are needed, including fine instruments for microsurgical anastomoses, 2 an operating microscope or good quality magnifying spectacles, and image intensifier equipment to allow x-rays to be taken on the operating table. A microscope will be required to examine fluid obtained from the epididymis or vas for the presence of spermatozoa. These should be prepared in advance so that they are readily available at the time of surgery. After induction of adequate anesthesia and careful skin preparation and draping, the scrotum is opened, and the testicles are delivered. For vasectomy reversal, oblique scrotal incisions are preferred so that they can be extended up to the inguinal region to obtain more length of vas if necessary. The superior end of the vas is identified first, cleaned of surrounding adhesions, and mobilized, taking care to preserve its blood supply, until it is free enough to allow it to meet the inferior end without tension. The inferior end is next identified, cleaned, and held in a suitable tissue forceps or clamp. Once the mobilizations are completed, an 0.5-cm linear incision is made in the inferior end of the vas just below the point of transection, and the tail of the epididymis is squeezed to express milky fluid from the vas. Ideally, this fluid should be examined for the presence of motile spermatozoa, although these are not always found. The superior end is similarly incised, and the lumen defined with a fine nylon probe. The two ends are overlapped and joined together by a side-to-side anastomosis using 6-0 Proline and no splint ( Fig. 59-1). Care should be taken to ensure that the anastomosis is as leakproof as any vascular anastomosis, to prevent spermatozoa from extravasating into the tissues and causing a sperm granuloma. Fibrin-based glue may be used to reinforce the anastomosis. 1B Alternatively, an end-to-end anastomosis may be done. However, the fine bore of the lumen and the disparity that often exists between the diameter of the upper and lower ends requires a microsurgical technique if an accurate union is to be obtained. A typical two-layer technique is shown in Fig. 59-2, and this method is favored by many microsurgeons; however, other experienced surgeons feel that a formal two-layer anastomosis is unnecessary and favor a careful one-layer method, using full-thickness luminal 9-0 alternating with seromuscular 8-0 sutures all around the circumference of the

reconstructed vas deferens. 15

FIG. 59-1. (A) Oblique scrotal incision, testis delivered. (B) Upper end of vas mobilized, transected, and vasal artery ligated. (C) Upper end spatulated by 0.5-cm linear back-cut. (D) Linear incision (0.5 cm) in lower end of vas; white efflux squeezed out by digital pressure. (E) Double-ended 6.0 Proline suture placed from inside lumen out, at top end of upper and lower parts of vas. (F) Suture tied at top end, and continuous stitch about to commence on each side. (G) Anastomosis completed.

FIG. 59-2. (A) Testis exposed. (B) Posterior layer of serosal stitches (8.0 nylon). (C) Mucosal stitches (10.0 nylon). (D) Anastomosis completed.

If there is no efflux from the testicular end of the vas, even with gentle squeezing of the tail of the epididymis, then attention should be turned to the epididymis. A section with distended tubules should be chosen and carefully incised. When it has been confirmed that sperm are present, vasoepididymostomy should be done (see Chapter 58), as there is no point in anastomosing a dry vas to a dry vas. If a vasal obstruction is found in the patient with obstructive azoospermia, it is generally best to make an additional incision in the groin to provide wide exposure of the vas so that the site of obstruction can be clearly defined with a view to reconstruction by vasovasostomy. However, if there is a coexisting caudal epididymal obstruction on the opposite testicle, or if the contralateral testis is atrophic, an alternative approach is transvasovasostomy ( Fig. 59-3).

FIG. 59-3. Trans-vasovasostomy.

Once the surgery is completed, the tunica vaginalis is closed (if it was opened), and the testes are returned to the scrotum on each side. After application of a little antibiotic spray and infiltration of 1 or 2 ml of local anesthetic such as Marcaine 0.5%, the dartos layer is closed with continuous catgut, and the skin is approximated with a subcuticular suture such as Vicryl or Maxon. The scrotum is wrapped in cotton wool and placed in a scrotal support, which should remain dry and undisturbed for 7 days. A Voltarol (diclofenac sodium) suppository will minimize postoperative pain, and a broad-spectrum antibiotic combination such as amoxycillin with flucloxacillin should be given for a few days postoperatively to minimize the chance of infection.

OUTCOMES
Complications Provided that care is taken to be gentle with the tissues, and perfect control of hemostasis is maintained throughout the procedure, there should be no complications. A hematoma or, worse still, wound infection is a complication that is likely to lead to failure of this operative procedure. Results The results of vasectomy reversal reported from a number of centers are shown in Table 59-1. Time elapsed between the vasectomy and the reversal procedure is the most important single factor in predicting success, possibly because of degenerative changes in the epididymis. 5 Among 1,469 men reported from five institutions in the United States over a 9-year period, there was a direct relationship between time since vasectomy and success rate, the pregnancy rates being 76% after less than 3 years, 53% after 3 to 8 years, 44% after 9 to 14 years, and 30% after more than 15 years. 3

TABLE 59-1. Results of vasectomy reversal

Silber pioneered a meticulous two-layer microscopic nonstented technique of end-to-end anastomosis and reported a pregnancy rate of 71% in the first 42 patients followed for over 1 year. 14 It has been calculated that it is unlikely that a success rate greater than this can be achieved. 13 Two large multicenter reviews have compared the results of one-layer and two-layer microsurgical anastomosis and found little difference (see Table 59-1). Because operating time is precious, it seems sensible to use a technique that is quick, robust, and reliable. 8 The author continues to favor overlapped side-to-side anastomosis with a meticulously careful microsurgical technique that provided patency for 93% of 104 men undergoing vasectomy reversal, with a pregnancy rate of 45%. 11B This method also provided good results with postinflammatory vasal obstructions that were encountered when operating for obstructive azoospermia. 6 With a good surgical technique, vasectomy reversal should successfully restore a man's fertility in over 50% of cases. If the operation is unsuccessful—i.e., if spermatozoa appear in the ejaculate only transiently or not at all—the scrotum should be reexplored. There are four distinct causes of failure of vasectomy reversal,11C the most common by far (approximately 50%) being stenosis or obstruction at the site of the previous vasovasostomy. Excellent results can be obtained by repeating the vasovasostomy. 13A The second most common cause (33%) of failure is rupture of epididymal tubules by back pressure, leading to local sperm granuloma formation within the epididymis and obstruction upstream of the original site of the vasectomy; this can be corrected by epididymo-vasostomy. 13B The patient with a very high antisperm antibody response to his vasectomy presents complex problems that may account for some failures, particularly in couples where there are some spermatozoa in the ejaculate but pregnancy does not ensue. There is an association between the presence of seminal plasma antibody and infertility. 8C In fact, the class of antibody on the spermatozoa is most important, IgA being far more inhibitory than IgG. 11A Following vasectomy reversal, the conception rate falls in proportion to the percentage of spermatozoa affected by antibody: from 86% with a pure IgG response to 43% when IgA is present as well to 22% when 100% are covered with IgA, and ultimately to zero when there is IgA on all the spermatozoa and a strong systemic immune response as shown by a high serum antibody titer. 10 Finally, spermatogenesis may have ceased. Although significant impairment of spermatogenesis following vasectomy is extremely uncommon, 1A the possibility should be considered among this population of patients, many of whom are entering middle life at the time that this surgery is carried out. CHAPTER REFERENCES
1. Amelar RD, Dubin L. Vasectomy reversal. J Urol 1979;121:547–550. 1A. Bagshaw HA, Masters JRW, Pryor JP. Factors influencing the outcome of vasectomy reversal. Br J Urol 1980;52:57–60. 1B. Ball RA, Steinberg J, Wilson LA, Loughlin KR. Comparison of vasovasostomy techniques in rats utilizing conventional microsurgical suture, carbon dioxide laser, and fibrin tissue adhesives. Urology 1993;41:479–483. 2. Belker AM. Principles of microsurgery (Review). Urol Clin North Am 1994;21:487–504. 3. Belker AM, Thomas AJ, Fuches EF, Konnak JW, Sharlip ID. Results of 1469 microsurgical vasectomy reversals by the vasovasostomy study group. J Urol 1991;145:505–511. 3A. Cos LR, Valvo JR, Davis RS, Cockett AJK. Vasovasostomy: current state of the art. Urology 1983;22:567–575. 4. Craft I, Tsirigotis M. Simplified recovery, preparation and cryopreservation of testicular spermatozoa. Hum Reprod 1995;10:1623–1627. 5. Flickinger CJ, Herr JC, Sisak JR, Howards SS. Ultrastructure of epididymal interstitial reactions following vasectomy and vasovasostomy. Anat Rec 1993;235:61–73. 5A. Fallon B, Jacob E, Bunge RG. Restoration of fertility by vasovasostomy. Journal of Urology, 1978;119:85–86. 5B. Fitzpatrick TJ. Vasovasostomy: the flap technique. Journal of Urology 1978;120:78–79. 5C. Hagan K, Coffey DA. The adverse effects of sperm during vasovasostomy. J Urol 1977;118:269–273. 6. Hendry WF, Levison DA, Parkinson MC, Parslow JM, Royle MG. Testicular obstruction: clinicopathological studies. Ann R Coll Surg Engl 1990;72:396–407. 7. Hendry WF, Parslow JM, Parkinson MC, Lowe DG. Unilateral testicular obstruction: orchidectomy or reconstruction? Hum Reprod 1994;9:463–470. 7A. Howard G. Who asks for vasectomy reversal and why? Br Med J (1982;285:490–492. 8. Kabalin JN, Kessler R. Macroscopic vasovasostomy re-examined. Urology 1991;38:135–138. 8A. Kessler R, Freiha F. Macroscopic vasovasostomy. Fertility and Sterility 1981;36:531–532. 8B. Lee L, McLaughlin MG. Vasovasostomy: a comparison of macroscopic and microscopic techniques at one institution. Fertility and Sterility 1980;33:54–55. 8C. Linnet L, Hjort T, Fogh-Andersen P. Association between failure to impregnate after vasovasostomy and sperm agglutinins in semen. Lancet 1981;1:117–119. 9. Matsuda T, Horn Y, Yoshida O. Unilateral obstruction of the vas deferens caused by childhood herniorrhaphy in male infertility patients. Fertil Steril 1992;58:609–613. 10. Meinertz H, Linnet L, Andersen PF, Hjort T. Antisperm antibodies and fertility after vasovasostomy: a follow-up study of 216 men. Fertil Steril 1990;54:315–321. 10A. Middleton RG, Smith JA, Moore MH, Urry RL. A 15-year follow-up of a non-microsurgical technique for vasovasostomy. J Urol 1987;137:886–887. 11. Parkhouse H, Hendry WF. Vasal injuries during childhood, and their effects on subsequent fertility. Br J Urol 1991;67:91–95. 11A. Reference not available. 11B. Reference not available. 11C. Reference not available. 12. Schlegel PN, Palermo GD, Alikani M, et al. Micropuncture retrieval of epididymal sperm with in vitro fertilization: importance of in vitro micromanipulation techniques. Urology 1995;46:238–241. 13. Sharlip ID. What is the best pregnancy rate that may be expected from vasectomy reversal? J Urol 1993;149:1469–1471. 13A. Silber SJ. Microscopic vasectomy reversal. Fertility and Sterility 1977;28:1191–1202. 13B. Silber SJ. Epididymal extravasation following vasectomy as a cause for failure of vasectomy reversal. Fertility and Sterility 1979;31:309–315.

14. Silber SJ. Pregnancy after vasovasostomy for vasectomy reversal: a study of factors affecting long-term return of fertility in 282 patients followed for 10 years. Hum Reprod 1989;4:318–322. 14A. Soonawalla FB, Lal SS. Microsurgery in vasovasostomy. Indian Journal of Urology 1984;1:104–108. 15. Thomas AI, Howards SS. Microsurgical treatment of male infertility. In: Lipshultz LI, Howards SS, eds. Infertility in the male, 2nd ed. St Louis: Mosby Year Book, 1991;357–369. 16. Urguhart-Hay D. A low-power magnification technique for reanastomosis of the vas. Br J Urol 1981;53:446–469.

Chapter 60 Varicocele Glenn’s Urologic Surgery

Chapter 60 Varicocele
Alain Jardin

A. Jardin: Hopital de Bicêtre, Université Paris Sud, 94275 Kremlin Bicêtre, France.

Diagnosis Indications for Surgery Alternative Therapy Surgical Techniques Inguinal (Ivanissevich) Approach Retroperitoneal (Palomo) Approach Outcomes Complications Results Chapter References

Varicocele is a dilation of the veins of the spermatic cord. It is almost always situated on the left side for anatomic reasons: the left spermatic vein is long, consists of a venous plexus at several points along its course, and drains at right angles to the left renal vein ( Fig. 60-1). Because of the upright posture of humans, the left renal vein is compressed between the aorta and the superior mesenteric artery, resulting in renal vein hypertension. The resulting hypertension promotes reflux of the left renal vein into the spermatic vein and is exacerbated by the valvular incompetence frequently seen in the spermatic vein.

FIG. 60-1. Venous anatomy of the retroperitoneum showing the relationship of the spermatic veins to the vena cava and renal veins.

Varicocele is rare in children and generally appears at the age of puberty, with development of the testes. It is estimated that varicocele affects 10% to 15% of postpubescent men. In the 1950s, a relationship was demonstrated between varicocele and male infertility. Although the frequency of sperm count abnormalities in patients with varicocele has not been clearly established, 20% to 40% of men consulting an infertility unit present with a varicocele. 9 The mechanisms explaining infertility related to varicocele are still controversial. The increased temperature of the scrotal contents induced by varicocele is universally accepted, and alteration of spermatogenesis by increased temperature has been demonstrated many times. 6 Also, it appears that varicoceles interfere with the normal growth of the testicle as evidenced by the fact that the left testis is smaller in more than one half of men with varicocele. 4

DIAGNOSIS
Varicocele is easy to diagnose clinically. It is generally visible or palpable in the upright position. During abdominal straining, the pressure in the left spermatic veins is increased as a result of the increased pressure in the inferior vena cava and therefore in the left renal vein. Descriptive grading of varicoceles is generally expressed as: Grade 1: palpable only during a Valsalva maneuver. Grade 2: palpable at rest but not visible. Grade 3: visible and palpable at rest. In practice, if the question of the presence or absence of a varicocele is raised during clinical examination, then the patient can be considered not to have a varicocele. The prevalence in a population of normal men of subclinical varicocele, demonstrable only by special tests such as a Doppler examination, is unknown. We do not consider varicoceles detected exclusively on Doppler examination to be real varicoceles, and Doppler examinations are not indicated in this disease. Because of its consequences in regard to fertility, the testicular volume is assessed clinically or by ultrasound. The left testis is generally smaller than the right testis in patients with varicoceles by a difference greater than 2 ml. 3,4 The FSH level is variable but usually greater than normal. A greater than normal response to gonadotropin hormones has been reported. 4 However, the best and most direct measure of the degree of impairment of fertility is the semenanalysis. Parameters associated with a varicocele include a reduction of the number of spermatozoa in the ejaculate, a reduction of the concentration of spermatozoa, decreased motility of spermatozoa, and an increased number of morphologic abnormalities. No test is available to determine whether varicocele is the primary cause of infertility in any given patient. It is therefore important, because of the high incidence of varicoceles in infertile men, to look for another factor in infertility (genital infection, prostatogenital abnormality, smoking, etc.).

INDICATIONS FOR SURGERY
The indication for surgical treatment of varicocele is still the subject of debate. Surgery for varicocele is not justified in asymptomatic adults over the age of 20 years with a normal sperm count. However, if such a man has not yet had children, it can appear reasonable to propose sperm storage because of the possibility of progressive alteration of the spermatozoa. 4 Surgical treatment of varicocele appears justified in an infertile man with an altered semenalysis, however severe the abnormalities are, provided any other identified cause of infertility has been treated, the patient has stopped smoking, and any gynecologic abnormalities of the partner have been investigated and treated. This protocol has actually been modified by the success obtained with medically assisted procreation and especially intracytoplasmic sperm insertion (ICSI). In women over the age of 35, it may be justified to perform medically assisted procreation immediately when the couple has been trying to become pregnant for more than 2 years. This age limit may be lower in women with gynecologic, particularly tubal, abnormalities. Adolescent varicocele has been the subject of many publications over recent years. 1,2,3 and 4 Studies with long follow-ups are still required in order to establish the definitive approach to this major public health problem, as an attitude of systematic surgery in adolescents would mean operating approximately 10% of the male adolescent population for varicocele. In adolescents (less than 20 years old), our philosophy is to offer surgery to those patients with a large, bothersome varicocele and in patients with a nonbothersome varicocele to offer annual surveillance when the two testes are symmetric and surgical treatment when the testes are

asymmetric.

ALTERNATIVE THERAPY
In the treatment of varicocele, we prefer the Palomo (retroperitoneal) approach. The other surgical options include other incisional surgery approaches, laparoscopic varicocele ligation, and percutaneous embolization. Regardless of the incision and surgical procedure used, the objective is practically always to interrupt the venous blood flow between the renal vein and the testis. Among the incisional approaches, some authors advocate the use of the operating microscope, 2 especially when the spermatic artery must be preserved. The benefits of microsurgery appear to be debatable. Other authors propose the inguinal (Ivanissevich) approach or the transscrotal approach. Varicocelectomy was one of the first urologic operations to be performed by laparoscopy. This relatively easy technique (but with greater risk) cannot claim to provide as complete a view of the retroperitoneal space, in which the spermatic vessels travel, as open surgery and does not appear to provide sufficient benefit to replace conventional surgery. 1,2,3,4,5,6,7 and 8 It is relatively easy to sclerose or embolize the spermatic vein(s) with various materials under radiologic guidance, and results are good (overall occlusion rate is 73% ranging from 50% to 100%) but inferior to those of surgery (98% in our experience), especially in the presence of anatomic variants, 5 which can be detected by open surgical intraoperative radiography.

SURGICAL TECHNIQUES
A very large number of techniques for the treatment of varicocele have been reported over the centuries. Transscrotal surgery has been virtually abandoned and replaced by inguinal and suprainguinal procedures. Most surgeons agree on the need to interrupt all of the spermatic venous drainage, but opinions differ concerning the need to preserve the spermatic artery. 10 Inguinal (Ivanissevich) Approach The incision is made 2 cm above the symphysis pubis. The external oblique aponeurosis is then carefully divided to avoid injuring the underlying ilioinguinal nerve. The cord is mobilized, and a Penrose drain is inserted beneath the cord and retracted to gain exposure of the cord. The spermatic fascia is then incised, and the dilated vessels are identified. Each vein is isolated, doubly ligated with nonabsorbable suture, and excised. The external oblique aponeurosis is then closed with interrupted absorbable or permanent sutures, and the skin is closed. Retroperitoneal (Palomo) Approach It seems logical, in our opinion, to approach the spermatic vessels at a site where a spermatic vein is most clearly visible, i.e., in the iliac region ( Fig. 60-2). At this level, the operation inevitably spares the other two arteries of the testis, and as a result, sacrifice of the spermatic artery has virtually no consequences. 4,5,6,7,8,9 and 10

FIG. 60-2. Modified Palomo suprainguinal approach for varicocelectomy. Location of incision. Internal spermatic vein (ISV) found on posterior aspect of the peritoneum isolated and divided between ligatures.

The patient is placed in the dorsal supine position on an operating table allowing radiography. A horizontal iliac incision equidistant from the umbilicus and anterior superior iliac spine is made (7 to 10 cm, depending on the patient's adiposity). The external oblique aponeurosis is incised obliquely. The internal oblique is split 1 cm off the lateral edge of the rectus abdominis, and the transversus abdominis is incised. The peritoneum is dissected free from the abdominal wall and retracted. The spermatic vessels appear adherent to the peritoneum, making it important to remain close to the peritoneum. Continued dissection along the abdominal wall would lead posteriorly to the psoas muscle. Retraction of the peritoneum allows easy identification of the spermatic veins, and in fewer than 10% of cases, the spermatic artery is clearly visible, isolated from the rest of the spermatic structures, identified, and preserved. The spermatic vein or the largest vein encountered is raised by two sutures and incised over half of its circumference. A catheter adapted to the size of the vein is inserted into the distal end, and another is inserted into the proximal end. Each catheter is maintained in place by ligation of the two sutures inserted previously, and 15 cc of contrast agent is gently injected into each catheter ( Fig. 60-3 and Fig. 60-4). These x-rays allow visualization of all of the spermatic vein and its branches and the renal vein, demonstration of left, right, and suprapubic anastomoses, accounting for the occasional associated right varicoceles, and, most importantly, visualization of anastomoses and demonstration of other spermatic veins with aberrant courses. Failure to treat these vessels is responsible for recurrent varicocele.

FIG. 60-3. Technique of intraoperative venography in which contrast material is injected through angiocatheter toward testicle into internal spermatic vein below the level of ligation.

FIG. 60-4. Insertion of catheters.

The rest of the operation depends on the radiologic findings. In the case of a single vein and no collateral, the artery is identified and will be preserved only when it is not accompanied by a plexus of small veins indissociable from the artery. In the case of multiple veins with anastomoses, the abnormal collaterals are identified, and all vessels from the ureter to the wall are ligated. Spermatic vessels are generally resected over a distance of 7 or 8 cm and ligated by resorbable suture material (Vicryl 3-0). After verification of hemostasis, the muscles are closed with resorbable suture material (Vicryl 2-0): a running suture on the internal oblique and transversus abdominis and a running suture on the external oblique aponeurosis. Fascia superficialis and subcutaneous fat are closed by a resorbable running suture (Vicryl 4-0). The skin is closed by a nonresorbable running suture or staples.

OUTCOMES
Complications Surgical varicocelectomy is a simple, straightforward procedure that can be easily performed on an outpatient basis, and complications are rare. The most common “complication” consists of recurrence, which can occur immediately if a branch draining into the left renal vein has been left in place, or later (several weeks or months), when small collateral veins were missed and subsequently reconstitute a venous channel supplying the varicocele from left renal venous blood. The recurrence rate (1% to 15%) is minimized by performing intraoperative radiography. Hydrocele is a frequent complication after varicocelectomy (8% in our experience) and can occur several years after surgery. Its pathogenesis has never been perfectly established. Preservation of the lymphatics, by using the operating microscope in the inguinal approach, 2 may have a certain value in this context. Testicular atrophy is quite exceptional (a few isolated cases have been reported) and is associated with anatomic abnormalities of the arterial supply. Fear of this complication has led some authors to propose the systematic use of microsurgery, 2 though this does not appear to be justified in our opinion. Results An early varicocele treatment may spare testicular atrophy, and furthermore, the benefit may be bilateral. 3 It is impossible to predict for a given couple the success rate of varicocelectomy in terms of pregnancy. Pregnancy rates after varicocelectomy ranging from 7% to 75% have been reported in the literature, with a mean rate of about 25% to 30% of term pregnancies, and our own experience shows a rate of 24%. However, the pregnancies obtained cannot be clearly attributed to varicocelectomy, and the debate between supporters and opponents of varicocele surgery for infertility is far from being closed. 3 Even apparently serious studies comparing two randomized series of operated and nonoperated patients are inconclusive. 7 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Enquist E, Stein BS, Sigman M. Laparoscopic versus subinguinal varicocelectomy: a comparative study. Fertil Steril 1994;61:1092–1096. Goldstein M, Gilbert BR, Dicker AP, Duvosh J, Gnecco C. Microsurgical inguinal varicocelectomy with delivery of the testis: an artery and lymphatic sparing technique. J Urol 1992;148:1808. Hargreave TB. Debate on the pros and cons of varicocele treatment in favour of varicocele treatment. Hum Reprod 1995;10(Suppl 1):151–157. Kass EJ, Reitelman C. Adolescent varicocele. Urol Clin North Am 1995;22:151–159. Lenz M, Hof N, Kersting-Sommerhoff B, Bautz W. Anatomic variants of the spermatic vein: importance for percutaneous sclerotherapy of idiopathic varicocele. Radiology 1996;198(2):425–431. Mieusset R, Bujan L. Testicular heating and its possible contribution to male infertility: a review. Int J Androl 1995;18:169–184. Nieschalg E, Hertle L, Fischedick A, Behre HM. Treatment of varicocele: counselling as effective as occlusion of the vena spermatica. Hum Reprod 1995;10(2):347–353. Parra RO. Laparoscopic surgery in urology: refining indications and techniques. J Urol 1995;153:1178. WHO. The influence of varicocele on parameters of fertility in a large group of men presenting to infertility clinics. Fertil Steril 1992;57:1289–1293. Yamammoto M, Tsuji Y, Ohmura M, Hibi H, Miyake K. Comparison of artery-ligating and artery-preserving varicocelectomy:effect on post-operative spermatogenesis. Andrology 1995;27:37–40.

Chapter 61 Simple Orchiectomy Glenn’s Urologic Surgery

Chapter 61 Simple Orchiectomy
Sherri M. Donat

S. M. Donat: Department of Surgery, Division of Urology, Memorial Sloan-Kettering Cancer Center, New York, New York 10021.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Simple Orchiectomy Scrotal Approach Suprapubic Approach Subcapsular Orchiectomy Subepididymal Orchiectomy Outcomes Complications Results Chapter References

Simple orchiectomy involves the removal of one or both testes at the distal cord, usually through an anterior transscrotal approach, although it has also been described through a transpubic approach. 6 It is used in the treatment of benign intrascrotal processes and as a method of hormonal ablative therapy in patients with advanced prostate cancer. Among men in the United States, prostate cancer is the most commonly diagnosed malignancy, with 317,000 new cases identified in 1996 and approximately 41,000 deaths attributed to it. 9 Since Huggins et al. first demonstrated the therapeutic benefit of hormonal ablation in the treatment of advanced prostate cancer, bilateral scrotal orchiectomy has been commonly utilized as a means of removing the testosterone-producing tissue, thereby bringing serum levels to castrate levels. 5 Although it has been traditionally used only for treatment in patients with advanced or metastatic disease, it is currently being evaluated for patients with presumed localized disease in both neoadjuvant and adjuvant settings to determine if it has any benefit in decreasing the chance of local or systemic recurrence and improving survival when used in combination with the traditional monotherapies of surgery or radiation. 2

DIAGNOSIS
The initial diagnosis of prostate cancer is usually made through a combination of digital rectal exam, prostate-specific antigen level, patient symptoms, and transrectal ultrasound-directed biopsy, although it is occasionally found incidentally on transurethral resection for obstructive benign disease. Advanced disease may involve lymph nodes, bone, or, less commonly, visceral or soft tissue lesions and may be documented by physical exam, elevated prostatic acid phosphatase levels, abnormal bone scan, CT scan, or MRI, plain film bone survey, or chest x-ray. Lymph node involvement may be determined by biopsy of enlarged nodes seen on imaging studies or unexpectantly during the pelvic lymph node dissection for a radical prostatectomy. Benign intrascrotal processes such as epididymo-orchitis or devitalization of testicular tissue by trauma or torsion are diagnosed by physical exam, patient symptoms, nuclear testicular exams, color Doppler ultrasonic exam, and/or surgical exploration. In general, inflammatory processes show increased flow on both nuclear scans and Doppler flow studies, and processes causing devascularization show decreased or no flow on nuclear and Doppler flow studies. However, if there is any question as to whether an acute scrotum represents a testicular torsion versus an epididymo-orchitis, it should be surgically explored immediately to answer the question and render the appropriate treatment.

INDICATIONS FOR SURGERY
The primary indication for bilateral scrotal orchiectomy is advanced prostate cancer requiring hormonal ablation as treatment. Other indications for simple orchiectomy include benign intrascrotal disorders such as a traumatic injury to the testis requiring partial or complete removal of the devitalized tissue, testicular necrosis following prolonged torsion, and severe epididymo-orchitis that is refractory to antimicrobial therapy. Removal of testicular neoplasms through a transscrotal approach is contraindicated because of the increased risk of local recurrence; therefore, if there is any question of a testicular mass, it should be approached through an inguinal incision.

ALTERNATIVE THERAPY
There are now several agonist analogs of gonadotropin-releasing hormone that inhibit the pituitary gonadal axis, resulting in the down-regulation of LH-RH receptors and subsequent decrease in gonadotropin secretion. These are equally effective in achieving hormonal ablation and provide an alternative to orchiectomy in patients for whom the psychological implications of the surgery are too great. 2 The cost effectiveness of surgery is certainly superior to chemical castration. Alternatives to the total removal of the testis and epididymis have been explored and include subcapsular orchiectomy and subepididymal orchiectomy. 1,4,9 These give the cosmetic effect of a testis being present but also achieve the therapeutic goal of androgen ablation. There has been some controversy 7 over the efficacy of subcapsular orchiectomy since it was first described in 1942 by Riba 9; however, multiple modern studies have demonstrated its efficacy and suggest that any residual testosterone production is most likely a result of incomplete removal of the intratesticular contents. 1 Testicular prostheses are also available for use in patients who desire a better cosmetic result.

SURGICAL TECHNIQUE
Simple Orchiectomy This procedure may be performed with local, regional, or general anesthesia. Local anesthetic sensory blockade is obtained by infiltrating the spermatic cord in the region of the vas deferens in the high scrotum just below the pubic tubercle with a 0.5% bupivacaine solution. Care must be taken to ensure that the block is not injected intravascularly by drawing back on the syringe prior to injecting the medication. The same solution is then injected subcutaneously at the site of the anterior scrotal incision. Sensory blockade should then be tested by pinprick before the beginning of the procedure. Scrotal Approach The anterior scrotal wall is shaved to remove existing hair, and the scrotum and genitalia are prepped in a sterile manner. A 2.5- to 3-cm midline incision is made just through the skin along the median raphe in the anterior scrotal wall using a #15 blade scalpel while the assistant pushes a testicle toward the incision between his thumb and index finger so that the testicle lies directly under the incision ( Fig. 61-1). By electrocautery, the incision is then carried down through the dartos and cremasteric layers until the parietal portion of the tunica vaginalis is incised directly over the testis. This is usually notable for a gush of fluid from the peritesticular space. The incision in the tunica vaginalis is lengthened in both directions far enough to allow exposure of the entire testicle through the wound. The surrounding tunics are freed from the spermatic cord by a combination of blunt and sharp dissection. Meticulous hemostasis may be obtained in each layer as it is entered with electrocautery. Once the spermatic cord is isolated, the vas deferens is separated, doubly clamped, divided, and ligated with 2-0 Vicryl ties ( Fig. 61-2). The remainder of the cord structures may be divided into one or more bundles and are doubly clamped on the proximal side and singly clamped on the distal side. Once divided, the proximal portion of the cord is ligated with an 0 Vicryl free tie behind the most proximal clamp which is then removed, and an 0 Vicryl suture ligature is placed just distal to the free tie ( Fig. 61-2). Before the cord is released to retract proximally, the tunics, dartos, and subcutaneous areas are again inspected for hemostasis. Once this is felt adequate, the cord is allowed to retract, and attention is turned to the opposite testicle. It is then removed through the same midline incision, in the same

manner. This leaves two openings in the tunica vaginalis and dartos layers, which are separated by a median septum. These deep layers are then closed in one layer using a 3-0 Vicryl running suture. Allis clamps are placed at either end of the median septum to facilitate the exposure ( Fig. 61-3). The skin is then closed with interrupted 3-0 chromic sutures, and a gauze dressing is applied. Drains are not required but can be considered if there is doubt about hemostasis. Compression or turban dressings may also be used if there is concern over postoperative hemostasis or edema.

FIG. 61-1. A midline scrotal incision over the median raphe allows access to both testes.

FIG. 61-2. Vas deferens is ligated separately from the vascular structures of the cord. The vascular structures are doubly ligated with both a free tie proximally and a suture ligature distally using 0 Vicryl.

FIG. 61-3. Closure of the deep layers of the scrotum including the dartos musculature and testicular tunics laterally and the midline septum in one running layer.

Suprapubic Approach This approach is advantageous in patients in whom you want to avoid a scrotal incision and/or place testicular prostheses at the time of the orchiectomy. The patient again is placed in the supine position, and the suprapubic area is shaved. A 4- to 5-cm transverse incision is made in the midline approximately 2 to 3 cm above the pubic symphysis (Fig. 61-4). This is extended down through the subcutaneous tissues to the level of the rectus fascia using electrocautery. The subcutaneous tissue is then swept bluntly toward the external inguinal ring, exposing the distal spermatic cord, which is then isolated just distal to the ring in the upper scrotum ( Fig. 61-5). Right-angle retractors are helpful to obtain exposure during this dissection by moving the incision toward the side being worked on. The spermatic cord is looped with a Penrose drain, and the testicle is delivered into the wound by placing upward pressure on the scrotum and testicle, with simultaneous upward traction on the cord with the Penrose drain. The gubernaculum is then divided, which mobilizes the testicle ( Fig. 61-5). The vas deferens and spermatic cord are then divided, as previously described. If testicular prostheses are desired, they are then placed into the empty scrotum after adequate hemostasis is ensured, and a pursestring suture of 3-0 silk is used to close the neck of each hemiscrotum ( Fig. 61-6). Scarpa's fascia is closed with 3-0 Vicryl suture, and the skin is closed with a 4-0 Vicryl subcuticular suture reinforced with benzoin and Steri-Strips.

FIG. 61-4. Midline suprapubic incision 2 to 3 cm above the pubis allows access to both testes at a level just below the external ring.

FIG. 61-5. Isolation of the cord with a Penrose drain. Division of the gubernaculum allows complete mobilization of the testis from the scrotum.

FIG. 61-6. If a prosthesis is desired, it is placed into the empty hemiscrotum. To prevent migration, a pursestring suture is used to close the neck of the hemiscrotum, eliminating the need for an anchoring suture.

Subcapsular Orchiectomy This approach is used in patients who desire the cosmetic effect of testicles being present without the use of testicular prosthesis. The operation is approached through the anterior scrotum as previously described ( Fig. 61-1). Once the testicle is delivered to the wound, the tunica albuginea is opened in midline in a cephalad-to-caudad fashion. Hemostats are placed on the edges of the capsule to provide traction, and an index finger is placed behind the capsule to invert it ( Fig. 61-7). This maneuver facilitates the removal of the parenchymal contents, which are swept to the midline using a gauze sponge. The midline attachment of the parenchyma is divided using electrocautery, and the remainder of the interior capsule is cauterized to ensure hemostasis and complete destruction of all testicular parenchyma (Fig. 61-8). This technique has also been described using a CO 2 laser. 10 The tunica albuginea is then closed using a running 3-0 Vicryl suture ( Fig. 61-9), and the residual testicular tunics, adnexa, and cord are returned to the scrotum. The deep layers of the scrotum are closed in one layer as previously described, as is the skin ( Fig. 61-3).

FIG. 61-7. Tunica albuginea is opened in midline opposite the epididymis, and the contents are swept bluntly to the midline attachment.

FIG. 61-8. The parenchymal tissue is then removed at its midline attachment with electrocautery, and the internal surface of the capsule is cauterized.

FIG. 61-9. The tunica albuginea is reapproximated with a running 3-0 Vicryl suture.

Subepididymal Orchiectomy This is another procedure offered as an alternative to simple scrotal orchiectomy for a more acceptable cosmetic result. Again, the initial exposure to the testicles is the same as previously described for the anterior scrotal approach ( Fig. 61-1). Once the testicle is delivered to the wound, a vasectomy is performed including double ligation and division to minimize the possibility of postoperative epididymitis. 4 A line of dissection between the cleavage plane of the testis and the epididymis is utilized (Fig. 61-10). Dissection is started at the head of the epididymis, and the epididymal tissue is clamped for hemostasis. Care must be taken to secure the spermatic artery entering the testis at a point between the midportion and the tail of the epididymis. The clamped epididymal tissue is ligated using 3-0 Vicryl suture. Meticulous hemostasis is then obtained using electrocautery, and a running 3-0 Vicryl suture on a tapered needle is used to approximate the edges of the tunica albuginea over the raw surface of the epididymis ( Fig. 61-11). The remaining spermatic cord and epididymis are replaced into the scrotum, which is closed as previously described.

FIG. 61-10. Dissection of the testis from the epididymis, ligating the epididymal side over clamps with Vicryl suture.

FIG. 61-11. After removal of the testis and further hemostasis with electrocautery, the tunica of the epididymis is closed over the raw surface using a running suture.

OUTCOMES
Complications Complications can include infection, hematoma, edema, and the inadvertent removal of a testicular neoplasm through the scrotal approach. Of these, hematoma can be a significant problem because of the distensible nature of the scrotum, which prevents any tamponade effect. Reexploration with evacuation of the hematoma and placement of a drain may be required but can be ineffective when the bleeding dissects sub-cutaneously in the dartos layer. This has led to the development of several preventative measures to achieve adequate compression of the area postoperatively, including turban-type dressings or compression of the scrotal wall over a gauze bolster as described by Oesterling. 8 These techniques should not be substituted for being meticulous in obtaining hemostasis at the time of the procedure. Infection, depending on the degree of severity, is managed by incision and drainage of any abscess pockets. This is followed by local wound care, which may include wet-to-dry dressing changes, sitz baths, and whirlpool with debridement of any devitalized tissue. Antibiotics can be used in cases where there is induration only and no abscess to drain, or in combination with the incision and drainage procedure if needed. If antibiotics are used, they should be directed by the results of wound cultures. If a testicular neoplasm has been inadvertently removed through a scrotal approach, the remaining inguinal spermatic cord should be removed, as well as a wide excision of the scrotal scar. The hemiscrotum should also be removed if there was any known tumor spillage during the transscrotal procedure. This usually results in cure rates similar to those for conventional initial radical inguinal orchiectomy. 3 Results The effectiveness of scrotal orchiectomy in terms of achieving adequate hormonal ablation in patients with prostate cancer can be easily determined by measuring serum testosterone levels postcastration, which are reduced by approximately 90%. Castrate levels of testosterone have reportedly been achieved by 2 and 15 hours after surgery.1,2 Prostate-specific antigen levels, patient symptoms, and patient survival may be followed to determine its effectiveness in terms of disease control. Side effects may include loss of libido, impotence, and hot flashes. For benign intrascrotal processes such as epididymo-orchitis unresponsive to antibiotics, or devitalized tissues secondary to torsion or trauma, simple orchiectomy is curative. CHAPTER REFERENCES
1. Arcadi JA. Rapid drop in serum testosterone after bilateral subcapsular orchiectomy. J Surg Oncol 1992;49:35. 2. Cassady JR, Hutter JJ, Whitesell LJ. Prostate cancer. In: Vogelzang NJ, Scardino PT, Shipley WU, Coffey DS, eds. Comprehensive textbook of genitourinary oncology. Baltimore: Williams & Wilkins, 1996;557–828. 3. Giguere JK, Stablein DM, Spaulding JT, et al. The clinical significance of unconventional orchiectomy approaches in testicular cancer. A report from the Testicular Cancer Intergroup Study. J Urol 1988;139:1225. 4. Glenn JF. Subepididymal orchidectomy: the acceptable alternative. J Urol 1990;144:942. 5. Huggins C, Hodges CV. Studies on prostatic cancer: the effect of castration, of estrogen, and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1941;1:293. 6. Klein EA, Herr HW. Suprapubic approach for bilateral orchiectomy and placement of testicular prosthesis. J Urol 1990;143:765. 7. O'Conor VJ, Chaing SP, Grayhack JT. Is subcapsular orchiectomy a definitive procedure? J Urol 1963;89:236. 8. Oesterling JE. Scrotal surgery: a reliable method for the prevention of postoperative hematoma and edema. J Urol 1990;143:1201. 9. Riba LW. Subcapsular castration for carcinoma of the prostate. J Urol 1942;48:384. 10. Wishnow KI, Johnson DE. Subcapsular orchiectomy using the CO 2 laser: a new technique. Lasers Surg Med 1988;8:604.

Chapter 62 Inguinal Orchiectomy Glenn’s Urologic Surgery

Chapter 62 Inguinal Orchiectomy
David A. Swanson

D. A. Swanson: Department of Urology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas 77030.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Testicular tumors are relatively rare (only two to three cases per 100,000 men; an estimated 7,400 new cases in the United States in 1996). 6 When they occur, 94% are germ-cell tumors, and the rest are tumors of the gonadal stroma and secondary tumors of the testis. In black men throughout the world, germ-cell tumors occur infrequently, but they do occur, so this diagnosis cannot be excluded on the basis of race alone. Although the etiology of testicular tumors is not known, there is a relatively high association (reported in up to 12% of tumors) with a history of cryptorchidism; in 20% of such cases, the tumor is in the normally descended testis. Orchidopexy does not prevent the subsequent development of tumor; it simply makes the diagnosis easier to establish. Carcinoma in situ (CIS) is also known to be associated with cryptorchidism, and data support the hypothesis that at least some, if not all, germ-cell tumors originate as CIS. 2

DIAGNOSIS
Most testicular tumors present as a palpable nodule or painless swelling of the testis, often discovered incidentally by the patient or his sex partner. The differential diagnosis of a testicular mass includes tumor, epididymitis, and epididymo-orchitis (the two most common diagnoses other than cancer), torsion (the diagnosis of which also requires surgery), and, less commonly, hernia, hydrocele, spermatocele, varicocele, hematoma, and hematocele. The patient with cancer may complain of a dull ache or sense of heaviness. Acute onset of pain is relatively rare and usually indicates bleeding within the tumor or associated epididymitis. Signs and symptoms may be secondary to metastatic spread. When careful bimanual examination of the testis reveals an intratesticular mass, with or without an associated epididymal mass or tenderness, testicular tumor must be suspected. A transscrotal ultrasound is a widely available, rapid, sensitive, noninvasive, and inexpensive way to determine whether there is a solid mass within the tunica albuginea. 9 Color Doppler ultrasound might occasionally help differentiate testicular torsion. Magnetic resonance imaging can also demonstrate the presence of an intratesticular mass but has no established advantage over a careful physical examination plus ultrasound exam. With rare exceptions, all solid masses within the testis should be considered malignant until proven otherwise, and require surgical intervention. Patients with testicular tumors commonly have elevated tumor markers, particularly b-human chorionic gonadotropin and a-fetoprotein. However, the presence of normal marker levels does not exclude malignancy, and there is generally no advantage to waiting for the marker results before operating, although blood should always be drawn for these tests before orchiectomy. 9

INDICATIONS FOR SURGERY
The presence of an intrascrotal mass that cannot be clearly localized to outside the tunica albuginea is sufficient indication for surgical exploration through an inguinal approach. Although a sense of urgency is appropriate, it is not necessary to consider inguinal orchiectomy an emergency procedure; it can be scheduled during regular operating hours.

ALTERNATIVE THERAPY
For the patient who has equivocal findings on physical examination and ultrasound that make the diagnosis of epididymitis tenable, a short course of antibiotics may be tried. If there is no prompt improvement (less than 10 to 14 days), or if the serum tumor markers are elevated, inguinal exploration should be performed. Although radical (inguinal) orchiectomy is standard therapy for a solid lesion within the tunica albuginea, partial orchiectomy might be appropriate in the patient with a solitary testis or in the patient whose preoperative evaluation makes the diagnosis of an epidermoid cyst of the testis highly likely. 1,3,4,10 The safety of this approach has not been confirmed by extensive experience or long-term follow-up. 8

SURGICAL TECHNIQUE
Although regional anesthesia is acceptable, general anesthesia is preferred because of the short duration of the surgery and the possible reflex response to traction on the testicle and cord. With the patient in the supine position, and after adequate anesthesia, the lower abdominal wall, penis, and scrotum are cleaned with a surgical scrub and draped in a sterile fashion so that the palpable testicular mass and ipsilateral hemiscrotum are accessible in the surgical field. Traditionally, an oblique skin incision is made approximately 2 cm superior to the inguinal ligament, extending parallel to that ligament from just above the pubic tubercle laterally approximately 8 to 10 cm to a point overlying the internal inguinal ring. I prefer a more horizontal incision, extending approximately 5 to 8 cm from one fingerbreadth above the superior aspect of the internal inguinal ring laterally toward the internal inguinal ring, which might be slightly more cosmetic ( Fig. 62-1).

FIG. 62-1. An almost horizontal incision from just cephalad to the external inguinal ring laterally almost to the internal ring provides adequate exposure and an excellent cosmetic result. Alternatively, the incision may be made parallel to the inguinal ligament.

Deepen the incision with the knife or electrocautery through the subcutaneous tissue until the external oblique aponeurosis is reached. There are usually one or two significant veins that traverse this incision. They should be isolated and secured with hemoclips or 3-0 plain or chromic catgut sutures before being cut. When the external oblique aponeurosis is cleaned sufficiently to be well visualized, it is helpful to place one or two small self-retaining Gelpi (or similar) retractors in the wound to improve exposure. Next, use the scalpel to make a small incision in the external oblique aponeurosis midway between the internal and external inguinal rings, in the direction of its fibers. Insert Metzenbaum scissors through this opening and push the closed scissors with slight upward pressure underneath the aponeurosis to

the external inguinal ring and then laterally toward the internal ring ( Fig. 62-2A). This helps ensure that the ilioinguinal nerve will not be cut during the next step, which is to push the partially opened Metzenbaum scissors from the point of incision into the external inguinal ring and then laterally toward the internal ring as required, thus splitting the aponeurosis and opening the roof of the inguinal canal ( Fig. 62-2B). Careful inspection will usually reveal the ilioinguinal nerve, which should be freed up carefully by blunt and sharp dissection for the length of the incision. It can then be retracted out of the surgical field by passing two small hemostats underneath the nerve, grasping the superior edge of the aponeurosis, and retracting it in a cephalad direction. It may also help to grasp the inferior edge of the aponeurosis and retract it as well to fully expose the inguinal canal.

FIG. 62-2. (A) The external oblique fascia is entered and tented up over the length of the inguinal canal before cutting to help ensure that the ilioinguinal nerve is not injured. (B) The external oblique fascia is incised over the spermatic cord, and the incision is extended into the external ring.

This will expose the spermatic cord, although it may not appear very distinct because of the cremasteric muscle fibers surrounding the cord, which merge into the internal oblique muscle. Using a gauze sponge wrapped around the fingertip ( Fig. 62-3A) or a peanut (Küttner) sponge (Fig. 62-3B), bluntly develop the plane between the spermatic cord and floor of the inguinal canal until it can be encircled with a thumb and forefinger. It is usually easiest to both initiate the dissection and to completely encircle the cord at the level of the pubic tubercle. Once you have ascertained that all components of the cord are included, pass a ½-inch Penrose drain around the cord (Fig. 62-3C), elevate it with gentle traction, and free up the cord laterally to the internal ring using predominantly blunt dissection, although some sharp dissection may be required. Be careful as you approach the internal ring that the inferior epigastric vessels are not injured, and inspect the spermatic cord carefully to ensure that an indirect inguinal hernia, which could contain bowel or bladder, is not present. At this point, occlude the spermatic cord firmly with either a soft rubber-shod clamp or, as I prefer, the ½-inch Penrose drain encircled twice around the cord, tightened in a tourniquet fashion, and secured with a Kelley or right-angle clamp (Fig. 62-4). Be sure to leave enough spermatic cord distal to the internal ring to permit it to be double-clamped later without first removing the tourniquet.

FIG. 62-3. (A) The spermatic cord is bluntly mobilized from the inguinal ligament and floor of the inguinal canal with a gauze-sponge-wrapped finger, starting near the pubic tubercle. (B) The blunt dissection is performed superiorly and inferiorly to the cord and may be facilitated by using a peanut sponge. (C) When the entire spermatic cord is free, a ½-inch Penrose drain is passed around the cord.

FIG. 62-4. The Penrose drain is encircled twice around the cord just distal to the internal ring and clamped to act as a tourniquet. A finger outside the scrotum helps push the testis into the surgical field. The gubernaculum is clamped and cut, and the scrotal side of the gubernaculum is tied.

It is now possible to mobilize the testis from the scrotum and through the opened external inguinal ring into the inguinal canal and surgical field. Upward pressure on the external skin of the hemiscrotum and testis, coupled with gentle traction on the spermatic cord, will generally define the circumferential fibromuscular attachments that need to be cut or, better, electrocoagulated to completely free up the testis ( Fig. 62-4). At the most inferior aspect of the testis, there may be a well-defined gubernaculum, which needs to be clamped and cut, with care taken to exclude scrotal skin, and tied with 2-0 or 3-0 chromic catgut. At this point, isolate the testis and spermatic cord, now free to the level of the internal ring, with sterile towels, and carefully inspect it. If the diagnosis is still in doubt, you may open the tunica vaginalis and expose the tunica albuginea of the testis. If doubt still persists, which should happen only very rarely, a small incision may be made in the tunica albuginea to permit insertion of a finger for palpation of the testicular parenchyma. If all of these maneuvers fail to exclude tumor, the surgeon should proceed with radical orchiectomy rather than risk returning a testis with tumor to the scrotum. I want to emphasize that it is only rarely necessary to open even the tunica vaginalis, and far more rare to open the tunica albuginea or perform a biopsy. To complete the orchiectomy, double-clamp the spermatic cord at the level of the internal ring and proximal to the Penrose tourniquet with two Kelley or heavy right-angle clamps; add a third clamp to occlude the cord just distal to the Penrose tourniquet ( Fig. 62-5). Transect the cord and remove the testicle, with attached spermatic cord, from the surgical field. The cord is tied behind the most proximal clamp with a 0 silk tie, and a suture ligature of 0 silk is placed behind the most distal clamp. Leave one of the two sutures long for later identification of the stump of the cord if a retroperitoneal lymph node dissection is performed. Some surgeons prefer to tie the cord in two portions, separating the spermatic vessels and the vas deferens, and double-tying both portions of the cord. In either case, after the cord is securely tied, allow it to retract through the internal ring into the retroperitoneum.

FIG. 62-5. The spermatic cord is triple-clamped just distal to the internal ring (two clamps proximal and one distal to the Penrose tourniquet) and cut after careful inspection of the testis, which has been isolated from the surgical field on a sterile towel.

Next, carefully inspect the entire floor of the inguinal canal as well as the scrotal compartment by everting the scrotal wall into the surgical field with upward external pressure on the most dependent portion of the hemiscrotum. Control all sites of bleeding, even very tiny ones, with the electrocautery and then irrigate with sterile water. It is prudent to perform a final inspection for complete hemostasis at this point before closure. Begin the closure by careful inspection of the inguinal floor. If it seems weak, it can be reinforced with several interrupted sutures in a standard hernia repair. If not, release the ilioinguinal nerve and close the external oblique aponeurosis with interrupted 2-0 silk or 2-0 Prolene sutures, placing the sutures at varying distances from the aponeurotic edge to prevent a linear tear and taking care to exclude the ilioinguinal nerve ( Fig. 62-6). The closure should begin at the level of the internal ring and extend medially as close to the pubic tubercle as possible because there is no longer any reason to have an external inguinal ring. A drain is not necessary or advisable. Irrigate the wound once again, and close the skin incision with clips or with a running subcuticular suture of 4-0 Vicryl. Cover the wound with a dry sterile dressing and gently compress the scrotum with either fluffed gauze sponges held in place with an athletic supporter or by gently wrapping the scrotum with a loosely applied turban dressing of Kling, Kerlix, or Coban. Avoid using an ice pack because it has little impact on swelling and is a source of considerable discomfort.

FIG. 62-6. The stump of the cord retracts back through the internal ring into the retroperitoneum; one suture is left long. The external oblique fascia is closed with interrupted sutures placed at varying distances from the cut edge and completely closing the external ring.

Patients may resume a regular diet and ambulation when completely awake and may be discharged. Most patients require oral narcotic analgesics for pain control for several days.

OUTCOMES
Complications Many would consider the most serious complication to be scrotal violation, which in the past required hemiscrotectomy and now requires irradiation to the scrotum (for seminoma). In truth, no data demonstrate reduced survival following scrotal contamination. 5 Nonetheless, there is virtually no reason why the testis should be approached transscrotally if the diagnosis of possible testicular tumor has been even considered. At the very least, a transscrotal approach prevents early control of venous outflow in the cord before manipulation of the tumor, it leaves spermatic cord behind, it potentially alters lymphatic drainage of the testis, it risks tumor contamination of the scrotum, and it may preclude consideration of surveillance as a treatment option. 7 The most common actual complication is probably an intrascrotal hematoma. Because the scrotum is such an expansile organ with loose areolar tissue beneath the dermis, bleeding may continue because of a lack of tamponade, and the resulting hematoma may grow quite large. These scrotal hematomas usually become organized and quite firm and may even raise the question of residual or recurrent tumor. Nonetheless, if the hematoma does not become infected (which would require surgical drainage), it can almost always be followed expectantly and will eventually regress. Ideally, this complication should be prevented. The depths of the inguinal canal and entire inner surface of the scrotal wall should be thoroughly and compulsively inspected and electrocoagulated to ensure hemostasis. The surgeon can facilitate this, as described earlier, by everting the scrotal wall with a finger positioned on the most dependent portion of the external scrotal wall. After hemostasis appears adequate, irrigate with sterile water and inspect again. Although the turban dressing used in bilateral orchiectomy, which completely collapses the scrotum, is not possible, it is possible to use a modified turban dressing that is wrapped firmly enough to collapse the hemiscrotum ipsilateral to the orchiectomy but not so tight as to cause pain because of pressure on the remaining testis. It is also possible to get a retroperitoneal hematoma if the ligature(s) on the spermatic cord pull off or if there is an injury to one of the inferior epigastric vessels. The first cause can be prevented by a properly tied ligature with adequate spermatic cord distal to it so it can't slip off. A suture ligature on the spermatic cord offers a measure of security. Injury to the epigastric vessels can be avoided by careful dissection of the proximal spermatic cord at the level of the internal inguinal ring. This complication is usually discovered incidentally at the time of further staging evaluation with a CT scan, although an unexplained and occult blood loss may prompt investigation. If bleeding has stopped when the problem is discovered, it virtually never requires specific treatment; the hematoma should ultimately be reabsorbed. Results Properly performed inguinal orchiectomy with the spermatic cord taken at the level of the inguinal ring is potentially curative if the tumor is still confined to the testis. Except for treatment of a complication, reoperation is virtually never required, although additional surgical procedures may be performed later to remove regional lymph nodes. CHAPTER REFERENCES
1. Berger Y, Srinivas V, Hajdu SI, Herr HW. Epidermoid cysts of the testis: role of conservative surgery. J Urol 1985;134:962. 2. Giwercman A, Rajpert-Meyts ERD, Skakkebaek NE. Carcinoma in situ of the testis: a new biological concept of urologic relevance and implications for detection and management. In: Vogelzang NJ, Shipley WU, Scardino PT, Coffey DS, eds. Comprehensive textbook of genitourinary oncology. Baltimore: Williams & Wilkins, 1996;941–952. 3. Heidenreich A, Bonfig R, Derschum W, von Vietsch H, Wilbert DM. A conservative approach to bilateral testicular germ cell tumors. J Urol 1995;153:10. 4. Heidenreich A, Engelmann UH, Vietsch HV, Derschum W. Organ preserving surgery in testicular epidermoid cysts. J Urol 1995;153:1147. 5. Leibovitch I, Baniel J, Foster RS, Donohue JP. The clinical implications of procedural deviations during orchiectomy for nonseminomatous testis cancer. J Urol 1995;154:935. 6. Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, CA 1996;46:5. 7. Pizzocaro G. Editorial comment. J Urol 1995;154:939. 8. Shapeero LG, Vordermark JS. Epidermoid cysts of testes and role of sonography. Urology 1993;41:75. 9. Watson DL, Kantoff PW, Richie JP. Staging and imaging of testis cancer. In: Vogelzang NJ, Shipley WU, Scardino PT, Coffey DS, eds. Comprehensive textbook of genitourinary oncology.

Baltimore: Williams & Wilkins, 1996;981–991. 10. Weissbach L. Organ preserving surgery of malignant germ cell tumors. J Urol 1995;153:90.

Chapter 63 Retroperitoneal Lymphadenectomy Glenn’s Urologic Surgery

Chapter 63 Retroperitoneal Lymphadenectomy
Michael A. S. Jewett

M. A. S. Jewett: Division of Urology, University of Toronto, Toronto, Ontario M5G 2C4, Canada.

Diagnosis Indications for Surgery Nonseminoma Stage I Nonseminoma Stage II Stage III Seminoma Alternative Therapy Surgical Technique Primary Procedure (No Previous Treatment) Postchemotherapy Procedure Outcomes Complications Results Chapter References

Carcinoma of the testis arises from the germinal epithelium and is the most frequent malignancy in young men aged 20 to 35 years. Virtually all patients are potentially curable with appropriate management. 5 There are two histologic types, seminoma and nonseminoma (the latter include embryonal carcinoma, teratocarcinoma, teratoma, yolk sac tumor and choriocarcinoma). After radical inguinal orchiectomy, seminomas are usually treated by radiation and/or chemotherapy. Retroperitoneal lymphadenectomy (RPL) is indicated in patients with nonseminoma for staging and treatment of early-stage disease. The etiology of these tumors is unknown, but there is a strong association with cryptorchidism. Trauma may draw attention to a mass but is not causative. There is a frequent association with infertility. For unexplained reasons the incidence is increasing in some parts of the world, notably Scandinavia and Northern Europe.

DIAGNOSIS
Characteristically the patients present with a painless mass in one testis that on examination is confined to the parenchyma. Ultrasound of the scrotum is employed to confirm the presence of the mass within the testicular parenchyma. Rarely the tumor will have an inflammatory component that produces pain and tenderness that can be mistaken for epididymitis. Up to 5% of tumors are bilateral, but they are usually asynchronous in presentation. Nonseminomas have a high propensity for metastases. The majority of metastases are lymphatic, although up to 10% appear to be hematogenous. The regional nodes that are involved are in the retroperitoneum, where the testis formed embryologically. Primary landing zones for nodal metastases differ between sides. 2 Right-sided tumors first metastasize to the interaortocaval nodes with less frequent involvement of the high preaortic and paracaval nodes. The left-side tumors metastasize to the left para-aortic region and the interaortocaval nodes. They rarely cross to the right side unless there is extensive metastatic disease. Extension via the thoracic duct to the supraclavicular node and systemic circulation occurs in approximately 40% of patients at presentation, who will characteristically have lung metastases. Clinical staging is defined as follows: stage I (or A) for tumor confined to the testis; stage II (or B) for metastatic disease confined to the retroperitoneal lymph nodes; and stage III (or C) for those with systemic metastases, usually to lung. The retroperitoneum is assessed by a CT scan that includes the pelvic area with special attention to the renal hilar area down to the inferior mesenteric artery, which is the level at which most metastases will occur. Lung metastases are identified by chest x-ray, reserving CT scan for special cases. The tumor markers a-fetoprotein (AFP) and human chorionic gonadotropin (b-hCG) are elevated in 80% of patients before orchiectomy and will usually remain elevated if there is metastatic disease. The half-lives are up to 5 days and up to 24 hours, respectively, for the two markers so that repeated determinations should be taken to discriminate between naturally decaying markers and ongoing production after orchiectomy. Lymphangiography is no longer used. The overall accuracy for staging in stage I is 70%; i.e., all tests are negative, but occult metastasis has occurred, usually to the retroperitoneum. For stage II, accuracy is 70% to 80% with approximately equal false-positive and false-negative rates of 20% to 30%. Patients with pure embryonal carcinoma are more likely to have metastatic disease, particularly if vascular invasion is seen in the primary tumor. Choriocarcinoma rarely occurs in its pure form and usually metastasizes hematogenously.

INDICATIONS FOR SURGERY
Nonseminoma Stage I The principal treatment alternatives after orchiectomy are surveillance or retroperitoneal lymphadenectomy. Patients with pure embryonal carcinoma plus vascular/lymphatic invasion have more than a 50% chance of metastases, and RPL should be considered. Patients with persistent elevation of markers are considered as a distinct stage subset and are generally treated with primary chemotherapy because of the high rate of systemic metastases, although RPL is an option. Nonseminoma Stage II Retroperitoneal lymphadenectomy is the standard, but if the nodal disease is >5 cm in diameter, or if there are multiple small nodes, primary chemotherapy should be considered because of the high rate of concurrent systemic metastases. Some centers use initial chemotherapy for disease greater than 2 cm in diameter. Stage III Patients are managed with initial chemotherapy, but all residual disease should be excised after the induction phase of therapy. Large retroperitoneal masses rarely completely disappear. The possibility of residual carcinoma or teratoma that can expand are the indications for surgery if a persistent mass is seen on CT of the retroperitoneum after three or four cycles of chemotherapy. The histology is useful to determine future treatment, as residual carcinoma may be an indication for further chemotherapy. 12 Preoperative bleomycin therapy creates a risk of pulmonary complications postoperatively. Low FiO2 and fluid restriction intraoperatively and postoperatively are mandatory to avoid an adult respiratory distress picture, which can be fatal. Seminoma Occasionally persistent retroperitoneal disease occurs after radiation or chemotherapy, often because there are nonseminomatous elements despite the fact that the primary was pure seminoma. These should be considered for retroperitoneal lymphadenectomy. It must be remembered that retroperitoneal masses in seminoma disappear slowly over as long as a 1 year after therapy.

ALTERNATIVE THERAPY
Patients with tumors confined to the testis (stage T1) who do not have vascular or lymphatic invasion and have elements of teratoma are less likely to have occult metastases and can be considered for surveillance. 10 Up to 85% to 90% of patients will require no further therapy, but follow-up must be close, particularly during the first 2 years. Surveillance protocols should include at least two monthly visits for the first 2 years with a CT scan of the abdomen every 4 months. Each visit should include history, physical examination, markers, and a chest x-ray. The mean time to failure is 6.5 months. Primary chemotherapy has been tested and is very

successful but has an associated morbidity that we consider to be greater than surveillance or RPL for low-risk patients. For stage II, primary chemotherapy when retroperitoneal metastatic disease is identified risks overtreatment, as 20% to 30% of these patients have false-positive CT scans. We believe surgical staging obtained by RPL to be important, and the procedure is therapeutic in most cases as well. There is clearly a tumor volume cut-point beyond which occult systemic metastases are so likely that primary chemotherapy is indicated. I believe that patients with a 5-cm CT diameter mass and/or marker levels that are significantly elevated at more than 100 units should be considered for primary chemotherapy. Radiation rarely has a role in nonseminoma management. Infertility may occur as a result of RPL because of damage to the sympathetic postganglionic nerves in the retroperitoneum. Therefore, patients should be advised to consider sperm banking. Infertility may also be preexistent in approximately 25% of patients. 8 Even poor-quality semen should be stored, as new reproductive technologies can use the DNA or immotile sperm. Bowel preparation is optional unless there is very extensive disease and one feels that an enterotomy may be necessary. Preoperative antibiotics are also optional. There have been a few reports of laparoscopic retroperitoneal lymphadenectomies, but the experience is extremely small and by surgeons with considerable laparoscopic experience. 7

SURGICAL TECHNIQUE
Primary Procedure (No Previous Treatment) We use a midline abdominal incision for the surgical approach, but a thoracoabdominal approach may be used, particularly with early-stage left-sided tumors, when the dissection may be limited to the left periaortic and preaortic nodal nodes. An anterior transabdominal midline approach is more commonly practiced and is described in detail. Patients may be given a mechanical bowel prep with 500 ml of 10% mannitol by mouth or the local protocol of choice. To ensure good hydration, an intravenous of normal saline is begun before the procedure. The RPL is usually performed through a midline abdominal incision, from xiphoid to midlower abdomen around the left side of the umbilicus to avoid the falciform ligament (Fig. 63-1A). Laparotomy is then carefully performed to assess the extent of retroperitoneal disease and to rule out other pathology. The small bowel and right colon are mobilized so that they can be rotated out of the abdomen and placed on the chest. The hepatic flexure is taken down, followed by incision of the posterior peritoneum along the right colic gutter to the cecum ( Fig. 63-1B). Care is taken to identify the right ureter during mobilization of the right colon, and the rest of the small bowel mesentery is incised cephalad to the ligament of Trietz ( Fig. 63-1C). Lymphatics are coagulated or preferably ligated when divided. The inferior mesenteric vein is routinely divided. Division of the vein allows rotation of the pancreas anteriorly to allow better access to the retroperitoneum. Using blunt and sharp dissection, the plane is developed between the bowel including the pancreas and the anterior leaf of the perinephric fascia. These structures are then placed into a bowel bag and rotated out of the abdomen onto the chest.

FIG. 63-1. (A) Midline incision from xyphoid to symphysis. (B) Incision of Toldt's line to allow reflection of right colon and exposure of retroperitoneum. (C) Continuation of line of peritoneal incision around root of bowel mesentery.

With the retroperitoneum now fully exposed, the limits of dissection can be defined. There is some controversy regarding the extent of dissection required for a given patient, but the following are reasonable recommendations: 1. For the grossly normal retroperitoneum at laparotomy, it seems reasonable to remove a template of nodes 3 that for right-sided tumors would include nodes to the left of the aorta to the ureter inferior to the para-aortic upper border of the renal vein, all the intra-aortocaval nodes, the precaval and paracaval nodes; below the aortic bifurcation, the nodes along the right common iliac should be removed to its bifurcation with the right ureter as the right lateral limit ( Fig. 63-2A). The sympathetic nerves can be identified and preserved within the template. For left-sided tumors, dissection would include para-aortic nodes from the renal vein to the bifurcation of the common iliac medial to the left ureter and nodes on the front of the aorta inferior to the renal vein to the inferior mesenteric artery ( Fig. 63-2B). Removal of the interaortocaval nodes is optional, almost always unnecessary, and should be done only by surgeons familiar with nerve identification and preservation. Any suspicious node outside these boundaries should be biopsied, and, if positive, a bilateral dissection performed.

FIG. 63-2. (A) Template for right-sided tumor accounting for crossover to left side. (B) Template for left-sided tumor. (C) Full retroperitoneal dissection.

2. In the presence of multiple gross nodes but fewer than five positive and none greater than 2 cm, use the same limits, recognizing that with right tumors, there may be left periaortic nodal involvement in up to 4% to 17%, and left iliac in 4% to 7%. 4 Two percent of patients might have a suprahilar node in the interaortocaval region, but that might well be grossly recognized and removed. It would seem reasonable, therefore, to omit this region for this stage of disease. If one of the grossly enlarged nodes is greater than 2 cm in diameter but smaller than 5 cm, or if there are more than five nodes, a full bilateral RPL should be done (Fig. 63-2C). There will be no significant additional patient morbidity if more dissection is performed unless the additional dissection involves dividing further lumbar arteries, which may contribute to some back pain postoperatively. The most important decision is when not to proceed with RPL but to use initial chemotherapy. This should be done if there is more extensive disease. Fortunately, this does not occur very often with improved staging modalities. For bilateral dissection, the limits are the ureters laterally; superiorly, the inferior border of the superior mesenteric artery extended bilaterally at a level 1 to 2 cm above the renal arteries; and inferiorly, to at least the midpoint of the ipsilateral external iliac artery and to the bifurcation of the contralateral common iliac artery including the aortic bifurcation ( Fig. 63-2C), the anterior spinal ligament and psoas muscles are cleared. Templates, as suggested, can be defined within these areas. Full bilateral dissection is begun by incising the retroperitoneal tissue in the midline anterior to the inferior vena cava (IVC). The left side of the cava is exposed from

the left renal vein to its origin at the left common iliac vein by peeling the adipose and lymphatic tissue medially, taking care to cauterize or ligate small vessels ( Fig. 63-3A). The IVC is carefully rolled laterally with ligation and division of the left lumbar veins as required ( Fig. 63-3B). The right sympathetic chain lies posterior to the midline of the cava (Fig. 63-4). A review of the anatomy of the retroperitoneal sympathetics is important because nerve-sparing techniques specifically identify and preserve these nerves while still performing a thorough lymphadenectomy.

FIG. 63-3. (A) Dissection between the vena cava and aorta. (B) Retraction of the great vessels allows exposure of the lumbar vessels, which should be ligated.

FIG. 63-4. Relationship between the great vessels and sympathetic trunks.

Seminal emission results from sympathetic stimulation of the seminal vesicles and vas deferens with deposition of semen in the posterior urethra. The sympathetic fibers arise from the sympathetic trunk as postganglionic nerves (specifically T12–L3). The paired sympathetic trunks run along the medial margin of the psoas, lateral to the lumbar vertebrae. The ganglia that comprise the sympathetic chain in this area usually number four or five (range two to six) and are rounded or fusiform in shape with diameter 1 to 10 mm. Variation of number with fusion of adjacent ganglia is common. The overlying aorta partly conceals the left trunk, with the right trunk being partly covered anteriorly by the IVC ( Fig. 63-4). The lumbar sympathetic nerves arise from the ganglia opposite the first, second and third lumbar vertebral bodies ( Fig. 63-5).6 These nerves course in a variable fashion anteromedially to form a network of fibers lying on the anterior surface of the aorta. In addition, one often sees small groups of ganglion cells. Condensation of these nerve fibers and ganglia occurs with the formation of several nerve plexuses: superior mesenteric plexus, inferior mesenteric plexus and hypogastric plexus. There is a high degree of variability in location of these plexuses, with their nomenclature approximating adjacent aortic branches (intermesenteric nerves, which have no constant pattern, connect the plexuses). The lowermost plexus, the hypogastric plexus, can usually be found anterior to the bifurcation of the aorta and in the interiliac angle. From this plexus, the right and left inferior hypogastric nerves run inferiorly to form the pelvic nerves that will eventually supply sympathetic fibers to the bladder, prostate, urethra, periurethral glands, vas deferens, and seminal vesicles. The right lumbar veins can be followed posteriorly behind the cava as a guide to identify it. Although these lumbar veins normally run medial to the sympathetic chain, occasionally they can be lateral and even branch around the sympathetic chain. From the fusiform ganglia, fine postganglionic nerve fibers can be seen coursing anteromedially and inferiorly to the right side of the aorta. These fibers are the lumbar splanchnic nerves and can be individually skeletonized from the underlying interaortocaval lymphatic tissue, which is ultimately mobilized posteriorly from the anterior spinal ligament and withdrawn inferiorly, leaving the web of sympathetic nerves.

FIG. 63-5. Anatomy of the lumbar sympathetic nerves.

Before the surgeon proceeds further, the left ureter is visualized by creating a plane across the midline anterior to the retroperitoneal lymphatic tissue but behind the inferior mesenteric artery, mesocolic fat, and other soft tissues. This dissection may require division of the inferior mesenteric artery, but it should be several centimeters distal to its origin. The left ureter can be exposed to the perinephric tissue and reflected laterally to visualize the left psoas muscle, leaving the left para-aortic retroperitoneal adipose tissue containing the lymph nodes intact ( Fig. 63-6). This bulk of tissue is carefully reflected medially to expose the left sympathetic chain. The lumbar vessels on the left are also identified with this dissection as they course directly medially past the sympathetic chain. Again, as with the right-sided vessels, they occasionally are lateral or bifurcate around the chain. These vessels can be sacrificed as needed. The lumbar splanchnic nerves are identified originating from the sympathetic ganglia coursing medially and anteriorly. Superiorly, there may be a branch anterior to the renal artery coming from a higher ganglia, but this may be difficult to preserve.

FIG. 63-6. Reflection of left ureter and adipose tissue.

At this point, the surgeon has a sense of the individual patient's sympathetic anatomy, which is variable. The aorta can now be exposed in the midline by mobilizing the left renal vein and by splitting the soft tissue over it. Dissection is carried close to the aorta. The individual sympathetic nerve branches, having been identified on both right and left sides from preceding dissection, are seen on the anterior and lateral surfaces of the aorta. They are skeletonized, with care taken to preserve those branches that form variable plexuses on the anterior aorta. This allows withdrawal of the interaortocaval lymphatic tissue and the left para-aortic tissue from between the anterior spinal ligament posteriorly and the nerves anteriorly. The plexus at and below the level of the inferior mesenteric artery condenses to form the two hypogastric nerves that pass over the aorta and proceed inferiorly into the pelvis. Identification and preservation of these two nerves allow all lymphatic tissue over the aortic bifurcation and common iliac veins down to the sacral promontory to be removed without sacrificing ejaculation. Inferiorly, it is our practice to limit the dissection to a point approximately midway along the ipsilateral external iliac artery and to the bifurcation of the contralateral common iliac artery. The remnant of the spermatic cord is removed ipsilaterally with as much of the vas deferens as is easily removable along with the spermatic vessels to their attachments in the retroperitoneum. Frequently nodal metastases or tumor masses are immediately adjacent to or involve nerves. It is not necessary to preserve all postganglionic nerves to ensure ejaculation. Therefore, the surgeon should not hesitate to sacrifice nerves unilaterally or even in a limited manner bilaterally to ensure complete removal of disease. The retroperitoneum defined by the selected surgical margins should now be clear of all lymphatic and fatty tissue surrounding the great vessels, with the postganglionic sympathetic nerves remaining beside and on the aorta. The contents of the bowel bag are placed back into the abdomen, and the large and small bowel peritoneal reflections are closed in their normal relationships using a running suture of 1 chromic gut. The abdominal incision is closed using a running 1 polyglycolic suture with interrupted figure-of-eight sutures. Drains are not used. The management of the patient postoperatively is usually straightforward and uncomplicated. The aggressive replacement of isotonic fluids and colloid intraoperatively and immediately postoperatively is required because of high third-space losses into the large raw retroperitoneal space. Postoperatively, the patient frequently has a positive fluid balance. Central venous pressure readings and hourly urine output give a reasonable estimation of intravascular volume. Nasogastric suction is continued until the postoperative ileus resolves, usually within 2 to 4 days. Routinely the indwelling urethral catheter is removed on the first postoperative morning to decrease the risk of urethral stricture and infection. Diet is increased stepwise following removal of the nasogastric tube. Chest physiotherapy and early mobilization of the patient help to minimize postoperative complications. The average patient is discharged from hospital on the sixth postoperative day. Postchemotherapy Procedure Patients with residual retroperitoneal disease pose particular problems. A full bilateral dissection remains the procedure of choice, although many patients appear to have a localized mass. Some centers advocate “lumpectomy” rather than a more extensive dissection, but this controversy is not resolved. Microscopic disease may occur in nodes adjacent to the gross disease. The potentially extensive nature of the tumor may require nephrectomy en bloc resection of part or all of the aorta, and even resection of the inferior vena cava. Vascular surgical skill may therefore be necessary. Small lacerations or punctures of the great vessels are frequent and require suture with fine monofilament suture such as 5-0. The approach and limits of dissection are similar. Nerve sparing may be possible in up to 50% of these patients if the mass is confined. The areas of greatest challenge are at the points that the masses adhere to the great vessels. A combination of sharp and blunt dissection is required with good proximal and distal control where the vessels are normal. The actual plane of dissection may appear to be under the adventitia. The locations of small neovascularization vessels are difficult to predict but manageable. Comfort with normal anatomy is mandatory, and this surgery is not for the inexperienced. Distortion of normal anatomy and the frequent anomalies of retroperitoneal vessels can be challenging. Large blood loss is not uncommon, so patients need close monitoring intraoperatively by an experienced anesthetist. Finally, intraoperative biopsy can be misleading and is not recommended for treatment decisions.

OUTCOMES
Complications Intraoperative complications are unusual and are mainly confined to vascular damage of the renal or great vessels. 1 Topical papaverine to reduce vessel spasm is used throughout the procedure. It is now rare to experience vascular injury without significant volume of disease. In removing a bulky mass from the aorta, care is taken to define the appropriate plane of dissection. It is very easy to dissect in a subadvential plane on the aorta leaving a weakened vessel for potential aneurysmal dilation. If an appropriate plane of dissection cannot be developed, it may be safer to resect part of the vessel wall with primary closure or placement of a synthetic graft. In addition, vein or Gore-Tex patches have been used to close caval defects. The meticulous obliteration of lymphatics by ligature or cautery will decrease the incidence of lymphocele and ascites postoperatively. Occasionally a lymphocele can become symptomatic because of a local mass effect resulting in vague abdominal or flank discomfort. Rarely obstruction of the small bowel or ureter can occur secondary to the lymphocele mass. Management usually is by the percutaneous drainage if symptomatic, with ultrasound or CT scan used for guidance. Most resolve spontaneously, however. Early postoperative complications with RPL are similar to those for any major abdominal procedure. Aggressive chest physiotherapy should be commenced immediately postoperatively in all patients to minimize the risk of atelectasis and pneumonia, which occurs in approximately 1% of patients. A potential pulmonary complication seen in testis cancer patients is related to pulmonary fibrosis secondary to previous bleomycin chemotherapy. Intraoperatively the patient is maintained on a lower inspired oxygen concentration, and a slight negative fluid balance is maintained. Following the recognition of this association with chemotherapy, only the rare patient will have a serious complication from his chest. Finally, late bowel obstruction which occurs in about 1% of patients, can be minimized with careful reapproximation of the retroperitoneum during closure or by employing the thoracoabdominal approach. Results It is anticipated that 70% of patients with clinical stage I are pathologic stage I and do not need further treatment. The majority of the remaining 30% have retroperitoneal metastases that can be sterilized by surgery. Patients selected with low-volume stage II disease who have complete resections now fail in approximately 30% of cases without adjuvant chemotherapy. Therapy may therefore be reserved for those patients who progress. Alternatively, all positive-node patients may have adjuvant treatment. Stage I and II patients should have cure rates of virtually 100%. 5,9,11 Patients with very bulky retroperitoneal disease or systemic disease have a lower survival rate and, with multidisciplinary management and use of initial chemotherapy followed by salvage surgery, approximately 70% should be curable.

CHAPTER REFERENCES
1. Baniel J, Foster RS, Rowland RG, Bihrle R, Donohue JP. Complications of primary retroperitoneal lymph-node dissection for low stage testicular cancer. World J Urol 1994;12(3):139–142. 2. Donohue JP, Maynard B, Zachary M. The distribution of nodal metastases in the retroperitoneum from nonseminomatous testis cancer. J Urol 1982;128:315–320. 3. Donohue JP, Thornhill JA, Foster RS, Rowland RG, Bihrle R. Retroperitoneal lymphadenectomy for clinical stage A testis cancer (1965–1989): modifications of technique and impact on ejaculation. J Urol 1993;149:237–247. 4. Donohue JP, Thornhill JA, Foster RS, Bihrle R, Rowland RG, Einhorn LH. The role of retroperitoneal lymphadenectomy in clinical stage B testis cancer: the Indiana University experience (1965 to 1989). J Urol 1995;153(1):85–89. 5. Horwich A, Norman A, Fisher C, Hendry WF, Nicholls J, Dearnaley DP. Primary chemotherapy for stage II nonseminomatous germ cell tumors of the testis. J Urol 1994;151(1):72–78. 6. Jewett MA, Kong YS, Goldberg SD, et al. Retroperitoneal lymphadenectomy for testis tumor with nerve sparing for ejaculation. J Urol 1988;139(6):1220–1224. 7. Klotz L. Laparoscopic retroperitoneal lymphadenectomy for high-risk stage 1 nonseminomatous germ cell tumor: report of four cases. Urology 1994;43(5):752–756. 8. Lange PH, Chang WY, Fraley EE. Fertility issues in the therapy of nonseminomatous testicular tumors. Urol Clin North Am 1987;14:731–745. 9. Motzer RJ. Adjuvant chemotherapy for stage II nonseminomatous testis cancer: what is its role? Semin Urol Oncol 1996; 14:30–35. 10. Read G, Stenning SP, Cullen MH, et al. Medical Research Council prospective study of surveillance for stage I testicular teratoma. J Clin Oncol 1992;10:1762–1768. 11. Richie JP, Kantoff PW. Is adjuvant chemotherapy necessary for patients with stage B1 testicular cancer? J Clin Oncol 1991;9:1393–1396. 12. Steyerberg EW, Keizer HJ, Fossa SD, et al. Prediction of residual retroperitoneal mass histology after chemotherapy for metastatic nonseminomatous germ cell tumor: multivariate analysis of individual patient data from six study groups. J Clin Oncol 1995;13(5):1177–1187.

Chapter 64 Torsion of the Testicle Glenn’s Urologic Surgery

Chapter 64 Torsion of the Testicle
Giovanni Grechi and Vincenzo Li Marzi

G. Grechi (Deceased) and V. Li Marzi: Department of Urology, Clinica Urologica II, Università degli Studi di Firenze, 50139 Firenze, Italy.

Diagnosis Torsion of Testicular Appendages Indications for Surgery Alternative Therapy Surgical Technique Chapter References

Torsion of the testicle is an infrequent disorder and is classified as a surgical emergency because failure to act may result in testicular atrophy. As the testis rotates around its cord, it causes strangulation of the venous blood supply with secondary edema and hemorrhage and, if severe enough, will cause arterial obstruction and gangrene of the testis and epididymis ( Fig. 64-1, Color Plate 2).

FIG. 64-1. Gangrene of the right testicle in a boy 16 years old who had undergone a 720-degree twist of the cord. (See also Color Plate 2.)

Testicular torsion is possible at any age. Many cases occur in the neonatal period, although it is most common during adolescence, and the incidence slowly decreases with increasing age. 6 Lee et al. found that 26% of patients with torsion were over age 21. 7 The increased incidence during adolescence is likely because the testis enlarges approximately five- to sixfold at puberty. It is also thought that spasm of the cremaster muscle may be an initiating factor. The pathophysiology of torsion also differs by age because the site of torsion can be supravaginal, mainly occurring in the neonatal period, or intravaginal, the most common site during adolescence. In the newborn period, at the end of testicular descent into the scrotum, the gubernaculum has still not become wholly fixed to the scrotal wall. 3 Therefore, the testicle, epididymis, and tunica vaginalis are free to rotate within the scrotum in a vertical axis ( Fig. 64-2). In pubertal testicular torsion, the gubernaculum is attached to the scrotal wall, and the testis twists intravaginally ( Fig. 64-3). The left testis rotates counterclockwise, and the right one clockwise.

FIG. 64-2. Supravaginal torsion of the testicle.

FIG. 64-3. Intravaginal torsion of the testicle.

The position of the testicle is inherently unstable because its major axis is inclined forward and up, allowing the testis to fall forward. Additionally, the testis is mobile because of the oblique insertion of the tunica vaginalis and the cremaster muscle fibers into the tunica vaginalis, which may allow the testis to rotate. The cord, gubernaculum testis, and the mesenteric attachment from the cord onto the testis and epididymis should fix the testicle, preventing rotation. Several congenital abnormalities have been mentioned as predisposing factors for testicular torsion. Cryptorchid or retractile testes are prone to undergo torsion, as are testes with a voluminous tunica vaginalis inserting high up on the cord, which allows the testis to rotate within it. Anomalies of the spermatic cord, such as a short mesenteric attachment onto the testis and epididymis, are also considered as an etiology for testicular torsion. Testes prone to undergo torsion lie in the horizontal position with the patient standing, an abnormality identified in boys who had suffered, in the past, from transient testicular torsion followed by spontaneous detorsion.

5

DIAGNOSIS

Sudden severe pain in one testicle, followed by testicular swelling and reddening of the scrotal skin, is the classic presentation of testicular torsion. The patient may also present with lower abdominal pain, nausea, and vomiting. However, in some patients only moderate scrotal swelling and little or no pain may accompany torsion of the cord. Symptoms may be absent in the infant, and the patient may present only with lack of appetite, instability, and restlessness. The temperature is usually normal, and urinalysis is clear, although moderate fever and leukocytosis may develop within a few hours. Physical examination reveals a swollen, soft testicle; shortening of the cord by volvulus retracts the testis upward. Lifting the testicle up over the pubic symphysis may increase the pain, which may allow differentiation from epididymitis in which the pain is usually alleviated by this maneuver. Acute epididymitis is the most common differential diagnosis confused with testicular torsion. In the early stages of torsion, the diagnosis can be made if the epididymis can be palpable in an anterior position. However, within a few hours, swelling of the testis and scrotal edema make it impossible to distinguish the epididymis from the testis by palpation. Additionally, acute epididymitis is unusual before age 16 and is often complicated by urethral discharge. Other differentials include mumps orchitis, which is usually accompanied by parotitis and is unusual in the prepubescent period, traumatic orchitis, strangulated hernias, hematoceles, tumors, and torsion of a testicular appendage. 8 When the diagnosis is in doubt, real-time scrotal ultrasound, 9 color Doppler sonography, 1 or a testicular technetium-99m pertechnetate nuclear scan 10 can be helpful to differentiate torsion from epididymo-orchitis, although they should not be regarded as definitive discriminators. Scrotal ultrasound is increasingly used for the investigation of scrotal pain, although the sonographic pattern in the presence of acute and subacute torsion is variable. An inhomogeneous testis with both increased and decreased echogenicity ( Fig. 64-4) is the most frequently reported pattern and results in little morphologic information for detecting torsion. Color Doppler sonography is now widely available and seems to overcome some of drawbacks of diagnosing testicular torsion by detecting morphologic information as well as blood flow ( Fig. 64-5, Color Plate 3). However, the growing number of false-negative patterns in patients with twisted testes seems to dampen the enthusiasm for Doppler sonography. The cause for this failure is that in testicular torsion the cessation of blood flow is generally not instantaneous; strangulation of the venous blood supply occurs first, then secondary edema and swelling occur, and only later, as the process advances and the interval of testicular viability is short, arterial obstruction finally becomes apparent.

FIG. 64-4. Sonographic pattern of testicular torsion with both increased and decreased echogenicity.

FIG. 64-5. Color Doppler sonography: avascular pattern of the testis 10 hours after torsion in a 16-year-old boy. (See also Color Plate 3.)

Testicular blood flow scans using technetium-99m pertechnetate appear to be the most definitive tests and are thought to be accurate in 90% to 100% of cases. In the torsed testis the pattern is avascular, while the epididymis has normal flow. In pediatric patients the nuclear scans are difficult to interpret because they have smaller vessels with relatively low blood flow. Another limitation is that they provide poor morphologic information and are often difficult to perform in many settings on short notice. Torsion of Testicular Appendages On the upper pole of the testis there is a small vestigial appendage, the appendix testis, which is a remnant of the Müllerian duct. It may be sessile or pedunculated and in the latter form may spontaneously undergo torsion. The disorder leads to ischemic necrosis followed by inflammation and resorption. The event usually affects adolescent patients, although reports of instances of torsion in adults have been reported. 2 Sudden onset of testicular pain is the usual clinical presentation that is comparable to that seen in a patient with true testicular torsion. However, in a careful physical examination, it may be possible to palpate, shortly after onset, a small tender lump at the upper pole of the testicle, which appears to be blue if the skin is kept fastened over the mass. Lidocaine administered to the external inguinal ring may make possible a more conclusive investigation. Later after onset, the physical examination may show that the entire scrotum is swollen, making it indistinguishable from testicular torsion.

INDICATIONS FOR SURGERY
The presence of an acute testicular mass in the absence of definitive signs of an infectious process is an indication for surgery. Because the prompt diagnosis of torsion is important for possible testicular salvage, it ought to be performed on the basis of history and physical examination. The diagnosis can be corroborated with imaging, but the clinician should rely on his best judgment in the situation and have a low threshold for surgical exploration.

ALTERNATIVE THERAPY
Because this is considered to be a surgical emergency, there are few alternatives. Delay in treatment will result in loss of the testis. Attempts to manually reduce the testis are usually not successful, are difficult because of the pain the patient experiences with manipulation of the testicle, and do not address the underlying pathology, which must be repaired to prevent further episodes of torsion. Torsion of the testicular appendage is usually treated by local excision of the twisted appendix, although if a careful clinical judgment shows the problem to be a twisted appendage, no surgical exploration is required, and the pain will gradually decrease and the scrotal swelling subside.

SURGICAL TECHNIQUE
In the newborn period early surgical intervention has been the rule; however, most testicles appear to be necrotic at the time of exploration and result in orchiectomy. Some authors have indicated that a necrotic testis left in situ after testicular torsion may cause an immunologic injury to the normal opposite organ to occur, resulting in abnormal sperm count at adulthood. In an adolescent, time is a critical element in the management of testicular torsion; the disorder is a urologic emergency, and a swollen painful scrotum having a sudden onset in an adolescent must wisely be considered to indicate torsion. A negative exploration of the scrotum is far more acceptable than the loss of a testicle that might have been salvaged. After 4 to 6 hours, necrosis commonly will have become evident in those testicles that have undergone a 720-degree torsion of the cord. However, the more resistant interstitial cells may be not destroyed even though destruction of the seminiferous tubules has occurred. If the adolescent is seen within a few hours of onset, manual detorsion may be attempted. The left testis rotates counterclockwise, and the right one clockwise; therefore the torsed organ should be rotated in the contrary direction. Lidocaine administered to the spermatic cord near the external inguinal ring may make the procedure easier. Surgical exploration is performed through an inguinal incision, incising the skin, subcutaneous fat, and external oblique in a line from 2 cm lateral to the pubic tubercle and 2 cm above and parallel to the inguinal crease. Care is taken to avoid injury to the branch of the genitofemoral nerve that traverses the inguinal canal. The cord is isolated, and the testicle delivered through the incision. The testicle is untwisted and observed for viability. This may be assessed by observing the return of color, the return of Doppler flow, or by incising the tunica and observing for signs of arterial blood flow. If the testicle is viable, it is pexed to the scrotal wall by invaginating the scrotum and placing 3 triangulating sutures of 2-0 Vicryl in the scrotal wall and the tunica of the testicle ( Fig. 64-6). An incision is made either in the raphe of the scrotum or in the contralateral scrotal compartment, and the contralateral testis is pexed to prevent torsion of the testicle. The incision is closed with 3-0 silk in the external oblique aponeurosis and 3-0 Vicryl in the subcutaneous tissue and subcuticular in the skin.

FIG. 64-6. Triangular fixation technique.

An alternative approach would be a scrotal incision, which has the advantages of exploring the testicle and pexing its contralateral mate through one incision. However, if testicular tumor is a possible diagnosis, it is better to approach via the inguinal approach. Testicular torsion can occur also after a previous fixation if the tunica vaginalis is left intact and the organ is allowed to rotate intravaginally. vaginalis is recommended, and fixation of the tunica albuginea to the scrotal wall should be performed. CHAPTER REFERENCES
Allen TD, Elder JS. Shortcomings of color Doppler sonography in the diagnosis of testicular torsion. J Urol 1995;154:1508. Altaffer LF III, Steele SM Jr. Torsion of testicular appendages in men. J Urol 1980;124:56. Backhouse KM. Embryology of the normal and cryptorchid testis. In: Foukalsrud EW, Mengel W, eds. The undescended testis. Chicago: Year Book Medical Publishers, 1981;5. Chinegwundew FI. Acute testicular torsion following testicular fixation. Br J Urol 1995;76:268. Greaney MG. Torsion of the testis: a review of 22 cases: improved diagnosis and earlier correction. Br J Surg 1975;62:57. Guiney EJ, McGlinchey J. Torsion of the testis and spermatic cord in the newborn. Surg Gynecol Obstet 1981;152:273. Lee LM, Wright JE, McLoughlin MG. Testicular torsion in the adult. J Urol 1983;130:93. Odabas O, Aydin S, Yilmaz Y. Torsion of spermatocele. J Urol 1995;154:2143. Pryor JL, Watson LR, Day DL, Abbitt PL, Howards SS, Gonzalez R, Reinberg Y. Scrotal ultrasound for evaluation of subacute testicular torsion: sonographic findings and adverse clinical implications. J Urol 1994;151:693. 10. Stage KH, Schoenvogel R, Lewis S. Testicular scanning: clinical experience with 72 patients. J Urol 1981;125:334. 1. 2. 3. 4. 5. 6. 7. 8. 9.
4

Eversion of the tunica

Chapter 65 Scrotal Trauma and Reconstruction Glenn’s Urologic Surgery

Chapter 65 Scrotal Trauma and Reconstruction
Gerald H. Jordan

G. H. Jordan: Department of Urology, Eastern Virginia Medical School, Norfolk, Virginia 23510.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Scrotal Hematomas and Blunt Testicular Injury Penetrating Injuries to the Genitalia Avulsion Injuries Genital Burns Radiation Injuries Fournier's Gangrene Outcomes Complications Results Chapter References

The anatomy of the male genitalia is quite complex, particularly in the scrotum, where there are multiple fascial layers ( Fig. 65-1). From the standpoint of trauma, however, most of the fascial layers are relatively unimportant. Buck's fascia is related to the deep penile structures and is important in the containment of periurethral processes or the occasional hematoma associated with injury to the corpora cavernosa. In the scrotum, the analogous fascia to Buck's fascia—the external spermatic fascia—is essentially uninvolved with scrotal trauma. The dartos fascia of the penis becomes the tunica dartos in the scrotum and then reflects onto the perineum as Colles' fascia. Those fascial layers are in continuity with Scarpa's fascia over the abdomen and reflect over the thighs to become contiguous with the fascia lata ( Fig. 65-2).1,3

FIG. 65-1. (A) Cross section of the penis illustrating the fascial and structural components of the shaft of the penis. (B) Sagittal section of the pelvis demonstrating the fascial layers and structural component.

FIG. 65-2. (A) Perineal view of a trauma patient illustrating a hematoma contained by Colles' fascia (classic butterfly hematoma). (B) Same trauma patient illustrating the containment of the hematoma by the extended Colles' layer, i.e., the fascia lata of the thigh and extending onto the abdomen beneath Scarpa's fascia.

The vascular supply to the deep structures beneath the scrotum is a continuation of the vasculature to the hilum and crura of the penis and to the spermatic cords and testicles, i.e., the deep internal pudendal arteries. These arteries give off the labial–scrotal and perineal branches to continue on as the common penile artery ( Fig. 65-3). These labial–scrotal branches arborize to become the fascial plexuses of the tunica dartos of the scrotum, and in reconstruction, large skin islands can be carried on this blood supply ( Fig. 5-4). There also are often vessels within the scrotal raphe that further add to the dependability of these scrotal fascial flaps. Further blood supply to the scrotum is via the perineal artery, which arborizes to become the fascial blood supply and is the basic blood supply to the skin lateral to the scrotum, the basis for the “perineal axial fasciocutaneous flap” (Singapore flap). The superficial external pudendal vessels, the vessels of the medial circumflex artery, and the deep external pudendal arteries likewise send branches onto the scrotum to arborize there and become a portion of the fascial blood supply ( Fig. 65-4).

FIG. 65-3. Illustration of the distribution of the common penile artery.

FIG. 65-4. Illustration of the blood vessels felt to contribute to the fascial blood supply of the scrotum.

The venous anatomy of the deep structures of the penis is essentially in reverse of the arterial supply, with the subcoronal plexus becoming the deep dorsal vein and, in about 60% of cases, the superficial dorsal vein of the penis. The deep dorsal vein drains to the median periprostatic plexus, and the superficial veins drain laterally into the iliofemoral system. The crural and cavernosal veins, those that are oriented medially, drain medially, to Santorini's plexus, and those departing laterally drain to the iliofemoral system (Fig. 65-5).

FIG. 65-5. The venous drainage of the deep structures of the penis.

Culp has classified injuries to the external male genitalia into five categories: nonpenetrating, penetrating, avulsion, burns, and radiation injuries (both direct and indirect). 3 Nonpenetrating injuries result from either a crushing or sudden deforming force to the scrotum. These forces can cause severe damage to the internal structures without disrupting the skin. With any nonpenetrating trauma to the scrotum or perineum, one must suspect and rule out injury to the corpus spongiosum and urethra. Penetrating injuries to the genitalia frequently involve the urethra, and most require exploration, irrigation, and removal of foreign body material, if any, with anatomic repair and drainage. Because of the position of the urethra beneath the scrotum and perineum, penetrating injuries to the scrotum will often miss the urethra. Avulsion injuries of the genital skin most frequently involve the scrotal skin and occur when the patient's clothing becomes entangled in machinery. The loose scrotal tissues entangle and as the clothing is ripped off, so is the loose genital skin. These avulsion injuries vary from minimal injuries, which are essentially nothing more than lacerations, to emasculating injuries, which take not only the skin but also the deep structures. It is not unusual for a testicle to be avulsed along with the skin. The deep penile structures, however, are usually not avulsed. Avulsion, fortunately, usually takes only the skin and the tunica dartos, leaving the underlying Buck's fascia of the penis and the fascial layers surrounding the testicle intact. In most of these cases, bleeding is not a problem, as the skin and fascia are avulsed in a plane between the fascial structures related to the deep structures and the superficial fascia. Burns to the genitalia are usually not isolated injuries but reflective of a wider area of body burn. Flame and chemical burns are generally only superficial and involve the skin, but electrical burns can be devastating. Electrical burns disseminate via the deep vascular and neurologic structures, and what may appear to be a minimal burn to the skin and scrotum may have significant deep injury associated with it. Radiation injuries to the genitalia can occur either from direct exposure to the genitalia or from the effects of radiation on the venous and lymphatic drainage. Now, by and large, we do not see complications of direct radiation to the genitalia because radiation for penile lesions is rarely undertaken. The secondary effects of radiation, as seen in the scrotum, are usually manifested by chronic lymphedema, chronic lymphangiectasia, and, in some cases, chronic recurring cellulitis. To this list must be added the patient who has required significant debridement because of rapidly progressive, multiorganism fasciitis (Fournier's gangrene). Fournier's gangrene is often seen with processes of the rectum, such as missed perianal abscess, or with processes involving the urethra, such as periurethral abscess, many times associated with other comorbidities such as diabetes mellitus. Many of these patients present late, and excision, while lifesaving, usually leaves a significant defect. If the fasciitis is associated with pathophysiology of the urethra and/or anus, then these situations must be managed and resolved before reconstruction.5

DIAGNOSIS
A contusion of the scrotum, i.e., scrotal hematoma, can be confused with a fracture of the testicle. The latter injury implies disruption of the tunica albuginea of the testicle. With a scrotal contusion, the hematoma usually is noted to be posterior and lateral to the testicle ( Fig. 65-6), whereas with fracture of the testicle, the parenchyma of the testis is not normal and is often associated with a hematocele ( Fig. 65-7). In the presence of penetrating trauma to the genitalia, a retrograde urethrogram is always indicated because of the close proximity of the urethra. In all cases of genital avulsion, other than a simple scrotal avulsion, a complete evaluation of the urethra must be done, along with a rectal examination and possibly flexible sigmoidoscopy.

FIG. 65-6. B-scan ultrasound demonstrating the classic distribution of a scrotal contusion (scrotal wall hematoma). Note the testis displaced anteriorly, with the hematoma contained in the multiple fascial layers of the scrotum.

FIG. 65-7. B-scan ultrasound of a patient with a fracture of the testicle. Notice the disrupted parenchymal pattern of the testicle, the hematocele, and the demonstration of extruded seminiferous tubules within the hematocele.

INDICATIONS FOR SURGERY
Exploration of the scrotum would be indicated if there are indications of injury to the testicle (blunt or penetrating). Scrotal reconstruction would be required following complex lacerations or penetrating injury to the scrotum, avulsion injuries, excisions for chronic cellulitis, lymphangiectasia and lymphedema, and following significant burns to the scrotal skin.

ALTERNATIVE THERAPY
Contusions of the scrotum are usually treated with bed rest, analgesia, and scrotal elevation. Scrotal elevation can be accomplished very efficiently with a Bellevue-style bridge or with scrotal support.

SURGICAL TECHNIQUE
Scrotal Hematomas and Blunt Testicular Injury There is little to be accomplished in exploring the scrotal hematoma secondary to external blunt trauma, as the hematoma is usually disseminated throughout the highly elastic layers of the scrotum and does not form a drainable hematoma per se. On the other hand, if there is a testicular injury, exploration and repair of the testicle is indicated ( Fig. 65-8). Exploration consists of exposure of the testicle. The space of the tunica vaginalis is opened, the hematocele is drained, seminiferous tubules are debrided, and the tunica albuginea is closed. In the past PGA suture was used; however, now we use polydioxanone suture, although Monocryl is also an option. The parietal tunica vaginalis is left open and is reflected as would have been done for a hydrocele. The scrotum is drained and closed with 3-0 PGA suture, chromic suture, or Monocryl.

FIG. 65-8. (A) Patient with fractured testicle; the skin has been opened, as has been the vaginal space. Notice the draining hematocele fluid with clots affixed to the seminiferous tubules. (B) The testicle has been delivered. Note the clot attached to the extruded tubules. (C) Appearance of the testicle following repair.

Penetrating Injuries to the Genitalia Penetrating injuries have been subclassified as either simple or complicated. By and large, complicated injuries imply urethral involvement, amputation, or near-total amputation, and, with regards to the scrotum, imply amputation or injury of the testicle, with or without amputation of the overlying scrotal skin ( Fig. 65-9). Simple lacerations are managed with primary closure and drainage if indicated. In the case of amputation of the testicles, testicular microreplantation has been performed including vasovasostomy, along with reapproximation of the vasculature of the spermatic cord ( Fig. 65-10). The testicle is placed, as quickly as possible, in a sterile bag in saline-soaked gauze, and that bag is placed in a second bag filled with saline slush for cold preservation. Unlike amputation of the penis, where successful replantation has been done as long as 18 hours from the injury, the testicle must be replanted by 6 to 8 hours because of the very high metabolic rate of testicular tissue.

FIG. 65-9. (A) Young patient with complex penetrating trauma to the thigh and genitalia. Retrograde urethrogram, rectal examination, and flexible sigmoidoscopy were normal. The patient's scrotum was explored. (B) The patient's right testicle as delivered; note the intact seminiferous tubule pattern with virtual complete disruption of the tunica albuginea. (C) Appearance of the same testicle as in B following reconstruction of the tunica albuginea with primary closure. (D) Appearance of the genitalia with the gunshot wounds debrided and the right hemiscrotum drained.

FIG. 65-10. (A) Appearance of a patient following bilateral testicular amputation and scrotal amputation. The patient presented with only his right testicle. The left testicle could not be found at the trauma scene. (B) The right testicle is reanastomosed to the left (longer) spermatic cord. (C) Notice the appearance of the debrided spermatic cord and the debrided distal spermatic cord going to the testicle. Vasovasostomy was performed with a two-layer microscopic technique. Microscopic coaptation of the spermatic artery and multiple spermatic veins was performed. (D) Appearance of the replanted testicle before closure of the scrotum.

There can be some difficulty in identifying the vessels in the spermatic cord, though they are somewhat compartmentalized. Identifying the artery proximally is not difficult, and identification of the distal artery in the severed organ can be aided by examining the relationship of the compartments to the vas deferens. Coaptation of the artery and a number of veins is optimal using 9-0 or 10-0 Proline, depending on the size of the respective vessels. Vasovasostomy can be done using 9-0 or 10-0 Proline or nylon suture either with a classic microscopic two-layer technique or a single-layer “tricorner” technique, depending on the preference of the surgeon. If possible, the testicle should not be covered with a graft but either placed in a thigh pouch and later liberated or, if there is some remaining redundancy of the scrotum, covered primarily with reapproximation of the remaining scrotal tissues. Obviously, if the patient arrives without his amputated testicle, then hemostasis must be assured. Usually the vessels are in spasm, but they clearly can come out of spasm later. The wound should be irrigated and, if contaminated, packed to be closed by secondary intention. If the wound is clean, then primary closure or primary grafting can be performed. Avulsion Injuries Small scrotal avulsions are managed as simple lacerations with either primary or delayed closure and drainage, as would be indicated for any laceration ( Fig. 65-11). For larger injuries, the emergency management consists of allowing the injury to completely declare ( Fig. 65-12). The area of the avulsed scrotum should be managed with cold saline packs and observed over 12 to 24 hours. Clear demarcation will occur and allow the surgeon to debride only the tissue that is nonviable. Debridement with closure is then performed.

FIG. 65-11. (A) Appearance of a left scrotal avulsion injury. Patient was injured in a motorcycle accident in which the patient's trousers were ripped off as he departed the motorcycle. (B) Appearance after closure. Primary closure is performed with drainage.

FIG. 65-12. (A) Large avulsion injury of the genitalia. Patient was injured when his clothing was ensnared in the power takeoff mechanism of a tractor. Notice the exposed shaft of the penis and the exposed testicles bilaterally. (B) The appearance after reconstruction with a split-thickness skin graft. The patient was observed for 24 hours, allowing the wounds to demarcate. In this case, both the shaft of the penis and the scrotum were reconstructed with a sheet split-thickness skin graft.

If a primary closure cannot be accomplished with the remaining scrotal tissue, then the surgeon has several options. One option is to place the testicles in thigh pouches to be later liberated and replaced to the area of the scrotum. The preferable option is to perform a primary reconstruction of the scrotum using a mesh split-thickness skin graft. The graft should be harvested 0.016 to 0.018 inches thick and then meshed on a 1½-to-1-inch meshing template. The testicles must be fixed in position so that they do not migrate beneath the graft using permanent suture or absorbable suture that is slowly absorbed. The meshing of the graft allows for escape of serum and blood products from beneath the graft but also allows the graft to configure to the complex contours of the underlying testicles. The vaginal space is left open, and the parietal tunica vaginalis is reflected to fix the testicle in place. The graft is then applied immediately to the testicles, suturing it to the surrounding skin. Unless the wound is markedly contaminated, cases so managed have been very successful. The grafts should be bolstered using Xeroform gauze applied directly to the graft with dacron batting soaked in saline and mineral oil placed over the fine meshed gauze. The bolster is held in place with 2-0 chromic sutures. Unless associated with a urethral injury, the patient can be “diverted” with a soft Foley catheter. In patients in whom the avulsion injury extends near to the anus, colostomy may be required. It must be emphasized that local skin flaps are not recommended for primary closure in these cases. Should the testicles be avulsed, replantation is not an option. During the avulsion injury, the vasculature is stretched before giving way to the force, and the endothelial damage can be significant and unpredictable. In patients in whom the avulsion injury is tantamount to emasculation, these injuries are often associated with significant injuries to the adjacent tissue. Reconstruction

assumes a very secondary position, as these patients require life-saving steps. The vast majority of these patients will require colostomy, suprapubic cystostomy, multiple dressing changes over the posttrauma course, and often present with significant bleeding. Genital Burns The emergency therapy of genital burns is similar to that for any burn. The scrotum can be dressed open with topical antibiotic ointments such as Silvadene, or a closed antibiotic dressing regimen can be used. The integrity of the urethra must be determined when the patient presents. A Foley catheter can be placed in the patient who has burns to his genitalia. If there is evidence of urethral burn, then most would suggest diversion with suprapubic cystostomy. If the burns to the genital and perineal area are extensive, then an occasional patient will require colostomy. If the subscrotal urethra is involved in the burn, no attempt at initial reconstruction should be made. The genital tissues are remarkably vascular; debridement of the genitalia, in general, should be accomplished carefully. Aggressive debridement should be avoided, as many of the tissues will recover and are nonreproducible. Whirlpools and tank soaks are very useful for gentle debridement and cleansing and may be done early on, two to three times per day. Chemical burns rarely involve the structures deep to the skin and are managed with copious irrigation and then as with a thermal burn. If a patient has evidence of an electrical burn to the area of the genitalia, the patient must be observed for 12 to 24 hours and then explored. In electrical burns, the initial management is aimed at debridement, and reconstruction can be offered later as the situation dictates ( Fig. 65-13).

FIG. 65-13. (A) The appearance of the genitalia in a patient who was burned in a steam-line accident. Notice the burns to the glans, the dorsum of the shaft of the penis, and the right hemiscrotum. (B) In this particular case, reconstruction of the glans was accomplished with a small split-thickness skin graft. Reconstruction of the shaft of the penis was accomplished with a penile skin island flap; the patient was uncircumcised, and the ventral skin was mobilized to the dorsum. The scrotal burn was completely excised, and primary reconstruction of the right hemiscrotum was accomplished. This particular patient demonstrates all of the possibilities for reconstruction following burn debridement.

Radiation Injuries Just as there can be lymphedema of the scrotum in men, there can be lymphedema in the vulva of women. Although it is not the topic of this chapter, some mention should be made, in that the management of lymphedema of the vulva and secondary reconstruction is different from the management of lymphedema of the scrotum with reconstruction. In women, the most common cause of lymphedema is idiopathic. The treatment for lymphedema of the labia is vulvectomy. If the vulvectomy involves excision of the vaginal tissues, then the lateral hair-bearing tissue should not be sutured into the vagina. Instead, reconstruction of the labia can be accomplished using either the rectus abdominis flap, gracilis flaps, or posterior thigh flaps. One should select the flap so that the flap drainage is away from the lymphatic distribution affected by the radiation. In the case of the man with scrotal lymphedema and recurrent cellulitis, all layers of the scrotum should be excised, down to the level of the external spermatic fascia (Fig. 65-14). It is not uncommon for the patient also to have large hydroceles. 2

FIG. 65-14. Appearance of a young man with chronic genital lymphedema following irradiation therapy for Hodgkin's lymphoma. (A) Appearance of the massively lymphedematous scrotum. (B) B-scan ultrasound demonstrating the large hydrocele. Notice the normal testis posteriorly. (C) Appearance of the genitalia after debridement of the lymphedematous tissue. Note orchidopexy has been performed. (D) Immediate appearance following reconstruction of the shaft of the penis with a split-thickness skin graft and reconstruction of the scrotum with a meshed split-thickness skin graft. (E) Appearance of that same patient 6 months postoperatively. Notice the small amount of residual lymphedema of the preputial cuff. Further notice the redundant appearance of the scrotal graft.

The vaginal space should be opened, and the parietal tunica vaginalis then reflected and incorporated in the orchidopexy. Reconstruction can be accomplished, using split-thickness skin grafts, as already discussed, and on the scrotum these grafts can be meshed as already discussed earlier. The meshing allows for the graft to better comply to the contours of the testicle and to allow the rather significant serous drainage noted in these patients to escape from beneath the graft. Additionally, the meshing seems to improve the cosmetic result by giving a pseudorugated appearance to the tissue. Neither full-thickness skin grafts nor local skin flaps should be used. The cosmetic results achieved in these patients so reconstructed are excellent. Fournier's Gangrene If the primary process has resulted in extensive debridement of the scrotum, the testicles can be placed in thigh pouches with the intention of later replacing the testicles in their normal anatomic area and for scrotal reconstruction ( Fig. 65-15). The lateral defects on the thigh can often be closed per primam or can be grafted.

FIG. 65-15. Technique after McDougal for liberation of testicles that have been placed in thigh pouches. Note that the testicles are mobilized with random thigh skin flaps and transposed to the midline to reconstruct the scrotum. The lateral defects can be closed per primam or can be grafted.

For most patients, however, the scrotal excisions do not necessitate placing the testicles in thigh pouches, but instead they can be dressed in the wound. These patients often require multiple dressing changes and wet-to-dry debridements. When reconstruction is undertaken, the testicles must be fixed in their normal anatomic position. It is my custom to open the parietal tunica vaginalis and reflect it, and, as already described, graft techniques can be utilized to restore the scrotum. Although local flaps can be used, the cosmetic results achieved with local flaps are usually less than optimal when compared to reconstruction techniques utilizing split-thickness skin grafts. Interestingly, hese mesh split-thickness skin grafts not only remain supple but, in many cases (as has been shown), will actually become redundant.

OUTCOMES
Complications Complications of scrotal trauma and reconstruction are primarily related to inadequate appreciation of the degree of injury with subsequent necrosis of additional skin. It is important for the surgeon to adequately debride the devitalized skin, which may require more than one trip to the operating room. The failure of a skin graft to take is usually a result of technical errors such as accumulation of fluid under the graft, inadequately prepared graft bed, or continual slippage of the graft over the bed, which does not allow capillary ingrowth. Results Generally, the scrotal skin is highly distensible, and even a small fragment can be expanded to cover a large defect with a good anatomic result. There has been concern regarding the effect of implantation of the testes in thigh pockets on spermatogenesis, though there are little clinical data to support this concern. Furthermore, studies utilizing thermocouples to monitor testicular temperature in testicles implanted in thigh pockets have shown that there is not a significant difference from the temperature of testicles in the scrotum. CHAPTER REFERENCES
1. 2. 3. 4. 5. Arneri V. Reconstruction of the male genitalia. In: Converse J, ed. Reconstructive plastic surgery, 2nd ed. Philadelphia: WB Saunders. 1977;7:3902–3921. Charles RH. The surgical technique and operative treatment of elephantiasis of the generative organs based on a series of 140 consecutive successful cases. Ind Med Gaz 1901;36:84. Jordan GH, Gilbert DA. Male genital trauma. Clin Plast Surg 1988;15:431. Jordan GH, Gilbert DA. Management of amputation injuries of the male genitalia. Urol Clin North Am 1989;16(2):359–367. Jordan GH, Schlossberg SM, Devine CJ. Surgery of the penis and urethra. In: Walsh PC, et al, eds. Campbell's urology, 7th ed. Philadelphia: WB Saunders, vol. 3, 1997;3316–3394.

Chapter 66 Penectomy for Invasive Squamous Cell Carcinoma of the Penis Glenn’s Urologic Surgery

Chapter 66 Penectomy for Invasive Squamous Cell Carcinoma of the Penis
James L. Mohler and John A. Freeman

J. L. Mohler and J. A. Freeman: Division of Urology, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599-7235.

Diagnosis Indications for Surgery Alternative Therapies Surgical Technique Partial Penectomy Total Penectomy Outcomes Complications Results Chapter References

Penile carcinoma occurs rarely in the United States. It is most commonly diagnosed in the sixth and seventh decades of life but can occur at any age, including childhood. Throughout the world penile carcinoma occurs more frequently in societies where hygiene is poor or where circumcision is not routinely practiced. Indirect evidence supports a carcinogenic effect of smegma, whose accumulation is favored by poor hygiene and phimotic foreskin. The duration of exposure to smegma may explain the decreasing preventive effect of neonatal versus pubertal versus adult circumcision. Controversy continues to surround the issue of neonatal circumcision in the United States, where the American Academy of Pediatrics Task Force on Circumcision maintains that newborn circumcision has potential medical benefits and advantages as well as disadvantages and risks. 12 More recently, penile carcinoma has been associated with sexually transmitted human papilloma virus type 16 4; approximately two-thirds of untreated carcinomas were positive for human papillomavirus DNA, which was also found in lymph node metastases of penile carcinoma.

DIAGNOSIS
Penile carcinoma most frequently occurs on the glans, the coronal sulcus, or the prepuce. It is less commonly found on the shaft or at the meatus. It has a variable appearance ranging from an area of induration or erythema to a wart-like growth to a shallow or deeply excavating ulcer to a large exophytic lesion. The lesions are rarely painful and, when large, can become necrotic and associated with penile discharge. Fixation of the foreskin may prevent direct examination. Delayed presentation occurs frequently; the mean delay to diagnosis is 8 months, and 15% of patients postpone medical evaluation for 1 year. 7 Metastases to the superficial and bilateral inguinal nodes may occur when penile carcinoma is confined to the glans and penile skin. Involvement of the corporal bodies may produce metastases to the inguinal and iliac lymph nodes. Careful examination is required to assess the extent of disease locally as well as to search for pelvic masses or inguinal adenopathy. Biopsy of the primary is required for histologic diagnosis as well as to determine the depth of invasion. Metastases are infrequent unless the primary tumor involves the corpora cavernosa or exceeds grade 1.14 Patients often exhibit hypercalcemia, which may resolve with surgical treatment. Staging evaluation should include CT or MRI of the pelvis and groin to search for evidence of adenopathy. Lymph nodes, visible by imaging studies or palpable on physical exam, should be aspirated following treatment of the primary and 3 weeks of antibiotic therapy.

INDICATIONS FOR SURGERY
Small, well-differentiated primary tumors may be treated by partial penectomy and monthly follow-up of at least 1 year. The inguinal lymph nodes may be safely observed if clinically negative and the primary tumor is low grade, stage T 1 or less, and without evidence of microvascular invasion. However, patients who have large or moderately to poorly differentiated primary tumors should undergo partial or total penectomy and immediate inguinal lymphadenectomy. 1

ALTERNATIVE THERAPIES
Mohs' microscopically guided surgery has successfully eradicated the primary in 92% of carefully selected patients with penile carcinoma. Following local anesthesia, a fixative paste of 40% zinc chloride chemically fixes tissues, which are excised at intervals of 4 to 24 hours. Each section is examined microscopically completely, and resection is continued until all areas of microscopically identified cancer are removed. Healing by secondary intention is rapid and uneventful. 8 Mohs' surgery is most ideally suited for small, distally located carcinomas, especially those that are less than 1 cm in diameter. Laser surgery using CO 2 or Nd:YAG lasers has been used successfully for treatment of T 1s tumors.6 Both CO2 and Nd:YAG lasers have been used as an adjunct to excision of T 1 and T2 tumors with excellent results. 5 Both external beam and interstitial brachytherapy 2,10 have provided local control in up to 85% of patients, with penile preservation in up to 80% of patients, although late side effects, including meatal stenosis, urethral fistula, penile curvature, and impotence, occur in a majority of patients. 2 In addition, distinguishing local recurrence from postradiation ulcer scar and fibrosis presents a diagnostic problem often necessitating repeated biopsy.

SURGICAL TECHNIQUE
Tumors involving the foreskin may be treated by circumcision alone, and lesions involving the glans or shaft may be treated by local wedge excision; however, local recurrence may occur in up to 50% of patients. 9 Standard therapy requires excision of 2 cm of penile shaft proximal to the extent of the tumor by palpation with microscopic confirmation of a tumor-free margin of resection. The reason for the 2-cm margin is that penile carcinoma, as well as other cancers of the skins, often sends out tentacles of cancer cells for large distances beyond the clinically visible or palpable margins of the neoplasm. For lesions involving the glans and distal shaft, partial penectomy should leave a residual penile stump suitable for upright micturition and sexual function. Involvement of the proximal shaft requires total penectomy. Extension to the perineal body, pubis, or scrotum in the absence of metastatic disease may require more extensive resection and adjuvant chemotherapy. Partial Penectomy The technique of partial penectomy (Fig. 66-1) consists of amputation with a 2-cm skin margin followed by corporal closure, skin closure, and creation of a neomeatus. Under local, regional, or general anesthesia, the patient is prepped and draped in the standard sterile fashion. A 14-Fr Foley catheter is placed for drainage, and a Penrose drain is used as a tourniquet at the base of the penis to control corporal bleeding. The primary tumor may be covered with a rubber glove to reduce the risk of wound seeding.

FIG. 66-1. Partial penectomy. (A) Circumferential incision made 2 cm proximal to the tumor margin. (B) A soft rubber drain serves as a tourniquet, and the corpora cavernosa are amputated 0.5 to 1.0 cm proximal to the original skin incision. Additional urethral length allows spatulation. (C and D) The corpora cavernosa are closed. (E) The dorsal shaft skin is closed longitudinally. (F) The spatulated urethra is approximated to the shaft skin.

A circumferential skin incision is made 2 cm proximal to the tumor margin. The corpora cavernosa are amputated 1 cm proximal to the original line of the skin incision in order to facilitate tension-free skin closure. The urethra is transected 1 cm distal to the point of amputation of the corpora cavernosa and, hence, at the same level as the original skin incision. The corpora cavernosa are closed horizontally with a running suture of 2-0 polyglycolic acid. The tourniquet is released, and additional sutures are placed, if necessary, to provide corporal hemostasis. The penile skin is closed longitudinally from dorsum to ventrum with a running suture of 3-0 or 4-0 polyglycolic acid until the site of the new external urethral meatus is reached. The urethra is spatulated posteriorly to allow inclusion of a broad-based V-shaped wedge of ventral penile skin to prevent meatal stenosis. Simple interrupted sutures of 4-0 polyglycolic acid are used to approximate skin and urethra. Subcutaneous drains are not used. A gentle pressure dressing is applied overnight with the penis and catheter affixed in the vertical position. The Foley catheter is removed 3 to 5 days postoperatively after complete healing of the neomeatus. Preoperative antibiotics should be administered if the penile carcinoma is frankly infected or if the patient is at risk for systemic infection (cardiac valvular disease or indwelling prosthetic equipment). Oral antibiotics, usually a cephalosporin, may be continued postoperatively, although in the absence of significant risk of or damage from infection, they are probably not necessary. Total Penectomy The patient is placed in the modified lithotomy position, and the operative field is prepared in standard fashion. A 14-Fr Foley catheter is placed, and the primary tumor is covered with a rubber glove ( Fig. 66-2). A diamond-shaped or elliptical incision is made around the base of the penis and extended through the subcutaneous tissues until the surface of the pubis is reached. Vascular structures are ligated to maintain hemostasis and control lymphatics. The suspensory ligament of the penis is isolated with a right-angle clamp and sharply divided. The underlying deep dorsal vein and penile arteries are isolated, ligated with absorbable suture, and divided. The penis is deflected cephalad, and the urethra is isolated ventrally from the proximal corpora cavernosa using blunt and sharp dissection. The corpora cavernosa are amputated in their crural portions near the sites of their bony insertions, and the stumps are closed with running horizontal mattress sutures of 2-0 polyglycolic acid. Sufficient urethral length must be achieved to allow a tension-free perineal urethrostomy. However, a 2-cm urethral margin relative to the primary tumor must be maintained, especially if the tumor is large, because invasion of the corpora spongiosum may occur.

FIG. 66-2. Total penectomy. (A and B) A diamond-shaped incision is made around the base of the penis. (C) The dissection is extended superiorly through the subcutaneous fat, lymphatics, and superficial veins to the surface of the pubis. The suspensory ligament of the penis is divided, and the vessels are transected between ligatures. (D) Cephalad deflection of the penis allows ventral isolation of the urethra. Proximal crural amputation and division of the urethra are accomplished. The corpora are closed. (E and F) Adequate bulbar urethral length is achieved, and a perineal urethrostomy is created by spatulating the urethra and using a V inlay of skin to prevent stenosis. The diamond-shaped defect is closed.

Immediately ventral to the perineal body, a 2-cm circle of skin is outlined, and sufficient skin is excised to leave behind a 2-cm defect with a broad-based V of skin. The Foley catheter is brought out through the urethrostomy site. The urethra is spatulated for 1 to 2 cm and approximated to the perineal skin with simple, interrupted sutures of 4-0 polyglycolic acid. Inclusion of a V of skin into the spatulated anastomosis should reduce the likelihood of stenosis. The penectomy skin defect is closed with a running suture, and a gentle pressure dressing and scrotal support are applied overnight. The Foley catheter is removed when the urethrostomy is well healed, usually after 3 to 5 days. Preoperative antibiotics are indicated if there is frank evidence of infection or the patient is at risk for seeding of abnormal heart valves or prosthetic material. A cephalosporin may be continued postoperatively if the patient is at increased risk of or danger from infection. The perineal urethrostomy should be observed carefully for evidence of stenosis, which, if it occurs, may require dilation or revision. Total penectomy may be psychologically devastating, especially to the younger patient. Preoperative psychiatric evaluation and postoperative psychiatric support is recommended. If the patient remains without evidence of recurrent disease for a period of approximately 2 years, penile reconstruction may be entertained. 3,11

OUTCOMES
Complications The complications of both partial and total penectomy are psychological and physical. The fear of loss of body image is a major factor in the delay that many patients have before seeing a physician and should not be underestimated. It is very important that the patient receive counseling and psychological support. The physical complications include necrosis of the flaps, which can be avoided by using broad-based, well-vascularized tissue. Stenosis of the urethral opening has been reported to be as high as 20% and can be avoided by using a V-inlay technique with the normal skin. Should meatal stenosis occur, it is important to be sure that the stenosis is not caused by recurrent carcinoma. Results Conventional treatment using partial penectomy for T 1–2 disease produced tumor-specific survival of 77%. Partial or total penectomy, alone or in combination with radiation to inguinal nodes after penectomy, produced 3-year or longer survival in only two of nine patients, whereas treatment by early extended excision of both primary lesion and the inguinal lymph nodes produced 3-year or longer survival in 11 of 13 patients. Patient survivorship is dependent on nodal status; patients without nodal metastases have a 5-year survival of 77% in contrast to patients with untreated nodal metastases, 95% of whom are dead at 3 years.1,7 Death from carcinoma of the penis is almost universally a result of metastatic disease, and almost all patients who die will first relapse in the inguinal lymph nodes. Bulky adenopathy is generally associated with a poor prognosis; however, clinical experience suggests that 5-year survival is surprisingly common (12% to 80%) among patients in whom clinically negative inguinal lymph nodes are resected and found positive on histologic examination of the adenectomy specimen.
13

Hence, treatment principles can be summarized as follows: the primary neoplasm should be treated with the intent of cure, but with attention to cosmesis and function. The decision to perform a concomitant lymphadenectomy is based on the extent and grade of the cancer. Finally, addition of iliac lymphadenectomy in the presence of inguinal lymph node metastases remains controversial but undoubtedly increases the morbidity of lymphadenectomy. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Fraley EE, Zhang G, Sazama R, Lang PH. Cancer of the penis. Prognosis and treatment plans. Cancer 1985;55:1618–1624. Gerbaulet A, Lambin P. Radiation therapy of cancer of the penis. Indications, advantages, and pitfalls. Urol Clin North Am 1992;19:325–332. Hanash KA, Tur JJ. One-stage plastic reconstruction of a totally amputated cancerous penis using a unilateral myocutaneous gracilis flap. J Surg Oncol 1986;33:250–253. Iwasawa A, Kumamoto Y, Fujinaga K. Detection of human papillomavirus deoxyribonucleic acid in penile carcinoma by polymerase chain reaction and in situ hybridization. J Urol 1993;149:59–63. Kriegmair M, Rothenberger KH, Spitzenpfeil E, Schmeller NT, Hofstetter AG. Neodymium-YAG-laser treatment for carcinoma of the penis (Abstract). J Urol 1990;143:351A. Malloy TR, Wein AJ, Carpiniello VL. Carcinoma of the penis treated with neodymium YAG laser. Urology 1988;31:26–29. McDougal WS, Kirchner FK, Edwards RH, Killion LT. Treatment of carcinoma of the penis. The case of primary lymphadenectomy. J Urol 1986;136:38–41. Mohs FE, Snow SN, Messing EM, Kuglitsch ME. Microscopically controlled surgery in the treatment of carcinoma of the penis. J Urol 1985;133:961–966. Narayana AS, Olney LE, Loening SA, Weimar GW, Culp DA. Carcinoma of the penis: Analysis of 219 cases. Cancer 1982;49:2185–2191. Rozan R, Albuisson E, Giraud B, et al. Interstitial brachytherapy for penile carcinoma: a multicentric survey (259 patients). Radiother Oncol 1995;36:83–93. Santi P, Berrino P, Canavese G, Galli A, Rainero ML, Badellino F. Immediate reconstruction of the penis using an inferiorly based rectus abdominis myocutaneous flap. Plast Reconstr Surg 1988;81:961–964. Schoen FJ, Anderson G, Bohon C, Hinman F Jr, Poland RL, Wakeman EM. Task force on circumcision. Report of the Task Force on Circumcision. Pediatrics 1989;84:388–391. Shellhammer PF, Jordan GH, Schlossberg SM. Tumors of the penis. In: Walsh PC, Retik AB, Stamey TA, Vaughan ED Jr, eds. Campbell's urology, 6th ed. Philadelphia: WB Saunders, 1992;1275. Solsona E, Iborra I, Ricos JV, et al. Corpus cavernosum invasion and tumor grade in the prediction of lymph node condition in penile carcinoma. Eur Urol 1992;22:115–118.

Chapter 67 Inguinal Lymphadenectomy for Penile Carcinoma Glenn’s Urologic Surgery

Chapter 67 Inguinal Lymphadenectomy for Penile Carcinoma
John A. Freeman and James L. Mohler

J. A. Freeman and J. H. Mohler: Division of Urology, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599-7235.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Anatomy Radical Lymphadenectomy Modified Inguinal Lymph Node Dissection Outcomes Complications Results Chapter References

The role of lymphadenectomy in the management of patients with penile carcinoma is controversial. Traditionally, the clinician has balanced the therapeutic benefit of lymphadenectomy against the significant morbidity of the procedure. Most patients with tumors that did not invade the corpora cavernosa or spongiosum were observed if the lymphatics were clinically uninvolved. Patients with clinically enlarged inguinal lymph nodes were treated with a 6-week course of antibiotics and then recommended for lymphadenectomy if the nodes remained enlarged. 10 Emerging evidence suggests that treatment of the tumor when it is micrometastatic and subclinical improves survival over delaying treatment until the time metastases become palpable. Defining which patients are at high risk of harboring micrometastases (i.e., those who would ultimately develop clinical lymph node metastases and require delayed lymphadenectomy) has become clearer.

DIAGNOSIS
The diagnosis and staging of penile cancer are covered in Chapter 66.

INDICATIONS FOR SURGERY
Patients with moderately or poorly differentiated tumors or invasion of the corporal tissues should be considered at high risk for subsequent relapse and considered for prophylactic dissection because of the potential survival benefit and lack of effective alternative therapy if lymphadenectomy does not cure the disease. Patients with clinically involved lymph nodes should undergo the classic radical lymph node dissection. In these patients, beginning the procedure with a midline incision to stage the pelvic lymph nodes (as in prostate cancer) is prudent because patients with pelvic metastases proximal to the iliac bifurcation are essentially incurable. If tumor is found in this location, inguinal lymphadenectomy is not performed except for palliation. A less morbid modified lymph node dissection described by Catalona is preferable for patients undergoing prophylactic lymphadenectomy for presumed occult micrometastatic disease. Patients with histologically confirmed nodal metastases in the groin following the modified dissection should be considered for pelvic lymph node dissection also, depending on the volume of metastases and clinical judgment. Surgical cure is rare, but possible, with small-volume pelvic metastases.

1

ALTERNATIVE THERAPY
An alternative to lymphadenectomy is a watch-and-wait approach, which has been the recommended course of treatment for invasive carcinoma of the penis until recently because equal survival was believed possible if surgery was delayed until the lymph nodes became clinically enlarged during the period of observation. 6 Unfortunately, 20% of patients with clinically normal lymph nodes can be found to have pathologic micrometastases, whereas only two-thirds with clinically enlarged lymph nodes harbor metastases. Furthermore, the survival of patients is adversely affected in the conservative approach, as noted above. Therefore, observation is recommended only in those patients with a true low risk of progression. Radiation and chemotherapy have been used in patients with metastatic penile carcinoma 3 but are inferior therapy to surgery in resectable disease. In most cases where these therapies have been used, adenopathy has been advanced, and results poor. Evaluation of efficacy is difficult because of small series, lack of randomization, and poor follow-up in patients with penile carcinoma. Agents showing objective response include methotrexate, bleomycin, cisplatin, 5-fluorouracil, adriamycin, vincristine, and vinblastine. Although anecdotal cases of complete response and long-term survival are noted, the results from larger series are bleak, with complete and partial responses of 6% each reported. 7 Neoadjuvant chemotherapy, with or without radiation, has been used for metastatic disease and in patients with fixed or locally invasive adenopathy. 11,12 Although this group has poor overall survival, long-term cure has occasionally been achieved.

SURGICAL TECHNIQUE
Anatomy Anatomic considerations are important if the inguinal lymph node dissection is to be adequately performed with minimal morbidity. Daseler's classic description of the groin anatomy, based on 450 dissections, is the foundation of anatomic considerations in inguinal lymphadenectomy. 4 The inguinal lymph nodes are divided into superficial and deep nodes, separated by the fascia lata of the thigh. The superficial lymph nodes reside in the subcutaneous fat, in the membranous portion of Camper's fascia, which is indistinguishable from Scarpa's fascia in the adult. The deep inguinal lymph nodes are located medial to the femoral artery below the fascia lata, around the femoral vein. The femoral sheath, deep to the fascia lata, must be opened anteriorly to skeletonize the femoral vessels and remove the deep inguinal lymph nodes. The proximal lymphatic, the lymph node of Cloquet, lies on the anterior surface of Cooper's ligament beneath the shelving edge of the inguinal ligament. Both the radical and modified lymph node dissections are designed to remove these two sets of lymphatics, but to a different extent. The blood supply of the skin of the thigh is supplied by branches of the femoral artery, which are necessarily ligated in the performance of the dissection. These arteries, the superficial epigastric, the superficial external pudendal, and the superficial circumflex iliac, all have corresponding veins that drain into the greater saphenous vein as it joins the femoral vein at the fossa ovalis, an opening in the fascia lata ( Fig. 67-1). Because the arteries are ligated, skin flap viability depends on anastomosing vessels within the superficial globular fat of Camper's fascia that course from lateral to medial along the skin lines, parallel to the inguinal ligament. A transverse skin incision disrupts these pathways the least.

FIG. 67-1. Anatomic considerations. (A) Superficial vaginal lymph nodes and saphenous vein and its tributaries at the fossa ovalis. (B) Deep inguinal lymph nodes and their relation to the femoral vessels and iliac lymph nodes.

Camper's fascia in the thigh is continuous with Camper's fascia of the abdominal wall. Scarpa's fascia from the abdomen becomes confluent with the superficial muscular fascia of the thigh, the fascia lata. It is critically important for the correct performance of the inguinal lymph node dissection to distinguish between the globular fat of Camper's fascia, where the superficial anastomosing blood vessels responsible for viable skin flaps reside, and the membranous fat layer that contains the lymphatics. Dissection must occur exactly between these two layers to raise viable skin flaps and still excise all potentially involved lymphatics. The deep margin of the dissection should not extend beneath the iliacus fascia because the femoral nerve courses there and may be injured. No significant lymphatics are in this region. Radical Lymphadenectomy For patients with clinically enlarged lymph nodes, a 6-week course of oral antibiotics (cephalexin) is recommended before surgery to determine if the adenopathy is inflammatory or metastatic. If adenopathy persists following the course of antibiotics, radical lymphadenectomy is recommended. The patient is adequately hydrated preoperatively, and the legs are placed in antiembolism stockings. The ipsilateral hip is abducted, and the ipsilateral foot is secured to the contralateral knee, effectively frog-legging the patient for dissection ( Fig. 67-2). Daseler's 4 boundaries of dissection are defined superiorly by a line drawn from the anterior superior iliac crest to the external inguinal ring, laterally by a line drawn inferiorly from the anterior superior iliac spine for 20 cm, medially by a line drawn inferiorly from the pubic tubercle for 15 cm, and inferiorly by a horizontal line joining the medial and lateral boundaries. An incision is made 4 to 5 cm inferior and parallel to the inguinal crease over the medial thigh from the lateral to medial limits of the dissection. The palpable femoral vessels should be in the medial aspect of the incision.

FIG. 67-2. Patient position and skin incisions for bilateral pelvic lymphadenectomy and unilateral inguinal lymphadenectomy.

Skin flaps are developed superiorly and inferiorly as the first step of the dissection. Elevation of the skin edges with skin hooks allows dissection within the superficial fatty fascia of the thigh. At the junction of the superficial globular fat and the deeper membranous fat, the tissues are separated ( Fig. 67-3). The skin and globular fat are elevated off of the deep membranous fascia and Scarpa's fascia cephalad to a point 4 cm above the inguinal ligament. Gentle elevation of the skin flaps with Deaver retractors over moist sponges protects the flaps. Inferior traction on the lymphatic package with a small sponge under the left hand provides countertraction to facilitate dissection in the proper plane.

FIG. 67-3. Inguinal lymphadenectomy. Superficial skin flaps are developed superiorly and inferiorly to expose the saphenous vein with its investing fat, lymph nodes, and tributaries.

Dissection then is carried down through Scarpa's fascia onto the external oblique aponeurosis. The external inguinal ring and emerging spermatic cord are identified medially and retracted as dissection extends to the pubic tubercle. The fat and lymphatics are separated from the spermatic cord and base of the penis medially. Vascular and lymphatic channels are meticulously ligated to prevent fluid accumulation under the skin flaps. The fatty lymph packet is elevated off of the external oblique to the inferior border of the inguinal ligament (Poupart's ligament), where the femoral vessels are identified within the femoral sheath. The medial (adductor longus muscle) and lateral (sartorius muscle) borders of the dissection are identified next, and the fascia lata is incised over the muscles. The muscles are traced to their confluence at the apex of the femoral triangle, representing the inferior limit of dissection. The resulting triangular packet of lymphatics and fat needs only to be elevated off its deep margin to complete the dissection. The saphenous vein is identified medially and preserved if possible. Although this vessel is traditionally sacrificed during radical lymphadenectomy, no therapeutic purpose is achieved, and morbidity is increased. If massive lymphadenopathy exists, the saphenous should be removed ( Fig. 67-4). The femoral sheath is incised over the femoral artery and vein. Medial dissection isolates the lymph node of Cloquet between Cooper's ligament and Poupart's ligament, lateral to the lacunar ligament. The femoral sheath is stripped inferiorly to the apex of the femoral triangle, as the fascia overlying the sartorius and adductor longus is stripped distally. The deep lymph nodes are removed from their location between the femoral artery and vein. The superficial cutaneous perforating arteries and veins are ligated as they are encountered on the surface of the femoral artery and saphenous vein, leaving the intact saphenous to join the femoral vein in the area of the now absent fossa ovalis. It is important to limit the lateral aspect of dissection along the femoral sheath to the anterior surface of the femoral artery. Dissection laterally on the femoral sheath can injure the femoral nerve beneath the iliacus fascia (deep to the fascia lata), and posterolateral dissection can injure the profunda femoris artery.

FIG. 67-4. The tributaries to the saphenous vein are ligated. The saphenous is sacrificed if bulky adenopathy exists but is preserved otherwise.

When this dissection is complete, the superficial and deep lymphatics are removed together as an en bloc specimen. The wound is liberally irrigated. The exposed femoral vessels are covered by mobilizing the sartorius from its insertion on the anterior superior iliac spine, transposing it over the vessels, and securing its cephalad margin to the inguinal ligament with 2-0 PGA suture ( Fig. 67-5). Blood supply to the sartorius arises from its medial and inferior aspects, and care must be taken to protect these vessels during mobilization The medial edge of the muscle can be tacked to the adductor longus to guarantee coverage of the femoral vein. If pelvic lymphadenectomy has been performed simultaneously, this dissection should join with the distal limit of the pelvic dissection and allow free communication with the pelvis. To prevent herniation, Cooper's ligament should be secured to the shelving edge of Poupart's ligament with permanent suture (2-0 Proline) without compromising the lumen of the femoral vein.

FIG. 67-5. After the specimen has been removed, the femoral vessels are covered by transferring the proximal insertion of the sartorius muscle to the medial aspect of the inguinal ligament. Fixation of the subcutaneous tissue to bare muscle achieves dead space obliteration before skin closure.

An ampule of fluorescein is given intravenously, and the edges of the skin flaps are inspected with a Wood's lamp for viability. Nonviable edges are trimmed to prevent slough. The skin flaps are tacked to the surface of the exposed muscles with 3-0 PGA suture to prevent seroma formation. A closed suction drain is brought out through the inferior and superior flap, the subcutaneous tissue reapproximated with 3-0 PGA, and the wound closed with staples. A light pressure dressing is applied for the first 12 hours. Care is taken not to apply excessive pressure, which might further compromise blood supply in the skin flaps. Parenteral antibiotics are continued for 48 hours and then switched to oral agents. The patient is left at bed-rest with deep venous thrombosis prophylaxis by subcutaneous heparin for 5 days, with the legs in thromboembolic stockings. The drain is removed when the patient is ambulatory and drainage is less than 30 cc/day. Modified Inguinal Lymph Node Dissection The modified inguinal lymph node dissection is designed to decrease the morbidity associated with inguinal lymphadenectomy while maintaining therapeutic efficacy in patients with clinically negative lymph nodes. It is important to appreciate that this approach is not recommended for patients with clinically suspicious lymph nodes.2 As described, the modified procedure differs from radical lymphadenectomy in that the skin incision is smaller, the limits of the dissection are reduced, the saphenous vein is preserved, and sartorius transposition is not performed. The skin is incised transversely 2 cm below the groin crease for a distance of 10 cm. Skin flaps are raised in the same manner as described above, for a distance of approximately 8 cm cephalad and 6 cm caudally. Cephalad dissection onto the external oblique is performed as in radical lymphadenectomy, and the medial extent of dissection is identical. The lateral dissection is more limited, however. After opening the femoral sheath, dissection lateral to the femoral artery is not performed. The sartorius is not exposed, and dissection inferiorly on the fascial lata and femoral sheath extends only to the caudal edge of the fossa ovalis ( Fig. 67-6). The saphenous vein is preserved in the superficial nodal package, and the deep lymphatics below the fascia lata between the femoral vessels and medial to the femoral vein along the adductor longus fascia up to Cooper's ligament are removed. Postoperative management is similar to that for radical lymphadenectomy.

FIG. 67-6. Comparison of limits of dissection of modified inguinal lymphadenectomy ( dashed line) with classical groin dissection ( dotted line) as described by Daseler and associates. 4 Black, superficial inguinal nodes; white, deep inguinal nodes; gray, external iliac nodes. (From Catalona WJ. Modified inguinal lymphadenectomy for carcinoma of the penis with preservation of saphenous veins: technique and preliminary results. J Urol 1988;140:306–310.)

OUTCOMES
Complications The most dreaded complications of inguinal lymphadenectomy are flap necrosis and lower extremity lymphedema. Older series report flap necrosis rates of up to 30%13 and various wound complications in up to 72% of patients undergoing groin dissection. 9 These rates have been considerably diminished by transverse skin

incisions, development of flaps in the proper subcutaneous fat plane, and the use of pre- and perioperative antibiotics, closed suction drains, postoperative bed rest, anticoagulation, and lower extremity support stockings. McDougal et al. 9 reported 100% of their patients developed lymphedema, and 16% found it incapacitating. Even after modified lymphadenectomy, lymphedema was reported in all patients, but none was debilitating. 1 Femoral artery and vein breakdown have been eliminated by sartorius transfer techniques. Operative mortality has been reported to be as high as 3.3% 5 but approaches zero in the modern surgical era. Results In an analysis of 76 patients, McDougal found that poorly differentiated or corporal invasive tumors were at very high risk of having lymph node metastases. Nodal metastases were found in 81% of these cases, including 78% of patients in this group with clinically negative lymph nodes. In contrast, if the tumor was both superficial (not invading corporal tissue) and well or moderately differentiated, only 4% of patients had nodal metastases. Furthermore, survival in the group of patients with clinically uninvolved nodes undergoing prophylactic dissection was 92%, compared to 33% in the group undergoing delayed lymph node dissection at the time of developing clinically enlarged lymph nodes. 8 Theodorescu et al.14 found that 62% of patients with clinically negative nodes at the time of treatment of the primary tumor ultimately developed relapse in the inguinal lymph nodes. This contrasts sharply to the 20% incidence of positive nodes found in prophylactic lymphadenectomy series, suggesting a significant pathologic false-negative rate caused by occult micrometastases. The actuarial risk of relapse was 45% for patients with well-differentiated tumors, whereas all others had a 100% risk of relapse. Taken together, these series demonstrate that the only patients at very low risk for nodal metastases are those with noninvasive and well-differentiated tumors. H4>CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Catalona WJ. Modified inguinal lymphadenectomy for carcinoma of the penis with preservation of saphenous veins: technique and preliminary results. J Urol 1988;140:306–310. Catalona WJ. Re: Modified inguinal lymphadenectomy for carcinoma of the penis with preservation of saphenous veins: technique and preliminary results. Editorial. J Urol 1988;140:836. Connell CF, Berger NA. Management of advanced squamous cell carcinoma of the penis. Urol Clin North Am 1994;21(4):745–756. Daseler EH, Hanson BJ, Reimann AF. Radical excision of the inguinal and iliac lymph glands. Surg Gynecol Obstet 1948;87:679–694. Ekstrom T, Edsmyr F. Carcinoma of the penis. A clinical study of 229 cases. Acta Chir Scand 1958;115:25. Frew IDO, Jefferies JD, Swiney J. Carcinoma of the penis. Br J Urol 1967;39:398. Kattan J, Culine S, Drop JP, et al. Penile cancer chemotherapy: twelve years experience at Institut Gustave-Roussy. Urology 1993;42:559–562. McDougal WS. Carcinoma of the penis: improved survival by early regional lymphadenectomy based on the histological grade and depth of invasion of the primary lesion. J Urol 1995;154:1364–1366. McDougal WS, Kirchner FK Jr, Edwards RN, Killian LT. Treatment of carcinoma of the penis: the case for primary lymphadenectomy. J Urol 1986;136:38. Mukamel E, deKernion JB. Early versus delayed lymph-node dissection versus no lymph-node dissection in carcinoma of the penis. Urol Clin North Am 1987;14:707–711. Pedrick TJ, Wheeler W, Riemenschneider H. Combined modality therapy for locally advanced penile squamous cell carcinoma. Am J Clin Oncol 1993;16:501–505. Shammas FV, Ous S, Fossa SD. Cisplatin and 5-fluorouracil in advanced cancer of the penis. J Urol 1992;147:630–632. Skinner DG, Leadbetter WF, Kelley SB. The surgical management of squamous cell carcinoma of the penis. J Urol 1972;107:273. Theodorescu D, Russo P, Zhang A-F, et al. Outcomes of initial surveillance of invasive squamous cell carcinoma of the penis and negative nodes. J Urol 1996;155:1626–1631.

Chapter 68 Peyronie's Disease Glenn’s Urologic Surgery

Chapter 68 Peyronie's Disease
Kenneth S. Nitahara and Tom F. Lue

K. S. Nitahara and T. F. Lue: Department of Urology, University of California, San Francisco, California 94143–0738.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Multiple Parallel Plication Plaque Incision and Venous Graft Outcomes Complications Results Chapter References

Peyronie's disease is a fibrous thickening in the tunica of the penis causing abnormal curvature or narrowing in the erect state. The relative inelasticity of the fibrous plaque precludes normal elongation of the tunica albuginea during the transition from the flaccid to the erect state. The plaque is usually located on the dorsal surface of the tunica albuginea, causing curvature in that direction. However, the scar may also be located laterally or ventrally, causing curvature in the direction of the scar. Circumferential scarring of the tunica or involvement of the septum may cause a shortening of the entire erect penis. The severity of curvature may be mild, such that it has no impact on sexual intercourse, or it may be more severe such that sexual intercourse is painful or even physically impossible.

DIAGNOSIS
The evaluation of such patients should begin with a thorough history and physical examination with emphasis on erectile function. Information on specific sexual or traumatic events leading to Peyronie's disease should be identified, as well as the duration of symptoms. In the acute phase, an inflammatory reaction may persist for 6 months or more, causing changes in the severity and even direction of angulation; surgical treatment should be postponed until this dynamic phase is complete. Other causes of penile scarring should be identified such as chronic penile intracavernosal injections for impotence. Also, a family history of Peyronie's disease or of Dupuytren's contracture should be identified. Finally, changes in libido or emotional stress related to impotence should be discussed. Although the psychological impact of such a change in sexual function may vary among men, reassurance will always help to establish rapport and confidence in the urologist. The focused physical evaluation begins with home photos of the erect penis. This allows the urologist to assess the character of curvature in the most natural setting. Palpating the stretched penis enables the physician to identify the extent and location of scarring. Plain films or sonography may also be helpful in identifying calcifications within scars. Duplex ultrasonography measures the vascular supply, collateral circulation, and venous drainage before and after intracavernous injection of vasodilators. 1 This also helps to characterize the angulation of the erect penis. Patients with adequate erections after pharmacologic injections are candidates for surgical repair. If venous leakage is suggested, cavernosometry and pharmacocavernosography may be indicated. Patients with identified venous leakage may undergo simultaneous repair.

INDICATIONS FOR SURGERY
The indications for surgery in Peyronie's disease include: (a) persistent pain or deformity for more than 12 months after the onset of the disease; (b) poor response to conservative therapy; (c) severe curvature, narrowing, shortening, or distal softening of the penis with or without erectile dysfunction. In the surgical treatment of Peyronie's disease we prefer multiple parallel plication (MPP) on the opposite side of the curvature or plaque incision with venous grafting. Incising rather than excising the fibrous plaque, we have found, results in a stronger repair with a lower incidence of impotence. During informed consent, patients are advised about both procedures. The MPP has the advantages of a relatively short and minimally invasive procedure; however, the penis may be shortened. Conversely, whereas the plaque incision with venous graft procedure is longer and more invasive, the length of the penis is usually maintained. The deep dorsal and saphenous veins are usually used as the venous graft; for the latter, the lower portion is usually chosen because of its superficial location.

ALTERNATIVE THERAPY
Alternative therapy to the MPP and incision and venous graft procedure would include observation, which is indicated in mild and evolving Peyronie's disease, medical management (vitamin E 400 to 1,000 IU/day, protaba, or colchicine 2.4 mg/day), or intralesional injections (steroid, verapamil, or a-interferon). Radiation therapy may damage penile erectile tissue and is indicated only in patients with severe penile pain that is resistant to other therapies. There are several approaches to the operative treatment of Peyronie's disease, such as Nesbit wedge resection, in which ellipsoids of tunica albuginea are excised on the side opposite the plaque. The cut edges of the tunica are then closed primarily to decrease the shaft length on the opposite side of the curved penis to allow a straightened penis. 8 For curvatures from Peyronie's disease, the plication procedure can be performed as long as the patient accepts the consequence of a shorter penis. Plaque incision and graft placement are a better procedure for the patient with severe curvature, a short penis, an hourglass deformity, distal flaccidity, penile narrowing or shortening without curvature, and persistent penile pain. Plaque excision and dermal graft placement have also been described; this technique involves excising the tunica involved by scar, followed by grafting a section of dermal tissue on to the surgically created defect. 5 Other autologous tissue and inert substances have been used as grafts. 2,3 and 4 Also, placement of a penile prosthesis has been described by some authors as an option for treating such patients. 6

SURGICAL TECHNIQUE
Multiple Parallel Plication This procedure (MPP) is performed under local anesthesia with the patient in the supine position. After the genitalia are prepped and draped in a sterile fashion, an artificial erection is induced with 10 µg of PGE 1 or 30 mg of papaverine using a 28-gauge needle. For dorsal curvature, a longitudinal ventral incision is made to the level of Buck's fascia so that the corpus spongiosum and cavernosum are identified ( Fig. 68-1). Using 2-0 Ticron nonabsorbable braided sutures, full-thickness stitches are placed 2 to 3 mm away from either side of the spongiosum at a point opposite the maximal curvature. Two to three pairs of sutures are usually needed along the length of the shaft. It is important not to grasp too much tissue in each suture because this will result in bunching of the tissue.

FIG. 68-1. Dorsal curve plication. After the tunica has been exposed, place sutures 2 to 3 mm away from the spongiosum in a location opposite the point of maximal curvature. Multiple parallel sutures are recommended because grasping large lengths of tissue in a single bite may cause tissue bunching.

For ventral curvature, a circumcising incision is followed by reflecting the skin of the penis to its base ( Fig. 68-2). With special care taken not to injure the neurovascular bundle, 2-0 Ticron sutures are placed opposite the maximal point of curvature between the deep dorsal vein and the dorsal arteries. Multiple bites of tissue with each suture are used to avoid bunching of the tissue about the suture. If the penis becomes flaccid during the procedure, 0.9% normal saline may be infused with the base of the penis compressed to assess curvature. If the penis remains rigid at the end of the procedure, detumescence with aspiration and/or instillation of phenylephrine HCl (250 to 500 µg) should be performed. The skin is reapproximated with 4-0 chromic sutures, and the penis is wrapped with a lightly compressive dressing, with care taken not to occlude blood flow to and from the glans.

FIG. 68-2. Ventral curve plication. After the tunica and neurovascular structures have been identified, carefully place sutures between the dorsal vein and artery on either side opposite the point of maximal curvature. This should be done under magnified vision.

Plaque Incision and Venous Graft Before surgery, the patient's lower extremities should be evaluated for quality of lower saphenous vein. The decision over which side to use should also be based on size of the vein, history of lower extremity surgery, and chronic pain or edema. The procedure may be performed under local, regional, or general anesthesia in the supine position. After a circumcising incision is made, the skin is reflected to the base of the penis. For patients with a dorsal curvature, the neurovascular bundle is identified, followed by mobilization of the deep dorsal vein along the length of the shaft ( Fig. 68-3). Small tributary veins should be ligated with 3-0 or 4-0 silk sutures and then sharply divided. Perforating veins left unligated will cause troublesome bleeding at the end of the procedure. Mobilization of the dorsal vein should be limited proximally to within about 1 cm of the pubis; more aggressive mobilization will result in postoperative scarring and shortening of the penis. Next, the dorsal vein is ligated with a 2-0 silk suture and sharply divided at the ends of the dissected portions; this portion of vein is saved in normal saline.

FIG. 68-3. Mobilization of the deep dorsal vein. After the neurovascular structures have been identified, incise Buck's fascia directly over the dorsal vein. Further mobilization of the vein with ligation and division of its branches will allow it to be removed and preserved in saline.

The next and most critical part of the procedure is dissection and mobilization of the neurovascular bundle ( Fig. 68-4); this should be done under loupe or magnifying scope vision. The bundle is separated from the surface of the tunica albuginea, with care taken to avoid piercing the tunica. In nonaffected areas of the tunica there is normally a clear plane allowing dissection of the bundle off of the tunica; this plane is routinely obliterated at the area of Peyronie's plaque, making dissection more difficult. Mobilization of the neurovascular bundles should be sufficient that there is some slack with the penis stretched; this is the limiting factor to the eventual length of the penis.

FIG. 68-4. Mobilization of the neurovascular bundle. Using magnified vision, carefully dissect the neurovascular bundles on either side off of the tunica albuginea. These bundles should be mobilized sufficiently that there is some slack when the penis is pulled in the stretched position.

With a 21-gauge needle, normal saline solution is infused into the corpora via a lateral tunical puncture to characterize the abnormal curvature. With this anatomic site noted, the tunica overlying the point of maximal curvature is marked and sharply incised ( Fig. 68-5). The initial incision is transverse through the middle of the scar. These are connected by perpendicular incisions at either end of the initial incision to form an “H” configuration. Cases in which there is concomitant narrowing of the corporal body from scar contracture may require an extra incision at each tip of the “H” configuration; this allows side grafts to be placed to widen the penis. After the incision is made, the size of the defect to be covered is measured by pulling the penis to the stretched position longitudinally and transversely.

FIG. 68-5. Tunical incision. Sharply incise the marked area directly across from the scarred area in an “H” fashion. This will increase the length of the incised side of the penis. Vein graft anastomosis. After the venous segments have been harvested and incised lengthwise, place them side by side with their endothelial side down. These are then anastomosed in a watertight fashion either with 5-0 Maxon or staples. After the graft has been created, it is placed over the defect and sewn into place with a continuous 4-0 Maxon suture.

With this measured area as a guide, the venous graft may now be considered. In general, one length of vein opened lengthwise will provide a stretched width of 1 cm. The preserved dorsal vein may be considered for use, but this is often not sufficient. Additional vein may be obtained from the previously prepped lower saphenous vein by sharply incising the skin directly overlying the vein and then mobilizing the vein and dividing its branches. Be careful with the peripheral nerve branch that runs alongside the vein; blunt injury to this may cause pain, and division may cause numbness. After a sufficient length of vein has been harvested, it can be cut into pieces to fit the specifications. Sharply incise each vein segment lengthwise to create a rectangle of tissue. With the endothelial sides faced down, the pieces of vein are lined up side by side and connected with either a vascular stapler or a 5-0 Maxon suture in a running, locking fashion ( Fig. 68-5). This graft is then sutured to the surgically created defect in the tunica using 4-0 Maxon in a running, locking fashion (Fig. 68-5). This should create a watertight closure. To test the adequacy of repair, inject normal saline into the cavernosal bodies as described above. It is occasionally necessary to make another incision in the tunica and apply a second vascular graft. Placing plication sutures on the ventral surface of the tunica may also be helpful during the healing process, but this is generally not necessary. The neurovascular bundles are repositioned in the dorsal midline with 5-0 Maxon. For ventral plaques, it is important to identify the neurovascular bundles, but their dissection is generally not necessary. Exposure of the plaque mandates dissection of the spongiosum off of the cavernosum. Troublesome bleeding may be encountered if the spongiosum is entered; this is usually resistant to electrocoagulation but may be more easily controlled with suture ligation with 5-0 Maxon. Penile skin closure is done in two layers using 4-0 Vicryl for fascia and 4-0 chromic for skin in an interrupted, simple fashion. A Foley catheter is left to drain the bladder. Xeroform gauze covers the incision, and Coban tape is used as a light compression dressing for control of swelling and hematoma. A scrotal support and ice pack are also helpful in this regard. The leg incision is closed in two layers, using 4-0 Vicryl for both fascia and skin. A dry sterile dressing is used to cover this incision. Patients who have undergone vein grafting are usually admitted to the hospital for an overnight stay. On the first post operative day, Foley catheters are removed, and the patients are discharged to home after demonstrating the ability to urinate. Patients are taught to change their penile dressing at home by reapplying Coban tape in a lightly compressive fashion daily for 10 days. They are specifically instructed to avoid wrapping too tightly, which might compromise blood flow to the glans and penile skin. They are also instructed to abstain from sexual relations for at least 5 weeks.

OUTCOMES
Complications Immediate complications may include bleeding and hematoma from the incision sites, persistent edema, wound infection, and necrosis of tissue from excessively tight dressings. The risks of these complications may be minimized by careful attention to basic principles of operative and perioperative antisepsis and postoperative care. Keeping the surgical sites clean is essential, as is maintaining a lightly compressive but not constrictive dressing around the penis for 10 days. We have not found that use of antibiotics for longer than the perioperative period has been of any increased benefit. Patients may also describe a painful erection for up to 2 months and penile numbness for as long as 6 months. These findings are normal in the postoperative period, and patients should be reassured that normal sexual relations will return after the tissue has healed. Patients should also be informed that about 2% of cases have recurrence of Peyronie's disease after this procedure. Finally, erectile dysfunction following surgery may occur in patients who have marginally poor vascular supply to the penis. This is predictable, and at-risk patients should be identified with preoperative evaluation including duplex ultrasonography so that additional management may be planned. Results We have performed more than 50 MPP and 200 plaque incisions combined with venous grafts in the past 6 years. Worsening curvature at the same site was noted in two patients who underwent MPP 6 months and 1 year after surgery, respectively. Three patients had new curvature after plaque incision and venous graft. Interestingly, the sites of deformity were away from the venous graft, suggesting a recurrence rather than progression of the disease. From our animal experiments and histologic studies, we have noted that the venous graft thickened and transformed into a fibroelastic structure similar to the tunica after 3 months. After surgery, all patients had a straight penis or curvature of less than 15 degrees during erection with the exception that less than a handful of cases still had curvature of approximately 30 degrees. Penile narrowing is more difficult to correct. However, many patients still had mild indentations after surgery. In conclusion, Peyronie's disease may be functionally and emotionally disabling to men. An understanding of the concepts and familiarity with the surgical options are important for urologists because up to 1% of the United States population may be affected by this. 7 For patients who are operative candidates, surgical repair may enable them to resume normal sexual activity without pain. We think that the continued use of these surgical techniques is warranted. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. Brock G, Nunes L, von Heyden B, Martinez-Pineiro L, Hsu GL, Lue TF. Can a venous patch graft be a substitute for the tunica albuginea of the penis? J Urol 1993;150:1306. Das S. Peyronie's disease: excision and autografting with tunica vaginalis. J Urol 1980;124:818. Fallon B. Cadaveric dura mater graft for correction of penile curvature in Peyronie's disease. Urology 1990;35:127. Ganabathi K, Dmochowski R, Zimmern P, Leach GE. Peyronie's disease: surgical treatment based on penile rigidity. J Urol 1995;153:662. Jordan GH, Devine PC, Schlossberg SM, Gilbert DA, Horton CE, Devine CJ Jr. The dermal graft procedure for Peyronie's disease. J Urol 1987;137:220A. Knoll LD. Use of penile prosthetic implants in patients with penile fibrosis. Urol Clin North Am 1995;22:857. Lindsay MB, Schain DM, Grambsch P, et al. The incidence of Peyronie's disease in Rochester, Minnesota, 1950 through 1984. J Urol 1991;146:1007. Nesbit RM. Congenital curvature of the phallus: report of three cases with description of corrective operation. J Urol 1965;93:230.

Chapter 69 Priapism Glenn’s Urologic Surgery

Chapter 69 Priapism
Kenneth S. Nitahara and Tom F. Lue

K. S. Nitahara and T. F. Lue: Department of Urology, University of California, San Francisco, California 94143–0738.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Glans–Cavernosal Shunt Al-Ghorab Procedure Quackles Cavernosum–Spongiosum Shunt Cavernosal–Venous Shunts Cavernosum–Dorsal Vein Shunt Postoperative Considerations High-Flow Priapism Outcomes Complications Results Chapter References

Priapism is a persistent, often painful erection that does not subside after ejaculation. Priapism is considered a surgical emergency until proven to be otherwise; in low-flow or ischemic states, patients must be treated immediately. Untreated cases will lead to ischemia, necrosis, and severe scarring of the erectile tissue. After 24 hours there is evidence of irreversible smooth muscle necrosis, destruction of epithelium, and exposure of the basement membrane with thrombocytes adherent to it. Priapism may be classified as high flow or low flow, depending on its etiology. High-flow priapism is often caused by an arteriovenous fistula circulating well-oxygenated blood; this may not need urgent surgical attention. Low-flow priapism is caused by a trapping of poorly oxygenated blood in the cavernosal tissue with associated ischemic damage; these patients must be decompressed as soon as possible.

5

DIAGNOSIS
On initial evaluation, patients should have a complete history and physical, with emphasis on possible causative factors. Physical examination should assess for subtle neurologic defects, pelvic masses, and perineal abnormalities as well as the characteristics of the penis. Patients with significant pain are more likely to have low-flow priapism, but this is certainly not diagnostic. Laboratory tests should include complete blood count, urinalysis, and sickle cell screen as well as evaluation for coagulopathy if indicated. Before any therapeutic attempts, informed consent should be obtained. Patients should understand that there is about a 50% chance of erectile dysfunction regardless of duration of priapism or method of management. Diagnosis of the type of priapism and the management of it should commence as soon as possible ( Fig. 69-1). Aspiration of penile blood may be diagnostic and therapeutic; acidotic, hypoxic blood is likely to indicate veno-occlusion.

FIG. 69-1. Algorithm for the diagnosis and management of priapism.

INDICATIONS FOR SURGERY
If medical management is unsuccessful in allowing detumescence of the penis, surgical management should be considered. The immediate goal is to allow blood to flow readily in and out of the penis to prevent ischemic damage.

ALTERNATIVE THERAPY
Occasionally, aspirating blood will allow resolution of the erection. a-Adrenergic therapy is often needed and may be used if the priapism has lasted less than 36 hours; after that, damaged smooth muscle will no longer respond to the pharmacologic stimulus. A variety of a agonists have been used; we have had the most success with aspiration plus diluted phenylephrine, 250 to 500 µg every 3 to 5 minutes until detumescence. This therapy is not recommended for high-flow priapism; it is ineffective and may have systemic side effects because the venous channels are wide open.

SURGICAL TECHNIQUE
Surgical correction involves connecting the engorged cavernosal tissue with the glans, corpus spongiosum, or dorsal or saphenous vein. This surgically created fistula will allow blood to drain from the corpora cavernosa until the pathologic process has resolved. Ideally, the surgically created fistula will spontaneously close after the priapism-causing factors have resolved. On the basis of these principles, many techniques have been used for the urgent treatment of priapism. We describe the procedures with which we have been most successful, including glans–cavernosum, spongiosum–cavernosum, and cavernosum–venous anastomoses. Distal shunts are attempted first and are followed by more proximal ones if penile detumescence is not achieved. Any shunt used must be tested intraoperatively to assess its success. This is best done by assessing the intracavernosal pressure for 10 to 15 minutes after the completion of the shunt. Placing a pressure-monitoring needle within the corpora will measure the pressure ( Fig. 69-2). A pressure of less than 40 mm Hg for longer than 10 minutes is necessary for a successful shunt. If this is not achieved, then a more invasive procedure must be performed.

FIG. 69-2. After multiple glans–cavernosal shunts, monitoring intracavernosal pressure for 10 minutes after closure to ensure a pressure of <40 mm Hg is essential for a successful repair.

Glans–Cavernosal Shunt This is the least invasive technique for priapism management. Under local or general anesthesia, a #11 knife blade is inserted through the coronal sulcus to penetrate the tip of the corpus cavernosum and glans. A pointed knife with a 5- to 10-mm cutting edge is inserted parallel to the dorsal surface of the penis into the corona glans and then twisted 90 degrees to incise the tunica albuginea close to the tip of the cavernous body. 2 After this anastomosis has been made, the glans skin is closed with fine chromic catgut. Good results are suggested by a jet of dark blood after the incision and swelling of the glans after closure. Al-Ghorab Procedure Under local anesthesia, a 2-cm transverse incision is made on the dorsum of the glans penis 1 cm distal to the coronal ridge. 6 The distal end of the glans is then hinged forward to expose the ends of the bulging corporal bodies. Using sharp dissection, a 5-mm segment of tunica at the tip of the corporal bodies is excised, including a portion of the septum. Dark blood should be draining from the corporal bodies. When detumescence has occurred, the skin is repositioned with chromic sutures in a way so as not to obliterate the spongy vascular space of the glans. Quackles Cavernosum–Spongiosum Shunt Under regional or general anesthesia, an 18-Fr Foley catheter is placed into the bladder for ease in identifying the urethra. 7 A vertical incision is made in the perineal skin just posterior to the scrotum and overlying the bulbous urethra ( Fig. 69-3). The bulbocavernous muscle is reflected off of the urethra and preserved. Identify the junction between the spongiosal and cavernosal bodies; the spongiosum should be easily compressible and flaccid. Corresponding 1-cm longitudinal incisions are made through the tunica of the spongiosal and unilateral cavernosal bodies close enough to each other so that they may be sewn together. Excising an elliptical segment of tunica rather than just incising may provide a better fistula. After one cavernosal body is opened, the old blood should be evacuated with manual milking until bright red blood is expressed. The shunt is completed by suturing the posterior wall followed by the anterior wall in a watertight fashion with 5-0 PDS. If the intracaver-nosal pressures stay below 40 mm Hg after 10 to 15 minutes, then the skin may be closed. If the pressure is more than 40 mm Hg, then the other side should be done.

FIG. 69-3. Cavernosal–spongiosal shunt. (A) Incisions on the tunica of the spongiosum and cavernosum. (B) Anastomosis of the openings.

Cavernosal–Venous Shunts In 1964, a venous bypass using saphenous vein was described for the treatment of priapism. 3 Under general anesthesia, the patient is placed in the supine position with his legs slightly froglegged. The first incision is made over the saphenofemoral junction 3 to 4 cm below the inguinal ligament; this incision may be extended distally along the course of the saphenous vein ( Fig. 69-4). Extending distally from the fossa ovalis, the saphenous vein is mobilized for 8 to 10 cm, enough to be tunneled subcutaneously to the root of the penis without tension. At the distal extent of vein mobilization, it is ligated with 2-0 silk and sharply divided.

FIG. 69-4. Cavernosal–saphenous vein shunt.

The second incision is a 1-inch vertical one over the lateral aspect of the penile shaft near the root of the penis and extending through all layers to the tunica albuginea. Blunt dissection is used to tunnel between the two incisions, after which the cut length of vein is passed to the medial incision without tension, twisting, or excessive angulation. With the ipsilateral corporal body identified, excise a small ellipse of tunica. Manually express and irrigate the old blood; then spatulate the vein in preparation for anastomosis. The vein is sewn to the aperture in the tunica with 5-0 PDS in a running, watertight fashion. Cavernosum–Dorsal Vein Shunt

This shunt 1 is based on the principle that the dorsal vein is not involved in priapism. A 4-cm skin incision is made at the base of the penis, extending through Buck's fascia. The superficial or deep dorsal vein is mobilized sufficiently that it will reach the desired site of tunical ellipse incision. After the cavernosal body has been opened and the vein is spatulated on its ventral surface, the anastomosis may be done using 5-0 PDS. Postoperative Considerations After surgery, patients are admitted as inpatients for intravenous antibiotics and supportive care. If treatment is successful, patients may be discharged on the first postoperative day. These shunting procedures will allow drainage through the dorsal or saphenous veins, glans, or corpus spongiosum. Circular compressive dressings should be avoided because they may decrease venous drainage and perpetuate the process. Intermittent manual squeezing and milking of tissue will help keep the shunt open and prevent a recurrence of priapism. The penis will likely appear partially erect even with an effective shunt because of the edematous process. If there is any question, blood gasses may determine ischemic versus nonischemic situations. Recurrent priapism may be treated with oral or self-injected a-adrenergic agonists such as terbutaline or phenylephrine. High-Flow Priapism A history of perineal trauma with a nonpainful, intermittent erection is suggestive of nonischemic priapism. Cavernous blood gases will reflect arterial levels. Color duplex ultrasonography is the best diagnostic tool; it can identify the high-flow state and site of the ruptured artery. After the diagnosis has been made, these patients may be observed for as long as several months to allow the fistula to spontaneously close. If this does not occur, then angiographic embolization of the ruptured artery should be performed. If the fistula persists, then suture ligation of the ruptured artery via a perineal approach should be done under ultrasound guidance.

OUTCOMES
Complications Early complications include recurrence, bleeding, infection, skin necrosis, abscess, cellulitis, gangrene, urethral damage, urethrocutaneous fistula, and urethral stricture. Late complications include fibrosis of erectile tissue and failure of venous shunt to close spontaneously, leading to venogenic impotence. Careful handling of tissue, close immediate postoperative management, and judicious use of antibiotics will best minimize the risk of these complications. Pulmonary embolism has been reported to occur following cavernosa–saphenous shunt. This serious complication is thought to occur from early rethrombosis of the shunt and should be considered in any patient with suggestive symptoms. Results The most common complication of priapism is loss of potency, which may be a function of the disease as well as a result of the procedure. Postmanagement potency rates range from 54% to 57%, and the impotence is generally considered to be a result of fibrosis of the corpora from the inflammatory response. The most critical factor in maintaining potency in patients with priapism is the duration of the priapic episode. Patients treated within 15 to 36 hours will more likely have a favorable response than those who delay treatment. Patients with prolonged or recurrent episodes are more likely to suffer impotence as a result of fibrosis. Other risk factors are older hypertensive patients and those with repeated episodes from sickle cell anemia. Traumatic priapism appears to have the best prognosis for future erectile function. The success in treatment of priapism is a function of the method chosen and the underlying cause of the priapism. Low-flow priapism from sickle cell anemia is usually successfully treated by hydration, alkalinization, analgesia, and transfusion. The results with other forms of priapism, however, are difficult to assess because of the lack of any series of large numbers of patients and the many causes of priapism. Success rates of 0% to 50% in detumescence have been reported with aspiration/irrigation. Potency rates following treatment for priapism range from 50% to 80%. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. Barry JM. Priapism: treatment with corpus cavernosum to dorsal vein penis shunts. J Urol 1976;116:754. Ebbehoj J. A new operation for priapism. Scand J Plast Reconstr Surg 1975;8:241. Goulding FJ. Modification of cavernoglandibular shunt for priapism. Urology 1980;15:64. Odelowo EO. A new cavernospongiosum shunt with saphenous vein patch for established priapism. Int Surg 1988;73:130. Spycher MA, Hauri D. The ultrastructure of erectile tissue in priapism. J Urol 1986;135:142. Wendell EF, Grayhack JT. Corpora cavernosa–glans penis shunt for priapism. Surg Gynecol Obstet 1981;153:586. Reference not available.

Chapter 70 Penile Prosthesis Glenn’s Urologic Surgery

Chapter 70 Penile Prosthesis
John J. Mulcahy

J. J. Mulcahy: Department of Urology, Indiana University Medical Center, Indianapolis, Indiana 46202.

Diagnosis Alternative Therapy Indications for Surgery Description of the Procedure Three-Piece Inflatable Prosthesis: Infrapubic Approach Three-Piece Inflatable Prosthesis: Scrotal Approach Two-Piece Inflatable Prothesis: Scrotal Approach Unitary Penile Prosthesis Outcome Complications Results Chapter References

Erectile dysfunction resulting from a disease process or injury or as a concomitant occurrence of the aging process is a common problem. It is estimated that about 25% of the population is unable to have an erection suitable for penetration on a consistent basis. By age 65, half of men are experiencing this problem. Atherosclerosis, trauma to the erectile bodies, aberrations of innervation to the penis, or incomplete corporal smooth muscle relaxation with venous leakage may each individually or in combination contribute to the suboptimal penile rigidity. When Kinsey chronologized the sexual habits of men in 1948, he concluded that 90% of erectile problems were psychological or situational and that only a small percentage were organic in nature. 4 Today, with advances in our ability to test for the causes of erectile disability, the percentages are reversed, with the majority of problems seen by physicians as having a basic physical component.

DIAGNOSIS
The most helpful element in determining the etiology of the erectile difficulty is the history. A detailed discussion of the problem highlighting the presence of diseases, use of medications, prior injuries or surgical procedures, libido, environmental influences, and partner's attitudes will most often indicate the source of the dysfunction. Serum testosterone and prolactin levels may be assessed to ascertain if hormonal abnormalities are present. Nocturnal penile tumescence monitoring can be used to determine the quality of sleep erections if there is a true question as to whether the problem is situational or physical. Dynamic infusion cavernosometry and cavernosography or penile duplex ultrasound evaluation will assess the status of cavernosal blood flow and detect the presence of venous leakage when appropriate. Testing for erectile dysfunction is often not necessary and should be done selectively to direct the treatment appropriately.

ALTERNATIVE THERAPY
Counseling the couple regarding environmental influences and the partner's participation in the arousal process should be the first step in restoring sexual compatibility. Altering medications that have an adverse influence on the erectile mechanism has occasionally been beneficial. Adjusting the hormonal milieu or trying some oral smooth muscle relaxant vasodilators has proven helpful in some mild cases of erectile disfunction. More effective oral and intraurethral medications are being developed for boosting waning erections. For patients with significant physical impotence, three modalities are in common use: medications, vacuum erection devices, and penile implants. Medication comes in three forms: intracorporal injections, an intraurethral pellet, and a recently introduced pill, Sildenfil. This new medication has been well received because of its simplicity of administration, efficacy, and lack of significant side effects. Vacuum devices have become popular because of their relatively low cost and noninvasive nature. Although the erection achieved is limited, successful results can be gained by patience and practice. Vascular procedures to augment penile blood flow have limited application. Modest success has been achieved in young patients with few vascular risk factors in whom blockage of inflow into the corporal bodies has been bypassed by rerouting of arterial blood. 3 Procedures to stop venous leakage have had poor long-term success and have been abandoned except in unusual cases of isolated leakage such as traumatic corporal–saphenous fistula. 5

INDICATIONS FOR SURGERY
Despite the arrival of less invasive and aggressive treatment modalities for erectile dysfunction, penile implants are still used frequently. With the advent of medical clinics treating erectile dysfunction and advertising heavily in major cities throughout the country, a large number of men are now coming in for treatment of this problem. Although the majority of these men will choose an injection program or vacuum erection device, sales of implants in most regions of the country are still stable, with a slight increase in use seen in some areas. We are noting a shift in the type of patient choosing an implant and for whom an implant is most suitable. In those with Peyronie's disease where soft erections and curvature combined to make intercourse impossible, an implant will both straighten and strengthen the erection. Vacuum devices will not conform to the curved penis, and an intracorporal injection will exaggerate the curve, propagate the scar, give unbalanced firmness to the erection, and prove difficult to administer in an area of dorsally located plaque. After an episode of significant priapism, corporal bodies are completely scarred, and blood cannot flow into the sinusoids adequately with any stimulus. An implant is the best solution, but the surgery will be difficult in the face of total corporal scar tissue. After significant trauma to the penis or when a penile prosthesis has been removed for infection or erosion, a subsequent implant may be the only solution. The enthusiasm of a patient who is satisfied with his implant is sometimes transmitted to a co-worker on the assembly line who opts for an implant as the treatment of his problem on his friend's recommendation.

DESCRIPTION OF THE PROCEDURE
Three-Piece Inflatable Prosthesis: Infrapubic Approach The patient is placed in the supine position, and the lower abdomen, genitalia, and upper thighs are shaved, prepped, and draped. A midline incision is made just above the dorsal surface of the neck of the penis approximately 5 cm in length. This is carried down through the subcutaneous tissues to expose the neck of the penis. Buck's fascia and areola tissue are skeletonized from the penis on the dorsal surface lateral to the dorsal neurovascular bundle at the 10-o'clock and 2-o'clock positions. Two 3-0 Proline stay sutures are placed in each corporal body parallel to the shaft of the penis lateral to the dorsal neurovascular bundle. The tunica albuginea of the right corporal body is incised between the two stay sutures using a #15 blade. The corporotomy is then broadened to approximately 3 cm using the Mayo scissors. The Metzenbaum scissors are used to dilate the spongy tissue distal to the subglandular area and proximal to the ischial tuberosities. The plane of dissection is carried out just under the tunica albuginea. The cavity in each direction is then broadened using dilators starting with size 9 mm up to size 12 mm. Corporal length is then measured in each direction, and the measured distances are added to determine the total corporal length. The same procedure of incision, corporal dilation, and measurement is then performed on the left side. The appropriate size cylinder with rear tip extenders equal to total corporal length is chosen so that the input tube exits directly from the corpus cavernosum at the site of its attachment to the cylinder. The corporotomy is then extended proximally so that this input tube exit can appropriately occur ( Fig. 70-1). Care must be taken not have the input tube exiting too distally, as prepubic pressure during intercourse may have a shearing effect on this tube at its attachment to the cylinder. By use of a Keith needle and Furlow introducer, the cylinder with its rear tip extender is introduced into the corporal body. The corporotomy is closed with a running 3-0 PDS or Maxon suture beginning at the exit of the input tube and progressing distally. Cylinder insertion and closure are accomplished on the opposite side in a similar fashion.

FIG. 70-1. Input exit from proximal end of coporotomy where it is attached to cylinder.

A midline incision is then made in the linea alba in its most inferior portion. The rectus muscles are separated in the midline with a blunt clamp, and the paravesical space on the right or left side is developed with the index finger. A balloon reservoir is then placed in this location and filled with isotonic saline solution. Care should be taken to make sure that an adequate reservoir cavity is present to prevent autoinflation. The volume of saline filling solution used should be about 5 cc less than the capacity of the balloon and at least 10 cc more than the total volume contained in both of the cylinders when they are filled to maximum inflation. The hiatus in the linea alba is then closed with a running 0 Dexon suture. A Stille clamp is then placed through the subcutaneous tissues just above the abdominal wall fascia from the inferior portion of the midline prepubic incision to the right of the penis and down into the scrotum to dilate a tract for the pump. This tract is then broadened with the size 18 mm dilator. The pump is then manually placed through this tract to the most inferior portion of the right hemiscrotum in the subcutaneous tissues. The pump is then secured in place with a Babcock clamp while tubing connections are made. The appropriate tubing for the pump is connected to the tubing from the balloon reservoir using a straight connector. The appropriate tubing for the pump is then connected to the tubing from each penile cylinder using right-angle connectors if this tubing is not preconnected. All connections are made in the prepubic space. The device is then cycled a number of times to assure easy inflation and deflation, good rigidity to the erection, and good flaccidity in the deflated state. The prepubic incision is closed with two layers of running 3-0 chromic on the subcutaneous tissue and 4-0 subcuticular Dexon or Vicryl reenforced by SteriStrips on the skin. Copious irrigations of antibiotic solution are used throughout the case. Three-Piece Inflatable Prosthesis: Scrotal Approach This procedure differs from that of the prepubic approach as follows. A midline scrotal incision is made about 4 cm in length inferior to the penoscrotal junction. Self-retaining retractors are placed, and the corpus spongiosum is identified after areolar tissue is sharply dissected from the corporal bodies. Two 3-0 Proline stay sutures are placed in each corporal body on either side of the corpus spongiosum, and the corporal bodies are incised, dilated, and measured as per the infrapubic approach. Cylinders are sized, filled with saline solution, rear tip extenders are added, and the cylinders are inserted with the Furlow introducer as described previously. A Richardson retractor then pulls the skin on the right side of the penis upward to expose the external inguinal ring. The spermatic cord is identified and retracted laterally. The Metzenbaum scissors are used to open the transversalis fascia by cutting against the pubic tubercle ( Fig. 70-2). The index finger is passed through the inguinal canal, and the right paravesicle space is developed. The balloon reservoir is placed around the index finger and inserted through the inguinal canal so that it rests in the right paravesicle space. The reservoir is then filled to a volume 5 cc less than its capacity and at least 10 cc more than the combined volume of both fully inflated cylinders. The tubing is then pulled so that the balloon reservoir rests at the level of the internal inguinal ring. A pocket is developed in the right hemi-scrotum for placement of the pump in the subcutaneous tissues. The appropriate tubing from the pump is connected to the tubing from the balloon reservoir using a straight connector in the midscrotal area.

FIG. 70-2. Metzenbaum scissors cutting through transversalis fascia against pubic tubercle.

If the tubing from the pump to the cylinders is not preconnected, one of the tubes from the pump to the cylinders is passed through the scrotal septum and attached to the tubing from the left cylinder using a straight connector. The remaining tubing from the pump is then connected to the right cylinder using a straight connector in the midscrotal area as well. Passing the tubing through the scrotal septum will help prevent migration and twisting of the pump. The scrotal wound is then closed with two layers of running 3-0 chromic in the subcutaneous tissue and interrupted 3-0 chromic on the skin. Two-Piece Inflatable Prothesis: Scrotal Approach The cylinders and rear tip extenders are inserted into the corporal bodies as described for the three piece inflatable prosthesis using the scrotal approach. With a Mark II prosthesis, a pocket is then developed in the midscrotum between the testicles for placement of the pump. This location is necessary because of the relatively large size of the pump. With the Ambicor two-piece prothesis, the pump may be placed in subcutaneous tissues of either the left or the right scrotum. Unitary Penile Prosthesis A ventral penile incision is made in the midline on the proximal half of the penile shaft just distal to the penoscrotal junction. The incision is carried down to identify the corpus spongiosum. The foreskin of the penis at the most distal portion of this incision is retracted distally using a vein retractor ( Fig. 70-3). The areolar tissue is dissected from the tunica albuginea of the corpus spongiosum at the distal two-thirds of the penile shaft. Two 3-0 Proline stay sutures are used to identify the edges of the corporotomy on either side of the corpus spongiosum, and the erectile bodies are opened, dilated, and measured as for the three-piece inflatable prothesis. With the unitary hydraulic prosthesis (Dynaflex), the Furlow introducer is used to introduce the cylinder with its rear tip extenders into the corporal body. For semirigid mechanical or malleable prostheses, the vein retractor lifts the distal edge of the corporotomy over the end of the cylinder ( Fig. 70-4). The corporotomy is closed with a running 3-0 PDS or Maxon suture. The subcutaneous tissue is closed with one layer of running 3-0 chromic, and the skin is closed with interrupted 3-0 chromic suture.

FIG. 70-3. Ventral penile incision for unitary prosthesis. Proximal skin incision and distal corporotomies.

FIG. 70-4. Vein retractor lifting distal end of corporotomy over end of cylinder.

OUTCOME
Complications The most dreaded problem following prosthesis placement is an infection in contact with the device. In most circumstances conservative therapy with high doses of systemic antibiotics has not been effective. The entire device should be removed, and one may return in 6 months after the infection has cleared and the wound healed to place another device if the patient wishes. The penis will be noticeably shorter (1 to 2 inches), and reimplantation will be more difficult in the face of significant corporal scarring. An alternative approach is to remove all foreign material, irrigate the wound copiously with antibacterial solutions, and replace a new prosthesis at the same sitting, a concept termed a salvage procedure. This option is gaining popularity, and success rates approaching 90% have been reported. 1 Decreased blood supply, location of prosthesis parts close to the skin surface, weakened tissue planes, and the presence of scar tissue may contribute to penile curvature or the migration of prosthesis parts from their normal location. This may lead to erosion through the penile and scrotal skin, corporal body hernias, hypermobile glans, or other complications. As physicians become more experienced in implantation technique, these problems have become less common, although they have not been eliminated. Results Mechanical reliability of the various devices has improved as manufacturers have reinforced or eliminated areas that have tended to wear. The repair rate of most penile prosthesis models is within the range of 5% to 10% in the first 5 years. 6 By 12 to 15 years, most hydraulic models would need replacement. The life of these devices may be prolonged by avoiding pumping to pressures beyond those needed to provide adequate rigidity to the penis. The malleable semirigid rods have had fewer mechanical problems than the hydraulic devices. Patient's and partner's satisfaction with penile implants has been the greatest of all the treatments for erectile disfunction. Eighty percent of patients seemed pleased with the results. 2 The fact that the outcome, the patient's erection, is now predictable and reliable provides a great boost to his confidence and his willingness to socialize. It has resulted in many positive personality changes and helped cement many marital relationships. CHAPTER REFERENCES
1. Brant MD, Ludlow JK, Mulcahy JJ. The prosthesis salvage operation: immediate replacement of the infected penile prosthesis. J Urol 1996;55:155. 2. Fallen B, Ghanen H. Sexual performance and satisfaction with penile prostheses in the impotence of various etiologies. Int J Implant Res 1990;2:35. 3. Janssen O, Sarramon JP, Rischmann P, Bennis S, Malavund B. Microsurgical arterioarterial and arteriovenous penile revascularization in patients with pure arteriogenic impotence. Br J Urol 1994;73:561. 4. Kinsey AC, Pomesoy WB, Martin CE. Sexual behavior in the human male. Philadelphia: WB Saunders, 1948. 5. Lewis RW. Venous ligation surgery for venous leakage. Int J Impot Res 1990;2:1. 6. Lewis RW, McLaren R. Reoperation for penile prosthesis implantation problems. Urology 1993;7:381.

Chapter 71 Penile Venous Surgery Glenn’s Urologic Surgery

Chapter 71 Penile Venous Surgery
Mark R. Licht and Ronald W. Lewis

M. R. Licht: Section of Sexual Dysfunction and Prosthetic Surgery, Department of Urology, The Cleveland Clinic Florida, Fort Lauderdale, Florida 33309. R. W. Lewis: Section of Urology, Medical College of Georgia, Augusta, Georgia 30912.

Diagnosis Indications for Surgery Alternative Therapy Description of Procedure Outcomes Complications Results Chapter References

Although the exact incidence of vasculogenic impotence is not known, arterial insufficiency and/or venous leakage of varying etiology probably account for the majority of cases. Sustaining a rigid erection depends on both adequate perfusion pressure of the erectile bodies via arterial inflow and maintenance of intracavernosal pressure by increased venous outflow resistance. The trapping of blood within the expanding corporal bodies during erection by direct compression of subtunical venules as they exit through the tunica albuginea is known as the corporal veno-occlusion mechanism. Venous leak impotence refers to the inability of an individual to maintain a rigid erection because of abnormal venous outflow from the corpora cavernosa secondary to failure of the corporal veno-occlusive mechanism. In this chapter, we discuss the diagnosis of venogenic impotence and detail the surgical correction of this form of erectile dysfunction by penile venous dissection and ligation.

DIAGNOSIS
A history and a physical examination help to identify patients who may have venous leak impotence. Patients present with either complete loss of erection, decreased penile rigidity, or rapid loss of erection during intercourse. Medication effect, significant cardiovascular disease, psychological disorders, and tobacco use should be excluded as contributing causes of impotence. Penile trauma, surgery for priapism, Peyronie's disease, and previous endoscopic incision of urethral strictures can all lead to focal defects in the corporal veno-occlusive mechanism. 7 The first diagnostic test that we employ in the diagnosis of venogenic impotence is penile duplex Doppler ultrasonography. This test allows for the evaluation of both penile arterial inflow and veno-occlusion as well as the erectile response to the intracavernosal injection of a vasodilating agent. Patients with measured end-diastolic velocities over 3 cm/sec for up to 15 to 20 minutes after the administration of the vasodilating agent despite normal arterial inflow measured as peak systolic velocities greater than 30 cm/sec, may have venogenic impotence. 3 Patients who obtain a full rigid erection within 10 minutes of injection that lasts for 30 minutes probably have no clinically significant vascular disease. Patients who achieve tumescence only or who rapidly obtain a rigid erection that dissipates within 15 to 20 minutes are suspected of having a venous leak. Infusion pharmacocavernosometry and cavernosography is the definitive test for diagnosing venous leak impotence and visualizing the sites of leakage. Knowledge of the anatomy of the venous drainage of the penis is critical in interpreting this study ( Fig. 71-1). The technique we employ has been described in detail elsewhere. 6 After injection of a vasodilating agent, need for a flow rate of >50 ml/min of saline to maintain a rigid erection at an intracorporeal pressure of at least 90 mm Hg is indicative of corporal veno-occlusive dysfunction. A maintenance flow rate between 30 and 50 ml/min is considered to be borderline for venous leakage. Contrast material is infused into the corpora at the defined maintenance flow rate, and then AP and oblique radiographs of the penis are taken. The most common sites of venous leakage seen in patients with veno-occlusive dysfunction are the deep dorsal vein, the cavernosal veins, and the circumflex veins at the base of the penis.

FIG. 71-1. Penile venous anatomy.

INDICATIONS FOR SURGERY
Patients must meet strict criteria to be selected for venous surgery. Candidates must first have a history that is consistent with venous leak impotence, corroborated by duplex Doppler sonography and intracavernosal test injection findings. Other causes of impotence should be ruled out. Normal penile arterial inflow must also be documented in response to an intracavernosal injection agent because patients with concomitant arterial disease often have a poor outcome after venous surgery. Cavernosometry and cavernosography must confirm veno-occlusive dysfunction and outline the sites of leakage. Patients should have no medical contraindications to surgery. There is no strict age limit, but we prefer to perform venous surgery on patients less than 65 years old. Patients need to eliminate all tobacco use at least 6 months before surgery. Finally, patients must select venous surgery after presentation of alternative forms of therapy and discussion of expected success rates. We also perform venous surgery in conjunction with penile arterial revascularization in select patients with both focal arterial disease and veno-occlusive dysfunction. 8

ALTERNATIVE THERAPY
Patients with venous leak impotence have some effective surgical and nonsurgical options to consider along with penile venous dissection and ligation. Vacuum erection devices create an adequate erection by drawing blood into the penis with negative pressure in the vacuum tube. A constriction band placed at the base of the penis prevents outflow of blood and substitutes for a faulty veno-occlusive mechanism. Self-injection of intracavernosal vasodilating agents at high doses can often produce a functional erection in patients with mild to moderate venous leak. Patients with severe leakage, however, will most often not respond to injection therapy. Combining self-injection with a constriction band is also helpful in maintaining an erection in some patients. Implantation of a penile prosthesis effectively replaces the natural veno-occlusive mechanism with a mechanical device capable of producing a sufficiently rigid erection. A goal-directed approach is used to help the patient select an appropriate form of therapy.

DESCRIPTION OF PROCEDURE
Patients are admitted to the hospital on the day of surgery. They receive a dose of intravenous cephalosporin antibiotic 1 hour before surgery. Surgery is performed under either general intubated or spinal anesthesia. Patients are positioned supine with legs abducted to allow easy access to the perineum. If crural ligation and banding are planned as part of the operative procedure, then a dorsal lithotomy position is preferred. A lighted suction device can facilitate illumination of the deep

infrapubic dissection. An intraoperative Doppler probe can be helpful in localizing small arteries in this region. We do not routinely use optical magnification for the dissection. The operative field is prepped and draped in a sterile fashion from the umbilicus to the perineum, and an 18-Fr Foley catheter is placed for the purpose of bladder drainage and improved urethral identification. An infrapubic curvilinear anterior peripenile scrotal incision is made with a #15 blade ( Fig. 71-2). The superior extent of the incision is the inferior border of the pubis, and the inferior extent is the median raphe of the scrotum below the penile shaft. Superficial tissue is dissected free of the corporal bodies with sharp and blunt dissection. Communicating veins joining the deep and superficial drainage systems are isolated, ligated with 3-0 plain gut suture on a reel, and divided. The penile skin is then stripped away from the shaft, and the penis is inverted into the wound to gain exposure to the superficial and deep venous systems ( Fig. 71-3). Any other venous trunks of the superficial system that receive tributaries from the corpora are ligated with absorbable suture material and divided at this time ( Fig. 71-4).

FIG. 71-2. Peripenile scrotal incision for penile venous surgery.

FIG. 71-3. Inversion of the penile shaft into the wound.

FIG. 71-4. Ligation of superficial veins with connections to the corpora.

A 19-gauge butterfly needle is placed into the base of the corpus cavernosum and fixed in place to the tunica albuginea with a 3-0 chromic suture ( Fig. 71-5). The cavernosal tissue receives an injection of 30 mg of papaverine, followed 10 minutes later by indigo-carmine-colored saline (12 ml in 250 ml of saline) to help visualize abnormally effluxing veins. The butterfly needle tubing is clamped for the duration of the procedure and is used again after the dissection to perform intraoperative cavernosometry. A $-inch Penrose drain is looped around the penile shaft between the corpora. The penile skin near the glans is clamped to the Mayo stand to retract the penile skin, elongate and stabilize the penile shaft, and afford exposure for the proximal dissection. The superficial fundiform ligament is identified at the base of the penis and is divided to expose the suspensory ligament. The suspensory ligament is then sharply divided close to the underside of the pubic symphysis ( Fig. 71-6). The suspensory ligament must be completely taken down to expose the deep infrapubic region. Care is taken to identify and divide small veins emanating from the underside of the pubis and joining the superficial drainage system as well as veins perforating Buck's fascia to connect the deep and superficial systems at this level. Failure to ligate these vessels can lead to significant bleeding, which can be difficult to control and can obscure exposure for the proximal portion of the venous dissection.

FIG. 71-5. Placement of a butterfly needle for intraoperative cavernosometry.

FIG. 71-6. The suspensory ligament is divided to expose the base of the penis.

Deep in the infrapubic region, Buck's fascia is opened in the midline over the deep dorsal vein. The vein usually has a single large main trunk at this level. The deep dorsal vein is dissected free of the tunica albuginea, ligated with 0 silk ties, and divided ( Fig. 71-7). Inferior to the deep dorsal vein, the cavernosal veins can be identified in the penile hilum at this time. They may be divided if they are a major source of leakage. Great care is taken to preserve the cavernosal arteries and nerve trunks that lie lateral to these veins. If the deep dorsal vein is a significant source of abnormal penile drainage, then it is dissected from the infrapubic region along the penile dorsal midline under Buck's fascia distally toward the glans. It is important to stay in the midline during the dissection to avoid the laterally located dorsal arteries and nerves. Circumflex and emissary veins encountered on either side of the deep dorsal vein are ligated with 3-0 plain gut suture and divided ( Fig. 71-8). Bipolar electrocoagulation on low setting can be used to cauterize some small vessels along the shaft. Sometimes the deep dorsal vein is composed of two trunks along the penile shaft, and each must be dissected separately. Dissection continues until several fanning tributaries constitute the deep dorsal vein approximately 1 cm from the glans. Rarely, a large vein arises from the tunica albuginea and penetrates Buck's fascia to join the deep dorsal vein. Ligation of this vein creates a sinusoidal defect in the tunica, which must be closed with a 3-0 chromic figure-of-eight suture ligature. The junction between the corpora cavernosa and the spongiosum is carefully inspected as well, and circumflex veins connecting the two structures are ligated and divided.

FIG. 71-7. Division of the deep dorsal vein in the infrapubic region.

FIG. 71-8. Division of circumflex and emissary veins on both sides of the deep dorsal vein.

After vein dissection and ligation are completed, 30 mg of papaverine is injected into the corpora via the butterfly needle, and cavernosometry is performed 10 minutes later. If the abnormal draining veins have been eliminated, then a rigid erection is easily maintained at a flow of saline considerably less than 5 ml/min ( Fig. 71-9). Following this, the suspensory ligament is reapproximated with a 0 silk suture ligature between the infrapubic periosteum and the penile shaft. A #10 Jackson-Pratt fenestrated bulb suction drain is then placed in the wound with the tubing exiting via a separate stab incision lateral to the surgical incision. The subcuticular tissue is closed with a running 3-0 chromic suture, with care taken to approximate equal tissue planes to minimize the chance of scar formation resulting in fixation of the base of the penis. The skin edges are then reapproximated with a running subcuticular 3-0 Monocryl suture. The wound is covered with a standard sterile dressing, and the penis is snugly wrapped with a self-adherent Coban wrap. Care is taken to avoid glanular edema from a dressing that is too tight. The Foley catheter and the dressing are removed the day after surgery. The drain is removed as soon as drainage is negligible, usually in 24 to 48 hours. Patients are discharged from the hospital on the second or third postoperative day. They are advised against engaging in intercourse for 6 weeks.

FIG. 71-9. Cavernosometry after completion of the dissection confirms correction of the venous leak.

If the crural veins are found to be the only source of major venous leakage by cavernosography, we perform a crural banding procedure. The crura are exposed near the bulb of the urethra via a perineal incision, and a $-inch Mersilene ribbon (Ethicon, Inc.) is used to band the crura ( Fig. 71-10). Veins draining from the edge of the crura are ligated as well. Crural banding is not routinely performed at the time of deep dorsal vein ligation and is usually a secondary procedure. Another secondary

procedure that is rarely employed is spongiolysis. Via a penile scrotal incision, the corpus spongiosum is exposed and stripped away from the ventral surface of the corpora cavernosa. All communications between the two structures are ligated or coagulated.

FIG. 71-10. Crural banding to correct venous leakage from crural veins.

OUTCOMES
Complications Table 71-1 lists the complications that have been encountered after penile venous surgery. 9 Complications can be divided into immediate and long-term. Most patients experience some superficial bruising of the shaft and scrotum. Penile edema is usually moderate and resolves within 2 to 3 weeks. The incidence of penile edema has decreased since our use of a compression dressing and a closed wound drainage system. Painful nocturnal rections often occur for the first 24 to 48 hours after surgery. Infrequently, they may last for longer than 1 week. Wound infection and true hematoma rarely occur. Care taken during the infrapubic portion of the dissection can eliminate the risk of postoperative hematoma.

TABLE 71-1. Complications of penile venous surgery

Despite careful reapproximation of the suspensory ligament after the vein dissection is complete, approximately 20% of patients complain of penile shortening. The amount of perceived loss of length, however, is rarely clinically or functionally significant. Hypoesthesia or numbness of the glans or shaft of the penis is a common occurrence after surgery. Patients who report a loss of sensation often experience a diminished ability to achieve orgasm as well. In most cases, though, penile sensation returns completely within 7 to 9 months. Infrequently, wound scar contractures occur that lead to true penile tethering. In these cases, revision surgery is necessary, consisting of release of scar tissue and skin Z-plasty. Results Although individual reports of successful vein ligation procedures date back to the early 1900s, the modern era of penile venous surgery did not begin until the development of accurate and appropriate diagnostic and surgical techniques. A number of surgeons have since reported on the initial and long-term success rates for this procedure. Donatucci and Lue reported on 100 patients operated on between 1986 and 1988. 2 Forty-four patients (44%) had an excellent result, defined as a complete return of spontaneous erections rigid enough for penetration, and 24 patients (24%) noted some improvement in rigidity. All patients were followed for longer than 1 year after surgery. Knoll et al. reported a 46% excellent response to surgery in 41 patients followed for an average of 28 months. 4 Claes and Baert similarly reported a return of normal erectile function in 30 of 72 patients (42%) and a partial response in 23 others (32%). 1 Patients were all followed for more than 1 year. Lewis recently reported on 60 patients, of whom 16 (27%) initially had return of normal erections. 8 Seventeen other patients (28%) experienced improved rigidity and were able to have intercourse with the aid of intracavernosal pharmacologic injections for a combined success rate of 55%. All patients were followed for at least 2 years. Over time, 13 of the 33 patients (39%) who initially experienced a successful result later reported failure of the procedure. Kropman et al. also reported a 40% late failure rate at a mean follow up of 28 months in 10 of 20 patients who initially experienced a successful result from surgery. 5 Several different factors can account for the approximately 40% failure rate of penile venous surgery. Inability to accurately diagnose concomitant arterial disease and less extensive venous dissection probably account for many of the early patient failures in the series reported above. With the use of stricter diagnostic inclusion criteria and a more aggressive surgical approach, many of these early failures could have been avoided. The development of collateral venous circulation is the most likely cause of late failure in patients who experience immediate postoperative success. Collateralization has limited the long-term success rates of other types of venous surgery as well. A final reason for failure is that the ligation of penile veins may not address the true underlying pathologic disorder in many patients. Sinus smooth muscle disease that prohibits the expansion of the tunica albuginea and the subsequent compression of subtunical venules have been postulated as a major cause of veno-occlusive dysfunction. 10 To date, though, no practical test is available to accurately diagnose this entity. Penile venous surgery remains a reasonable surgical option for highly selected patients with venous leak impotence. CHAPTER REFERENCES
1. Claes H, Baert L. Cavernosometry and penile vein resection in corporeal incompetence: An evaluation of short-term and long-term results. Int J Impotence Res 1991;3:129. 2. Donatucci CF, Lue TF. Venous surgery: Are we kidding ourselves? In: Lue TF, ed. World book of impotence. London: Smith-Gorthdon and Co, 1992;221–227. 3. King BF, Lewis RW, McKusick MA. Radiologic evaluation of impotence. In: Bennett AH, ed. Impotence. Diagnosis and management of erectile dysfunction. Philadelphia: WB Saunders, 1994;52–91. 4. Knoll LD, Furlow WL, Benson RC. Penile venous ligation surgery for the management of cavernosal venous leakage. Urol Int 1992;49:33. 5. Kropman RF, Nijeholt AABL, Giespers AGM, Swarten J. Results of deep penile vein resection in impotence caused by venous leakage. Int J Impotence Res 1990;2:29. 6. Lewis RW. Diagnosis and management of corporal veno-occlusive dysfunction. Semin Urol 1990;8:113. 7. Lewis RW. Venogenic impotence. Diagnosis, management, and results. Probl Urol 1991;5:567. 8. Lewis RW. Venous surgery in the patient with erectile dysfunction. Atlas Urol Clin North Am 1993;1:21. 9. Petrou S, Lewis RW. Management of corporal veno-occlusive dysfunction. Urol Int 1992;49:48. 10. Wespes E, Moreira De Goes P, Sattar AA, Schulman C. Objective criteria in the long-term evaluation of penile venous surgery. J Urol 1994;152:888.

Chapter 72 Penile Arterial Reconstruction (Penile Revascularization) Glenn’s Urologic Surgery

Chapter 72 Penile Arterial Reconstruction (Penile Revascularization)
John Mulhall and Irwin Goldstein

J. Mulhall: Department of Urology, Loyola University Medical Center, Maywood, Illinois 60153. I. Goldstein: Boston Medical Center, Boston, Massachusetts 02118.

Diagnosis Indications for Surgery Operative Technique Dorsal Artery Dissection Penile Inversion Preparation of Recipient Dorsal Arteries Harvesting of the Inferior Epigastric Artery Microvascular Anastomosis Outcomes Complications Results Chapter References

Erectile dysfunction is defined as the consistent inability to obtain or maintain a penile erection satisfactory for sexual relations. Community epidemiologic studies have revealed that 52% of men age 40 to 70 years have self-reported (17%), moderate (25%), and complete (10%) forms of impotence. Nonsurgical treatment options for impotence include psychotherapeutic, hormonal, pharmacologic, and external device interventions. 4 Surgical treatment options consist primarily of penile prosthesis insertion, microvascular arterial bypass surgery, and, until recently, surgery for corporovenous occlusive dysfunction. 1 This latter surgery has been demonstrated to have poor long-term success rates and we no longer perform venous leak surgery. Penile revascularization is currently the only modality of therapy that has the potential to permanently cure patients, i.e., allow return of spontaneously developing erections without the necessity for any internal or external devices. This procedure has undergone many refinements since its first description by Michal in 1973. 5,6 Many variations have been described by workers such as Virag, Hauri, Crespo, and Goldstein. 3 We are currently utilizing one primary procedure involving anastomosis of the inferior epigastric artery to one or both dorsal arteries of the penis. The overall goal of penile revascularization surgery is to bypass obstructive arterial lesions in the hypogastric-cavernous arterial bed. The specific objective of the surgery is to increase the cavernosal arterial perfusion pressure and blood inflow in patients with vasculogenic erectile dysfunction secondary to pure arterial insufficiency. Young men, without other vascular risk factors, who have erectile dysfunction of a pure arteriogenic nature represent the ideal patient population for this procedure, and the investigation and preoperative evaluation of these patients are aimed at ensuring adequacy of hormonal status, neurologic function, and corporovenous occlusive function.

DIAGNOSIS
The diagnostic algorithm is aimed at ensuring that this operation is performed on the ideal candidate, i.e., one in whom there is erectile dysfunction purely on the basis of arterial insufficiency. All young patients with a history suggestive of trauma-associated impotence (pelvic fractures and perineal trauma) undergo a comprehensive history and physical examination. They have a routine endocrinologic evaluation to ensure adequate circulating levels of testosterone. These patients undergo a nocturnal penile tumescence test in an attempt to rule out neurogenic and psychogenic erectile dysfunction. Finally, they are required to have hemodynamic assessment by dynamic infusion cavernosometry/cavernosography (DICC), although other workers have used duplex Doppler ultrasonography for the same purpose. The criteria for the definition of pure arteriogenic erectile dysfunction are beyond the scope of this chapter and have been outlined elsewhere. 2 The purpose of the testing is to rule out corporovenous occlusive dysfunction. Following hemodynamic diagnosis, if the patient has pure arterial insufficiency, a selective internal pudendal arteriogram is performed to define the arterial anatomy and confirm the location of the obstructive lesion, which is generally in the common penile or cavernosal artery(ies). Only following this complete evaluation do we perform a penile revascularization procedure.

INDICATIONS FOR SURGERY
The success of this operation is based on the selection of the correct candidate and the microsurgical capabilities of the surgeon. To this end, we have developed a list of criteria that the patient and surgeon must meet to ensure optimum results. The criteria are as follows: The patient's history is characterized by (a) strong libido, (b) a consistent reduction in erectile rigidity during sexual activity, (c) increased erection rigidity during morning erections, (d) variable sustaining capability with the best maintenance of the rigidity during early morning erections, and (e) poor spontaneity of erections, taking much effort and excessive time to achieve the poorly rigid erectile response. Normal hormonal evaluation. Normal neurologic evaluation. Increased arterial gradients during cavernosal artery occlusion pressure determination at the time of DICC, indicative of arterial insufficiency. Normal venoocclusive parameters (flow-to-maintain values, pressure decay values) during DICC. The presence of an occlusive lesion in one or both hypogastric-cavernous arterial beds, located in the common penile artery or cavernosal artery that is amenable to distal bypass. The presence of an inferior epigastric artery of sufficient length to allow anastomosis to the dorsal artery. The presence of a communication branch(es) between the dorsal artery and the cavernosal artery distal to the occlusion that will allow the inflow of new blood flow and the development of increased intracorporal pressure. The development of surgical skills allowing strict adherence to operative microsurgical principles, namely, (a) avoidance of any mechanical trauma to the donor or recipient arteries due to twisting or excessive stretching; (b) avoidance of any thermal changes due to electrocautery, exposure to cold irrigants, or drying from exposure to air; (c) avoidance of ischemia resulting from compression or excessive adventitial dissection; and (d) meticulous microsurgical anastomotic technique. Avoidance of any penile trauma, such as may occur during masturbation or a sexual encounter, for a period of 6 weeks postoperatively to avoid disruption of the anastomosis. As this is elective surgery, options include observation and continued impotence or the insertion of penile prostheses.

OPERATIVE TECHNIQUE
As the candidates are young men, there is no necessity for preoperative testing other than that outlined above. In the operating room, the patients are placed supine on the operating table and their legs are placed in the frog-leg position. As this operation may last in excess of 6 hours, great care must be taken in the positioning and padding of the limbs, particularly the neurovascular points on the upper limbs. We have had one patient who developed a transient postoperative ulnar palsy due to protracted pressure on his medial epicondylar area. It is our policy now to instruct the anesthesiologist to move the arms around and alternate the position of the blood pressure cuff periodically throughout the procedure. Either general endotracheal or regional anesthesia may be used. It is our policy also to have the arteriograms in the operating room so that we can refer to them intraoperatively if necessary. Once in the operating room the patient's abdomen and genitalia are carefully shaved and he is given one dose of preoperative antibiotics. Once the patient is prepped and draped, a 16-Fr Foley catheter is placed using sterile technique and the bladder is drained of urine. From a technical standpoint, the operation can be divided into three stages: dorsal artery dissection, inferior epigastric artery harvesting, and microsurgical

anastomosis. Dorsal Artery Dissection A curvilinear incision is made, generally on the side opposite to the planned abdominal incision for inferior epigastric artery harvesting ( Fig. 72-1). The incision is made 2 fingerbreadths from the base of the penis, from a point opposite the ventral root of the penis, to the scrotal median raphe. This incision is carried down through the dartos layer using cautery. The advantages of this incision are that it offers (a) excellent proximal and distal exposure of the penile neurovascular bundle, (b) the ability to preserve the fundiform and suspensory ligaments, and (c) the absence of unsightly postoperative scars on the penile shaft or at the base of the penis. Use of a ring retractor with its elastic hooks maximizes operative exposure of the penis with a minimum of assistance.

FIG. 72-1. Inguinoscrotal incision.

Penile Inversion The ipsilateral tunica albuginea is subsequently identified at the midpenile shaft. With the penis stretched, blunt finger dissection along the tunica albuginea is performed in a distal direction deep to the spermatic cord structures along the lateral aspect of the penile shaft avoiding injury to the fundiform ligament. The penis is then inverted through the skin incision, with care taken to push the glans in fully ( Fig. 72-2). The penis must not be tumesced during this maneuver. If a partial erection is present, intracavernosal a-adrenergic agonist (100 µg phenylephrine) should be administered. Blunt finger dissection around the distal penile shaft enables a plane to be established between Buck's fascia and Colles' fascia, and a Penrose drain is secured in this plane.

FIG. 72-2. Penile inversion.

Preparation of Recipient Dorsal Arteries Exposure of the neurovascular bundle and, in particular, the right and left dorsal penile arteries is now performed. The arteries are usually obvious, located on either side of the deep dorsal vein. Isolation of the dorsal penile arteries for such arterial bypass surgery requires limited dissection at this time in the procedure; thus ischemic, mechanical, and thermal trauma to the dorsal penile arteries may be minimized. To avoid injurious vasospasm, topical papaverine hydrochloride irrigation is applied frequently. In this way, reservation of endothelial and smooth muscle cell morphology during dorsal artery preparation is ensured. This is very critical as the room temperature of the operating room, the use of room temperature irrigating solution, and even the skin incision can induce vasoconstriction, spasm, and possible endothelial cell damage. For intraluminal irrigation, we utilize a dilute papaverine, heparin, and electrolytic solution believed to be capable of inhibiting the early development of myointimal proliferative lesions during surgical preparation. The right and left dorsal penile arteries are identified first in the midpenile shaft. Their course is followed proximally underneath the fundiform ligament, with care being taken to leave the fundiform ligament intact. A fenestration is fashioned in the fundiform ligament proximally, usually near the junction of the fundiform and suspensory ligaments at a location where the pendulous penile shaft becomes fixed proximally ( Fig. 72-3). Blunt dissection is performed under the proximal aspect of the fundiform ligament above the pubic bone toward the external ring. This dissection enables the inferior epigastric artery to pass from its abdominal location to the appropriate location in the penis while simultaneously preserving the fundiform ligament.

FIG. 72-3. Fenestration of the fundiform ligament.

Harvesting of the Inferior Epigastric Artery Abdominal Incision The inguinoscrotal incision is temporarily closed with staples. A unilateral abdominal incision is made and can be transverse or paramedian in its disposition ( Fig. 72-4). The transverse incision provides excellent operative exposure of the inferior epigastric artery and heals with a more cosmetic scar compared to those observed

with paramedian skin incisions. The starting point of the transverse incision is approximately three-quarters of the total distance from the pubic bone to the umbilicus in the midline. It extends laterally along the skin lines for approximately 4 fingerbreadths. The rectus fascia is incised vertically. The junction between the rectus muscle and underlying preperitoneal fat is identified and the preperitoneal space is entered. The rectus muscle is reflected medially.

FIG. 72-4. Abdominal incision for harvesting epigastric artery.

Inferior Epigastric Artery Dissection The inferior epigastric artery and its two accompanying veins are located beneath the rectus muscle in the preperitoneal plane. The ring retractor is again utilized to optimize operative exposure. It is critical to harvest an inferior epigastric artery of sufficient length to prevent tension on the microvascular anastomosis. Application of topical papaverine is utilized on the inferior epigastric artery throughout the dissection. Thermal injury is avoided using low current microbipolar cautery set at the minimum level necessary for adequate coagulation and the vasa vasora are preserved by dissecting the artery en bloc with its surrounding veins and fat. Dissection of the inferior epigastric is required from its origin at the level of the external iliac artery to a point at the level of the umbilicus ( Fig. 72-5). It is at this point that the artery bifurcates. We make every effort to use the bifurcation where possible to allow anastomoses to both dorsal arteries.

FIG. 72-5. Inferior epigastric artery dissection.

Inferior Epigastric Artery Transfer The transfer route of the neoarterial inflow source is prepared from the abdominal perspective prior to transecting the vessel distally (the penile transfer route has previously been dissected) ( Fig. 72-6). The temporary scrotal staples are removed and the penis is reinverted. The internal ring on the side of the harvested artery is identified lateral to the origin of the inferior epigastric artery. Using blunt finger dissection through the inguinal canal, a long fine vascular clamp is passed through the fenestration in the fundiform ligament, the external and internal inguinal rings, and a Penrose drain is placed to protect this transfer route.

FIG. 72-6. Inferior epigastric artery transfer.

The donor vascular bundle is transected at the level of the umbilicus between two Ligaclips and is carefully inspected for any proximal bleeding points. The long fine vascular clamp is brought through the internal inguinal ring again, this time to grasp the end of the transected inferior epigastric artery. The inferior epigastric vascular bundle is transferred to the base of the penis. It should be briskly pulsating and of adequate length. The origin of the inferior epigastric artery should be inspected for kinking or twisting. Following the achievement of complete hemostasis, closure of the abdominal wound is performed in two layers. The rectus fascia is closed utilizing a running 0 polyglycolic acid suture, one suture started at either end of the incision. The skin edges are apposed using skin staples. Microvascular Anastomosis Vessel Preparation A ring retractor and the associated elastic hooks are utilized once again on the inguinoscrotal incision and the fenestration of the fundiform ligament to gain exposure of the proximal dorsal neurovascular bundle. The pulsating inferior epigastric artery is placed against the recipient dorsal penile arteries and a convenient location is selected for the vascular anastomosis. The anastomosis (or anastomoses) is(are) created based on the arteriographic findings. An end-to-side anastomosis is best under conditions whereby dorsal penile artery communications exist to the cavernous artery. Furthermore, an end-to-side anastomosis protects arterial blood flow in the distal direction to the glans penis. It has been our experience, however, that ligation of both dorsal penile arteries to perform bilateral proximal end-to-end anastomoses has not ever caused ischemic injury to the glans penis. The appropriate dorsal penile artery segment is freed from its attachments to the tunica albuginea, with care being taken to avoid injury to any communicating branches to the cavernosal artery. Vascular hemostasis of this segment of the dorsal penile artery may be achieved with either gold-plated (low-pressure) ( Fig. 72-7) aneurysm vascular clamps or vessel loops under minimal tension for the minimal of operating time. The only location where the adventitia must be carefully removed

is at the site of the vascular anastomosis, i.e., the distal end of the inferior epigastric artery and the selected region of the dorsal penile artery, in order to avoid causing subsequent thrombosis. If segments of adventitia enter the anastomosis patency of the anastomosis is in jeopardy, as adventitia activates clotting factors from the extrinsic clotting system. The remaining adventitia should be preserved in the vessels as the vasa vasorum provide a nutritional role to the vessel wall. The preservation of the adventitia is additionally important in terms of vessel innervation.

FIG. 72-7. Preparation of vessel for microvascular anastomosis.

Anastomotic Technique Under microscopic control at 5–10× magnification, a 10-0 suture (single-armed, 100-µm, 149 degree curved needle) is placed along the longitudinal axis of the dorsal penile artery in a 1-mm segment in the region of the intended anastomosis. After placing tension on the suture, an oval section of the artery wall is excised with curved microscissors resulting in a 1.2- to 1.5-mm horizontal arteriotomy ( Fig. 72-8). We use a plastic colored background material to aid in vessel visualization under the microscope. A temporary 2-Fr silastic stent is placed within the arteriotomy for clearer definition of the vessel lumen.

FIG. 72-8. Microsurgical anastomosis with sutures in place.

An end-to-side anastomosis is performed between the inferior epigastric artery and the dorsal artery using interrupted 10-0 nylon sutures under the 5-10× magnification. The sutures are placed initially at each apex of the anastomosis and then subsequently three to five interrupted sutures are placed into each side wall (Fig. 72-8). All sutures used to complete the anastomosis are inserted equidistant from each other to avoid an uneven anastomosis. One side of the anastomosis is completed prior to commencing the other side. If a temporary vascular stent is used, it is removed following placement of all sutures. The use of a temporary vascular stent enables careful inspection of the vessel back wall. Following release of the temporary occluding vascular clamps (or vessel loops) on the dorsal penile artery, the anastomosed segment should reveal arterial pulsations along its length and retrograde into the inferior epigastric artery. Such an observation implies a patent anastomosis. At this time, the inferior epigastric artery gold-plated aneurysm clamp may be removed. The intensity of the arterial pulsations in the anastomosis usually increases. Occasionally, the application of a small amount of hemostatic material may be needed to aid in promoting hemostasis from suture needle holes in the vessel walls. After complete hemostasis has been achieved and correct instrument and sponge counts are assured, closure of the inguinoscrotal incision may begin. The dartos layer is reapproximated using a 3-0 polyglycolic acid suture in a running fashion. The skin edges are closed with skin staples. The Foley catheter is left to closed-system gravity drainage overnight. Modifications of the above-described procedure may be utilized. The most common alternative arterial anastomosis is an end-to-end anastomosis between the inferior epigastric artery and the ligated proximal end of the dorsal penile artery. Depending on the site of the arterial communication from the dorsal penile artery to the cavernosal artery, the end-to-end anastomosis may also be anastomosed to the distal ligated end of the dorsal penile artery. It is also most common to anastomose the opposite dorsal penile artery to the inferior epigastric artery. This can be done with an appropriately sized distal branch of the inferior epigastric artery end-to-side as previously described.

OUTCOMES
Complications Mechanical disruption of the microvascular anastomosis and subsequent uncontrolled arterial hemorrhage may occur from blunt trauma in the first few postoperative weeks following coitus, masturbation, or from accidents. We recommend abstention from sexual activities involving the erect penis until 6 weeks postoperatively. Other complications include penile pain and diminished penile sensation from injury to the nearby dorsal nerve. 7 Loss of compliance of the suspensory and fundiform ligaments postoperatively may lead to diminished penile length. Preserving the two ligaments has markedly minimized those complications in our series. Glans hyperemia, once a complication seen when inferior epigastric artery to deep dorsal vein anastomoses (dorsal vein arterialization) were performed, is no longer seen because we no longer perform this form of anastomosis.3 Results We have reported on the objective postoperative hemodynamic status including steady-state equilibrium intracavernosal pressures, venoocclusive and arterial function testing parameters, in patients who experienced successful as well as unsuccessful clinical results following microvascular arterial bypass surgery for impotence. Of the 226 patients who underwent penile microvascular arterial bypass surgery from 1985 to 1992, 68 (30%) (mean age 34 ± 10 years) underwent both preoperative and postoperative pharmacocavernosometry/graphy. The mean duration between the bypass procedure and follow-up postoperative testing was 8 ± 6 months. Surgical bypasses in these 68 patients included 65 inferior epigastric artery to dorsal penile artery including 30 with dual dorsal arterial anastomoses. There were, in addition, 9 artery-to-deep-dorsal-vein anastomoses including 6 performed in conjunction with an arterial anastomosis. Forty-nine patients (72%) had concomitant venous surgery for venoocclusive dysfunction (46 deep dorsal vein excisions, 35 crural placation, 19 cavernosal vein ligations, and 3 spongiolyses). Twelve patients (21%) with pure arteriogenic impotence had a postoperative mean increase in steady-state equilibrium intracavernosal pressure of 25 ± 12.3 mm Hg (range 13 ± 45 mm Hg). Of the remaining 56 patients, 49 had concomitant venous surgery. In this latter group, there was no significant change in mean steady-state

equilibrium intracavernosal pressure (38 ± 20 mm Hg preoperative and 43 ± 17 mm Hg postoperative), mean pressure decay in 30 seconds (67 ± 9 mm Hg preoperative versus 64 ± 26 mm Hg postoperative) or mean flow-to maintain values (38 ± 36 ml/min preoperative versus 28 ± 28 mm Hg postoperative). The remaining 7, who did not have venous surgery, had normal venoocclusive function pre- and postoperatively; however, the postoperative intracavernosal pressure did not increase (54 ± 22 mm Hg preoperatively; 58 ± 27 mm Hg postoperatively). 3 On review of the recent literature, there has been reported a variable success rate following penile revascularization, ranging from 30% to 74%. One of the problems in interpreting these data is that there exists no standardization of the definition of success following this operation. Success has been variably defined by questionnaire, patient interview, and hemodynamic evaluation. Furthermore, the patient populations have been heterogeneous with varying numbers of patients with venogenic impotence. We no longer offer this operation to patients with venous leak. The third problem with the literature to date is the inclusion in reported series of dorsal vein arterialization procedures, which we believe to have success rates inferior to procedures utilizing pure arterial anastomoses. Finally, most of the studies have involved short periods of follow-up ranging from 16 to 36 months. The most recent study had a short-term success rate of 80% and long-term (up to 5 years) success rate of 64%, utilizing strict patient selection criteria. To evaluate this operation as a valid management strategy, prospective analysis of patient outcomes is essential. There should be a standardization of success definition, follow-up techniques, and patients need to be followed long-term postoperatively. The goal of the operation is to restore natural, spontaneously occurring erections without the aid of any internal or external means to young men with erectile dysfunction. Much research is needed to define why a significant number of men continue to fail to respond to penile revascularization. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Goldstein I. Arterial revascularization procedures. Semin Urol 1986;4:25.2. Goldstein I. Vascular diseases of the penis. In: Pollack HM, ed. Clinical urography. 3rd ed. Philadelphia: WB Saunders, 1990;2231–2252. Hatzichristou DG, Goldstein I. Arterial bypass surgery for impotence. Curr Opin Urol 1991;1. Krane RJ, Goldstein I, Saenz de Tejada I. Impotence. N Engl Med J 1989;321:1648–1659. Michal V, Kramer R, Popischal J, Hejhal L. Direct arterial anastomosis on corporal cavernosa penis in therapy of erectile dysfunction. Rozhl Chir 1973;52:587–590. Michal V, Kramer R, Popischal J. Femoropudendal bypass, internal iliac thrombendarterectomy and direct arterial anastomosis to the cavernous body in the treatment of erectile impotence. Bull Soc Int Chir 1974;33:341–345. 7. Zorgniotti AW, Lizza EF. Complications of penile revascularization. In: Zorgniotti AW, Lizza EF, eds. Diagnosis and management of impotence. Philadelphia: BC Decker, 1991.

Chapter 73 Penile Trauma Glenn’s Urologic Surgery

Chapter 73 Penile Trauma
David M. Nudell, Allen F. Morey, and Jack W. McAninch

D. M. Nudell: Department of Urology, University of California, San Francisco, California 94140. A. F. Morey: Department of Surgery (Urology Service), Uniformed Services University of the Health Sciences, and Brooke Army Medical Center, San Antonio, Texas 78258. J. W. McAninch: Department of Urology, University of California, and San Francisco General Hospital, San Francisco, California 94110. The opinions expressed herein are those of the authors and are not to be construed as reflecting the views of the Armed Forces or the Department of Defense.

Penile Rupture Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Penile Skin Loss Indications for Surgery Alternative Therapy Surgical Technique Outcomes Penetrating Penile Trauma Diagnosis Indications for Surgery and Alternative Treatments Surgical Technique Outcomes Chapter References

Trauma to the penis occurs from both blunt and penetrating injury. Such injuries present unique and difficult management problems to the urologic surgeon, including overall cosmesis, micturition, and future potency. Major blunt penile injuries include penile rupture and skin loss from strangulation or degloving injuries. Penetrating penile trauma seldom occurs in the absence of associated genital or major organ injury, except in the event of bites and self-inflicted wounds.

PENILE RUPTURE
The most common blunt injury involving the penis is rupture of the corpora cavernosa, or penile fracture. This almost invariably occurs when the erect penis is forced to bend in an irregular fashion, such as when it accidentally impinges on the pubis or perineum after slipping out of the vagina during sexual intercourse. 9 The remainder of cases are caused by falls out of bed with an erect penis, masturbation, or manipulation of the erect penis. The patient often reports a cracking or popping noise at the time of injury, leading to immediate detumescence and rapid onset of discoloration and swelling over the site of injury. Diagnosis The diagnosis of penile rupture is easily made by physical examination along with the appropriate history. Swelling and discoloration may or may not be limited to the penis, depending on the integrity of Buck's fascia. If Buck's fascia is intact, the hematoma will be contained and will not usually spread below the base of the penis, resulting in the typical “eggplant” deformity ( Fig. 73-1). However, if the laceration in the tunica albuginea involves Buck's fascia, extravasation will be contained by Colles' fascia, and ecchymosis will extend in a “butterfly” distribution over the perineum, scrotum, and lower abdomen. Examination may reveal angulation of the penis away from the side of rupture because of the mass effect of the hematoma. In addition, focal tenderness and a palpable defect in the tunica albuginea may help localize the fracture site.

FIG. 73-1. Fractured penis displaying the pathognomonic “eggplant deformity” with swelling and discoloration extending to the base of the shaft. The penis usually bends away from the side of injury because of the hematoma.

Penile rupture can occur anywhere along the shaft including the base of the penis, where the corpora are fixed by the penile suspensory ligament. Generally, only one corporal body is injured, although both corpora and the corpus spongiosum can be affected, depending on the severity of the injury. Most patients are able to urinate normally. Failure to void spontaneously may signify compression of the urethra by hematoma but should lead to evaluation of urethral injury by retrograde urethrography (RUG). Urethral injury occurs in approximately 20% of cases and usually consists of partial disruption, although complete transection can result. 8 Retrograde urethrography is mandatory in all patients with blood at the urethral meatus, hematuria of any extent, or inability to void. 9 However, because RUG is easy to perform and provides reliable results, we perform it routinely in all cases of suspected penile rupture. Imaging of the corporal bodies is generally not necessary because prompt surgical exploration is usually advised when penile rupture is suspected. Although magnetic resonance imaging has recently been shown to be highly sensitive in the diagnosis of rupture, cost and inaccessibility limit routine use of MRI in this setting. 7,9 Indications for Surgery Although penile fractures can be managed nonoperatively, the literature shows a clear advantage to early operative repair. evacuation of hematoma and primary repair of the laceration. Alternative Therapy Conservative treatment consists of cool compression dressings, anti-inflammatory agents, and sedatives to reduce erections. Surgical Technique
8,9

The goals of acute exploration are

The patient is placed in a supine position, and a Foley catheter is placed to facilitate identification of the urethra and urinary drainage. Exposure is usually obtained through a subcoronal circumferential incision, and the penile skin is degloved down to the base. The distal circumferential incision is favored because it allows both exposure of the ruptured corpus and adequate assessment of the contralateral corpus and corpus spongiosum. The corpus spongiosum is carefully inspected to evaluate for potential urethral injury. Inspection of the fracture site usually reveals a transverse laceration, between 0.5 and 2.0 cm long, in the tunica albuginea of the proximal penile shaft. 4 After evacuation of the hematoma and irrigation, minimal debridement of nonviable wound edges may be necessary before closure with interrupted 4-0 Maxon sutures (Fig. 73-2). The surgeon should not probe the exposed cavernous tissue unnecessarily, as this may elicit troublesome bleeding. A tourniquet may be used intraoperatively to control hemorrhage. Lacerations may run directly under the dorsal neurovascular bundle located on the dorsal surface of the corpora at approximately the 10- and 2-o'clock positions ( Fig. 73-3). This necessitates careful dissection of these structures off the corpora to allow a safe, watertight closure. Division of the deep dorsal vein facilitates unilateral dissection of the neurovascular bundle off the underlying corpus cavernosum. The penile skin is then replaced, and the subcoronal incision is closed with interrupted 4-0 chromic sutures. Postoperatively, a loose compression dressing (Coban) is gently placed, and the urethral catheter may be removed on the first postoperative day. Systemic antibiotics, anti-inflammatory agents, and fibrinolytics are unnecessary. Most patients can be discharged home within 1 to 2 days of surgery. Sexual activity can be resumed at about 4 to 6 weeks.

FIG. 73-2. Identification and repair of penile fracture. A distal circumferential subcoronal incision is made, and skin and soft tissue are mobilized off the underlying corporal bodies down to the base of the penis. This maneuver exposes the transverse laceration in the tunica albuginea. The laceration is repaired using interrupted 4-0 Maxon with the knots buried. Exposed corporal erectile tissue should not be probed or explored, as this may cause troublesome bleeding.

FIG. 73-3. Penile fracture extending beneath dorsal neurovascular bundles. Elevation of the ipsilateral dorsal neurovascular bundle facilitates repair of lacerations and protects these structures from inadvertent injury. Division of the deep dorsal vein in the midline provides access to the correct surgical plane beneath the ipsilateral neurovascular bundle.

When urethral transection occurs in the context of penile rupture, we advocate primary repair with interrupted 5-0 or 6-0 Maxon sutures over a 16-Fr silicone catheter. In cases of complete urethral transection, additional urinary diversion through a percutaneous suprapubic cystostomy tube may be prudent. 9 Outcomes Complications Many patients treated conservatively or with delayed repair have some form of sexual dysfunction, such as painful erection, curvature, or even impotence secondary to cavernous–venous occlusive disease. 1,10 Results Patients who have operative repair within 48 hours of the injury have excellent functional results. In two relatively large studies, none of the patients with early operative repair experienced impotence, penile curvature on erection, or pain at coitus. 4,9

PENILE SKIN LOSS
Penile skin loss can occur from infection, burns, constrictive bands, or degloving injuries from blunt or penetrating trauma. When the skin loss is secondary to infection, repeated debridement with antibiotics and moist dressing changes must be instituted to prepare the underlying tissue for delayed reconstruction. Avulsions are most often caused by power tool injuries or motor vehicle accidents. 1 Immediate repair is frequently possible in cases of traumatic skin loss. Indications for Surgery Partial penile skin loss, especially in the distal shaft, is best managed by rotational mobilization of a local skin flap. Extensive skin loss, whether from the injury itself or surgical debridement, usually requires tissue transfer for repair. In impotent patients, the penis can be buried under a scrotal flap with the glans left exposed to allow micturition (Fig. 73-4).3 In sexually active patients, a thick (0.016 to 0.018 inch), nonmeshed split-thickness skin graft is used. Thick split-thickness grafts are preferred because they are not hair-bearing, have minimal contraction, and offer excellent cosmesis and viability. In impotent patients, a thinner or meshed split-thickness graft, despite a higher rate of contraction, may be preferred to increase the chance of graft survival. Avulsed skin that is still attached on a viable pedicle can be gently washed and reapplied with the knowledge that it may need to be debrided at a later time. Skin that is no longer attached is usually unsuitable for free tissue graft. 6

FIG. 73-4. Scrotal tunnel maneuver for penile skin coverage. The penis shaft may be buried beneath a flap of scrotal skin to provide skin coverage, leaving the glans exposed. This is a viable option in older patients who either were impotent before their injury or who have sustained severe associated injuries.

Alternative Therapy There are no alternatives to surgery. Surgical Technique The patient is placed supine, and both the genitals and a carefully chosen donor site are prepped into the field. The anterolateral thigh provides thickness, texture, and color resembling penile skin and is therefore the preferred donor site. A Foley catheter is placed to prevent postoperative urinary contamination. The shaved donor site is coated with sterile mineral oil, and a Brown or Paget dermatome (10 cm wide strip) is used to harvest the graft in standard fashion. Once the graft is harvested, the donor site is covered with fine mesh gauze under gentle pressure. The penile wound is sharply debrided of all devitalized tissue and any chronic granulation tissue. It is imperative to prepare the recipient site so that the graft will have adequate blood supply. Debridement of the glans should be avoided, but penile skin left distal to the site of planned skin grafting should be excised up to the coronal sulcus to prevent chronic lymphostasis in this area. Hemostasis at the graft site is essential to prevent hematoma formation under the graft. Once prepared, the penis is stretched, and the graft applied circumferentially such that the seam is in the midline ventrally, providing the cosmetic appearance of a midline raphe ( Fig. 73-5). Chordee formation has generally not been a problem because the graft will have minimal longitudinal contraction. The graft is secured to itself and along the shaft with interrupted 5-0 chromic sutures. The proximal and distal margins are secured with 4-0 silk sutures, some of which are left long to fashion a bolster dressing. A Xeroform dressing is placed directly on the graft, followed by cotton soaked in mineral oil and fluffs. The whole dressing is secured in place using the bolster sutures, leaving the glans visible for inspection. To keep the penis in a vertical position, a splint is placed around the bolstered dressing.

FIG. 73-5. Penile split-thickness skin graft. A thick (0.016 to 0.018 inch) split-thickness skin graft is applied to the denuded penile shaft. The distal skin is discarded to just beneath the corona when a circumferential graft is indicated. The graft is placed with the seam in the midline ventrally and secured with 5-0 chromic sutures to itself and along the shaft, while 4-0 silk sutures placed proximally and distally are left long to secure a bolster dressing.

Postoperatively, the patient is kept at strict bed rest until the dressing is removed, usually after 5 days, when the Foley catheter is also removed. The operatively placed fine mesh gauze is allowed to separate from the donor site wound on its own, which usually occurs within 7 to 10 days. Once the penile dressing is removed, twice-daily sitz baths can be started. We do not discourage erections, as they provide natural tissue expansion and may reduce the rate of graft contracture. Outcomes Complications The common causes of early failure are infection, shearing forces during the critical first 5 postoperative days, and underlying hematoma. It is imperative that the graft bed be free of infected granulation tissue and any necrotic tissue. Shearing forces disrupt the blood supply to the new graft and are prevented by the penile splint, provided the patient is cooperative with strict bed rest for 5 days. Hematoma causes failure by creating poor contact between the graft and the recipient bed. Meshed grafts allow better dissipation of hematoma fluid but are discouraged in potent patients because of their increased degree of contraction. Long-term results of reconstruction have been excellent. Patients may lose superficial sensation but maintain deep touch sensation and potency. Results Graft take exceeds 90%. Most patients have satisfactory intercourse after reconstruction. Cosmetic and functional results of nonmeshed, thick split-thickness penile grafts are superior.

PENETRATING PENILE TRAUMA
Penetrating trauma to the penis is most often caused by firearms but can also result from stab wounds, industrial accidents, self-mutilation attempts, and bites. In all cases, general principles of management include hemo-stasis, judicious wound debridement, and assessment of urethral and corporal injury. Most civilian penile gunshot wounds are caused by low-velocity missiles, which cause damage only in the path of the bullet. Associated wounds of the thigh and pelvis are common. Diagnosis Genital injury is determined by careful physical examination, with special attention paid to the trajectory of the bullet and initial hemostasis. The exam should include a vascular (glanular capillary refill) and penile sensory assessment. 7 Urethral injury, which occurs in 25% to 40% of penetrating injuries to the penis, should be excluded with retrograde urethrography (RUG) in all cases. 5 The triad of no blood at the meatus, absence of hematuria, and normal voiding suggests that there is no urethral injury; however, penetrating trauma can cause urethral injury without clinical signs of damage. Cystography, intravenous pyelography, and scrotal ultrasonography may be necessary to evaluate associated urologic injuries. Indications for Surgery and Alternative Treatments

Penetrating injury to the penis most often requires surgical exploration. The exceptions are single pellet wounds with small entrance sites and superficial stab wounds in which there is no active bleeding or hematoma. 2 In addition, patients with unstable major organ injury will be unable to undergo immediate exploration. In these cases, initial treatment consists of hemostasis and packing of major wounds. Penetrating injury causing major skin loss will require tissue transfer for satisfactory coverage, but associated corporal and urethral injuries must be repaired before the skin grafting. Immediate primary closure or reconstruction should take place only with a clean wound that is generally less than 8 hours old. Surgical Technique Operation consists of judicious debridement of devitalized tissue and hemostasis. The wound must be copiously irrigated to remove all foreign bodies, including powder from shotgun pellets and pieces of clothing. Bleeding almost always occurs from a lacerated corporal body but may also be from disrupted superficial veins. The corpora are well vascularized, and extensive debridement is usually unnecessary and will hinder future potency. Hemostasis is obtained by gentle compression and watertight closure of the tunica albuginea alone, usually with interrupted 4-0 Maxon sutures. Deep sutures or clamping within the corpora is discouraged, as delicate erectile tissue is damaged and bleeding is usually exacerbated. Urethral injuries are repaired with 5-0 Vicryl suture over a silicone catheter. Devitalized urethra must be carefully debrided, and primary repair with a tension-free anastamosis can usually be accomplished. Associated scrotal and spermatic cord injuries are treated with debridement and, if necessary, orchiectomy or ligation of the vas deferens. The skin can be closed primarily unless viable skin edges cannot be approximated. In contaminated wounds or those encountered after 8 hours, immediate skin closure or grafting is not recommended, and the wound is packed instead. Once the wound is clean, delayed primary closure, staged reconstruction, or healing by secondary intention may be selected. Penile bites deserve special mention, as they can rapidly progress to severe infection. Wounds should be copiously irrigated, and all devitalized tissue debrided. All wounds should be left open, and prophylactic antibiotics administered. Antibiotic treatment should cover gram-positive and gram-negative organisms as well as anaerobic gram-negative rods. Hospitalization with frequent wound inspection and intravenous antibiotics is necessary in those with delayed presentation or with increased risk factors such as steroid use, diabetes, or immunodeficiency syndromes. Close follow-up is mandatory in all outpatients. Outcomes Complications Early complications of penetrating penile trauma include rebleeding and infection. Because the corpora are heavily vascularized, breakdown of repair in the tunica albuginea is rare. A small minority will report superficial sensory loss, pain with erection, and rapid detumescence. Results Excellent functional results can be expected except in those cases of high-velocity injuries where massive tissue destruction has occurred. Most patients report retained potency without penile curvature and with satisfactory cosmetic results. 3,6 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Armenakas NA, McAninch JW. Use of skin grafts in external genital reconstruction. In: McAninch JW, ed. New techniques in reconstructive urology. New York: Igaku-Shoin, 1996;127–141. Goldman HB, Dmochowski RR, Cox CE. Penetrating trauma to the penis: functional results. J Urol 1996;155:551. Gomez RG. Genital skin loss: reconstructive techniques. In: McAninch JW, ed. Problems in urology. Philadelphia: JB Lippincott, 1994;290–301. Gomez RG. Genital injuries: presentation and management. In: McAninch JW, ed. Problems in urology. Philadelphia: JB Lippincott, 1994;279–289. Gomez RG, Castanheira AC, McAninch JW. Gunshot wounds to the male external genitalia. J Urol 1993;150:1147. McAninch JW. Management of genital skin loss. Urol Clin North Am 1989;16:387. Miller KS, McAninch JW. Penile fracture and soft tissue injury. In: McAninch JW, ed. Traumatic and reconstructive urology. Philadelphia: WB Saunders, 1996;693–698. Nicolaisen GS, Melamud A, Williams RD, McAninch JW. Rupture of the corpus cavernosum: surgical management. J Urol 1983;130:917. Orvis BR, McAninch JW. Penile rupture. Urol Clin North Am 1989;16:369. Volz LR, Broderick GA. Conservative management of penile fracture may cause cavernous-venous occlusive disease and permanent erectile dysfunction. J Urol 1994;151(5):358A.

Chapter 74 Penile Replantation Glenn’s Urologic Surgery

Chapter 74 Penile Replantation
Farhad Parivar, Allen F. Morey, and Jack W. McAninch

F. Parivar: Department of Urology, University of California, San Francisco, California 94140. A. F. Morey: Department of Surgery, Uniformed Services University of the Health Sciences, and Brooke Army Medical Center, San Antonio, Texas 78258. J. W. McAninch: Department of Urology, University of California, and San Francisco General Hospital, San Francisco, California 94110. The opinions expressed herein are those of the authors and are not to be construed as reflecting the views of the Armed Forces or the Department of Defense.

Diagnosis Indications for Surgery Surgical Technique Outcomes Complications Results Chapter References

Traumatic amputation of the penis is an uncommon injury. Its incidence is not well known, as it is believed to be underreported. There are approximately 100 reported cases of injury in the English literature. By far the most common cause of injury is genital self-mutilation in men with psychiatric disorders. The incidence of psychiatric disease has been reported to be 65% to 87%. 1,2,3,4 and 5 Patients with a history of substance abuse or those with gender misidentification (transvestites, homosexuals, and transsexuals) constitute the remainder of the population. Other causes of genital mutilation include accidental injuries such as dog bites, ballistic wounds, and industrial injuries. Because of the nature of the injury, these usually carry a worse prognosis than self-inflicted wounds. Partial amputations in general have better surgical outcome than complete amputations.

DIAGNOSIS
The diagnosis of penile amputation is self-evident. No radiologic study is necessary to evaluate the urethra or penile vasculature. A plain x-ray of genitalia may be necessary in patients with shrapnel or shotgun blast injuries to evaluate for the presence of foreign bodies.

INDICATIONS FOR SURGERY
Because of the reported good cosmetic and functional results of reconstructive surgery, attempt at replantation is recommended in all cases of penile amputation. There is no consensus on the timing of surgery after complete amputation. The majority of reported series have not documented the time elapsed between injury and surgery, but replantation up to 8 hours after amputation has been reported. 3 The amputated distal penis should be wrapped in a saline/water-soaked gauze and placed in a plastic bag, which is in turn placed in a second ice-containing plastic bag and transported to the hospital as soon as possible (“bag-in-a-bag” technique).

SURGICAL TECHNIQUE
Following partial or complete amputation of the penis, there is often profuse blood loss, and therefore blood should be obtained for cross-match. General or continuous regional anesthesia is necessary. For the surgeon, optical magnification and familiarity with microvascular techniques are mandatory. Broad-spectrum antibiotics and tetanus immunization are instituted. After preparation of the surgical field, a tourniquet is placed on the penile stump to prevent further blood loss. The overlying blood clot is then removed. It is usually at this stage that a thorough examination can be carried out and the extent of injury assessed. Dorsal arteries, veins, and nerves are identified at this stage and tagged with bulldogs and sutures ( Fig. 74-1). The amputated part is separately cleaned with saline and appropriate antiseptic solution and examined on the field. The dorsal arteries, veins, and nerves are similarly identified.

FIG. 74-1. The replantation procedure begins with identification and isolation of the neurovascular structures, corpora, and urethral mucosa on the penile stump.

A 16-Fr silastic Foley catheter is passed per meatus and through the stump into the bladder ( Fig. 74-2). Urethral anastomosis is first carried out over this Foley. We prefer an end-to-end anastomosis using 10 to 12 sutures of 5-0 polydeoxanon (Maxon). The cut edges may need to be trimmed, and some urethral mobilization may be necessary to achieve a tension-free anastomosis. When possible, the corpus spongiosum should be closed with interrupted sutures of 5-0 Maxon as a second layer. However, these injuries usually occur in the region of pendulous urethra where a one-layer urethral anastomosis is more practical. The corpora cavernosa are next sutured with interrupted sutures of 3-0 polyglycolic acid (Dexon/Vicryl) starting from the septum. It is not necessary to anastomose the central cavernous arteries. The deep dorsal vein and artery are next anastomosed under magnification using interrupted sutures of 9-0 nylon. For best results, at least one vein and artery should be anastomosed to provide both adequate blood supply and venous drainage of the distal segment. If a large segment of the artery or vein is missing, vascular interposition using saphenous vein will be necessary. Use of synthetic grafts is not recommended. The dorsal nerve is finally anastomosed using sutures of 9-0 nylon. The Buck's fascia is closed with interrupted sutures of 3-0 chromic, and penile skin with interrupted sutures of 4-0 chromic.

FIG. 74-2. A 16-Fr Foley catheter is placed through the urethral meatus and then through the stump into the bladder. Repair of the urethra with interrupted 5-0 Maxon sutures is the first anastomosis to be completed.

If penile skin is inadequate for closure, the penile shaft may be temporarily covered with vaselinated gauze in preparation for skin grafting at a later stage. Furthermore, if there is inadequate tissue to cover the vascular anastomoses, the denuded shaft may be temporarily buried under the scrotal skin in the midline. The bladder is then filled with saline through the Foley, and a percutaneous suprapubic tube is inserted for urinary diversion until the Foley is removed in 3 weeks. At the end of the procedure, the penis needs to be elevated and immobilized to enhance venous and lymphatic drainage ( Fig. 74-3). This can be accomplished by wrapping the penis in loose gauze and placing it in a plastic irrigation fluid bottle cut and well padded on both ends. The glans and distal penis should be inspected regularly for evidence of venous congestion and skin loss. Five days of bed rest is prescribed, during which patients should have some form of prophylaxis against thromboembolism.

FIG. 74-3. The reconstructed phallus is immobilized in a plastic housing that allows inspection of the glans. Urethral and suprapubic catheters are placed.

OUTCOMES
Complications Common to most reports of penile replantation is the high incidence of partial or complete skin necrosis. This in turn is a function of extent of injury, delay in repair, and success of vascular anasomosis. Restoration of venous return is extremely important, as venous congestion of the distal penis may lead to eventual necrosis and graft loss. Loss of distal cutaneous sensation is invariable if dorsal nerve is not anastomosed. Urethral stricture and urethrocutaneous fistula have been reported in some of the earlier cases, but in most of these, urethral anastomosis was done using chromic sutures. We believe stronger, longer-lasting, monfilament sutures such as Maxon or PDS facilitate better results. Reduction in penile rigidity may require subsequent intervention. Results In general, all patients with penile amputation should be considered for replantation unless the distal segment is completely mutilated. Microvascular techniques attain the best results, but even simple reimplantation results in adequate cosmetic and functional restoration of the penis in a majority of cases. 4,6 Nearly all patients undergoing either form of reconstruction are capable of subsequent intromission. Surgical repair is thus a worthwhile exercise whenever possible. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Aboseif S, Gomez R, McAninch J. Genital self-mutilation. J Urol 1993;150:1143. Bhanganada K, Chayavatana T, Pongnumkul C, et al. Surgical management of an epidemic of penile amputation in Siam. Am J Surg 1983;146:376. Bux R, Carroll P, Berger M, Yarbrough W. Primary penile reanastomosis. Urology 1978;11:500. Carroll PR, Lue TF, Schmidt RA, Trengrove-Jones G, McAninch JW. Penile replantation: Current concepts. J Urol 1985;133:281. Greilsheimer H, Groves JE. Male genital self-mutilation. Arch Gen Psychiatry 1979;36:441. Heymann AD, Bell-Thomson J, Rathod DM, Heller LE. Successful reimplantation of the penis using microvascular techniques. J Urol 1977;118:879.

Chapter 75 Hydrocele and Spermatocele Glenn’s Urologic Surgery

Chapter 75 Hydrocele and Spermatocele
Theodros Yohannes and James I. Harty

T. Yohannes and J. I. Harty: Division of Urology, University of Louisville School of Medicine, Louisville, Kentucky 40292.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique: Hydrocele Lord Procedure Jaboulay Procedure Surgical Technique: Spermatocele Outcomes Complications Results Chapter References

A hydrocele is an abnormal accumulation of serous fluid between the layers of the tunica vaginalis. Embryologically, the tunica vaginalis is an extension of the peritoneal sac that has a serous surface that secretes and absorbs fluid. The fluid collection that occurs is thought to be due to an imbalance between production and absorption within the layers of the tunica vaginalis. Hydroceles may be congenital or acquired. Congenital hydroceles result from persistence of the processus vaginalis (peritoneal sac) and are treated by ligation of the sac at the internal inguinal ring through an inguinal incision if they persist after 1 year of life. Most acquired hydroceles are idiopathic, although some are related to trauma, infection, testicular tumors, or inguinal operations that cause lymphatic obstruction, such as ipsilateral renal transplantation. A spermatocele is a cystic structure that usually arises from the head of the epididymis and rete testis. It is usually filled with a milky fluid that contains spermatozoa. Spermatoceles are usually located on the superior aspect of the testicle but can occur anywhere on the epididymis. Although the exact etiology is unknown, obstruction and trauma have been implicated. Spermatoceles are found most commonly in middle-aged men, and their incidence increases with age.

DIAGNOSIS
The patient with a hydrocele usually presents with a smooth scrotal swelling, which may be painful or cause embarrassment due to its appearance. Upon examination, the swelling is confined to the scrotum distinguishing it from an inguinal hernia. A hydrocele will transilluminate to varying degrees, depending on the thickness of the wall. The testis is not usually palpable within the hydrocele. In this instance, or when there is any suspicion that there may be an underlying testicular tumor, a scrotal ultrasound should be performed. The patient with a spermatocele also usually presents with a scrotal mass. Physical examination reveals a mass superior to and separate from the testicle. This gives the impression that the patient has three testes, the so-called pawnbroker's sign. The spermatocele transilluminates and may give a Chinese lantern appearance due to septae within the spermatocele.

INDICATIONS FOR SURGERY
No surgery is indicated unless the hydrocele or spermatocele is causing pain, social embarrassment, or a tumor is suspected based on the ultrasonographic findings.

ALTERNATIVE THERAPY
Transcrotal needle aspiration, with or without instillation of sclerosing agents such as tetracycline, may be indicated in elderly patients with severely symptomatic hydroceles who may be poor surgical risks. This form of therapy is contraindicated in younger, healthier patients because of the high recurrence rate and the risk of persistent discomfort after sclerosis. Occasionally, the spermatocele may also be aspirated and injected with a sclerosing agent, such as tetracycline, though this is rarely required.

SURGICAL TECHNIQUE: HYDROCELE
The surgery can be performed under local, spinal, or general anesthesia. The entire scrotum is shaved immediately prior to the procedure, but the pubic hair is left intact. The scrotum and penis are cleaned with betadine and the area is draped with sterile towels, including one beneath the scrotum to elevate it. The incision in the scrotum can be made along the median raphe or transversely in an avascular area between the blood vessels that run transversely in the scrotal wall, while an assistant grasps the scrotum and compresses the hydrocele against the skin. The incision in the skin is made with a knife while the remainder of the incision through the subcutaneous layers and dartos muscle is made with the electrocautery in order to achieve satisfactory hemostasis. Once the parietal layer of the tunica vaginalis is exposed, a decision is made as to which type of procedure should be performed, i.e., Jaboulay or Lord. 1,2 The Jaboulay procedure is preferred in cases involving a thick-walled sac, whereas the Lord procedure is more suitable for thin-walled hydroceles. With either procedure, the testicle and appendages should be examined carefully for pathology. Hydrocele fluid is not usually sent for examination unless the fluid appears purulent or bloody. Lord Procedure The Lord procedure is performed by directly opening the parietal layer of the hydrocele sac without dissecting it free from the dartos layer. The testicle is then extruded into the surgical field and examined. The parietal layer of the tunica vaginalis is then plicated using a 3-0 chromic catgut suture by taking small bites at 1-cm intervals (Fig. 75-1A). Eight to 10 of these sutures are placed approximately 1 cm apart, and then all are tied so as to accordion the sac into a collar surrounding the testis and epididymis ( Fig. 75-1B).

FIG. 75-1. (A) The Lord operation. The testis is extruded through a small incision in the middle of the sac. (B) The sac is plicated with multiple sutures.

Jaboulay Procedure In the Jaboulay procedure (Fig. 75-2), the hydrocele is freed from the dartos layer using blunt dissection with a dry gauze sponge before the sac is opened and the testicle delivered. The excess sac is excised leaving a 2- to 3-cm remnant around the testicle, and the edges are oversewn with a running locked 3-0 chromic suture. The remnant of the sac is then wrapped posteriorly around the spermatic cord and sutured with a 3-0 chromic catgut suture, with care being taken not to strangulate the cord.

FIG. 75-2. Jaboulay technique. Most of the sac is excised and a running suture closes the free edges loosely about the cord. This is a rapid method of controlling troublesome bleeding.

SURGICAL TECHNIQUE: SPERMATOCELE
The skin preparation and incision is identical to that described above for a hydrocele repair. The tunica vaginalis is opened and the testicle is delivered with the spermatocele. The spermatocele is dissected from the epididymis using electrocautery ( Fig. 75-3), and if the attachment of the spermatocele to the epididymis can be found, it is ligated with a 4-0 chromic catgut suture. Using this technique, the spermatocele can usually be removed intact. The edge of the tunica vaginalis is sutured with a running 4-0 chromic catgut suture for hemostatic purposes, and the tunica vaginalis is left open to avoid the formation of a hydrocele. The scrotal incision is then approximated in two layers, as outlined for the hydrocele closure.

FIG. 75-3. Spermatocelectomy. The spermatocele is being separated from the head of the epididymis with electrocautery.

After either type of repair has been carried out, drains are not usually necessary. However, if hemostasis is difficult to attain, a Penrose drain should be placed and brought out through a separate stab incision in the inferior aspect of the scrotum. The incision is then closed in two layers using 3-0 chromic catgut sutures, with the first layer approximating the dartos muscle in a running fashion, and the second layer closing the skin with interrupted horizontal mattress sutures tied loosely without tension. A fluff dressing is applied and held in place with a scrotal support. An ice pack is kept on the scrotum for the first 24 hours to reduce pain and swelling, and appropriate oral analgesics are prescribed. Antibiotics are not routinely used.

OUTCOMES
Complications The most common complication is usually a hematoma. A wound infection, scrotal abscess, and recurrent hydrocele or spermatocele are less common. These complications are seen less frequently when the Lord procedure is performed. 3 Results The success rate of hydrocelectomy or spermatocelectomy should approach 100%. 3 CHAPTER REFERENCES
1. Jaboulay M. In: Chirurgie des Centres Nerveux, des Viscres et des Membres. Vol. 2. Lyon: Storck, 1902;192. 2. Lord PH. A bloodless operation for the radical cure of idiopathic hydrocele. Br J Surg 1964;51:914. 3. Rodriguez WC, Rodriguez DD, Fortuno RF. The operative treatment of hydrocele: a comparison of four basic techniques. J Urol 1981;125:804.

Chapter 76 Ureterosigmoidostomy and the Mainz Pouch II Glenn’s Urologic Surgery

Chapter 76 Ureterosigmoidostomy and the Mainz Pouch II
Margit Fisch and Rudolf Hohenfellner

M. Fisch and R. Hohenfellner: Department of Urology, Johannes Gutenberg University Medical School, 55131 Mainz, Germany.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Ureterosigmoidostomy Rectosigmoid Pouch (Mainz Pouch II) Serous Lined Extramural Tunnel Surgical Tricks for the Recto Sigmoid Pouch Outcomes Complications Results Chapter References

Since the introduction of internal urinary diversion 140 years ago by Simon, more than 60 modifications of ureterosigmoidostomy have been published. 2,7 Critics of ureterosigmoidostomy tend to quote publications dealing with complications in patients operated on before the 1950s. 3 The development of new absorbable suture material, modern ureteric stents, antibiotics, and alkalinizing drugs served in solving many of the traditional shortcomings of ureterosigmoidostomy and rekindled the interest in this technique. However, high intraluminal pressures caused by circular bowel contractions toward the anus during peristalsis can result in anal incontinence and dilation of the upper tract. Implantation into the so-called false loop, especially when the sigmoid colon is extremely mobile, may result in kinking and dilation of the ureters. Fixation of the sigmoid colon after ureteral implantation might be difficult and can damage the blood supply of the mesentery. Pertaining to these points, the technique of the rectosigmoid pouch (Mainz pouch II) optimizes classical ureterosigmoidostomy and has replaced it at our institution. 4,5 Dilated ureters, a contraindication for classical ureterosigmoidostomy, can also be implanted into a rectosigmoid pouch, preferably by the technique of the serous lined extramural tunnel. 1,2

DIAGNOSIS
A competent anal sphincter is prerequisite for ureterosigmoidostomy/rectosigmoid pouch. This can be checked by a tap water enema (rectal instillation of 200 to 300 ml saline), which the patient should keep at least for 3 hours without leakage. Also, a rectodynamic investigation should show no incontinence during measurement and an anal sphincter profile of a baseline closure pressure greater than 60 cm H 2O and a closure pressure under stress greater than 100 cm H 2O.

INDICATIONS FOR SURGERY
Indications for a sigma rectum pouch include the need for a primary urinary diversion, a revision of ureteral implantation after ureterosigmoidostomy, or a conversion of an incontinent diversion with a nonrefluxing ureteral implantation (colon conduit). Contraindications include an incompetent anal sphincter, prior irradiation of the pelvis, sigmoid diverticula or polyposis, or a creatinine concentration greater than 1.5 mg %.

ALTERNATIVE THERAPY
Alternatives to the Mainz pouch II include other forms of continent urinary diversion, ileal conduit or other incontinent diversions, or ureterosigmoidostomy.

SURGICAL TECHNIQUE
Instruments and suture material required include a basic kidney set, additional instruments for abdominal surgery, suction, and a basin containing prepared iodine solution for disinfection. Additionally, disposable supplies will include two ureteral stents (6 Fr), one rectal tube, and suture material. We prefer chromic catgut 5-0 for the mucosa and polyglycolic acid 4-0 in the seromuscular layer of the bowel and pouch walls, and we recommend 5-0 and 6-0 chromic catgut for the ureteral reimplantation. Ureterosigmoidostomy The initial incision is a median laparotomy. An incision is made in the peritoneum lateral to the ascending and descending colon, allowing identification of the right and left ureters, respectively. Both ureters are dissected, taking care to avoid injury to the longitudinal vessels running inside the Waldeyer's sheath down to the ureterovesical junction. At this point, the ureters are divided as distally as possible and a stay suture placed at the 6 o'clock position while the ureteral stumps are ligated. The colon is slightly elevated at the rectosigmoid junction by 4 stay sutures and a 4-cm incision is made in the anterior sigmoid wall in the area of the taenia libera to facilitate ureteral reimplantation. Four mucosal stay sutures are placed and an incision is made in the mucosa between the proximal stay sutures ( Fig. 76-1). A buttonhole of musculature is removed and a straight or slightly curved hemostat advanced into the opening ( Fig. 76-2). A 3-cm tunnel is created by blunt dissection just below the visceral peritoneum of the mesosigmoid and the ureter is pulled into the lumen of the intestine through this tunnel, with care taken to avoid torsion of the ureter (Fig. 76-3). The anterior wall of the ureter is spatulated for a length of 1 cm ( Fig. 76-4A). The final anastomosis is performed by first placing a 5-0 chromic catgut anchor suture at the 6 o'clock position (incorporating intestinal mucosa and musculature) and mucosa-to-mucosa sutures of 6-0 chromic catgut. A 6-Fr silastic stent is inserted and fixed to the mucosa by a 5-0 catgut suture ( Fig. 76-4B).

FIG. 76-1. Open transcolonic ureterosigmoidostomy: both ureters have been cut at their entrance into the bladder and mobilized. The site of the planned left ureteral implantation in the posterior sigmoid wall is outlined by stay sutures.

FIG. 76-2. After incision of the mucosa and excision of a buttonhole from the musculature at the point where the ureter will be brought through the intestinal wall, a tunnel is modeled bluntly from this point to the left incision in the peritoneum. The hemostat is advanced precisely below the peritoneum.

FIG. 76-3. The ureter has been pulled into the bowel and through a submucosal tunnel reaching from the proximal to the distal stay suture.

FIG. 76-4. Spatulation of the anterior wall of the ureter (A) and mucosa-mucosal anastomosis between ureter and intestinal wall. The ureter is stented (B).

The contralateral ureter is reimplanted in the same technique about 3 cm lateral and proximal or distal to the first anastomosis ( Fig. 76-5). The ureteral stents are led out perianally with the rectal tube and the anterior sigmoid incision is closed by seromuscular single stitches of 4-0 polyglycolic acid or two-layer running suture (5-0 chromic catgut for the mucosa and 4-0 polyglycolic acid for the seromuscularis). The peritoneal incisions are then closed. At the end of the operation, a rectal tube is reinserted and separately fixed from the ureteral stents near the anus with a suture of nonabsorbable material.

FIG. 76-5. Identical implantation of the right ureter 3 cm lateral and proximal or distal of the first anastomosis.

Rectosigmoid Pouch (Mainz Pouch II) The abdomen is entered via a median laparotomy and the ureters are prepared as for a ureterosigmoidostomy. The left ureter is pulled through the sigmoid mesentery above the inferior mesenteric artery ( Fig. 76-6). The junction between the sigmoid colon and rectum is identified and the intestine is opened at the taenia libera starting from the recto-sigmoid junction over a total length of 20 to 24 cm distal and proximal of this point. Two stay sutures are placed at the summit of the rectosigmoid, giving the split intestine the shape of an upside-down U ( Fig. 76-7). A side-to-side anastomosis of the medial margins of the U is created by a two-layer running suture of 4-0 polyglycolic acid for the seromuscular layer and 4-0 chromic catgut for the mucosa ( Fig. 76-8).

FIG. 76-6. Opening of the rectosigmoid at the taenia libera starting from the rectosigmoid junction over a total length of 20 to 24 cm distal and proximal of this point. By placing two stay sutures at the summit of the rectosigmoid the split intestine receives the shape of an upside-down U.

FIG. 76-7. Side-to-side anastomosis of the medial margins of the U by two-layer running sutures using 4-0 polyglycolic acid for the seromuscular layer and 4-0 chromic catgut for the mucosa.

FIG. 76-8. The left ureter is pulled through retromesenterically to the right side.

The ureteral reimplantation is performed by placing 4 mucosal stay sutures parallel to the right and left of the medial running suture. The mucosa and the seromuscular layers are excised to create a wide buttonhole between the two cranial stay sutures as an entrance of the ureter into the pouch. A submucosal tunnel is dissected starting from this incision over a length of 2 to 2.5 cm. The mucosa is incised at its distal end of the tunnel to allow a pull-through of the ureter. The ureter is resected to an adequate length and the reimplantation is completed by placing two anchor sutures at the 5 o'clock and 7 o'clock position and several single-stitch mucosa-mucosal sutures. The cranial mucosal incision is closed by a running suture of 6-0 chromic catgut ( Fig. 76-9). The contralateral ureter is reimplanted in an identical fashion. Then 8-Fr ureteral stents are placed to secure the ureteral implantation and are led out via the rectal tube.

FIG. 76-9. Creation of a wide buttonhole as an entrance of the ureter into the pouch and preparation of a submucosal tunnel (2 to 2.5 cm in length). Incision of the mucosa at its distant end and pull-through of the ureter. Ureteral implantation by two anchor sutures at the 5 o'clock and 7 o'clock positions and single-stitch mucosa-mucosal sutures. Closure of the cranial mucosal incision by a running suture with chromic catgut 6-0.

The pouch is fixated to the anterior longitudinal cord of the sacral promontory by two Bassini sutures in the area of the proximal end of the medial running suture ( Fig. 76-10). The anterior pouch wall is closed by two-layer sutures of 5-0 polyglycolic acid for the seromuscular and 4-0 chromic catgut for the mucosal layer. Alternatively, seromuscular single stitches can be used. The peritoneal incisions are closed and the anastomotic site of the pouch is covered by omentum ( Fig. 76-11). The ureteral stents and rectal tube are fixed as for ureterosigmoidostomy.

FIG. 76-10. After having placed two 8-Fr ureteral stents, which are led out with the rectal tube, the pouch is fixed to the anterior longitudinal cord of the promontory in the area of the proximal end of the medial running suture to by two Bassini sutures.

FIG. 76-11. Closure of the anterior pouch wall by seromuscular single stitches with polyglycolic acid 4-0. Closure of the peritoneal incisions.

Serous Lined Extramural Tunnel In cases of a short sigmoid colon, a left paracolonic incision is made mobilizing the descending colon including the left colonic flexure. The left ureter is identified and dissected free. A right paracolonic incision is made and continued along the root of the mesentery of the small bowel, allowing the right ureter to be identified and dissected free (Fig. 76-12). The ureters are divided to maximize the length and the left ureter is pulled through the sigmoid mesentery to the right side ( Fig. 76-13). An S-shaped sigmoid segment is marked by stay sutures, with each limb of 10 to 12 cm length resulting in a total length of 30 to 36 cm. An antimesenteric incision is made in the tenia libera of the S-shaped pouch ( Fig. 76-14). A side-to-side adaptation of the limbs is created by two serous running sutures close to the mesentery (non-absorbable suture, e.g., 4-0 silk or prolene) forming two serous-lined grooves.

FIG. 76-12. Right and left paracolonic incision, identification of the ureters.

FIG. 76-13. The ureters are cut and the left ureter is pulled through retromesenterically to the right side.

FIG. 76-14. An S-shaped sigmoid segment is marked by stay sutures, length 2 times 10 to 12 cm starting at the rectosigmoid junction (standard technique) plus additional 10 to 12 cm of the ascending colon and opened in the area of the taenia libera. Excision of a mesenteric window to pull the left ureter through.

On the right side an entrance for the right ureter is left at the cranial aspect of the running suture. The mesentery is incised cranial to the left running suture and the left ureter is pulled through the hiatus. The ureters are then brought into their respective grooves ( Fig. 76-15) and the respective borders of the bowel are anastomosed over the ureter by a running suture grasping all layers of the bowel wall. The length of the tunnel should be approximately 4 times the diameter of the ureter. The ureter is cut at its required length and spatulated. Four 4-0 chromic catgut anchor sutures are placed at the 11 o'clock and 1 o'clock and the 5 o'clock and 7 o'clock position grasping all layers of the ureter as well as all layers of the bowel wall. Mucosa-mucosal stitches of 5-0 chromic catgut are placed between the anchor sutures to complete the anastomosis ( Fig. 76-16). Two ureteral stents are inserted and led out with the rectal tube after being fixed to the bowel mucosa with 5-0 catgut. The anterior pouch wall is closed by seromuscular single sutures of 4-0 polyglycolic acid ( Fig. 76-17).

FIG. 76-15. Side-to-side adaptation of the limbs of the S by two serous running sutures close to the mesentery (nonabsorbable suture 4-0). Thereby two serous lined grooves are created. On the right side an entrance for the right ureter is left at the cranial aspect of the running suture and the ureter is pulled through. The ureters are laid down in the respective groove.

FIG. 76-16. Definitive ureteral implantation in the area of the continuous suture line. Anastomosis is secured by means of ureteral stents.

FIG. 76-17. The stents are led out transanally and the pouch is closed.

Surgical Tricks for the Recto Sigmoid Pouch When the anastomosis reaches deep down to the rectum, it is easier to suture the pouch starting caudally, as the deepest point of the anastomosis is the most critical part and can be reached easier at the beginning of the anastomosis. To facilitate fixation of the pouch to the promontory, one sutured end of the dorsal running suture can be pulled through dorsally outside of the pouch and be tied with the fixation suture placed in the anterior chord of the promontory. A Bassini needle facilitates placement of the fixation suture into the anterior chord. During ureteral implantation extensive spatulation of the ureters is of utmost importance to avoid a cuff-like protrusion of the ureteral borders. To avoid a hematoma of the submucosal tunnel, the mucosa over the tunnel can be carefully incised at different points.

OUTCOMES
Complications At our institution, we have long-term follow-up (more than 5 years) in 46 children who have undergone ureterosigmoidostomy. The indication for ureterosigmoidostomy had been bladder exstrophy in 40 patients, incontinent epispadias in 5, and neurogenic bladder dysfunction in 1. Seven early postoperative complications occurred: severe pyelonephritis developed in one patient resulting in nephrectomy. Three patients with unilateral dilation required reimplantation within 3 months after the initial operation. Three patients underwent late conversion of ureterosigmoidostomy because of upper urinary tract problems. One patient underwent a percutaneous stone operation on one kidney and nephrectomy of the infected contralateral kidney had to be performed 20 years after ureterosigmoidostomy. Episodes of pyelonephritis developed in five patients with nondilated upper tracts. Between 1990 and July 1994 a rectosigmoid pouch was performed in 87 patients (69 adults and 18 children). Mean age was 40.2 years. Indications were malignancy (n = 65), bladder exstrophy and incontinent epispadias ( n = 17), trauma (n = 4), and a sinus urogenitalis ( n = 1). Of the 87 patients, 83 were followed with a mean follow-up of 26.4 months (5 months to 4.3 years). Seven patients died during follow-up due to their primary malignant tumor. Seven early complications were encountered in 6 patients (6.9%): one dislodged ureteral stent requiring temporary nephrostomy, one pulmonary embolism, two pneumoniae, one suture dehiscence, and one ileus requiring operative intervention. One patient developed severe complications as suture insufficiency and leakage required revision and ended with a colostomy due to a pouch fistula. Eight late complications occurred followed by an intervention (9.6%), with stenosis at ureteral implantation site being the most common complication (7.2%). During follow-up eight patients developed pyelonephritis (9.6%). Results In the patients with ureterosigmoidostomies, the daytime continence rate was 97.4% and the complete continence rate was 92.3%. In the patients undergoing the rectosigmoid pouch, 76 of the 83 patients were completely continent postoperatively; 2 children are still too young for final judgment (daytime continence 93.8%). Four of the five patients suffered from stress incontinence grade I–II. Five patients are incontinent during the night (nighttime continence 93.8%). Thirteen patients irregularly lose some drops of urine during the night; seven of them use pads prophylactically. The majority of the patients take alkalinizing drugs to avoid metabolic acidosis.

CHAPTER REFERENCES
1. Abol-Enein H, Ghoneim MA. A novel uretero-ileal reimplantation technique: the serous lined extramural tunnel. Preliminary report. J Urol 1993;1193–1197. 2. Abol-Enein H, Ghoneim MA. Optimization of uretero-intestinal anastomosis in urinary diversion: an experimental study in dogs. III. A new antireflux technique for ureterointestinal anastomosis: a serous-lined extramural tunnel. Urol Res 1993;21:135–139. 3. Connor JP, Hensle TW, Lattimer JK, Burbige KA. Long-term follow-up of 207 patients with bladder exstrophy: an evolution in treatment. J Urol 1989;142:793. 4. Fisch M, Hohenfellner R. Der Sigma-Rektum Pouch: Eine Modifikation der Harnleiterdarmimplantation. Akt Urol 1991;22:I–IX. 5. Fisch M, Wammack R, Miller SC, Hohenfellner R. The Mainz pouch II (sigma rectum pouch). J Urol 1993;149:258–263. 6. Hinman F, Weyrauch HM Jr. A critical study of the different principles of surgery which have been used in uretero-intestinal implantation. Trans Am Assoc Genitourin Surg 1036;29:15. 7. Simon J. Ektropia vesica (absence of the anterior wall of the bladder and pubic abdominal parieties): operation for directing the orifices of the ureters into the rectum; temporary success; subsequent death; autopsy. Lancet 1852;2:568.

Chapter 77 Conduit Urinary Diversion Glenn’s Urologic Surgery

Chapter 77 Conduit Urinary Diversion
Lesley K. Carr and George D. Webster

L. K. Carr: Department of Surgery Division of Urology, University of Toronto, Wellesley Hospital, Toronto M4Y IJ3, Ontario, Canada. G. D. Webster: Department of Surgery, Division of Urology, Duke University Medical Center, Durham, North Carolina 27710.

Diagnosis Indications for Surgery Alternative Therapy Description of Procedure Patient Preparation Surgical Technique Postoperative Care Outcomes Complications Results Chapter References

The ileal conduit, as first described by Bricker in 1950, continues to be the most common form of urinary diversion performed worldwide. Although continent urinary diversion to a catheterizable abdominal stoma and orthotopic bladder replacement are gaining popularity and may offer the patient greater freedom to continue social and leisure aspects of their life, circumstances exist when simple conduit diversion is optimal.

DIAGNOSIS
This procedure is done as part of a reconstruction following functional or actual loss of the urinary bladder, and all diagnostic studies are directed to the underlying pathology in the bladder.

INDICATIONS FOR SURGERY
Patients should be thoroughly counseled regarding options for diversion, both continent and incontinent. Coexistent medical illness such as renal compromise (serum creatinine more than 2.5 mg %) predicting electrolyte disturbance or bowel disease (inflammatory bowel, malignancy) may mitigate against continent diversion. In very elderly or infirm patients who would tolerate additional operative time or possible revisions of the diversion poorly, a simple conduit may be wise. Finally, preoperative confirmation of the ability and reliability of the patient to perform self-catheterization is an essential requirement for any continent diversion.

ALTERNATIVE THERAPY
When a conduit diversion is selected, options exist for the segment of intestine used and the type of ureteroenteric anastomosis. Other potential conduit segments include colon and jejunum, the latter of which is associated with a high incidence of metabolic disturbances. The most commonly employed alternative to the ileal conduit is some form of continent diversion or orthotopic neobladder.

DESCRIPTION OF PROCEDURE
Patient Preparation Proper stoma siting is critical for a successful outcome in appliance-dependent urostomy surgery. The entero-stomal therapist should examine the patient in the supine, erect, and sitting positions and will consider preferred clothing styles. The stoma site is generally in the right lower quadrant just medial to the rectus on a line between the umbilicus and anterior superior iliac spine. The site should avoid these landmarks, along with scars and creases to enable face plate adherence. Bowel preparation should begin 2 days prior to surgery with clear fluids. Polyethylene glycol electrolyte solutions taken by mouth the day before surgery will achieve an adequate preparation in most cases. Volumes of 4 L are generally sufficient and may be augmented by tap water enemas per rectum until efflux is clear. Intravenous hydration during this final phase of preparation may be advisable to avoid dehydration. The addition of oral antibiotics is debatable but broad-spectrum intravenous antibiotic coverage is indicated at the time of surgery. Surgical Technique Incision and Preparation of the Field A long midline incision is optimal for dissection and an abdominal ring retractor (Bookwalter type) facilitates exposure. The ureters should be mobilized with care to protect their blood supply and transacted distally as close to the bladder as possible. This is especially important on the left side where the ureter must have adequate length to traverse a hiatus beneath the sigmoid mesocolon and approximate the right ureter for ureteroenteric anastomosis. Isolation of the Intestinal Segment The terminal ileum remains the standard bowel segment for conduit urinary diversion. Relative contraindications to the use of ileum are significant radiation changes, coexistent bowel disease such as active Crohn's disease, short bowel syndrome, and insufficient mesenteric length to allow proper delivery of the loop through the abdominal wall for stoma creation. In the patient with a normal body habitus, a 15-cm length of ileum is usually sufficient. Conduits that are too short jeopardize proper stoma formation and conduits that are excessively long may aggravate metabolic and electrolyte abnormalities and empty poorly. The distal mesenteric incision is made in the avascular area between the ileocolic artery and the right colic artery ( Fig. 77-1). This location, which is generally 15 cm from the ileocecal valve, allows for a long distal mesenteric incision facilitating mobility of the bowel for its transfer through the abdominal wall yet maintains sufficient terminal ileum to avoid bile salt malabsorption. At least four vascular arcades should supply the loop to avoid ischemic complications.

FIG. 77-1. (A) Selection of the ileal segment. Note distal mesenteric window located in the watershed area between the right colic artery and the ileal branch of the ileocolic artery. (B) Bowel anastomosis performed cephalad to the isolated ileal segment with mesenteric defect closed.

Ileoileal reanastomosis is performed cephalad to the conduit using either standard staple or suture techniques. The mesenteric window is closed to avoid internal bowel herniation. The isolated ileal loop is then irrigated free of fecal contents using antibiotic solution and a sump suction. Use of Colon There are several theoretical advantages to the use of colon for urinary conduits. The colon is more amenable to tunneled antireflux ureteroenteric anastomosis, is less prone to stomal stenosis, and is potentially less likely to result in electrolyte disturbances. Despite this, colon conduits generally only find their use in patients with pathology of the small intestine or heavy prior pelvic irradiation. The transverse and sigmoid colon are the most commonly employed segments. The transverse colon has the advantage of being out of the field of pelvic irradiation, having good mobility, and occupying a more cephalad position for use with short ureters. The segment of colon is isolated in a similar fashion to that described for the ileum ( Fig. 77-2). Leadbetter-type tunneled antireflux ureteroenteric anastomosis or simple end-to-side anastomosis is then performed.

FIG. 77-2. Transverse colon conduit. The large bowel is anastomosed and intestinal continuity is restored. The ureters are brought intraperitoneally below the duodenum next to the proximal portion of the colon conduit. Anastomoses of ureters to bowel are accomplished in a routine manner. Stents may be employed.

Use of Jejunum Jejunum has been recommended for use as a urinary conduit when the stoma must be placed higher on the abdominal wall or when high diversion is needed because of lower ureteral pathology. It may also be less affected by prior pelvic irradiation than ileum. The disadvantage of jejunum is its marked absorptive capacity, which is associated with large fluid shifts and a syndrome characterized by hyponatremic, hypochloremic, hyperkalemic metabolic acidosis. The situation is aggravated by compromised renal function. Ureteroenteric Anastomosis A good outcome from ureteroenteric anastomosis is dependent on intact blood supply to the conduit and distal ureter along with avoidance of tension. If cystectomy was performed for malignancy, frozen sections of the ureteral margins should be free of tumor, carcinoma in situ, and severe dysplasia. A surgical clip or ligature left on the distal ureter from the time of its transection will allow the ureter to dilate gently and facilitate anastomosis. The most commonly used anastomotic techniques are the Bricker end-to-side and the Wallace conjoined spatulated end-to-side anastomosis, both of which are freely refluxing. Alternatively, a split-cuff nipple or the technique of Le Duc may be used to prevent reflux. When colon is used, Leadbetter antirefluxing tunneled implants through the tenia are generally advocated. The true value of nonrefluxing ureteroenteric anastomosis remains unknown, and it may be that the potential advantages of reflux avoidance are offset by the increased risk of ureteral obstruction. Atraumatic handling of the ureter and careful suture placement using 4-0 or 5-0 absorbable material are essential for a good outcome. Urinary leakage with resulting fibrosis may be minimized with the use of stents. Silastic single J stents may be brought out through the conduit stoma and offer the additional benefit of preventing poor conduit drainage from early bowel edema. Regardless of anastomotic technique, it is customary to retroperitonealize the proximal conduit end and the region of the ureteroenteric anastomosis by fixing the cut edges of the peritoneum to the conduit. This also limits future risk to the anastomotic area from intraperitoneal surgery or pathology, and may contain any postoperative anastomotic urine leak. Most surgeons advocate the use of closed suction drains placed alongside the ureteroenteric anastomosis during the initial preoperative period. Bricker End-to-Side Ureteroileal Anastomosis With this technique, the butt (proximal) end of the conduit is closed using two layers of absorbable suture. A small full-thickness plug of ileal wall is sharply excised at a site where the end of the ureter naturally lies. The mucosal defect need not be as large as the seromuscular defect. The terminal ureter is spatulated and a single-layer anastomosis is performed using interrupted sutures of absorbable 4-0 or 5-0 material. Sutures should encompass the full thickness of the ureteral and bowel wall. Further sutures approximating ureteral adventitia and bowel serosa will act to reinforce the anastomosis but must not cause kinking ( Fig. 77-3).

FIG. 77-3. Ileal conduit showing the ureters sutured from beneath the isolated segment into the antimesenteric border of the loop. The reconstituted mesentery is demonstrated.

Wallace Anastomosis With this technique, the proximal end of the conduit is left open for anastomosis to the conjoined ureters. The ureters are spatulated over a length of 2 to 3 cm and the adjacent sides arc approximated using a running 4-0 absorbable suture. The open ureteral plate is then anastomosed end-to-end to the open proximal end of the ileal

conduit using running absorbable suture. The advantages of this technique are its speed and the wide ureteral anastomosis, which is less prone to obstruction. A suggested disadvantage when diversion is being performed for bladder malignancy is the possibility of ureteral recurrence causing bilateral obstruction ( Fig. 77-4).

FIG. 77-4. Wallace technique with anastomosis of the conjoined ureteral plate to the proximal end of the ileal conduit.

Leadbetter Tunneled Reimplant into Colon For tunneled anastomosis, the ureter is spatulated and a 2- to 3-cm trough is made in an antimesenteric tenia. The grove is full-thickness through the serosa and muscularis but leaves the mucosa intact. A small opening is made in the mucosa at the inferior end of the trough and the ureter is anastomosed in this location using fine absorbable suture. The seromuscular layer is then approximated over the proximal ureter, thus creating a submucosal tunnel to prevent reflux. The groove must be wide enough that the ureter is not obstructed by reapproximation of the muscular layer ( Fig. 77-5).

FIG. 77-5. Leadbetter ureterocolic nonrefluxing anastomosis. A submucosal tunnel is created along an antimesenteric tenia and reapproximated in a nonobstructing fashion over the ureter.

Construction of the Stoma A circular plug of skin and subcutaneous fat is excised at the preoperatively marked site. A cruciate incision in the anterior rectus fascia is made and the rectus muscle bluntly split. The hiatus created into the peritoneal cavity should readily admit two fingers. This will avoid compression of the conduit mesentery and reduce the risk of parastomal hernia formation. The distal conduit is grasped and delivered through the abdominal wall defect, ensuring that the mesentery is not twisted or stripped. The conduit is secured by fixing the anterior rectus fascia to the serosa of the bowel in four quadrants. The border of the stoma is matured to the skin edge using three-way bites incorporating dermis, serosa 2 cm from the edge, and full-thickness bowel margin ( Fig. 77-6). Four such sutures are generally sufficient to achieve eversion of the stoma and additional sutures may be placed between the dermis and bowel edge, as needed.

FIG. 77-6. A rosebud stomal nipple is created by using interrupted absorbable stitches between subcutaneous tissue, lateral bowel wall, and the terminal edge of the segment. This avoids puckering of the skin and permits a smooth surface for the appliance.

In cases where the bowel has insufficient mobility or the patient is morbidly obese, a Turnbull loop stoma may offer benefit ( Fig. 77-7).9 In this technique, the distal end of the conduit is closed using two-layer absorbable suture. A knuckle of the terminal loop is delivered through the stomal defect and the bowel is incised on the antimesenteric border. The serosa is secured to the anterior rectus fascia and the cut edge of the bowel is matured to the skin edge. The blind end of the stoma should be positioned cephalad.

FIG. 77-7. The modified Turnbull loop stoma.

Postoperative Care The major factor dictating duration of hospital admission is return of bowel function. Nasogastric or percutaneous gastrostomy bowel decompression may be advocated during the initial postoperative period of ileus depending on the magnitude of the associated surgical procedures. Ureteral stents are maintained for approximately 7 days or longer. The suction drain may be removed once drainage is negligible or shown to be serous based on a low-drain fluid creatinine. Before discharge the patient should be taught about all aspects of stoma care by an enterostomal therapist. Routine follow up is generally at 4 to 6 weeks and then 3 months postoperatively. An intravenous pyelogram and serum renal profile is obtained at the 3-month visit. Unless problems arise, annual surveillance of the upper urinary tract, renal and metabolic profile, and periodic serum B 12 and folate are levels sufficient thereafter. If the reason for cystectomy was malignancy, urine cytology is also indicated.

OUTCOMES
Complications The perioperative mortality and morbidity for patients undergoing ileal conduit urinary diversion has reduced remarkably as a result of improved surgical technique along with better supportive care both intraoperatively and postoperatively. Current mortality rates including cystectomy can be expected between 1% and 3%. The most common urology-specific early postoperative complications are ureteroenteric anastomotic urine leak or obstruction and urosepsis. With the availability of percutaneous nephrostomy drainage, surgical intervention for ureteral anastomotic obstruction due to edema is rarely required. Delayed complications of conduit urinary diversion are common, but generally are mild and do not require surgical revision. Urinary colonization can be expected in at least 65% of patients, but perhaps as few as 20% develop pyelonephritis as a result of reflux of this colonized urine. Antibiotic therapy is usually warranted only for recurrent pyelonephritis or if Proteus becomes chronic. Approximately 10% of patients will develop renal calculi but their management has been dramatically facilitated by percutaneous and lithotripsy techniques. Stomal complications, namely, stenosis, prolapse, or parastomal hernias combined, may occur in up to 25% of cases with prolonged surveillance. These complications generally require operative revision. Ureteral obstruction has been reported with an incidence of 5% to 10%. 7 Most commonly it is a result of fibrosis, but recurrent malignancy and stones may be contributing factors. Endoscopic or fluoroscopic balloon dilation of the strictured area is often successful at relieving the obstruction. Deterioration in renal function must be monitored on a lifelong basis. In some cases it is chronic and progressive with no identifiable cause, but in many cases an etiology may be identified and corrected. Results Conduit urinary diversion will continue to enjoy a major role even with increasing popularity of continent options. The simplicity of surgical technique and postoperative stoma care renders conduit diversion the most appropriate choice in many instances. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Bricker EM. Bladder substitution after pelvic evisceration. Surg Clin North Am 1950;30:1511. Golimbu M, Morales P. Jejunal conduits: technique and complications. J Urol 1975;112:787. Leadbetter WF, Clarke BG. Five years experience with ureteroenterostomy by the “combined” technique. J Urol 1954;73:67. Leduc A, Camey M, Teillac P. An original antireflux ureteroileal implantation technique: long-term follow up. J Urol 1987;137:1156. Morales P, Golimbu M. Colonic urinary diversion: 10 years of experience. J Urol 1975;113:302. Regan JB, Barrett DM. Stented versus non stented ureteroileal anastomoses: is there a difference with regard to leak and stricture? J Urol 1985;134:1101. Schmidt JD, Hawtrey CE, Flocks R, Culp DA. Complications, results and problems of ileal conduit diversion. J Urol 1973;109:210. Stone AR, Macdermott JPA. The split-cuff ureteral nipple reimplantation technique: reliable reflux prevention from bowel segments. J Urol 1989;142:707. Turnbull RB, Fazia V. Advances in surgical technique in ulcerative colitis surgery. In: Nyhus L, ed. Surgery annual. New York: Appleton-Century Crofts, 1975;315. Wallace DM. Ureteric diversion using a conduit: a simplified technique. Br J Urol 1968;38:522.

Chapter 78 Kock Pouch Continent Urinary Diversion Glenn’s Urologic Surgery

Chapter 78 Kock Pouch Continent Urinary Diversion
John A. Freeman

J. A. Freeman: Division of Urology, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599–7235.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Preparation Operative Technique Postoperative Care Outcomes Complications Results Chapter References

Innovative surgeons have struggled for centuries with the question of how best to replace a diseased bladder. During this time the technique of urinary diversion has evolved along three themes: cutaneous fistulas (nephrostomy, pyelostomy, and ureterostomy), implantation of ureters into segments of bowel in continuity with the fecal stream (ureterosigmoidostomy), and implantation of ureters into segments of bowel isolated from the fecal stream (bowel conduits and continent diversion). While all of these forms of diversion are successful in eliminating urine, they do not achieve similar success in recapitulating the detrusor's inherent properties of providing a large-capacity, low-pressure reservoir that is continent and nonrefluxing, allows volitional voiding, and prevents significant resorption of urinary electrolytes. These properties are best mimicked by continent forms of urinary diversion. Modern advances in surgical technique, perioperative care, and anesthesia have lessened the morbidity of exenterative surgery, allowing the surgeon and patient to consider reconstructive options. In 1982, Nils Kock reported his results with a continent cutaneous ileal reservoir for urinary diversion in 12 patients. 4 This was the first documentation of the simultaneous use of the preeminent concepts of modern continent urinary diversion, which must now be considered to be the gold standard in urinary reconstruction: (a) detubularization of the bowel to provide a low-pressure, large-capacity reservoir; (b) an antireflux mechanism to protect the upper urinary tract; and (c) urinary continence. With more than a decade of experience with continent diversion, reconstructive surgeons have proven that continent forms of diversion can replace conduit diversion in most cases, are technically sound and durable, and are clearly the preferred choice of most patients who are given the option.

DIAGNOSIS
The use of this procedure is as an adjunct for reconstruction of the urinary tract, usually following cystectomy. The diagnostic modalities therefore are relevant only as to the primary diagnosis that resulted in the plan for the cystectomy.

INDICATIONS FOR SURGERY
Any patient who requires removal of the bladder or diversion of the urinary stream away from the bladder is a potential candidate for a Kock pouch cutaneous urinary reservoir. Today, however, the number of patients undergoing cutaneous diversion (continent or incontinent) continues to decline. This is a function primarily of the success and utilization of orthotopic urinary diversion, which is applicable to approximately 90% of male bladder cancer patients and 75% to 85% of female bladder cancer patients. 2,10 To be a candidate for continent cutaneous diversion, the patient should have the intelligence, maturity, and manual dexterity to manage the urinary diversion, as well as a life expectancy great enough to recover from the operation and enjoy the lifestyle benefits of the reconstructive surgery. Some patients are less than ideal candidates. Morbid obesity presents significant technical difficulties in forming the pouch, securing the cutaneous continence limb, and achieving catheterization. Patients with prior small bowel resections, bowel symptoms, or neurogenic bowel dysfunction are at risk for development of bowel dysfunction following surgery. Impaired renal function, unless severe (serum creatinine more than 2.5–3.0), is not a contraindication to continent urinary diversion. Initial concerns over electrolyte disturbances caused by urine resorption in the face of compromised renal function have not proven clinically significant. Only rarely does a patient develop a clinically significant metabolic acidosis requiring therapy because the bowel mucosa loses much of its absorptive capacity after chronic exposure to urine.

ALTERNATIVE THERAPY
The alternatives to continent cutaneous urinary diversion are incontinent cutaneous urinary diversion (percutaneous nephrostomy, pyelostomy, ureterostomy, and bowel conduit urinary diversion), ureterosigmoidostomy, and orthotopic urinary diversion (neobladder construction). The patient must understand that the choice of a particular form of urinary diversion consists primarily of a quality-of-life decision and has essentially no impact on the course of the disease necessitating bladder replacement. It is the responsibility of the surgeon who undertakes urinary reconstruction to fully educate the patient of all available forms of reconstruction, and their relative benefits and risks. Having been so educated, the patient is prepared to make a truly informed decision. Currently, continent urinary diversion is the gold standard against which any other form of urinary diversion must be measured.

SURGICAL TECHNIQUE
Preoperative Preparation The patient is admitted 1 day prior to surgery for bowel preparation, intravenous hydration, and antibiotics. The modified Nichols bowel preparation, which has been previously described, consists of a clear liquid diet and 120 cm 3 of oral castor oil emulsion (Neoloid) administered on arrival, followed by oral neomycin and erythromycin base over the course of the day. 8 This is a well-tolerated regimen that gently and reliably cleanses and sterilizes the small bowel. The enterostomal therapist marks an appropriate abdominal stoma site and reinforces patient education concerning reconstruction alternatives. If the patient is undergoing conversion of an existing ileal conduit to a cutaneous Kock ileal reservoir, it is important to radiographically assess the patency of the ureteroileal anastomoses. If free reflux is demonstrated, the ileal conduit can be incorporated into the afferent antireflux segment of the reservoir. If obstruction is demonstrated, the ureteroileal anastomoses should be revised prior to incorporation of the conduit into the antireflux segment. Intravenous hydration overnight prevents dehydration, which can complicate any cathartic bowel preparation. Operative Technique The patient is placed in the hyperextended supine position. A midline incision from the pubis is carried left of the umbilicus and extended to the epigastrium. Construction of the urinary reservoir is begun after completion of the exenterative portion of the operation. The small bowel is divided distally in a freely mobile portion of the mesentery, 15 to 20 cm proximal to the cecum in the avascular plane of Treves between the terminal branch of the superior mesenteric artery and the ileocolic artery. If small bowel has been resected previously, then the enteroenterostomy site should be utilized if at all possible. Locating the distal division in a mobile portion of the mesentery is critical because this will later be secured to the abdominal wall as the cutaneous continence mechanism. The length of small bowel to be used for the reservoir must be measured initially. Segments of ileum to be utilized for specific components of the reservoir are measured and marked with silk sutures as one moves proximally along the 78 cm of small intestine that will be required for the reservoir: 17 cm distally for the efferent cutaneous continence nipple mechanism, two 22-cm segments for the reservoir, and a second 17-cm segment for the proximal antireflux nipple mechanism ( Fig. 78-1). A shallower mesenteric division is made at the proximal end of the bowel, assuring a broad vascular pedicle to the segment for the reservoir. An additional 5-cm segment of bowel is discarded proximal to this proximal mesenteric division to allow free mobility of the reservoir separate from the remaining small bowel. Continuity of the gastrointestinal tract is restored by performing an enteroenterostomy. The use of a stapled or hand-sewn anastomosis is the operating surgeon's

choice. If staples are utilized, the staples at the proximal end of the segment of intestine to be utilized for the reservoir should be excised and the bowel should be closed with absorbable suture (3-0 chromic Parker-Kerr stitch), reinforced with 4-0 silk Lembert sutures to prevent stone formation within the reservoir. The mesenteric window is closed with running, locking 3-0 chromic suture, with care taken not to compromise the blood supply to the reservoir.

FIG. 78-1. A 78-cm segment of small bowel is isolated for the Kock pouch. The distal 17 cm is utilized for the cutaneous continence valve, two 22-cm segments are used for the lumen of the reservoir, and a proximal 17-cm segment is used for the antirefluxing valve mechanism. A 5-cm segment of small bowel is excised proximal to the isolated segment to ensure adequate mobility between the reservoir segment and the small bowel reanastomosis. The proximal mesenteric division should be shallow to provide a broad vascular pedicle to the reservoir segment. The distal mesenteric division is normally located in the avascular plane between the terminal branch of the superior mesenteric artery and the ileal colic artery.

The small bowel segment to be utilized for the reservoir is then isolated in the right lower quadrant ( Fig. 78-2). It is easiest to lay the bowel into a U shape in the right lower quadrant on top of a sterile towel and drape a second towel over the remaining bowel above the reservoir segment. This leaves only the reservoir segment visible in the surgeon's field. The serosal surfaces of the inner edges of the two 22-cm segments that form the U are then apposed to each other 0.5 cm above the mesentery with a running 3-0 polyglycolic acid (PGA) suture, with care taken not to damage the mesentery or move too far away from the mesenteric edge ( Fig. 78-2). The 22-cm segments are then opened along the antimesenteric edge of the 3-0 PGA suture line, exposing the lumen of the bowel and “back wall” of the reservoir. It is important to continue these incisions some 2 to 3 cm onto the 17-cm segments so that when the nipple valves are later constructed they will be separated along the back wall of the reservoir (Fig. 78-3). The raw mucosal edges of the back wall resulting from incising the serosa adjacent to the original 3-0 PGA suture are then anastomosed with two layers of running, locking 3-0 PGA suture ( Fig. 78-4).

FIG. 78-2. The 78-cm segment is isolated in the right lower quadrant with the two 22-cm segments forming a U directed into the right lower quadrant. The 17-cm segments lie as wings at the top of the U. The serosal surfaces of the 22-cm segments are sewn together with 3-0 polyglycolic acid suture 0.5 cm above the mesenteric junction.

FIG. 78-3. The reservoir segments are incised with electrocautery adjacent to the polyglycolic acid suture line. The incisions are extended 2 to 3 cm along the 17-cm wings, allowing the nipple valves to be separated on the back wall of the reservoir.

FIG. 78-4. The back wall of the reservoir is closed in water tight fashion with a running, locking two-layer 3-0 polyglycolic acid suture.

The next task is to create the nipple valve mechanisms responsible for continence in the efferent (distal) segment and preventing reflux in the afferent (proximal) segment. Both nipple mechanisms are constructed in similar fashion, although the proximal antireflux nipple can be shorter than the distal continence nipple. Beginning where the 17-cm segments join the 22-cm segments, the mesentery of the 17-cm wings of ileum is divided for 7 to 8 cm (windows of Deaver) with electrocautery adjacent to the ileum without injuring the bowel serosa ( Fig. 78-5). Limiting adherent mesenteric fat on the bowel serosa facilitates the intussusception required for nipple construction and prevents later extussusception. Three vascular arcades are spared and a separate small window of Deaver is made in the

mesentery to allow passage of a 2- to 3-cm doubled strip of PGA mesh soaked in tetracycline (250 mg in 10 cm 3 normal saline). Marlex mesh (as originally described) should not be used because it can erode into the bowel, causing infection, stone formation, and incontinence. The PGA mesh is used to fix and stabilize the nipple, and will assist in anchoring the efferent limb to the abdominal fascia.

FIG. 78-5. To allow intussusception of the nipple valve mechanisms, the mesentery adjacent to the 17-cm segments must be mobilized. An 8-cm mesenteric window is developed using electrocautery for hemostasis. One centimeter of mesentery is spared where each 17-cm segment joins the 22-cm segments. A second 1-cm window in the mesentery is created one or two vascular arcades beyond the 8-cm mesenteric window to accommodate the polyglycolic acid mesh collars.

The nipple valves are created by intussuscepting bowel from the 17-cm segments into the lumen of the reservoir. The intussusception is performed by passing two Allis clamps two-thirds of the way into the lumen of the 17-cm segment toward the PGA mesh, grasping the mucosa, and withdrawing the Allis clamps. A third Allis clamp on the cut edge of the ileum facilitates this maneuver ( Fig. 78-6). The resulting nipple valve should be at least 5 cm long, and it is important to leave a lip of bowel serosa (where the third Allis clamp is located) to facilitate closure of the reservoir. The nipple intussusception is secured by stapling of the intussuscepted bowel. This can be performed with a standard TA-55 stapler, but I prefer a custom gastrointestinal anastomosis (GIA) device that has no knife. The TA-55 that aligns with a pin in the stapling head occasionally leads to a pinhole fistula in the valve mechanism and should be avoided. Pinless TA-55 devices are available. Two rows of staples are applied in the anterior 180 degrees of the nipple, making sure to utilize the entire length of the stapling device to achieve a 5-cm nipple mechanism ( Fig. 78-7). Since the staples at the tip of the nipple do not significantly contribute to its stability, the distal six staples can be removed from the cartridge prior to firing. This eliminates redundant staples that are exposed at the tip of the nipple and contribute to stone formation.

FIG. 78-6. Tetracycline-soaked mesh is positioned in each distal mesenteric window. The nipple valves are intussuscepted into the lumen of the reservoir by passing two Allis clamps two-thirds of the way into the lumen of the 17-cm segment, grasping the mucosa, and pulling the limb 5 cm into the reservoir.

FIG. 78-7. The nipple valve mechanisms are secured by stapling the anterior 180° of the valve mechanism with two staple lines from a pinless TA-55 or knifeless GIA. The distal 6 staples at the tip of the nipple are removed prior to firing the stapler.

To prevent extussusception of the valve, with resultant incontinence and reflux, a third full-length row of staples secures the nipple to the back wall of the reservoir. The easiest and most reliable method to do this is by passing the customized GIA (without a knife) or a TA-55 device along the serosal surface of the intussusception. Firing the device in this way attaches the back wall of the reservoir to a single thickness of the nipple with a full row of staples ( Fig. 78-8). The nipple is further secured to the pouch wall by scoring the apposing mucosal surfaces with electrocautery and suturing the nipple tip to the back wall with 2-0 chromic suture ( Fig. 78-9). Finally, the tetracycline-soaked PGA mesh is secured circumferentially to the ileal serosa with 2-0 chromic suture, and additionally to the base of the reservoir where the intussusception enters the reservoir. Care is taken not to narrow the lumen of the ileum by performing the fixation over a 30-Fr Medina (ileostomy) catheter (Fig. 78-10). Careful construction and fixation of the valves is crucial to maintain continence, prevent reflux, avoid fistulas, and allow easy catheterization; this portion of the operation cannot be rushed.

FIG. 78-8. The valves are secured to the back wall of the reservoir by passing a pinless TA-55 or knifeless GIA along the serosal surface of the back wall and serosal surface of the intussuscepted ileum, adjacent to the mesentery. This fixes one fold of the intussusception to the back wall of the reservoir. Alternatively, a hole can be

made in the back wall of the reservoir adjacent to the tip of the valve and the arm of the stapler can then be passed inside the lumen of the valve. This then fixes all three serosal layers of the bowel (two from the intussusception and one from the back wall). If this technique is used, the hole in the back wall needs to be oversewn with 3-0 polyglycolic acid suture.

FIG. 78-9. The tips of the nipples are secured to the back wall of the reservoir with 2-0 chromic after scoring the mucosal surface of the nipple and the reservoir to increase scar formation and nipple fixation.

FIG. 78-10. The polyglycolic acid mesh collars are sutured to the 17-cm segments and base of the 22-cm segments over a 30-Fr catheter.

The reservoir is then closed by folding the bottom of the U up to the free mucosal edge near the base of the valve mechanisms. The center portion of the bottom of the U is attached to the midpoint of the free edge between the nipples, creating a spherical reservoir. Running 3-0 PGA suture closes this anterior wall in two layers ( Fig. 78-11). Closely placed sutures that invert the mucosa prevent urinary leaks.

FIG. 78-11. The reservoir is closed in spherical fashion with two layers of a running, locking 3-0 polyglycolic acid suture.

The ureters are spatulated and anastomosed to the reservoir in end-to-side fashion on opposite sides of the afferent limb using interrupted 4-0 PGA sutures ( Fig. 78-12). The base of the afferent segment is secured to the sacral promontory. The ureteroileal anastomoses are stented with radiolucent no. 8 infant feeding tubes that are passed proximally into the renal pelvis. The distal ends are passed across the antireflux valve into the lumen of the reservoir. When the ureteroileal anastomoses are completed, the sigmoid mesenteric trap is closed by suturing the edge of the mesentery to the PGA collar at the base of the afferent limb. In this way, the left ureter is left in a retroperitoneal location.

FIG. 78-12. The ureters are anastomosed to the afferent limb in end-to-side fashion, and the base of the efferent limb is secured to the sacral promontory. The stoma site is selected and three horizontal mattress sutures of no. 1 polyglycolic acid (PGA) are fixed to the lateral, medial, and inferior aspects of the anterior fascial opening. These are secured to the corresponding sites of the PGA mesh collar on the efferent limb. A 1-cm Marlex mesh sling is passed through the mesenteric window of Deaver on the superior aspect of the fascial opening. The cutaneous continence limb is pulled through the stoma, the PGA sutures are tied, and the Marlex sling is secured to the posterior abdominal fascia.

Finally, the efferent limb is prepared for cutaneous anastomosis. It is my preference to taper the efferent limb segment over an 18-Fr catheter down to the PGA mesh. This removes some of the redundancy that can make catheterization difficult at times. The tapering is easily performed by excising the antimesenteric border of the

ileum with a standard GIA stapler. The location of the stoma site is chosen to allow the efferent limb to rise to the skin in an essentially perpendicular fashion, preventing angulation which may later make catheterization difficult. A small skin plug is excised using the flat end of a 10-cm 3 syringe plunger as a template. The anterior rectus fascia is incised vertically for 2 to 3 cm and the muscle fibers retracted. Three no. 1 PGA horizontal mattress sutures are placed through the anterior rectus fascia on the lateral, inferior, and medial portions of the incision, and affixed in their corresponding spots on the PGA collar of the efferent limb. Care must be taken to ensure that the efferent limb is not twisted when the sutures are placed and that the sutures do not become crossed while being passed through the abdominal wall and secured to the collar. A narrow 2- to 3-cm strip of Marlex mesh is secured with no. 1 PGA to the abdominal wall superior and lateral to the cephalad extent of the fascial incision. This is then passed through the mesenteric window of the efferent limb (window of Deaver) which accommodates the PGA mesh collar, forming a sling that supports the mesentery to the efferent limb. A Babcock clamp guides the free distal end of the efferent limb through the stomal opening, with care taken not to cross sutures or wrap the ileum in suture. The PGA sutures are tied, fixing the PGA mesh to the anterior rectus fascia. The medial free edge of the Marlex “strut” is then sutured to the abdominal wall medial to the mesentery, completing a four-quadrant fixation of the efferent limb of the reservoir that facilitates catheterization and prevents parastomal herniation. The catheterizable ileal limb is stretched to the skin level to guarantee a straight catheterization channel, and the redundant ileum above skin level is excised. A flush stoma is constructed circumferentially with 3-0 PGA suture. Drainage of the reservoir during the healing phase can be performed in several ways. A 30-Fr Medina tube through the stoma into the reservoir has been traditional. In a tapered efferent limb, a 30-Fr tube is precluded and instead a capped 18-Fr catheter is left through the stoma to fix the efferent limb in a straight position during the scar formation of the early postoperative period. The ureteral stents are sutured to a 22-Fr catheter that is exteriorized through the skin and anterior pouch wall, draining the reservoir. Alternative methods of managing the distal ureteral stents include exteriorization by bringing them through the anterior wall of the reservoir and through a skin stab wound, creating a dry reservoir during the healing process or leaving them within the lumen of the reservoir for later endoscopic retrieval. Mucous irrigation can be performed in an in-and-out fashion if needed between the two catheters. The catheters are secured to the skin with nonabsorbable suture after assuring easy irrigation of the reservoir prior to, and subsequent to, abdominal closure. A doubled 1-in. Penrose drain is placed through a separate stab wound in the lower abdomen and is positioned in a dependent portion of the pelvis. The drain should not lie on suture lines of the reservoir because it may migrate into the lumen of the pouch. Fixing it to the psoas muscle or peritoneum with a 4-0 chromic will prevent migration of the drain. Postoperative Care The most important aspect of early postoperative care is adequate, unobstructed drainage. The reservoir is irrigated with 60 to 120 cm 3 of normal saline every 4 hours to prevent mucous obstruction of the catheters. Patients learn to manage their catheters, Penrose drain, and irrigation and are discharged when bowel function returns. They return 3 weeks after surgery for removal of catheters and drains if intravenous pyelogram and reservoir x-rays reveal no leakage. The initial catheterization regimen is every 2 hours, increasing by 1 hour each week until the goal of every 6-hour catheterization is achieved. A small absorbent pad worn over the stoma site prevents mucous soiling of clothing.

OUTCOMES
Complications The early complication rate (10% to 15%) is similar for patients undergoing single-stage cystectomy and Kock pouch construction, conversion of an ileal conduit to a Kock pouch, or construction of an ileal conduit at the time of cystectomy. The overall incidence of late complications related to the reservoir is 10% to 15%. 6,7 It is convenient to segregate reservoir-related complications into those related to the cutaneous continence limb, the reservoir itself, and the afferent antireflux limb. The most common and confounding problem associated with the cutaneous continence limb remains incontinence. With all of the modifications to nipple construction over the years, incontinence can be expected to occur in 5% to 10% of patients due to an extussuscepted nipple, a fistula, a nipple of inadequate length, or other rare causes. Parastomal herniation rarely occurs if the four-quadrant technique described here is utilized. The problem of Marlex erosion described in the early literature has disappeared since Marlex mesh is no longer used to stabilize the valves. Difficult catheterization remains a problem in the obese patient, but tapering the catheterization limb should minimize this difficulty. Problems related to the storage portion of the reservoir can be divided into metabolic and nonmetabolic. Metabolic complications include solute absorption disorders due to urine coming into contact with the absorptive bowel mucosa for long periods between catheterizations, and metabolic alterations related to the removal of segments of bowel from the gastrointestinal tract. Although these topics are beyond the scope of this discussion, they have been well reviewed. 5 The bowel mucosa atrophies over chronic exposure to urine so that long-term metabolic disturbances, such as metabolic acidosis with its potential demineralizing effects on bone, are rare if renal function is normal. Vitamin B 12 replacement may be necessary after 5 years. The most common nonmetabolic complication associated with the reservoir is calculus formation, seen in 5% to 10% of patients. Vigilance is needed to diagnose these calculi, and they can be managed by outpatient endoscopic techniques. 3 Spontaneous rupture has been reported, and this uncommon complication can be fatal if not recognized. Complications associated with the afferent antireflux limb includes ureteroileal anastomotic stricture in 3%, stenosis of the valve in 3%, and asymptomatic reflux. afferent valve, however, has been remarkably free of significant complications. Results Patients are encouraged to return to a normal lifestyle with minimal restrictions after complete healing has occurred. Patient satisfaction has been tremendous but depends on realistic expectations from the outset. 1 Patient counseling should include a discussion that makes clear that the choice of urinary diversion is a lifestyle choice unlikely to significantly impact survival (which will be determined by the disease necessitating bladder replacement). The patient should not expect the reservoir to perform like a native bladder. Nonetheless, quality-of-life surveys demonstrate an advantage for patients with continent diversion compared to conduit diversion in terms of body image, sexual life, and personal satisfaction. A properly motivated patient with realistic expectations is likely to benefit most from the procedure. CHAPTER REFERENCES
1. Boyd SD, Feinberg SM, Skinner DG, Lieskovsky G, Baron D, Richardson J. Quality of life survey of urinary diversion patients: comparison of ileal conduits versus continent Kock ileal reservoirs. J Urol 1987;138:1386–1389. 2. Elmajian DA, Stein JP, Esrig D, et al. The Kock ileal neobladder: updated experience in 295 male patients. J Urol 1996;156:920–925. 3. Ginsberg D, Huffman JL, Lieskovsky G, Boyd SD, Skinner DG. Urinary tract stones: a complication of the Kock pouch urinary diversion. J Urol 1991;145:956–959. 4. Kock NG, Nilson AE, Nilsson LO, Noden LJ, Philipson BM. Urinary diversion via a continent ileal reservoir: clinical results in 12 patients. J Urol 1982;128:469–475. 5. McDougal WS. Metabolic implications and electrolyte disturbances. In: Webster GD, Goldwasser G, eds. Urinary diversion: scientific foundations and clinical practice. 1st ed. Oxford: Isis Medical Media, 1995;32–44. 6. Skinner EC, Lieskovsky G, Boyd SD, Skinner DG. Continent cutaneous diversion and total bladder replacement using the Kock principles. Rec Adv Urol Androl 1991;5:135–147. 7. Skinner DG. Intussuscepted ileal nipple valve; development and present status. Scand J Urol Nephrol (Suppl) 1992;142:63–65. 8. Skinner DG, Lieskovsky G. Technique of radical cystectomy. In: Skinner DG, Lieskovsky G, eds. Diagnosis and management of genitourinary cancer. Philadelphia: WB Saunders, 1988;607. 9. Stein JP, Freeman JA, Esrig D, et al. Complications of the afferent antireflux valve in the Kock ileal reservoir. J Urol 1996;155:1579–84. 10. Stein JP, Stenzl A, Esrig D, et al. Lower urinary tract reconstruction following cystectomy in women using the Kock ileal reservoir with bilateral ureteroileal urethrostomy: initial clinical experience. J Urol 1994;152:1404–1408.
9

The

Chapter 79 Right Colon Reservoir Glenn’s Urologic Surgery

Chapter 79 Right Colon Reservoir
Jorge L. Lockhart

J. L. Lockhart: Division of Urology, Harborside Medical Center, Tampa, Florida 33606.

Anatomic, Physiologic, and Urodynamic Considerations Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Care Reservoir Preparation Ureteral Intestinal Reimplantation Antiincontinence Mechanism Conversion From an Ileal Conduit to A Continent Urinary Reservoir Postoperative Care Outcomes Complications Results Chapter References

Attempts to develop a continent urinary reservoir have been studied since 1950, when Gilchrist and Merricks described their original cecal bladder. 2 An important step for the popularization of continent urinary reservoirs followed Lapides's experience with intermittent catheterization, demonstrating both safety and patient acceptance.4 Following the original report of 12 patients undergoing the Kock procedure, Skinner and colleagues focused on improvement of the surgical technique. Ten other authors, including ourselves, adopted other modified techniques utilizing the ileocecal segment (Indiana, Mainz, Florida, Miami).

ANATOMIC, PHYSIOLOGIC, and URODYNAMIC CONSIDERATIONS
The extended right colonic segment includes the cecum, ascending colon, hepatic flexure, and right half of transverse colon. The blood supply is provided by branches of the superior mesenteric artery. These arteries are the middle colic (midtransverse colon), right colic (hepatic flexure), and ileocolic (cecum and distal ileum). The venous drainage is characterized by veins that accompany the arteries and drain into the portal system (superior mesenteric vein). This colonic segment has two primary functions: storage of chyme and water, and electrolyte absorption. The amount of water absorbed represents a small amount of the total absorbed in the intestinal tract. The motility of the right colon is characterized by both a segmental and a mass type of activity; however, the transit time is much slower than in the small bowel, which emphasizes the value of right colon as a storage organ. This organ harbors a large number of bacteria, among which the Bacteroides fragilis (anaerobe) and Escherichia coli (aerobe) are the most common, emphasizing the need of an adequate bowel preoperative preparation. Postoperative urinary continence depends on a suitable balance between reservoir pressure and outlet resistance. Incontinence occurs if there is an imbalance between both systems and ensues if there is a high-pressure, small-capacity reservoir or an inadequate outlet resistance. For the above reasons, detubularized intestine should be used to create a large-capacity reservoir and decrease its intraluminal pressure. In our patients, the long-term follow-up of reservoir pressure recording has demonstrated that involuntary bowel contractions tend to diminish in amplitude. Therefore, incontinence can improve with time if these patients are properly followed provided there is adequate resistance at the antiincontinence segment level.

DIAGNOSIS
The right colonic reservoir is usually performed in conjunction with a cystectomy and as such the diagnostic modalities relate to the indications for cystectomy (see Chapter 23 and Chapter 24). All physiologic and medical abnormalities should be diagnosed and treated preoperatively. The patient should completely understand the magnitude and risks of the procedure and should be assessed for their compliance and dexterity. The rigors of time, intermittent catheterization and periodic pouch irrigation should be explained.

INDICATIONS FOR SURGERY
Indications for a right colonic reservoir include the absolute or functional loss of the bladder. Contraindications for the procedure include the following: (a) poor understanding of or compliance with the procedure; (b) inability to properly maneuver upper extremities; (c) high risk from the surgical point of view; (d) inflammatory bowel disease, large bowel malignancy, or previous history of multiple bowel ablative procedures; (e) low-weight myelomeningocele, with history of bowel ablative procedures, with or without history of diarrhea. In adult patients with bowel disease or previous bowel resection, I obtain a gastroenterology consultation for colonoscopy and occasionally for a barium enema. A patient with an ileal conduit diversion, without another bowel segment previously resected, can be converted to a continent colonic diversion. 2,4,8,10 I do not routinely perform a barium enema on children unless there is a specific indication. Children with myelomeningocele and low body weight or young adults with previous bowel ablation procedures associated with the elimination of the ileocecal valve present a well-known risk for the development of postoperative diarrhea with utilization of the ileocecal segment. In some of these patients we have recommended diversion utilizing a composite gastrointestinal segment. 6

ALTERNATIVE THERAPY
Alternatives to the right colonic reservoir include other forms of urinary diversion including ureterosigmoidostomy ( Chapter 76), ileal or colonic conduits ( Chapter 77), continent urinary diversions ( Chapter 78, Chapter 80), or orthotopic bladders (Chapter 81, Chapter 82 and Chapter 83).

SURGICAL TECHNIQUE
Preoperative Care The patient should arrive at the operating suite in a satisfactory nutritional and hydroelectrolytic balance. An intravenous central line should be initiated the night before the surgical procedure. Bowel preparation should include a low-residue diet, antibiotics (neomycin, erythromycin), and a mechanical clean-out. Golytely solution the day prior to surgery accomplishes the latter function. Preoperative stomal position is not as essential as with an incontinent appliance diversion. An area where the patient would perform intermittent catheterization without difficulty should be selected for stomal placement. Reservoir Preparation With the patient in the supine position, a midline incision should be made between 5 cm above the umbilicus and the symphysis pubis. The incision may be extended

upward if difficulties are encountered with the colonic mobilization or if changes due to multiple adhesions or radiation therapy in the lower abdomen exist. After exploration of the abdominal contents, any necessary pelvic cavity ablative procedure should be performed first. Both ureters should be isolated and mobilized, taking special care to preserve the adventitia. The left ureter is brought behind and above the mesosigmoid to the right hemiabdomen; it is important to avoid bringing the ureter below the sigmoid vessels to prevent its kinking. This maneuver is facilitated if a 4-0 silk suture is left attached to its lower end. Attention is then focused on the cecum and distal ileum, where 10 to 12 cm of distal ileum is freed from adhesions while the bowel is transacted proximally utilizing GIA-60 nonabsorbable staples. The blood supply of the distal ileum is provided by the ileal branch of the ileocecal artery. This intestinal segment is used for the construction of the antiincontinence mechanism. The right colon, hepatic flexure, and right half of transverse colon are mobilized after the incision of the parietal peritoneum and gastrocolic ligament. Care should be taken to preserve its blood supply while transecting the right colon (ileocolic and right colic arteries) with similar GIA-60 staples. The middle colic artery should remain with the left transverse mesocolon to prevent ischemia in the left half of the transverse and the splenic flexure in the event of future need of left colon resection ( Fig. 79-1). The bowel stream is reconstituted through a laterolateral ileotransversostomy using routine stapling techniques. Care should be taken to suture the mesentery to the transverse mesocolon to prevent internal hernias postoperatively.

FIG. 79-1. Division of transverse colon. Care is taken to preserve middle colic artery.

The end of the transected transverse colon is folded toward the right gutter and shaped as an inverted U. Both bowel limbs can be grasped with Allis clamps or approximated with silk sutures as depicted in Figure 79-2. The staple line from the transverse colon end is removed and an opening of approximately 3 in. is created in the cecum (Fig. 79-3). A Polysorb 75 (absorb-able) stapler is utilized to suture both bowel limbs together ( Fig. 79-4).7 With the reservoir mucosa exposed, a small incision (2 cm) is made at the end of the staple lines to allow the insertion of the absorbable stapler and continue the bowel limbs suturing ( Fig. 79-5). The Polysorb 75 stapler is again inserted and the suturing between both bowel limbs is completed. As a rule, three Polysorb staplers are utilized to complete the detubularization of the reservoir (Fig. 79-6 and Fig. 79-7). With the reservoir open at its lower end, the ureters are brought in for reimplantation.

FIG. 79-2. Approximation of limbs of bowel as part of detubularization.

FIG. 79-3. Opening in bowel approximately 3 in. from cecum.

FIG. 79-4. Stapling of bowel limbs with Polysorb GIA stapler.

FIG. 79-5. A small incision must be made at the end of the staple line to allow further stapling.

FIG. 79-6. Usually three Polysorb staplers are utilized to complete detubularization.

FIG. 79-7. Completion of detubularization.

Ureteral Intestinal Reimplantation Authors differ on the ureteral intestinal reimplantation technique. The main concern is ureteral intestinal obstruction, which can occur after pouch distention, lower ureteral stretching, and fibrosis in the weeks following surgery. Reflux has proven to be non-deleterious in these large-capacity, low-pressure reservoirs. 3 In one of the larger series, the Leadbetter combined technique is recommended. 9 The technique is extraintestinal and combines a direct mucosa-to-mucosa anastomosis with a submucosal tunnel. We have utilized a direct transintestinal reimplantation as described by Goodwin. 3 In some situations, however, the extraintestinal approach can be utilized at the discretion of the surgeon. For creation of a direct anastomosis, and with the bowel mucosa exposed, the reservoir wall is perforated with a small hemostatic clamp. The ureter is brought intraluminally (Fig. 79-8). Care should be taken to avoid ureteral kinking or twisting, mainly during the left ureteral manipulation. The ureter is spatulated for at least 2 cm and fixed to the submucosa with four quadrant sutures using 5-0 absorbable material. Four or five sutures are placed between the end of the ureter and the bowel mucosa (Fig. 79-9). Single-J ureteral stents, 7.5 mm in diameter, are placed in both ureters. Both catheters are fixed to the bowel mucosa with 3-0 absorbable sutures. A 22-Fr Malecot catheter is left indwelling in the reservoir and is brought out through a separate incision in the reservoir's anterior wall. The purpose of leaving a Malecot catheter is for postoperative irrigation of clots and mucus, and for urine drainage until the reservoir suture lines are completely healed. A purse string suture with 3-0 chromic material is placed where the Malecot exits from the pouch. The ureteral catheters can be brought out through the same Malecot catheter opening, through the stoma, or can be transected at 4 cm from the exit from the ureteral orifices and sutured to the Malecot catheter with nonabsorbable suture material. This maneuver would eliminate the need for postoperative care of the ureteral stents. However, in some difficult cases in which renal failure occurs, the lack of access to the ureteral catheter prevents injection of contrast to determine the real cause for the problem. The reservoir can be closed as shown on Figure 79-10 or with a running and locking 3-0 Vicryl suture.

FIG. 79-8. Reimplantation of ureter.

FIG. 79-9. Anastomosis of ureter to bowel mucosa.

FIG. 79-10. Closure of reservoir (absorbable staples).

Antiincontinence Mechanism The Indiana group pioneered the bowel plication concept as an antiincontinence mechanism for continent urinary reservoirs. 9 The main idea is to reinforce the ileocecal valve using bowel wall infolding sutures. Using a similar concept, we reported our plication technique. 5 Similar to Rowland's technique, the last 10 cm of ileum is left attached to the reservoir and plicated around a 10- or 12-Fr Robinson catheter. In our procedure, the bowel infolding is performed with interrupted, nonabsorbable sutures placed longitudinally in the ileum, ileocecal valve, and cecal wall ( Fig. 79-11). A second plicating suture row is placed in the opposite bowel side at 12 o'clock from the first one ( Fig. 79-12). A 10- or 12-Fr Robinson catheter is passed several times to test the ease of intubation and the catheter is withdrawn. An alternative preparation of the antiincontinence mechanism utilizing staples has been reported. 1 With a red Robinson no. 14 Fr catheter in place, the antimesenteric bowel edge is grasped with Allis clamps; two GIA-60 stapling devices are utilized to transect the bowel between catheter and clamps leaving the catheterizable segment around the red Robinson catheter ( Fig. 79-13 and Fig. 79-14). Other alternatives to strengthen the ileocecal valve have been utilized by different authors. Rowland et al. utilized plicating sutures in the cecum and opposite to the stapled antiincontinence segment ( Fig. 79-15). When this technique is utilized, we also place infolding sutures at the base of the antiincontinence segment in an attempt to further reinforce the ileocecal valve ( Fig. 79-16 and Fig. 79-17).

FIG. 79-11. Bowel infolding technique to achieve continence. A layer of nonabsorbable sutures is placed longitudinally in the ileum, ileocecal valve, and cecal wall.

FIG. 79-12. A second plication layer is placed opposite to the first.

FIG. 79-13. Allis clamps are used to guide GIA stapling of catheterizable segment over a red Robinson catheter.

FIG. 79-14. Construction of tubular segment of catheterizable segment using GIA stapler.

FIG. 79-15. Completion of catheterizable segment with red Robinson catheter in place.

FIG. 79-16. Plication of cecum over catheterizable segment.

FIG. 79-17. Completion of plication.

More recently, we initiated an ileocecal valve reinforcing technique in reoperations for a variety of failed antiincontinence systems (appendix, reimplanted and tapered ileum, ileocecal valve). Following the stapling of the antiincontinence mechanism, a small window is created in the mesentery for a distance of approximately 2 to 3 in. (Fig. 79-18). The cecum is grasped at both sides of the base of the tapered ileal segment with Allis clamps and is brought around the ileocecal valve ( Fig. 79-19). Then both ends are sutured with four 3-0 silk sutures ( Fig. 79-20). This technique has been utilized in reoperations and is now routinely performed in the new cases with excellent results.

FIG. 79-18. Window in mesentery created 2 to 3 in.

FIG. 79-19. Cecal segment brought through mesenteric window.

FIG. 79-20. Cecal segments sewn to encircle catheterizable segment.

The integrity of the antiincontinence mechanism is tested after pouch filling through the Malecot with 250 to 300 ml of saline. The ileal segment is brought out through the skin through the abdominal wall. Some surgeons prefer to bring out the ureteral stents through the stoma in order to facilitate postoperative care because urine will be collected in an urostomy bag. The ileal segment is brought out to the skin through the rectus sheath or umbilicus ( Fig. 79-21). Bringing out the ileal segment through the abdominal wall weak area (lateral to the rectus sheath through Spiegel's line) increases the incidence of parastomal hernias. Patients with a wide, well-vascularized distal ileal opening can be brought out to the skin through a circular stoma. If the distal ileum's diameter is relatively small or suspected to be poorly vascularized, the bowel is spatulated and the stoma is constructed utilizing a Y-V plasty technique. A laterally based skin triangle is elevated and defatted ( Fig. 79-22 and Fig. 79-23). The antiincontinence mechanism construction is completed with both ureteral stents coming out per stoma ( Fig. 79-24 and Fig. 79-25).

FIG. 79-21. Possible sites for stoma.

FIG. 79-22. Anatomic landmark relations to location of stoma in the right lower quadrant.

FIG. 79-23. Skin triangle developed for stoma.

FIG. 79-24. Catheterizable segment brought through the stoma site.

FIG. 79-25. Completed stoma.

Conversion from an Ileal Conduit to a Continent Urinary Reservoir A similarly prepared detubularized right colonic segment is positioned on top of the conduit and is sutured to the conduit utilizing Polyabsorb 75 GIAs as needed. Figure 79-26 shows the colon already stapled to the conduit and the antiincontinence mechanism to be brought to the abdominal wall. Figure 79-27 shows the final coloileal preparation.

FIG. 79-26. Conversion of ileal conduit to continent reservoir, with colon stapled to conduit.

FIG. 79-27. Conversion of ileal conduit to continent reservoir. Completed configuration.

Postoperative Care Immediately after surgery the patient is admitted to the intensive care unit. Precautions are taken to maintain circulating volume, hemoglobin levels, and hydroelectrolytic balance, which are essential in the immediate postoperative period. Prophylaxis for pulmonary infections and venous embolism are carried out. Intravenous antibiotics are maintained until the initiation of oral feedings; then they are switched to oral antimicrobials. I prefer to use third-generation cephalosporins with adequate gram-negative and anaerobe coverage. Patients undergoing a simultaneous cystectomy, who are expected to have a greater morbidity from the procedure, are initiated on patyrenteral hyperalimentation. The Malecot and ureteral catheters are irrigated twice daily to facilitate urine drainage. It is essential to irrigate the mucus from the reservoir cavity in the next two or three postoperative weeks. After return of bowel function, the ureteral stents are removed. Patients who live in our community are sent home and readmitted in 1 week for pouch activation. Other patients are maintained in the hospital until they perform intermittent catheterization satisfactorily. Pouch activation represents Malecot catheter clamping and initiation of intermittent catheterization. A baseline pouchogram and intravenous pyelogram are obtained. This is initially performed by well-trained nurses who also begin the patient's follow-up instruction. The next day, the Malecot catheter is removed and the patient is discharged from the hospital.

OUTCOMES
Complications The most ominous complication with this procedure has been the development of ureteral-intestinal obstruction. When this complication occurs, initial percutaneous nephrostomy and internal stenting for 6 to 8 weeks is all that is required in approximately 50% of cases. The rest will need re-reimplantation, and in this situation a

trans-reservoir approach facilitates the procedure. Prior radiation therapy to the pelvis, though not a contraindication to this procedure, is associated with a higher incidence of wound infection and ureteral obstruction. Other postoperative complications include persistent postoperative diarrhea and metabolic abnormalities. Patients with persistent diarrhea will generally require gastroenterologic assistance for its management. The presence of intermittent acidosis has required treatment with Bicitra solution. Problems specific to the reservoir found in long-term follow-up include parastomal hernias and stone formation. Parastomal hernias require surgical correction and for that purpose we have achieved better results with takedown of the antiincontinence segment, internal closure of the hernia, and stoma repositioning to the abdominal wall. Stone disease in the reservoir can be managed better with percutaneous lithotripsy and, in the case of large stones, through an open lithotomy. Results We reported our initial results in 1990. A total of 92 patients underwent continent urinary diversion with an extended, detubularized right colonic segment as the urinary reservoir and the distal ileum as a continent catheterizable efferent system. In this series, 65 patients were followed for 6 to 46 months (average 17 months). Our reservoir allows the accommodation of a large volume of urine; urodynamic studies in 28 patients demonstrated a maximum reservoir capacity varying between 550 and 1200 cm3 (average 747 cm3). Maximal reservoir pressures ranged from 10 to 58 cm H 2O (average 35 cm H2O). Of the 127 ureterocolonic reimplantations, 4 ureters were initially reimplanted with a modified Le Duc procedure, 26 ureters were managed subsequently with the Goodwin transcolonic approach, and 91 reimplantations were done with a direct (nontunneled) mucosa-to-mucosa anastomosis. The overall success rates with each of the three techniques (absence of reflux and obstruction) were 75%, 88.6%, and 90.1%, respectively. Six megaureters underwent imbrication and direct reimplantation, and three of these (50%) became obstructed. Two converted ileal conduits were opened at the antimesenteric edge and were patched to the reservoir whereas the ureteroileal anastomosis was left undisturbed. One patient (1.5%) died of pulmonary embolism. Medical and surgical complications occurred only in the group who underwent simultaneous cystectomy and the overall rate of complication was comparable to previous series with ileal conduits. The double-row plication of the distal ileum and ileocecal valve allows for easy catheterization every 4 to 6 hours, and 63 patients (97%) remain continent between catheterization. Four patients (6%) required reoperation for correction of incontinence or another complication. Our satisfactory experience with these patients makes this technique an excellent approach to achieving continent urinary diversion. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Bejany DE, Politano VA. Stapled and non-stapled tapered distal ileum for construction of a continent colonic urinary reservoir. J Urol 1988;140:491–494. Gilchrist RK, Merricks JW, Hamlin HH, Rieger IT. Construction of substitute bladder and urethral. Surg Gynecol Obstet 1950;90:752–760. Helal MA, Pow-Sang JM, Lockhart JL, Sanford EJ, Figueroa TE. Direct (non-tunnelled) uretero-colonic reimplantation in association with continent reservoirs. J Urol 1993;150:835–837. Lapides J, Diokno AC, Gould FR, Lowe BS. Further observations on self catheterization. J Urol 1976;116:169–171. Lockhart, JL. An alternative method for continent supravesical diversion. Soc Pediatr Urol Newslett 1987;3:18–20. Lockhart JL, Davies R, Cox C, McAllister E, Helal M, Figueroa TE. The gastroileal pouch: an alternative continent urinary reservoir for patients with short bowel, acidosis and/or extensive pelvic radiation. J Urol 1993;150(l):46–50. Olsson CA, Kirsch AJ, Whang MIS. Rapid construction of right colonic pouch. Curr Surg Tech Urol 1993;6:3:1–7. Pow-Sang JM, Helal MA, Figueroa TE, Sanford EJ, Persky L, Lockhart JL. Conversion from external appliance wearing or internal urinary diversion to continent urinary reservoir (Florida Pouch I and II): surgical technique, indications and complications. J Urol 1992;147:356. Rowland RG, Mitchell ME, Birhle R. The cecoileal continent urinary reservoir. World Urol 1985;3:185–188. Skinner DG, Boyd SD, Lieskovsky G. Clinical experience with the Kock continent ileal reservoir for urinary diversion. J Urol 2985;132:1101–1107.

Chapter 80 Mitrofanoff Continent Urinary Diversion Glenn’s Urologic Surgery

Chapter 80 Mitrofanoff Continent Urinary Diversion
Hubertus Riedmiller and Elmar Werner Gerharz

H. Riedmiller: Department of Urology, Julius Maximilians-University Medical School, 97080 Wurzburg, Germany. E. W. Gerharz: Department of Urology, Julius Maximilians-University Medical School, 97080 Wurzburg, Germany, and The Institute of Urology and Nephrology, University College London Medical School, London W1P 7PN, United Kingdom.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

In 1980 the French urologist Paul Mitrofanoff rediscovered the vermiform appendix as a catheterizable channel for emptying a urinary reservoir and for the first time described its attachment with a flap-valve antirefluxing technique to the bladder in children with meningomyeloceles. With its vascular mesenteric pedicle preserved, the proximal end of the isolated appendix was brought out to the skin, allowing clean intermittent catheterization (CIC). 7 But Mitrofanoff not only initiated an up-to-date persisting renaissance of the appendix in urologic reconstructive surgery; he also lent his name to a far-extended, well-defined concept applicable to a wide variety of urologic conditions. 13 The Mitrofanoff principle encompasses two distinct requirements: 1. A small-caliber supple conduit, most frequently the appendix, is brought to the skin as a catheterizable stoma facilitating catheterization by avoiding kinking and coiling catheters. 2. An antirefluxing connection of this conduit to the reservoir with a portion of the conduit placed in a submucosal tunnel provides continence by a highly effective flap-valve mechanism. To allow an adequate function those requirements have to be combined with the factors formulated by Snyder including a low-pressure urinary reservoir with adequate storage volume, nonrefluxing ureteral reimplantation, and successful emptying of the reservoir by CIC. The high-capacity reservoir is important to maintain a socially acceptable interval between catheterizations and is created using detubularized gastrointestinal segments for either augmentation cystoplasty or de novo construction if the native bladder is inadequate or absent. The nonrefluxing ureteral implantation is important for protection of the upper urinary tract integrity and renal function. Reliable low-pressure and complete evacuation of the reservoir by CIC permits the storage cycle to recommence. 11 Only 6 years after Mitrofanoff's report the original Mainz pouch technique was presented, 12 and the introduction of the submucosally embedded in situ appendix as the continent outlet since 1990 has simplified the surgical technique. This has greatly increased the acceptance of the Mainz procedure, developing it into an almost ideal realization of the above-described reconstructive “philosophy.” 6 Being the most intriguing modification of the Mainz pouch, its appendiceal configuration finally contributes to the wide array of successful Mitrofanoff variations: shorter excluded bowel segments, significant reduction of operation time (simple procedure), minimized risk of stone formation (no staples), and perfect continence with easy catheterization. 8 With a significant decrease of specific morbidity and mortality, continent intestinal reservoirs with cathe-terizable cutaneous stoma are no longer surgical curiosities but rather standardized indispensable constituents of the reconstructive urologist's armamentarium with large personal and institutional series throughout the world. 1,9,13

DIAGNOSIS
As this technique is for urinary diversion, it is usually an adjunct to the treatment of the underlying pathology that resulted in the loss of the bladder and/or sphincteric function. The diagnostic modalities therefore should be directed at the underlying pathology and definitive treatment. Since in every case one should attempt to use any previously existing material, such as conduits or native bladder remnants, preoperative anatomic evaluation must result in a maximum of information. Excretory urography, voiding cystourethrography, loopogram, urethrography, endoscopy, and retrograde injection of ureteral stumps (if feasible) are therefore mandatory investigations especially in children with complex abnormalities and in patients with multiple prior surgeries. The functional assessment should include isotope renography, filling and voiding cystometrography, and upright cystography. An overall assessment of intelligence, manual dexterity, education, and social support is of paramount importance.

INDICATIONS FOR SURGERY
Avoiding the need for external collection devices, continent urinary reservoirs have proved to be advantageous with respect to all issues directly related to the stoma, when compared to a conduit diversion. Significant superiority of continent diversion in the patients' global self-assessment of their quality of life, physical strength, mental capacity, leisure time activities, and social competence as indicators of enhanced vital power support our understanding that especially younger women and men do benefit from a Mitrofanoff solution and therefore should be offered this surgical option. 3 The appendix seems to be the most satisfactory structure for the creation of a catheterizable stoma and the placement in the umbilicus is recommended for cosmetic and functional reasons. 1,2,7,8,10,12,13 Whereas the most common indication for continent urinary reconstruction in adults has been in bladder replacement after anterior pelvic exenteration for malignant disease with curative intent, the predominant applications of the Mitrofanoff principle in children are inaccessible urethral orifice, discomfort in catheterization, or difficult incontinence due to simple or complex genitourinary abnormalities. 2,9,13 Moreover, continent urinary diversion increasingly becomes a consideration for patients in whom an abnormality of the lower urinary tract has resulted in renal failure or a functioning bladder is lacking for other reasons. 4 We use the Mitrofanoff principle as the continence mechanism in a primary therapeutic approach as well as in conversion and salvage maneuvers after failure of previous reconstructive surgery. The major advantage of the submucosally embedded in situ appendix, highly effective continence, may turn out to be its main drawback: once the bladder outlet is closed the patient is entirely dependent on CIC of the reservoir. If not emptied in a timely fashion, sufficiently high pressures may be generated within the reservoir to cause rupture with potentially fatal consequences. To achieve satisfactory results careful patient selection, continuing education, and meticulous compliance on the patient's part are essential. It is therefore desirable to have highly motivated patients with realistic expectations and normal intelligence who are physically and emotionally capable of dealing with the strict regimen of CIC.

ALTERNATIVE THERAPY
The appendix may not be available, if surgically removed beforehand or unsuitable for use in reconstructive surgery. In these cases the surgeon must be familiar with appropriate alternative methods. In 6 cases out of our 95 nonappendectomized patients (6.3%), the vermiform appendix could not be used to create a continence mechanism due to complete fibrotic obliteration or insufficient diameter. Other authors observed higher rates of unsuitability (17% to 31%). In these cases, several alternatives have been described including ureter, tapered ileum, and fallopian tube. 13 In the ileocecal reservoir there are basically two surgical options: forming an isoperistaltic ileoileal intussusception valve ( Chapter 78), a technique with a long individual learning curve and a relatively high complication rate. Another option would be to mimic the appendix by construction of either a seromuscular bowel flap tube (pedicled island flap of large bowel) or a full-thickness bowel flap tube as described by Lampel et al. 6

SURGICAL TECHNIQUE
In contrast to Mitrofanoff's original technique, in the appendiceal configuration of the Mainz pouch the appendix is left in situ. The excluded bowel segment is divided on the antimesenteric border leaving the lower 5 cm of the cecum (cecal pole) tubularized and intact. The opened bowel loops are sewn together thus creating a low-pressure and high-capacity reservoir. Both ureters are implanted in the large bowel segment of the pouch plate, forming long submucosal tunnels for reflux prevention. In some cases the appendix must be dilated to accommodate a 16- to 18-Fr catheter to assure sufficient postoperative drainage of mucus. After cutting of the distal end of the appendix, dilation is easily achieved using a set of slightly curved bougies. Analogous to the Lich-Gregoir procedure for vesicoureteral reflux, the seromuscular layer of the intact cecal pole is incised along the tenia libera for approximately 5 cm in length down to the mucosa. The incision should reach directly to the appendicular base and is extended like an inverted Y to both sides of the appendix, thus creating small seromuscular flaps. By careful dissection of the seromuscular tissue with small scissors, a broad submucosal bed (2 to 2.5 cm wide) for the appendix is created. The appendicular mesentery is freed of its excessive fatty tissue. Windows in the mesoappendix are created between the branches of the appendicular artery without compromising the blood supply ( Fig. 80-1). Anatomic variations of the appendicular artery ( Fig. 80-2) have to be respected and an additional branch of the anterior or posterior cecal artery supplying the base of the appendix should be preserved. After the appendix is correctly positioned into its submucosal bed the seromuscular layer is closed over the embedded in situ appendix through the mesoappendicular windows with interrupted 4-0 polydioxanone sutures ( Fig. 80-3 and Fig. 80-4). The seromuscular flaps described above are closed over the appendicular base and guarantee a safe and firm coverage. A short, mobile portion of the distal appendix remains for creation of the appendicoumbilical stoma (Fig. 80-5). The free portion of the appendix is pulled through the abdominal wall via a small incision at the umbilical area, where even in obese patients the thickness of the ventral abdominal wall is minimal. Anastomosis to the deepest point of the umbilical funnel is performed with six interrupted 4-0 polydioxone sutures on a cutting needle. If no umbilicus is present (as in the case of exstrophy), it is created by tubularizing a V-shaped cutaneous flap connected to the appendicular stump. The pouch is carefully attached to the abdominal wall with nonreabsorbable sutures to prevent the appendix from rotating and kinking and to keep the ileocecal pouch in its original position.

FIG. 80-1. Seromuscular incision along the tenia libera. Creation of the mesenteric windows in the mesoappendix.

FIG. 80-2. Anatomic variations of the appendicular blood supply.

FIG. 80-3. Closure of the seromuscular layer through the mesenteric windows over the embedded in situ appendix.

FIG. 80-4. Intraoperative view of the ileocecal reservoir with continent appendix outlet (mobile portion still has to be shortened).

FIG. 80-5. Correctly embedded in situ appendix with short mobile portion for creation of the appendicoumbilical anastomosis.

Placement of the stoma within the umbilicus is cosmetically very satisfactory and obviates the need of much length of an appendix to bridge a thick abdominal wall. In comparison to other variations of the Mitrofanoff principle in ileocecal reservoirs, submucosal embedding of the in situ appendix avoids excision, reimplantation, and especially 180 degrees rotation of the appendix with the risk of compromising the blood supply. Preserving the anatomic connection between appendix and cecal pole and positioning of the continence mechanism inside the pouch with an extremely short appendicular portion (less than 0.5 cm) outside the reservoir ensures ease of catheterization. In children, of course, one must minimize the use of bowel by appropriately scaling down the length of the intestinal segments recommended for adults.2,9

OUTCOMES
Complications In 93 patients with appendiceal continence mechanisms in a continent urinary pouch followed for a mean of 36.1 months (2.9 to 67.8 months), two major stoma-related complications (2.1%) were observed: due to complete necrosis the appendix had to be replaced by an intussuscepted ileal nipple. Stomal stenosis was observed in 14 patients (15.1%). Although the intermittent passage of the catheter through the conduit dilates the lumen repetitively and may successfully prevent, decrease, or delay encroachment of the appendiceal lumen, it has been recognized that stomal stenosis at the appendicoumbilical junction may be troublesome, especially in the absence of a natural umbilicus such as in patients with bladder exstrophy. In most cases, a gradually stenosing stoma requires simple revision (see below) and remains of adequate caliber. In these patients 20 reinterventions—either simple removal of scar tissue or star-shaped incision (Sachse appendicotomy)—were required. All procedures could be performed with local anesthesia on an outpatient basis. An indwelling catheter was left in place for 1 to 3 days. Although these simple revisions usually lead to good results, the prevention of recurrence is desirable, particularly in patients with a neoumbilical stoma. For that reason, we have developed a cone-shaped metal dilator with a maximum caliber of 24 to 26 Fr. 5 The effective length was chosen to cover only the known critical segment of the tunnel. It is currently applied by 14 of our patients, including 2 men with a neoumbilicus. Directly before inserting the catheter for evacuating the low-pressure reservoir the stoma is gently dilated for a few minutes once or twice per day. With a mean follow-up of 18 months, none of these patients has suffered from recurrent stenosis. A modification of the stomal implantation, described by Woodhouse, avoids stomal stenosis by creation of an oval anastomosis using a V-shaped flap of the umbilical funnel instead of the original circular anastomosis. The long-term complications inherent in continent urinary diversion lined with enteric epithelium are well described. This issue, however, is gaining importance with the number of years a patient has to cope with the physical and mental consequences of an artificial nonphysiologic combination of the gastrointestinal and urinary tract. Whereas in the adult population survival is often determined by the underlying malignant condition, except for a genitourinary abnormality healthy children are facing a close to normal life expectancy. Only a minority of the cancer patients will experience the whole range of late complications of urinary diversion, but in many of the children those complications are perhaps inevitable and may involve serious physical and psychological morbidity, or, even in the worst case, become a limiting factor quo ad vitam. Results At the Department of Urology in Marburg within the last 5 years, 375 patients (352 adults and 23 children) underwent urinary diversion with creation of continent reservoirs in more than 70%. The spectrum of applied techniques included orthotopic bladder substitution, sigma rectum pouch, and ileal or colonic conduits. In 193 patients an ileocecal pouch with umbilical stoma was fashioned. As a continence mechanism the submucosally embedded in situ appendix could be utilized in 93 cases; in 100 cases an ileal intussusception valve was established. In terms of outcome, 97.9% patients with appendicoumbilical stoma are completely continent day and night; easy CIC was initially achieved in 100%. In 11 patients with functional or morphologic bladder loss, a continent urinary reservoir has been created in preparation of kidney transplant. Five patients (aged 11, 14, 16, 31, and 52 years) received a cadaveric donor kidney 5, 6, 8, 9, and 17 months after urinary reconstruction with excellent renal function (creatinine 0.9 to 1.7 mg/dl) at a follow-up of 1 to 48 months (mean 13.8 months). Performed as a staged procedure with urinary diversion as the first step, this strategy successfully combines the advantages of two clinically well-established therapeutic principles obviating external urine collecting devices and chronic hemodialysis. Both in situ appendix and intussuscepted ileum valve are very satisfactory techniques for providing continence in an ileocecal urinary reservoir. Besides the known advantages of the appendix in the primary reconstructive approach—easy and relatively quick procedure, minimized risk of stone formation, and reduced loss of bowel—the treatment of subsequent complications usually is simple. Thus, whenever an appropriate appendix is encountered it should be the intestinal segment of choice in forming a continence mechanism. CHAPTER REFERENCES
1. Duckett JW, Lofti AH. Appendicovesicostomy (and variations) in bladder reconstruction. J Urol 1993;149:567–569. 2. Dykes EH, Duffy PG, Ransley PG. The use of the Mitrofanoff principle in achieving clean intermittent catheterisation and urinary continence in children. J Pediatr Surg 1991;26:535–538. 3. Gerharz EW, Weingärtner K, Dopatka TH, Köhl U, Basler HD, Riedmiller H. Quality of life following cystectomy and urinary diversion: results of an interdisciplinary retrospective study. Urol 1997;158:778–785. 4. Gerharz EW, Köhl U, Melekos MD, Weingärtner K, Riedmiller H. Kidney transplantation into continent urinary reservoirs in patients with absent or non-functioning bladder. Abstracts of the 31st Congress of the European Society for Surgical Research, Southampton, United Kingdom, March 3–4, 1996. 5. Köhl U, Gerharz EW, Weingärtner K, Riedmiller H. Point of technique: a simple device in the prevention of recurrent appendico-umbilical stenosis. Br J Urol 1996;77:603–604. 6. Lampel A, Hohenfellner M, Schultz-Lampel D, Thüroff JW. In situ tunneled bowel flap tubes: two new techniques of a continent outlet for Mainz pouch cutaneous diversion. J Urol 1995;153:308–315. 7. Mitrofanoff P. Cystostomie continente transappendiculaire dans le traitement des vessies neurologiques. Chir Pediatrics 1980;21:297–305. 8. Riedmiller H, Bürger R, Müller SC, Thüroff J, Hohenfellner R. Continent appendix stoma: a modification of the Mainz pouch technique. J Urol 1990;143:1115–1117. 9. Riedmiller H, Gerharz EW. The in situ-appendix modification of the Mitrofanoff principle in ileocecal reservoirs. Dialog Pediatr Urol 1996;19:3–5. 10. Riedmiller H, Gerharz EW. The Mitrofanoff principle in continent urinary diversion. In: Webster GD, Goldwasser B, eds. Urinary diversion. 1st ed. Oxford: Isis Medical Media, 1995;236–243. 11. Snyder HM. Principles of pediatric urinary tract reconstruction: a synthesis. In: Gillenwater JY, Grayhack JT, Howards SS, Duckett JW, eds. Adult and pediatric urology. Vol. 2. Chicago: Year Book, 1987;1726–1781. 12. Thüroff J, Alken P, Riedmiller H, Jacobi G, Hohenfellner R. 100 cases of Mainz pouch: continuing experience and evolution. J Urol 1988;140:283–288. 13. Woodhouse CRJ, MacNeily AE. The Mitrofanoff principle: expanding upon a versatile technique. Br J Urol 1995;74:447–453.

Chapter 81 Orthotopic Urinary Diversion Using an Ileal Low-Pressure Reservoir with an Afferent Tubular Segment Glenn’s Urologic Surgery

Chapter 81 Orthotopic Urinary Diversion Using an Ileal Low-Pressure Reservoir with an Afferent Tubular Segment
Hansjörg Danuser and Urs E. Studer

H. Danuser and U. E. Studer: Department of Urology, University of Berne, 3010 Bern, Switzerland.

Diagnosis Indications Alternative Therapy Surgical Technique Cystectomy Preparation of the Ileal Segment for the Bladder Substitute Construction of the Bladder Substitute and Anastomosis to the Urethra Postoperative Education Outcomes Complications Results Chapter References

This form of orthotopic bladder substitution offers several significant advantages over other forms of diver-sion. One is the ease of surgery, which can be done by any urologist experienced in performing a radical prostatectomy or a cystectomy and ileal conduit. Anatomically, the terminal ileum as well as the ileocecal valve is preserved. The reservoir is spheric, thus achieving a maximum volume/surface area ratio with maximum capacity from a given bowel segment. To avoid metabolic disturbances from reabsorption of urine metabolites, a small surface of intestinal mucosa and a short contact time of the urine with the neobladder mucosa is important. The short ileum segment, 54 to 60 cm long, used to construct this bladder substitute minimizes intestinal malabsorption. On the other hand, according to Laplace's law, a certain reservoir size is mandatory to reduce tension on the neobladder wall and to keep intraluminal pressure low, aiding patients to achieve continence. Another advantage of this bladder substitute is the isoperistaltic tubular afferent segment with the ureteroileal anastomosis at its proximal end. It allows a resection of the distal ureters, including the paraureteral lymphatics, at a safe distance from the bladder cancer and reduces the risk of leaving behind distal ureters that may contain carcinoma in situ. The shorter the ureters are the better the blood supply can be preserved and the lower is the risk of ischemic stricture of the distal ureter. The peristalsis of the afferent ileal segment acts as a dynamic antireflux mechanism. In addition, the simple end-to-side technique of the ureteroileal anastomosis and the omission of an additional antireflux mechanism with a potentially high rate of strictures keeps this risk low. Even if a stricture occurs, the distensibility of the afferent tubular segment allows for bridging unilateral ureteral strictures or necrosis by reanastomosis to the more proximal ureter. In the case of complicated urethral strictures or tumor recurrence in the urethra, the afferent tubular segment can easily be transformed into an ileal conduit.

DIAGNOSIS
In patients with invasive bladder cancer the workup includes bladder biopsies, CT scan of the pelvis and abdomen, chest x-ray, and bone scan to exclude lymph node or systemic metastasis. Urography is done to detect possible tumors of the upper urinary tract. Endoscopy of the urethra should exclude strictures or tumors and verify a normally functioning external urethral sphincter. In addition, biopsies from the prostatic urethra beneath the verumontanum are taken to exclude carcinoma in situ of the urethra, which would force one to perform a urethrectomy and choose another type of urinary diversion. Independent of the indication to cystectomy, renal function must be sufficient with a serum creatinine below 150 to 200 µmol/L (or when it can be expected to fall below this level after cystectomy in cases of obstruction of the upper urinary tract). Otherwise severe metabolic acidosis will occur.

INDICATIONS
The most common indication to use the orthotopic bladder substitute is in patients after radical cystectomy for invasive bladder cancer. This reservoir, with slight modifications, is also useful for bladder augmentation after subtotal cystectomy in patients with a contracted bladder due to neurologic disorders or to replace a shrunken bladder and ureter after radiotherapy, particularly in female patients. Patients who want this orthotopic bladder substitute should be willing to cooperate and to accept and follow the postoperative education instructions. Only then will they achieve urinary continence, void without residual urine, and avoid urinary infections and metabolic disturbances.

ALTERNATIVE THERAPY
Alternative therapies include other types of continent bladder substitutes and pouches, the ureterosigmoidostomy, and the ileal conduit.

SURGICAL TECHNIQUE
All patients receive peri- and postoperative antibiotic prophylaxis consisting of an aminoglycoside and ornidazol for 48 hours and amoxicillin/clavulanate given until all drains are removed 12 to 14 days postoperatively. Heparin is given subcutaneously peri- and postoperatively as thrombosis prophylaxis. Cystectomy Pelvic lymphadenectomy and cystectomy are performed according to standard technique with slight modifications. 3 The external iliac vessels, the obturator fossa, and the hypogastric vessels are freed of all lymphatic, fatty, and connective tissue. Having divided the dorsolateral bladder pedicles containing the superior and inferior vesical vessels along the hypogastric arteries, the pelvic floor fascia is incised and Santorini's plexus is ligated. The ureters are divided where they cross the iliac vessels. This allows en bloc removal of the distal ureters and paraureteral lymphatic vessels, together with the cystectomy specimen. The dorsome-dial pedicle is resected along the pararectal/presacral plane on the tumor-bearing side. Whenever possible, however, care is taken to preserve the hypogastric fibers and the pelvic plexus situated dorsolaterally to the seminal vesicle on the contralateral non-tumor-bearing side. On this side the dissection along the dorsolateral wall of the seminal vesicle is stopped at the base of the prostate. Santorini's plexus is then divided and the membranous urethra transected as close as possible to the apex of the prostate by excavating it from the donut-shaped apex. The neurovascular bundles dorsolateral to the prostate are also preserved. Preparation of the Ileal Segment for the Bladder Substitute For construction of the reservoir, an ileal segment 54 to 60 cm long is isolated approximately 25 cm proximal to the ileocecal valve and bowel continuity is restored (Fig. 81-1). The length of the ileal segment is measured with a ruler in portions of 10 or 15 cm along the border of the mesoileum without stretching the bowel. Irritation of the bowel as well as peridural anesthesia can increase smooth muscle tone and activity and “shorten” the length of the bowel, which will then be too long after muscle relaxation. The distal mesoileum incision transects the main vessels, whereas the proximal mesoileum incision must be short in order to preserve the main vessels perfusing the future reservoir segment ( Fig. 81-1). The mesoileal borders are adapted with a running suture (2-0 polyglycolic acid) in which the mesoileum of the bladder substitute is included ( Fig. 81-2). The stitches must be applied superficially, with care taken to preserve the blood supply to the bladder substitute. Both ends of the isolated ileal segment are closed by seromuscular running sutures (4-0 polyglycolic acid). The distal end of the ileal segment, approximately 40 to 44 cm long, is opened along its antimesenteric border ( Fig. 81-2).

FIG. 81-1. Preparation of the 54- to 60-cm-long ileum segment for the bladder substitute. Note the different incision depth of the mesoileum proximally and distally, in order to preserve the circulation.

FIG. 81-2. Closure of the mesoileum incision. Avoid deep sutures in the area joining the mesoileum of the terminal ileum to the mesoileum of the bladder substitute in order not to compromise circulation.

Ureteroileal End-to-Side Anastomosis The ureters are split over a length of 1.5 to 2 cm and anastomosed by two running sutures using the Nesbit technique in an open end-to-side fashion to two longitudinal 1.5- to 2-cm-long incisions along the paramedian antimesenteric border of the afferent tubular ileal segment, which is 14 to 16 cm long ( Fig. 81-3). The ureters are stented with 7-Fr or 8-Fr catheters. To prevent dislocation of the stents, a rapidly absorbable suture (4-0 polyglycolic acid) is placed through the ureter and stent together 3 to 4 cm proximal to the anastomosis. It is tied very slackly, not compromising the ureteral vasculature. The most distal periureteral tissue is sutured to the afferent ileal segment to neutralize tension on the anastomosis and to cover it. The ureteral stents are passed through the wall of the most distal end of the afferent tubular segment, where it is covered by some meso-ileum. This provides a “covered” canal in the reservoir wall when withdrawing the ureteral stents 5 to 7 days postoperatively.

FIG. 81-3. Ureteroileal anastomosis using a simple end-to-side Nesbit technique with a 4-0 running suture assures a low stricture rate.

Construction of the Bladder Substitute and Anastomosis to the Urethra To construct the reservoir itself, the two medial borders of the opened U-shaped distal part of the ileal segment are oversewn with a single-layer seromuscular continuous 2-0 polyglycolic acid suture ( Fig. 81-4). The bottom of the U is folded over between the two ends of the U (Fig. 81-4 and Fig. 81-5), resulting in a spherical reservoir consisting of four cross-folded ileal segments. After closure of the lower half of the anterior wall and part of the upper half ( Fig. 81-5), the surgeon's finger is introduced through the remaining opening to determine the most caudal part of the reservoir. There a hole, 8 to 10 mm in diameter, is cut out of the pouch wall, outside the suture line (Fig. 81-6). Six polyglycolic acid 2-0 seromuscular sutures are placed between the hole in the reservoir wall and the membranous urethra ( Fig. 81-7). An 18-Fr urethral catheter is inserted before tying the six sutures ( Fig. 81-8). Before complete closure of the pouch, a 10-Fr cystostomy tube is passed through the wall of the pouch where it is covered by some mesoileum ( Fig. 81-9). The cystostomy tube is withdrawn 10 days postoperatively after exclusion of any leakage by pouch radiography. The indwelling catheter is left on continuous drainage for 2 more days before removal to allow for closure of the cystostomy canal in the reservoir wall.

FIG. 81-4. The two medial antimesenteric borders of the opened U-shaped distal ileum segment are oversewn with a single-layer seromuscular running suture. The bottom of the U is folded over and tied between the two ends of the U.

FIG. 81-5. The caudal half of the remaining reservoir opening is closed completely, the cranial half partially by a running seromuscular suture.

FIG. 81-6. A hole is cut into the most caudal part of the reservoir, close to the mesoileum and 2 to 3 cm away from the edge that resulted from cross-folding the ileum segment.

FIG. 81-7. Six seromuscular sutures are placed between the reservoir and the membranous urethra.

FIG. 81-8. The six prelayed sutures are tied after inserting an 18-Fr urethral catheter.

FIG. 81-9. After insertion of a cystostomy tube into the reservoir, the pouch is closed completely.

Postoperative Education Meticulous postoperative surveillance and teaching of the patient are paramount for good long-term results. After catheter removal, any bacteriuria is treated until the urine is sterile. Patients are instructed to void every 2 hours, first while sitting, by relaxing the pelvic floor and, if necessary, by abdominal straining. 2 Patients are encouraged to drink 2 to 3 L of fluid per day and to take additional dietary salt. The reservoir will secrete sodium and chloride if the urine is hypoosmolar. The additional salt intake prevents a salt-losing syndrome that might result in hypovolemia and acidosis. 5 Therefore, body weight and blood gases are controlled regularly. Patients without metabolic acidosis (negative base excess of more than 4 mmol/L), or with metabolic acidosis compensated by oral intake of sodium bicarbonate, are then instructed to retain the urine for 3 and later for 4 hours (even if they become incontinent earlier) until the maximal voiding volume is increased to 500 ml. Patients must be told that when the reservoir is full, increased reservoir pressure may cause urinary incontinence, but that it is essential to maintain the elevated pressure in order to expand the reservoir to the desired capacity. If a patient voids as soon as he or she becomes incontinent the reservoir will hardly ever expand; nighttime

incontinence in particular will be inevitable due to lack of capacity and the unfavorable pressure characteristics of a low-capacity reservoir. Residual urine is repeatedly ruled out and any bacteriuria treated.

OUTCOMES
Complications Early complications are dehydration due to salt loss syndrome and/or metabolic acidosis, as well as those of extended abdominal surgery. complications of the bladder substitute such as intestinal obstruction, pouch necrosis, ureteral strictures, or necrosis are rare. Results Capacity The median functional capacity of the bladder substitute increases from 120 ml immediately postoperatively to 350 ml after 6 months and 500 ml after 12 months postoperatively by extending the micturition interval from 2 to 4 hours. 2,5,6 With the advice to maintain regular voiding intervals of 4 to 6 hours and to avoid micturition volumes exceeding 500 ml, overdistention of the reservoir will be prevented and its capacity will remain stable for years. Spontaneous voiding with abdominal straining is possible in 98% of patients. The necessity of intermittent self-catheterization is rare (1%). 5,6 Continence Urinary continence is poor just after removal of the catheter 10 to 14 days postoperatively but improves within 3 months with the increase of functional bladder capacity. Younger patients and those in whom a nerve-sparing cystectomy was performed achieve continence faster than older patients and patients with previous radiation therapy of the pelvis. 11 Daytime continence is achieved in 92% of the patients 1 year postoperatively, nighttime continence in 80% of the patients 2 years postoperatively. 1,2,5,6 Urinary Infection Positive urinary cultures 6 months postoperatively are found in 12% of patients, usually in combination with residual urine. The risk of pyelonephritis within a year is less than 2% for patients with this bladder substitute. 6 Ureteral Obstruction Ureteral strictures, usually on the left side distal ureter, occur in 2% of patients. 1,6. This low stricture rate is the result of a simple end-to-side anastomosis with short ureters and preservation of the ureter circulation, and absence of any additional antireflux mechanism. 7,8 An antireflux nipple valve in this type of bladder substitute is not necessary for the following reasons: 1. The bladder substitute is a low-pressure reservoir. 2. During voiding, abdominal straining simultaneously produces the same pressure increase in the bladder substitute, the abdomen, and the renal pelvis. Without pressure differences retrograde urinary flow is impossible. 7,8 3. The peristalsis of the afferent tubular ileal segment and the peristalsis of the ureter provide a dynamic antireflux mechanism, as has been demonstrated in a video showing simultaneous pressure recordings in the bladder substitute and in the renal pelvis. 7,8 4. The long-term results of a randomized prospective study including 70 patients with either an antireflux mechanism or an afferent isoperistaltic ileal segment support the hypothesis that an antireflux nipple is not necessary. There were no major differences between these two groups except for an increased rate of upper urinary tract obstruction in the patients with antireflux nipples. 7 Metabolic Disturbances The incidence of postoperative metabolic acidosis is related to the length of ileum used to construct the reservoir, excluding the length of ileum used for the afferent tubular segment. Metabolic acidosis occurs predominantly within the first 3 months postoperatively. If the problem is recognized, resolution is made easy by draining the bladder and administering saline infusions and sodium bicarbonate if necessary. Permanent sodium bicarbonate substitution is only necessary in acidotic patients with reservoirs constructed from ileum segments 50 to 60 cm long. 5,6 A major risk of long-term acidosis would be impaired bone metabolism with osteopenia and/or osteomalacia. However, those patients with a 5- to 8-year follow-up still have bone densities in the normal range. 10 In summary, this orthotopic bladder substitute constructed from 40- to 44-cm ileum with a 14- to 16-cm-long afferent tubular ileal segment is a low-pressure urinary reservoir with excellent urodynamic properties providing good daytime and nighttime continence. It has a low stricture rate of the ureteroileal anastomosis and an adequate dynamic antireflux mechanism, preventing kidney damage. Furthermore, this urinary diversion is well accepted by the patients and promises good results, provided that the patients are carefully selected, well instructed, and meticulously followed. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Benson MC, Seaman EK, Olsson CA. The ileal ureter neobladder is associated with a high success and low complication rate. J Urol 1996;155:1585. Casanova GA, Springer JP, Gerber E, Studer UE. Urodynamic and clinical aspects of ileal low pressure bladder substitutes. Br J Urol 1993;72:728. Skinner DG. Technique of radical cystectomy. Urol Clin North Am 1981;8:353. [Deleted in proofs.] Studer UE, Danuser H, Hochreiter W, Springer JP, Turner WH, Zingg EJ. Summary of ten years experience with a ileal low-pressure bladder substitute combined with an afferent tubular isoperistaltic segment. World J Urol 1996;14:29. Studer UE, Danuser H, Merz VW, Springer JP, Zingg EJ. Experience in 100 patients with an ileal low pressure bladder substitute combined with an afferent tubular isoperistaltic segment. J Urol 1995;154:49. Studer UE, Danuser H, Thalmann GN, Springer JP, Turner WH. Antireflux nipples or afferent tubular segments in 70 patients with ileal low pressure bladder substitutes: long term results of a prospective randomized trial. J Urol 1996;156:1913. Studer UE, Spiegel T, Casanova GA, et al. Ileal bladder substitute: antireflux nipple or afferent tubular segment? Eur Urol 1991;20:315. Studer UE, Thalmann G, Springer JP, Zingg EJ. Complications and functional results in 100 patients with an ileal bladder substitute. J Urol 1993;149:326. Tschopp AB, Lippuner K, Jaeger P, Merz VW, Danuser H, Studer UE. No evidence of osteopenia 5 to 8 years after ileal orthotopic bladder substitution. J Urol 1996;155:71. Turner WH, Danuser HJ, Moehrle K, Studer UE. The effect of nerve sparing cystectomy technique on postoperative continence after orthotopic bladder substitution. J Urol 1997;158:2118.
5,6,9

Specific late

Chapter 82 Orthotopic Urinary Diversion Using a Colonic Segment Glenn’s Urologic Surgery

Chapter 82 Orthotopic Urinary Diversion Using a Colonic Segment
Daniela Schultz-Lampel and Joachim W. Thüroff

D. Schultz-Lampel: Adult Department of Urology and Pediatric Urology, Klinikum Wuppertal GmbH, University of Witten/Herdecke Medical School, Wuppertal, Germany. J. W. Thüroff: Department of Urology, Johannes Gutenberg University Medical School, 55131 Mainz, Germany.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique The Mainz Pouch Le Bag Right Colon for Orthotopic Bladder Substitution Left Colon for Orthotopic Bladder Substitution Outcomes Complications Results Chapter References

Orthotopic urinary diversion refers to total replacement of the bladder with bowel and preservation of transurethral voiding, with an attempt to restore the storage capacity and volitional control of emptying. 13 For patients with benign diseases of the bladder and for selected cases of bladder cancer, orthotopic bladder substitution can achieve normal functional control of urinary storage and emptying and can contribute to a tremendous improvement of quality of life in patients who must undergo total cystectomy. For successful restoration of function, surgical reconstruction of a urinary bladder from bowel must achieve the following: 1. Creation of a reservoir with adequate capacity and storage pressures less than 40 cm H 2O 2. Ureteral reimplantation avoiding reflux and obstruction 3. Reliable control of continence and emptying of the reservoir A variety of surgical techniques have been developed whereby almost all segments of the gastrointestinal tract are used in different configurations to reconstruct a neobladder: stomach, ileum, ileocecum, ascending colon, sigmoid colon, and rectum. 9 Each surgical technique addresses certain advantages but the variety of available techniques obviates the ability to gain widespread and broad experience by a standardized approach. We prefer the large bowel for orthotopic bladder substitution because it has several advantages over small bowel: 1. 2. 3. 4. A larger diameter requires a bowel segment of limited length. The length of the ileocecal artery allows transfer of the reservoir into the pelvis. A standard technique of antirefluxive ureteral implantation using a submucosal tunnel can be applied. Metabolic complications are fewer than after isolation of long segments of small bowel.

DIAGNOSIS
Urinary diversion is generally done in situations in which there is an actual or functional loss of bladder. The diagnostic modalities utilized are more specific to the cause of the operation (e.g., cystectomy for cancer) than the type of urinary diversion utilized.

INDICATIONS FOR SURGERY
Patient selection for orthotopic bladder replacement is critical and remains controversial. 10 Generally, an intact urethra and urethral sphincter mechanism as well as the feasibility of transurethral self-catheterization are regarded as prerequisites for orthotopic urinary diversion. Impaired renal function constitutes a contraindication for orthotopic bladder substitution because those patients may suffer from severe water and electrolyte imbalance. According to the literature, the main indication for orthotopic bladder substitution is bladder cancer in male patients requiring radical cystoprostatectomy. However, in our practice we apply a restricted indication for orthotopic total bladder substitution in malignant disease. Since patients with bladder cancer demonstrate an involvement of the urethra in 11% to 48% of all cases and the urethral recurrence rates are 6% to 15% in patients in whom the urethra is preserved as part of the cystectomy, we developed the following philosophy with regard to orthotopic bladder substitution: We regard transitional cell carcinoma in the urethra, prostate, or bladder neck as well as multifocal tumor occurrence or carcinoma in situ as contraindications for orthotopic bladder replacement. In females, the feasibility of an orthotopic bladder replacement has to consider not only the 40% incidence of primary tumor invasion of the urethra but also possible difficulties in transurethral self-catheterization. Despite recent promising results of orthotopic bladder replacement in females, urethrectomy with cutaneous urinary diversion must be considered the standard approach for females requiring cystectomy for bladder cancer. 9,10 On the other hand, orthotopic urinary diversion is an alternative to cutaneous urinary diversion in a variety of benign diseases, such as bladder contracture due to interstitial cystitis, tuberculosis, bilharziasis, or irradiation damage. In addition, orthotopic bladder substitution to the bladder neck is a therapeutic option for a functionally reduced bladder capacity in severe cases of detrusor hyperreflexia, bladder hypersensitivity, or low-compliance bladder, if these conditions do not respond to conservative therapy and if subtrigonal cystectomy with ureteral reimplantation is required. In patients with neurogenic bladder dysfunction, bladder substitution to the bladder neck can be an alternative to continent cutaneous diversion. This would be the case when intermittent self-catheterization through the urethra is feasible, standard bladder augmentation is not suitable, and ureteral reimplantation is required. However, pharmacotherapy and less invasive surgical techniques, such as neuromodulation and sacral deafferentation with implantation of an anterior sacral root nerve stimulator, should be considered first. 9 Orthotopic bladder substitution may also be applied for undiversion in selected patients who had previously undergone supravesical diversion.

ALTERNATIVE THERAPY
As ileum, ileocecum, cecum, and different segments of the colon have been used for orthotopic bladder replacement with similar results, any of these surgical techniques may be used. Usually, the individual preference of the surgeon will determine the choice of the bowel segment. Most frequently, orthotopic bladder replacement is performed by using the ileum. 8 Such ileal reservoirs include the Camey II procedure for total bladder replacement, the Hautmann ileal neobladder, the Melchior ileal neobladder, the orthotopic Kock pouch, the Studer cross-folded ileal reservoir, the ileal S pouch, and other procedures. The variety of techniques indicates that there is no standardization regarding how to perform an ileal neobladder. The main differences between these techniques are the length of the ileum used and the antireflux technique of ureter implantation. The main problem after ileal neobladder over all techniques is a nighttime incontinence rate of at least 30%.8,10 In all patients with contraindications for orthotopic bladder replacement and especially in females, continent cutaneous urinary diversion should be considered as an alternative to orthotopic bladder substitution. 9,10 Incontinent urinary diversion may be reserved for patients with severely impaired renal function and patients with

impaired general health conditions who would not tolerate continent urinary diversion. In patients with benign diseases, bladder augmentation is an alternative for those in whom sparing of the bladder is possible.
9

SURGICAL TECHNIQUE
The Mainz Pouch The ileocecal region and appendix was first utilized experimentally for supravesical urinary diversion by Verhoogen in 1908. Since that time different authors have successfully performed continent cutaneous urinary diversion and bladder substitution using the ileocecal region. 9 Since 1983, the Mainz pouch technique has combined detubularized cecum, ascending colon, and two ileal loops to create a spherical reservoir. 11,12 Light and Engelmann7 used a slightly modified pouch (“Le Bag”) for complete bladder substitution after cystoprostatectomy. The Mainz pouch technique of constructing a spherical pouch from 10 to 15 cm detubularized cecum and ascending colon and two ileal loops of the same length has been used since 1983 without major changes. A high-capacity reservoir with low pressures, a standard technique of ureteral implantation, and a universal applicability for bladder augmentation, bladder substitution, and continent cutaneous diversion are main advantages of this procedure. Diagnostic procedures for evaluation of the urinary tract include urodynamic studies of the bladder and the urethral sphincter mechanism in those patients with functional loss of bladder capacity. Evaluation of the colon by radiologic (contrast enema) and/or endoscopic (colonoscopy) examinations is performed to exclude polyposis, diverticulosis, and other bowel pathology. The bowel is rinsed the day before surgery by antegrade irrigation with 8 to 10 L of saline via a nasogastric tube. Intraoperatively the nasogastric tube is substituted by a gastrostomy catheter, and a rectal drainage tube is inserted in addition. Serum electrolyte concentrations are checked immediately after irrigation and, if necessary, are substituted intravenously. Prophylactic antibiotics are started the day before surgery and continued for at least 6 days postoperatively, e.g., piperacillin/tazobactam (Tazobac) 3 × 450 mg. After a median laparotomy, the cecum, ascending colon, and right colic flexure are detached from the abdominal wall and mobilized. The mesentery is divided between the right colic and the ileocolic arteries ( Fig. 82-1A). 10- to 15-cm of cecum and ascending colon and two segments of terminal ileum of same length each are isolated. Bowel continuity is restored by spatulated end-to-end ileoascendostomy using a single row of running 4-0 polyglyconate (Maxon) sutures of the seromuscularis layer only. Recently, reconstruction of the ileocecal valve has been introduced as a choice for selected cases. 2 As the appendix is not needed as a continent outlet for continent cutaneous urinary diversion, appendectomy is performed.

FIG. 82-1. Operative technique of Mainz pouch bladder substitution. (A) Intestinal segment to be isolated for the Mainz pouch procedure. (B) Antimensenteric splitting of intestine. (C) Side-to-side anastomosis of terminal and next proximal ileal loop. Line at antimesenteric splitting of the cecum for completing the posterior wall (pouch plate). (D) Antirefluxive implantation of ureters through a submucosal tunnel of 3 to 5 cm in length. (E) Buttonhole incision and anastomosis of the cecal pole to the membranous urethra for complete bladder substitution after radical cystoprostatectomy. (F) Completed Mainz pouch bladder substitution after radical cystoprostatectomy. Two 6- to 8-French ureteral stents and a 10-Fr pigtail cystostomy catheter are brought out through the abdominal wall. A transurethral 20-Fr Foley catheter is used for drainage and stenting of the urethral anastomosis.

The isolated bowel segment is irrigated clean with neomycin dissolved in saline. To create the reservoir, the isolated bowel segments are opened antimesenterically (Fig. 82-1B) and anastomosed side-to-side, connecting colon with terminal ileum and the latter with the next proximal segment of ileum, by a single row of running sutures of all layers using a straight needle and 4-0 polyglyconate (Maxon) sutures to create the posterior pouch plate ( Fig. 82-1C). For ureteral implantation, the left ureter is tunneled to the right side retroperitoneally and both ureters are implanted at the open end of the large bowel via a 3- to 5-cm-long submucosal tunnel with 6-0 polyglycolic acid (Dexon) sutures. The ureters are stented with 8-Fr ureteral catheters ( Fig. 82-1D). If the ureters are grossly dilated, implantation of the ureters in a serous-lined extramural tunnel in the Abol-Enein technique is preferable ( Fig. 82-2).3

FIG. 82-2. Operative technique of ureteral implantation in Mainz pouch bladder substitution through a serous-lined extramural tunnel (Abol-Enein technique). (A) Two times 12 to 15 cm of cecum and ascending colon and one ileum segment of the same length are isolated and incised antimesenterically. The lateral limbs are joined together by seromuscular continuous suture to create two serous-lined intestinal troughs. The right ureter is pulled through an opening of the seromuscular trough; the left ureter is pulled through a window of the mesentery. (B) Ureters are inlayed within the corresponding trough and the tunnel is closed over the ureters. (C) Ureters are spatulated and anastomosed to the intestinal mucosa. (D) Completion of ureteral implantation. (Modified from Fisch M, Abol-Enein H, Hohenfellner R. Ureterimplantation mittels serösem extramuralem Tunnel in Mainz-Pouch I und Sigma-Rektum-Pouch (Mainz-Pouch II): Akt Urol 1995;26:I–X.)

For the urethral anastomosis, a buttonhole incision of approximately 0.5 cm in diameter is made at the lowest aspect of the cecal pole and the mucosa is everted by 4 to 5 catgut sutures to ensure mucosal adaption with the urethral stump. Closure of the anterior wall of the pouch is completed by running 4-0 polyglyconate (Maxon) sutures. At the site of ureteral implantation, the mucosa only is sutured to prevent ureteral obstruction at the entry into the pouch. The pouch is then anastomosed to the membranous urethra by six everting 5-0 polyglycolic (Dexon) sutures ( Fig. 82-1E, F). The reservoir is tested for leakage by filling the reservoir through the catheter. A 20 F transurethral Foley catheter and an additional 10-Fr pouchostomy tube are used for stenting the urethral anastomosis and drainage. Two 20-Fr silicone drains are placed into the pelvis and at the site of ureteral implantation. Beginning with the second postoperative day, the pouch is irrigated twice daily to prevent mucous retention. The ureteral stents are removed after 10 to 12 days and

an intravenous pyelogram (IVP) is performed. The transurethral catheter is removed after 4 weeks, when a pouchogram confirms absence of extravasation. The cystostomy catheter is removed when the patient is able to reliably empty the pouch spontaneously or by self-catheterization. Le Bag Another variant of a detubularized ileocolonic pouch is called “Le Bag,” which was first described by Light and Engelmann in 1986 7 for bladder augmentation and total substitution in four patients (three male and one female) after cystectomy due to neuropathic bladder and bladder cancer. This ileocolonic pouch also offers a low-pressure reservoir for orthotopic replacement of the bladder in selected patients, but it also has the disadvantage of a relatively high incontinence rate and is often associated with hyperchloremic metabolic acidosis. 5,9 After a midline abdominal incision, the cecum and ascending colon are mobilized up to the hepatic flexure and an appendectomy is performed. A segment of at least 20 cm of ascending colon and a corresponding segment of terminal ileum are isolated on a common vascular pedicle. Bowel continuity is restored between the ileum and transverse colon ( Fig. 82-3A).

FIG. 82-3. Operative technique of ileocolonic bladder substitution Le Bag. (A) Selection of ileocolonic segment with single vascular pedicle. (B) Incision along antimesenteric border to create two flat sheets of bowel. (C) Suturing of ileal and colonic segments. (D) Completion of pouch for anastomosis to the urethral stump. (E) Pouch rotated 180 degrees into the pelvis for anastomosis to the urethral stump, leaving an opening for later ureteral implantation. (F) Ureteral implantation into colonic segment of the pouch through hiatus in anterior wall. (From Light JK, Engelmann UH. Reconstruction of the lower urinary tract. Observations on bowel dynamics and the artificial urinary sphincter. J Urol 1985;135:215–224.)

For bladder substitution to the urethra, the ileocecal segment is opened along the antimesenteric border commencing at the terminal ileum 2 in. distal to the cut end and continuing through the ileocecal valve and into the anterior tenia of the ascending colon ( Fig. 82-3B). In this way, two flat sheets of small and large bowel are obtained that are in continuity at the incised ileocecal valve. The two outermost incisions of the large and small bowel segments are joined with horizontal mattress sutures of 1-0 chromic catgut (Fig. 82-3C). The ileocolonic pouch is rotated 180 degrees into the pelvis. Thus, the initial anterior wall, which has already been sutured, comes to lie posteriorly ( Fig. 82-3D). Similarly, the open posterior wall then lies anteriorly and is closed in a similar fashion from the inferior to the superior aspect, leaving an opening for later implantation of the ureters ( Fig. 82-3E). The short tubular segment of intact ileum allows end-to-end anastomosis to the urethra. The urethral remnant is mobilized and anastomosed to the ileal loop with 6 all-layer sutures of 1-0 chromic catgut ( Fig. 82-3E). Implantation of the ureters into the colon is performed once the ileocolonic pouch is situated and sutured in the pelvis from inside the pouch through the defect left previously at a convenient site. A submucosal tunnel is created from inside the pouch for a length of approximately 1 in., the ureters are pulled through, and the ureteral end is spatulated and sutured to the colonic mucosa with 4-0 catgut sutures ( Fig. 82-3F). Stenting of the ureters is not performed routinely but is recommended if there is any concern regarding tension at the ureteral implantation. The remaining portion of the anterior wall of the pouch is closed with a running horizontal mattress suture of 1-0 chromic catgut. A large catheter is placed through the urethra into the completed pouch, and the pouch is filled with water to estimate capacity and examine for leaks. The indwelling catheter is maintained for 3 weeks and a cystogram is performed before removal to ensure that there is no urinary leakage. Right Colon for Orthotopic Bladder Substitution The use of the detubularized right colon for bladder replacement was first described by Goldwasser et al. in 1980. According to the authors' opinion, the main advantage of the use of the right colon is the long vascular pedicle that allows easy mobilization of the bowel segment to the urethra. 4 The construction of the bladder substitution is commenced with division of the posterior peritoneum lateral to the right colon, around the cecum and along the root of the small bowel mesentery up to the ligament of Treitz. After incision of the hepatocolic ligament, the entire right colon including cecum, ascending and right transverse colon can be mobilized on their mesentery. At that stage of the operation, the surgeon must decide whether the lowermost part of the cecum can reach the pelvic floor. If this is not possible, the procedure should be changed to a technique using an alternative bowel segment. After identification of the ileocolic, the right colic, and middle colic arteries by palpation and transillumination, the right colon is isolated by incision of the ileal and colonic mesenteries ( Fig. 82-4A). Intestinal continuity is reestablished by an ileocolostomy.

FIG. 82-4. Operative technique of right colonic bladder substitution. (A) Isolation of right colon with vascular pedicle: ICA = ileocolic, RCA = right colic arteries, MCA = middle colic artery, SMA = superior colic arteries. (B) Splitting of bowel segment through anterior tenia. Resection of ileal stump and closure of ileocecal valve. Resection of appendix if present. (C) Approximately 5 cm of proximal cecum remains tubular-shaped. Antirefluxive implantation of the ureters through submucosal tunnels in the proximal half of the segment. Buttonhole incision is performed at the cecal pole. (D) Urethrocecal anastomosis over a urethral catheter. (E) Suprapubic catheter and ureteral stents are passed through anterior wall of the neobladder. (F) Distal opened bowel segment is folded over to become the anterior wall of the neobladder. (From Goldwasser B, Mor Y. Bladder replacement using the detubularized right colon. In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;166–174.)

The isolated bowel segment is then irrigated clean with saline. Taking care of the ileocolic artery, the ileal stump is resected completely and the opening of the ileocecal valve is closed with two layers of 3-0 polyglactin (Vicryl) sutures. If present, the appendix is removed ( Fig. 82-4B). The right colonic segment is then opened through the anterior tenia, leaving the distal 3 to 5 cm of the cecum tubularized.

The ureters are implanted through submucosal tunnels of 3 to 4 cm length according to Goodwin's antireflux technique into the proximal half of the bowel segment (Fig. 82-4C). The ureters are stented with 5- to 8-Fr feeding tubes. A 4-0 chromic catgut suture between the ureteral serosa and the bowel wall at the site of entry into the neobladder preserves the intended length of the submucosal tunnel. A buttonhole is incised at the lowest pole of the cecum, which later allows anastomosis to the urethra (Fig. 82-4D). The mucosa of the bowel is everted by a few interrupted 4-0 chromic catgut sutures. The distal opened bowel segment is folded over to become the anterior wall of the neobladder ( Fig. 82-4E). If this is difficult, the right colonic artery may be sacrificed. The pouch is closed over the two ureteral stents and a 20-Fr Foley cystostomy catheter, which are brought through separate openings in the anterior wall of the neobladder, with two layers of running 3-0 Vicryl sutures leaving an opening of 3-4 cm in length at the right upper corner of the cystoplasty. For anastomosis to the urethra, six 2-0 chromic catgut sutures are placed. These sutures are then tied over a transurethral 20-Fr silastic Foley catheter. The remaining opening of the neobladder is closed ( Fig. 82-4F). Finally, the bowel used for bladder replacement is retroperitonealized and omentum is wrapped over the suture lines. Two suction drains are placed behind and anterior to the neobladder. The urethral catheter and the suprapubic tube are flushed with 20 ml of sterile saline every 4 hours for the first week. Thereafter, irrigation is performed twice daily. If the cystogram performed after 10 days does not show any extravasation, the ureteral stents are removed. The urethral catheter remains in place for 21 days and the cystostomy is removed when the patient can void spontaneously without significant residual urine. Left Colon for Orthotopic Bladder Substitution The use of the left colon (detubularized sigmoid colon) for total bladder substitution was described by Reddy and co-workers in 1991 and an update of the surgical technique was published in 1995. 1 Candidates for bladder substitution using the left colon have to undergo a barium enema or colonoscopy. Whereas detection of an occult malignancy is an absolute contraindication for using the left colon for bladder replacement, diverticula of the sigmoid colon are not. However, chronic diverticulitis usually leaves the left colon unsuitable for reconstructive applications. The left colon is mobilized up to the splenic flexure to allow the bowel segment to reach the pelvic floor. A 30- to 35-cm segment of sigmoid and descending colon is then isolated on a broad mesenteric pedicle of the sigmoid arterial branch of the inferior mesenteric artery ( Fig. 82-5). Colonic continuity is reestablished by end-to-end anastomosis of the colon with a single layer of interrupted 3-0 Vicryl sutures or recently by using the Valtract BAR biofragmented anastomosis ring (Davis and Geck, Medical Device Division, American Cynamid Corp, Wayne, NJ). 1 The isolated colonic segment is irrigated with antibiotic solution and positioned in a U shape in the pelvis. The lowermost part of the U is marked for later anastomosis to the urethral stump. The segment is completely detubularized by incision of the medial tenia close to the mesenteric border ( Fig. 82-6A). The posterior layer of the colon segment is aligned by a few interrupted 3-0 Vicryl sutures and then closed by a single continuous suture of 3-0 Vicryl ( Fig. 82-6B). After ureteral implantation, the anterior layer is closed in a similar way. A 24-Fr Malecot catheter is placed in the reservoir and the cephalad opening of the reservoir is closed with a single layer of 3-0 Vicryl. For further reduction of reconstruction time, recently the Auto Suture Poly GIA 75-060 (United States Surgical Corp., Norwalk, CT) disposable surgical stapler is used for detubularization and simultaneous placement of the posterior and anterior suture lines. 1

FIG. 82-5. Operative technique of left colonic bladder substitution. Isolation of left colon on a broad vascular pedicle based on the sigmoid arteries. (From Goldwasser B, Mor T: Bladder replacement using the dehulularized right colon. In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;175–182.)

FIG. 82-6. Operative technique of left colonic bladder substitution. (A) Colon segment folded in a U shape. Most dependent part is marked for later anastomosis to the urethral stump. (B) Ureteral implantation anteriorly into the lateral tenia. (C) Splitting colonic segment along the medial tenia. Ureteral implantation posteriorly through submucosal tunnels. (D) Approximation of posterior walls of the sigmoid. (E) Completion of anterior wall of the neobladder. (F) Anastomosis of the neobladder to the urethral stump over a Foley catheter. (From Bernard PH, Iseri A, Reddy PK. Sigmoid bladder. In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;175–182.)

Implantation of the ureters into the reservoir can be performed either from an external anterior approach using a seromuscular technique or from inside by a submucosal technique and a posterior approach. In the first technique, the ureters are implanted before detubularization of the colonic segment. The ureters are implanted along the lateral tenia on either side by the Leadbetter technique using 5-0 Vicryl sutures ( Fig. 82-6B). The ureters are stented with 8-Fr infant feeding tubes. In the alternative technique, the colonic segment is already detubularized and the posterior pouch plate is formed. The ureters are implanted through submucosal tunnels in the posterior colonic wall using 5-0 Vicryl sutures and stented with 8-Fr feeding tubes ( Fig. 82-6C, D). For anastomosis to the urethra, a small opening is made into the lowest aspect of the colonic reservoir and the neobladder is sutured to the urethral stump over a 20-Fr Foley catheter using 6 interrupted 2-0 Vicryl sutures ( Fig. 82-6E, F). The neobladder is irrigated with saline every 8 hours to prevent mucus obstruction. Ureteral stents are removed on days 13 and 14. If a cystogram on day 14 does not reveal extravasation, the transurethral catheter is removed. Urinary drainage via the Malecot catheter is continued until day 21 and the catheter is removed when successful voiding has been achieved without major amounts of residual urine.

OUTCOMES
Complications

Mainz Pouch Among all 561 patients treated with the Mainz pouch I procedure, early and late complications were encountered in 12% and 37%, respectively. 6 Most of the complications (9%) were general surgical ones, such as thrombosis, pulmonary embolism, or pneumonia, whereas pouch-related early complications only occurred in 17 of 561 patients (3%). Complications that were specifically related to the technique of orthotopic bladder substitution occurred in 3 of 61 patients (5%), all of whom had leakage at the urethral anastomosis ( Table 82-1). Of the late complications, 50% were related to the continence mechanism in continent cutaneous diversion. 6 Eleven of 61 patients with orthotopic bladder substitution (18%) developed stenosis at the urethral anastomosis, which could easily be managed by endoscopic incision or resection of the anastomosis ( Table 82-2). Surprisingly, in one patient resection of the stenosis at the urethral anastomosis revealed tumor recurrence at the site of the anastomosis. In this patient, urethrectomy and conversion into a continent cutaneous urinary diversion was performed.

TABLE 82-1. Early complications of orthotopic bladder substitution using a colonic segment

TABLE 82-2. Late complications of orthotopic bladder substitution using a colonic segment

Le Bag In the early series, renal function and electrolyte balance remained normal, 7 but later series revealed hyperchloremic metabolic acidosis in most of the patients, which was related to the pouch length (Table 82-1 and Table 82-2).5 Right Colon Early complications occurred in 10 of 41 patients (24%). Four patients developed cardiovascular complications, which were fatal in two patients. Three patients had prolonged intestinal obstruction which was successfully treated by conservative means. Three patients had extravasation of urine from the neobladder after 12 to 14 days, which stopped spontaneously after prolonged transurethral drainage ( Table 82-1). Late complications occurred in only four patients (10%) who developed ureteral obstruction. In one patient ureteral obstruction occurred at the site where the left ureter was passed through the sigmoid mesentery and nephrectomy was performed because of poor kidney function. In the other case, fibrous tissue caused right ureteral obstruction that necessitated ureterolysis. Two patients developed urethral strictures that were successfully treated by dilation ( Table 82-2). Left Colon Early complications occurred in 2 of 27 patients (7%) ( Table 82-2). These were a pulmonary embolism in one patient and urinary extravasation in another. Late complications in 8 of 27 patients (30%) comprised ureteroenteric obstruction requiring ureteral reimplantation (1), progressive renal failure (1), grade I reflux (3), and asymptomatic urinary tract infection (3) (Table 82-2). Results The Mainz Pouch From 1983 to July 1994, a Mainz pouch procedure has been performed in 561 patients in Mainz and Wuppertal. 6 Sixty patients underwent Mainz-pouch procedure for bladder augmentation or bladder substitution to the bladder neck, 61 for bladder substitution to the urethra, and 440 for continent cutaneous urinary diversion. The mean follow-up for all patients is 57 months (range: 3 to 127 months); in the group with substitution to the urethra follow-up is 46.3 months (range: 3 to 90 months). Fifty-eight of 61 male patients (95%) with orthotopic bladder substitution to the urethra after radical cystoprostatectomy are continent during daytime. Eight of 58 patients (14%) have leakage during the night if they do not empty the bladder at regular 4-hour intervals. Another 8 of 61 patients (13%) are not able to void spontaneously and have to perform clean intermittent catheterization ( Table 82-3).

TABLE 82-3. Results of orthotopic bladder substitution using a colonic segment

Postoperative urodynamic studies in the patients with orthotopic bladder substitution revealed approximately the same pressure/volume characteristics as measured under the same conditions in the normal urinary bladder. Mean capacity of the bladder substitutes was 500 ml (range: 250 to 1000 ml). The residual volume ranged from 0 to 80 ml (mean: 30 ml). Total reservoir pressures in sitting position ranged from 28 cm H 2O (50% filling) to 45 cm H 2O (100% filling). Peristaltic contractions were recorded in the early postoperative phase in most of patients with filling volumes ranging from 130 to 570 ml (mean: 317 ml) and a mean maximum amplitude of 10 cm H2O. After follow-up periods exceeding 6 months, contraction waves subsided in the majority of cases or shifted to the right on the cystometric curve to non-physiologic high filling volumes. Le Bag In an early series, 6 weeks postoperatively three patients were continent, two of whom were assisted by the placement of an artificial sphincter. Water cystometry revealed high compliance and adequate bladder capacity in all four patients. Two patients showed low-amplitude contractions at maximum capacity with total pressures staying below 45 cm H2O.7 In a later series of 11 patients, 5 of 9 men achieved continence, whereas the remaining required implantation of an artificial sphincter (Table 82-3).9 In a recent series of 38 cases with Le Bag orthotopic bladder substitution to the urethra, daytime continence was 91%, whereas only 46% achieved nighttime continence. 5 Right Colon Between 1985 and 1993, 41 men underwent orthotopic urinary diversion using the right colonic segment after radical cystoprostatectomy for invasive transitional cell carcinoma of the bladder. 4 During the follow-up, 38 of 41 patients (93%) were completely continent during the day. However, 2 of 41 patients (5%) suffered from severe daytime and nighttime incontinence that was successfully treated by implantation of an artificial urinary sphincter, and 1 patient had mild diurnal stress incontinence. Continence at night was not so favorable in that 13 of 41 patients (32%) had nocturnal incontinence using protective pads at night and the other two-thirds achieved nighttime continence if they emptied their bladders every 2 to 3 hours ( Table 82-3). Large amounts of residual urine or voiding difficulties were not a problem in any of the patients. Urodynamic studies were performed in the first nine patients at least 3 months after surgery. The capacity ranged from 500 to 1600 ml and residual urine ranged from 0 to 70 ml. Sensation of bladder fullness was present in about half of the patients. At 500 ml capacity, the mean basal pressure was 11 cm H 2O (range: 6 to 20 cm H2O) and the maximum reservoir pressure was 22 cm H2O (range: 12 to 49 cm H2O). Peristaltic contractions occurred in about 70% of patients at a filling volume between 100 and 900 ml (mean 270 ml) with a maximum amplitude of less than 40 cm H2O. The maximum urethral closure pressure ranged between 35 and 60 cm H 2O (mean 48 cm H2O). Left Colon All 27 patients with a sigmoid orthotopic bladder substitution were completely continent during the day and 18 of 27 (67%) also achieved nighttime continence after a follow-up period of 26.5 months (range: 13 to 38 months). 1 The remaining nine patients (33%) have to get up more than twice each night to maintain continence or have urinary leakage at night requiring pads or condom catheter ( Table 82-3). Capacity after 3 months was 450 ml (range: 230 to 720 ml) and increased to 600 ml (380 to 780 ml) after 12 months, allowing mean voiding intervals of 4.5 hours (range: 3 to 6 hours). After 12 months, reservoir pressure during filling was 12 cm H 2O (range: 10 to 16 cm H2O) and at capacity 16 cm H 2O (range: 10 to 22 cm H2O). Residual urine was 40 ml (range: 4 to 80 ml) and no patient required intermittent catheterization. Recurrence of Tumor The incidence of recurrence of the cancer can be seen from our experience with the right colon orthotopic bladders, in which we performed the procedure in 41 patients with cancer. Eleven of 41 patients (27%) died during the follow-up period of 30 months (range: 6 to 42 months). Two patients died in the early postoperative period and 1 patient died of myocardial infarction 6 months after surgery. An additional 8 patients died from disseminated disease or local recurrence. However, none of these patients had recurrence in the urethra. Orthotopic bladder substitution is a highly appreciated therapeutic modality for reconstruction of the urinary tract after total removal of the bladder when certain prerequisites are fulfilled. The colonic segments fulfill all requirements of an intestinal urinary reservoir for orthotopic bladder substitution. Complications relate to general risks of major surgery or to the specific surgical techniques and are comparable to techniques using ileum for bladder substitution. As in some cases the final choice of the bowel segment to be used for construction of a neobladder can only be made at the time of surgery, it is useful to be able to use both small and large bowel segments in different techniques. CHAPTER REFERENCES
1. Bernard PH, Iseri A, Reddy PK. Sigmoid bladder. In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;175–182. 2. Fisch MM, Wammack RE, Hohenfellner R. The Mainz pouch procedure (mixed augmentation, ileum and cecum). In: Webster G, Kirby R, King L, Goldwasser B, eds. Reconstructive urology. Vol. 1. London: Blackwell Scientific, 1993;459–475. 3. Fisch M, Abol-Enein H, Hohenfellner R. Ureterimplantation mittels serösem extramuralem Tunnel in Mainz-Pouch I und Sigma-Rektum-Pouch (Mainz-Pouch II): Akt Urol 1995;26:I–X. 4. Goldwasser B, Mor Y. Bladder replacement using the detubularized right colon. In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;166–174. 5. Kolettis PN, Klein EA, Novick AC, Winters JC, Appell RA. The Le Bag orthotopic urinary diversion. J Urol 1996;156:926–930. 6. Lampel A, Fisch M, Stein R, et al. Continent diversion with the Mainz pouch. World J Urol 1996;14:85–91. 7. Light JK, Engelmann UH. Reconstruction of the lower urinary tract. Observations on bowel dynamics and the artificial urinary sphincter. J Urol 1985;135:215–224. 8. Mark SD, Webster GD. Ileal reservoirs (Camey II, Hautmann, Kock, Studer). In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;140–147. 9. Schultz-Lampel D, Lampel A, Thüroff JW. Ileocolonic reservoir (Mainz, Le Bag). In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;148–165. 10. Seemann O, Jünemann KP, Alken P. Patient selection criteria for orthotopic bladder replacement. In: Webster GD, Goldwasser B, eds. Urinary diversion: scientific foundation and clinical practice. Oxford: Isis Medical Media, 1995;128–139. 11. Thüroff JW, Alken P, Engelmann U, Riedmiller H, Jacobi GH, Hohenfellner R. The Mainz pouch (mixed augmentation ileum and coecum) for bladder augmentation and continent urinary diversion. Eur Urol 1985;11:152–160. 12. Thüroff JW, Alken P, Riedmiller H, Jacobi GH, Hohenfellner R. 100 cases of Mainz pouch: continuing experience and evolution. J Urol 1988;140:283–288. 13. Thüroff JW, Mattiasson A, Andersen JT, et al. Standardization of terminology and assessment of functional characteristics of intestinal reservoirs. Neurourol Urodynam 1996;15:499–511.

Chapter 83 Orthotopic Bladder Replacement in Women Glenn’s Urologic Surgery

Chapter 83 Orthotopic Bladder Replacement in Women
John P. Stein and Donald G. Skinner

J. P. Stein and D. G. Skinner: Department of Urology, University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, California 90033.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Evaluation Radical Cystectomy Construction of the Kock Ileal Neobladder Urethral Anastomosis Outcome Complications Results Chapter References

Since the early 1900s innovative surgeons have persistently pursued how best to replace the original bladder removed for either benign or malignant disease. Currently, the ultimate goals of lower urinary tract reconstruction have become more than simply a means to divert urine and to protect the upper urinary tracts. Contemporary objectives of lower urinary tract reconstruction should include eliminating the need for a cutaneous stoma, urostomy appliance, or the need for intermittent catheterization, while maintaining a more natural voiding pattern that allows volitional micturition through the intact native urethra. These advances in urinary diversion have been made in an effort to provide patients a more normal lifestyle, with an improved self-image following removal of the bladder. Over the past 45 years the evolution of urinary diversion has developed along three distinct paths: a noncontinent cutaneous form of urinary diversion; a continent cutaneous form of urinary diversion; and, most recently, an orthotopic form of diversion to the native urethra. In 1950, Bricker introduced the ileal conduit, which established a reliable form of urinary diversion. This form of urinary diversion remained, even until the early 1990s, the gold standard to which all other forms of urinary diversion were compared. Concurrent with Bricker's introduction of the ileal conduit, Gilchrist independently reported on the concept of a continent cutaneous form of diversion, utilizing the ileocecal valve as the continence mechanism and the distal ileum as a catheterizable stoma. However, Gilchrist's ileocecal reservoir garnered little support, whereas the Bricker ileal conduit became the urinary diversion of choice for the next several decades. The concept of a continent cutaneous diversion was eventually popularized in the 1980s at several large institutions. 9 This revolutionized lower urinary tract reconstruction to a continent cutaneous form. Patients were relieved from the problems of an external collection device but still required catheterization of an abdominal stoma. In 1979, Camey and Le Duc reported their pioneering work with orthotopic neobladders to the native urethra. This landmark accomplishment demonstrated the feasibility of lower urinary tract reconstruction to the urethra, with reasonable continence rates, in carefully selected male patients following cystectomy. From 1982 to 1986, the continent cutaneous Kock ileal reservoir was the procedure of choice in all patients requiring urinary diversion at our institution. Beginning in 1986 we began to perform orthotopic reconstruction to the urethra in carefully selected male patients undergoing cystectomy with excellent functional results. 10 However, prior to 1990, orthotopic reconstruction was limited to male patients and considered to be a contraindication in the female subject undergoing cystectomy. It was previously thought necessary to remove the urethra during cystectomy in order to provide an adequate surgical cancer margin in women. In addition, it was believed that the female patient would be unable to maintain a continence mechanism if orthotopic diversion was performed following cystectomy. However, based on an extensive pathologic review of female cystectomy specimens removed for transitional cell carcinoma of the bladder, it was shown that the urethra could be safely preserved in the majority of women undergoing cystectomy.13 This study provided sound pathologic criteria that helped to safely identify appropriate female candidates for orthotopic diversion following cystectomy for bladder cancer. In addition, elegant anatomic dissection of the female pelvis provided a better understanding of the continence mechanism in women, suggesting that continence could be maintained in women following cystectomy. 2 These important discoveries provided a foundation on which to offer women lower urinary tract reconstruction to the urethra. Our initial clinical experience in women demonstrated outstanding functional results. 14 This achievement marked another significant step forward in the evolution of lower urinary tract reconstruction. It is our firm opinion that the orthotopic neobladder currently represents the most ideal form of urinary diversion available today and should be considered the gold standard to which other forms of diversion are compared. In fact, in 1993 at the Fourth International Consensus Conference on Bladder Cancer in Antwerp, Belgium, consensus opinion was that in the properly selected bladder cancer patient, lower urinary tract reconstruction to the urethra is the procedure of choice in most centers worldwide. We believe that this form of diversion can be safely performed in nearly 80% of women undergoing cystectomy for bladder cancer. This chapter will focus on the indications and the technique of the orthotopic diversion in women.

DIAGNOSIS
This reconstructive technique is performed following cystectomy, usually for malignant disease. As such, proper diagnostic studies for the underlying disease are important and are discussed in Chapter 24.

INDICATIONS FOR SURGERY
Indications for orthotopic urinary diversion in women are following actual or functional loss of the bladder. These modifications are further modified by patient selection criteria specific to orthotopic diversion, which include the following: (a) the external sphincter must remain functionally intact to provide continence and to allow for volitional voiding per urethra and (b) the cancer operation must under no circumstance be compromised by the orthotopic reconstruction at the ure-throenteric anastomosis, the retained urethra, or the surgical margins. If these two criteria can be safely maintained, the patient may then be considered for orthotopic diversion following cystectomy. In general, the majority of women undergoing cystectomy are potential candidates for orthotopic reconstruction. Contraindications to this form of diversion include (a) a noncompliant patient with a mental or physical handicap, (b) patients with impaired renal function whose serum creatinine is greater than 2.5 mg/dl, and (c) those patients with inflammatory bowel disease. However, patients with an elevated serum creatinine secondary to ureteral obstruction should undergo upper urinary tract decompression (via a percutaneous nephrostomy tube) allowing recovery, and determination of the true baseline renal function prior to urinary diversion. Although controversial, patient age alone should not be a contraindication to orthotopic diversion. A differentiation between physiologic and chronologic age should be made. In a recent review of 295 male patients undergoing orthotopic diversion, the overall percentage of patients with good and satisfactory continence was statistically similar in older (more than 70 years) and younger age groups ( p < 0.001).4 These findings, at least in male patients, underscore the notion that patient age alone should not preclude orthotopic urinary diversion. Furthermore, body habitus has not been an exclusive factor for orthotopic reconstruction. In fact, the obese patient is an ideal candidate for this form of diversion, as the need to negotiate a thick abdominal wall for intermittent catheterization is eliminated. Currently, most patients requiring lower urinary tract reconstruction undergo cystectomy for transitional cell carcinoma of the bladder. Despite progress in chemotherapeutic regimens, radical cystectomy remains the treatment of choice for high-grade, invasive bladder cancer. Furthermore, as the incidence of bladder cancer rises, the number of patients requiring cystectomy and subsequent urinary diversion for transitional cell carcinoma of the bladder can also be expected to increase. A critical issue in women undergoing cystectomy and orthotopic diversion for a pelvic malignancy is ensuring that the cancer operation is not compromised by the

reconstruction. Concerns for orthotopic diversion in the female patient arise from the fact that the pathologic implications of sparing the female urethra had not been well studied previously. In addition, urethrectomy was routinely performed in women without sound scientific data. In contrast to male patients with prostatic tumor involvement, risk factors that may predict for urethral tumor involvement in women have only recently been identified. In an extensive retrospective analysis of female cystectomy specimens we have helped define the incidence of carcinoma involving the bladder neck and urethra in women with transitional cell carcinoma of the bladder. 13 This pathologic study identified important risk factors for urethral tumor involvement in women that could help select appropriate candidates for orthotopic diversion following cystectomy for bladder cancer. A total of 67 consecutive female cystectomy specimens removed for biopsy proven transitional cell carcinoma of the bladder between 1982 and 1990 were pathologically reviewed. 13 Histologic evidence of tumor (carcinoma in situ or invasive carcinoma) involving the urethra was present in 9 cystectomy specimens (13%). In all cases, tumor was confined to the proximal or midurethra, and in no patient was the distal urethra involved with tumor. Most importantly, all patients with carcinoma involving the urethra had concomitant tumor involving the bladder neck. A total of 17 patients (25%) had tumor involvement of the bladder neck; all patients with an uninvolved bladder neck also had an uninvolved urethra. Tumor involving the bladder neck and urethra was more commonly associated with high-grade and high-stage tumors, as well as lymph node–positive disease. In addition to bladder neck involvement, anterior vaginal wall involvement with tumor was also identified as a major risk factor for urethral tumor involvement in these female cystectomy specimens. All patients with tumor extending into the anterior vaginal wall were also found to have bladder neck involvement, and 50% of these specimens also demonstrated urethral tumor involvement. However, if the bladder neck was histologically free of tumor, then no patient demonstrated any urethral or vaginal wall tumor. This pathologic study suggested that female patients without tumor involvement of the bladder neck and anterior vaginal wall may be considered appropriate candidates for orthotopic diversion. However, not all specimens with tumor involving the bladder neck demonstrated urethral tumor involvement. This is an important issue because although bladder neck involvement with tumor is a risk factor for urethral tumor involvement, approximately 50% of patients with tumor involving the bladder neck will have a urethra free of tumor. Accordingly, these female patients could be considered appropriate candidates for orthotopic diversion. Therefore, although bladder neck involvement with tumor is a risk factor for concomitant urethral tumor involvement, pathologic evaluation of the proximal urethra appears to be the most critical determinant for orthotopic diversion. These initial pathologic guidelines were subsequently evaluated prospectively in 29 consecutive women undergoing orthotopic diversion following cystectomy for transitional cell carcinoma of the bladder. 12 A total of 23 cystectomy specimens were without evidence of tumor involvement of the bladder neck and were also free of tumor at the urethra. Overall, 6 specimens demonstrated tumor involvement at the bladder neck, with only one of these demonstrating any urethral tumor involvement. These results appear to support our previously established pathologic criteria that identify bladder neck tumor involvement as a risk factor for urethral tumor involvement. In addition, in our clinical series of 29 women undergoing orthotopic diversion, we found intraoperative frozen section analysis of the distal surgical margin (proximal urethra) to be an accurate and reliable method to evaluate the proximal urethra for urethral tumor involvement. Intraoperative frozen section analysis accurately evaluated the proximal urethra in all 29 specimens removed for transitional cell carcinoma, including 28 cases without tumor involvement and 1 specimen with carcinoma in situ involving the proximal urethra. In all 29 cases the frozen section analysis was correctly confirmed on permanent section of the cystectomy specimen. Currently, we believe that intraoperative frozen section of the proximal urethra is the most decisive method to determine if a female patient is an appropriate candidate for orthotopic diversion. Furthermore, because of the potential risk of injuring the continence mechanism with preoperative biopsy of the bladder neck and urethra, coupled with a confirmed method to reliably evaluate the proximal urethra (intraoperatively), we now rely on intraoperative frozen section analysis of the proximal urethra for proper patient selection in women considering orthotopic lower urinary tract reconstruction.

ALTERNATIVE THERAPY
Alternatives to orthotopic diversion include any other form of urinary diversion including ileal conduit or other bowel conduit, ureterosigmoidostomy, varied forms of continent urinary diversion, and varied segments of bowel described for orthotopic diversion.

SURGICAL TECHNIQUE
Although the ideal bladder substitute remains to be developed, the orthotopic neobladder most closely resembles the original bladder in both location and function. The orthotopic neobladder eliminates the need for a cutaneous stoma or cutaneous collection device. This form of diversion relies on the intact rhabdosphincter continence mechanism, eliminating the need for the often-plagued efferent continence mechanism of most continent cutaneous reservoirs and need for intermittent catheterization. The majority of patients undergoing orthotopic reconstruction are continent and able to void to completion without the need for intermittent catheterization. 4,14 Certain principles of all orthotopic urinary reservoirs should include a large-capacity, low-pressure, nonrefluxing (to protect upper urinary tract), nonabsorptive surface that allows the patient to volitionally void per urethra. The continence mechanism is maintained by the external striated sphincter muscle (rhabdosphincter muscle) of the pelvic floor, whereas voiding is accomplished by concomitantly increasing intraabdominal pressure (Valsalva), along with relaxation of the pelvic floor. The literature is replete with particular opinions on which bowel segment and/or reservoir is optimal for construction of the orthotopic neobladder. The small intestine, terminal ileum and cecum, large intestine, or a combination of these have all been utilized to construct a urinary reservoir. It is the authors' preference to use the small bowel (ileal reservoir) as it appears to offer less contractility, greater compliance, and improved continence rates compared to large bowel neobladders. There is evidence to suggest that the muscular-walled colon is less compliant than ileum and may store urine at higher pressures than ileum. 7 In addition, several clinical studies have demonstrated that the urodynamic characteristics of ileum appear superior to those of the colon. 1,3,8 Furthermore, mucosal atrophy with less reabsorption of urinary constituents appears to be more reliable in small bowel than in large bowel reservoirs. For these reasons, in addition to the ease with which the small bowel can be surgically manipulated, it is our preference to use ileum in the construction of the neobladder. Preoperative Evaluation All women considering orthotopic diversion should have the understanding that if the bladder or pelvic tumor involves the proximal urethra (diagnosed on intraoperative frozen section analysis), then lower urinary tract reconstruction should not be performed. In this case, a cutaneous form of diversion should be performed based on the patient's desires, as discussed preoperatively. It is therefore important to involve the enterostomal therapy nurse in the preoperative period, to mark for an appropriate cutaneous stoma, and to instruct the patient on how to self-catheterize should it be necessary. Radical Cystectomy The technique of en bloc radical cystectomy (anterior exenteration) with bilateral pelvic lymphadenectomy has remained standard and is described in Chapter 24.11 However, preparation of the anterior urethra in women undergoing orthotopic diversion deserves specific mention. This portion of the procedure is critical in maintaining the continence mechanism, and to the successful outcome of the procedure. Elegant neuroanatomic and histologic studies of the female pelvis and urethra in fetal specimens have provided a better understanding of the female urethra and continence mechanism.2 These anatomic dissections have identified three layers of smooth muscle in the proximal two-thirds of the urethra. The innervation of this proximal urethral segment can be traced back to the pelvic plexus coursing along the lateral aspect of the uterus, vagina, and bladder neck. A gradual transition with intermingling smooth muscle to striated pelvic floor muscle can be identified in the mid- to lower third portion of the urethra. This striated pelvic floor muscle, the so-called rhabdosphincter muscle, with its major portion on the ventral aspect of the urethra, is innervated from branches off the pudendal nerve that course along the pelvic floor posterior to the levator muscles. These findings suggest that preservation of the distal half of the urethral musculature, together with the corresponding nerve supply, is crucial in maintaining the continence mechanism in females. Furthermore, a properly performed cystectomy with en bloc removal of the uterus and cervix effectively denervates the bladder neck and proximal urethral sphincter mechanism, rendering them ineffective as a continence mechanism. This unique anatomic study supports the complete removal of the bladder neck with transection of the proximal urethra just beyond the urethrovesical junction because continence is maintained solely by the rhabdosphincter muscle of the lower urethra. In addition, minimal dissection should be performed anterior to the

proximal urethra, which could injure the pudendal innervation to the rhabdosphincter and possibly damage the continence mechanism. When developing the posterior vascular pedicles during the cystectomy in women, the posterior vagina is incised at the apex just distal to the cervix. This incision is carried anteriorly along the lateral and anterior vaginal wall forming a circumferential incision. The anterior lateral vaginal wall is then grasped with a curved Kocher clamp. This provides countertraction and facilitates dissection between the anterior vaginal wall and the bladder specimen. Development of this posterior plane and vascular pedicle is best performed sharply with the use of hemoclips and carried just distal to the vesicourethral junction. Palpation of a previously placed Foley catheter balloon in the bladder assists in identifying this region. This dissection effectively maintains a functional vagina. Furthermore, an intact anterior vaginal wall may help support the proximal urethra through a complex musculofascial support system that extends from the anterior vagina. The vagina is then closed at the apex and suspended to Cooper's ligament to prevent vaginal prolapse or the development of an enterocele postoperatively ( Fig. 83-1).

FIG. 83-1. View of the female pelvis from above. Note that no dissection is performed anterior to the urethra along the pelvic floor. This helps prevent injury to the rhabdosphincter region and corresponding innervation, which are critical to the continence mechanism in women undergoing orthotopic diversion.

Alternatively, in the case of a deeply invasive posterior bladder tumor with concern of an adequate surgical margin, the anterior vaginal wall can be removed en bloc with the cystectomy specimen. After dividing the posterior vaginal apex, the lateral vaginal wall subsequently serves as the posterior pedicle and is divided distally. This leaves the anterior vaginal wall attached to the posterior bladder specimen. Again, the Foley catheter balloon facilitates identification of the vesicourethral junction. The surgical plane between the vesicourethral junction and the anterior vaginal wall is then developed distally at this location. Dissection is carried downward just distal to the proximal urethra, while the remaining urethra distally is left intact with the anterior vaginal wall. Vaginal reconstruction by a clam shell (horizontal) or side-to-side (vertical) technique is required. Other means of vaginal reconstruction may include a rectus myocutaneous flap, detubularized cylinder of ileum, a peritoneal flap, or an omental flap. Regardless, a well-vascularized omental pedicle graft is placed between the reconstructed vagina and neobladder, and secured to the levator ani muscles to separate the suture lines and prevent fistulization ( Fig. 83-2).

FIG. 83-2. Sagittal section of the female pelvis. Note that a well-vascularized omental pedicle graft is placed between the reconstructed vagina and the neobladder. This pedicle graft is secured to the levator ani muscles to separate the suture lines and prevent fistulization. Note also that the vagina is closed at the apex and will be suspended to Cooper's ligament to prevent vaginal prolapse or the development of an enterocele postoperatively (not shown).

It is critical that no dissection be performed anterior to the urethra along the pelvic floor in women undergoing orthotopic diversion. This prevents injury to the rhabdosphincter region and corresponding innervation, which are necessary in maintaining the continence mechanism. Any dissection performed anteriorly may injure these nerves and compromise the continence status. Some reports suggest that a sympathetic nerve–sparing cystectomy is important in maintaining continence in these women. We have routinely sacrificed the autonomic nerves coursing along the lateral aspect of uterus and vagina and relied on the pudendal innervation of the rhabdosphincter region. With this approach we have observed excellent continence results in women undergoing orthotopic diversion. 12 In fact, extensive urodynamic studies in women undergoing orthotopic diversion have identified this rhabdosphincter region as the area that provides the continence mechanism in these women. 5 Furthermore, it is possible that preservation of the sympathetic nerves may contribute to the high incidence of hypercontinence and urinary retention requiring continuous intermittent catheterization reported by Hautmann and associates. 6 Following completion of the posterior dissection (ensuring dissection just distal to the vesicourethral junction), a large Statinsky vascular clamp is placed across the bladder neck. With gentle traction the proximal urethra is divided anteriorly for 270 degrees circumferentially just distal to the bladder neck and clamp. Approximately 6 to 8 anterior urethral sutures (2-0 polyglycolic acid) are then placed. The distal portion of the catheter is then drawn into the wound through the urethrostomy and divided. The Statinsky vascular clamp placed across the catheter at the bladder neck prevents any tumor spill from the bladder. Gentle cephalad tract on the clamped catheter allows placement of 2 to 4 posterior urethral sutures. The posterior urethra is then transected and the specimen is removed. Frozen section analysis is performed on the distal urethral margin of the cystectomy specimen to exclude tumor. Construction of the Kock Ileal Neobladder The Kock ileal neobladder is constructed from approximately a 61-cm segment of terminal ileum ( Fig. 83-3). The reservoir portion of the neobladder is constructed from two 22- to 25-cm segments of distal ileum, whereas a single 17-cm segment of more proximal ileum is used to form the antireflux afferent nipple valve. The distal mesenteric division of the terminal ileum is performed along the avascular plane of Treves, between the ileocolic artery and the terminal ileal branches of the superior mesenteric artery, extending to the base of the small bowel mesentery. The proximal mesenteric division is short to ensure a broad vascular supply to the proximal ileal segment. Usually a 5-cm portion of proximal ileum is discarded proximal to the overall segment in order to ensure adequate mobility of the pouch and the small bowel anastomosis. The proximal end of the isolated ileal segment is closed with a running Parker-Kerr suture (3-0 chromic) and imbricated with a layer of 3-0 silk sutures. The small bowel continuity is reestablished with a standard small bowel anastomosis and the mesenteric trap closed.

FIG. 83-3. The ileal segment used to construct the orthotopic Kock ileal neobladder. The distal mesenteric division is made approximately 20 cm proximal to the cecum along the avascular plane of Treves. A total length of 61 cm of ileum is required in the construction of the neobladder, two 22-cm segments for the reservoir portion of the pouch, and a proximal 17-cm segment for the intussuscepted afferent nipple valve.

The two isolated 22-cm ileal segments are positioned adjacent to one another in a U fashion with the apex directed caudally ( Fig. 83-4A). These 22-cm ileal segments are sewn together along the serosa (side-to-side) approximately 1 cm from the mesentery with a 3-0 polyglycolic acid running suture. The ileum is opened just lateral to this serosal basting suture with electrocautery, The incised mucosa is approximated and oversewn the entire length with a 3-0 polyglycolic acid suture in two layers (Fig. 83-4B). This suture line forms the posterior wall of the reservoir.

FIG. 83-4. Creation of the pouch. (A) The two 22-cm segments are approximated in a U formation (directed to the right lower quadrant) with a serosal basting suture 1 cm from the mesentery (bottom). The ileum is then opened with electrocautery adjacent to the basting suture line. (B) The posterior wall of the reservoir is then sutured in two layers with a continuous 3-0 polyglycolic acid suture ( top). (C) A 5- to 7-cm antireflux valve is created by intussusception of the afferent limb with Allis forcep clamps.

Attention is then directed to the 17-cm ileal segment and construction of the intussuscepted afferent nipple valve. Windows of Deaver are developed alone, with the afferent limb mesentery adjacent to the serosa of this proximal 17-cm segment. The mesentery is divided away from the pouch, extending for 5 to 6 cm. This is best performed with electocautery. This maneuver effectively strips the mesentery from the serosa of the ileum, ensuring that the mesentery is not incorporated into the intussuscepted nipple valve. Furthermore, this also helps prevent future prolapse or extussusception of the afferent nipple valve. One additional window of Deaver is created beyond the previously stripped mesentery, while preserving at least one vascular arcade in between. A 2.5-cm strip, tetracycline-soaked, of doubled polyglycolic acid mesh is then brought through this smaller window. This mesh will serve as an anchoring collar at the base of the nipple valve. The intussuscepted antireflux afferent nipple valve is then created ( Fig. 83-4C). The afferent nipple is intussuscepted by passing two Allis forcep clamps toward the anchoring collar, grasping the mucosa, and inverting the ileum into the pouch. This positions the mesh adjacent to the wall of the pouch at the base of the intussuscepted nipple. Next, two single rows of parallel 4.8-mm nonhemostatic staples (applied with a custom no-knife GIA 50 stapler), with the proximal four staples removed from each row, are applied to the medial and anterior aspect of the nipple ( Fig. 83-5A). A third parallel staple line is applied from outside the pouch between the posterior wall of the pouch along the mesenteric border and one layer of the intussuscepted nipple (applied with a TA-55 stapling device; Fig. 83-5B). The pinhole created by the TA-55 is oversewn from within the pouch to prevent formation of a fistula. This maneuver effectively fixes the afferent nipple valve to the posterior wall of the reservoir.

FIG. 83-5. Fixation of the afferent limb. (A) Two rows of staples (with the distal 4 to 6 staples removed) are placed within the limb to stabilize the two leaves of the valve. (B) The valve is then fixed to the back wall from outside the reservoir. This fixes the intraluminal valve, which becomes more efficient as the reservoir fills, thus preventing reflux.

The anchoring mesh collar is fixed to the base of the afferent nipple valve. The mesh is circumferentially sutured to the serosa at the base of the nipple with 2-0 chromic. This helps prevent prolapse of the nipple valve. The reservoir is closed by folding the ileum in the opposite direction back onto itself, creating a totally detubularized reservoir ( Fig. 83-6). The anterior aspect of the pouch is closed in two layers with 3-0 polyglycolic acid sutures in a water-tight fashion. It should be noted that the anterior suture line is stopped just prior to reaching the end of the right side, allowing easy entry of one finger. This is the most mobile and dependent portion of the reservoir and will be the portion of the reservoir anastomosed to the native urethra.

FIG. 83-6. After the afferent valve is constructed, the reservoir is completed by folding the ileum upon itself and suturing it with a 3-0 polyglycolic acid. The most dependent comer is left open for the urethral anastomosis. The urethroileal anastomosis is completed in a tension-free, mucosa-to-mucosa fashion with interrupted 2-0 polyglycolic acid. Note that the ureteroileal anastomosis is completed prior to the urethral anastomosis.

The ureteroileal anastomosis is then performed end-to-side with interrupted 4-0 polyglycolic acid sutures. We routinely stent the ureters with 8-Fr pediatric feeding tubes. These feeding tubes will be sutured to a 24-Fr hematuria catheter that is passed per urethra and positioned in the pouch. Urethral Anastomosis Following completion of the neobladder, the urethral sutures are placed in the neobladder and tied down to form a tension-free, mucosa-to-mucosa, urethroileal anastomosis (Fig. 83-6). The pelvis is drained by a 1-in. Penrose drain for 3 weeks; the drain is removed along with the urethral catheter at that time.

OUTCOME
Complications Since 1990, 34 women aged 31 to 86 years (median age 66) have undergone orthotopic lower urinary tract reconstruction following cystectomy at our institution. 5 The median follow-up in this group of patients was 21 months (range 7 to 53). The indication for cystectomy in 29 women (85%) was for transitional cell carcinoma of the bladder. There were no perioperative deaths. Three patients (9%) suffered a total of 4 early com-plications: 1 pouch-related (3%, urine leak) and 3 pouch-unrelated. Late complications requiring rehospitalization or reoperation occurred in 3 patients (9%). However, only 1 of these late complications was pouch-related in a woman who developed a Kock pouch stone. To date, there have been no urethral recurrences in this group of female patients. Results Continence status was evaluable in 33 of 34 patients. Complete daytime continence is reported in 28 of 33 (85%) women, whereas complete nighttime continence is reported in 27 (82%) women. The ability to void to completion by Valsalva and relaxation of the pelvic floor is reported in 28 (85%) patients. Only 5 patients require some form of intermittent catheterization to empty their neobladder. All patients are completely sat-isfied. The key to a successful outcome with orthotopic diversion in women is appropriate patient selection and attention to surgical detail with minimal dissection performed along the anterior urethra. Overall, most women undergoing orthotopic diversion are continent, have the luxury of voiding every 4 to 6 hours with excellent voided volumes, retain a more routine micturition pattern, avoid the need for a cutaneous stoma or external urostomy appliance, and live a more normal lifestyle with an improved self-image. Careful preoperative counseling in all female patients considering orthotopic lower urinary tract reconstruction should include the possible need for clean intermittent catheterization in the rare patient unable to void with pelvic floor relaxation and Valsalva. In addition, patients should understand the potential risk of a urethral recurrence and the need for continued long-term surveillance of the retained urethra. Careful follow-up will be necessary to define the true risk of urethral or vaginal wall recurrence in women and to diagnose urethral recurrences in all patients at an early curable stage. Presently, meticulous monitoring of the retained urethra includes careful palpation of the urethroenteric anastomosis on vaginal examination. In addition, voided urine cytology should be performed on a regular basis in all patients at each follow-up visit. The development of orthotopic lower urinary tract reconstruction has been a significant step in the continued progression of urinary diversion. We believe that the orthotopic reservoir should now be considered the gold standard to which other forms of diversion are compared. Orthotopic diversion can now be safely offered to female patients undergoing cystectomy. With this form of diversion we hope that patients with bladder cancer, as well as their physicians, may be encouraged toward an earlier and more aggressive form of therapy with cystectomy, when cure and ultimately survival are greatest. CHAPTER REFERENCES
1. Berglund B, Kock NG, Norlen L, Philison BM. Volume capacity and pressure characteristic of the continent ileal reservoir used for urinary diversion. J Urol 1987;137:29. 2. Colleselli K, Strasser H, Moriggl B, Stenzl A, Poisel S, Bartsch G. Hemi-Kock to the female urethra: anatomical approach to the continence mechanism to the female urethra. J Urol 1994;part 2, 151:500A, Abstract 1089. 3. Davidsson T, Poulsen AL, Hedlund H, et al. A comparative urodynamic study of the ileal and the colonic neobladder. Scand J Urol Nephrol 1992;142 (Suppl.):143. 4. Elmajian DA, Stein JP, Esrig D, et al. The Kock ileal neobladder: update experience in 295 male patients. J Urol 1996;156:920. 5. Grossfeld GD, Stein JP, Bennett CJ, et al. Lower urinary tract reconstruction in the female using the Kock ileal reservoir with bilateral ureteroileal urethrostomy: update of continence results and fluorourodynamic findings. Urology 1996;48:383. 6. Hautmann RE, Paiss T, Petriconi R. The ileal neobladder in women: 9 years of experience with 18 patients. J Urol 1996;155:76. 7. Hinman F Jr. Selection of intestinal segments for bladder substitution: physical and physiological characteristics. J Urol 1988;139:519. 8. Lytton B, Green DF. Urodynamic studies in patients undergoing bladder replacement surgery. J Urol 1989;141:1394. 9. Rowland RG, Mitchell ME, Bihrle R. The cecoileal continent urinary reservoir. World J Urol 1985;3:185. 10. Skinner DG, Boyd SD, Lieskovsky G, Bennett C, Hopwood B. Lower urinary tract reconstruction following cystectomy: experience and results in 126 patients using the Kock ileal reservoir with bilateral ureteroileal urethrostomy. J Urol 1991;146:756. 11. Skinner DG, Lieskovsky G. Technique of radical cystectomy. In: Skinner DG, Lieskovsky G, eds. Diagnosis and management of genitourinary cancer. Vol. 1. Philadelphia: WB Saunders, 1988;605–621. 12. Stein JP, Grossfeld G, Freeman JA, et al. Orthotopic lower urinary tract reconstruction in women using the Kock ileal neobladder: updated experience in 27 patients. J Urol 1996;155(2):399A, Abstract 353. 13. Stein JP, Cote RJ, Freeman JA, et al. Lower urinary tract reconstruction in women following cystectomy for pelvic malignancy: a pathological review of female cystectomy specimens. J Urol 1995;part 2, 154:1329. 14. Stein JP, Stenzl A, Esrig D, et al. Lower urinary tract reconstruction following cystectomy in women using the Kock ileal reservoir with bilateral ureteroileal urethrostomy: initial clinical experience. J Urol 1994;152:1404.

Chapter 84 Neuroblastoma Glenn’s Urologic Surgery

Chapter 84 Neuroblastoma
Yves L. Homsy and Paul F. Austin

Y. L. Homsy: Department of Pediatric Urology, University of South Florida College of Medicine, Tampa, Florida 33607. P. F. Austin: Indiana University School of Medicine, James Whitcomb Riley Hospital for Children, Indiana University Medical Center, Indianapolis, Indiana 46202.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Neuroblastoma represents the most common extracranial malignant solid neoplasm of childhood. The first successful excision of a neuroblastoma was performed in 1916. Surgical therapy remained the primary mode of treatment despite a poor prognosis until adjuvant treatment modalities were later added. Radiation therapy was introduced in 1928 and chemotherapy in 1951; however, little improvement in survival was achieved. Each modality has evolved in its influence and impact on prognosis, but surgery remains the best mode of therapy for localized, nonadvanced disease. For locally advanced or metastatic disease, the role of surgery is not as clear. Nevertheless, total or near-total resection of the primary tumor has recently been shown to be of prognostic significance. 1,20 Neuroblastoma arises along the sympathetic ganglia or from neuroblasts that may have migrated from the mantle layer of the developing spinal cord. Neuroblastomas exhibit a wide spectrum of morphologic differentiation and may be quite primitive (neuroblastoma), more differentiated (ganglioneuroblastoma), or well differentiated (ganglioneuroma). Two-thirds of these tumors present in the retroperitoneum with the majority occurring in the adrenal medulla (more than 65%). Cervical and mediastinal lesions tend to present in younger patients less than 1 year of age. Half of the patients present in the first 2 years of life with the peak incidence occurring between 12 to 18 months of age. The clinical presentation of neuroblastoma is dependent on the site of the primary tumor, presence of metastatic disease, and production of metabolically active substances.6 The catecholamine urinary metabolites vanillylmandelic acid (VMA) and homovanillic acid (HVA) are elevated in more than 80% of neuroblastoma patients. An increased VMA/HVA ratio has been shown to have a better prognosis in localized disease. 3 Other metabolic products that may be elevated include lactate dehydrogenase (LDH), serum ferritin, and serum neuron-specific enolase (NSE), and these have been associated with a poor prognosis. Typically children with neuroblastoma appear more ill than children with Wilms' tumor. At least half of the children with neuroblastoma will have metastatic disease, particularly to the bone, bone marrow, liver, and skin, at the time of diagnosis. Because of the multiple sites of potential origin and involvement, the presentation may vary from paresis/paralysis to Horner's syndrome to numerous subcutaneous tumor deposits (“blueberry muffin babies”).

DIAGNOSIS
A complete history and physical should be performed as well as a meticulous diagnostic workup that should include the tests in Table 84-1. Of particular importance is the presence of multiple copies of the N- myc oncogene (genomic amplification) because N- myc amplification is associated with an unfavorable tumor and poor prognosis. 6 Other molecular studies such as ploidy and chromosome analysis may be performed as well but are optional as the role they play in diagnosis and staging is not yet well defined.

TABLE 84-1. Neuroblastoma diagnostic/staging workup

Multiple staging systems have evolved in the management of neuroblastoma. The Evans-D Angio staging system consists of a clinical assessment that describes the initial tumor distribution and whether the tumor crosses the midline. Both the St. Jude Children's Research Hospital (SJCRH) and the Pediatric Oncology Group (POG) staging systems are based on surgical resectability of the primary tumor and the findings at the time of surgery. The International Neuroblastoma Staging System (INSS) incorporates many of the important criteria from each of the staging systems and includes initial tumor distribution as well as its surgical resectability (Table 84-2). It is important to realize that accurate staging cannot be determined until the time of surgical exploration.

TABLE 84-2. International staging system (INSS)

INDICATIONS FOR SURGERY
If the patient has stage I or II disease, then surgical resection can be undertaken with reasonable success. In a prospective study by the POG in 1988, an 89% 2-year

disease free survival was obtained in 100 patients with localized neuroblastoma when treated by surgery. chemotherapy or radiation even in the presence of residual disease. 10

14

In addition, no advantage is gained by the addition of

A multimodal approach utilizing chemotherapy, radiation, and surgery is used for advanced disease. Patients with stage IV disease and older than 1 year of age unfortunately have a poor outlook; however, intensive multiagent chemotherapy has increased the 5-year survival for patients less than 1 year of age to 75%. 16 Radiation has also been shown to be of benefit in advanced disease and has been demonstrated to give superior initial and long-term disease control when administered together with chemotherapy rather than treating with chemotherapy alone. 4 After chemotherapy is administered, surgical intervention is usually performed 13 to 18 weeks afterward. 1 Surgical resection seems to be somewhat easier around vital structures and major blood vessels because the tumors tend to have become smaller and more firm. This rubbery consistency also decreases the risk of rupture and spillage of tumor that would otherwise be encountered. Radiation therapy may then be delivered to the tumor bed and regional lymph node areas. Intraoperative radiation may be utilized and is helpful in reducing the exposure to adjacent normal structures. Supralethal chemotherapy and total body irradiation followed by bone marrow transplantation has shown promise and a 72% 3-year disease-free survival has been reported in advanced neuroblastoma. 12 The role of surgery is controversial, particularly regarding the extent of surgical resection. Some reports have been unfavorable 10,17 whereas others have reported significant improvement in survival with gross complete resection or near-total (95%) resection. 5,20 Stage IV-S patients may be followed conservatively due to their high rate of spontaneous regression and 60% to 90% survival rate. Resection of the primary tumor has been associated with a survival of 96%, 9 though in the current philosophy operative intervention is felt to be diagnostic and supportive in nature rather than therapeutic. Respiratory insufficiency may develop in these infants with massive hepatomegaly and the use of a silastic pouch to enlarge the abdomen has been recommended. Chemotherapy and radiation have also been advocated for extensive metastatic disease in these patients. Stage IV-S patients have been stratified into high- and low-risk groups based on the patient's age and the pattern of metastatic sites. Low-risk patients benefit more from a conservative approach that includes primary tumor resection when possible followed by careful observation. Because of their poorer prognosis, high-risk patients may benefit more from an aggressive approach that includes multimodal therapy. 18

ALTERNATIVE THERAPY
Due to the extraordinarily complex therapy involved and the relative rarity of this tumor, patients should be managed according to the protocols established by the POG. The therapy is multimodal and there is no accepted alternative therapy available.

SURGICAL TECHNIQUE
Unlike pheochromocytoma where choice of general anesthetic plays an important role in the operative case, no specific optimal anesthetic regimen is recommended and intraoperative hypertension is rare. 7 The patient is placed in the supine position and at the discretion of the surgeon an appropriate incision is made that should provide adequate exposure of the abdomen and retroperitoneum. We prefer an upper abdominal transverse incision, although a chevron incision may also be used for tumors involving the upper abdomen/retroperitoneum. The viscera are reflected to the midline and placed in an intestinal bag. Exploration of the abdomen and retroperitoneum is performed with particular attention to the relationship of the primary tumor to the surrounding vital structures ( Fig. 84-1). Regional lymph nodes are assessed and the liver is biopsied.

FIG. 84-1. CT scan (A) and operative view (B) of a stage I neuroblastoma. Note relationship of the tumor with the sympathetic chain ( small arrow) and the inferior vena cava (large arrow). The vena cava can be seen to be compressed on the CT scan.

Neuroblastoma often invades the tunica adventitia of large blood vessels and consequently resection of these tumors should be regarded as a vascular procedure with appropriate instrumentation and sutures. 8 After adequate exposure of the surrounding vasculature, the dissection begins distal to the tumor typically along the external or common iliac artery. The adventitia is incised longitudinally and once the subadventitial plane is entered, the dissection proceeds proximally to encounter the tumor. Utilization of bipolar cautery as well as the Cavitron Ultrasonic Surgical Aspirator (CUSA) is beneficial during the dissection. The CO 2 laser has likewise been reported to be helpful in the resection of neuroblastoma but the cytotoxic effect seems to be related to its increased tumor immunogenicity rather than to increased tumor kill. 11 Maintaining the subadventitial plane dissection proximally, the anterior aorta is cleared and the origins of the inferior mesenteric, superior mesenteric, and celiac arteries are identified and cleared as well. The renal artery is similarly dissected; when there is tumor present deep in the hilum, nephrectomy is sometimes necessary. The main attachments to the tumor are then divided and the tumor is removed. The surgical management is determined at the time of this staging laparotomy. If primary tumor resection cannot be adequately accomplished, then an open wedge biopsy is obtained for histopathologic and genetic analysis. Neuroblastomas are hypervascular and therefore adequate control of hemorrhage at the biopsy site as well as avoidance of tumor spillage is essential. Tissue should be placed in saline prior to freezing and should be of adequate volume for histopathology, immunohistochemistry, and genetic studies.

OUTCOMES
Complications The morbidity associated with surgical intervention for neuroblastoma has primarily been related to vascular injuries occurring during the dissection. Any major branch originating from the aorta might be injured, as might tributaries of the inferior vena cava. Other reported injuries include a splenic injury, brachial plexus injury, and a chylothorax.2 Acute renal failure has also been reported secondary to renal artery spasm. This may be induced by excessive traction on the renal vessels during the dissection or by stimulation of the autonomic nerve plexi surrounding the origin of the renal arteries with resultant vasospasm. 15 Intravenous digital subtraction arteriography utilizing a central venous catheter has been used to evaluate renal blood flow following surgery. 19 If renal blood flow is compromised, then vasodilator drips such as lidocaine, papaverine, or prostaglandin may be initiated locally to reverse the vasospasm. Results Neuroblastoma is a unique tumor that violates many of the treatises of surgical oncology, being one of the few tumors where a surgeon may violate the capsule or pseudocapsule leaving residual tumor yet still achieve a good overall outcome. Even though surgical therapy has been reported to affect the prognosis of advanced neuroblastoma, it is the biological and immunologic properties that determine much of the eventual outcome. Low-risk tumors tend to occur in lower stage patients

whereas higher risk tumors tend to occur in the more advanced staged patients and henceforth have a poorer prognosis. Although as surgeons we prefer to believe that we provide a significant impact with our intervention, it most likely is the intrinsic properties of the tumor itself that seems to be a greater determinant of the eventual outcome in neuroblastoma. CHAPTER REFERENCES
1. Azizkhan RG, Haase GM. Current biological and therapeutic implications in the surgery of neuroblastoma. Semin Surg Oncol 1993;9:493. 2. Azizkhan RG, Shaw A, Chandler JG. Surgical complications of neuroblastoma resection. Surgery 1985;97:514. 3. Berthold F, Hunneman DH, Harms D, Kaser H, Zieschang J. Serum vanillylmandelic acid/homovanillic acid contributes to prognosis estimation in patients with localized but not with metastatic neuroblastoma. Eur J Pediatr Surg 1992;28A:1950. 4. Castleberry RP, Kun LE, Shuster JJ, et al. Radiotherapy improves the outlook for patients older than 1 year with Pediatric Oncology Group stage C neuroblastoma. J Clin Oncol 1991;9:789. 5. DeCou JM, Bowman LC, Rao BN, et al. Infants with metastatic neuroblastoma have improved survival with resection of the primary tumor. J Pediatr Surg 1995;30:937. 6. Grosfeld JL. Neuroblastoma in infancy and childhood. In: Hays DM, ed. Pediatric surgical oncology. Orlando: Grune and Stratton, 1986;63–85. 7. Kain ZN, Shamberger RS, Holzman RS. Anesthetic management of children with neuroblastoma. J Clin Anesth 1993;5:486. 8. Kiely EM. The surgical challenge of neuroblastoma. J Pediatr Surg 1994;29:128. 9. Martinez DA, King DR, Ginn-Pease ME, Haase GM, Wiener ES. Resection of the primary tumor is appropriate for children with stage IV-S neuroblastoma: an analysis of 37 patients. J Pediatr Surg 1992; 27:1016. 10. Matthay KK, Sather HN, Seeger RC, Haase GM, Hammond GD. Excellent outcome of stage II neuroblastoma is independent of residual disease and radiation therapy. J Clin Oncol 1989;7:236. 11. McCormack CJ, Naim JO, Rogers DW, Ziegler MM, Hinshaw JR. Beneficial effects following carbon dioxide laser excision on experimental neuroblastoma. J Pediatr Surg 1989;24:201. 12. Mugishima H, Iwata M, Okabe I, et al. Autologous bone marrow transplantation in children with advanced neuroblastoma. Cancer 1994;74:972. 13. Nakagawara A, Ikeda K, Tsuda T, Higashi K. N-myc oncogene amplification and prognostic factors of neuroblastoma in children. J Pediatr Surg 1987;22:895. 14. Nitschke R, Smith EI, Shochat S, et al. Localized neuroblastoma treated by surgery: a Pediatric Oncology Group study. J Clin Oncol 1988;6:1271. 15. Ogita S, Tokiwa K, Takahashi T. Renal artery spasm: a cause of acute renal failure following abdominal surgery for neuroblastoma. J Pediatr Surg 1989;24:215. 16. Paul SR, Tarbell NJ, Korf B, Kretschmar CS, Lavally B, Grier HE. Stage IV neuroblastoma in infants. Long-term survival. Cancer 1991;67:1493. 17. Shorter NA, Davidoff AM, Evans AE, Ross AJ III, Zeigler MM, O'Neill JA Jr. The role of surgery in the management of stage IV neuroblastoma: a single institution study. Med Pediatr Oncol 1995;24:287. 18. Stephenson SR, Cook BA, Mease AD, Ruymann FB. The prognostic significance of age and pattern of metastases in stage IV-S neuroblastoma. Cancer 1986;58:372. 19. Yamagiwa I, Obata K, Saito H, Washio M. Intravenous digital subtraction angiography for the evaluation of renal artery blood flow following the removal of a neuroblastoma. Surg Today 2994;24:973. 20. Yokoyama J, Ikawa H, Endow M, et al. The role of surgery in advanced neuroblastoma. Eur J Pediatr Surg 1995;5:23.

Chapter 85 Wilms' Tumor Glenn’s Urologic Surgery

Chapter 85 Wilms' Tumor
O. Lenayne Westney and Michael L. Ritchey

O. L. Westney: Division of Urology, University of Texas Health Science Center, Houston, Texas 77030. M. L. Ritchey: Division of Surgery and Pediatrics, University of Texas Medical School, Houston, Texas 77030.

Diagnosis Indications for Surgery Surgical Technique Bilateral Tumors Postoperative Treatment Results Complications Results Chapter References

Wilms' tumor, or nephroblastoma, represents 5% to 6% of all childhood cancers in the United States and is the most common primary malignant renal tumor of childhood. The incidence rate of Wilms' tumor is 1 in 10,000 children with a new case rate of 450 to 500 annually in the United States. 1 The mean age at diagnosis for unilateral nephroblastoma is 36.5 months for males and 42.5 months for females. The age of peak incidence for bilateral tumors is lower for both sexes: 29.5 months for males and 32.6 months for females. Children with Wilms' tumor may have associated anomalies, including aniridia, hemihypertrophy, and genitourinary tract malformations (hypospadias, cryptorchidism, and renal fusion anomalies). 3 The constellation of Wilms' tumor, aniridia, genitourinary malformations, and mental retardation (WAGR syndrome) occurs in association with a constitutional deletion of chromosome 11. About one-third of sporadic Wilms' tumors show tumor-specific loss of heterozygosity for polymorphic DNA markers on 11p13.6 The sequence of this putative Wilms' tumor gene, WT1, has now been determined. Mutations of WT1 have also been found in patients with the Denys-Drash syndrome. The Denys-Drash syndrome is of particular relevance to urologists due to the association of male pseudohermaphroditism and renal mesangial sclerosis. Wilms' tumor also has an increased incidence in children with Beckwith-Wiedemann, hemihypertrophy, and Perlman syndromes. Beckwith-Wiedemann syndrome (BWS), which consists of macroglossia, omphalocele, and visceromegaly, is associated with a 10% to 20% risk of tumor development, including nephroblastoma, adrenocortical neoplasms, and hepatoblastoma. This syndrome has also been linked to a second Wilms' tumor locus ( WT2) 11p15.5.6.

DIAGNOSIS
More than 90% of children with Wilms' tumor present with an abdominal mass found incidentally by a family member or physician. The mass may be extremely large relative to the child and not necessarily confined to one side. Approximately 25% of children will have hematuria at diagnosis. However, gross hematuria is less common and warrants further evaluation to rule out tumor extension into the collecting system. Children may present more acutely with abdominal pain, leading to exploration for assumed appendicitis. Tumor rupture into the peritoneal cavity or bleeding within the tumor are the common reasons for presentation with an acute abdomen. A persistent varicocele in the supine position or hepatomegaly may be reflective of inferior vena caval obstruction from tumor thrombus. 8 The preoperative evaluation of a child with an abdominal mass can be accomplished in 24 to 48 hours in most medical centers. The laboratory evaluation should include a complete peripheral blood count, differential white blood cell count, platelet count, liver function tests, and renal function tests. There is an 8% incidence of acquired von Willebrand's disease in newly diagnosed Wilms' tumor patients. 4 This defect can be corrected preoperatively with the administration of 1-desamino-8-D-argiuine-vasopressin (DDAVP). Serum calcium should also be checked as this can be elevated in both congenital mesoblastic nephroma and rhabdoid tumor of the kidney. Children with abdominal masses require radiographic evaluation. The first study that should be obtained is an abdominal ultrasound, which can differentiate between solid and cystic masses. 1 For children with renal tumors, real-time ultrasonography of the inferior vena cava is necessary to exclude intracaval tumor thrombus, which occurs in 4% of patients with Wilms' tumor. 8 If this study is inconclusive, magnetic resonance imaging (MRI) is an excellent modality to assess the venous system. There is controversy regarding the need for additional imaging studies if ultrasound yields the information listed above. 1 Additional data required in a patient suspected of having Wilms' tumor is the presence of a contralateral functioning kidney and detection of pulmonary metastases. The question is whether advanced imaging studies, e.g., computed tomography (CT) or MRI, can provide useful staging information. Definition of local tumor extent and assignment of tumor stage is determined by surgical and pathologic findings. CT can suggest extrarenal extension into the perirenal fat or adjacent organs (e.g., liver) and regional adenopathy, but must be confirmed at surgery. Lungs represent the most common site of metastases and plain chest radiographs should be obtained to exclude pulmonary lesions.

INDICATIONS FOR SURGERY
The initial management of a child with Wilms' tumor is abdominal exploration. In most cases, a radical nephrectomy can be performed. Histopathology and tumor stage have been demonstrated to be the key determinants of prognosis in patients with Wilms' tumor. Assignment of tumor stage ( Table 85-1) is based on intraoperative and pathologic findings. Therefore, the surgeon has an essential role in determining the treatment.

TABLE 85-1. Staging system of the National Wilms Tumor Study

The International Society of Pediatric Oncology (SIOP) advocates primary hemotherapy for all patients with Wilms' tumor regardless of extent of disease. Preoperative treatment can produce dramatic reduction in the size of the primary tumor facilitating surgical excision. The SIOP trials have utilized preoperative treatment for Wilms' tumor since the early 1970s, and their studies have demonstrated that the incidence of tumor rupture is lower after preoperative therapy. 10 It should be noted that there was no survival advantage over a primary surgical approach. One notable difference between SIOP and the National Wilms' Tumor Study Group (NWTSG) is that SIOP investigators use the postchemotherapy stage to determine the amount of postoperative therapy, which may inadequately define the risk of intraabdominal

recurrence in unirradiated patients. The NWTSG recommends preoperative chemotherapy only in children with bilateral tumors, tumors inoperable at surgical exploration, and inferior vena cava extension above the hepatic veins. All other patients should undergo primary excision of the tumor. This will allow precise staging of patients with modulation of treatment for each individual, thereby decreasing the intensity of treatment when possible while maintaining excellent overall survival.

SURGICAL TECHNIQUE
Nephrectomy is routinely performed via a generous transverse abdominal incision. The patient is positioned in a supine fashion with some flexion of the lumbar spine to facilitate the exposure of retroperitoneal structures. The incision is made approximately 2 fingerbreadths above the umbilicus. The incision begins in the midaxillary line on the side of the neoplasm. The extent to which the incision is extended across the midline will vary with the size of the tumor and amount of exposure needed. The incision may be extended into a thoracoabdominal approach by continuing through the bed of the 9th or 10th rib, if necessary. The muscle layers are divided sequentially to facilitate exposure. The peritoneal space should be opened very carefully. The tumor will compress the colon and/or small bowel up against the anterior abdominal wall, which can inadvertently lead to enterotomy. A thorough exploration of the abdomen is then performed. The peritoneal cavity is assessed for evidence of preoperative tumor rupture, tumor implants, or drop metastases in the pelvis. The liver is carefully examined, as many of the liver metastases are not identified on preoperative imaging studies, and an assessment of tumor extent is then performed, including palpation of the inferior vena cava, assessment of regional lymphadenopathy, perinephric extension, and tumor mobility. Prior to proceeding with nephrectomy, the contralateral kidney is examined. The colon is reflected medially by incising the white line of Toldt. Gerota's fascia is opened to allow inspection as well as palpation of the anterior and posterior surfaces of the kidney. Any suspicious lesions should be biopsied for frozen section to exclude Wilms' tumor or nephrogenic rests. The nephrectomy now proceeds by reflection of the colon in a similar fashion as to expose the contralateral kidney. The colonic mesentery is mobilized with care taken to preserve the colonic vessels that are draped over the tumor. The colon can then be retracted medially to expose the renal vessels ( Fig. 85-1). For right-sided tumors, the posterior peritoneum can be incised up to the base of the mesentery. This will allow reflection of the entire colon and small bowel, which provides excellent exposure of the retroperitoneal vessels.

FIG. 85-1. Descending colon retracted medially after incising of the line of Toldt and mobilizing of the mesocolon off the anterior surface of the tumor.

If possible, the renal vessels should be ligated at the beginning of the operation. Once the renal vein has been identified, a vessel loop is placed around the vein. Any nodal tissue around the renal vein may be sent as part of the permanent specimen. The renal vein and inferior vena cava should be carefully palpated for the presence of tumor thrombus. The artery can be identified with careful retraction of the vein. Prior to ligation of the vessels, the contralateral renal vessels and superior mesenteric artery are identified to avoid injury to these structures. The vessels are then doubly ligated and divided. An alternative for management of the renal vein is to place a Satinsky clamp on the vena cava just proximal to the insertion of the renal vein. This is of great value when the vein is short or if there is tumor extension through the renal vein. The venous stump in the Satinsky is oversewn with continuous 5-0 prolene in two layers after the vein is divided ( Fig. 85-2A, B).

FIG. 85-2. Mobilization of left Wilms' tumor by blunt dissection after ligation and division of the renal artery (A) and vein (B).

If a tumor thrombus is present in the inferior vena cava, additional surgical exposure is necessary. In order to extract tumor thrombus from within the vena cava, both proximal and distal vascular control are necessary. For minimal extension well below the hepatic veins, the inferior edge of the liver can be retracted to expose the infrahepatic vena cava. For a tumor that extends more cephalad, mobilization of the liver is required. Division of the triangular and coronary ligaments of the liver allows rotation and exposure of the retrohepatic vena cava. Additional exposure can be gained by dividing the lesser hepatic veins. The contralateral renal vein and infrarenal IVC are controlled with vessel loops. The vena cava is then vertically incised just medial to the entrance of the renal vein. If the thrombus is free-floating, it may easily be milked out at this point. In many cases, however, the thrombus is adherent to the wall of the inferior vena cava. A Fogarty or Foley balloon catheter is passed beyond the level of the hepatic veins, the balloon inflated and pulled inferiorly, displacing the thrombus into the cavotomy. The vena cava is allowed to fill by releasing the vessel loops on the distal cava and contralateral renal vein. This will displace the air from within the cava. The cavotomy is then clamped with a Satinsky and oversewn in a continuous fashion with a 5-0 prolene suture (Fig. 85-3).

FIG. 85-3. Surgical technique to manage tumor extension through the renal vein. (A) Into the vena cava (limited to the infrahepatic level). (B) After exposure of the vessels, the infrarenal vena cava and contralateral renal vein are controlled with vessel loops and the vena cava is incised vertically at the intersection with the renal vein. (C) A Fogarty catheter is passed superior to the tumor thrombus and the balloon inflated. (D) The vena cava is flushed of air and a Satinsky clamp is placed to allow closure of the cavotomy.

After the vessels are controlled, a dissection plane is established outside of Gerota's fascia by sharp and blunt dissection. The perforating vessels can be quite large and should be ligated individually. Gentle handling of the tumor is needed to avoid rupture of the tumor during this dissection. Wilms' tumors are very soft, and it is easy to enter the tumor resulting in either local or diffuse tumor spill. The ureter is divided as low as possible after palpation of the ureter to rule out intraureteral extension. The lymphatic tissue in the renal hilum and adjacent precaval and preaortic area is generally removed with the specimen. Formal lymph node dissection is not required, but all suspicious lymph nodes should be biopsied. After removal of the tumor, the wound is irrigated with saline and hemostasis is assessed. A drain is not routinely left in place unless a portion of the pancreas or liver has been resected. The displaced colon is placed back in the tumor bed. Bilateral tumors Radical nephrectomy should not be performed at the initial operation of a child with bilateral Wilms' tumors. Children treated with preoperative chemotherapy have an equivalent survival to those undergoing initial radical surgery, but more renal units can be preserved in those given preoperative chemotherapy and partial nephrectomy(ies).7 Initial exploration of the abdomen and biopsy of both kidneys are performed to verify the histologic type of each tumor, although sampling errors may still occur. Partial nephrectomy or wedge excision can be employed at the initial operation only if all tumor can be removed with preservation of the majority of renal parenchyma on both sides. Grossly abnormal lymph nodes or other lesions suggestive of extrarenal spread should be biopsied and a surgical stage assigned to each kidney. The patient is given preoperative chemotherapy appropriate to the stage and histology of the tumor. The response of the tumors is evaluated by CT after week 5. This can assess the reduction in tumor volume and the feasibility of partial resection. At the time of the second-look procedure, partial nephrectomy or wedge excision of the tumor is performed, but only if it will not compromise tumor resection and negative margins can be obtained. If complete excision of tumor from one kidney can be performed leaving a viable and functioning kidney, then radical nephrectomy is performed to remove the contralateral kidney with extensive tumor involvement. Enucleation of the tumor should be considered in lieu of a formal partial nephrectomy only if removal of a margin of renal tissue would compromise the vascular supply to the kidney. If the tumor is not amenable to partial resections, repeat biopsies should be performed. Patients with persistent viable tumor that cannot be resected should be treated with a different chemotherapeutic regimen. The patient should be reassessed after an additional 12 weeks of chemotherapy to determine the feasibility of resection. If there is extensive tumor involvement precluding partial resection in a solitary kidney, radiation therapy can then be instituted to effect tumor shrinkage. Postoperative Treatment Current recommendations have been given for treatment to be utilized in the recently opened intergroup study NWTS-5 ( Table 85-2) in which biological features of the tumors will be assessed in patients who will not be randomized for therapy. This study will attempt to verify the preliminary findings that (LOH) for chromosomes 16q and 1p are useful markers in identifying patients who will relapse. 6 If these variables are found to be predictive of clinical behavior, then this information will be used in subsequent clinical trials to further stratify patients for therapy.

TABLE 85-2. Protocol for National Wilms' Tumor Study–5

The treatment for patients with stage I or II favorable histology (FH) and stage I anaplastic Wilms' tumor is the same. They will receive a pulse-intensive regimen of vincristine (VCR) and dactinomycin (AMD) for 18 weeks. A select group of patients under two years of age with stage I FH tumors weighing under 550 g will be selected for management with surgery alone in NWTS-5. 5 Careful postoperative surveillance of this group of children will be necessary so that any relapses can be detected early. Patients with stage III FH and stage II–III focal anaplasia are treated with AMD, VCR, and doxorubicin (DOX) and 1080 cGy abdominal irradiation. Patients with stage IV FH tumors receive abdominal irradiation based on the local tumor stage and 1200 cGy to both lungs.

RESULTS
Complications Despite improvements and refinement of surgical technique, the removal of a large nephroblastoma is still prone to intraoperative and postoperative complications. Review of the charts of 1910 children with unilateral Wilms' tumor enrolled in NWTS-3 revealed a complication rate of 19.8%. 9 The most common complication was intestinal obstruction (6.9%) secondary to intestinal adhesions or intussusception. This was followed closely by extensive hemorrhage (5.9%), defined as intraoperative blood loss exceeding 50 ml/kg of body weight ( Table 85-3).

TABLE 85-3. National Wilms' Tumor Study–3: Most commonly reported surgical complications after unilateral nephrectomy

A mortality rate of 0.5% (intraoperative 0.05%) related to surgical complications is reported from NWTS-3. However, a higher intraoperative mortality rate of 1.5% has been reported from other centers. The risk factors associated with surgical complications are higher tumor stage, tumor size greater than 10 cm, incorrect preoperative diagnosis, thoracoabdominal incision, extrarenal intravascular tumor extension, and resection of other visceral organs. Results With current multimodal therapy, the overall 4-year survival for patients with favorable histology is approximately 90%. However, the regimens have not been as successful, with clear cell sarcoma of the kidney and rhabdoid tumor of the kidney patients demonstrating 4-year survival rates of 75% and 25%, respectively. Children with stage II–IV diffuse anaplasia and stage I–IV clear cell sarcoma and rhabdoid tumor of the kidney will be treated with new chemotherapeutic regimens in an attempt to further improve the survival of these high-risk groups. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Babyn P, Owens C, Gyepes M, D'Angio GJ. Imaging patients with Wilms' tumor. Hematol Oncol Clin North Am 1995;9:1217–1252. Breslow N, Olshan A, Beckwith JB, Green DM. Epidemiology of Wilms' tumor. Med Pediatr Oncol 1993;21:172–181. Clericuzio CL. Clinical phenotypes and Wilms' tumor. Med Pediatr Oncol 1993;21:182–187. Coppes MJ, Zandvoort SWH, Sparling CR, et al. Acquired von Willebrand disease in Wilms' tumor patients. J Clin Oncol 1993;10:1–7. Green DM, Beckwith JB, Weeks DA, et al. The relationship between microsubstaging variables, tumor weight, and age at diagnosis of children with stage I/favorable histology Wilms' tumor. A report from the National Wilms' Tumor Study. Cancer 1994;74:1817–1820. Grundy P, Coppes MJ, Haber DA. Molecular genetics of Wilms' tumor. Hematol Oncol Clin North Am 1995;9:1201–1216. Horwitz J, Ritchey ML, Moksness J, et al. Renal salvage procedures in patients with synchronous bilateral Wilms' tumors: a report of the NWTSG. J Pediatr Surg. 1996;31:1020–1025. Ritchey ML, Kelalis PP, Breslow N, et al. Intracaval and atrial involvement with nephroblastoma: review of National Wilms' Tumor Study–3. J Urol 1988;140: 1113–1118. Ritchey ML, Kelalis PP, Breslow N, et al. Surgical complication following nephrectomy for Wilms' tumor: a report of National Wilms' Tumor Study–3. Surg Gynecol Obstet 1992;175:507–514. Tournade MF, Com-Nougue C, Voute PA, et al. Results of the Sixth International Society of Pediatric Oncology Wilms' Tumor Trial and Study: a risk-adapted therapeutic approach in Wilms' tumor. J Clin Oncol 1993;11:1014–1023.

Chapter 86 Renal Fusion and Ectopia Glenn’s Urologic Surgery

Chapter 86 Renal Fusion and Ectopia
Ross M. Decter

R. M. Decter: Department of Surgery, Section of Urology, Pennsylvania State Geisinger Health System, Hershey, Pennsylvania 17033.

Embryology Horseshoe Kidney Renal Ectopia Diagnosis Ectopic Kidney Indications for Surgery Alternative Therapy Surgical Technique Pyeloplasty Ureterocalicostomy Surgery for Tumors Stone Surgery in the Horseshoe Kidney Surgical Options for the Ectopic Kidney Outcomes Complications Results Chapter References

Although clinical problems associated with abnormali-ties of renal fusion and ectopia present infrequently in urologic practice, because these anomalies predispose to infection, hydronephrosis, stone disease, and, in some instances, neoplasia, a clear understanding of these anatomic variants and the deviation from standard urologic techniques to address them is important.

EMBRYOLOGY
The ureteral bud branches from the Wolffian duct and extends toward the metanephric blastema during the fourth and fifth week of gestation, inducing the metanephric blastema to form the functioning kidney. The exact mechanism of renal ascent is not known but during normal development the kidneys ascend and rotate. The renal pelvis rotates from an initial anterior position 90 degrees toward the midline until it reaches its final medial position. Migration and rotation occur simultaneously between the fourth and eighth or ninth week of gestation. During renal ascent, the blood supply to the kidney is derived from successively higher levels of the aorta and its branches. The most common anomaly of renal position is incomplete rotation of the kidney to its final position. The renal pelvis in these malrotated kidneys generally lies anterior to the parenchyma as opposed to its normal medial location. Malrotation may be seen in kidneys that are otherwise normally positioned and malrotation is commonly observed in ectopic kidneys. Simple malrotation of a normally positioned kidney is often an incidental finding. These pyelocaliceal systems are morphologically abnormal, but functionally they usually drain without impairment. Close approximation of the proliferating renal blastemas prior to significant ascent is a normal embryologic finding. 5 If there is any disturbance of separation of the closely approximated renal blastemas, fusion anomalies of the kidneys may develop. The most common fusion anomaly is the horseshoe kidney. It is unknown as to whether this is caused by an abnormality of the arterial supply to the developing kidney, abnormal unfolding of the tail of the embryo, or some teratogenic agent. The horseshoe kidney generally ascends until the upper border of the isthmus is at the level of the inferior mesenteric artery. The second most common fusion anomaly is crossed-fused ectopia. This occurs when the developing kidney crosses from one side to the other during its ascent or when the ureteral bud from one side crosses to the contralateral side and induces abnormal development of that metanephric blastema. Crossed ectopia with fusion may occur in a variety of forms (Fig. 86-1). Although crossed ectopia occurs most frequently with fusion, the anomaly may occur without fusion ( Fig. 86-2).

FIG. 86-1. Six types of crossed renal ectopia with fusion. (A) Ectopic kidney superior. (B) Sigmoid or S-shaped kidney. (C) Lump kidney. (D) L-shaped kidney. (E) Disk kidney. (F) Ectopic kidney inferior. (Modified from McDonald JH, McClellan DS. Crossed renal ectopia. Am J Surg 1957;93:995.)

FIG. 86-2. Types of crossed renal ectopia. (A) Fused. (B) Nonfused. (C) Solitary. (D) Bilateral. (Modified from McDonald JH, McClellan DS. Crossed renal ectopia. Am J Surg 1957;93:995.)

HORSESHOE KIDNEY
Horseshoe kidney is the most common of the fusion anomalies. It occurs in 1 n 400 to 1 in 1800 births. 8 There is a male predominance for the condition. 10 The fusion in horseshoe kidney almost always occurs at the lower poles. Only rare cases of upper pole fusion are recorded. 10 The isthmus of the horseshoe kidney lies as

mentioned above, just below the inferior mesenteric artery at the L4 vertebral level. The blood supply to these kidneys is often variable ( Fig. 86-3).

FIG. 86-3. Three common variants of blood supply in horseshoe kidney. (A) Single renal arteries arising from the aorta. (B) Multiple aortic arteries. (C) Multiple aortic and iliac arteries.

Horseshoe kidney is associated frequently with abnormalities of other systems, suggesting that a teratogenic factor causing widespread abnormalities may be etiologically important. About one-third of patients with these kidneys have such anomalies. Predominant among these are nonrenal genitourinary problems, including hypospadias and undescended testicles. These occur in about half of the patients with associated anomalies. Abnormalities of the musculoskeletal and cardiovascular systems each occur in about one-third of patients who have associated anomalies. Cardiovascular anomalies include ventriculoseptal defect (VSD). Musculoskeletal anomalies include spina bifida occulta, polydactyly, cleft lip and palate, hemivertebra, and scoliosis. Gastrointestinal anomalies include anorectal imperforation, rectosigmoid duplication, and malrotation of the kidney. In addition, horseshoe kidney is clearly associated more frequently with certain neurologic conditions including myelodysplasia. Chromosomal abnormalities have been associated with an increased risk of horseshoe kidneys. In Turner's syndrome, horseshoe kidney is a common occurrence and the diagnosis of a horseshoe kidney in a female suggests the need for a karyotype. Horseshoe kidneys are seen more frequently in patients who have trisomy 18 and in patients with long-arm deletions of chromosome 18. 1 The horseshoe kidney seems to be at no greater risk for renal malignancies than an orthotopically positioned kidney, but the distribution of the types of renal tumors in these kidneys is remarkably different from that of normal kidneys. In horseshoe kidneys, the proportion of transitional cell carcinoma and Wilms' tumor is much higher than expected. In fact, in one series of renal tumors occurring in horseshoe kidneys, 20% were transitional cell tumors, 25% Wilms', and only 50% were renal cell carcinomas. 2 It seems reasonable to suggest serial ultrasonographic follow-up in children with horseshoe kidneys to allow for early detection of a Wilms' tumor.

RENAL ECTOPIA
An ectopic kidney lies outside of its normal position in the renal fossa. In simple ectopia, the kidney lies in the ipsilateral retroperitoneal space at a position that is generally lower than normal ( Fig. 86-4). It may be pelvic, iliac, lumbar, or thoracic. In crossed ectopia, the kidney crosses the midline and it is frequently fused to its contralateral mate. The autopsy incidence of renal ectopia is about 1 in 1000 cases. In screening studies, these kidneys were discovered in about 1 of 5000 cases and ectopic kidneys become clinically recognized in about 1 of 10,000 patients. Therefore, the condition often is totally asymptomatic. 6

FIG. 86-4. Location of lumbar and pelvic ectopically positioned kidneys in relation to the normally positioned kidney.

Reviews of renal ectopia show that the left and right kidneys are ectopic with close to equal frequency. Ectopic kidneys occur bilaterally around 10% of the time. The most common position of the ectopic kidney is pelvic. Pelvic ectopia was found in about 55% of the patients in one series; crossed-fused ectopia occurred in 27%; lumbar ectopia in 12%; noncrossed fused ectopia about 5% of the time; and thoracic kidney about 1% of the time. 6 Rarely, a solitary pelvic kidney occurs. This kidney suffers the risk of injury during pelvic surgical procedures and occasionally has been reported as an unusual cause of giant hydronephrosis. Ectopic kidneys are smaller than their contralateral mates. 3 The blood supply to the pelvic kidney, the commonest of the ectopic kidneys, is variable. The arterial supply may arise from the distal aorta or bifurcation, the ipsilateral common iliac, or the hypogastric vessels. In general, the lower the kidney is in its pelvic location, the greater the likelihood that multiple arterial vessels are supplying it. 4 Patients with renal ectopia frequently have a variety of other associated conditions. Musculoskeletal anomalies including vertebral abnormalities and nonrenal genitourinary conditions including hypospadias, undescended testicles, and anomalies of the vagina and Müllerian ducts occur in about a quarter of patients. Appropriate gynecologic evaluation is therefore mandated. 3 Significant cardiovascular conditions including tetralogy of Fallot, patent ductus arteriosus, VSD, and aortic coarctation have been recorded in association with renal ectopia. In addition, gastrointestinal tract anomalies including imperforate anus, cloacal anomalies, and Bochdalek's hernia are described. 6

DIAGNOSIS
Patients with horseshoe kidneys seem to present in two groups. The first group consists of neonates or stillborns who have severe anomalies of other organ systems. Death in these children results from the anomalies in other systems. 10,11 The second group of patients comprises those who survive beyond the newborn period. The horseshoe kidney is asymptomatic in between a quarter and a third of these patients. 10 In children, typical symptoms leading to presentation include urinary tract infection in half; an abdominal mass, hematuria, or abdominal pain each occurring in about 10%; and other assorted causes in the remainder. The diagnostic evaluation of the clinical problem presenting in the horseshoe kidney generally proceeds along standard lines. Most children will have been evaluated initially with an ultrasound and many subsequently have intravenous pyelography (IVP). Most adults have IVP as their initial study. The intravenous pyelographic features of the horseshoe kidney are typical. One observes the abnormality of the renal axis with the more vertical orientation or lateral tilt of the renal axis. The renal pelves tend to be located anteriorly and the ureters course ventral to the isthmus. The lower calyces are oriented caudally or even medially as opposed to laterally. Kidneys with fusion anomalies are subject to a high incidence of reflux, variably reported between 20% and 50%. Voiding cystourethrography (VCUG) is therefore mandated during the evaluation of these patients. At times the diagnosis of ureteropelvic junction (UPJ) obstruction in a horseshoe kidney is straightforward; the patient's symptoms and IVP that reveals significant pyelocaliectasis can lead to the diagnosis. In other instances, with less severe dilation and especially when there is coexistent stone disease, we find the diuretic

renal scan valuable in helping to assess the drainage of these systems and deciding whether the hydronephrosis should be surgically addressed. Ectopic Kidney Patients who are symptomatic from their ectopic kidney frequently present with urinary tract infection. An ectopic kidney may also be discovered in the evaluation of abdominal pain. The workup of a palpable abdominal mass and discovery of the abnormal renal position in the workup of other associated anomalies each occur in about one-fifth of the cases. Hematuria, incontinence, renal insufficiency, and nephrolithiasis are less common presenting complaints. It is important to emphasize that the majority of patients who have ectopic idneys will be asymptomatic. The evaluation of the presenting symptoms generally proceeds along standard lines. Often in children an ultrasound evaluation leads to the recognition of the condition. In older patients, IVP is frequently performed. The ectopic kidney can be difficult to detect on the intravenous pyelogram as the pyelocaliceal system often overlies the bony pelvis. Special scrutiny needs to be directed to these areas to observe these systems. Functional evaluation of the ectopically positioned kidney is routinely performed with a diuretic renal scan. We find this testing very helpful in trying to assess whether or not a hydronephrotic ectopic kidney is truly obstructed. Ectopic kidneys have a high incidence of vesicoureteral reflux and consequently VCUG should be a routine part of the evaluation of these patients. Reviews of symptomatic patients with ectopic kidneys reveal that over half of patients present with hydronephrosis. The hydronephrosis is due to obstruction, most frequently at the UPJ but at times at the ureterovesical junction. In about one-quarter of cases the hydronephrosis in ectopic kidneys is a consequence of reflux; in a similar number, it is due to the configuration of extrarenal calyces that are neither obstructive nor refluxing. 6 The extrarenal calyces seen with ectopic kidneys look clubbed on IVP, as if they are affected by obstruction or chronic infection. In most instances, however, they are not affected by these processes but are simply an anatomic variant seen in association with the abnormal position of the kidney. 3

INDICATIONS FOR SURGERY
The indications for surgical intervention in the horseshoe kidney are similar to those in the normally positioned kidney. Pyeloplasty is required in patients with symptomatic UPJ obstruction or when it is considered that the abnormality at the UPJ may impact on ultimate renal function. Symptomatic stone disease needs to be cleared by either open, endoscopic, or extracorporeal technique. The evaluation of infections in horseshoe kidney includes a VCUG and ureteral reimplantation may be mandated if reflux is of high grade, persists, or if prophylaxis fails to prevent infection.

ALTERNATIVE THERAPY
The alternative to surgical intervention is nonoperative management, usually including no active therapy and antibiotics. In cases of severe obstruction, stones, and tumors, surgery is the only viable option. Endopyelotomy has recently been utilized to treat UPJ obstruction in horseshoe kidneys. The initial results of endo-pyelotomy performed by experienced surgeons are encouraging; however, we currently prefer pyeloplasty as the initial procedure on UPJ obstructions. Whereas pyelolithotomy has in past decades been utilized to clear calculi from horseshoe kidneys, more recently percutaneous and extracorporeal techniques have been employed. Extracorporeal shock wave lithotripsy (ESWL) in horseshoe kidneys has not enjoyed the success rates that it provides in orthotopically positioned kidneys. Most series note utilization of an increased number of shocks, the need for an increased retreatment rate, and an overall decreased stone clearance rate in horseshoe kidneys as compared to stones in conventionally positioned kidneys. One series recorded a 73% stone-free rate using ESWL after multiple treatments in horseshoe kidneys. 9 One of the reasons for difficulties treating stones with ESWL is that the anterior position of the stone makes it harder to position the stone at the F2 focus; often the surgeon will have to employ the blast path to try to fragment the stone. Some investigators have used prone positioning to overcome this problem. Percutaneous access to the horseshoe kidney seems readily achievable by most experienced percutaneous surgeons. Most investigators note that the use of a midto posterior calyx allows good access to the renal pelvis and state that the percutaneous access is not generally problematic. 7 Reports comparing ESWL of stones in horseshoe kidneys to percutaneous nephrostolithotomy suggest that the latter procedure provided superior stone clearance rates.

SURGICAL TECHNIQUE
Pyeloplasty Pyeloplasty is the most common procedure performed on the horseshoe kidney. Division of the isthmus with nephropexy was thought in the past to be an important part of the procedure, but recent experience suggests that isthmus division or symphysiotomy is rarely necessary in the correction of UPJ obstruction. The surgical exposure of the horseshoe kidney can be achieved through a midline transperitoneal, anteriorly positioned flank extraperitoneal, or transverse transperitoneal approach. The transverse transperitoneal exposure seems to provide the widest exposure with a cosmetically acceptable scar and so we prefer it for pyeloplasty. The incision extends from the anterior axillary line on the affected side crossing the midline several centimeters below the umbilicus. It can be extended laterally in either direction if necessary. Depending on the position of the affected UPJ, the posterior peritoneum may be incised medial to the inferior mesenteric vein up to the ligament of Treitz, inferior and laterally along the small bowel mesentery around the cecum, and up along the line of Toldt on the right side. The small bowel and cecum can then be reflected upward out of the operative field and packed in the upper abdomen. Exposure is maintained with a ring retractor. Repair of UPJ obstruction in the horseshoe kidney can be performed by a Foley Y-V plasty or a dismembered pyeloplasty. Although the Foley Y-V repair is nicely suited to the typical high-insertion obstruction seen in horseshoe kidneys ( Fig. 86-5), we prefer the dismembered technique because it seems to provide more flexibility. During conduct of the pyeloplasty care must be taken to avoid inadvertent division of small vessels to the parenchyma and excessive dissection of the ureter or pelvis. In general, as much adventitial tissue is left on the ureter as possible and no vessels to the ureter are sacrificed unless their division is absolutely necessary to provide for adequate mobilization.

FIG. 86-5. The Foley Y-V technique is illustrated in A–C, the dismembered technique in D–G. (A) Dashed lines indicate inverted Y-shaped incision. Stay sutures of 5-0 chromic help define the margins of the incision. (B) A¢ indicates the tip of the renal pelvic flap. A indicates the inferior margin of the ureteral incision. (C) A and A¢ are sutured together with 7-0 Vicryl, creating a dependent, widely patent anastomosis. (D) The ureter is divided distal to the ureteropelvic junction after stay stitches are positioned. The inverted V, indicating the outline of the pelvic incision, is indicated by a dashed line. (E) The ureter has been spatulated and the pelvic flap developed. The initial stitch of 7-0 Vicryl is positioned at the heel of the anastomosis. (F) The interrupted sutures around the heel are completed. (G) The running locking sutures extending between the spatulated ureter and pelvis are completed creating a widely patent anastomosis.

After the proximal ureter and renal pelvis are adequately exposed using sharp dissection, two stay stitches of 5-0 chromic are positioned in the ureter just below the UPJ (Fig. 86-5). The ureter is divided between these stitches and carefully mobilized. A flap is then created by orienting an inverted V-shaped incision on the renal pelvis with the apex of the inverted V at the UPJ. The flap is designed so that when it is opened it will provide an adequate dependent portion of pelvis for the ureteral anastomosis. It is important that the base of the V be wide to avoid ischemia of the flap. The flap is opened with Wescott tenotomy scissors and the tip is trimmed

minimally to smooth the point of the V. The ureter is then positioned so that one can judge the length of the spatulation. The fact that the ureter is dismembered allows one to position it in such a way that the ureteral spatulation can extend in a relatively wide portion of the ureter and simultaneously orient it so that there is no redundancy that might kink the ureter below the UPJ repair. In addition, the anastomosis seems technically easier than the Foley Y-V as the ureter is not fixed at two points. The spatulation is created using Potts scissors on the posterior aspect of the ureter such that when it is laid on the dependent pelvic flap it will not be twisted. The anastomosis and dissection are performed with the aid of 2.5 power optical magnification. The anastomosis is performed using 7-0 Vicryl in children; in adults, 5-0 chromic or polydioxunone suture (PDS) is employed. We begin the anastomosis at its heel, suturing the most dependent portion of the V-shaped incision to the apex of the ureteral spatulation. The initial portion of the anastomosis is performed using interrupted sutures, generally one at the apex and two on either side of the apex. Each stitch must be precisely positioned to avoid postoperative leakage and/or stricturing. After the apex is anastomosed, the remainder of the pyeloplasty is performed using a running locking 7-0 Vicryl suture up one side of the spatulated ureter and then up the other side. Prior to complete closure, the anastomosis is tested for patency with five and eight feeding tubes. A double-J stent is placed in adults; no stent or diversion is generally employed in children. A Penrose drain is positioned near the anastomosis and made to exit extraperitoneally through a separate stab wound. The posterior peritoneum is approximated over the repair and the abdominal wall closure is performed using running 3-0 or larger PDS. We generally close the skin with a subcuticular pull-out stitch of 3-0 prolene. Most children are discharged 1 or 2 days after surgery. The skin suture is removed between 5 and 7 days postoperatively and the drain is removed at that time if drainage is minimal. Ureterocalicostomy A ureterocalicostomy may be performed to salvage a failed prior pyeloplasty, to correct UPJ obstruction when there is a small intrarenal collecting system, or in other instances when the lower pole parenchyma is extremely thinned ( Fig. 86-6). The ureter is separated from the pelvis as described above. The parenchyma over the lower pole calyx is incised and is resected to allow adequate exposure of the calyx. After hemostasis is achieved using cautery and/or sutures of 3-0 or 4-0 chromic through the edge of the resected parenchyma, the pelvis is incised from the region of the UPJ down into the exposed lower pole calyx. The ureter is spatulated and the anastomosis between the ureter and the opened calyx and pelvis is performed as previously described. It is important to resect enough parenchyma so that it does not impinge on the anastomosis. In these instances, diversion by either a nephrostomy tube (a 10- or 12-Fr Malecot) and a stent (usually a 5-Fr feeding tube) or a double-J stent are employed.

FIG. 86-6. Ureterocalicostomy for correction of ureteropelvic obstruction complicated by a small extrarenal pelvis. (A) Incision in the proximal ureter, through the stenotic junction into a wider portion of the ureter. (B) The 3-0 chromic sutures compress the resected renal parenchyma. (C) The opened edges of the ureter sutured to the cut edges of the calyx with 4-0 or 7-0 absorbable sutures. (D) Completed closure. A nephrostomy tube and ureteral stent should be used.

Surgery for Tumors Wilms' tumor commonly presents in the horseshoe kidney. In general, the involved kidney and isthmus are resected in the course of removal of the tumor. If the tumor occurs in the isthmus, some authors have recommended bilateral lower pole heminephrectomy. If the Wilms' tumor is bilateral at presentation, it is managed the same as bilateral Wilms' tumors in orthotopically positioned kidneys. Tumor surgery in the isthmus of the horseshoe kidney deserves special mention because this will involve division of the isthmus. If the isthmus is composed of a band of fibrous tissue it can be readily divided using cautery; however, if it is functioning parenchyma it must be carefully addressed to avoid excessive blood loss and necrosis of remaining parenchyma with risk of secondary bleed and urinary fistula. The area must be carefully dissected and arteries to the isthmus sequentially occluded with bulldogs to assess the line of demarcation. Once this line is established the capsule is divided sharply and the parenchyma divided. Bleeding from the cut parenchyma is controlled with 4-0 chromic suture ligation. Any exposed calyces are closed with running locking 4-0 or 5-0 chromic, and the capsule and parenchyma are closed with carefully positioned horizontal mattress suture of 1-0 or 2-0 chromic ( Fig. 86-7).

FIG. 86-7. Division of the isthmus of a horseshoe kidney with a right-sided renal tumor. The isthmus blood supply is from the left iliac. (A) After identification of the line of demarcation an incision is made around the capsule of the isthmus. (B) The capsule is peeled back. (C) The parenchyma of the isthmus is transected in a wedge fashion to facilitate closure. (D) Horizontal mattress sutures of absorbable 2-0 material are used to close the parenchyma for hemostasis. (E) The capsule is closed over the retained parenchyma with a continuous absorbable suture.

Stone Surgery in the Horseshoe Kidney When calculus disease complicates obstruction of the UPJ in a horseshoe kidney the stone is removed at the time of pyeloplasty. In these instances, there may be considerably more reaction around the pelvis and ureter, so that the use of a nephrostomy tube and ureteral stent is prudent. An antegrade study can be performed 10 to 12 days postoperatively prior to nephrostomy tube removal to confirm drainage through the UPJ and integrity of the repair. Surgical Options for the Ectopic Kidney The ectopic kidney can be affected by any of the processes that occur in a normally positioned kidney. Overall, the evaluation and surgical management of these conditions will follow the lines of those discussed with horseshoe kidney. Reflux, if it mandates treatment, is generally dealt with by a standard Cohen ureteral reimplantation.

Occasionally, one has to address the problem of a failed pyeloplasty in a patient who has an ectopic pelvic kidney. Ureterocalicostomy is one alternative in management of this problem; another is the use of a pyelovesicostomy. Pyelovesicostomy has been performed in renal transplant recipients after ureteral loss due to ischemia and/or rejection, and has proven to be a viable salvage procedure.

OUTCOMES
Complications Problems common to UPJ repair regardless of kidney position, such as prolonged urine leakage and poor anastomotic drainage, occur more frequently in the horseshoe kidney. The risk of renal ischemia caused by damage of a aberrant vessel is increased in the horseshoe or ectopic kidney. Results Pyeloplasty in the horseshoe kidney is generally a successful procedure, although surgery on this anatomic variant does have a somewhat higher rate of complications than in normally positioned kidneys especially if division of the isthmus is employed. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Boatman DL, Kölln CP, Flocks RH. Congenital anomalies associated with horseshoe kidney. J Urol 1982;107:205. Buntley D. Malignancy associated with horseshoe kidney. Urol, VIII:146, 1976. Dretler SP, Olsson C, Pfister RC. The anatomic, radiologic and clinical characteristics of the pelvic kidney: an analysis of 86 cases. J Urol 1971;105:623. Dretler SP, Pfister R, Hendren WH. Extrarenal calyces in the ectopic kidney. J Urol 1970;103:406. Friedland GW, DeVries P. Renal ectopia and fusion: embryologic basis. Urology 1975;5:698. Gleason PE, Kelalis PP, Husmann DA, Kramer SA. Hydronephrosis in renal ectopia: incidence, etiology and significance. J Urol 1994;151:1660. Jones DJ, Wickham JEA, Kellett MJ. Percutaneous nephrolithotomy for calculi in horseshoe kidneys. J Urol 1991;145:481. Kölln CP, Boatman DL, Schmidt JD, Flocks RH. Horseshoe kidney: a review of 105 patients. J Urol 1972;107:203. Locke DR, Newman RC, Steinbock GS, Finlayson B. Extracorporeal shock-wave lithotripsy in horseshoe kidneys. Urology 1990;35:407. Pitts WR Jr, Muecke EC. Horseshoe kidneys: a 40-year experience. J Urol 1975;113:743. Segura JW, Kelalis PP, Burke EC. Horseshoe kidney in children. J Urol 1972;108:333.

Chapter 87 Transureteroureterostomy Glenn’s Urologic Surgery

Chapter 87 Transureteroureterostomy
Anthony J. Casale

A. J. Casale: Department of Urology, Indiana University School of Medicine, and Pediatric Urology Division, James Whitcomb Riley Hospital for Children, Indianapolis, Indiana 46202.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Transureteroureterostomy (TUU) is a renal salvage procedure necessitated by severe disease or injury of the lower half of the ureter. TUU was first performed clinically by Charles Higgins in 1934. Conditions that disable the distal ureter are rare and include trauma, malignancy, severe reflux, and distal ureteral obstruction. 6 As a rule, a ureteroureterostomy or ureteroneocystostomy are preferable procedures, but when the loss of ureter exceeds the ability of the surgeon to reanastomose the two ends of the ureter and the proximal ureter is too short to reach the bladder even with the aid of nephropexy and Boari bladder flap, then TUU is a useful option. TUU is occasionally utilized in complex reconstruction when the distal ureter is harvested to be used as a urethral replacement, for a catheterizable continent stoma, or for ureterocystoplasty. 3 Patients with advanced pelvic malignancies and a dilated ureter may be diverted utilizing a TUU and a cutaneous ureterostomy. 5

DIAGNOSIS
Transureteroureterostomy is used as a salvage reconstructive procedure in conditions that would require resection or loss of the distal ureter. Any unilateral pelvic malignancy or trauma may result in unilateral ureteral loss. Complex reconstruction for congenital urologic anomalies and neurogenic bladder conditions may also require harvesting and re-use of a distal ureteral segment.

INDICATIONS FOR SURGERY
Conditions that should be present to consider TUU include a salvageable kidney with a viable upper ureter of at least one-half its original length and a viable contralateral ureter with adequate drainage. The contralateral kidney is not necessary; in fact, the absence of the contralateral kidney facilitates the procedure and allows for an end-to-end anastomosis. A long segment of proximal ureter on the affected side will make the anastomosis easier and more successful by allowing a tension-free, straight anastomosis. Contraindications to TUU include impediments such as inadequate length of the donor ureter, a previous injury to either ureter from surgery or radiation, genitourinary tuberculosis, urothelial tumors, chronic pyelonephritis of either kidney, and a large size disparity between ureters. 7 A history of renal stones is considered to be a relative contraindication. In general, if one kidney is subject to a pathologic process that has spared the contralateral kidney and there is any possibility that the process can affect the uninvolved kidney via the shared ureter, then the procedure should not be performed.

ALTERNATIVE THERAPY
Nephrectomy, nephrostomy, ureterostomy or pyelostomy; ureteral replacement with bowel; Boari bladder flaps and nephropexy to allow ureteroneocystostomy or ureteroureterostomy; and autotransplantation all stand as alternatives to TUU. Our preferences in reconstruction include keeping the drainage of each kidney into the bladder separate and the use of only urothelium in the urinary tract. Obviously, these goals cannot always be realized.

SURGICAL TECHNIQUE
The surgical position for transureteroureterostomy is supine with the patient flat or with the back slightly flexed to elevate the retroperitoneum. The Trendelen-berg position will aid the surgeon by keeping the bowels out of the lower abdominal cavity during the procedure. If feasible, a preliminary cystoscopy may allow the urologist to place ureteral catheters in one or both ureters, which may aid in identifying the ureters and be used for postoperative stenting. A midline incision is best for this procedure, and because the ureters are closest to one another just above the bifurcation of the aorta the incision should afford the best visibility in this area. 4 The ureters must be easily visualized as they pass over the iliac vessels and into the pelvis in order to facilitate adequate mobilization of the donor ureter. Upon entering the peritoneal cavity, the bowel is packed into the upper abdomen or, if necessary, placed in a bowel bag. There are two options in opening the retroperitoneum. Two vertical incisions 5 cm long may be made over the distal ureters where they cross the iliac vessels and extending cephalad opening a window in the posterior peritoneum on each side ( Fig. 87-1). Alternately, a single curved incision may be used that opens the retroperitoneum from over the left distal ureter and extends across the midline along the small bowel mesentery and the cecum and up the right side along the line of Toldt (Fig. 87-2). This incision allows the bowels to be mobilized in an upward direction and exposes the retroperitoneum widely. On the left side the inferior mesenteric artery is a potential impediment to exposure of the distal ureter. If the blood supply of the left colon is otherwise normal, the inferior mesenteric artery can be ligated safely; however, this is seldom necessary.

FIG. 87-1. Transureteroureterostomy (TUU) using two posterior peritoneal incisions. (A) Plan for a left-to-right TUU. (B) The donor left ureter is spatulated and aligned with an incision made in the medial aspect of the recipient right ureter. (C, D) The anastomosis of the ureters may be performed with either interrupted or running (illustrated) absorbable sutures. (E, F) A catheter may be used during the anastomosis to separate the front and back wall of the anastomosis and removed at the last stitch. (G) Finished anastomosis visible through the right retroperitoneal window.

FIG. 87-2. Exposure of the retroperitoneal space using a circular self-retaining retractor and a curvilinear incision in the posterior peritoneum for mobilization of the bowel and mesentery.

It is advisable to expose and examine both ureters prior to beginning dissection. The recipient ureter should be mobilized only enough to allow an easy end-to-side anastomosis. Two 4-0 chromic stay sutures, one proximal and one dista to the anastomotic site, may be used to elevate and stabilize the ureter. Alternatively, small vessel loops may be carefully passed around the ureter at the same positions. Surgeons agree that dissection of the recipient ureter must be kept at a minimum to protect its blood supply. The donor ureter must be extensively mobilized to allow it to swing across the midline and lie in a gradually curved and tension-free course. It is important to maintain the ureter's associated blood supply and dissection should include mobilization of the gonadal vessels and periureteral adventitia along with the ureter. The dissection should begin just lateral to the gonadal artery and vein, and sweep all of the adventitia lying anterior to the psoas muscle with the ureter medially. The ureter should be divided and ligated with 3-0 chromic as distal as possible. The gonadal vessels should be divided and ligated with 3-0 silk near the length of the associated donor ureter. A long 4-0 chromic stay suture on the distal end of the divided donor ureter will allow it to be handled with minimal injury and facilitate its being passed through the retroperitoneal tunnel. Once the ureter has been mobilized the tunnel can be created through the retroperitoneal space. The tunnel should course beneath the posterior peritoneum, anterior to the great vessels, and superior to the inferior mesenteric artery to avoid trapping the ureter in the angle between the artery and the aorta. This is best done by dissecting medially from each side under direct vision. The midline portion of the tunnel may need to be created blindly with blunt dissection to connect the two sides. The surgeon's finger should pass through the tunnel to assure that it is wide enough for the ureter. A hemostat may then be used to grasp the stay suture on the end of the donor ureter and pull the ureter into position. Care must be taken to prevent twisting or kinking of the donor ureter. The donor ureter should easily reach the recipient ureter and lie in position against it without tension. The donor ureter should then be trimmed obliquely and may be spatulated if necessary to provide a wide anastomosis. The recipient ureter should be opened on its medial wall facing the donor ureter for a distance of at least 1.5 cm. The exact method of sewing the anastomosis may vary and either interrupted or running suture techniques are appropriate. We prefer to place a 5-0 chromic suture at each end of the anastomosis and to tie the two ureters in position ( Fig. 87-3). The sutures then can be sewn in a simple continuous running stitch to the opposite end and tied to the tail of the other end suture. The end sutures hold the ureters straight in position and allow some retraction during the anastomosis. We sew the back wall first, and once this step is done the ureteral catheter or stent can be placed under direct vision through the anastomosis.

FIG. 87-3. Ureteroureteral anastomosis may be performed with interrupted sutures. (A) After spatulating the donor ureter a suture is placed at the apex of each end of the anastomosis. (B) Once these sutures are tied the ureteral alignment should be straight. (C) Additional sutures are placed on each side of the anastomosis by dividing the distance between previous sutures evenly in half. (D) The finished anastomosis. (E) The TUU should not significantly alter the course of the recipient ureter.

Stenting of the transureteroureterostomy offers many options. We prefer to leave a stent across the newly created anastomosis and two ureteral catheters or double-J stents would be optimal. This is not often possible, however, because of the size of the recipient ureter especially at the ureterovesical junction. Stents may be placed in an antegrade fashion through the kidneys and the ends left in the distal ureter below the anastomosis and above the bladder. Finally, a single ureteral catheter or stent may be left in place that passes up the distal recipient ureter, passes through the anastomosis, and ends in the donor kidney. It is important that the stent have multiple side holes to facilitate drainage from both kidneys. All urologists agree that retroperitoneal drainage should be established. Dissection laterally from the ret-roperitoneal window over the anastomosis to the body wall allows placement of a drain to the skin through a separate stab incision. Some urologists prefer suction drains but we have used Penrose drains with success. The correct placement of the drains is more important than their type. The posterior peritoneum is then closed with a running 4-0 chromic so that any urine drainage cannot enter the peritoneal cavity. The peritoneal cavity is not drained. A Foley catheter is left in place unless bladder surgery was also performed and a suprapubic tube is utilized. If double-J stents are used the bladder is drained for 5 days and the stents left in place for 3 to 4 weeks. If ureteral catheters or antegrade stents are utilized, then the catheters or stents may be injected with contrast in x-ray to check for the patency and intactness of the anastomosis at 5 days and the tubes removed if the x-rays are acceptable. The bladder catheter may be removed 1 day after the ureteral catheters or antegrade stents. The retroperitoneal drains may be removed 5 days later. Patients often go home and return to have catheters and drains removed in clinic. Patients remain on prophylactic antibiotics until all tubes are removed and the urine is sterile.

OUTCOMES
Complications Ureteral obstruction due to stricture may occur in 1% to 3% of patients and is the most serious possible complication. Therefore, it is imperative to carefully follow the patient with a transureteroureterostomy with imaging studies on a frequent basis. We prefer an IVP or Lasix renogram 2 weeks after stent removal. These studies should be repeated at 6 and 12 months after surgery, and then every 12 months. In addition, interval ultrasounds may be used in between these functional studies. Strictures may present either silently with gradually increasing hydronephrosis or with symptoms such as flank pain or infection. While the donor ureter is most at risk

because of mobilization, the recipient ureter may also be injured. Ureterocutaneous fistula might also occur and will become obvious when urine continues to drain from the Penrose site. Like strictures, fistulas may result from poor healing due to diminished vascular supply to the ureters or from inflammation, tumor, or distal obstruction. The initial management of both strictures and fistulas is long-term stenting and may include nephrostomy drainage in some cases. Finally, the most serious result of transureteroureterostomy occurs when both kidneys become obstructed due to a process at or below the anastomosis. The most common problem is an impacted stone in the distal ureter, but tumors or structure can also be to blame. In this instance the patient may become anuric and a true urologic emergency is present. Often use of percutaneous nephrostomy tubes is the best initial management to establish adequate drainage of urine until the primary obstruction can be evaluated and definitively treated. Results Transureteroureterostomy has been a highly useful technique with success rates of over 90%. In order to achieve these high levels of success, however, it is important to follow well-documented principles, such as (a) adequate mobilization of the donor and minimal mobilization of the recipient ureter, (b) spatulation of the donor ureter and anastomosis to the medial aspect of the recipient ureter, and (c) adequate drainage and prevention of urinary extravasation. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. Ehrlich RM, Skinner DG. Complications of transureteroureterostomy. J Urol 1975;113:467–473. Higgins CC. Transuretero-ureteral anastomosis. Report of a clinical case. J Urol 1935;34:349. Mitrofanoff P. Cystostomie Continent Trans-appendiculaire dans le Traitement des Vessaies Neurologiques. Chir Pediatrics 1980;21:297. Netto NR Jr. Transureteroureterostomy. In: Glenn JF, ed. Urologic surgery. 4th ed. Philadelphia: JB Lippincott, 1993;306–310. Rainwater LM, Leary FJ, Rife CC. Transureteroureterostomy with a cutaneous ureterostomy: a 25-year experience. J Urol 1991;146:13. Rushton HG, Parrott TS, Woodard JR. The expanded role of transureteroureterostomy in pediatric urology. J Urol 1987;138:357. Sandoz IL, Paull DP, Macfarlane CA. Complications with transureteroureterostomy. J Urol 1977;117:39.

Chapter 88 Pyeloplasty Glenn’s Urologic Surgery

Chapter 88 Pyeloplasty
Eugene Minevich and Jeffrey Wacksman

E. Minevich and J. Wacksman: Department of Surgery, University of Cincinnati College of Medicine, and Division of Pediatric Urology, Children's Hospital Medical Center, Cincinnati, Ohio 45229-3039.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique: Dismembered Pyeloplasty Outcomes Complications Results Chapter References

The ureteropelvic junction (UPJ) is the most common site of obstruction in the upper urinary tract. The pelvis is usually a single hollow structure forming a cone-like transition to the upper ureter, which is fortunate for the performance of most surgical procedures. Urine is collected in the pelvis and transmitted through the UPJ to the upper ureter, where it travels via a peristaltic wave down the ureter to the bladder. Histologically, the renal pelvis is composed of three main layers: an inner transitional cell mucosal lining, a middle smooth muscle layer oriented in circular fashion (with a few internal and external bundles), and an outer fibrous coat. The renal vessels are usually situated superior and anterior to the UPJ. Contained within the outer layer of the renal pelvis is a rich arterial and venous anastomosis that provides an abundant blood supply to the pelvis. The blood supply of the upper ureter is contained within the surrounding adventitial layer of the ureter and outer ureteral wall. Therefore, skeletonizing of the ureter during dissection may cause devascularization of this segment. Lesions causing obstruction to the UPJ are divided into those resulting from intrinsic or extrinsic causes. Of the intrinsic factors (calculus, ureteral valve, fibroepithelial polyp, primary carcinomas), the adynamic segment is associated most often with classical UPJ obstruction. This obstruction is thought to be a congenital absence or abnormal arrangement of the muscular fiber at the transition zone of the upper ureter and renal pelvis. Extrinsic causes include fibrous bands, aberrant accessory vessels, and various organ-compressing factors (e.g., retroperitoneal fibrosis, carcinomatous nodal disease, inflammatory bowel disease).

DIAGNOSIS
Prior to the advent of widespread ultrasonic screening during pregnancy, most children presented with abdominal or flank pain, hematuria, urinary tract infection (UTI), or gastrointestinal symptoms. Currently, most cases of UPJ obstruction are diagnosed in utero, although such patients are initially asymptomatic. Ultrasound is the test of choice to determine the degree of renal pelvic dilation, parenchymal thickness, and associated abnormalities of the bladder and ureter. The voiding cystourethrogram (VCUG) and occasional excretory urogram (IVP) provide necessary anatomic details. We use the diuretic renogram (MAG3) and occasional Whitaker antegrade pressure perfusion studies to define UPJ obstruction. 6 Retrograde pyelography is used occasionally to rule out other causes of obstruction and is usually performed just before surgery. 5

INDICATIONS FOR SURGERY
In our institution, indications for surgery comprise the following: 1. Symptomatic UPJ obstruction with intermittent flank pain 2. Hematuria or pyelonephritis. 3. Renogram Lasix clearance T ½ (the time taken to clear half of the accumulated radioactive agent from the renal pelvis following Lasix administration) in excess of 20 minutes. 4. Renal pelvis pressure exceeding 20 cm H 20. 5. Differential renal function of the obstructed kidney less than 40%

ALTERNATIVE THERAPY
Alternative treatments, including endopyelotomy (percutaneous or retrograde) and laparoscopic pyeloplasty, are beyond the scope of this chapter but are discussed in Chapter 116 and Chapter 128 of this book.

SURGICAL TECHNIQUE: DISMEMBERED PYELOPLASTY
Although many repair techniques are available, we prefer a dismembered Anderson-Hynes pyeloplasty ( Fig. 88-1) for obstructed UPJ repair because it allows the removal of an adynamic segment and reduces the renal pelvis, thereby creating a more efficient pelvis. 1 Optical magnification (×3.5) is beneficial in ensuring precise suture placement and a watertight anastomosis. We have successfully used a standard subcostal extraperitoneal flank approach for several decades.

FIG. 88-1. Dismembered pyeloplasty after subtotal resection of an excessively large renal pelvis. (A) The line of excision provides an adequate margin of pelvis to enable closure without tension. (a) A vertical incision is made in the lateral aspect of the ureter to spatulate it. (B) The apical sutures are carefully placed. (C) Approximation and closure of the pelvis are completed.

The procedure starts with careful but limited mobilization of the upper ureter, being certain to keep all periureteral tissue attached to the ureter. In most cases this can be done without placing any type of tape around the ureter, which could damage the blood supply coming up through its adventitia. Next, a 4-0 or 5-0 chromic traction suture on a small taper needle is placed at the UPJ. A second suture is usually placed above this and the upper ureter transected. The transected ureter with its traction suture is carefully mobilized for a short distance. With use of the traction suture through the UPJ, the pelvis is completely mobilized and prepared for reduction. Next, the renal pelvis is incised (usually below the UPJ) and excess pelvis is trimmed away. At this point, the surgeon should be careful not to cut across

any extrarenal calyces. A 5-0 chromic traction suture is used to mark the most dependent position of the UPJ. We prefer to close the trimmed upper pelvis next, which usually is accomplished with a 4-0 or 5-0 chromic running, interlocking suture placed halfway down the renal pelvis, in a watertight fashion. The ureter-to-pelvis anastomosis is performed next. The suture on the upper ureter is placed on traction and the upper ureter is spatulated for 2 to 3 cm on either the posterior or the lateral aspect of the ureter. Care is taken not to twist or spiral this incision. Next, the spatulated end of the ureter is anastomosed to the dependent portion of the renal pelvis with interrupted 6-0 monofilament polyglyconate (Maxon) sutures, or 6-0 or 7-0 Vicryl sutures. These are usually through-and-through inverting sutures, with knots secured on the outside of the anastomosis. After placement of 4 or 5 sutures, the KISS catheter (Kidney Internal Split-Stent, Cook Urological) is placed. 4 The remainder of both sides of the anastomosis is closed, with either a running suture of 6-0 Maxon or 6-0 or 7-0 Vicryl. An external drain (Penrose or Jackson-Pratt) should be placed after the anastomosis is completed. The type of suture material used is not critical, but the type of surgical dissection and mobilization are most important. If the ureter is compressed against a distended pelvis by crossing polar vessels, we prefer to divide the UPJ and bring the pelvis and ureter anterior to the crossing vessel (Fig. 88-2). Having accomplished this, the rest of the procedure is similar to the dismembered pyeloplasty.

FIG. 88-2. Dismembered pyeloplasty with accessory polar vessels. (A) Vessels compressing the ureteropelvic junction. (B) Ureter divided from renal pelvis. (C) Reduction pyeloplasty with ureter brought anterior to polar vessels for anastomosis.

The most common site of UPJ obstruction in a duplicated kidney is usually the lower pole segment ( Fig. 88-3). Since the upper pole is separate and not obstructed, we usually perform a simple dismembered pyeloplasty to the lower pole. On occasion, we have anastomosed the lower pole renal pelvis to the upper ureter, especially in long-segment disease. For obstruction of the upper pole segment alone, either a routine pyeloplasty to the upper pole or, in this case, a pyeloplasty to the lower pole pelvis can be performed ( Fig. 88-4).

FIG. 88-3. Ureteropelvic junction obstruction of the lower pole ureter in a duplicated system. (A) Ureteropelvic stenosis of lower pole. (B) Dismembered pyeloplasty with resection of stenotic segment. (C) Reduction pyeloplasty and anastomosis of lower pole pelvis to upper ureter.

FIG. 88-4. Pyelopyelostomy. (A) Adjacent surfaces of the unobstructed normal lower segment of pelvis and the obstructed (abnormal) upper segment of pelvis. (B) A window is created between the two pelves, and the posterior margins are joined. (C) Anastomosis is completed. An optional resection of a portion of the upper segment of the ureter is illustrated.

The anatomies of the horseshoe kidney and pelvic or ectopic kidney with a UPJ obstruction are similar. In both, the renal pelvis is usually anterior, with several crossing vessels. In patients with horseshoe kidneys, we usually recommend a transabdominal approach. Although not absolutely necessary, we prefer to divide the renal isthmus to try to establish a more inferior and dependent position to the new UPJ. After this, a standard dismembered pyeloplasty is preferred over a flap procedure.

OUTCOMES
Complications Urinary leakage may occur during the first few postoperative days. Usually a minor amount of leakage is not problematic if the area is drained adequately. Therefore the drain should only be removed after one has made sure that the anastomosis is intact. If intraoperative drainage of the renal pelvis was not performed, leakage can be handled by percutaneously placed nephrostomy or by ureteral stent. With the use of meticulous technique and a KISS catheter, immediate postoperative excess urinary leakage is virtually non-existent. Obstruction at the UPJ is usually secondary to traumatic dissection, devascularization of the ureter, excessive traction on the ureter resulting in ischemia or creation of an anastomosis that is too tight, or extravasation from an undrained anastomotic leak with subsequent fibrosis. This can be managed initially by intubation or balloon dilation of the strictured area and formal reoperation (open or endoscopic) if the initial step fails. Postoperative bleeding (usually from the nephrostomy tract) can jeopardize the repair by the formation of obstructive clots. Other postoperative problems include

acute pyelonephritis, wound infection, or incisional hernia. Results Experimental and clinical data clearly support our view that obvious obstruction at the UPJ should be surgically corrected expeditiously, even in the neonatal period. Dismembered pyeloplasty with careful attention to the details outlined is successful in more than 95% of cases. 2 Surgical complications are rare and can be managed conservatively in most cases. In our 15-year experience of more than 200 pyeloplasties with nephrostomy tube or KISS catheter drainage, no patient required reoperation for persistent obstruction, increased hydronephrosis, or decreased split renal function. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Anderson JC, Hynes W. Retrocaval ureter: case diagnosed preoperatively and treated successfully with plastic operator. Br J Urol 1949;21:209. Bernstein GT, Mandell J, Lebowitz RL, Bauer SB, Colodny AH, Retik AB. Ureteropelvic junction abstraction in the neonate. J Urol 1988;140:1216. Conway J. Well tempered diuresis renography. Semin Nucl Med 1992;22:74. Ritchie E, Reisman EM, Zaontz MR, Hatch DA, Wacksman J, Maizels M. Use of kidney internal split/stent (KISS) catheter in urinary diversion after pyeloplasty. Urology 1993;42(l):55. Rushton HG, Salem Y, Belman AB, Maid M. Pediatric pyeloplasty: is routine retrograde pyelography necessary? J Urol 1994;152(2 part2):604. Whitaker RH. The Whitaker test. Urol Clin North Am 1979;6:529.

Chapter 89 Megaureter Glenn’s Urologic Surgery

Chapter 89 Megaureter
Edmond T. Gonzales, Jr.

E. T. Gonzales, Jr.: Scott Department of Urology, Baylor College of Medicine, Houston, Texas 77030-2399.

Diagnosis Indications for Surgery Surgical Technique Outcomes Complications Chapter References

A megaureter is, simply stated, a ureter that is wider than normal. By convention, this is usually a ureter greater than 7 mm in diameter. 3 However, the extent of dilation varies considerably and in some cases the ureter may be as wide as 3 cm or more in cross-section diameter. In nearly all cases, the dilation is present along the entire length of the ureter. In the past, most patients with megaureter presented because of urinary tract infection, pain, palpable abdominal or flank mass, or were found incidentally when an intravenous urogram or renal ultrasound was obtained for unassociated symptomatology. Today most children with megaureter are found on prenatal ultrasound, before associated symptoms or infections develop. There are multiple causes for megaureter and the abnormalities are considerably more varied than the pathology associated with ureteropelvic junction (UPJ) obstruction. Megaureter may be acquired or congenital. Acquired abnormalities might include a postsurgical stricture, ureteral calculus, or ureteral occlusion associated with an inflammatory process or a tumor. In the pediatric age group, acquired causes for megaureter are distinctly uncommon. Almost all children with megaureter have some form of congenital anomaly associated with the development of the ureter, the ureterovesical junction, or abnormally high intravesical pressures. Megaureter is generally classified as (a) obstructed megaureter, (b) nonobstructed, nonrefluxing megaureter, or (c) refluxing megaureter. Each of these categories is subclassified into primary and secondary disorders. Primary causes for megaureter involve anomalies of the ureterovesical junction or the ureter itself. Secondary causes would result from abnormally high intravesical pressures or abnormally high urinary flow rates (e.g., diabetes insipidus). This classification emphasizes the observation that all widened ureters are not associated with obstruction at the ureterovesical junction. It behooves the surgeon contemplating an operative repair to evaluate all aspects of urinary pathophysiology that might be associated with a megaureter. For instance, high intravesical pressures associated with posterior urethral valves, neurogenic dysfunction, or dysfunctional voiding can result in substantial dilation of the ureter, which would resolve if the bladder pathology were corrected. In most cases, if the primary pathology is associated with high intravesical pressures, one expects bilateral ureteral changes as well as some degree of detrusor thickening. However, the degree of ureteral dilation is not always symmetric. A voiding cystourethrogram is, therefore, essential as part of the initial screening evaluation of any child with a megaureter. In some children, vesical urodynamics will also be necessary to fully elucidate the possibility of bladder pathology. In other cases, massive vesicoureteral reflux will be found at the time of voiding cystourethrography. If there is no obvious bladder pathology and no findings to suggest high intravesical pressures, this confirms the diagnosis of a primary refluxing megaureter. In this situation, it is likely that the extreme dilation of the ureter represents some form of primary ureteral muscle abnormality in association with the reflux. Ureteral dilation can be associated with ureteral abnormalities without reflux or obstruction. Examples of primary ureteral dysfunction are most obvious and exaggerated in the prune belly syndrome. Isolated examples of this form of ureteral dilation, though, are also thought to exist and have been described in the past as a forme fruste of the prune belly syndrome. Primary obstructive megaureter is about one-sixth as common as primary UPJ obstruction. Truly obstructive megaureter will ultimately require surgical repair and is the primary topic of this discussion. In a primary obstructed megaureter, the most distal portion of ureter for 1 to 2 cm just outside of the bladder is usually of normal or slightly smaller caliber. This is the obstructing portion. The obstruction appears to result from faulty peristalsis of this segment, and histologic studies have demonstrated increased circular muscle and fibrosis in that segment. 7 In the refluxing megaureter, the ureter is generally dilated all the way to the detrusor and the typical changes seen in the obstructed megaureter are not present. There is also a histologic difference between the obstructed and refluxing megaureter. Whereas the ureteral wall in obstructed megaureter may be thickened and hypertrophied, the smooth muscle-to-collagen ratio is similar to normal control ureters. In the refluxing megaureter there is an increased percentage of collagen to smooth muscle in the ureter, suggesting that there is an inherent developmental abnormality in the refluxing megaureter. 5

DIAGNOSIS
The confirmation of significant ureterovesical junction obstruction as a cause for megaureter is not always straightforward. Evaluation and assessment of the child with hydroureteronephrosis must consider the extent of renal function in the involved kidney, the degree of dilation, level of bladder function, and presenting symptoms. Diagnostic evaluation might include renal ultrasonography, intravenous urography, radioisotope renal scan, percutaneous nephrostomy for temporary diversion that would allow evaluation of specific renal function and for antegrade pressure flow studies (Whitaker test), 13 as well as a thorough bladder evaluation (cystography, urodynamics) as discussed above. In most situations, the initial evaluation will consist of a renal ultrasound, voiding cystourethrogram, and diureticenhanced (Lasix) renal scan. The renal scan will provide an estimate of comparative renal function and washout times.

INDICATIONS FOR SURGERY
Indications for surgery would include evidence of deterioration (such as increasing ureteral dilation on subsequent ultrasounds or decreasing renal function on follow-up renal scans) or development of urinary infection or pain consistent with the obstruction. If the hydroureter was identified incidentally (e.g., prenatally or during a bone scan for evaluation of joint pain), and renal function is normal or near-normal on the side of the abnormality, an argument could be convincingly made to follow this child. This presentation differs significantly from the child who presents with pain and/or urinary infection. Perhaps the most controversial aspect of managing a child with megaureter and ureterovesical obstruction is the current increased recognition of the disorder in the fetus. Initially, this observation was met with great surgical enthusiasm on the expectation that progressive deterioration and loss of renal function will undoubtedly result. However, increasing experience has shown that a considerable number of newborns with hydroureter will maintain stable renal function and show substantial, gradual improvement in the degree of dilation, ultimately regaining near-normal renal and ureteral anatomy. 2,6,9 In one study reported by Baskin and associates, 25 neonates with primary megaureter (without reflux) and good ipsilateral renal function were followed for an average of more than 8 years. 1 In half of the patients, the hydronephrosis improved spontaneously, whereas in another 20% it remained stable. None of these children demonstrated loss of renal function as measured by radionuclide renal scanning.

SURGICAL TECHNIQUE

Surgical techniques for correction of obstructive anomalies at the ureterovesical junction involve a variation of ureteral reimplantation. If ureteral dilation is only mild to moderate, a simple ureteral reimplantation may be all that is necessary. The segment of abnormal ureter is usually short and must be excised in its entirety. The procedure can be done either intravesically or extravesically by any standard ureteral reimplantation technique as described elsewhere in this text and can be chosen at the surgeon's preference. More often, however, some degree of remodeling of the lower ureter is indicated during correction of a megaureter. The lower ureter is approached through a Pfannenstiel incision. The procedure can be performed intravesically or extravesically. If the ureteral diameter is substantial, my preference is to expose the ureter extravesically before opening the bladder. The principle of remodeling the lower ureter for correction of a megaureter anomaly is to reduce the circumference sufficiently to allow a more near-normal-caliber ureter to be implanted. The goal is to achieve a homogeneous caliber of lower ureter for the entire segment of submucosal tunnel as well as for a short distance outside of the detrusor. It is not necessary to extensively reduce the caliber of the ureter more proximally. Since these are obstructed ureters, once the obstruction is satisfactorily relieved the proximal dilation will resolve substantially. It is especially important to respect the blood supply when mobilizing the ureter. The blood supply to the lower portion of the ureter will be coming into the ureter laterally from the superior vesical artery as well as small branches that may still be coming from the umbilical vessels. These vessels will usually have to be sacrificed. The first significant medial branch is encountered coming off of the common iliac or internal iliac (hypogastric) artery, and every effort must be made to identify and preserve this vessel (Fig. 89-1). During dissection, a visible layer of adventitia is left along the ureter to minimize damage to the delicate blood supply that encompasses the ureter. Special care is taken to identify and preserve any longitudinal blood supply along the ureter.

FIG. 89-1. Diagrams of the major arterial supply to the ureter. It is not necessary to sacrifice any medially based blood supply for primary megaureter repair.

In most cases, the discarded segment of ureter will be a lateral wedge. However, the basic principle is to respect obvious intrinsic blood supply, and at times the resection may follow a more circuitous path if a clearly defined longitudinal vessel is evident. At this point, a choice is made regarding whether a formal surgical excision and reduction ureteroplasty is to be performed or whether a plication procedure would be satisfactory. This decision is generally based on the caliber and thickness of the ureter as well as the preference of the surgeon to some extent. Plicating a very large, thick-walled ureter leaves a considerable bulk of defunctionalized ureter that can be difficult to bring through a submucosal tunnel and results in considerable edema that may require a lengthy period of ureteral diversion in some instances. From a practical point of view, postoperative management is similar for a plication versus a reduction ureteroplasty in my hands, and I do not feel that there is a substantial difference between the two procedures. Formal excision and reduction ureteroplasty will be described first, followed by techniques for plication of the ureter. After adequate dissection and mobilization of the ureter in the perivesical space, a 12-Fr catheter is passed up to the renal pelvis (in a newborn or very young infant, the catheter would be 10-Fr). Occlusive clamps are placed loosely around the catheter, visually preserving longitudinal blood supply along the course of the segment to be preserved. As the clamps are placed proximally, the length of segment needed for passage into the bladder and through the submucosal tunnel is kept at a homogeneous caliber. As soon as this length is felt to be sufficient, more proximal clamps are gradually moved farther and farther away from the catheter so that a gentle tapering of this portion of the ureter results. The clamps are placed so that the discarded segment is what is secured by the clamps. Every effort is made to avoid damage to the small vessels of the preserved ureteral strip ( Fig. 89-2). The ureter is closed in two layers. The musculomucosal layer is closed with a running, locking 7-0 chromic catgut suture. The thin adventitia that was preserved at the time of ureteral mobilization is then closed as a second layer with running, nonlocking, 7-0 Vicryl or 7-0 PDS suture. Near the distal portion of the ureteral strip, for a length of about 1.5 cm, interrupted sutures are placed. This is because a segment of ureter may have to be excised at the time of reimplantation and this prevents cutting back into the running suture line. At this point, care is taken to inspect the blood supply and extent of bleeding at the tip of the ureter.

FIG. 89-2. (A, B) The wedge of ureter to be excised is secured with Allis or Hendren clamps. (C) The outlined segment of ureteral wall is excised sharply. (D, E) The ureter is closed in two layers. Distally, interrupted sutures are placed to allow for trimming of the end of the ureter at the time of reimplantation.

The bladder is then opened anteriorly. A suitable location on the posterolateral side of the bladder is chosen for creation of a new ureteral hiatus, usually just cephalad to the original ureteral meatus. The ureter is brought into the bladder. A cross-trigonal tunnel is then fashioned, and the tapered segment of the ureter is brought into the tunnel and secured distally in the usual fashion. The mucosa overlying the location of the neohiatus is closed with submucosal sutures to evert the mucosal edges. This is done to reduce the likelihood that a ureteral bladder fistula could result that would subsequently eliminate the length of tunnel that was created. A 7-Fr silastic single-J catheter is placed in the ureter and secured just at the meatus with a single 4-0 chromic catgut suture. The catheter is then brought out through the bladder and through a separate puncture wound in the skin for external diversion and drainage. Vesical diversion is accomplished by a small urethral catheter in girls or a small suprapubic tube in boys. The wound is closed in layers in the routine fashion, and the perivesical space is drained with a small Penrose drain. The ureteral stent is generally left in place for 10 days. There are two techniques for ureteral plication: the Starr technique and the Kalicinski technique. The Starr technique involves imbrication and plication of the ureter along the course to be reimplanted as demonstrated in Figure 89-3.12 The Kalicinski technique differs by isolating the ureteral lumen from a defunctionalized portion by a running horizontal mattress suture and then folding this defunctionalized segment around the ureter to maintain a more homogeneous ureteral caliber ( Fig. 89-4).4 Respect for the ureteral blood supply is identical to what was emphasized for formal excision ureteroplasty. In both cases, reimplantation is accomplished in a similar fashion and a ureteral stent is similarly left in place. If the amount of tissue folded is not extensive, then the catheter is generally taken out after several days.

FIG. 89-3. (A) Starr plication of the ureter suture to infold the ureter. (B) Cross-section to show placement of Lembert-type suture. (C) Cross-section after ligation of suture.

FIG. 89-4. Kalicinski technique of ureteral imbrication. (A) Placement of cobbler's stitch to exclude a major portion of the ureteral lumen. (B) Same in cross-section. (C) After ligation. (D) Excluded portion of ureter is folded over and wrapped around the intubated ureter. (E) Final appearance in cross-section.

A totally extravesical approach for the management of a megaureter has been described, but it is not as popular as the transvesical approach.

8

OUTCOMES
Complications Reduction ureteroplasty is a very safe, reproducible, and successful procedure. 10 The major complications are those associated with any ureteral reimplantation: the development of ureteral obstruction (about 5% of cases) or vesicoureteral reflux (about 10% of cases). As suggested previously, one abnormality that can result in postoperative reflux is the development of a fistula along the tunnel fashioned for the reimplantation. Although a two-layer ureteral closure offers some additional assurance to avoid this complication, care should be taken to handle the mucosa during the development of the submucosal tunnel carefully and to keep it full thickness throughout the course of dissection. In most cases, obstruction is probably a result of ischemia of the preserved ureteral strip. Initial efforts at managing this complication would be by percutaneous dilation of the stricture, but many such instances will ultimately require formal open surgical revision. Many clinicians have experienced more difficulties when reconstructing a refluxing megaureter than a primary obstructed megaureter. 11 The specific reasons for the observation have not been clearly elucidated, although it is generally accepted that refluxing ureters are inherently more abnormal than primary obstructed systems and therefore are less likely to recover fully. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Baskin LS, Zderic SA, Snyder HM, Duckett JW. Primary dilated megaureter: long term follow-up. J Urol 1994;152:918–921. Cozzie F, Madonna L, Maggi E, et al. Management of primary megaureter in infancy. J Pediatric Surgery 1993; 28:1031–1033. Cussen LJ. Dimensions of the normal ureter in infancy and childhood. Invest Urol 1997;5:194–199. Kalicinski ZH, Kansy J, Kotarbinska B, Joszt W. Surgery of megaureters: modification of Hendren's operation. J Pediatr Surg 1977; 12:183–188. Lee BR, Partin AW, Epstein JI, Quinlan DM, Gosling JA, Gearhart JP. A quantitative histological analysis of the dilated ureter of childhood. J Urol 1992; 148:1482–1489. Liu HY, Dhillon HK, Young CK, Diamond DA, Duffy PG, Ransley PG. Clinical outcome and management of prenatally diagnosed primary megaureter. J Urol 1994;152:914–917. McLaughlin AP III, Pfister RC, Leadbetter WF, Salzstein SL, Kessler WO. The pathophysiology of primary megaloureter. J Urol 1973; 109:805–811. McLorie GA, Jayonthi VR, Kinaham TJ, Khoury AE, Churchill BM. A modified extravesical technique for megaureter repair. Br J Urol 1994;74:715–719. Mollard P, Foray P, Degoday JL, Valignat C. Management of primary obstructive megaureter without reflux in neonates. Eur Urol 1993;24:505–510. Peters CA, Mandell J, Lebowitz RL, et al. Congenital obstructed megaureters in early infancy: diagnosis and treatment. J Urol 1989;142:641–645. Rabinowitz R, Barkin M, Schillinger JF, Jeffs RD, Cook GT. The influence of etiology on the surgical management and prognosis of the massively dilated ureter in children. J Urol 1978;119:808–813. 12. Starr A. Ureteral plication: a new concept in ureteral tailoring for megaureter. Invest Urol 1979;17:153–157. 13. Whitaker RH. Methods of assessing obstruction in dilated ureters. Br J Urol 1973;45:15–22.

Chapter 90 Triad Syndrome Glenn’s Urologic Surgery

Chapter 90 Triad Syndrome
David B. Joseph

D. B. Joseph: Division of Urology, Children's Hospital of Alabama, and Department of Surgery, The University of Alabama, Birmingham, Alabama 35233.

Diagnosis Presentation Evaluation Indications for Surgery Urinary Tract Reconstruction Urinary Diversion Definitive Urinary Reconstruction Reduction Cystoplasty Abdominal Wall Reconstruction Orchiopexy Outcomes Complications Results Chapter References

Triad syndrome—the clinical association of a thin flaccid abdominal wall, undescended testes, and bladder hypertrophy with hydroureters—was originally described in 1895 by Parker. 9 Shortly thereafter, Osler presented a similar constellation of findings in a child he described as having the appearance of a wrinkled prune. 8 From that point, prune belly has unfortunately become synonymous with this syndrome. This clinical manifestation is also known as the Eagle-Barrett syndrome and the abdominal muscular deficiency syndrome. By classic description, the triad syndrome occurs in boys. However, 5% are girls presenting with similar physical findings with the obvious exception of the gonadal abnormality. The incidence of triad syndrome occurs in 1 of every 30,000 to 50,000 live births. Most cases are sporadic, although a familial occurrence has been described. There is no single theory that incorporates all aspects of the triad syndrome. Early theories were founded on a primary mesodermal defect or a primary obstructive process.3 The mesodermal defect theory is based on an abnormality occurring in embryonic mesoderm during the third week of gestation that results in an abnormally developed abdominal wall and urinary system. The obstructive theory is based on a lesion occurring in the region of the membranous urethra. This causes back pressure, which leads to dilation of the bladder, urinary ascites, and abdominal distention. The result is degeneration of abdominal wall musculature. The increased abdominal pressure is transmitted to the upper urinary tract resulting in hydroureteronephrosis and renal dysplasia. The presence of intra-abdominal testes with either of these theories has not been supported by experimental investigation. Approximately three-quarters of the children with classic triad syndrome will have associated anomalies. Most often there is a thoracic deformity resulting in protrusion of the upper sternum, depressed lower sternum, and splayed ribs. Other skeletal deformities include talipes equinovarus, congenital hip dislocation, calcaneus valgus, polydactyly, syndactyly, arthrogryposis, scoliosis, and lordosis. Midgut malformations, most often due to defective fixation or malrotation of the midgut, have been noted in approximately one-third of children. Cardiac problems (atrial and ventricular septal defects) have been reported in approximately 15% of children.

DIAGNOSIS
Presentation Fetal sonography has had a major impact on early identification of genital urinary abnormalities. With sophisticated equipment and operator expertise, the identification of a child with the triad syndrome can often be established in utero. A similar constellation of findings can be seen in a fetus with posterior urethral valves or the megacystis-megaureters syndrome. Close inspection of the abdominal wall musculature will often hedge the differential diagnosis to that of the triad syndrome. In utero diagnosis allows for a planned neonatal investigation spear-headed by the pediatric urologist. At birth, the diagnosis of triad syndrome is often obvious based on the pathognomonic physical findings of a loose, lax, wrinkled abdominal wall, flared chest, and undescended testes. Several classifications of the triad syndrome have been established based on severity and clinical presentation. There is no single classification system that incorporates the total spectrum of this syndrome. For practical purposes, children can be grouped into severe, moderate, or mild presentations. With a severe presentation, survival is often limited by significant respiratory compromise due to pulmonary immaturity and dysplasia, as well as extensive renal dysplasia, resulting in a Potter-like syndrome. Children described with moderate involvement have combined renal and respiratory insufficiency mandating close observation and early intervention to minimize the sequelae of pulmonary and renal compromise. Monitoring of the urinary system is necessary in order to prevent progressive renal deterioration due to stagnation of urinary flow, urinary tract infections, and possible urinary tract obstruction. Children with mild involvement do not suffer from respiratory or renal compromise. While long-term follow-up is necessary, operative intervention is often limited to orchiopexy and abdominal wall reconstruction. Evaluation A team approach consisting of a pediatric urologist, neonatologist, nephrologist, pulmonologist, and cardiologist is required in order to maximize the outcome. The initial cardiorespiratory status of the neonate must be established. The baby should undergo a chest x-ray and, when indicated, cardiac sonography. Treatment may be required to maintain adequate pulmonary toilet for oxygenation and prevention of upper respiratory tract infections. Urologic evaluation commences with abdominal sonography. Both the upper and lower urinary tract should be assessed. Attention should be placed on the degree of hydronephrosis, the volume of renal parenchyma, and its echogenicity. Oftentimes there will be a disproportionate degree of lower ureteral and urinary tract dilation when compared to the proximal ureter and kidney. On occasion a marked transition of ureteral dilation is noted. If the infant is clinically stable and voiding per urethra or draining through a patent urachus, further diagnostic testing can be placed on hold. Children with renal insufficiency must undergo further testing, which may include a voiding cystourethrogram, to determine whether the insufficiency is related to renal dysplasia, stagnant urine flow or true obstruction. It is of utmost importance that any invasive lower urinary tract imaging be performed in a sterile environment with the child receiving pre- and postprocedural antibiotics. The abnormally dilated urinary system in a child with triad syndrome results in stagnant urine flow that is very susceptible to bacteriuria and often difficult to clear. When functional renal information is needed, renal scintigraphy provides greater objective information than conventional intravenous urography.

INDICATIONS FOR SURGERY
There is no other pediatric urologic pathologic process that requires as much individualized patient care as the triad syndrome. Each child presents with a unique constellation of problems resulting in its own set of considerations. Therefore, no one treatment plan is appropriate for all children. In general, operative management can be divided into three broad areas: 1. Reconstruction of the urinary system 2. Reconstruction of the abdominal wall 3. Transfer of the intraabdominal testes to the scrotum

These three areas of operative management are not exclusive of each other, but for simplicity's sake each will be described independently. H4>Urinary Tract Reconstruction Controversy surrounds the need for aggressive urinary tract reconstruction. Early aggressive operative intervention in all cases is countered by the fact that renal dysplasia may be inherent, thus preventing any intervention from improving the functional status. In addition, imaging studies depicting significant hydroureteronephrosis do not correlate with obstruction. Hydroureteronephrosis does not by itself mandate reconstruction. Urinary tract reconstruction is beneficial in a child who has a component of obstructive uropathy and has been shown to have improved renal function with decompression of the urinary system. Reconstruction is also of benefit in the child who has progressive hydroureteronephrosis associated with increasing renal compromise and in the child who has urinary tract infections due to stagnant urine flow. Urinary Diversion Urinary diversion can play a temporary initial role in the management of acute renal failure or sepsis. Oftentimes children with urethral atresia or obstruction will present with a patent urachus, effectively emptying their lower tract. Infants with associated urethral abnormalities resulting in obstruction or poor bladder decompression, who are not candidates for intermittent catheterization, benefit from a vesicostomy. A vesicostomy, however, may not adequately drain the upper urinary tract due to a relative obstruction of the ureter at the level of the bladder or poor urinary transport due to a highly compliant adynamic ureter. Temporary diversion of the upper urinary tract may be undertaken with nephrostomy tube drainage. When formal upper urinary tract diversion is necessary, there is a theoretical advantage in performing the diversion as proximal as possible. This should maximally relieve stress to the kidney and limit stagnation of urine in a dilated tortuous ureter. However, there is often a disproportionate degree of proximal versus distal ureteral dilation that prevents easy access of the proximal ureter. When there is minimal proximal dilation, a low distal cutaneous ureterostomy provides adequate decompression with relief of stagnated urine flow and stabilization of renal function. The ureter can be approached from a small (2.5-cm) incision placed in a lower inguinal location. The muscles are split to enter the retroperitoneum. The ureter can have the appearance of bowel due to its large size. When in doubt, a 21-gauge needle should be passed, aspirating contents to confirm urine. Once identified, the ureter is sacrificed at the level of the obliterated umbilical artery. The size of the ureter usually prevents postoperative stenosis, allowing for either an end or loop ureteral anastomosis. When definitive urinary reconstruction is undertaken following a distal ureterostomy, the proximal urinary system will have remained uncompromised allowing for easier mobilization and greater flexibility if tailoring of the ureter is required. Definitive Urinary Reconstruction When definitive urinary reconstruction is necessary, the initial approach to the ureter can be extravesical. The ureter is isolated at the level of the bladder and proximal dissection ensues. If a transitional phase exists between the dilated distal ureter and the normal proximal ureter, the dissection should be continued proximal past that point, taking care not to devascularize the ureter. All of the distal ureter is excised when there is adequate length for the proximal ureter to be reimplanted in the bladder in a standard fashion or with the assistance of a psoas hitch. If total proximal and distal ureteral tailoring is necessary, full mobilization of the ureter is required, which can be accomplished via a retroperitoneal approach. However, it is often helpful to enter the peritoneum and reflect either the descending or ascending colon along the white line of Toldt. The dilated ureter is often exceedingly redundant and tortuous. Straightening of the ureter without devascularization is required. The functional capability of the ureter for peristalsis and transmission of urine into the bladder parallels the degree of hydroureter. Therefore, ureteral tapering may enhance urinary flow into the bladder. Multiple techniques exist for ureteral tailoring that include ureteral imbrication and formal ureteral excision 4,5,11 (Fig. 90-1). Ureteral imbrication is appropriate for marginally dilated ureters. But when massive ureteral dilation is present, formal excision is preferred, eliminating the bulky tissue that results from the large imbricated ureter.

FIG. 90-1. (A) The tortuous dilated ureter is carefully straightened without compromising blood supply. The redundant portion is excised and the remaining distal segment tapered if necessary. (B) Ureteral folding over a 10- or 12-Fr ureteral catheter. (C) Formal ureteral tapering with excision and closure. Note: The continuous running closure stops 1 to 2 cm from the end of the segment, followed by interrupted suture placement allowing for excision of the distal end of the ureter without compromise of the running closure.

The ureter should be tapered loosely over either a 10- or 12-Fr catheter depending on the child's age and size. With excision, the excised ureteral segment might take an unconventional course in order to preserve adequate blood supply to the tailored ureter. After excision, the ureter is closed in a two-layer technique. The first running suture line of 5.0 or 6.0 chromic gut directly opposes the mucosa and muscularis of the ureter. The second layer reapproximates the adventitial tissue with the same suture material. Both running layers are discontinued a few centimeters from the distal end of the ureter. The very distal portion of the ureter is closed with interrupted sutures. This allows for excision of the distal ureter without interruption of the running suture line. Having enough ureteral length should not be a problem, allowing for a tunneled antirefluxing ureteroneocystostomy in all cases. Ureteral stents remain in place for 5 to 10 days for postoperative management. If a large, redundant, renal pelvis is present in association with a dilated proximal ureter, a reduction pyeloplasty should be performed in line with the ureteral excision. Preservation of the proximal ureteral blood supply is mandatory. Reduction Cystoplasty Reduction cystoplasty is a very compelling component to urinary reconstruction in a child with triad syndrome. However, long-term follow-up in children having undergone reduction cystoplasty has shown no objective advantage. 1,6 With time the bladder will regain its large size, lose its tone, and lead to inadequate emptying. For these reasons it is not practical to proceed with reductive cystoplasty as the primary indication for urinary reconstruction. If a large, poorly contracting bladder results in inadequate urinary emptying, intermittent catheterization would be a more appropriate form of management. However, when formal urinary reconstruction is required for ureteral tailoring, reductive cystoplasty can be performed and may provide limited improved bladder emptying. Reductive cystoplasty should incorporate the urachus and majority of the dome of the bladder (Fig. 90-2). A 2- to 3-cm strip of mucosa is removed from one side of the bladder wall, allowing for a reinforced overlapping suture line. The bladder is closed in three independent layers using a running suture of 3.0 chromic gut. A suprapubic tube is inserted for postoperative monitoring regarding the effectiveness of bladder emptying.

FIG. 90-2. Reduction cystoplasty. The dome of the bladder, including any urachal remnant, is removed. A 2- to 3-cm mucosal strip is then removed from one portion of the bladder allowing for an overlapping suture line.

Abdominal Wall Reconstruction Several techniques have been devised to maximize the cosmetic benefits of abdominal wall reconstruction in children with triad syndrome. 2,10 There is evidence indicating that the muscular defect is more pronounced centrally and caudally. Initial reconstructive efforts were based on removal of this abnormal tissue. While the appearance of the abdomen was improved, it was not ideal and resulted in loss of the umbilicus. Monfort originally described preservation of the umbilicus. 7 Based on this approach, abdominal wall reconstruction now allows for an excellent cosmetic and functional outcome. The benefit of abdominal wall reconstruction is dependent on the degree of abdominal wall laxity. The timing for this procedure should be based on the need for other operative intervention. If it is obvious that the child will not require upper urinary tract reconstruction, abdominal wall reconstruction can be undertaken at any time. If, however, there is the potential for upper urinary tract reconstruction, abdominal wall reconstruction should be deferred until the time of that intervention. The Monfort approach begins with a midline incision from the tip of the xyphoid process carried inferiorly, circumscribing the umbilicus, leaving an adequate margin of umbilical tissue, and ending at the symphysis pubis ( Fig. 90-3). A full-thickness skin flap is created bilaterally, elevating the subcutaneous fat from the underlying fascia. The dissection is carried laterally to the anterior axillary line. Oftentimes there will be variability and asymmetry of muscular development. Care must be taken not to enter the peritoneum while mobilizing the skin flaps, particularly in areas where the fascia is relatively thin. A lateral incision is then made through the fascia entering the peritoneum. The first incision is made lateral to the superior epigastric artery. Once entrance has been made into the peritoneum, the superior epigastric artery is located and the incision is continued lateral and parallel to its course from the costal margin to the symphysis pubis. Having entered the peritoneum on one side, the fascia is elevated and the contralateral superior epigastric artery identified. A second lateral parallel incision can then be made. The central fascial bridge with the umbilical island is supported by the superior epigastric arteries. The two lateral incisions provide excellent exposure for orchiopexy and major urinary tract reconstruction.

FIG. 90-3. (A) An incision is begun at the xyphoid, circumscribing the umbilicus, and carried down to the pubis. (B) Skin flaps are then elevated, dissecting between the subcutaneous fat and the fascial layer. The lateral extension is the anterior auxiliary line. (C, D) The umbilicus is supported by the central fascial bridge. Incisions will be made into the peritoneum lateral to the epigastric vessels. The central fascial bridge is easily manipulated to allow for excellent intraabdominal exposure. (E) At the time of closure, a line is scored on the peritoneal surface of the fascia. (F) The central fascial strip is then secured laterally to the scored fascia line with a running suture of 2-0 or 3-0 polyglactin. (G) The lateral fascia is then secured in the midline above and below the umbilicus with 2-0 or 3-0 polyglactin. Centrally, the fascia is secured directly to the umbilicus. This allows for an overlapping reinforced fascial wall closure. Subcutaneous tissue is closed with 3-0 or 4-0 plain gut and the skin with a running subcuticular 4-0 or 5-0 polyglactin.

At the time of closure, the lateral fascia wall is secured to the central fascial strip with a running 2-0 or 3-0 polyglactin. The suture line is scored on the peritoneal side of the lateral fascia to enhance adherence. The lateral fascia is then secured in the midline with figure-of-8 suture placement using 2-0 or 3-0 polyglactin. This pants-over-vest closure provides additional ventral support. Two flat 7-Fr suction drains are placed between the fascia and the subcutaneous space. The skin flap is then tailored, removing the excess, allowing for a midline and periumbilical closure. The skin flap is closed in multiple layers, securing the subcutaneous tissue with 4-0 plain gut sutures. The epithelial edge is reapproximated with a running subcuticular suture of 5-0 polyglactin. The drains remain in place for 2 or 3 days for decompression of the dead space. Orchiopexy The timing for orchiopexy can be individualized based on the child's need for urinary reconstructive surgery. If urinary reconstructive surgery is required, orchiopexy can be performed in the same procedure. If urinary reconstructive surgery is not required, then timing is variable. Placement of the testes in the scrotum is obviously important for psychological and hormonal factors but, unfortunately, fertility does not appear to be improved. Biopsies of testes have shown a Sertoli-cell-only feature prohibiting future fertility. 12 Sacrifice of the gonadal artery may be required to obtain adequate length for the testicle to be delivered in the scrotum as described by Fowler and Stephens. 3 If orchiopexy is undertaken early, particularly within the first year of life, there is often adequate vascular length to deliver the testicle directly into the scrotum without transection of the testicular artery. A staged approach, using laparoscopic-assisted ligation of the gonadal artery, followed by a 6-month delay in delivering the testicle in the scrotum, has also recently been described. It appears appealing if urinary reconstruction and abdominal wall reconstruction are unnecessary. Exposure for the orchiopexy is often dependent on other urinary or abdominal wall reconstructive efforts. When the gonad is first identified it may be found closely associated with a dilated distal ureter. To determine whether the testicle can be delivered into the scrotum without sacrifice of the gonadal artery, the testicle should be released from the ureter. A lateral peritoneal incision can be made to the proximal gonadal artery. The incision is continued medial to the artery and carried down inferiorly. It is important to not disrupt the vascular supply of the peritoneal pedicle running on both sides of the vas deferens. After mobilization, if it becomes apparent that the testes will not reach into the scrotum, the gonadal artery is sacrificed. The blood supply to the testes is maintained by the vasal artery and small anastomotic channels in the peritoneal flap. A tunnel is then made into the scrotum and an incision placed inferiorly in the scrotum to create a dartos pouch. A clamp is passed from the scrotum to the inguinal canal. The testicle is grasped, pulled down through the tunnel, and delivered to the scrotum. Care must be taken not to twist or place the peritoneal pedicle on tension. If desired, the testicle can be secured to the dartos tissue with 5-0 PDS suture.

OUTCOMES

Complications Ureteral devascularization resulting in ischemia and subsequent obstruction can occur if attention has not been paid to the ureteral blood supply. The risk of bowel obstruction is present as in any intraabdominal procedure. Testicular ischemia and atrophy due to a Fowler-Stephens procedure has been reported to occur in 15% of children. Results The results of urologic reconstruction can be very gratifying in the initial postoperative period, particularly with removal of significant redundancy and relief of stagnated urine. However, with time, there can be an increase in both bladder size and ureteral dilation. This is often due to ineffective voiding and is independent of bladder reduction. For those reasons, the urinary tract must be closely monitored for an extended period. Patients should be prepared for the potential need for intermittent catheterization. Because of normal sensation, children are often unwilling to cooperate with urethral catheterization. If catheterization appears to be a realistic possibility at the time of urinary reconstruction, placement of an appendicovesicostomy should be considered. This provides a very easy mechanism for catheterization if required. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Bukowski TM, Perlmutter AD. Reduction cystoplasty in the prune-belly syndrome: a long term follow up. J Urol 1994;152:2113–2116. Ehrlich RM, Lesavoy MA, Fine RN. Total abdominal wall reconstruction in the prune-belly syndrome. J Urol 1986;136:282–285. Fowler R, Stephens FD. The role of testicular vascular anatomy in the salvage of high undescended testis. Aust N Z J Surg 1959;29: 92–106. Hendren WH. Operative repair of megaureter in children. J Urol 1969;101:491. Kalicinski ZH, Kansy J, Kotarbinska B, et al. Surgery of megaureters modification of Hendren's operation. J Pediatr Surg 1977;12:183. Kinahan TJ, Churchill BM, McLorie GA, et al. The efficiency of bladder emptying in the prune-belly syndrome. J Urol 1992;148:600–603. Montfort G, Guys JM, Boccoardo, et al. A novel technique for reconstruction of the abdominal wall in the prune belly syndrome. J Urol 1991;146:639–640. Osler W. Congenital absence of the abdominal muscles with distended and hypertrophied urinary bladder. Bull Johns Hopkins Hosp 1901;12:331–333. Parker RW. Absence of abdominal muscles in an infant. Lancet 1895;1:1252–1254. Randolph JG. Total surgical reconstruction for patients with abdominal muscular deficiency (“prune-belly”) syndrome. J Pediatr Surg 1977;12:1033–1043. Starr A. Ureteral plication. A new concept in ureteral tailoring for megaureter. Invest Urol 1979;17:153. Uehling DT, Zadina SP, Gilbert E. Testicular histology in triad syndrome. Urology 1984;23:364. Wheatley JM, Stephens FD, Hutson JM. Prune-belly syndrome: ongoing controversies regarding pathogenesis and management. Semin Pediatr Surg 1996;5:95–106.

Chapter 91 Supravesical Urinary Diversions Glenn’s Urologic Surgery

Chapter 91 Supravesical Urinary Diversions
Byron Joyner and Antoine Khoury

B. Joyner: Division of Urology, The Hospital for Sick Children, Toronto, Ontario M5G 1X8, Canada. A. Khoury: University of Toronto, and Department of Surgery, Division of Urology, The Hospital for Sick Children, Toronto, Ontario M5G 1X8, Canada.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Cutaneous Pyelostomy and High Cutaneous Ureterostomy Pelvioureterostomy-En-Y (Sober Loop Ureterostomy) Closure of Pyelostomies and Ureterostomies End-Cutaneous Ureterostomies Outcomes Complications Results Chapter References

Over the last century, various forms of cutaneous urinary diversions have been developed. The inception of these developments began in the mid-1800s when Rayer described the association between hydronephrosis and renal failure. Surgical intervention for this problem was not documented until the 19th century when percutaneous drainage of a distended renal pelvis was performed. In 1878, Robert Weir performed the first open drainage of a hydronephrotic kidney. Twenty years later, the nephrotomy was espoused by Henry Morris whose patient with a solitary kidney drained for 6 years postoperatively. These seminal attempts at urinary diversion have inspired multiple techniques that have all been neatly categorized as either intubated or nonintubated urinary diversions. Intubated diversions typically are employed in the acute setting of urinary obstruction or difficulty voiding. The tube remains in place for brief periods of time to facilitate urinary drainage. Intubated diversions can be divided further into vesical or supravesical diversions. The Foley catheter and suprapubic tube account for the majority of intubated vesical diversions. A variety of nephrostomy tubes and nephrovesical (double-J ureteral) stents account for the supravesical diversions. Nonintubated, vesical urinary diversions are surgically constructed in situations requiring prolonged periods of diversion. These would include perineal urethrostomies or cutaneous vesicostomies. The non-intubated, supravesical diversions include cutaneous pyelostomies, loop ureterostomies, and end-cutaneous ureterostomies.

DIAGNOSIS
Indications for nonintubated urinary diversions and, specifically, supravesical urinary diversions have been better defined and are performed rarely now compared to 20 years ago. New technical advances in radiology and urology have resulted in fewer temporary urinary diversions and more definitive reconstructive surgery for those children with obstructive uropathies. The increased use and clarity of ultrasonography has had a significant impact on the detection of antenatal hydronephrosis, as well as the subsequent management of its postnatal course. Refined endoscopic equipment for the treatment of posterior urethral valves, improved anesthesia, standardized intensive care, and better operative techniques have made extensive reconstructive efforts more common earlier in infancy. The most effective management of these children is to progress through a logical algorithm in order to establish adequate urinary drainage ( Fig. 91-1). Initially, when posterior urethral valves are suspected, a 5-Fr pediatric feeding tube is inserted into the urethra to drain the bladder. The response to catheter drainage will be determined by the level of the serum creatinine. This response, in association with the physical examination and radiographic findings, will direct further management. Radiographic studies may provide further insight into the quality of the bladder. A good bladder has a normal capacity (ml) for age (in infants: weight in kg × 7) and normal compliance. In contrast, a bad bladder is characterized as thick and noncompliant. The detrusor muscle hypertrophies in response to infravesical obstruction, with subsequent trabeculation and sacculation. These distinct findings are extremely useful in managing the infant who may need some type of urinary diversion.

FIG. 91-1. Management algorithm for infravesical urinary obstruction.

INDICATIONS FOR SURGERY
Despite new technical advances and earlier reconstructive surgery, supravesical urinary diversions are still occasionally indicated. They are, however, typically restricted to newborns with infravesical obstruction who subsequently have progressive hydroureteronephrosis, poor urinary drainage, and renal compromise, despite initial urethral intubation. In this situation, the most common diagnoses are posterior urethral valves associated with a thickened, stiff bladder; high-grade vesicoureteral reflux; or obstructive megaureter. These babies require stabilization in an intensive care unit where their fluid and electrolyte status can be optimized, their acid–base problems corrected, and sepsis addressed. The next step is to decide on the best method of urinary diversion. The first diagnostic, as well as therapeutic, step in the posterior urethral valve management algorithm is to insert a 5-Fr urethral catheter. If the urethral catheter causes the creatinine to decrease daily by 10% to a nadir of less than 80 mol/L by day 5, then transurethral resection of the posterior urethral valves is performed. Primary valve ablation is the preferred treatment of choice. But temporary urinary diversion may be necessary if the creatinine continues to rise after valve ablation. Children who may be too small (<2.5 kg) or too sick for a technically satisfactory endoscopic procedure may be best managed by a cutaneous vesicostomy. Then, when the urethra is larger, endoscopic valve ablation can be performed simultaneously with vesicostomy closure. It should be emphasized that a 5-Fr pediatric feeding tube is preferred to a 6-Fr Foley catheter. An indwelling Foley catheter with an inflated balloon could potentiate renal failure by at least two mechanisms. First, the weight and size of the balloon could actually occlude the delicate ureteral orifices. Additionally, the balloon might aggravate detrusor instability with subsequent elevated vesical pressures. Both of these problems could result in progressive hydroureteronephrosis.

ALTERNATIVE THERAPY
Babies who are candidates for a cutaneous vesicotomy should have a voiding cystourethrogram (VCUG) in order to determine the bladder capacity and configuration, as well as to establish the presence and grade of vesicoureteral reflux. Then, a vesicotomy can be performed as a temporary diversion that not only drains the

bladder but also decompresses the entire urinary tract. Still, there are those babies who do not improve in spite of a cutaneous vesicostomy. They have persistent renal failure, severe hydronephrosis, and sepsis. In this case when immediate urinary tract reconstruction would not be appropriate, a temporizing supravesical urinary diversion and a renal biopsy may be beneficial. Review of the radiographic studies is essential and can direct further management. For example, an infant whose VCUG demonstrates massive bilateral reflux or unilateral reflux associated with ipsilateral renal dysplasia (VURD syndrome) may benefit from a pyelostomy or loop ureterostomy and renal biopsy. This maneuver may serve to decompress the bladder and protect the normal kidney. When the renal ultrasound reveals hyperechoic renal parenchyma, consistent with renal dysplasia, a supravesical diversion may provide the best decompression as well as the best method of monitoring the urine output from each renal unit. A supravesical diversion allows the urine to bypass the long, tortuous ureter and the intramural tunnel, both of which could delay, if not impede, urinary drainage. Theoretically, a more proximal diversion should provide for better drainage. Supravesical diversions are useful in the management of vesical and urethral anomalies that are associated with persistence of upper tract obstruction, unrelieved by either valve ablation or vesicostomy. Additionally, the stable child with massively dilated ureters may be decompressed with a supravesical ureteral diversion, making ureteral tapering unnecessary and allowing interval improvement of the renal function. Duckett claims that infants with ipsilateral renal functions of less than 15% benefit from supravesical diversions because this maneuver may allow the kidney to regain some of its function. In this setting, supravesical diversions have a threefold benefit: First, a biopsy of the kidney can be obtained during the procedure. A renal biopsy might be helpful in sorting out whether the baby whose creatinine continues to rise despite adequate infravesical diversion is adequately drained or whether he simply has poorly functioning kidneys. It may also have prognostic value. Second, loop cutaneous ureterostomy affords sequential monitoring of the affected kidney. Third, in the interim, the surgeon can decide on the appropriate, definitive surgical management. However, the proximal approach should not be the first choice of urinary diversion in the majority of neonates with severe hydroureteronephrosis, poor urinary drainage, and renal compromise. Some urologists uti-lize high diversions, selectively, in the sickest children with progressive uropathy and severe hydroureterone-phrosis. Others claim that there are no indications for supravesical diversions whatsoever, citing that they are prone to ureteral stricturing, difficult to reconstruct, and detrimental to the bladder. The debate regarding the last claim has existed for many years and is unlikely to be resolved soon. Proximal urinary diversions are further criticized for their inability to preserve bladder cycling. That is to say that high diversions prevent bladder storage and expulsion of urine and may result in a defunctionalized bladder. This is an attractive, although controversial, theory. Some clinicians believe that postdiversion bladder defunctionalization is related to issues that are intrinsic to the bladder. This could include, for example, posterior urethral valves or neurogenic bladder disease. Still, others believe that postdiversion bladder contraction occurs following severe, chronic infections or fibrosis secondary to multiple surgical procedures. But Jayanthi et al. concluded in a retrospective review of temporary diversions that most patients can undergo primary reversal of the temporary cutaneous diversion with preservation of bladder function. The only certainty is that prudent indications should guide the surgeon who plans to create a supravesical diversion.

SURGICAL TECHNIQUE
Cutaneous Pyelostomy and High Cutaneous Ureterostomy Loop ureterostomies or pyelostomies continue to have clinical application. They are rarely initial procedures and are often performed only after a cutaneous vesicostomy does not improve progressive hydroureteronephrosis associated with azotemia, urosepsis, high-grade reflux, and/or incomplete bladder emptying. The cutaneous ureterostomy and the pyelostomy provide adequate urinary drainage and they both have the additional benefit of allowing a diagnostic renal biopsy to be obtained. A renal biopsy may have prognostic value in that the degree of renal dysplasia may be assessed. Another benefit to performing either a loop cutaneous ureterostomy or a pyelostomy is that, if indicated, radiographic evaluation can be performed in a retrograde or an antegrade fashion. The cutaneous pyelostomy (Fig. 91-2) is preferred only when the renal pelvis is of sufficient size to reach the skin without tension or distortion of the ureteropelvic junction. A pyelostomy has the additional advantage of maintaining ureteral integrity. This makes subsequent closure and ureteral reimplantation easier by avoiding devascularization of the ureter. Pyelostomies or loop ureterostomies are the urinary diversions of choice for the child with a chronically dilated, obstructed, redundant ureter who has failed vesical or infravesical diversion.

FIG. 91-2. Technique of loop cutaneous pyelostomy in the dorsal lumbotomy position.

When performing a loop ureterostomy ( Fig. 91-3) there are several technical pearls to remember in order to avoid ureteral stenosis, which is the most common problem following this procedure:

FIG. 91-3. Technique of loop cutaneous ureterostomy.

1. Large ureters with diameters greater than 8.0 mm should be considered in order to prevent compromise of the blood supply and subsequent stricture formation.

2. For the same reason, limited dissection of the ureter and its attendant adventitia should be performed. 3. Careful selection of cutaneous stomal site is extremely important. The stoma should not be placed in a position that might cause occlusion of urinary flow when the child is in the seated or supine positions. There are a variety of approaches to the ureter, including subcostal midline extraperitoneal or posterior lumbotomy. We prefer the posterior lumbotomy approach ( Fig. 91-4A,B), which offers several advantages over the other approaches. It provides excellent, rapid exposure bilaterally without the necessity for turning the child to access the opposite side. Unlike the subcostal approach, access to the kidney is obtained by dividing the posterior fascial layers, without transecting muscles. Therefore, postoperative pain is considerably reduced. Accordingly, this approach allows for strong closures as the natural interposition of the muscles between the fascial layers discourages the development of incisional hernias. The only caveat is that the ureteral anatomy must be precisely defined prior to embarking on a technique that provides limited exposure.

FIG. 91-4. (A) Dorsal lumbotomy position in the child. The skin incision is transverse and the fascial incision vertical. (B) Course of dorsal lumbotomy lateral to the sacrospinalis and quadratus lumborum muscles. The incision does not require division of any muscle tissue.

The patient is anesthetized in the standard supine position and then carefully turned to the prone position. All pressure points are padded. Cushioning rolls should be placed under the chest, and at the hips (just superior to the anterior superior iliac spines) to prevent lordosis. Finally, the child should be taped into position ( Figure 91-5).

FIG. 91-5. Positioning for the bilateral dorsal lumbotomy approach.

The surface landmarks of the posterior lumbotomy incision include the 12th rib superiorly, the lateral border of the sacrospinalis muscle medially, and the iliac crest inferiorly ( Fig. 91-6). A transverse incision is made along Langer's lines, approximately one-third the distance from the 12th rib to the iliac crest. The incision should start just medial to the lateral border of the sacrospinalis muscle and run laterally for a distance equivalent to that between the 12th rib and the iliac crest. The skin with its attendant subcutaneous tissue is elevated for approximately 4 to 5 cm superiorly and inferiorly in order to expose the lumbodorsal fascia. This posterior lamina of the lumbodorsal fascia is incised vertically, 1.5 to 2 cm lateral to the midline, just below the 12th rib with a sweeping lateral curve that should hook just superior to the iliac crest. When making the superior part of this incision, the subcostal neurovascular bundle should be avoided. The sacrospinalis muscle is retracted medially to expose the middle lamina of the lumbodorsal fascia. This layer is incised at the lateral border of the quadratus lumborum muscle vertically. When making the superior part of this incision, the subcostal neurovascular bundle should be avoided. The quadratus lumborum muscle is retracted medially. At this point, preserve the iliohypogastric and ilioinguinal nerves, which are identified as they run across and superficial to the thin anterior lamina of the lumbodorsal fascia. After opening this fascial layer, apply a baby Balfor retractor.

FIG. 91-6. The surface landmarks of the dorsal lumbotomy incision include the 12th rib superiorly; the lateral border of the sacrospinalis muscle medially; and the iliac crest inferiorly. The length of the incision is measured from the border of the 12th rib and the sacrospinalis muscle, cephelad, to the superior border of the iliac crest, in an oblique direction.

The kidney normally lies in the superomedial aspect of the wound and should be palpated before incising Gerota's fascia. This will prevent violation of the peritoneum. Sweep the kidney laterally to expose the renal pelvis. Then, insinuate a small narrow Deavor to retract the kidney laterally. If the renal pelvis is large enough, a pyelostomy should be performed. Mobilize the pelvis just enough to reach the skin, making sure to avoid tension and angulation of the ureteropelvic junction (UPJ). If the pelvis is intrarenal or cannot be mobilized to the surface without tension, then the ureter should be brought to the skin level. Interrupted, 4-0 polyglactin or polyglycolic sutures are used to create the stoma by securing the ureteral adventitia to the external oblique fascia. Similar suture can be used to mature the stoma to the skin. Pelvioureterostomy-en-Y (Sober Loop Ureterostomy) The pelvioureterostomy-en-Y technique introduced by Sober in 1972 is just one of several modifications of the original loop ureterostomy ( Fig. 91-7). It is a more extensive procedure and therefore is not recommended when expediency is essential. An advantage of the Sober modification is that the closure is quite simple and

is not associated with high morbidity. The Sober technique also allows efflux of urine into the bladder, which may avoid the postdiversion bladder contracture associated with a dry bladder. The approach of the Sober technique is identical to that of the previous procedure. The proximal ureter is mobilized and brought out to the skin in a tension-free manner. The distal ureter is anastomosed to the renal pelvis in an end-to-side fashion using fine (6-0) polyglactin or polyglycolic sutures.

FIG. 91-7. The pelvioureterostomy-en-Y Sober technique.

The creation of the stoma is similar in all cases. For the high-loop ureterostomy and pyelostomy, a 5-0 prolene traction suture can be used to facilitate a 2-cm longitudinal incision, with care taken not to perforate the back wall. Interrupted, fine (4-0) polyglactin or polyglycolic sutures are used to create the stoma by securing the ureteral adventitia to the external oblique fascia. Similar suture is used to mature the stoma to the skin. Removal of the Balfor retractor will allow reapproximation of the muscles. Muscle and fascia should not be sutured between the limbs of the loop as this may cause obstruction. One or two 4-0 polyglactin or polyglycolic sutures are placed in the posterior layer of the lumbodorsal fascia. Closure of Pyelostomies and Ureterostomies Closure of a temporary pyelostomy begins only after careful consideration has been made of all clinical, radiographic, and laboratory data. Once all of these parameters have been satisfied, reconstruction of the urinary system is performed. An elliptical incision is made around the stoma, which is adequately mobilized. The exteriorized portion of the stoma is resected so that only healthy renal pelvic edges can be approximated in a transverse (Heineke-Mikolitz) fashion with fine (6-0) polyglactin or polyglycolic acid sutures. This technique increases the diameter of the ureter and discourages subsequent stenosis at the stoma site. The ureteropelvic junction, which subtends the stoma site, should always be checked to ensure that it is free of angulation or obstruction that could cause subsequent morbidity. Closure of the Sober pelvioureterostomy-en-Y ureterostomy is easier than the pyelostomy closure. The stoma at the level of the skin is excised to the level of the renal pelvis. The patent pelvis is oversewn with 6-0 polyglycolic or polyglactin sutures and the skin is reapproximated. Closure of a loop ureterostomy is technically more demanding than the original operation and can be frought with significant morbidity ( Fig. 91-8). Closure is usually delayed for at least 6 months to a year. This would give sufficient time for renal function to improve and for collateral ureteral circulation to develop. Like closure of the pyelostomy, an elliptical incision is made around the stoma. Adequate mobilization of the proximal and distal ureteral limbs should be accomplished so as not to create any angulation of the ureter. Leave the back wall intact and excise only the exteriorized portion of the stoma. This portion usually undergoes metaplastic changes and is typically adynamic. The remaining healthy ureteral margins are then spatulated and closure is performed in a transverse fashion, using interrupted fine polyglactin or polyglycolic acid suture. Urinary diversion is optional but may prevent anastomotic urinary leak. A small drain can be placed.

FIG. 91-8. Takedown of loop cutaneous ureterostomy. The stoma and adjacent fibrotic ureter are excised, leaving the back wall intact. The remaining, healthy ureteral margins are then spatulated and closure is performed in a transverse fashion.

End-Cutaneous Ureterostomies The neonate who would be a candidate for a supravesical diversion typically has severely dilated ureters and azotemia. The goal of any diversion is to improve urinary drainage. Coaptation is the method by which a bolus of urine is formed and carried distally into the bladder. Because the ureters in these babies are so dilated, they do not coapt and, therefore, urine transport primarily depends on the hydrostatic forces generated by the kidney and gravity. This method of drainage is ineffective and urine simply pools and moves in an aimless Brownian motion. If the child is supine, then the urine will not drain and eventually will become infected. Bacterial toxins released during a urinary tract infection further paralyze the already nonfunctioning ureter, thus complicating the problem. For all of these reasons, end-cutaneous ureterostomies have little theoretical or clinical application in pediatric urology.

OUTCOMES
Complications Complications of supravesical diversion are primarily stomal stenosis. Meticulous attention to the creation of the stoma should obviate this complication. Hernias are very rare using the above techniques and closures are relatively free of complications. It should be reemphasized, however, that these are temporizing procedures performed primarily in the newborn and are not recommended for long-term diversion. Despite the fact that end-cutaneous ureterostomies allow fairly easy construction with a low incidence of stomal stenosis, there are at least two disadvantages compared to other forms of supravesical diversion: First, this procedure does not allow the kidneys to be biopsied, which could be of prognostic value. Second, the distal ureter, having been disconnected from the bladder at the original procedure, is now dependent on its proximal blood supply. This becomes obvious at the time of closure. In order to reestablish vesicoureteral continuity, it must be mobilized again, causing further compromise of its vascular supply. This is not an ideal situation for reimplantation of the ureters and it places them at an unnecessarily high risk for ischemic change and ureteral stenosis. Results Supravesical urinary diversions are performed much less frequently today than 20 years ago. Although reconstructive surgery and new technical advances direct treatment more commonly at the primary pathology, supravesical diversions are still an effective, though controversial, option for temporary or permanent urinary

transport in children. Neonates who may benefit most are those with significant hydroureteronephrosis, progressive renal failure, and uncontrolled infection. These babies are usually extremely sick and require an expedient procedure to divert their upper tracts. A therapeutic algorithm is a good way to process the information. But the ultimate therapeutic result depends on matching the patient with the most appropriate surgical technique. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Burstein JD, Firlit CF. Complications of cutaneous ureterostomy and other diversions. Urol Clin North Am 1983;3:433–443. Glassberg KI. Current issues regarding posterior urethral valves. Urol Clin North Am 1985;12:175–185. Hendren WH III. Complications of ureterostomy. J Urol 1978;120:269–281. Johnston JH. Temporary cutaneous ureterostomy in the management of advanced congenital urinary obstruction. Arch Dis Child 1963;38:161–166. Krueger RP, Hardy BE, Churchill BM. Growth in boys with posterior urethral valves: primary valve resection vs upper tract diversion. Urol Clin North Am 1980;7:265–272. Mitchell ME, Rink RC. Urinary diversion and undiversion. Urol Clin North Am 1985;12:111–122. Perlmutter AD, Patil J. Loop cutaneous ureterostomy in infants and young children: late results in 32 cases. J Urol 1972;107: 655–659. Sober I. Pelvioureterostomy-en-Y. J Urol 1973;107:473–475. Williams DI, Rabinovitch HH. Cutaneous ureterostomy for the grossly dilated ureter of childhood. Br J Urol 1967;39:696–699.

Chapter 92 Surgery for Childhood Rhabdomyosarcoma Glenn’s Urologic Surgery

Chapter 92 Surgery for Childhood Rhabdomyosarcoma
Bruce Broecker

B. Broecker: Department of Surgery, Division of Urology, Emory University School of Medicine, Atlanta, Georgia 30329.

Pathology and Biology Diagnosis Clinical Presentation Clinical Grouping and Staging Evaluation Indications for Surgery Alternative Therapy Surgical Technique Bladder/Prostate Vagina/Uterus Paratesticular Outcomes Complications Results Chapter References

Rhabdomyosarcoma (RMS) is a malignant tumor that arises from undifferentiated mesodermal cells but bears a resemblance to normal fetal muscle prior to innervation. It is the most common soft tissue sarcoma of childhood, representing more than half of such tumors in the pediatric population (approximately 250 cases per year in the United States). 9 It occurs in a variety of characteristic locations throughout the body with about 25% arising in genitourinary locations (bladder, prostate, vagina, uterus, paratesticular area). 3 Its rarity and prior poor prognosis have led to cooperative trials to determine and refine new therapies for children with this tumor. In North America, the Intergroup Rhabdomyosarcoma Study (IRS) formed in 1972 to pool patients and investigative resources; its protocols are the foundation of contemporary treatment of RMS in the United States. In Europe, the Society of Pediatric Oncology (SIOP) has performed a similar function. Contemporary treatment of RMS is multimodal, employing surgery, radiation, and chemotherapy in combinations tailored to the location and stage of the tumor. Long-term survival today is approximately 70%, in sharp contrast to cure rates of less than 30% prior to 1970. 3,4,12 Importantly, these contemporary cure rates encompass patients in whom organ preservation (i.e., bladder salvage in patients with bladder/prostate RMS) has steadily increased. Pathology and Biology Rhabdomyosarcoma is one of a broader category of small, blue, round cell tumors of childhood. Its name derives from its resemblance to fetal skeletal muscle and consistent with this relationship rhabdomyoblasts share patterns of expression of muscle-specific genes, including the MyoD family of regulatory factors. 5,7,8 Classification of a tumor as RMS depends on identification of its myogenic lineage. Typically, this consists of light microscopic identification of cross-striations typical of skeletal muscle or characteristic rhabdomyoblasts, electron microscopic identifications of sarcomeres with Z bands and actin and myosin filaments, and/or immunohistochemical expression of muscle-associated proteins such as myoglobin, desmin, and actin, in addition to MyoD proteins. The first formal histologic classification of RMS was proposed by Hom and Enterline in 1958. They described three histologic variants: embryonal, alveolar, and pleomorphic. This system was adopted by both the IRS Committee and the World Health Organization and remains widely used today. Recently, an international group of pathologists has attempted to improve the classification of RMS with one that is more reproducible among pathologists and more prognostic of outcome. Their modified classification system proposes three risk-based categories: favorable prognosis, which includes botryoid and spindle cell histologic subtypes; intermediate prognosis, which includes the embryonal subtype; and unfavorable prognosis, which includes the alveolar and undifferentiated sarcoma subtypes. 2 This system is being evaluated in current IRS clinical trials. Several cytogenetic abnormalities have been found in RMS. In alveolar tumors a translocation involving chromosome 2 and 13, t(2:13) (q35:q14), is frequently found. Embryonal tumors have been found to have a loss of heterozygosity at the 11p15 locus. Both embryonal and alveolar tumors may have p53 mutations. This mutation is characteristically found in patients with the Li-Fraumeni syndrome, who have a 20- to 30-fold risk of developing bony or soft tissue sarcoma.

DIAGNOSIS
Clinical Presentation Presenting symptoms of patients with RMS depend on the location and size of the tumor. Tumors arising in the pelvis (bladder, prostate, vagina, and uterus) most commonly present with a palpable lower abdominal mass. Other symptoms or findings may include voiding difficulties (including urinary retention), hematuria, and an introital mass. Paratesticular tumors present as a painless, firm mass in the scrotal or inguinal region. Clinical Grouping and Staging Staging systems for RMS categorize the extent of disease and form the basis of the risk-based stratification of patients used to determine treatment and compare outcomes. The original clinical grouping system developed by the IRS ( Table 92-1) is based on the extent of residual disease following surgical resection. The limitations of this system have become apparent as the treatment philosophy incorporated in later IRS protocols has shifted to include organ preservation. A pretreatment TNM staging classification system has been developed ( Table 92-2) to address these limitations. 6 Current IRS treatment protocols stratify patients based on both TNM stage and clinical group.

TABLE 92-1. IRS clinical groups

TABLE 92-2. TNM pretreatment staging classification of the IRS IV

Evaluation The radiographic evaluation of a patient with a suspected or proven RMS is most commonly a computed tomogram (CT scan) of the region of the primary tumor. Evaluation of the primary tumor is done to assess the size and location of the tumor, the presence or absence of regional lymph node enlargement, and the involvement of any adjacent organs. Occasionally, a magnetic resonance imaging (MRI) scan will improve this assessment. Metastatic spread of RMS most commonly involves regional lymph nodes, lung, bone marrow, and bone, and consequently a CT scan of the chest, a bone scan, and bone marrow aspirate are the appropriate studies to evaluate this concern. The diagnosis itself requires histopathologic evaluation of tissue obtained either by biopsy or at the time of surgical excision.

INDICATIONS FOR SURGERY
Surgery in patients with RMS is both diagnostic and therapeutic. The current IRS surgical guideline for patients with newly diagnosed RMS is initial complete removal of all tumor with a margin of normal tissue when such removal is possible and does not result in the removal of important organs or unacceptable loss of organ function (IRS-4 Surgical Guidelines, personal communication).

ALTERNATIVE THERAPY
When complete removal is not possible or would result in the loss of important organs, biopsy alone is performed. All patients then receive further treatment with combination chemotherapy with or without radiotherapy. This approach has evolved through successive IRS studies 1,2,3 and 4 and is the result of an appreciation of the responsiveness of RMS to current chemotherapy and radiotherapy. Further surgery is frequently employed in these patients both to establish by biopsy what response has been achieved by adjuvant therapy (chemotherapy/radiotherapy) and to achieve a complete response (CR) when possible by further resection.

SURGICAL TECHNIQUE
Bladder/Prostate Complete surgical excision with preservation of bladder function may be possible in tumors arising in the dome of the bladder. However, most bladder tumors arise from the bladder neck and these, along with tumors arising in the prostate, should have biopsy alone as the initial surgical procedure. Biopsy may be done percutaneously, endoscopically, or by laparotomy. Patients presenting in urinary retention or with bilateral upper tract obstruction will require intubated decompression as part of the initial surgical treatment. All patients initially biopsied should have a second-look operation (SLO) after the completion of all adjuvant therapy, generally week 46/47 of treatment. At this time, if there is evidence of residual disease further resection, including cystoprostatectomy, should be done when this can result in a CR. In rare patients with prostatic primaries, residual disease confined to the prostate may be removed by prostatectomy with preservation of bladder function. Great care should be taken in such situations to ensure that bladder neck and urethral margins are negative before considering this option ( Table 92-3).

TABLE 92-3. Surgical procedures for children with genitourinary rhabdomyosarcomas

Radical Cystoprostatectomy (Males) The patient is placed supine with the lower extremi- ties distracted with stirrups or spreader bars ( Fig. 92-1). The pelvis can be opened by placing a rolled towel beneath the sacrum to increase the pelvic tilt and provide better visualization of the pelvic structures. The incision is made in the midline, beginning at the level of the symphysis pubis and carried around the umbilicus and into the upper abdomen to the xyphoid process.

FIG. 92-1. Patient position for cystoprostatectomy.

It may be convenient to enter the peritoneal cavity at the level of the umbilicus. The posterior aspect of the rectus sheath is incised, and the peritoneum is grasped with Halstead clamps. An opposing pair of Halstead clamps permits the surgeon to palpate the anterior peritoneum with the thumb and the forefinger to determine that the bowel does not underlie the intended area of incision. The peritoneal cavity should be sharply incised between the Halstead clamps and an examining finger placed into the peritoneal cavity. Using this finger to provide tension, the anterior peritoneum can be lifted and sharply incised from the umbilicus to the upper aspect of the incision. Attention is turned to preparing the anterior peritoneum for development of the anterior peritoneal flap. The intent is to provide an inverted V, with the point of the V based at the umbilicus and the arms of the V ending at the internal rings bilaterally ( Fig. 92-2). This is accomplished by dissecting the anterior peritoneum from the posterior rectus sheath. The urachus and the obliterated umbilical vessels are grasped with a Kelly clamp just below the umbilicus, and the anterior peritoneal V flap is developed. The incision is carried down toward the internal rings bilaterally, using the epigastric vasculature as the lateral margins of the distal aspect of the flap (Fig. 92-3).

FIG. 92-2. Lines of incision of the peritoneum from superior (apex of V) to lower abdomen. The arms of the V are directed to the proximal femoral vessels.

FIG. 92-3. Creation of anterior peritoneal flap.

The surgeon's finger can be passed beneath the peritoneum covering the pelvic floor. The peritoneum is incised, exposing the iliac vessels, and the posterior peritoneal flap is developed as the peritoneal incision is carried parallel and medial to the external iliac vasculature to a point approximately 2 cm above the bifurcation of the right and left common iliac ( Fig. 92-4). During this dissection the vas deferens is encountered. The vas deferens should be divided, with the proximal end (the end closest to the testicle) being controlled either with metallic surgical clips or absorbable sutures and a long traction suture of either silk or chromic catgut placed on the distal portion of the vas deferens, the portion that is carried with the specimen.

FIG. 92-4. Mobilization of the peritoneum inferiorly and posteriorly to provide access to the pelvic vasculature and ureters.

The ureters are identified above the bifurcation of the common iliac vasculature bilaterally, and they should be isolated with surgical tapes. Division of the ureters should be delayed until it can be determined that radical cystectomy is appropriate for the patient. Pelvic Lymphadenectomy Pelvic lymphadenectomy is begun at the bifurcation of the common iliac vasculature, carried down the interior margin of the external iliac vein to the pelvic floor, across the pelvic floor, and up the hypogastric vasculature. All lymphatic and fibrofatty tissue that lies within this triangle is removed, including the lymph node packages surrounding the obturator nerve. This dissection not only allows assessment of the stage of the cancer but also facilitates the anatomic dissection necessary for cystectomy and better visualization of the vascular supply of the specimen. Completion of the Cystectomy in Males The endopelvic fascia on either side of the prostate should be incised from a point just inferior to the puboprostatic ligaments posteriorly to the inferior border of the prostate (Fig. 92-5). Once this has been accomplished, the surgeon can pass a finger between the prostate and the bony pelvis to the apex of the prostate.

FIG. 92-5. Point of incision in the endopelvic fascia for mobilization of the prostate.

The ureters should be divided deep in the pelvis. The proximal ureters may be tagged with a 3-0 black silk suture and packed beneath the posterior peritoneum behind a moist sponge, protecting the ureter from harm during completion of the dissection. With traction on the Kelly clamp previously placed on the obliterated urachus, the specimen can be elevated in the pelvis. Traction allows easy identification of the vascular pedicles (Fig. 92-6 and Fig. 92-7). The superior, middle, and interior vascular pedicles can be controlled either by large metallic surgical clips or by heavy absorbable sutures (Fig. 92-8). After the lateral pedicles are divided, palpation in the pouch of Douglas reveals a ridge across the deep pelvis, which identifies the vas deferens and the appropriate line of incision for development of the posterior peritoneal flap ( Fig. 92-9). Allis clamps should be placed on the patient side of this incision, and the surgeon, with blunt digital dissection in the midline, can develop a surgical space behind the bladder and anterior to the rectum. This is best accomplished by moving the fingers in an anterior posterior direction, sweeping the prerectal fascia away from the posterior aspect of the bladder, seminal vesicles, and prostate. The surgeon should be able to develop the space in the midline to the apex of the prostate. This will expose the posterior pedicles, which are serially clipped and divided to the apex of the prostate ( Fig. 92-10).

FIG. 92-6. Peritoneal reflection in rectovesical cul-de-sac has been incised and the bladder elevated from the rectum. Tips of seminal vesicles are seen. (Modified from Schlegal PN, Walsh PC. Radical cystoprostatectomy with preservation of sexual function. J Urol 1987;138:1402–1406.)

FIG. 92-7. Demonstration of the avascular plane behind the ureter with definition of lateral vascular pedicles to the bladder.

FIG. 92-8. An avascular plane exists posterior to the ureter. With either a finger or a Kelly clamp passed behind the ureter, this avascular plane can be developed. The superior and middle vesicle pedicles can be controlled either with surgical clips or with suture ligature.

FIG. 92-9. After division of the superior and middle vascular pedicles, the peritoneum in the cul-de-sac can be incised sharply. The transverse line of incision into the

peritoneum should be at the margin of the tips of the seminal vesicles. After the peritoneum is incised, the posterior plane between the rectum and bladder is developed bluntly, allowing identification of the posterior pedicles.

FIG. 92-10. The posterior pedicles on either side are divided between clips.

The specimen should be held in only by the urethra and by the puboprostatic ligaments superiorly. The rectum is protected by the hand of the surgeon behind the bladder and the prostate. With the specimen being elevated from the pelvis by the assistant, the surgeon can sharply divide the puboprostatic ligaments. Further elevation of the specimen brings the membranous urethra from the muscular pelvic floor; it can be sharply divided and the specimen removed. A warm, moist pack should be placed in the pelvis and held in place with a curved Deaver retractor for about 5 minutes. During this time the area of dissection lateral to the rectum and along the hypogastric vasculature should be inspected to ensure that adequate hemostasis has been established. The pack should be removed and the veins that accompany the urethra identified. The assistant can assist in visualization by placing a closed fist in the perineum and pushing inward. This elevates the pelvic floor and permits the surgeon to identify any venous bleeding points, which then can be controlled by figure-of-8 sutures. In those instances in which diffuse venous bleeding is encountered and it is difficult to establish hemostasis, a small pack of absorbable gelfoam can be placed in the midline and held firmly in place behind heavy chromic catgut sutures that have been passed into the muscular pelvic floor lateral to the urethra. Radical Cystectomy in Females Females are placed in a position similar to that of males, and a similar line of incision into the abdomen is established. The anterior, inferior, and posterior peritoneal flaps are established as in males; in females, however, as this line of incision into the peritoneum is established, both round ligaments are encountered and should be divided. The tubes and ovaries should be carried with the specimen in the midline. The broad ligament should be incised, and the ovarian vasculature is divided and controlled with surgical clips. Lymphadenectomy is accomplished as in the male. After completion of the lymphadenectomy, the vascular supply of the bladder, the uterus, and the vagina can be identified. Vascular control can be established between surgical clips or absorbable sutures. The uterus is carried with the specimen, and the posterior peritoneal flap is developed across the base of the uterus at the level of the cervix. Attention then should be turned to release of the bladder and the vagina. The vagina and the bladder are bound down around the rectum laterally by dense connective tissue, which must be controlled with either absorbable sutures or metallic clamps, and divided lateral to the rectum as dissection proceeds. The dissection should be carried from proximal to distal. Such a maneuver releases the tissue from the posterior pelvic wall and rectum, and dissection can be continued to the level of the bladder neck. The assistant then can elevate the specimen by traction off the Kelly clamp, which has been placed across the urachus. The posterior portion of the vagina should be entered at the level of the cervix and the specimen is released by incising the anterior third of the vagina, moving from proximal to distal. As the bladder neck is reached, the incision in the vagina is moved more to the midline and carried parallel to the urethra and to the level of the urethral meatus. The urethra then can be released from either below or above, and the specimen is removed. Extensive bleeding may be encountered from the line of incision in the anterior third of the vagina. This can be controlled by grasping the vaginal margin between Allis clamps. After removal of the specimen, the vagina should be closed by moving from distal to proximal using absorbable sutures, noting that a locking suture provides adequate vascular control as well as vaginal closure. Completion of the Cystectomy In both males and females, after removal of the specimen and after confirmation that hemostasis is secure, attention can be turned to the urinary diversion. After urinary diversion, the surgeon is ready to close the abdomen. No attempt is made to reperitonealize the pelvic floor. The omentum should be brought into the pelvis and used to line the pelvic floor. Mobilization of the omentum may be necessary to accomplish this. This is desirable because the omentum functions as an internal drain and also protects against adhesion and fixation of the bowel deep in the pelvis. Vagina/Uterus Tumors arising in the vagina or uterus are managed in a similar fashion to those arising in the prostate and bladder neck with initial biopsy followed by chemotherapy. The response of vaginal tumors can be easily assessed by endoscopic vaginal exams and biopsy. Residual disease is managed by local resection including partial vaginectomy or hysterectomy/vaginectomy, depending on its extent. Paratesticular Radical inguinal orchiectomy is the initial operation for almost all patients with paratesticular RMS. If there has been a prior transcrotal orchiectomy, inguinal exploration with removal of the remaining spermatic cord and hemiscrotectomy are recommended. Until recently, a modified unilateral retroperitoneal lymph node dissection (RPLND) was recommended by the IRS in all patients with paratesticular RMS. This recommendation was based on a reported incidence of lymph node metastases of 40% in earlier IRS studies. 10 This recommendation has been modified based on SIOP experiences.11 Patients whose tumors were completely excised and had no clinical evidence of nodal metastasis were treated with chemotherapy alone. Their survival is comparable to similar patients in IRS studies treated with chemotherapy and RPLND. Current recommendations of both IRS and SIOP investigators is that patients with low-stage, completely excised paratesticular RMS do not need RPLND 14. Patients with clinical evidence of retroperitoneal lymph node involvement or a positive margin on the surgical specimen should have a modified unilateral retroperitoneal lymph node dissection.

OUTCOMES
Complications Complications of multimodal therapy for RMS are specific to the treatment utilized. The side effects of vincristine actinomycin D cytoxan (VAC) chemotherapy include bone marrow suppression (85% during induction dropped the absolute neutrophil count to less than 500/mm 3), gram-negative sepsis (11%), alopecia, gastrointestinal disturbances, venoocclusive disease of the liver (1.2%), stomatitis (24%), and death (7%). In the IRS-4 pilot study of etoposide, ifosfamide, and vincristine, the complications include (with 62 evaluable patients) life-threatening neutropenia seen in 55 of 60 with life-threatening infections in 27 of 60 and 3 fatal toxicities due to infection. Twenty-five patients (42%) developed neurotoxicity from vincristine. Eleven patients (18%) developed nephrotoxicity; of these, 7 cases were severe and over half were in patients younger than 2 years old. 1 Complications of surgery are dependent on the surgery performed but include hemorrhage and infection in the

perioperative period, bowel obstruction (10%), lower extremity edema (5%), and ejaculatory disturbances (8%) in patients undergoing retroperitoneal node dissection for paratesticular RMS. Results With the advent of the combined multimodal therapy for rhabdomyosarcomas of children, the results have been gratifying in terms of both the response and remission rates as well as the reduction in the long-term morbidity associated with the radical excisions of the past. This is mainly testimony of a well-run large-scale clinical trial that incorporates multiple disciplines and allows difficult questions to be resolved in a relatively uncommon tumor. An example of this is the evaluation of second-look operations (SLOs) that were done on patients in IRS-3. 13 The SLO changed the response status in a significant number of patients and improved survival in selected patients. Twelve percent of patients presumed to have a CR by clinical and radiographic evaluation were found to have residual disease and 74% of patients with no-remission (NR) or partial remission (PR) status were recategorized as CR after operation (biopsy and/or resection). Overall, the RMS trials have allowed identification of risk factors such as histology, site of disease, and extent of disease as markers of prognosis and have made significant impacts on the therapeutic algorithm. Included in these are the more common use of local versus radical surgery, the reduction in the need for nodal dissections, reduced need for radiation therapy in certain groups, increased numbers of patients who retain their native bladders after therapy (25% versus 60%), and the use of second-look surgery, which has enhanced the overall response and survival. The goal of treatment of patients with RMS, as with other tumors, is to achieve the highest cure rates with the lowest possible treatment-related morbidity and mortality. This requires an artful as well as fact-based blend of current treatment modalities that will continually be refined by data derived from continuing studies and the introduction of new therapies. Current treatment is derived from the efficacy of each modality balanced by its known toxicity. Much is known of the short-term toxicity of each modality, but much less is known of their long-term toxicities. CHAPTER REFERENCES
1. Arndt C, Tefft M, Gehan E, et al. A feasibility, toxicity, and early response study of etoposide, ifosfamide, and vincristine for the treatment of children with rhabdomyosarcoma: a report from the Intergroup Rhabdomyosarcoma Study (IRS) IV pilot study. J Pediatr Hematol Oncol 1997 Mar;19(2):124–129. 2. Asmar L, Gehan EA, Newton WA. Agreement among and within groups of pathologists in the classification of rhabdomyosarcoma and related childhood sarcomas: report of an international study of four pathology classifications. Cancer 1994:2579–2588. 3. Crist W, Gehan EA, Ragab AH, et al. The third Intergroup Rhabdomyosarcoma Study. J Clin Oncol 1995;13:610–630. 4. Flamant F, Hill H. The improvement in survival associated with combined chemotherapy in childhood rhabdomyosarcoma: a historical comparison of 345 patients in the same center. Cancer 1984;53:2417–2421. 5. Horn RC, Enterline HT. Rhabdomyosarcoma: a clinicopathologic study and classification of 39 cases. Cancer 1958;11:181–199. 6. Lawrence WJ, Gehan EA, Hays DM, et al. Prognostic significance of staging factors of the UICC staging system in childhood rhabdomyosarcoma: a report from the Intergroup Rhabdomyosarcoma Study (IRS-3). J Clin Oncol 1987;5:46–54. 7. Parham DM. Immunohistochemistry of childhood sarcomas: old and new markers. Mod Pathol 1993;6:133–138. 8. Parham DM, Webber MB, Holt H, et al. Immunohistochemical study of childhood rhabdomyosarcoma study project. Cancer 1991;67:3072–3080. 9. Raney RB, Hayes DM, Tefft M, et al. Rhabdomyosarcoma and the undifferentiated sarcomas. In: Pizzo PA, Poplack DG, eds. Principles and practice of pediatric oncology. Philadelphia: JB Lippincott, 1993;769–794. 10. Raney RB, Tefft M, Lawrence W, et al. Paratesticular sarcoma in childhood and adolescence; a report from the Intergroup Rhabdomyosarcoma Studies 1 and 2, 1973–1983. Cancer 1987;60:2337–2343. 11. Rodary C, Flamant F, Maurer LT, et al. Initial lymphadenectomy is not necessary in localized and completely resected paratesticular rhabdomyosarcoma. Med Pediatr Oncol 1992;20:430. 12. Sutow WW, Sullivan MP, Ried HL, et al. Prognosis in childhood rhabdomyosarcoma. Cancer 1970;25:1384–1390. 13. Wiener ES, Lawrence W, Hayes D, et al. Survival is improved in clinical group III children with complete response established by second-look operations in the Intergroup Rhabdomyosarcoma Study (IRS-3). Med Pediatr Oncol 1991;19:399. 14. Wiener ES, Lawrence W, Hays DM, et al. Retroperitoneal node biopsy in childhood rhabdomyosarcoma. J Pediatr Surg 1994;29:171–178.

Chapter 93 Vesicoureteral Reflux Glenn’s Urologic Surgery

Chapter 93 Vesicoureteral Reflux
Anthony Atala and R. Dixon Walker

A. Atala: Department of Surgery, Harvard Medical School, and Department of Urology, Children's Hospital, Boston, Massachusetts 02115. R. D. Walker: Department of Surgery, Division of Urology, University of Florida College of Medicine, Gainesville, Florida 32610.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Preparation Position and Prepping of the Patient Surgical Exposure of the Bladder and Preparation for Reimplantation Extravesical Ureteral Reimplantation Combined Intravesical and Extravesical Repair Intravesical Reimplant of the Ureter Duplicated Ureters Endoscopic Treatment of Reflux Laparoscopic Management Postoperative Evaluation Outcomes Complications Results Chapter References

Primary reflux is a congenital anomaly of the ureterovesical junction whereby a deficiency of the longitudinal muscle of the intravesical ureter results in an inadequate valvular mechanism, permitting the retrograde flow of urine from the bladder to the upper urinary tract. The factor most critical to a competent ureterovesical junction is the length of submucosal ureter relative to its diameter. Large-caliber ureters or those having short intravesical segments cannot be effectively shut by the junction's valvular mechanisms. A 5:1 tunnel length/ureteral diameter ratio has been determined to be necessary for children to have a normal ureterovesical junction, without reflux. 8 The interplay between the ureteral bud and renal blastema during the first few weeks of gestation determines the development of the uretero- vesical junction. A ureteral bud that is laterally (cranially) positioned from a normal trigonal takeoff offers an embryologic explanation for primary reflux. 6 Secondary reflux is that caused by bladder obstruction and its consequent elevated pressures. The obstructions may be anatomic (e.g., posterior urethral valves) or functional (e.g., neurogenic bladder), though the outcomes can be the same. Functional causes are more common in both sexes. The overall incidence of primary reflux in normal children has been observed to be as high as 18.5%. Reflux is found in up to 70% of infants who present with urinary tract infections (UTIs). Reflux is also a common finding in fetuses with antenatally diagnosed hydronephrosis. Although the vast majority of reflux occurs in females, males who present with urinary infection have a higher likelihood of having the anomaly. The detection of reflux is also influenced by the age of the child being evaluated, with the highest incidence found in younger children. An explanation for this finding has as its basis the spontaneous resolution of reflux that occurs in many children with interval growth of the bladder and elongation of ureteral tunnel length. Race also plays a role in the overall incidence of reflux. Caucasian girls are 10 times more likely to have reflux than their black American counterparts. There is an increased risk of reflux in fair-skinned children with blue eyes and blond or red hair. Like most genitourinary anomalies, vesicoureteral reflux appears to have a multifactorial etiology, although a genetic component undoubtedly exists. Siblings of patients with reflux are at much greater risk of having reflux than the normal population. Up to 45% of siblings have been noted to have reflux. Children from parents with a history of reflux are also at a much greater risk of having reflux than the normal population. Up to 66% of offspring have been noted to have reflux. While a polygenic mode of transmission is favored by some investigators, these types of data suggest a dominant inheritance pattern having variable penetrance. The high incidence of sibling reflux amplifies the need for some form of screening. Standard or radionuclide cystography is indicated as a screening test in babies and young children. In older children, urinary ultrasonography should be obtained. The appearance of upper tract scarring or hydronephrosis warrants further evaluation.

DIAGNOSIS
Most patients with reflux present initially with some symptom(s) that suggest a UTI, although newborns are typically nonspecific in this regard. Failure to thrive and lethargy are worrisome signs. Infants and young children may present with fever, malodorous urine, dysuria and urinary frequency, lethargy, and gastrointestinal symptoms including nausea, vomiting, and vague abdominal discomfort. A urine culture should be included in the evaluation of any infant or child who presents with fever or malaise. Vesicoureteral reflux will be found in 29% to 50% of children with UTI. 5 Approximately 30% of these already have some evidence of renal parenchymal scarring that is usually proportional to the severity of reflux. A diagnostic workup is recommended after the first urinary infection and is tailored to the individual according to age, gender, and clinical history. The presence of vesicoureteral reflux can be established with either fluoroscopic or radionuclide voiding cystography. Complete evaluations that include voiding cysto- urethrography (VCUG) and ultrasonography are required of three groups: any child under the age of 5 years with a valid documented UTI; children with a febrile UTI, regardless of age; and any boy with a UTI unless he is sexually active or has a past urologic history. Older children who present with asymptomatic bacteriuria or UTIs that manifest solely with lower tract symptoms can be initially screened with ultrasonography alone, reserving cystography for those with abnormal upper tracts or recalcitrant infections. Newborns with moderate to severe degrees of upper tract dilation should be fully evaluated with VCUG, as should any baby with intermittent hydronephrosis. The International Classification System devised by the International Reflux Study provides the current standard for grading reflux and is based on the appearance of contrast in the ureter and upper collecting system during VCUG ( Fig. 93-1).5

FIG. 93-1. International Classification of vesicoureteral reflux. Grade I, into the nondilated ureter; grade II, into the pelvis and calyces without dilation; grade III, mild to moderate dilation of the ureter, renal pelvis, and calyces with minimal blunting of the fornices; grade IV, moderate ureteral tortuosity and dilation of the pelvis and calyces; grade V, gross dilation of the ureter, pelvis, and calyces, and loss of papillary impressions and ureteral tortuosity.

Radionuclide cystography (RNC) is the scintigraphic equivalent of conventional cystography. While the technique does not provide the anatomic detail of fluoroscopic studies, it is a more accurate method for detecting and following reflux using a markedly decreased radiation dose than a standard VCUG. Indirect cystography utilizes intravenously injected technetium-99m ( 99mTc). This agent is cleared by glomerular filtration and is excreted into the bladder within 20 minutes. The child can then be scanned for reflux, thus avoiding catheterization. Unfortunately, the method has a high rate of false-negative studies. Ultrasonic cystography offers an ideal screening tool for reflux because there is no exposure to radiation. Sonicated human serum albumin, composed of 3–5 × 10 8 air-fired echogenic spheres per ml, can be detected by ultrasound and has been successfully used to detect reflux in animals and humans during preliminary studies. Until further studies are completed the modality must be considered a research tool. Ultrasonography has replaced the excretory urogram (intravenous pyelogram, IVP) as the diagnostic study of choice to initially evaluate the upper urinary tracts of patients with suspected or proven vesicoureteral reflux. Ultrasound alone cannot effectively rule out reflux. Renal scintigraphy using 99mTc-labeled dimercaptosuccinic acid (DMSA) is the best study to detect pyelonephritis and the cortical renal scarring that is sometimes present. Pyelonephritis impairs tubular uptake and causes areas of photon deficiency in the cortical outline. High-resolution single-photon emission computed tomography (SPECT), which involves 360-degree imaging and computer reconstruction, improves the ability to identify cortical defects. Ureteral orifice configuration has been found to be of little value in predicting the presence of reflux or prognosticating the likelihood of its spontaneous resolution. Therefore, cystoscopy has a limited role in the diagnosis of reflux. Urodynamics are indicated in any child suspected of having a secondary cause for reflux (valves, neurogenic bladder, nonneurogenic neurogenic bladder, etc.) and help direct therapy.

INDICATIONS FOR SURGERY
Both medical and surgical treatment bring attendant risks and benefits that must be weighed in light of the natural history of the anomaly and its implications for the patient. Many studies have attempted to understand certain aspects of the natural history of reflux in order to better define the parameters for treatment. The International Reflux Study in Children compared the medical and surgical management of high-grade (III–IV) reflux after randomization in children under 9 years of age in Europe and the United States. It was noted that surgery was more effective than medical therapy in preventing pyelonephritis but the overall incidence of urinary infections between the two was the same. Both modalities were equally effective in preventing new renal scarring, although scars did develop regardless of therapy. Cessation of reflux was seen in 34 of 151 (23%) of patients medically managed in the European arm, although children with bilateral reflux were less likely to resolve (10%). The Birmingham Reflux Study used a similar study design in treating 104 patients with severe reflux. Surgical and medical management were equally effective in preventing new renal scars. Between 50% and 80% of medically treated patients still had reflux after a 5-year period. 2 The prevention of renal scarring is the most important goal in the management of reflux. The early work of Hodson et al., using a minipig model of reflux with obstruction of the bladder outlet, implicated sterile reflux and a high-pressure “water hammer” effect as a significant cause of renal scarring. Additional significance is given to sterile reflux by the realization that occasional patients with scars and reflux have no prior history of infection. The pathophysiology of renal scarring was better defined by Ransley and Risdon, who were unable to create scars in a similar model to Hodson's if the bladder was left unobstructed and the urine kept sterile. The bulk of clinical and experimental studies underscore the important role of infection in the evolution of most renal scars. Although pyelonephritis can commonly occur in the absence of vesicoureteral reflux, reflux predisposes the kidney to ascending infections and may amplify the invasive effects of pathogens. There is a direct relationship between the grade of reflux and incidence of scarring. 11 Other factors also determine the severity of renal injury. These include the age of the patient, anatomic considerations, bacterial virulence factors, host susceptibility, and the inflammatory response to parenchymal infection. Early diagnosis and effective antibiotic treatment are instrumental to reducing the inflammatory response to acute pyelonephritis and decreasing the incidence and degree of renal scarring. Equally important is the early diagnosis of reflux with immediate initiation of prophylaxis. Papillary configuration also plays a role in protecting the renal parenchyma from urinary pathogens. Compound or concave papillae, which are flattened, are more likely to allow intrarenal reflux than simple or convex papillae. The former are mostly localized in the polar regions of the kidney, where reflux nephropathy is much more likely to occur. There are several more premises that anchor the decision making in the management of reflux. Most low grades (I–II) of reflux resolve spontaneously. Few high grades (IV–V) of reflux resolve spontaneously. A more aggressive surgical stance has usually been taken with the more severe grades of reflux. The results of the International Reflux Study appear to support that bias. In the European arm of the study, 82 patients were followed with bilateral grade III or IV reflux. Cessation of reflux occurred in only 7 (9%). Children with unilateral reflux alone had a more optimistic prognosis, with reflux abating in 23 of 38 (61%). The results from the 41 patients entered in the American arm of the study were similarly discouraging, with reflux persisting in 75% of patients. 11 The age of the patient also assumes importance in any discussion of the spontaneous resolution of reflux. Younger children are more likely to have reflux and also appear to be more likely to have it resolve spontaneously. If resolution is to occur, it usually does so within the first few years after diagnosis, most likely due to intervals of significant growth and beneficial urodynamic changes. The absence of improvement in the degree of reflux on serial cystograms is a concerning sign that resolution may not occur with interval growth. Resolution of reflux most often occurs before 5 years of age and is less likely to resolve thereafter. If reflux persists into puberty, other pathophysiologic factors become important, regardless of the grade. Women with a history of reflux and renal scarring have increased morbidity during pregnancy because of infection-related complications. The morbidity during pregnancy of women with persistent reflux without renal scarring remains poorly defined, but the tendency to urinary infection seems increased. Failure of renal growth, new renal scars, or deterioration of renal function on serial ultrasounds and/or scans should prompt surgical intervention. Finally, grade V reflux is unlikely to resolve spontaneously. Surgery is recommended after infancy, although a period of observation seems reasonable for perinatally diagnosed disease. The onset of puberty and cessation of longitudinal growth will alter the recommendations for adolescents whose reflux is being medically managed. Surgery is recommended for most girls with persistent reflux in order to avoid the implications of active reflux for future pregnancies. This is especially true if they have nephropathy or have shown a tendency to urinary infections in the past. Before recommending surgery for the prepubertal child, consideration is given to the severity of the reflux; possible underlying risk factors, including bladder dysfunction, age at presentation, and duration of the disorder; and the presence and quality of UTIs that may have occurred during medical management. Clinicians must gauge parental compliance with medical management and periodic radiologic follow-up, factor socioeconomic concerns, and weigh their own personal experience with the anomaly. The likelihood of resolution of reflux and expectations of medical management should be frankly discussed with the parents. The requirements of periodic radiologic follow-up and, in some cases, long-term prophylaxis cannot be fulfilled by certain families. If compliance is an issue, some children are best served by surgical correction of their reflux. Also, patients with breakthrough UTI despite prophylactic antibiotics and those with associated congenital abnormalities at the ureterovesical junction (e.g., bladder diverticula) should have surgical intervention. The expectations of surgery must be understood beforehand. Successful ureteral reimplantation should decrease but does not eliminate the incidence of pyelonephritis in children with reflux.

ALTERNATIVE THERAPY
Medical management consists of continuous low-dose prophylactic antibiotics until the expected surgical intervention or resolution of reflux occurs. Nighttime dosing in toilet-trained children is most effective because it precedes the longest period of urinary retention when infection is most likely to develop. Treatment of vesicoureteral reflux should be individualized. Some generalized recommendations can be made based on the likelihood of grade-related spontaneous resolution and the natural history of reflux. 10 Medical management is initially recommended for prepubertal children with grade I–III reflux under the assumption that most will resolve. A period of observation and medical therapy also seems warranted for most grade IV reflux, especially in younger children and those with unilateral

disease. Some trend to improvement should become evident within 2 or 3 years.

SURGICAL TECHNIQUE
A variety of techniques has been described for the correction of vesicoureteral reflux. These are anatomically categorized as extravesical, intravesical, or combined, depending on the approach to the ureter and suprahiatal or infrahiatal, in description of the position of the new submucosal tunnel in relation to the original hiatus. Common to each is the creation of a valvular mechanism that enables ureteral compression with bladder filling and contraction, reenacting normal anatomy and function. A successful ureteroneocystostomy provides a submucosal tunnel for reimplantation having sufficient length and adequate muscular backing. A tunnel length of 5 times the ureteral diameter is cited as necessary for eliminating reflux. 8 Deviation from this basic principle is the most common cause of failed reimplants and explains the lack of success seen with many earlier reimplantation techniques no longer employed. Preoperative Preparation A urine culture should ideally be done 4 to 6 weeks prior to surgery to ensure sterile urine and repeated at the time of the preoperative evaluation. At the time of surgery, the patient is usually started on intravenous Cefazol, which is continued as long as the intravenous line is in place. After the intravenous line is discontinued (48–72 hours postoperatively), the patient is started on oral cephalexin for 2 weeks and then converted to the same antibacterial that was used prior to surgery. Patients are maintained on prophylactic antibacterials until the postoperative voiding cystogram demonstrates resolution of reflux. Position and Prepping of the Patient The patient is placed in a supine position with legs slightly spread so that the external genitalia can be easily prepped and there is access to the urethra. The patient is prepared with appropriate antibacterial soap from the umbilicus to midthigh and draped so that there is exposure to the lower abdomen with access to the genitalia. Surgical Exposure of the Bladder and Preparation for Reimplantation Anesthesia with appropriate muscle relaxation is necessary for the duration of the case, as inadequate muscle relaxation of the patient, with the retractor in place, can tear the peritoneum, damage the bladder, and makes exposure more difficult. The use of nitrous as an anesthetic agent should be discouraged as it may dissipate into the bowel, thus interfering with surgical exposure. A transverse skin incision is made, curved upward on the end, about 1 cm above the symphysis pubis. The rectus fascia is divided in the line of the incision and freed from the underlying rectus muscle. The rectus muscle is divided in the midline as is the pyramidalis muscle so as to expose the bladder and prevesical space. The peritoneum is swept off the dome of the bladder by blunt dissection. If an extravesical approach is to be made, the lateral perivesical space is developed by blunt dissection, with care taken to sweep the peritoneum off the rectus muscle prior to placement of the retractor. We prefer the Dennis-Browne retractor with two lateral “scoop” or Richardson-type blades. Further extravesical dissection is individualized depending on whether one is doing a Lich-Gregoir or Paquin repair. If an intravesical repair is to be performed, the lateral perivesical space is only minimally developed by blunt dissection. Two traction sutures of 3-0 chromic or polyglycolic acid are placed in the bladder, and the bladder opened in the midline. Placement of the Dennis-Browne retractor with two scoop blades allows excellent intravesical exposure. Exposure of the trigone is facilitated by a Deaver-type blade with an underlying sponge placed at the dome of the bladder and countertraction with a scoop blade under the pubis. Extravesical Ureteral Reimplantation The extravesical reimplant was developed concurrently in Europe by Gregoir and in the United States by Lich, yet for decades this procedure never achieved the widespread popularity in this country that it enjoyed overseas. Since the bladder is left intact, risks from urinary contamination are minimized. Other advantages are that bladder spasms and hematuria are lessened because the trigone is minimally disturbed and hospitalization is shortened. The disadvantages include concerns about disrupting the innervation of the bladder, which has led some surgeons to defer the technique in patients with bilateral reimplants for risk of causing urinary retention. In addition, because the bladder is not opened, associated pathology might be difficult to appreciate if cystoscopy were not done, and the approach causes a relative inability to easily repair associated abnormali- ties (such as a bladder diverticulum) and surgical correction with small-capacity bladders. In performance of this procedure ( Fig. 93-2), the patient is placed in a supine position with a Foley catheter in the bladder. The catheter should be placed after the patient is prepped and draped so that access to the urethra can be maintained. Standard exposure of the prevesical space is obtained with a transverse incision, and with blunt finger dissection the lateral perivesical space is developed on each side. Development of this space allows one to place the Dennis-Browne retractor with two lateral scoop or Richardson blades. One must be careful in placement of these blades that the peritoneum is not torn. Proper placement allows one to easily identify the obliterated hypogastric vessel coming from the lateral pelvic wall across the dome of the bladder to the umbilicus. This vessel is frequently patent and should be suture-ligated with 3-0 silk. After ligation, dissection medial to the pelvic portion of the obliterated hypogastric will usually readily identify the ureter. The ureter should be gently freed and isolated with a vessel loop.

FIG. 93-2. Lich-Gregoir procedure.

Careful dissection toward the bladder with retraction of the bladder medially with a Deaver retractor and a moist sponge allows one to preserve the major vascular pedicles. The bladder muscular wall, in the cephalad line of the ureter, is then divided down to the mucosa for a distance of 4 to 5 cm ( Fig. 93-2A). Any small tears in the mucosa can be repaired with fine, adsorbable sutures such as 6-0 chromic or polyglycolic acid. In the classic Lich-Gregoir reimplant, the ureter is laid in the newly prepared muscular bed and the muscle approximated over the ureter with interrupted 2-0 to 3-0 chromic sutures. We like to secure the repair by placing an adsorbable suture of 4-0 chromic or 5-0 polyglycolic acid through the adventitia of the distal ureteral wall to the muscle of the trigone ( Fig. 93-2B). After reapproximation of the muscle over the ureter ( Fig. 93-2C), the perivesical space is drained with a Penrose or Jackson-Pratt. This drain is removed 1 to 2 days postoperatively. Ureteral stenting is not required, and the Foley catheter may be removed within 24 to 72 hours. Combined Intravesical and Extravesical Repair The nipple technique of preventing reflux was one of the first intravesical-extravesical repairs but was generally unsuccessful because over a period of time elevated bladder pressures with voiding reduced the size of the nipple. The Paquin repair is a combined intravesical and extravesical repair that has been used frequently. 8 It is particularly useful when the ureter needs to be tailored with or without a psoas hitch. The exposure of the bladder is similar to that used for the Lich-Gregoir procedure. The patient is not catheterized preoperatively, and opening the bladder in the midline with traction sutures of 3-0 chromic or polyglycolic acid on the edges permits evacuation of the bladder. This maneuver then facilitates identification, division, and ligation of the obliterated umbilical artery. Isolation of the ureter with a vessel loop and traction thereof will allow dissection

cephalad and caudad of the ureter, taking care to preserve ureteral mesentery. The distal ureter is clamped with a right angle clamp, divided, and the stump suture-ligated at the level of the bladder with a 2-0 chromic or polyglycolic acid suture. A traction suture can be placed on the distal ureter to further facilitate mobilization up to the level of the iliac vessels. If the ureter is to be retailored, the dissection may be carried above the vessels. If a psoas hitch is to be performed, extensive retroperitoneal dissection of the lower quadrant should be done. In males, this can be performed by isolating the testicular vessels with a vessel loop and mobilizing the retroperitoneum above and below the cord. In a female, the round ligament with associated peritoneal processus can be suture-ligated as it enters the internal inguinal ring, and this allows excellent visualization of the lower quadrant. In preparation for the intravesical part of the procedure, the peritoneum must be dissected off the posterior lateral bladder wall such that the hiatus will be located in a line with the ureter and in the most posterior portion of the bladder base. It is vital that the hiatus not traverse the peritoneal cavity and that it not be located on the side wall of the bladder. If a psoas hitch is to be done, it is preferable to do it at this stage (Fig. 93-3). This is done by placing 3 to 4 interrupted 2-0 chromic or polyglycolic acid sutures from the adventitia and muscle of the bladder wall (just lateral to the site of the new hiatus) to the tendon of the psoas muscle. Tying these interrupted sutures with a surgeon's knot is facilitated by a finger in the bladder displacing the bladder to the psoas muscle. The intravesical portion of the procedure is initiated by analyzing the interior of the bladder and the anticipated tunnel in relationship to ureteral diameter. The diameter of the ureter is measured with ophthalmic calipers and a tunnel planned that is approximately 5 times the diameter of the ureter. The tunnel is made by dissecting from the new hiatus submucosally to the point of the new orifice ( Fig. 93-3A). This dissection can be accomplished with Metzenbaum scissors or a long clamp. The new orifice is created with a small transverse incision. Following placement of a clamp through the hiatus, tunnel, and new orifice, a red Robinson catheter placed inside the ureter is grasped and brought through the tunnel with the top suture ligated to the distal ureter ( Fig. 93-3B). The ureter can then be brought through the tunnel with care taken not to twist it, and the catheter is used for traction, decreasing the necessity for handling the ureter with forceps. The ureter is divided as interrupted sutures are placed through the ureteral wall, bladder muscle, and mucosa ( Fig. 93-3B inset). A small polyethylene feeding tube should pass easily to the kidney and may be left in place as a stent at the discretion of the surgeon, exiting through a stab wound of the bladder and lower quadrant. Drainage of the bladder can be accomplished with a Foley catheter, although a suprapubic tube using a 16- to 18-Fr Malecot catheter may be preferable. Prior to closing the bladder, a suture is placed outside the bladder from the muscular hiatus to the adventitia of the ureter.

FIG. 93-3. Paquin repair.

The bladder is closed in two layers: a running 4-0 polyglycolic acid suture on the mucosa, and a running 3-0 polyglycolic acid suture on the muscle and adventitia. The rectus muscle and the fascia are closed separately with a 3-0 adsorbable suture. The subcutaneous tissue is closed with a 4-0 plain cat gut suture and the skin with 4-0 or 5-0 running subcuticular polyglycolic acid suture. Paper adhesive strips and tincture of benzoin with 4 × 4 gauze constitute the external dressing. Drainage of the perivesical space is accomplished with a Jackson-Pratt or Penrose drain brought out through the lower abdominal wall. This procedure enjoys a high success rate, comparable to that of other intravesical procedures. As mentioned, it lends itself to an associated psoas hitch and ureteral retailoring. It is also a procedure that can be readily used for reimplantation of the transplant ureter in children. Intravesical Reimplant of the Ureter Intravesical reimplantation is currently the most popular way to manage vesicoureteral reflux. The advantage of intravesical reimplantation is that in most hands it is less time consuming and less complicated than extravesical reimplantation. Once the bladder is opened and the Dennis-Browne retractor placed, the initial step in intravesical repair is the same regardless of procedure. This initial mobilization of the ureter is shown in Figure 93-4. The ureter is catheterized with a 3.5- or 5-Fr feeding tube. A figure-of-8 traction suture is passed through the mucosa, with an attempt made to catch a bit of ureteral wall to avoid tearing of the mucosa ( Fig. 93-4B). The incision around the ureter is made with a long-handled knife, maintaining a small cuff of mucosa (Fig. 93-4C). Dissection of the ureter using Metzenbaum scissors should be done by a combination of gentle spreading with minimal cutting (Fig. 93-4D). Caudal to the ureter the scissors can usually be passed in this manner through the thickness of the bladder wall without damaging the ureter. Once through the bladder wall the remaining muscle bundle attachments lateral and cephalad are easier to dissect with an assistant elevating the muscle bundles with forceps. Dissection outside the bladder ( Fig. 93-4E) is entirely blunt with a right angle clamp and Kittner dissector.

FIG. 93-4. Intravesical mobilization of ureter. (A) Low transverse incision. (B) 3.5- or 5.0-Fr polyethylene tube and traction suture. (C) Incision around meatus with mucosal cuff using long-handled knife. (D) Cutting and blunt dissection of muscle of superficial trigone. (E) Extravesical mobilization with right angle clamp and Kittner dissector.

Once the ureter is mobilized, the surgeon has a choice of procedures, some of which make a tunnel cephalad to the hiatus (suprahiatal), or those that make the tunnel lateral or caudad to the hiatus (infrahiatal). The suprahiatal repair most commonly used is the Politano-Lead- better repair, as shown in Figure 93-5 and Figure 93-6.9 The submucosal tunnel is made with Metzenbaum scissors, so that the tunnel length is 4 to 5 times the diameter of the ureter. It is vital that the peritoneum be swept off the ureter because a reported complication involves placement of the ureter through a tongue of peritoneum. Visualization can be enhanced by a vein retractor in the hiatus. The new hiatus is best created by making a transverse incision on top of a right angle or Kittner dissector.

FIG. 93-5. Politano-Leadbetter repair. (A) Mobilization of intravesical ureter. (B) Dissection of peritoneum with Kittner dissector. (C) Development of submucosal tunnel. (D) Creation of the new hiatus.

FIG. 93-6. Politano-Leadbetter repair and details of ureteral reimplant. (A) The ureter is brought through the new hiatus, and the old hiatus is closed with interrupted absorbable suture (B). The ureter is brought through the new tunnel (C), spatulated if necessary (D) and sewn in place with interrupted absorbable suture (E).

There have been a number of infrahiatal repairs developed that had high failure rates because of lack of mobilization of the ureter in the initial step. The repairs that have worked well have combined mobilization of the ureter with good tunnel length. The Glenn-Anderson repair develops a tunnel toward the bladder neck. 3 This concept is shown in Figure 93-7, with the details of surgery in Figure 93-8. Some individuals use a combination of a Politano-Leadbetter suprahiatally and a Glenn-Anderson infrahiatally to make an adequate tunnel length.

FIG. 93-7. Glenn-Anderson repair. (From Walker RW. Vesicoureteral reflux. In: Gillenwater JY, Grayhack JT, Howards SS, Duckett JW Jr., eds. Adult and pediatric urology. Vol. 2. St. Louis: Mosby–Year Book, 1996.)

FIG. 93-8. Glenn Anderson repair—details of ureteral reimplantation. (A) Tunnel made toward bladder neck. Ureter brought through tunnel (B) and sewn in place (C). Alternative, similar to Mathison technique, allows enlargement of hiatus (D) and results in longer tunnel (E, F).

The Cohen technique is a similar modification of the advancement technique that involves a ureteral course across the trigone or bladder, rather than toward the bladder neck (Fig. 93-9). This technique is particularly suited to small bladders and is in wide usage today.

FIG. 93-9. Cohen technique of ureteral reimplantation. Cross-trigonal tunnel, stippled area (A) sewn in place with interrupted sutures (B). (C) Bilateral reimplants easily ac-complished. (From Walker RW. Vesicoureteral reflux. In: Gillenwater JY, Grayhack JT, Howards SS, Duckett JW Jr., eds. Adult and pediatric urology. Vol. 2.

St. Louis: Mosby–Year Book, 1996.)

Duplicated Ureters Approximately 10% of children undergoing antireflux surgery have an element of ureteral duplication. The most common configuration is a complete duplication that results in two separate orifices. This is best managed by preserving a cuff of bladder mucosa that encompasses both orifices ( Fig. 93-10). Since the pair typically share blood supply along their adjoining wall, mobilization as one unit with a common sheath preserves vascularity and minimizes trauma. Reimplantation then follows a sequence similar to that used in any of the techniques described above. In occasional instances a Y or V ureteral configuration results in a single orifice that refluxes into both systems. If the common stem is long enough to be reimplanted alone, standard techniques can be used. Otherwise, to avoid obstruction at the union, the stem should be excised and the ureters converted to a complete duplication. A side-to-side reimplant can then be completed. Finally, in rare instances, an unexpected ureteral duplication is encountered at surgery. These can be associated with functioning renal tissue or end blindly. Small ureteroceles that are unappreciated on preoperative studies have also been described. Common sheath reimplantation rather than excision is recommended.

FIG. 93-10. Reimplantation of duplex ureters.

Endoscopic Treatment of Reflux The endoscopic treatment of vesicoureteral reflux was given its origins by otolaryngologists who treated patients with vocal cord paralysis by injecting their cords with polytetrafluoroethylene (Teflon) paste. In the 1970s, urologists realized the potential applications of this bulking agent and applied the technology as a solution to urinary incontinence. The concept of treating vesicoureteral reflux endoscopically is a valid one because an object or injectable material positioned behind the ureter provides the backing necessary to enable its coaptation during bladder filling and contraction. Success rates vary from 66% to 92% depending on the grade of reflux and number of repeat treatments, which are required of as many as a third of patients. However, Teflon has not fared well in the urinary tract. Teflon particles are phagocytized into the reticuloendothelial system and are able to migrate locally and to distant sites, including the lungs, brains, and lymph nodes of dogs and monkeys after periurethral injection. Concerns of a similar scenario in children have prevented the widespread use of this substance for the treatment of reflux in the United States. Multiple other substances have been proposed as viable alternatives for the endoscopic treatment of reflux. Textured silicone macroparticles suspended in hydrogel have been used clinically; however, like Teflon, particle migration to distant organs after their submucosal injection in the bladder has been noted. Collagen has been used as an injectable soft tissue substitute for years, yet the major flaw of collagen is that its volume decreases with time. Biodegradation explains the high frequency of recurrent reflux and need for retreatment. As opposed to incontinence, reflux is largely a silent disease whose presence can only be determined with invasive radiologic testing. The major problem with any bulking agent that loses its volume is not knowing when reflux might reappear. For these reasons, collagen has not been approved for the treatment of vesicoureteral reflux by the Food and Drug Administration. The Deflux system, which combines biodegradable dextranomer microspheres with sodium hyaluranan, a common polysaccharide, has also been used clinically for the treatment of reflux. Concerns about long- term efficacy, as with collagen, will determine the role of this agent in the treatment of reflux. Studies are in progress for injectable Bioglass, a ceramic composed of oxides of silicone, calcium, and sodium and calcium silicone that is biocompatible and bonds to animal soft tissues, and alginate, a biodegradable polymer, can be seeded with chondrocytes to serve as a synthetic substrate for the injectable delivery and maintenance of cartilage in vivo. Because it is a biodegradable, injectable material that is nonmigratory, nonantigenic, and conserves its volume, the alginate-chondrocyte mixture may be ideal for the treatment of reflux as well as urinary incontinence. Another attempt at endoscopic treatment of reflux has been a detachable, self-sealing silicone balloon developed for the endoscopic treatment of reflux. The balloon is cystoscopically maneuvered into the submucosa beneath the ureter and filled with hydroxyethylmethylacrylate (HEMA) through a catheter, which is then withdrawn leaving the membrane intact. Prior to endoscopic treatment of reflux, all patients are given a broad-spectrum preoperative antibiotic. The patient is placed in the dorsal lithotomy position. Routine cystoscopy is performed and the ureters are visualized. A 19- to 22-gauge needle is advanced through the working channel. The needle tip is inserted under direct visualization at the 6 o'clock position into the subureteric space, approximately 4 to 6 mm distal to the ureteral orifice ( Fig. 93-11). Occasionally, proper placement of the needle may be facilitated by a 3-Fr catheter in the ureter. The needle is then advanced proximally. The material is injected slowly until a bulge nearly obliterates the ureteral orifice. A precise single injection should be made because multiple puncture sites can allow extravasation of the material. The needle is kept in position for 2 to 3 minutes before being withdrawn in order to minimize extravasation of the injected material through the needle track. Clinical studies have shown that such methods can be performed in an outpatient setting, be completed in less than 15 minutes, and have a low morbidity.

FIG. 93-11. Principle of endoscopic treatment of reflux. Bulking agent is injected beneath ureteral orifice with needle. The buttress that is provided helps coapt distal ureter. (From Atala A, Keating M. Vesicoureteral reflux. In: Walsh PC, Retik AB, Wein A, Vaughan ED. eds. Campbell's urology. Vol. 2. Philadelphia: WB Saunders, 1997.)

Laparoscopic Management The advantages of laparoscopy over open surgery include smaller incisions, less discomfort, brief hospitalizations, and quicker convalescence. The drawbacks to a laparoscopic repair of reflux include the fact that it requires a team with at least two surgeons; the repair is converted from an extraperitoneal to intraperitoneal approach; many of the available instruments are less than ideal for use in children; operative time is greater than with open techniques; and cost is increased because of lengthier surgery and the expense of disposable equipment. These considerations explain why, after an initial flurry of interest, laparoscopic reimplantation has not been used widely. It may be that improved instrumentation and more experience will allow better acceptance of the laparoscopic approach to reflux. To perform a laparoscopic correction of reflux, a Verris needle is placed beneath the umbilicus and pneumoperitoneum is obtained by insufflation of CO 2 (rate 2 L/min) up to a pressure of 15 mm Hg. After adequate insufflation, two 10-mm trocars are placed: one in the side opposite the refluxing ureter, at the midclavicular line 1 cm above the umbilicus (for various instruments), and one in the infraumbilical midline (camera). Two 10-mm trocars are then positioned in the left and right midclavicular line, 2 cm above the level of the anterior superior iliac spine (dissecting instruments, refractors). The table is then laterally rotated with the refluxing side up, allowing the bladder and viscera to fall away from the area of repair. Ureteral mobilization is begun by identifying the obliterated umbilical artery along the pelvic side wall ( Fig. 93-12). The bladder is then shifted away from the operative side by grasping and retraction of its dome. This stretches the obliterated umbilical artery, which is traced deep into the pelvis until the ureter is seen passing beneath. After division of the artery, periureteral adventitial tissue is gently grasped to retract the ureter away from the bladder. Blunt dissection is used to mobilize 4 cm of the ureter proximal to the ureterovesical junc- tion (UVJ) to permit placement within a bladder trough. Small vessels are isolated and carefully fulgurated with cautery and the UVJ cleared of any bulky surrounding tissues.

FIG. 93-12. Laparoscopic reimplant. (A) Ureteral mobilization. The obliterated umbilical artery is identified and traced distally until the ureter is seen. The ureter is grasped gently and the periureteral tissue dissected bluntly toward the ureterovesical junction. (B-D) Creation of bladder wall trough. Bladder wall is incised with electrocautery 3 cm proximal to the ureterovesical junction. Muscle fibers are gently cut and spread. Dissection is complete when mucosal tissue bulges outward. (E) After placing the ureter in the trough grasping instruments, the superior aspect of the bladder wall is wrapped around the ureter and a suture placed proximally, immobilizing the ureter in the trough (left). Remaining sutures are placed throughout the length of the tunnel (right). (F) Completed repair. (From Atala A, Keating M. Vesicoureteral reflux. In: Walsh PC, Retik AB, Wein A, Vaughan ED. eds. Campbell's urology. Vol. 2. Philadelphia: WB Saunders, 1997.)

The bladder wall trough is created by incising the detrusor with electrocautery along a line from the UVJ superiorly for approximately 3 cm and inferiorly for 1 cm. Gentle spreading of the muscle fibers opens the trough until mucosa bulges outward. This level of dissection is developed for the entire length of the tunnel. Muscle fibers around the UVJ are also spread slightly, but no attempt is made to extensively dissect the UVJ free of the bladder wall. The ureter is advanced by placing a pexing suture that approximates lateral muscle at the distal end of the trough, distal ureter, and medial bladder muscle tissue at the end of the trough. Once the advancement is complete, detrusor is approximated over the ureter with a continuous polydioxanone suture (PDS). A Lapra-Ty (Ethicon) is extremely useful for this portion of the procedure. This device laparoscopically applies absorbable anchors onto sutures. The anchor secures the suture and acts as a knot. Two grasping instruments are used to wrap the superior aspect of the bladder wall trough around the ureter and a suture is placed at the distalmost point, bringing the edges of the bladder muscle together over the ureter. A Lapra-Ty anchor is placed at the end of the suture. This tie serves to maintain traction on the bladder wall trough as a running suture is initiated from the most distal portion of the trough. Additional Lapra-Ty anchors are placed along the anastomosis in order to keep the suture under appropriate tension during the closure. After the repair is complete, instruments and trocars are removed after checking for trocar site bleeding or visceral injury. The fascial defects and skin of each port are closed with absorbable sutures. A bladder catheter may be kept overnight. Postoperative Evaluation A renal ultrasound is obtained 6 to 12 weeks after surgery and should show good function and minimal ureteral dilation. Contrast or radionuclide cystography 6 months later checks the quality of the repair. Periodic visits with blood pressure measurements, urinalysis, and renal ultrasound at 18 months, 3 years, and 5 years are also recommended.

OUTCOMES
Complications The most common complication after reimplantation surgery is mild to moderate reflux, which can occur in either the index ureter(s) or a contralateral ureter that has been undisturbed. These changes may be secondary to trigonal edema or exacerbations of bladder dysfunction during the perioperative period. Most cases subside within 1 year and conservative management is all that is required. Reflux that persists or is high grade usually results from failure to achieve sufficient submucosal length or tailoring of a dilated ureter. Before considering reoperation, the etiology of the complication should be fully defined so that any error of earlier surgery will not be repeated. Urodynamics and renal pressure perfusion studies are sometimes helpful, especially when a functional cause is suspected. Cystoscopy with retrograde ureterography provides anatomic detail. If this exam shows a good tunnel, then observation or endoscopic treatment may be considered. If there is breakdown of the repair and a poor tunnel, then repeat reimplantation is performed under the same anesthesia in a manner similar to that described for primary ureteroneocystostomy. The technique of ureteral mobilization differs from that in standard reimplantation in terms of incising of the mucosa of the old submucosal tunnel and sharp removal of scarred ureteral attachments. Various degrees of obstruction can be expected of the reimplanted ureter early after surgery and may be due to edema, subtrigonal bleeding, bladder spasms, mucous plugs, or blood clots. Most postoperative obstructions are mild, asymptomatic, and resolve spontaneously. More significant obstructions are usually symptomatic. Renal scintigraphy or IVP usually shows marked delay in excretion, whereas severe hydroureteronephrosis is present on ultrasound. The large majority of perioperative obstructions subside spontaneously but placement of a nephrostomy tube or ureteral stent sometimes becomes necessary for unabating symptoms. Chronic post-reimplant ureteral obstruction varies in location and degree depending on its etiology. Complete obstructions usually occur from ischemia, angulation at the new hiatus, inadvertent passage through the peritoneum or viscera, or compromise at the hiatus or within an inadequately developed submucosal tunnel. Mild obstruction of a stenotic ureter may be treated with balloon dilation and double-J ureteral stenting, but more severe stenosis or obstruction due to displacement of the ureter will require reoperation and repeat reimplantation. Intravesical mobilization is futile with the obstructed ureter because it is usually irreparably damaged within the bladder wall. Virgin ureter is identified at the bifurcation of the iliac vessels. It is then freed as distally as possible and transacted outside the bladder. The intramural segment remains untouched unless it hampers the planned repair. If after proximal mobilization inadequate ureteral length is noted, a psoas hitch can be used. The bladder can be tacked to the psoas tendon above the common iliac

vessels. The technique provides lengths of as much as 8 to 10 cm. Ureters with defects that cannot be bridged by a psoas hitch or those that continue to reflux despite repeated efforts at reimplantation are ideal candidates for transureteroureterostomy (TUU). In order to maximize the success of a TUU, the recipient ureter should be mobilized distally, there should be no angulation of the donor ureter under the sigmoid mesentery, and a tension-free, widely spatulated anastomosis should be created. A combination of psoas hitch and TUU is especially useful in cases of bilateral failures where the feasibility of creating two adequately long tunnels is limited by the anatomy of most bladders. Bladder diverticula may occur at the site of reimplantation or in the area of the anterior bladder wall closure. If unassociated with reflux or obstruction and if the diverticula drain well, then a reasonable option is to simply follow the patient. Diverticula that do not drain well will likely require reoperation. Results The success rate with open ureteral reimplantation is greater than 95%. Regardless of the technique, in the most experienced hands, success rates are 99% when the ureter is of normal size and the bladder of normal thickness. Success rates decrease as the bladder wall becomes more abnormal or if the ureter enlarges. Therefore, success rates are much lower in diseases where these characteristics are most common, such as neurogenic bladder, posterior urethral valves, prune belly syndrome, or megaureter. Success rates with the endoscopic technique vary widely, depending on the technique and agent used. Success rates from 60% to 92% have been reported. Only a few case reports regarding laparoscopic repair have been published; therefore success rates for laparo- scopic repairs are unknown. CHAPTER REFERENCES
1. Atala A, Kavoussi LR, Goldstein DS, et al. Laparoscopic correction of vesicoureteral reflux. J Urol 1993;150:748. 2. Birmingham Reflux Study Group. Prospective trial of operative versus non-operative treatment of severe vesicoureteric reflux in children: 5 years observation. Br Med J (Clin Res Ed) 1987;295:237. 3. Glenn JF, Anderson EE. Technical considerations in distal tunnel ureteral reimplantation. J Urol 1978;119:194. 4. Gregoir W, Van Regermorter GV. Le reflux vesico-ureteral congenital. Urol Int 1964;18:122. 5. International Reflux Study Committee. Medical versus surgical treatment of primary vesicoureteral reflux. Pediatrics 1981;67:392-400. 6. Mackie GG, Stephens FD. Duplex kidneys: a correlation of renal dysplasia with position of the ureteral orifice. J Urol 1975;114:274. 7. O'Donnell B, Puri P. Treatment of vesicoureteric reflux by endoscopic injection of Teflon. Br Med J 1984;289:5–9. 8. Paquin AJ. Ureterovesical anastomosis. The description and evaluation of a technique. J Urol 1959;82:573. 9. Politano VA, Leadbetter WF. An operative technique for the correction of vesicoureteral reflux. J Urol 1968;79:932. 10. Walker RD. Vesicoureteral reflux update: effect of prospective studies on current management. Urology 1994;43:279. 11. Weiss R, Duckett J, Spitzer A, on behalf of the International Reflux Study in Children. Results of a randomized clinical trial of medical vs. surgical management of infants and children with grades III and IV primary vesicoureteral reflux (United States). J Urol 1992;148:1667.

Chapter 94 Ureterocele Glenn’s Urologic Surgery

Chapter 94 Ureterocele
Sami Arap and Amilcar Martins Giron

S. Arap and A. M. Giron: Division of Urology, Hospital das Clínicas, and University of Sao Paulo School of Medicine, Caixa Postal 11-273-9, CEP 05422-970 Sao Paulo, SP, Brazil.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Upper Pole Preservation Endoscopic Incision of the Ureterocele Excision of Ureterocele with Ureteral Reimplantation Ureteropyelostomy Pyelopyelostomy Upper Pole Ablation Ureterocele in Infants and Neonates Ureterocele with Sepsis Loop Ureterostomy Outcomes Complications Results Chapter References

Ureteroceles are congenital cystic dilations of the intravesical ureter found more frequently in female children. They are associated with either the upper segment of a duplex collecting system or with a single renal unit and ureter. The orifice of the ureterocele may be located in the normal position on the trigone (orthotopic ureteroceles), draining either a single ureter or the upper pole ureter of a completely duplicated system where it is frequently stenotic. More commonly, the orifice is ectopic, and it opens in the urethra or bladder neck (ectopic ureterocele). In some patients, the ureterocele may be bilateral. 3,6,9

DIAGNOSIS
The classical presentation of ureteroceles is in the presence of urinary tract infections in the first months or years of life, eventually associated with voiding symptoms. Ureteroceles in infants may present with sepsis, necessitating immediate intervention, resuscitation, and hydration including parenteral antibiotics and drainage of the obstructed unit. On occasion, a ureterocele presents as a palpable abdominal mass, representing a dilated renal unit. A prolapsing ureterocele causes urethral obstruction and sepsis in female children and is the commonest cause of urethral obstruction in girls. In boys, ureteroceles may prolapse into the posterior urethra, causing a distended bladder, and are in differential diagnosis with posterior urethral valves. More recently, antenatal maternal ultrasound has allowed detection of ureteroceles in the neonatal period, when treatment should be initiated to avoid urinary tract infection. Intravenous excretory urography has been the standard imaging modality to demonstrate the presence of ureteroceles but has been progressively displaced by abdominal ultrasound. Typical findings are a duplex collecting system, with upper pole ureterohydronephrosis at the renal level, associated with a cystic intravesical at the bladder base and a dilated retrovesical ureter. Voiding cystourethrography is essential to evaluate this anomaly. It may demonstrate reflux, a filling defect in the urethra compatible with a prolapsed ureterocele, and may also allow evaluation of the vesical wall surrounding the ureterocele. 10 Radioisotopic studies may be used to evaluate differential renal function between each kidney or each segment of each kidney, yet they are of no value in predicting recovery of renal function.

INDICATIONS FOR SURGERY
The therapeutic alternatives on ureteroceles are as varied as the presentation of this anomaly. The goals of surgical treatment are clear and well defined, i.e., preservation of the renal function through the establishment of an effective urinary drainage and prevention or correction of the ipsi and/or contralateral vesicoureteral reflux. The main factors that will guide the therapeutic option are as follows: 2 Patient's age Presence of reflux and urinary tract infection Function of the several renal segments Grade of ureteral dilation Ureterocele size Patient's clinical condition

ALTERNATIVE THERAPY
Among the several alternatives of surgical treatment, no single one is adequate to all cases, and the therapeutic option must be individualized for each patient ( Table 94-1). For example, in special circumstances, a small ureterocele with a moderate dilation of the upper urinary tract may be followed conservatively. At the opposite end of the spectrum would be the nonfunctioning renal segment that drains into the ureterocele, This makes the first priority the preservation of the function of the lower renal pole moiety. In this setting, the classical and conventional approach consists of resection of the affected upper pole, its entire ureter including the ureterocele, correcting the reflux and reconstructing the vesical base, thus performing a total urinary tract reconstruction. This therapeutic strategy is associated with a long surgical procedure and occasional postoperative complications due to its complexity in young children. This has been gradually replaced as new concepts and alternative methods of treatment have appeared, aiming at reducing the surgical procedures and morbidity. In turn, it has resulted in a decrease in the necessity of surgery in terms of both extent and number of procedures.

TABLE 94-1. Ureterocele therapeutic options

SURGICAL TECHNIQUE
In the patient presenting with a symptomatic ureterocele, the first goal is to attempt to preserve the upper pole moiety. If this is not a possibility, then resection of the upper pole is the next step. Upper Pole Preservation Classically, only one-third of patients presenting with ureteroceles have any remaining renal function in the upper pole of the duplex kidney. For this group, upper pole preservation may be attempted. This situation can be expected to occur more frequently in the future because prenatal ultrasound allows early diagnosis, often before a symptomatic urinary tract infection can occur. The easiest methods of preserving the upper pole moiety are ways of providing retrograde ureterocele decompression. Endoscopic Incision of the Ureterocele Endoscopic incision of the ureterocele has gained support during the past several decades as especially useful for newborn infants. 5 With high-quality optics available in the small pediatric cystoscopes it is possible to endoscopically drain the ureteroceles in almost 90% of cases, including infants. 8 Blyth et al. stressed the variations of the incision technique depending on whether the ureterocele is orthotopic or ectopic. 1 For the intravesical ureterocele, a small (3-Fr) hole should be made inside the bladder, at the ureterocele base. For the ectopic ureterocele, its urethral portion must be drained effectively, usually by a longitudinal incision from the urethra to the ureterocele base, to ensure that bladder neck closure does not occlude this opening. We use a 3-Fr Bugbee electrode, puncturing or incising the ureterocele at its lowest point above its base, on the intravesical portion. Pure cutting current is used until the Bugbee electrode penetrates the ureterocele, draining and decompressing it. When needed, we use a small transverse incision along its base, preserving the roof to act as a flap valve to prevent reflux ( Fig. 94-1). This endoscopic procedure is performed on a one-day hospital stay basis.

FIG. 94-1. (A) Endoscopic incision and puncture of the ureterocele are minimal procedures, using a 3-Fr Bugbee electrode or a small transverse incision at the lowest intravesical part of the ureterocele. The ureterocele roof is preserved to prevent reflux. For the ectopic ureteroceles, a longitudinal incision as far as the intravesical portion is important to effectively drain the system. (B) After drainage, the ureterocele collapses; the lower pole ureter lies over the collapsed ureterocele.

Excision of ureterocele with ureteral reimplantation The indications for open ureterocele excision and common sheath ureteral reimplantation is limited because a large dilation of the upper pole ureter demands tapering of this ureter, raising the possibility of recurrent reflux. Additionally, many avoid this approach because it will compromise the function of the lower pole ureter should reoperation be necessary. Ureteropyelostomy Drainage of the upper pole by an ureteropyelostomy (upper pole ureter anastomosed to the lower pole pelvis) is a lesser procedure than common sheath reimplantation. This approach does not require dissection of the entire ureter and establishes good drainage for the upper pole. The distal upper pole ureter, at the level of iliac arteries, is divided and left in situ, draining into the retroperitoneal space. This procedure allows decompression of the ureterocele by its aspiration and instillation of antibiotic solutions. We normally dissect both ureters through a small lumbar incision, make a 1-cm longitudinal incision in the lower pole pelvis, and anastomose the upper pole ureter to the lower pole pelvis with separate stitches of 6-0 PDS suture. The retroperitoneal space is drained through a Penrose drain. Pyelopyelostomy Pyelopyelostomy is a variation of the ureteropyelostomy that allows simultaneous decompression of the ureterocele and the upper pole moiety. This is an uncommonly performed procedure, and it is similar to ureteropyelostomy. An incision is made in the dependent portion in this pelvis and a corresponding incision is also made in the adjacent pelvis of the lower pole. The edges are anastomosed with a running 6-0 PDS suture, beginning with the posterior face and completing with the anterior face, without the employment of catheter. Upper Pole Ablation In patients with no significant renal function of the upper pole moiety, it is more likely that the procedure of choice would be an upper pole nephrectomy. Due to complications from excision of the entire ureter, a more simplified approach has been described. Simplified Approach Most of the ureteroceles in children are associated with a nonfunctional upper pole in duplex systems and the procedure of choice is upper polar heminephrectomy, removing the ureter by the same lumbar incision. Through a lumbar incision, the kidney is completely freed of its surrounding perinephric fat, remaining attached to its vascular pedicle, and both ureters are identified. The upper pole ureter is then isolated with vessel loop and dissected distally, preserving the blood supply of the lower pole ureter. At the junction of the renal pelvis with the renal parenchyma, renal vessels are seen crossing the pelvis and entering the parenchyma. At this level, the upper pole arterial blood supply is ligated near the kidney, preserving the main renal artery and vein. Demarcation between the nonfunctional upper pole and the remaining kidney is very clear. The renal capsule is incised and dissected to be subsequently sutured over the renal parenchyma. 7 The upper pole is then removed, along with ligation of any identified parenchymal vessels, which may be assisted by the use of an ultrasonic cavitron. Care must be taken to remove all of the calyces. Deep catgut sutures are passed through the renal parenchyma and tied over the preserved renal capsule thereby assuring

complete hemostasis (Fig. 94-2). The upper pole ureter is dissected to the iliac vessels and then divided, leaving the distal stump open. This allows free drainage of the ureterocele, which will then collapse. If there is reflux to this ureter, it can be managed through either ureteral ligation or, eventually, through lower urinary tract reconstruction.

FIG. 94-2. Upper pole remotion with preservation of ureterocele (primary renal approach or simplified approach). (A) Duplex system and ectopic ureterocele, a classic type of clinical presentation. (B) Classic lumbotomy incision; both ureters are identified at the renal pelvis level. The renal vessels of the upper renal pole with hydronephrosis are divided at the parenchyma level. (C) After remotion of the destroyed upper pole with all the calyces, deep sutures with chromic catgut 2-0 are performed for the complete hemostasis. The renal capsule dissected beforehand is used to cover the remaining lower pole. (D) Final aspect of the collapsed intravesical ureterocele after aspiration.

Complete reconstruction Whenever there is high-grade reflux involving the remaining renal units or when the ureterocele is very large, the possibility of additional surgery increases to approximately one-third of cases. In this situation, upper pole heminephrectomy, total ureterectomy, and lower tract reconstruction are indicated. After heminephroureterectomy, a hypogastric incision (Pfannenstiel) is made on the child's abdomen and the bladder is opened. A mucosal incision is performed around the ureterocele and the ureterocele is then carefully dissected from the detrusor and bladder neck fibers, leaving it attached to the ureteral hiatus and to the ureteral meatus (eventually in the posterior urethra), which must then be resected completely. It is very important to avoid leaving any remnant of the ureterocele attached to the urethra, which will eventually function as an obstructing urethral valve. If there is any question of incomplete resection of the ureterocele, we always endoscope the child, resecting any remnants left behind. Both ureters are dissected, pulled inside the bladder, and separated. Extreme care should be taken, intra- and extravesically, to avoid injuring the blood supply of the lower pole ureter (eventually removing most of the upper pole ureter, leaving a flap of it attached to the remaining ureter). The bladder base is then reconstructed to correct the weakness originated in the detrusor layer by the ureterocele. The detrusor muscle is approximated with interrupted sutures of 4-0 chromic catgut or 5-0 PDS sutures, with correction of the weak detrusor backing and avoidance of diverticulum formation. The lower pole ureter is reimplanted, with or without tailoring. We frequently use a cross-trigonal submucosal tunnel to avoid leaving the ureter lying over the sutured detrusor muscle ( Fig. 94-3).

FIG. 94-3. Open ureterocele excision. (A) Hypogastric incision with transvesical approach. The ureterocele is repaired and dissected from the detrusor; this dissection must reach distally the bladder neck or urethra to avoid remnants into the urethra. (B) After excision and remotion of the ureterocele, the detrusor hernia is repaired with interrupted sutures. The intra- and extravesical ureter is removed, preserving the blood supply and the integrity of the lower pole ureter. (C) At this step, the normal ureter is reimplanted by a cross trigonal submucosal tunnel. We use an ureteral stent transfixing through the bladder wall and exteriorized at the hypogastric region.

Ureterocele in Infants and Neonates Recently, the antenatal diagnosis of ureteroceles has become increasingly common, and treatment must be offered before urinary tract infection occurs to maximize recovery of renal function. Expectant management associated with prophylactic antibiotic therapy has been advocated in special cases for asymptomatic children, but there is no follow-up study to ensure that this is an adequate option. Complete reconstruction in neonates is a technically challenging procedure, and even upper pole nephrostomy and nephrectomy are perhaps overwhelming procedures in this setting. Endoscopic incision or puncture of the ureterocele has been proposed as the initial intervention and subsequently has proven to be the definitive treatment in 90% of the intravesical ureteroceles. Ectopic ureteroceles, however, needed a second stage in 50% of cases. There is a big controversy regarding the indication of this procedure when a duplex system is encountered in the absence of reflux into any moiety. 5 Although decompression by partial nephrectomy may be the only required surgery in this situation, it is a major procedure to be performed in a neonate. For these cases, we prefer to temporize with a high diversion through a loop cutaneous ureterostomy of the upper pole ureter. It allows drainage of the upper moiety of the kidney and ureterocele decompression with no risk of iatrogenic reflux. The child may be observed for 1 to 2 years, and during this period we will be able to follow the collapsed ureterocele, which eventually causes voiding abnormalities. Recovery of renal function in the upper pole of the kidney has been observed in approximately 50% of cases. Definitive treatment has been accomplished by a simple ureteropyelostomy afterward. Ureterocele with Sepsis In young infants and neonates with ureterocele and sepsis, the primary goal after fluid and electrolyte resuscitation, as well as antibiotics, should be to establish drainage of the upper pole. The preferred route is via an endoscopic incision, although direct incision when the ureterocele is prolapsed into the urethra is an effective and attractive method of urinary tract decompression. The direct approach to a prolapsing ureterocele would be uncapping or excision of the prolapsed part of the ureterocele. The area is closed with interrupted sutures of 4-0 chromic catgut surrounding the mucosa border and then the ureterocele is pushed into the bladder. A urethral catheter is necessary to drain the bladder and upper renal pole; the infection may be controlled with appropriate antibiotic. At a later date, it may be necessary to reoperate and perform a complete reconstruction of the bladder, with excision of the ureterocele and ureteral reimplantation of the inferior pole ureter. The follow-up renal function will be evaluated with radionuclide scans and will demonstrate function on the upper renal pole after uncapping

about 50% of the time. Loop Ureterostomy The loop ureterostomy is also a very attractive approach in septic infants. As described above, ureterostomy of the upper pole ureter decompresses all of the excretory system (upper and inner renal pole) and collapses the ureterocele. Additionally, the loop ureterostomy avoids vesicoureteral reflux and provides an opportunity to evaluate recovery of upper pole renal function and to observe the effects of the collapsed ureterocele. These two aspects will influence the long-term management of the ureterocele. If the renal pole function does not recover, then this diversion delineates the next approach: a partial heminephroureterectomy to the level of the cutaneous stoma. The upper pole loop ureterostomy is made through a small (2 cm) flank incision at the axillary line between the 12th rib and iliac crest. The retroperitoneal space is exposed and the dilated ureter or ureters can easily be identified. It is important to identify the upper pole ureter; generally it is wide, tortuous, and massively dilated. Conversely, when both ureters are of similar size, it may be necessary to use radiographs to identify both ureters. We try to separate both ureters for a 5- to 6-cm distance to avoid the kinking of the lower pole ureter during ureterostomy. The ureter is retracted to the skin, which must be reached without tension and without kinks, leaving the normal ureter in the retroperitoneal space. At the skin level the ureter is opened longitudinally for about 2 cm and the borders are sutured to the skin with 4-0 chromic catgut interrupted sutures ( Fig. 94-4).

FIG. 94-4. Ureterostomy of upper renal pole ureter. Small lumbar incision at the axillary line between 12th rib and iliac crest; the dilated ureter is commonly identified at the retroperitoneal space. Eventually both ureters are equally dilated and might be incorrectly identified; they are slightly separated with blunt dissection and then a 2-cm longitudinal incision is performed in the upper pole ureter. Urinary drainage occurs. A small stent is placed through the stump ureteral to the ureterocele and its content is aspirated to provide its collapse. Interrupted sutures between ureteral border and the skin maintain the urinary diversion.

OUTCOMES
Complications Endoscopic treatment of a ureterocele may result in reflux, though the reported incidence of reflux is less than 10% following this procedure. Failure to adequately excise the distal portion of a ureterocele may result in the creation of a flap valve that may cause bladder outlet obstruction. Excision of ectopic ureteroceles may leave a patulous bladder neck that may render the patient incontinent. It is important, therefore, to perform some method of vesical neck reconstruction or plication at the time of ureterocele excision. Results In most patients with a ureterocele and a functioning upper pole moiety, the best approach is via a minimally invasive method, specifically endoscopic ablation of the ureterocele. Its morbidity is very low, and although it is initially performed in neonates, long-term follow-up reveals that 66% of patients require no further surgery, especially for intravesical ureteroceles. Even for ectopic ureteroceles, in 50% of the cases the incision was the definitive treatment. The procedure is minimally invasive in infants, allowing recovery of function of the upper pole in many cases, as well as some resolution of reflux in both the ipsilateral and contralateral ureters. Reflux will occur or persist in 30% to 50% of patients and may demand secondary reconstruction. This procedure may be postponed for 1 to 2 years because the urinary system is already protected by the decompression. Extensive lower tract reconstruction is reserved for those patients in whom there is persistent reflux or large ureteroceles. More common is the simplified approach, which combines division of the ureter with heminephrectomy in patients with no moiety function. Ureteropyelostomy/pyelopyelostomy is the choice for those patients with functioning upper pole moieties. This simplified approach allows decompression of the ureterocele, with eventual resolution of vesicoureteral reflux in the other renal segments, such that in 75% of the patients no other procedure is necessary. In our experience, a second stage has been necessary in 25% of cases, due either to reflux, infection, or voiding abnormalities resulting from failure of the ureterocele to collapse, with consequent bladder neck obstruction. 4 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Blyth B, Passerini-Glazel G, Camuffo C, Snyder HM III, Duckett JW. Endoscopic incision of ureteroceles: intravesical versus ectopic. J Urol 1993;149:556. Coplen DE, Duckett JW. The modern approach to ureteroceles. J Urol 1995;153:166. Glassberg KI, Braren V, Duckett JW, et al. Suggested terminology for duplex systems, ectopic ureters and ureteroceles. J Urol 1984;132:1153. Husmann DA, Ewalt DH, Glenski WJ, Bernier PA. Ureterocele associated with ureteral duplication and a non-functioning upper pole segment: management by partial nephrourethrectomy alone. J Urol 1995;154:723. Monfort G, Morrison-Lacombe G, Coquet M. Endoscopic treatment of ureteroceles revisited. J Urol 1985;133:1031. Retik AB, Peters CA. Ectopic ureter and ureterocele. In: Walsh PC, ed. Campbell's urology. 6th ed. Philadelphia: WB Saunders, 1992;2:1743–1771. Schulman CC. Ureterocele. In: Glenn JF, ed. Urologic surgery. 4th ed. Philadelphia: JB Lippincott, 1991;381–389. Smith C, Gosalbez R, Parrott TS, Woodard J Jr, Broecker B, Massad C. Transurethral puncture of ectopic ureteroceles in neonates and infants. J Urol 1994;152:2110. Snyder HM III. Anomalies of the ureter. In: Gillenwater JY, ed. Adult and pediatric urology. 3rd ed. St. Louis: CV Mosby, 1996;2197–231. William DI, Fay R, Lillie JG. The functional radiology of ectopic ureterocele. Br J Urol 1972;44:417.

Chapter 95 Urachal Anomalies and Related Umbilical Disorders Glenn’s Urologic Surgery

Chapter 95 Urachal Anomalies and Related Umbilical Disorders
H. Norman Noe

H. N. Noe: Department of Urology, LeBonheur Children's Medical Center, Memphis, Tennessee 38120.

Diagnosis Patent Urachus Urachal Cyst Urachal Sinus Vesicourachal Diverticulum Nonurachal Umbilical Disorders Malignant Urachal Lesions Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

The urachus is a tubular structure that extends from the anterior dome of the bladder to the region of the umbilicus. Since the urachus normally closes or involutes at approximately 32 weeks of gestational age, disorders involving this structure represent an arrest of this normal involutionary process. The urachus is thought to develop from the allantois and cloaca, although the exact contribution of each structure to the formation of the urachus is still controversial. The allantois is a finger-like projection that arises from the area of the yolk sac that will eventually give rise to the cloaca ( Fig. 95-1). The extraembryonic end of the allantois extends into the body stalk (the anlage of the umbilical cord) and the embryonic segment remains continuous with the portion of the cloaca that eventually will form the urinary bladder. The urachus itself arises from the ventral cloaca where it joins the allantois. Ultimately, the allantois obliterates into a fibrous band by approximately the third month of gestation and the urachus continues to elongate as the fetus grows. Urachal growth continues as the distance between the bladder dome and body stalk (umbilicus) increases. By the fifth month of gestational age, the urachus narrows to a small-caliber epithelialized tube, which normally is completely obliterated or exists only as a 1-mm4 tubular structure filled, in part, with epithelial debris.

FIG. 95-1. (A) At 9 weeks of gestational age showing the allantois extending into the body stalk. (B) At 3 months gestation, the urachus connects to the dome of the bladder. (C) The urachus persisting as the median umbilical ligament in the adult.

Histologically, the urachus consists of three layers. An outer layer of muscle surrounds a connective tissue layer of blood vessels with the innermost layer being a small tubular structure lined by transitional or cuboidal epithelium. In the adult, the urachus is bound both anteriorly and posteriorly by a layer of fascia known as the umbilicovesical fascia. This fascia is in a pyramidal shape extending from the transversalis fascia anteriorly to the peritoneum posteriorly and laterally to each obliterated umbilical artery. Inferiorly the fascial compartment is marked by the dome of the bladder extending posteriorly over the hypergastric arteries and to the pelvic diaphragm anteriorly. This is an enclosed space completely separated from the peritoneum and, in early stages, tends to confine urachal disease such as infection or malignancy within this compartment. Abnormalities of the urachus are relatively rare and present predominantly in childhood, although some may become symptomatic in adult life. The basic disorders involving the urachus are (a) patent urachus, (b) vesicourachal diverticulum, (c) urachal cyst or sinus, and (d) alternating urachal sinus. 2,3 These abnormalities are depicted in Figure 95-2. Benign abnormalities are more common than malignant abnormalities, and all urachal disorders occur more commonly in males. The benign disorders occur almost exclusively in the pediatric population with autopsy series indicating patent urachus present in 1 of 7610 autopsies and urachal cysts existing in 1 of 5000 autopsies. 10 Malignant urachal lesions are confined almost exclusively in adult patients.

FIG. 95-2. The four types of abnormalities involving the urachus. (A) Complete patency to the umbilicus. (B) A blind-ending urachal sinus open at the umbilicus but not complete to the bladder. (C) A urachal cyst without communication to either the umbilicus or the bladder. (D) Simple diverticulum from the dome of the bladder.

DIAGNOSIS
Patent Urachus Failure of the epithelialized urachal tract to close results in a patent communication with the bladder that results in urinary leakage from the umbilicus. It was thought that failure of descent of the bladder contributed to the patency of the urachus, but this is now felt to be an associated finding with patency rather than a cause. 8 There has been a high percentage of children with prune belly syndrome as well as other obstructive uropathy reported to be associated with a patent urachus. Obstruction

of the lower urinary tract has been considered a contributing cause to patency of the urachus. 5,6 The theory of obstruction being causative has been challenged because it is noted embryologically that urethral tubularization occurs after the urachal lumen obliterates. Thus, it appears that distal obstruction is not necessarily causative of the urachus remaining patent as there are clinical examples of children with a patent urachus who have otherwise normal lower urinary systems. The condition has been described prenatally without associated abnormalities being present. Patent urachus usually becomes obvious due to urinary leakage from the umbilicus. A protruding mass of tissue is usually evident at the umbilicus and may be associated with an umbilical hernia. The urinary drainage may be increased by coughing, crying, or other causes of increased intraabdominal pressure ( Fig. 95-3).

FIG. 95-3. Patent urachus showing the usual protruding mass of tissue at the umbilicus with expression of urine through the patency.

The diagnosis can be confirmed by analyzing the fluid for urea and creatinine or by any one of several radiographic techniques. A voiding cystourethrogram or fistulagram may demonstrate connection through the urachus into the bladder. Indigo carmine or methylene blue can be placed into the bladder and observed to come through the umbilicus or, oppositely, placed in through the patent urachus and noted to be present in the bladder urine. Once the diagnosis has been established, voiding cystourethrogram and renal ultrasound should be obtained to rule out any distal obstruction or associated urinary abnormalities. Urachal Cyst Incomplete obliteration of the urachal lumen results in the potential for formation of a cystic lesion in this epithelial-lined space. The result is a urachal cyst that usually exists in the proximal or lower third of the urachal remnant adjacent to the bladder ( Fig. 95-4).9 The cysts are usually small and remain without symptoms, with only a third of the lesions being discovered in infancy or childhood. The cysts are usually lined with transitional epithelium and filled with serous fluid, but mucinous contents along with intestinal mucosal lining have been described.

FIG. 95-4. Urachal cyst located near dome of bladder. Right-angled clamp is under the obliterated urachal remnant as it courses toward the umbilicus.

Symptoms related to the cysts are usually associated with either large size or infection. A large cyst may present as an abdominal mass or give the feeling of abdominal pain or heaviness. Irritative voiding symptoms from bladder compression can also occur. Infected cysts usually have the signs and symptoms associated with inflammation. Pain, tenderness, and redness with localized swelling in the area underneath the umbilicus, in addition to constitutional symptoms such as chills and fever, are usually present. Once again, irritative bladder symptoms, hematuria, and pyuria can be present. The fascial compartment that contains the urachus will limit the extension of the infection and if left untreated will allow drainage either into the bladder or, alternately, into the umbilicus. The diagnosis usually requires a strong clinical suspicion but can be confirmed by ultrasound, CT scan, or cystogram. Urachal Sinus Incomplete patency of the urachal lumen opening at either the umbilical or vesical end is known as a urachal sinus. The origin of such a lesion is probably that of an original urachal cyst that became infected and then dissected either to the umbilicus distally or to the bladder proximally. In unusual circumstances, it may alternately first drain to the umbilicus and then into the bladder, creating the alternating urachal sinus. Symptoms usually include periumbilical redness and tenderness with intermittent drainage. The umbilicus may become alternately inflamed or exhibit granulation tissue. If the sinus drains primarily into the bladder, the signs and symptoms of urinary tract infection will predominate, sometimes being manifest by purulent drainage from the bladder. Diagnosis beyond clinical suspicion from physical findings can be confirmed by a fistulagram to help delineate the extent of the sinus externally. A cystogram may demonstrate the sinus, although it is more likely to show an irregular area in the dome of the bladder, which at cystoscopy will appear to be raised, reddish, and inflamed, possibly associated with some element of purulent drainage. Vesicourachal Diverticulum When the portion of the urachus adjacent to the bladder fails to obliterate, a wide-mouth diverticulum of varied size can result. This may or may not be associated with distal obstruction but has been reported frequently in boys with the prune belly syndrome ( Fig. 95-5). The diverticulum is most likely discovered during radiographic investigation for any associated abnormalities and, if it drains well, usually does not require any treatment. Treatment of any associated distal obstruction should be undertaken and excision be performed only when specific symptoms arise.

FIG. 95-5. (A) Cystogram showing large urachal diverticulum. (B) Bladder ultrasound of the same lesion.

Nonurachal Umbilical Disorders Some conditions exist that can be confused with urachal abnormalities. Omphalomesenteric duct anomalies are rare disorders of this variety. The portion of the yolk sac destined to become the foregut and midgut develops a communication between the gut and yolk sac. This tubular structure is called the omphalomesenteric duct, which usually disintegrates by the seventh fetal week. Incomplete disintegration of this connection can lead to a variety of abnormalities. These abnormalities are similar to those of the urachus and can be designated as a completely patent omphalomesenteric duct, an omphalomesenteric sinus, or an omphalomesenteric cyst. A completely patent duct is usually heralded by intestinal or fecal drainage. A sinus, cyst, or patent duct can usually be identified by either probing its depth or by fistulagram identifying the connection to the bowel instead of the bladder. A Meckel's diverticulum represents a proximal portion of a persistent omphalomesen-teric duct that may be connected by a fibrous band to the umbilicus and heralded by a granular-appearing umbilical polyp. It is important to distinguish these disorders from those of urachal origin to avoid therapeutic misadventure that can involve the intestinal tract. Umbilical hernias are common occurrences, particularly in black children. An umbilical hernia differs from an omphalocele in that a full-thickness layer of skin and subcutaneous tissue cover the protruding hernia. Occasionally, an umbilical hernia is encountered in which a urachal sinus or urachal diverticulum is included ( Fig. 95-6), and repair of this entity along with resection of the urachal abnormality is required.

FIG. 95-6. Smaller urachal diverticulum shown extending into an umbilical hernia.

Malignant Urachal Lesions Malignant lesions of the urachus are exceedingly rare. In the majority, these tumors are adenocarcinomas, although sarcoma and transitional cell carcinoma have been described. The most common sign of urachal cancer is hematuria. Additional findings can be that of a suprapubic mass, abdominal pain, irritative voiding symptoms, or passage of mucus in the urine. Umbilical discharge can be seen with urachal carcinoma but is not commonly encountered. The diagnosis of carcinoma of the urachus is usually made by cystogram showing a filling defect in the bladder dome with stippled calcifications. Computed tomography is also helpful in establishing the nature of a lesion in the dome of the bladder. Cystoscopy with transurethral biopsy can establish the histology of such a lesion.

INDICATIONS FOR SURGERY
In a patient with patent urachus, treatment is recommended because of the propensity for recurrent urinary tract infections, septicemia, and stone formation. Also to be considered is the prevention of continued umbilical excoriation, pain, and discharge. The most definitive method of treatment is excision of the patent urachus with a cuff of bladder by an extraperitoneal approach. A symptomatic type I variant has been described that likewise involved obvious umbilical signs and symptoms. The umbilical band was complete but obliterated, and caused inversion of the umbilicus with pain upon micturition. 7 Complete excision of the urachal band with a cuff of bladder was likewise curative. Treatment of infected urachal cysts requires initial drainage with subsequent resection of the cyst as well as the proximal and distal ends of the urachal remnant. Small uninfected cysts discovered incidentally can be likewise resected at the time of discovery. Treatment of a urachal sinus initially should be directed to the elimination of infection. Once that has been accomplished, excision of the entire urachal and sinus tract along with a cuff of bladder should be performed as definitive therapy. The treatment of urachal malignancies is, in general, that of any malignancy of the bladder (discussed elsewhere in this book). It is important to mention, however, that local control seems to be quite important in the natural history of urachal cancer. Excision of the urachal mass en bloc with the umbilicus, peritoneum, and posterior rectus fascia is important when one encounters an invasive lesion of the dome of the bladder. En bloc cystectomy, pelvic lymphadenectomy, and umbilectomy is recommended for all lesions unless they are known to be sarcomas or early stage I carcinomas.

ALTERNATIVE THERAPY
In general, the treatment of urachal and associated umbilical lesions is surgical. There is no alternative treatment per se but medical management, such as antibiotics for infection or chemotherapy for malignant diseases, is used as part of the overall treatment plan in which surgical management is central.

SURGICAL TECHNIQUE
In general, the surgical procedure to correct urachal and umbilical abnormalities is essentially surgery involving the dome of the bladder. The approach utilized for the procedure can either be through a suprapubic midline or a transverse infraumbilical incision. Either incision will give access to both the region of the umbilicus and the dome of the bladder to allow resection of a cuff of bladder along with the associated urachal or umbilical anomaly. If need be, the incision can be extended on into the

peritoneum for whatever associated procedure is necessary. With the bladder distended and palpable, a transverse infraabdominal incision is made roughly one-half to two-thirds of the way to the symphysis pubis and the umbilicus. If necessary, a catheter or probe can be placed in the umbilical portion of the urachal lesion for ease of intraoperative identification ( Fig. 95-7). The rectus fascia can either be opened longitudinally or transversely. The dome of the bladder is identified and the urachus isolated with either a vessel loop or a surgical clamp. As resection proceeds toward the umbilicus, the obliterated umbilical arteries are visible on either side. Once the proximal portion of the urachus has been delineated, a cuff of bladder, which includes the urachal insertion, is then removed and can be performed with the cautery to minimize blood loss. The bladder is then closed in a watertight two-layered fashion using running locking 3-0 plain and 3-0 chromic suture. The urachus is then dissected distally in an extraperitoneal fashion and removed or ligated in its obliterated portion. The goal is to remove all urachal tissue and leave the umbilicus intact. If an associated umbilical hernia is present, the fascial edges can be delineated and the peritoneal sac closed internally with the fascia and then sewn in an overlapping fashion using polyglycolic acid suture.

FIG. 95-7. (A, B) Typical transverse infraumbilical approach to the urachus. The catheter through the urachus can aid in identification of a patency. (C, D) The fascia is divided transversely and the rectus muscle is parted in the midline remaining preperitoneal. (E) The urachus or urachal remnant can be separated from the peritoneum and identified in its proximal and distal extent. (F, G) The urachus is resected with a cuff of bladder and the bladder closed in a watertight fashion. The urachus is removed completely out to the umbilicus if necessary.

Postoperative bladder drainage can be either by urethral or suprapubic catheter for 2 to 3 days depending on the surgeon's preference and the extent of bladder resection. The prevesical space can be drained with either a Penrose or a hemovac and can be removed prior to discharge of the patient from the hospital. Postoperative studies should be obtained as dictated by the patient's original condition.

OUTCOMES
Complications Complications can include those of hemorrhage and infection, which are inherent to all surgical procedures. Prolonged urinary drainage from the bladder incision can be treated by bladder catheter drainage, and infection can be treated with antibiotics or wound drainage if necessary. Results In general, the results for urachal and umbilical disorders can be expected to be quite good, particularly for the benign lesions. The simple resection of the urachus with a bladder cuff usually results in an excellent outcome barring local wound complications. The results for surgical treatment of malignant lesions of the urachus will be limited by the extent of the lesion, the ability to locally resect the lesion, and the aggressiveness of the tumor. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Arey LB. Developmental anatomy: a textbook and laboratory manual of embryology. 7th ed. Philadelphia: WB Saunders, 1974;308–312. Bauer SB, Retik AB. Urachal anomalies and related umbilical disorders. Urol Clin North Am 1978;5:195. Blichert-Toft M, Koch F, Nielson OV. Anatomic variants of the urachus related to clinical appearance and surgical treatment of urachal lesions. Surg Gynecol Obstet 1973;137:51–54. Hammond G, Yglesias L, Davis JE. The urachus, its anatomy and associated fasciae. Anat Rec 1941;80:271–287. Herbst WP. Patent urachus. South Med J 1937;30:711–719. Hinman F. Surgical disorders of the bladder and umbilicus of urachal origin. Surg Gynecol Obstet 1961;113:605–614. Knoll LD, Pustka RA, Anderson, JR, Kramer, SA. Periumbilical pain secondary to persistent urachal band. Urology 1988;22:526. Nix JT, Menville JG, Albert M, Wendt DL. Congenital patent urachus. J Urol 1958;79:264. Persutte WH, Lenke RR, Kropp K, Ghareeh C. Antenatal diagnosis of fetal patent urachus. J Ultrasound Med 1988;7:399–403. Rubin A. Handbook of congenital malformations. Philadelphia: WB Saunders, 1967;334. Sheldon CA, Clayman RA, Gonzalez R, Williams RD, Fraley EE. Malignant urachal lesions. J Urol 1984;131:1.

Chapter 96 Vesical Neck Reconstruction Glenn’s Urologic Surgery

Chapter 96 Vesical Neck Reconstruction
Stuart B. Bauer

S. B. Bauer: Department of Urology, Harvard Medical School, and Department of Urology, The Children's Hospital, Boston, Massachusetts 02115.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Rectus Fascial Sling for Bladder Neck Suspension Bladder Neck Reconstruction: Overview Young-Dees Procedure Leadbetter Procedure Kropp Procedure Pippi-Salle Procedure Outcomes Complications Results Chapter References

Recently, there has been a resurgence of interest in preserving the bladder as a useful organ for the storage of urine and the desire to maintain the urinary tract in an intact state to maximize a patient's quality of life. To this end, urologic surgeons have looked critically at alternatives to improve bladder capacity, detrusor compliance, and bladder neck resistance. The realization that urinary diversion is not a panacea but rather a Pandora's box of new clinical problems and the universal acceptance of clean intermittent catheterization as a simple means of easily emptying the bladder have prompted those who care for individuals with neurogenic bladder dysfunction to rethink their approaches to the problem of urinary incontinence. The high degree of success achieved with ureteral reimplantation and the advocates for bladder preservation have made the need for a reliable bladder neck reconstructive procedure mandatory.

DIAGNOSIS
Before reconstructive surgery on the bladder is contemplated, information regarding the status of the upper and lower urinary tract should be compiled. This includes a serum creatinine (in debilitated individuals a surface area–adjusted creatinine clearance is necessary); renal ultrasonography and a renal scan; conventional voiding cystourethrography to determine the capacity of the bladder, the presence of reflux, and the appearance, position, and competency of the vesical neck throughout filling and emptying of the bladder; and a urodynamic study. A comprehensive urodynamic study consists of a spontaneous void or a credé or Valsalva maneuver to empty the bladder followed by catheterization to measure urine residual. To ensure an accurate measurement of residual urine the catheter always should be aspirated with a syringe to remove all remaining urine from the bladder. A urethral pressure profile is performed by withdrawing the catheter at a constant rate as fluid is instilled through the side hole port per minute through the second channel of the urodynamic catheter. By constantly measuring urethral pressure, a resistance curve is obtained; then the catheter is positioned so that its side hole opening is adjacent to the maximum point of resistance. This resistance point is monitored as the cystometrogram is performed to see if any changes occur in response to filling and emptying of the bladder. The bladder is filled with warm saline (37°C) at a rate equal to 10% of the estimated or known capacity of the bladder. Intravesical pressure is continuously monitored during bladder filling and emptying through the third channel in order to assess the compliance of the detrusor. It is important to determine compliance because an augmentation procedure may be needed in addition to, or instead of, surgery at the bladder neck. Compliance of the bladder is influenced by many factors: the presence or recent occurrence of urinary infection; the type and temperature of the fluid being instilled; and the rate of infusion. Thus, a relatively slow infusion is used. Low urethral resistance may mask detrusor hypertonicity or hyperreflexia because the instilling medium escapes around the catheter as soon as intravesical pressure rises above the maximum level of resistance, thus failing to record the potential for higher than expected intravesical pressure. To measure detrusor compliance accurately in these situations, the bladder outlet needs to be occluded with a balloon catheter snugged up against the bladder neck. Alternatively, the bladder outlet may be closed off by compressing the penile urethra between two fingers or by compressing the distal female urethra against the undersurface of the pubis. Theoretically, this simulates the effect of increased bladder outlet resistance and its effect on detrusor function following surgery. In addition, it is useful to monitor the electromyographic activity of the striated muscle component of the external urethral sphincter in order to better understand the dynamics of neurogenic bladder dysfunction. This has provided insight into the changing neurourologic picture that is so often evident in children with myelodysplasia. The activity is easily measured by insertion of a concentric needle electrode paraurethrally in girls and perineally in boys, and advancing of the needle until crisp bioelectric action potentials are seen and heard on an electromyographic amplifier with an oscilloscope and an audio channel. Motor unit potentials are monitored while sacral reflexes are performed—the bulbocavernosus and anocutaneous reflex, credé, and Valsalva maneuvers, and voluntary tightening and relaxing of the sphincter. This electoral activity is recorded while the bladder is filled to capacity and during attempts at emptying.

INDICATIONS FOR SURGERY
Failure to achieve continence with relatively simple measures such as medication or intermittent catheterization augers the need for surgery. The type of surgery is dictated by the capacity, tonicity, and spasticity of the detrusor muscle; the presence of vesicoureteral reflux; the competence of the bladder neck mechanism; and the degree of denervation of the external urethral sphincter. The ability of the individual to empty the bladder and the desire to perform intermittent catheterization are other determining factors (Table 96-1). The objective of all bladder neck reconstructive procedures is to increase outlet resistance. As a result, individuals undergoing one of the operations to be described should be aware of and familiar with the techniques of intermittent catheterization. When catheterization is not a viable option, an artificial urinary sphincter can be implanted as an alternative but it must be determined that the bladder can be emptied by either a credé or Valsalva maneuver. A poorly compliant bladder in which the pressure rises steadily from the beginning of filling until capacity is reached, a hyperreflexic bladder that does not respond to high-dose anticholinergic medication, and a bladder with an age-adjusted smaller capacity than expected all warrant augmentation cystoplasty in addition to bladder neck surgery. Creation of a large enough organ for storage that fills at low pressures has greatly improved the chances of achieving a satisfactory result when surgery is performed on the bladder outlet. Depending on the need and type of surgery necessary for increasing bladder outlet resistance, an enterocystoplasty may be performed in conjunction with or before these procedures. 20

TABLE 96-1. Incontinence management

Urethral suspension operations are the least complicated of the operations designed to increase outlet re-sistance. The Marshall-Marchetti-Krantz procedure, popularized in 1949, attempted to improve stress incontinence due to an incompetent bladder neck mechanism. Pereyra and Lebherzil and then Stamey 18,19 simplified this procedure when they described a technique that elevated the bladder neck using sutures passed suprapubically alongside the most proximal portion of the urethra with specially designed needles and incorporating anterior vaginal wall tissue in the repair as a means of raising, angulating, and supporting the bladder neck. This has the effect of placing the latter in a more intraabdominal position and compressing it near the undersurface of the pubis. In patients with neurogenic bladder dysfunction, these operations do not seem to work as well as they do in individuals with normal innervation to the external urethral sphincter. Rather, they work best for females who have true stress incontinence secondary to incompetency and descent of the vesical neck without detrusor instability or hypertonicity. 3

ALTERNATIVE THERAPY
With increasing social pressure from parents and school settings, it is becoming imperative that at least preliminary measures be undertaken to achieve continence in handicapped children before they start school. For those children already on intermittent catheterization, anticholinergic medication is begun using oxybutynin hydrochloride, 1 mg/year of age, twice daily. If necessary, this dosage may be increased to 3 times a day as long as the child suffers no side effects. a-Sympathomimetic agents (ephedrine sulfate, 0.5 mg/kg twice a day, or phenylpropanolamine, 2.5 mg/kg twice a day) are added if urethral resistance is low or the child remains wet. The frequency of catheterization during the day and fluid intake may have to be adjusted accordingly to maximize the effects of medication and to maintain continence, but under no circumstances is it acceptable to perform catheterization more frequently than every 2.5 to 3.0 hours.

SURGICAL TECHNIQUE
Rectus Fascial Sling for Bladder Neck Suspension Raising and compressing the bladder neck against the undersurface of the pubis with a strip of rectus fascia has been described by both McGuire and colleagues 9,10 and Raz and associates. 14,15 The McGuire operation, which uses a strip of rectus fascia passed around the bladder neck and tied over the reapproximated rectus muscle and fascia, is performed through a Pfannenstiel incision, whereas the Raz technique of placing a square patch of rectus fascia along the undersurface of the urethra and tied suprapubically is approached transvaginally. This latter procedure is undertaken in older teenagers or adult women with a capacious vaginal vault, whereas the McGuire sling procedure is performed in prepubertal girls. Results have been reliable and sustainable in properly selected individuals. 1,14 This technique is reserved for female patients who have maximum resistance in the mid- or distal urethra of 20 to 30 cm H 20 and who are not leaking constantly but have stress incontinence whenever they cough or move about with a bladder that is at least half full. The incontinence may result from urethral instability (periodic relaxation of the external urethral sphincter during filling of the bladder), or failure of the urethra to increase its resistance correspondingly when there is a rise in abdominal pressure from a cough or Valsalva maneuver, or from progressive loss of bladder neck or proximal urethral resistance as the bladder is filled to capacity. Patients may not be able to empty their bladders completely and intermittent catheterization may be needed after surgery. Although the McGuire sling has been performed in pre-pubertal boys as well with reasonable rates of success, it is not widely accepted as a viable alternative for incontinence in postpubertal males. 15 Bladder Neck Reconstruction: Overview These procedures are useful when the bladder neck mechanism is incompetent and the external sphincter provides some resistance but is insufficient to prevent stress urinary incontinence. These techniques are usually performed in conjunction with ureteral reimplantation. The Young-Dees and Leadbetter 6,7 approaches have been used most often in cases of bladder exstrophy (secondarily after closure), epispadias, urogenital sinus abnormalities, bilateral single ectopic ureters, and after trauma to the bladder neck region. They also have been used in patients with neurogenic bladder dysfunction when the bladder neck is patulous and the external sphincter is only partially denervated but reacts to stimulation. Young-Dees Procedure The Young-Dees operation has been described classically to treat exstrophy of the bladder and epispadias. Ureteral reimplantation in a cross-trigonal manner is necessary to provide adequate room for the bladder neck reconstruction. A large-capacity bladder is mandatory preoperatively; otherwise an augmentation is needed to ensure a satisfactory size for the urinary reservoir. Normally, a Pfannenstiel incision is made when no previous surgery has been performed; however, if a midline vertical scar is present from prior surgery, that incision can be used again as long as it is extended to the symphysis pubis. Once the bladder neck and anterior surface of the bladder are exposed, the bladder wall is opened in the midline down to and including the patulous bladder neck area. The ureteral orifices are identified and may be reimplanted higher in the posterior wall of the bladder using the Cohen cross- trigonal technique. This procedure provides sufficient room for advancing the neourethra more superiorly and incorporating the trigone in the new bladder neck mechanism. Extending the ureteral hiatus somewhat more superiorly and laterally after the ureters have been dissected adequately facilitates the creation of new ureteral submucosal tunnels higher in the posterior detrusor wall without excessively angulating the ureters as they enter the bladder. The inferior aspect of each ureteral hiatus is closed with interrupted 3-0 Vicryl sutures. A 12- to 14-mm-wide strip of mucosa is outlined along the floor of the posterior urethra from an area just proximal to the external sphincter to a point immediately below the newly positioned submucosal ureters. Parallel incisions are made in the mucosa from the inferior to the superior aspect of the strip ( Fig. 96-1A). Each incision is then directed laterally at right angles to the parallel incision at its cephalad margin. This creates two triangular patches of bladder mucosa laterally that are removed from the underlying detrusor muscle using cautery to minimize blood loss. The midline strip of mucosa is isolated as it extends from the original posterior urethra to the new submucosal ureteral tunnels. The isolated strip of mucosa is closed over an 8-Fr catheter that has been passed into the bladder through the urethra, with a running submucosal suture of 4-0 Vicryl, beginning inferiorly and extending in a cephalad direction ( Fig. 96-1B). This closure must be tension-free. Next, the denuded triangular portion of detrusor on either side of the newly constructed tube is cut along its horizontal edge. The serosal side of this triangular area of muscle is freed from surrounding attachments to permit these lateral flaps to be brought around and over the neourethra. The increased mobility of each flap allows for a muscular enclosure of the neourethra in a vest-over-pants manner. The most mobile lateral muscular flap is used for the outside covering, whereas the less mobile flap is employed first as the inner covering. The inner flap is brought over and around, and sutured to the opposite posterolateral side of the neourethra, at the base of the contralateral flap, with a running 3-0 Vicryl suture. The contralateral, more mobile detrusor flap is then brought over the first muscle flap and sutured as far laterally as possible to create a tight and strong muscular covering ( Fig. 96-1C). Ureteral catheters may be left in place if there is concern about edematous tissue affecting the drainage of urine into the bladder postoperatively. The remaining cystotomy site is closed, first with a horizontal mattress suture of 3-0 chromic excluding the mucosa and then with a second running layer incorporating only the approximated muscle and serosal tissues. A cystotomy tube, which has been used to drain the bladder, and ureteral catheters (if necessary) are brought out through a separate opening in the anterior detrusor wall, before the bladder is completely closed. A 2-0 silk suture is secured to the end of the urethral catheter and brought through the anterior bladder and abdominal wall. The silk suture is tied over a dental pledget once the abdominal wound is closed to maintain the catheter in the fundus of the bladder and to minimize any pressure the catheter may exert on the posterior aspect of the neourethra and cross-trigonal submucosal uretral tunnels.

FIG. 96-1. (A) After the ureters are reimplanted, a midline strip of mucosa 12 to 14 mm wide is outlined. Adjacent lateral detrusor muscle areas are denuded of mucosa. (B) First the mucosa and then the muscle is rolled into a tube after the upper horizontal edge of the muscle has been incised. (C) The muscle flaps are secured to create a strong outer wall. (D) The neourethra and the bladder neck are reinforced with vertical mattress sutures to develop a competent angle of

continence.

After the bladder is completely closed, a clear demarcation between the neourethra and the new bladder neck can be seen. This angle is further reinforced by several vertical mattress sutures of 3-0 Vicryl (Fig. 96-1D). Finally, the neourethra is sutured to the undersurface of the symphysis pubis via placement of two or three pairs of zero or 2-0 Vicryl sutures in the muscle layer of the neourethra and the periosteum of the symphysis, as has been described in the Marshall-Marchetti-Krantz procedure. 8 One pair of sutures is placed at the bladder neck; all are then tied to ensure a good suspension of the new bladder neck. The ureteral catheters are left for 5 to 7 days, depending on the amount of urine draining around them and entering the bladder. The 8- Fr urethral catheter is removed after 7 to 10 days, but the suprapubic tube may be left as long as 3 to 6 weeks, especially if intermittent catheterization is resumed or initiated. A contrast cystogram may be obtained to look for incomplete healing or extravasation before the cystotomy tube is withdrawn. Leadbetter Procedure An alternative to the Young-Dees procedure is the Leadbetter technique. 6,7 After the ureters have been reimplanted transtrigonally higher in the posterior wall of the bladder, two parallel, full-thickness incisions through the entire detrusor wall (not just the mucosa), 12 to 14 mm apart, are made in the posterior trigonal wall and proximal urethra (Fig. 96-2A). This creates two lateral flaps ( Fig. 96-2B). The mucosa and muscle of this midline strip are each closed seperately with a running 3-0 Vicryl suture. The remaining lateral flaps of trigonal and bladder neck tissue are placed over the neourethra in a vest-over-pants fashion, creating a strong three-layer closure. The surface of these flaps must be denuded of mucosa before they can be sewn in place. The open bladder and the new bladder neck are closed in a fashion described previously. The catheters draining both ureters and the bladder are managed in a similar manner.

FIG. 96-2. The Leadbetter technique differs slightly in that a full-thickness strip of posterior bladder wall is created (A) and tubularized (B).

Kropp Procedure The Kropp operation has the advantage of creating an extremely efficient bladder outlet mechanism. It can be used when both the bladder neck and the external sphincter mechanisms are incompetent. The procedure, however, requires repositioning of the ureteral orifices, and invariably an enterocystoplasty, because so much bladder is used for the reconstruction. Thus, this reconstructive effort is reserved for those individuals who need (a) ureteral reimplantation because of reflux or ureterovesical junction obstruction, (b) augmentation as a result of small capacity or detrusor hypertonicity, and (c) who accept intermittent catheterization as the only means of emptying the bladder postoperatively. After the anterior bladder wall is exposed, traction sutures of 3-0 Vicryl or chromic are placed outlining the anterior flap of detrusor muscle that is 2.5 cm wide and 6.0 cm long, with the base of the flap originating at the lowest point of the bladder neck ( Fig. 96-3A). The flap runs superiorly in the midline with the base slightly wider than the tip. Additional traction sutures are placed at the superior corners of the flap and the detrusor muscle and mucosa are incised with electrocautery. The flap is retracted over the pubis. The Denis-Browne ring retractor is set in place inside the bladder. The ureters are dissected free of their attachments to the surrounding detrusor muscle and retroperitoneum, and are reimplanted more superiorly and laterally in the posterior bladder wall so that there is enough room to develop a midline submucosal tunnel for the anterior bladder tube. Each old ureteral hiatus is closed with interrupted 3-0 Vicryl sutures. A submucosal tunnel is fashioned either transversely or longitudinally, leaving enough room in the midline trigone to accommodate the eventual submucosal bladder flap tube. The ureters are brought through the submucosal tunnels and are sutured in place in a standard fashion using 4-0 chromic catgut.

FIG. 96-3. A long (6-cm) anterior flap (2.5 cm wide) of detrusor muscle based at the vesical neck junction with the urethra (A) is tubularized (B). A submucosal midline tunnel is created in the trigonal area after the ureters have been reimplanted more cephalad and lateral (C). The neourethra is pulled through this tunnel (D) and its open end sutured to surrounding bladder muscle and mucosa to produce a new bladder neck. The base of the new urethral tube is secured to the bladder wall both posteriorly and anteriorly with double-helical sutures (E). The bladder is either closed primarily or augmented (shown here), but not before the urethral catheter is held in place with a silk suture and a suprapubic tube is passed through the detrusor muscle to drain the urine (F).

Attention is then directed to creating a tube from the anterior bladder wall flap. A 14-Fr catheter is passed from the meatus and advanced sufficiently that the entire length of bladder flap can fit over it. The flap is tubularized around this catheter with two layers of a running Vicryl suture, using 4-0 to incorporate the mucosa and muscularis and then 3-0 to approximate the muscularis and serosa ( Fig. 96-3B). Next, the midline submucosal tunnel is made (Fig. 96-3C). The mucosa along the posterior wall of the original bladder neck, at the level adjacent to the base of the flap, is then incised horizontally from one edge of the open bladder wall to the other. Although Kropp and Angwafo's original description 5 suggests completely incising the posterior bladder wall at this level, it is easier to catheterize the patient subsequently, when the bladder neck is not entirely separated from the posterior urethra. 2 Stay sutures of 3-0 Vicryl are placed outside-in in the detrusor and posterior urethral walls just below the transverse mucosal incision line to help draw the base of the trigone downward and anteriorly. The needles are left attached to these sutures. A midline submucosal tunnel is made in the trigone and posterior bladder wall beginning at this incision to accommodate the new bladder neck tube ( Fig. 96-3C). The tunnel is widened sufficiently to ensure easy passage of the tube but without interfering with the drainage of the newly reimplanted ureters. The new bladder tube is pulled through the submucosal tunnel after the surgeon has passed a right angle clamp from above downward and grasped the stay sutures attached to the end of the tube ( Fig. 96-3D). The stay sutures at the base of the trigone are sutured

to the base of the tubularized flap to close the gap along the undersurface where a new tube begins. Additional sutures of 3-0 Vicryl are placed laterally to approximate completely the muscularis of the posterior urethral wall to the origin of the new tubularized tunnel ( Fig. 96-3E). This facilitates keeping the base of the trigone in an inferior position, minimizing any foreshortening of the tunneled tube, and ensuring that all portions of the new urethra are supported. This maneuver allows easy passage of a catheter into the bladder. Finally, the new bladder outlet at the superior end of the tubularized flap is secured to the detrusor muscle posteriorly and to the mucosal layer overlying the tunnel anteriorly, using 4-0 chromic sutures. The ability to catheterize the bladder is tested. When the surgeon is assured that catheterization will proceed without difficulty in the postoperative period, the inferior portion of the opened bladder is closed by suturing the edges of the detrusor to the anterior surface of the tubularized bladder flap distal to the point at which it enters the submucosal tunnel. If augmentation of the bladder is needed at this juncture, the midline lower abdominal incision is extended superiorly toward the xyphoid and the peritoneal cavity entered. Once the isolated segment of bowel is prepared for the augmentation, it is anastomosed to the remaining open edges of the detrusor ( Fig. 96-3F). The upper edge of the original cystotomy incision may be advanced over the dome of the bladder to allow for a superior placement of the augment. A cystotomy tube is passed through the detrusor muscle to drain the bladder. The urethral catheter is secured in place with a silk suture. Ureteral catheters are mandatory because periureteral swelling from manipulation of the trigone and posterior bladder wall may impair drainage for several days. Drains are placed in appropriate places in the region of the old bladder neck, the space of Retzius, and peritoneal cavity (if an augmentation has been performed). The ureteral catheters are usually left in place for a week until the cystotomy catheter demonstrate that sufficient amounts of urine are passing around the ureteral catheters and entering the bladder. The urethral catheter is left intact for 4 to 6 weeks. Only after a voiding cystogram demonstrates no suture line leak is the suprapubic tube clamped and the individual instructed in self-catheterization. The suprapubic tube is removed when it is easy to catheterize. Pippi-Salle Procedure The space of Retzius is developed through either a Pfannenstiel or vertical midline incision. With the bladder partially filled, a 7-cm-long vertical midline flap, with its base 2 cm wide at the bladder neck and 1.5 cm at its tip, is outlined ( Fig. 96-4A).13,16 Stay sutures are placed along the edges and ends of the flap. Using electrocautery for the muscle and sharp dissections for the mucosa, the flap is created leaving a slightly excessive amount of mucosa at the cephalic end to delay the maturation of the new bladder neck stoma. Next, the bladder is opened further by extending the midline incision and setting a self-retaining retractor in place. Two vertical incisions in the bladder mucosa outline a 7-cm-long × 1-cm-wide strip of trigonal mucosa ( Fig. 96-4A). The vertical mucosal incisions in the anterior and posterior bladder wall are extended distally into the proximal urethra until they join just slightly inferior to the old bladder neck. Once the ureters are reimplanted in a cross-trigonal fashion more cephalad to the isolated trigonal mucosal strip, the bladder mucosa lateral to each vertical incision is elevated sufficiently, using cautery to minimize bleeding, in order to bring the edges together in the midline without tension.

FIG. 96-4. (A) An anterior bladder wall flap is outlined 7.0 cm long × 2.0 cm at its base × 1.5 cm at its tip and created as the bladder is opened. Mucosal incisions are made in the trigone 1 cm apart after the ureters are reimplanted more cephalad. (B) The mucosal edges of the flap are sewn to mucosal edges of the trigonal strip to create an onlay extension of urethra. (C) The muscle layers of the flap and trigone are anastomosed to complete the onlay. (D) Lateral trigonal mucosal flaps are then developed and approximated in the midline to cover the new urethra. (E) The meatus of the new bladder neck is matured and the bladder closed. (F) Catheters are left in place as shown and an augment performed if needed to achieve an adequate-sized urinary reservoir.

The mucosa of the anterior bladder wall flap is sutured to the isolated strip of trigonal mucosa using a running subcuticular 4-0 Vicryl suture ( Fig. 96-4B). Uniform tubularization is created by progressing cephalad simultaneously on each side beginning at the bladder base. The muscle layer of the flap is sewn to the exposed posterior trigonal muscle on either side of the ucosal strip using interrupted 4-0 Vicryl sutures ( Fig. 96-4C). If a urethral catheter passes easily through this newly constructed tube, the two lateral mucosal flaps are brought together in the midline to cover the entire neourethra ( Fig. 96-4D). At the site of the original bladder neck, the excess mucosa at the cephalic extent of the flap is sutured to the reapproximated lateral mucosal flaps to mature the new “bladder neck” ( Fig. 96-4E). The lateral edges of the anterior bladder wall are sewn together in two layers; the mucosal edges are sewn to the approximated mucosa covering the flap and the muscle is stitched to the serosal surface of the flap as far inferiorly as possible to elongate the tunnel and the new bladder base. If bladder enlargement is necessary, the initial bladder incision is extended over the dome to make room for an augmentation. Ureteral stents, a suprapubic tube exiting through the detrusor muscle, and an 8-Fr feeding tube passed through the neourethra secured in place with a silk suture and tied over a dental pledget affixed to the abdomen are left in place postoperatively ( Fig. 96-4F). The ureteral catheters are removed in 7 days as long as sufficient amounts of urine drain around them. The urethral catheter is maintained for 3 weeks. Prior to removal of the suprapubic tube, a cystogram is obtained to ensure complete healing without any leaks from the bladder, and patency of the neourethra is assessed by passing a catheter into the bladder transurethrally.

OUTCOMES
Complications As with any bladder surgery, prolonged leakage along the cystotomy site can occur until the suture line heals. Although this is least likely to happen in children, it is particularly true for those who have had multiple procedures performed on their bladders in the past. In patients who have an augmentation, the leakage may take place along the anastomosis between the bowel segment and the bladder. If this area has not been retroperitonealized, the urine may extravasate into the peritoneal cavity, which can have devastating consequences if not recognized early. Electrolyte disturbances and a rising creatinine from the reabsorption of urine across peritoneal surfaces may be an insidious event if the urine output is not carefully measured. Abdominal distention, prolonged postoperative ileus, decreasing urine output, and sepsis are signs of a bladder disruption and urinary extravasation. The two most likely long-term complications to occur after bladder neck reconstruction are difficulty with catheterization leading to urinary retention or failure of the reconstructive effort to provide sufficient resistance to prevent wetting. These complications do not manifest themselves until the urethral catheter is removed and the patient is given a trial of catheterization with the suprapubic tube in place. Ease of catheterization should be obvious during the first few attempts at passing a catheter. If a straight catheter is difficult to pass beyond the reconstructed bladder neck, a Coudé tipped catheter can be substituted. On occasion, catheterization may be easy at first but become increasingly difficult after several months due either to progressive angulation of the new bladder neck or ischemia of the flap if that type of reconstructive procedure had been undertaken. Sometimes the children are dry in the early postoperative period because edema at the operative site provides adequate resistance. Over time, however, urinary incontinence becomes obvious as tissue swelling subsides. Results The Young-Dees and Leadbetter procedures are most effective for children who have an intact external urethral sphincter mechanism but who suffer from incontinence because of an incompetent bladder neck. The success rate approaches 90% or more when the patient has incontinence stemming from a prior transurethral resection or Y-V plasty of the bladder neck; 80% following repair of an exstrophied bladder or surgery for epispadias, bilateral single ectopic ureters, or postoperative excision of a sphincteric ureterocele. The success rates for neurogenic bladder dysfunction rarely exceed 25% because the nerve injury invariably affects the external urethral sphincter musculature as well as the bladder neck mechanism. Despite a significant increase in resistance at the bladder neck following

surgery in these patients, stress incontinence is still likely as a result of the denervated external sphincter muscle and its failure to react to increased abdominal pressure. Therefore, this technique for increasing outlet resistance is not recommended for children with a neurologically impaired bladder. The fascial sling is better suited for patients with neurogenic bladder dysfunction with success rates reaching 75% or higher when augmentation had been performed in conjunction with the sling. 12 The tubularized or onlay flap buried submucosally (Kropp or Pippi-Salle technique) has achieved a 75% to 80% continence rate. Difficulty with catheterization has been reported in 40% of patients with the Kropp procedure, but that incidence has been lowered when the posterior urethra is not totally transected but rather the mucosa alone is incised at the point at which the new submucosal tunnel starts. 2 Stones secondary to stasis of urine below the new bladder neck (when the bladder is not completely drained each time it is catheterized) have occurred in 10% to 15% of children within 2 years of surgery. Reflux may appear in as many as 45% of patients post-operatively when the ureters are not reimplanted following submucosal trigonal tunneling of the neourethra. Although surgeons have feared that urinary obstruction from edematous tissues surrounding the intramural ureters is a distinct possibility when they are left in situ during these procedures, this has not occurred with any regularity in reported series. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Bauer SB, Peters CA, Colodny AH, Mandell J, Retik AB. The use of rectus fascia to manage urinary incontinence. J Urol 1989;142:515. Belman AB, Kaplan GW. Experience with the Kropp anti-incontinence procedure. J Urol 1989;141:1160. Blaivas JG, Olsson CA. Stress incontinence: classification and surgical approach. J Urol 1988;139:727. Dees JE. Congenital epispadias with incontinence. J Urol 1949;62:513. Kropp KA, Angwafo FF. Urethral lengthening and reimplantation for neurogenic incontinence in children. J Urol 1986;135:533. Leadbetter GW Jr. Surgical correction of total urinary incontinence. J Urol 1964;91:261. Leadbetter GW Jr, Fraley EE. Surgical correction for total urinary incontinence: five years after. J Urol 1967;97:869. Marshall VF, Marchetti AA, Krantz KE. The correction of stress incontinence by simple vesico-urethral suspension. Surg Gynecol Obstet 1949;88:509. McGuire EJ, Lytton B. Pubovaginal sling for stress incontinence. J Urol 1978;119:82. McGuire EJ, Wang CC, Usitalo H, Savastano J. Modified pubovaginal sling in girls with myelodysplasia. J Urol 1986;135:94. Pereya AJ, Ledherz TD. Combined urethrovesical suspension and vaginal urethroplasty for correction of urinary stress incontinence. Obstet Gynecol 1967;30:537. Perez LM, Smith EA, Broecker BH, Massad CA, Parrott TA, Woodard JR. Outcome of sling cystourethropexy in the pediatric population: a critical review. J Urol 1996;156:642. Pippi Salle JL, deFraga JCS, Amarante A, et al. Urethral lengthening with anterior bladder wall flap for urinary incontinence: a new approach. J Urol 1994;152:803. Raz S, Ehrlich RM, Ziedman EJ, Alarcon A, McLaughlin S. Surgical treatment of the incontinent female patient with myelomeningocele. J Urol 1988;139:524. Raz S, McGuire EJ, Ehrlich RM, et al. Fascial sling to correct male neurogenic sphincteric incontinence. J Urol 1988;139:528. Rink RC, Adams MC, Keating MA. The flip flap technique to lengthen the urethra (Salle procedure) for treatment of neurogenic urinary incontinence. J Urol 1994;152:799. Ritchey ML, Kramer SA, Kelalis PP. Vesical neck reconstruction in patients with epispadias-exstrophy. J Urol 1988;139:1278. Stamey TA. Endoscopic suspension of the vesical neck for urinary incontinence. Surg Gynecol Obstet 1973;136:547. Stamey TA. Endoscopic suspension of the vesical neck for urinary incontinence in females: report on 203 consecutive patients. Ann Surg 1980;192:465. Strawbridge LR, Kramer SA, Castillo O, Barrett DM. Augmentation cystoplasty and the AS-800 artificial genitourinary sphincter. J Urol 1989;142:297. Young HH. An operation for the cure of incontinence of urine. Surg Gynecol Obstet 1919;28:84.

Chapter 97 Considerations in Pediatric Endoscopy Glenn’s Urologic Surgery

Chapter 97 Considerations in Pediatric Endoscopy
Scott A. Berkman and Edwin A. Smith

S. A. Berkman and E. A. Smith: Department of Surgery, Emory University School of Medicine, Egleston Children's Hospital, Atlanta, Georgia 30322.

Indications Cystourethroscopy Ureteroscopy Percutaneous Nephrolithotomy/Lithotripsy Laparoscopy Alternative Therapy Surgical Technique Instrumentation Cystourethroscopy Ureteroscopy Percutaneous Nephrolithotomy Laparoscopy Outcomes Complications Results Chapter References

Endoscopy in pediatric urology has evolved from lens technology and practice principles that were developed in the adult arena. The initial advance was the Hopkins lens system engineered in the 1970s. Subsequently, advanced video technology with high-resolution cameras and magnified images freed the surgeon from directly viewing through the lens. 12 Large-screen monitors also allowed for clear visualization of endoscopic maneuvers. With miniaturization of instrumentation, endoscopic procedures became applicable in the pediatric population. The extension of endoscopic techniques to the pediatric urologic patient continues to be defined. However, special considerations in this population have been realized. This chapter focuses on the basic principles of pediatric cystourethroscopy, ureteroscopy, laparoscopy, and percutaneous nephrolithotomy.

INDICATIONS
Cystourethroscopy Cystourethroscopy may be employed for pediatric investigation of hematuria, anatomic delineation prior to open procedures, confirmation and treatment of intraurethral or intravesical problems, or ureteral cannulation. If conventional radiographic imaging does not yield an explanation for nonglomerular-type hematuria or if hematuria persists for more than 3 months, then cystoscopic evaluation may be warranted. 9 Rarely, tumors of the lower urinary tract or urethral polyps may require cystoscopic inspection and biopsy. Cystoscopy, however, is most commonly employed in patients with ectopic ureters, ureteroceles, urethral valves, or vesicoureteral reflux that may necessitate intraoperative examination and intervention. Endoscopic injection of different substances behind the sub mucosal ureter for the correction of vesicoureteral reflux has been reported. Currently, polytetrafluoroethylene (PTFE, Teflon) seems to be the substance that produces the most reliable results. However, the risk of distal migration of Teflon particles has precluded widespread use of this material. 16 The search for a more suitable substance continues. Endoscopic delivery has also been used for submucosal bladder neck injection therapy for stress incontinence. Use of submucosal bladder neck injections of bovine collagen has offered improvement in approximately 45% of patients with stress incontinence of neurogenic origin and 65% of patients with exstrophy and epispadias. 14 Ureteroscopy Urolithiasis in children is a relatively rare phenomenon. As with adult patients, the position and size of the stone are central management considerations. However, spontaneous passage is much more difficult to predict in the pediatric age group. The percentage of different-sized stones that pass spontaneously in adults may be applicable to pubertal and postpubertal children. Overall the spontaneous passage of ureteral calculi in pediatric patients with urolithiasis is about 21%. 11 Ureteroscopic stone manipulation has been successful in children as young as 3 years. 1A Preliminary placement of an indwelling ureteral stent for a few days prior to ureteroscopy is generally unnecessary in the adult patient, yet for the pediatric patient with a limited ureteral caliber, this approach continues to be useful. Dilation of the transmural tunnel just prior to ureteroscopy may be suitable for the adolescent patient. Available pediatric-sized ureteral stents have contributed to adequate preoperative and postoperative drainage. Percutaneous Nephrolithotomy/Lithotripsy Indications for percutaneous nephrolithotomy/lithotripsy include bulky stone burden in the kidney. Also, dilation of a narrow infundibulum or ureteropelvic junction may be achieved through the same percutaneous access site. Laparoscopy The advantages of laparoscopy over open laparotomy are that it allows a shorter hospitalization, earlier return to unrestricted activity, and less postoperative pain. In the mid 1970s, diagnostic laparoscopy was useful in evaluating nonpalpable testes. Many applications have been described in the literature. Laparoscopic exploration for nonpalpable testes can help the surgeon plan the proper incision, can demonstrate blind ending vas deferens and spermatic vessels, or can facilitate a ligation of the spermatic vessels. Insufflation with or without laparoscopy has been mentioned for investigation of the contralateral asymptomatic inguinal hernia. 5,15 Gonadal examination and biopsy may be accomplished in cases of intersex. Lich-Gregoir extravesical antireflux procedures have been described laparoscopically. 5 Spermatic vein ligation, laparoscopic autoaugmentation, nephrectomy, nephroureterectomy, dismembered pyeloplasty, and decortication of renal cysts, among others, have been reported. 2 Diagnostic laparoscopy is now used widely, whereas the technical proficiency required for interventional laparoscopic maneuvers exists in selected institutions. Prior abdominal surgery constitutes a commonly encountered relative contraindication to laparoscopy. In addition, abdominal wall infection, generalized peritonitis, bowel obstruction, and an uncorrected coagulopathy serve as absolute contraindications to laparoscopy. 3 Patient and family should be informed that if a laparoscopic attempt cannot yield satisfactory results or if inadvertent injury to an intraabdominal organ occurs, then the procedure will be converted to an open approach.

ALTERNATIVE THERAPY
Endoscopic stone manipulation often provides the most expeditious form of therapy. However, extracorporeal shock wave lithotripsy (ESWL), open surgery, and chemolysis represent alternatives or adjunctive measures. Difficulty in predicting spontaneous passage and inadequate instrumentation for endoscopy of pediatric ureters may have led to a greater likelihood of open stone management in the past. Now, with miniaturization of ureteroscopic instruments and with the application of ESWL to pediatric ureteral calculi, open surgery has become the exception rather

than the rule. ESWL has become the recommended treatment for upper and midureteral stones. Lower ureteral stones are generally amenable to ureteroscopy. Although more distal stones may be successfully treated by ESWL, the female gonads may be at risk for shock wave damage in cases of ureteral stones in the true pelvis.7 Open reconstruction of the collecting system during a nephrolithotomy may be useful in patients with concomitant anatomic anomalies such as ureteropelvic junction obstruction or infundibular stenosis. The open approach to stone management may also be considered in myelodysplastic patients with a distorted body habitus, in whom percutaneous access may be dangerous or unreliable. Other scenarios that warrant an open surgical procedure involve ureteral stones refractory to ESWL and endoscopic means, or stones in ureteroceles and primary obstructed megaureters where reconstruction and reimplantation may be necessary. Chemolysis provides an alternative management for patients with cystine, uric acid, and struvite stones. Uric acid calculi respond to alkalinization of the urine with oral therapy or with direct contact of an irrigation solution such as 0.1 M sodium bicarbonate. For cystine stones, chemolysis may be achieved with urinary alkalinization, possibly including oral penicillamine. For struvite calculi, Suby's G, and hemiacidrin (Renacidin) appear to be effective by direct irrigation. Overall, chemolysis is most effective for pure stones, but days to months may be required if used as monotherapy.6,10 Usually, the chemolysis is used as adjuvant therapy in combination with ESWL or percutaneous nephrolithotripsy, and as a primary therapy for dissolution of small calculi. A child with urolithiasis may try chemolysis in the absence of significant pain, persistent obstruction, or unresolved urinary tract infection. Otherwise, expeditious intervention with the least morbidity should be undertaken whether it involves an endoscopic or open approach.

SURGICAL TECHNIQUE
Instrumentation Cystourethroscopes for neonates and infants are available in multiple sizes by different instrument companies ( Table 97-1). Manufacturers currently offer a wide array of ureteroscopes (Table 97-2). There are a variety of ancillary instruments such as wires, dilators, and ureteral catheters available from Microvasive Boston Scientific Corporation and Cook Urological; and whistle, round, and olive tip catheters as small as 3.0 and 4.0 Fr from Bard. In addition, two-prong and three-prong grasping forceps, as well as flat wire and spiral helical baskets, can be used to retrieve stones and stone fragments created by lithotripsy. Lithotripsy can be carried out with electrohydraulic or pneumatic devices within the ureter. 8

TABLE 97-1. Pediatric cystoscopes

TABLE 97-2. Pediatric ureterorenoscopes

For percutaneous cases, Storz has a 17-Fr operating sheath utilizing a zero degree telescope with and without an angled eyepiece. Circon ACMI offers 19.5-Fr, 22.5-Fr, and 26.1-Fr sheaths with 4-mm and 5-mm working ports. The 17- and 19.5-Fr sizes would be best suited for small children. Adolescents may accommodate the 22-Fr and larger nephroscopes. Cystourethroscopy The operating surgeon and operating room assistants should be attentive to using the adequate caliber of the instrument, to applying generous lubrication, and to introducing the instruments under direct visual control. The potential for hypothermia may be prevented by using warm irrigation and the risk of overdistention reduced by hanging the irrigation fluid no more than 40 cm above the patient. Metal sounds may be used to gauge the patency of and gently dilate the urethral meatus, the fossa navicularis, and the pendulous urethra. Overaggressive dilation or use of inappropriately large instruments exposes the urethra to destructive shearing forces. Usually the neonate and infant urethra can accept instruments up to 11 Fr. The small child between 1 and 10 years of age may accept up to 14 Fr. The shorter female urethra may accommodate 13.5 Fr soon after infancy. The instruments above 17 Fr should be reserved for postpubertal adolescents. Endoscopy of the lower urinary tract is performed in a systematic fashion with retrograde passage of the scope. The anterior and posterior components of the urethra and the bladder are carefully examined. In the anterior urethra, one may note findings of urethral diverticula, anterior urethral valves, or polyps. Bulbar urethritis would be manifested by red bulbar mucosa or by inflammatory pseudopolyps at the verumontanum. Upon passing gently through the membranous urethra, inspection of the posterior urethra may yield findings of posterior urethral valves or a prostatic utricle. While retracting the scope, the leaflets of the posterior urethral valves can be demonstrated with mild suprapubic pressure. To improve outlet resistance, submucosal bladder neck collagen injections should be directed to the 3 o'clock and 9 o'clock positions with the bevel of the needle facing medially. Two to three injection sessions may be necessary to provide sufficient submucosal bulk, narrowing of the urethral lumen, and, eventually, functional occlusion of the proximal urethra. Examination of the bladder may delineate the anatomy of a ureterocele. It is usually possible to discern whether the ureterocele is associated with a single or duplex system, whether the orifice is orthotopic or ectopic, whether the orifices are stenotic or patulous, or whether the ureterocele is tense or not. Also, in evaluation of the bladder, the degree of trabeculation, the orifices of refluxing ureters, and the configuration of the urachus can be assessed. Diverticula can be better characterized in terms of their location: whether paraureteral or not, whether the mouth of the outpouching is sphincteric or capacious, or whether any mass or stone is present in the

diverticulum. On the introitus or anterior vaginal wall of a female or prostatic urethra of a male, orifices to ectopic ureters or to cecoureteroceles may be encountered. Ureteroscopy During ureteroscopic manipulation, the fragile nature of the ureteral wall and the potential submucosal passage of instruments, wires, or stents should be considered. The ureteral mucosa, which is four to five cell layers thick, consists of transitional epithelium, and a lamina propria with loose or dense connective tissue, but not a distinct submucosa. The second layer—the muscularis—varies in its composition over the course of the ureter. The muscularis is less developed in the proximal ureter and pelvis rendering this region more vulnerable to perforation. Three narrow areas of the ureter—the ureterovesical junction, the pelvic brim, and the ureteropelvic junction—may cause difficult passage of a stone or of a ureteroscope. Flexible ureteroscopes have made endoscopic manipulation of stones above the iliac vessels feasible. Preoperative intravenous antibiotics are recommended. Sterile urine, general anesthesia with muscle paralysis, and the immediate availability of an assortment of baskets, forceps, and lithotripsy devices are absolute prerequisites. Initially, a retrograde study delineates the intraluminal pathology, using a ureteral catheter with a cone tip, flanged cone, tapered end, or open end. Injection of contrast should be gentle with minimal volume under fluoroscopy to prevent early proximal migration of the stone. Dilution of the contrast with saline to 50% or less reduces the likelihood of obscuring a small calculus. A floppy-tip PTFE Bentson-type 0.018- to O.038-in. guidewire is passed with cystoscopic and fluoroscopic guidance into the renal pelvis. The guidewire diameter of 0.018 in. is equivalent to 1.4 Fr; (0.025 in. = 1.9 Fr; 0.032 in. = 2.5 Fr; 0.035 in. = 2.7 Fr; and 0.038 in. = 2.9 Fr). Occasionally, a glidewire (coated with a hydrophilic polymer) and open-ended ureteral catheter may be used in concert, advancing the glidewire past the point of obstruction and then passing the ureteral catheter over the wire into the renal pelvis. The more slippery glidewire can be exchanged for the stiffer PTFE wire. Dilation of the ureteral orifice and intramural ureter can be carried at 2 atmospheres with a balloon or serially with ureteral dilators. The introduction of the ureteroscope with the bevel facing up may prevent the beak of the scope from catching on the upper lip of the ureteral orifice. Then, the ureteroscope can be rotated back 180 degrees to complete the procedure. Caution should be exercised not to advance unless the lumen appears circular on the screen. A crescentic ureteral lumen would indicate excessive forward force from the ureteroscope resulting in buckling of the ureter. Ureteral stents are positioned with endoscopic and fluoroscopic assistance after ureteroscopy in children. Indwelling double-pigtail ureteral or coil retention stents are available in diameters from 4.7 to 7.0 Fr, with lengths varying from 8 to 30 cm. One could measure the preoperative length either by making a note of the centimeter increments calibrated on the open-ended ureteral catheters or by measuring a portion of the wire between the renal pelvis and bladder. Percutaneous Nephrolithotomy The important adjacent anatomic structures in percutaneous renal stone removal are the diaphragm and pleura, liver, spleen, and right and left colon. Potential perforation of these organs by the puncturing needle and dilators must be considered. Puncturing just adjacent to the lateral border of the paraspinal muscles may offer a consistently safe route of entry into the renal collecting system. Difficulty arises in patients with spinal dysraphism, renal ectopia, or horseshoe kidneys. More commonly, the anterior calyces are irregularly arranged about 70 degrees anterior to the frontal plane of the kidney. The posterior calyces are more regular in position and lie about 20 degrees posterior to the frontal plane. The major segmental arteries lie in a relatively deep plane within the kidney, running close to the infundibula in the hilar area. The interlobar arteries travel between the medullary pyramids. 1 Cystoscopic and fluoroscopic placement of an open-ended or balloon occlusion ureteral catheter may be completed prior to turning the patient into the prone position. Percutaneous access may be provided by the urologist or by the radiologist. Close communication with the radiologist allows the urologist to convey the appropriate puncture site to best address the stone burden. Renal pelvic stones are accessible via a mid- or upper pole puncture. Additional access sites may be necessary to achieve a stone-free state. Percutaneous access ideally should enter the calyx end-on rather than side-on, avoiding the interlobar arteries, which can cross the infundibula. A retrograde ureteral stent or a percutaneous puncture with a 21-gauge trocar point needle may be used to opacify the collecting system. The appropriate calyx is accessed, and then a 0.038-in. floppy tip guidewire is passed through the needle into the collecting system and preferably down the ureter. Torque catheters may facilitate guidewire advancement. A stiffer working wire is passed in addition to the safety wire. Serial dilation or balloon dilation is achieved and a working sheath only large enough to accommodate the nephroscope is moved into position. The use of normal saline as irrigation fluid during the percutaneous nephrolithotomy is recommended. Electrohydraulic, ultrasonic, or pneumatic lithotripsy are options for stone fragmentation. The ultrasonic probe offers retrieval of debris with suction as well ( Fig. 97-1 and Fig. 97-2). Flexible nephroscopy can be used to investigate areas that are difficult to negotiate with the rigid scope. A nephrostomy tube with a Council-type distal port is placed at the termination of the procedure. A nephroureteral stent may be placed if one has a concern over postoperative edema at the ureteropelvic junction. A new guidewire may be replaced through the indwelling Council tip nephrostomy tube if a second-look procedure is necessary.

FIG. 97-1. Preoperative preparation with endoscopic instruments, needles, wires, catheters, baskets, forceps, lithotripsy devices, fluoroscopic equipment, and video should be sufficient to complete the designated stone retrieval and evacuation goals.

FIG. 97-2. A: A 3-cm right renal stone in a patient with a history of cysteine urolithiasis is shown. B: This is an intraoperative image after the successful evacuation of the entire stone burden using the pneumatic and ultrasonic lithotriptors; a ureteral balloon occlusion catheter is lodged at the ureteropelvic junction to prevent distal migration of stone fragments.

Laparoscopy After decompression of the stomach with a nasogastric tube and Foley drainage of the bladder, the patient is placed in 30 degree Trendelenberg. Obtaining the pneumoperitoneum may be achieved with the needle (Veress) or open (Hasson) method. Fewer complications have been reported with the Hasson technique, which may be easily accomplished by placing a 2-cm curvilinear incision within the inferior lip of the umbilicus. 15 The incision is then carried down to expose the rectus fascia just inferior to the umbilicus. Stay sutures are placed on each side of the midline, the fascia is elevated, and the linea alba is incised in the midline. The peritoneum is then elevated and incised. A blunt trocar is then directed through the defect, the retaining flange is activated, and insulation is initiated. Insufflation should progress at a rate of less than 1 L/min, while intraabdominal pressure is maintained below 15 mm Hg. 9A,17 The intraabdominal cavity is inspected to rule out any injury to the underlying viscera from needle or trocar placement. The urachus in the midline and the medial umbilical ligaments on either side of the urachus are identified. The vas deferens, if present, traverses the medial umbilical ligament and enters the internal inguinal ring at its medial border. The spermatic vessels may be seen traveling toward the internal inguinal ring laterally. Blind-ending spermatic vessels are pathognomonic for an absent testis ( Fig. 97-3). Additional working ports can be obtained if intentions extend beyond purely diagnostic laparoscopy, such as division and clipping of the spermatic vessels in a first-stage Fowler-Stephens orchiopexy. The CO 2 is allowed to escape from the port site after the fascial edges are grasped with a Kocher or Allis clamp. A figure-of-8 absorbable suture for the fascial closure and smaller absorbable suture for the subcutaneous tissue are used.

FIG. 97-3. Potential laparoscopic findings in the pelvis of a patient with a history of nonpalpable testis.

OUTCOMES
Complications Cystoscopy is an invasive procedure in neonates, infants, and children requiring anesthesia. Infection and damage to the urethra or bladder are potential complications. Complications of ureteroscopy may be immediate or delayed in recognition. Submucosal passage of guidewires or instruments may occur due to the lack of a distinct ureteral submucosa, especially in areas of inflammation secondary to an impacted stone. Ureteral perforation with extravasation of urine and irrigant is an indication for achieving proper drainage with a stent and returning to the operating room another day. During stone manipulation, when engaging a fragment in the forceps or baskets, telescoping of the ureter should be prevented to prevent a catastrophic ureteral avulsion injury. In adults, 3% to 11% of patients undergoing ureteroscopic stone procedures have developed strictures secondary to stone impaction, urinary extravasation, mechanical trauma, ureteral ischemia, and thermal injury. 13 A follow-up intravenous pyelogram 2 to 3 weeks postoperatively should rule out poorly draining renal units, residual fragments, or early ureteral strictures. Dilation of the ureteral orifice and transmural tunnel has in some cases produced vesicoureteral reflux or resulted in ureteral strictures. The reported risk of reflux has varied from 0 to 50%. 1A,8 Complications of percutaneous renal surgery include pneumothorax and hydrothorax, which may be evaluated by moving the fluoroscopic arm over the ipsilateral thorax intraoperatively and may require decompression with a thoracentesis or thoracostomy tube. Hemorrhage from the kidney may be managed with a Kaye balloon catheter for 20 to 30 minutes to tamponade the collecting system or parenchymal tract. If persistent hemorrhage occurs, angiography with possible embolization of an iatrogenic arteriovenous fistula may be necessary. An intraperitoneal injury of the colon leads to an exploratory laparotomy. A nephrostomy tube that has traversed the collecting system and terminated in the colon should be pulled back into the urinary tract; and the patient should receive hyperalimentation. The nephrostomy tube that traverses the colon and terminates in the renal pelvis should be removed, and proper stenting or percutaneous drainage should be completed. Stenosis of the ureteropelvic junction and sepsis should also be considered as possible complications of a percutaneous nephrolithotomy. Complications of laparoscopy include difficulties with ventilation, impaired venous return, hemorrhage secondary to vascular injury, hollow viscus or solid organ injury, air embolism, hypercarbia, hernia at the port sites, and subcutaneous or mediastinal emphysema. 9A,17 Results Advancements in pediatric endourology have been dramatic over the years. The development of equipment more compatible with the size limitations of the neonate, infant, and child has facilitated application of cystoscopic, ureteroscopic, nephroscopic, and laparoscopic procedures to the pediatric population. CHAPTER REFERENCES
1. Anplatz K. Percutaneous nephrostomy. In: Castenada-Zunisa WR, Tadavarthy SM (eds.), Interventional Radiology, 2nd ed. Baltimore: Williams & Wilkins, 1992, 814–835. 1A. Caione P, DeGennaro M, Capozza N, et al. Endoscopic manipulation of ureteral calculi in children by rigid operative ureterorenoscopy. J Urol 1990;144:484–485. 2. 3. 4. 5. 6. 7. 8. 9. Clayman RV. Pediatric laparoscopy: quo vadis? A view from the outside. J Urol 1994;152:730–733. [Deleted in proofs.]. [Deleted in proofs.] Colodny AH. Laparoscopy in pediatric urology: too much of a good thing. Semin Pediatr Surg 1996;5(1):23–29. El-Damanhoury H, Burger R, Hohenfellner R. Practical pediatric nephrology. Surgical aspects of urolithiasis in children. Pediatr Nephrol 1991;5:339–347. Harmon EP, Neal DE, Thomas R. Pediatric urolithiasis: review of research and current management. Pediatr Nephrol 1994;8:508–512. Hill DE, Segura JW, Patterson DE, Kramer SA. Ureteroscopy in children. J Urol 1990;144:481–483. Hoebeke P, Van Laecke E, Raes A, Vande Walle J. One hundred consecutive cystoscopic examinations in children: indications and results. Eur Urol 1996;30:112–118.

9A. Kavoussi LR. Pediatric applications of laparoscopy. In: Clayman RV, McDougall EM (eds.), Laparoscopic urology. St. Louis: Quality Medical Publishers, 1993, 209–224. 10. 11. 12. 13. 14. Laufer J, Boichis H. Practical pediatric nephrology. Urolithiasis in children: current medical management. Pediatr Nephrol 1989;3:317–331. Lim DJ, Walker D, Ellsworth PI, et al. Treatment of pediatric urolithiasis between 1984 and 1994. J Urol 1996;156:702–705. Lobe TE. Advances in pediatric endoscopy. Compreh Ther 1993;19(6):281–285. Motola JA, Smith AD. Complications of ureteroscopy: prevention and treatment. AUA Update Series 1992;9:21. Perez LM, Smith EA, Parrott TS, Broecker BH, Massad CA, Woodard JR. Submucosal bladder neck injection of bovine collagen for stress urinary incontinence in the pediatric population. J Urol 1996;156(2):633–636. 15. Peters CA. Laparoscopy in pediatric urology: challenge and opportunity. Semin Pediatr Surg 1996;5(1):16–22. 16. Rames RA, Aaronson IA. Migration of Polytef paste to lung and brain following intravesical injection for correction of reflux. Pediatr Surg Int 1991;6:239–240.

17. See WA. Selection and preparation of the patient for laparoscopic surgery. In: Clayman RV, McDougall EM (eds.), Laparoscopic urology. St. Louis: Quality Medical Publishers, 1993, 3–27.

Chapter 98 Pediatric Urethral Diverticulum Glenn’s Urologic Surgery

Chapter 98 Pediatric Urethral Diverticulum
Hrair-George O. Mesrobian

H.-G. O. Mesrobian: Division of Pediatric Urology, Medical College of Wisconsin, and Section of Urology, Children's Hospital of Wisconsin, Milwaukee, Wisconsin 53226.

Diagnosis Indications for Surgery Alternative Therapy Description of Procedures Endoscopic Open Procedures Outcomes Complications Results Chapter References

The topics of female urethral diverticula and megalourethra are discussed elsewhere in this textbook. The purpose of this chapter is to focus on congenital and acquired pediatric urethral diverticular disease. The former occurs primarily in conjunction with anterior urethral valves and, to a lesser extent, anorectal anomalies, whereas the latter is most commonly encountered following urethral surgery such as repair of hypospadias. Unlike the adult population, most pediatric urethral diverticula are found in boys. The diverticulum is usually posterior but an anterior location has been described in conjunction with penoscrotal transposition. 4 Because of the diverse etiology the pathophysiology is variable. A common theme is the absence (or thinning) of the ventral spongy tissue and thus lack of support leading to outpouching of the urothelium and diverticulum formation.

DIAGNOSIS
Ballooning of the ventral urethra during micturition with or without a urinary tract infection is one of the most common forms of presentation. A voiding cystourethrogram is the most expeditious and accurate means of establishing the diagnosis. In addition, the presence or absence of distal obstruction and the location and size of the diverticulum are ascertained. Fluoroscopy is desirable to allow for the dynamic evaluation of the effect of the diverticulum on urinary flow: the posterior leaflet of a congenital diverticulum can impinge on the urethra, leading to a valve-like obstruction. A renal ultrasound to assess the kidneys is indicated because of the occurrence of hydronephrosis in some patients with anterior urethral valves or diverticulum.

INDICATIONS FOR SURGERY
Most patients with a congenital urethral diverticulum present with symptoms the nature of which warrants surgical treatment. Some patients with an acquired diverticulum may be asymptomatic. Indications for surgery can include hydronephrosis and/or urinary retention (usually encountered in newborns), or problems related to postmicturition dribbling of urine, dysuria, infection, stone formation, and, in older patients, ejaculatory difficulties. The mere presence of a small asymptomatic diverticulum may not by itself represent an indication for surgery.

ALTERNATIVE THERAPY
When definitive treatment is indicated, endoscopic or open surgical procedures may be employed. In the newborn or infant presenting with renal failure and or urosepsis, a period of urethral catheter drainage may be required. Percutaneous nephrostomy tube insertion when hydronephrosis is present is rarely necessary. The principles governing the management of the newborn with anterior urethral valves are similar to those governing the management of the newborn with posterior urethral valves.

DESCRIPTION OF PROCEDURES
Endoscopic In cases of anterior urethral valvular obstruction, the distal lip of the diverticulum is hooked with a hot endoscopic knife and incised in the midline. Care is taken not to incise the floor of the diverticulum for fear of creating a urethrocutaneous fistula because the spongy tissue is deficient. The incision may relieve the obstruction but the diverticulum will remain. Further treatment, if necessary, will require an open surgical procedure. Open Procedures 1. Creation of a vesicostomy may be the procedure of choice in the newborn with an anterior urethral valve and a small urethral lumen, an approach not too dissimilar to patients with posterior urethral valves. Rushton et al. have demonstrated the efficacy of this approach in four patients in whom undiversion at a later date was combined with transurethral incision of the obstructing leaflet of the diverticulum 3. The technique of vesicostomy creation and closure is discussed in conjunction with the management of patients with posterior urethral valves. 2. Repair of the diverticulum can consist of excision of the redundant urethra and primary urethroplasty or a plication procedure as described by Heaton et al. 1. The latter approach is not indicated when distal obstruction is present. The site of the obstruction can be at the distal lip of the diverticulum or, in patients who have undergone a hypospadias repair, at an anastomotic site or at the level of the meatus. When obstruction is present or suspected, its repair can be incorporated with the repair of the diverticulum. In fact, the diverticular tissue can be mobilized and used as a vascularized flap to enhance the caliber of the stenotic area. 2 In general, following a preliminary cystoscopy, a circumcision type of incision is made and the avascular layer overlying the tunica albuginea is entered and developed. The skin is then reflected to the penopubic angle dorsally and the penoscrotal angle ventrally. Care must be taken during the dissection over the diverticulum so as to avoid entering it. A Foley catheter of appropriate size can help identify the landmarks. Although one can utilize a ventral midline incision to expose the diverticulum, the exposure obtained by reflecting the penile skin sets the stage for subsequent maneuvers during closure. For example, a deepithelialized skin flap can be incorporated in the closure over the urethral suture line to minimize the risk of a urethrocutaneous fistula developing 2. At this point, sterile saline is gently injected alongside the Foley catheter while an assistant applies perineal pressure. This allows delineation of the full extent of the diverticulum. In the absence of distal obstruction, the redundant diverticulum is incised and excised. Care must be taken not to excise too much so as not to risk narrowing the lumen. The edges of the urethra are then approximated transversely with a running suture of 6-0 or 7-0 PDS. Every attempt is made to invert the edges during the closure. A second imbricating layer of interrupted Lembert sutures is desirable but not essential. In the presence of distal obstruction, the diverticulum can be mobilized after incision of one side, thus raising a pedicled flap ( Fig. 98-1). The strictured area of the urethra is then incised in the midline for a length sufficient to expose a normal caliber of the lumen. The mobilized portion of the diverticulum can then be laid over the opened urethral stricture, thus enhancing its caliber. Care is taken to avoid devascularizing this flap.

FIG. 98-1. (A) Urethral diverticulum mobilized over pedicle and distal strictured area incised. (B) Diverticular flap incorporated into urethroplasty.

Another approach consists of plicating the diverticulum without excising it. This is done by exposing the diverticulum as described above. Inverted Lembert sutures are then placed in a vertical row, moving proximal to distal ( Fig. 98-2). A second layer can also be incorporated. The urethra is calibrated with bougie à boule to estimate the size of the resulting lumen. This is an important step because one cannot visually appreciate the size of the urethra resulting from the placement of the inverting sutures. The absence of a suture line eliminates the risk of a urethrocutaneous fistula. In addition, the imbricated tissue may provide some needed ventral support to the urethra.

FIG. 98-2. Urethral imbrication for acquired diverticular disease in the absence of distal obstruction.

With either approach, it is important to provide a well-vascularized layer of tissue to cover the primary closure or the imbricated area, while avoiding overlapping suture lines. A deepithelialized skin flap or a tunica vaginalis wrap is desirable. 2,5 The choice of the best alternative is usually dictated by the local anatomy and availability of tissue. The catheter is usually left indwelling for 5 days or so when a primary repair has been performed and for a much shorter time with the plication procedure. The procedures that are necessary for the repair of urethral diverticula associated with anorectal malformations are beyond the scope of this chapter and are generally performed in conjunction with the staged repair of these anomalies.

OUTCOMES
Complications The immediate postoperative complications of infection and bleeding can be avoided by adhering to general principles of hypospadias repair. The judicious use of antibiotics is desirable. The two most common complications consist of urethrocutaneous fistula formation or urethral stricture development. Results As these procedures apply to a heterogeneous group of diseases, the outcome is to a great extent dependent on the disease process itself. For example, the prognosis in the infant presenting with anterior urethral valves (diverticulum) may be largely determined by the level of renal function. The long-term results of the surgical procedures described in this chapter are excellent. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Heaton BW, Snow BW, Cartwright PC. Repair of urethral diverticulum by plication. Urology 1994;44(5):749. Mesrobian HG. The addition of three variants of subcutaneous flaps can improve the results of urethroplasty. Pediatr Surg Int 1991;6:122. Rushton HG, Parrott TS, Woodard JR, Walther M. The role of vesicostomy in the management of anterior uethral valves in neonates and infants. J Urol 1987;138:107. Shanberg AM, Rosenberg MT. Partial transposition of the penis and scrotum with anterior urethral diverticulum in a child born with the caudal regression syndrome. J Urol 1989;142(4):1060. Snow BW. Use of tunica vaginalis to prevent fistulas in hypospadias surgery. J Urol 1986;136:861. Winslow BH, Vorstman B, Devine CJ Jr. Urethroplasty using diverticular tissue. J Urol 1985;134:552.

Chapter 99 Posterior Urethral Valves Glenn’s Urologic Surgery

Chapter 99 Posterior Urethral Valves
Alan B. Retik

A. B. Retik: Department of Urology, Children's Hospital, Boston, Massachusetts 02115.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Posterior urethral valves (PUVs) are being diagnosed with increased frequency, primarily due to their prenatal detection. Improvements in endoscopic technology have permitted definitive treatment of posterior urethral valves in all age groups. This chapter reviews the surgery of valve ablation, along with results and potential complications. Posterior urethral valves were described by Young in 1919, 11 emphasizing the common type 1 valves (Fig. 99-1). He also described the rare type 3 valves, with diaphragmatic appearance, and type 2 valves, which are not functionally significant. Most valves are of the type 1 variety. Treatment of PUV was initially carried out endoscopically by Young through a cystostomy, followed by rupture of the valves with sounds, then endoscopic punch ablation. The beginning of electroresection is credited to Randall in 1921. 8 In children, valve ablation remained an open procedure as described by Gross until smaller endoscopic instruments became available. Primary endoscopic valve ablation became a more acceptable procedure as the risks of urethral injury from instrumentation were reduced.

4

FIG. 99-1. (top) Cystoscopic appearance of obstructing type 1 posterior urethral valves. The valve leaflets curve from the verumontanum distally and upward to meet as the anterior aspect of the valve leaflet. (bottom) Lateral view of the prostatic urethra with posterior urethral valves illustrating their three-dimensional orientation, including the lateral and anterior (dorsal) aspects of the leaflets.

DIAGNOSIS
The boy with obstructive valves may present at any age depending in part on the severity of the obstruction. This entity is diagnosed prenatally very often and provides the opportunity for intervention in the neonatal period. Sepsis and azotemia are common reasons for infants with PUV to come to medical attention; voiding dysfunction, hematuria, and urinary tract infections are common symptoms in older boys. Valves cause a varying degree of urinary tract obstructive changes, ranging from severe hydronephrosis with renal dysplasia to relatively normal upper urinary tracts. 5 Diagnosis is predominantly radiographic; a carefully performed voiding cystourethrogram is essential. The classic appearance is of a dilated and elongated posterior urethra, a cutoff of urethral caliber just proximal to the membranous urethra, often bladder neck hypertrophy, and occasionally the leaflets are visible as filling defects. Many radiographic variations have been described. 6

INDICATIONS FOR SURGERY
Primary valve ablation may be accomplished safely in most instances and certainly in the older child. In cases of an acutely ill newborn or infant, prompt management of acidosis, sepsis, and fluid and electrolyte derangements is essential. Preoperative bladder drainage with a fine feeding tube per urethra is often necessary in the azotemic patient. The extent of fallen serum creatinine with catheter drainage may be a useful prognostic indicator of renal reserve. Appropriate imaging studies are obtained when the child may safely tolerate them and include a voiding cystourethrogram and a functional/anatomic study of the kidneys.

ALTERNATIVE THERAPY
Other than temporizing urinary diversion in the critically ill patient, there is no alternative therapy.

SURGICAL TECHNIQUE
In general, the limiting size of the cystoscope employed is the caliber of the fossa navicularis. The urethra is dilated gently with sounds and the appropriately sized, well-lubricated cystoscope is passed under direct vision into the bladder. In general, a 7.5- or 8-Fr instrument is used in infants and a 10-Fr cystoscope may be used in the older child. Visualizing PUV is best done with the bladder full and the tip of the cystoscope at the external sphincter. The bladder is creded, which fills the valve leaflets to depict the obstruction. The position, degree of obstruction, and the thickness of the valves should be noted. The prostatic urethra is often dilated while the bladder neck appears to be thickened and the opening narrowed. Bladder trabeculation, diverticula, and ureteral orifice abnormalities should be sought and noted. There is less edema and viability if the valves are ablated prior to an extensive examination of the bladder. The goal of transurethral ablation of PUV is to disrupt the continuity of the valve leaflet rendering it nonobstructive. An appreciation of the three-dimensional anatomy of valves is important in achieving this. The medial edge of the valve provides the integrity of the leaflet that snaps into the urethral stream to obstruct it. This edge is an arc from one side of the verumontanum, up along the urethral wall angled distally, curving around anteriorly (at the roof), and returning in a symmetric curve to the verumontanum (Fig. 99-1). A significant component of the obstruction is anterior as well as lateral. Disrupting the continuity of the curve makes the entire leaflet incontinent, as would tearing a sail from its edge toward its attachment. Valve ablation is performed usually with the wire insert of a 3-Fr ureteral catheter. This is especially helpful in small infants because it may be passed through a 7.5or 8-Fr infant cystoscope and still permit flow of irrigant during the procedure. The wire stylet is cut in such a way as to protrude from the end of the catheter. The other end is clamped to the cautery electrode. The plastic ureteral catheter itself acts as the insulator. The catheter is advanced through the cystoscope and up to the valve before the wire is advanced. The wire is then advanced 2 to 3 mm and pushed just into the valve leaflet. Fulguration is carried out with the current set at 25 on pure cut (Fig. 99-2). On applying the wire at 5 o'clock and 7 o'clock, a segment of the leaflet is ablated, leaving the valve incompetent. The remnant leaflets flutter with the expressed urine flow (Fig. 99-3). Fulguration should be repeated if residual obstructive tissue is noted. Care must be taken not to injure the verumontanum or the sphincter. Incision at the 12 o'clock position also disrupts the anterior component of the valve and in some instances may be very effective. An alternative means of

valve ablation is with a no. 3 Bugbee electrode used in a similar manner.

FIG. 99-2. Valve ablation using the Bugbee electrode, shown engaging the valve leaflet just before current is applied. The electrode is pushed gently into the valve.

FIG. 99-3. Appearance of the ablated valve. Disrupting the continuity of the arc of the leaflet has relieved the obstruction.

During the procedure, it is important to carefully inspect the bladder noting the usual characteristics of the bladder and the presence of diverticula, configuration of ureteral orifices, and so forth. After removal of the cystoscope, an expression stream should be full. An expression cystourethrogram may be done in the operating room at the time of fulguration if there is any doubt as to the adequacy of the procedure. A small catheter is left in place for several days. Mild or minimally obstructive posterior urethral valves may be treated on an ambulatory basis without the placement of a catheter. Voiding cystourethrography is performed after the catheter is removed to confirm the adequacy of the valve ablation. Perineal urethrostomy may be necessary for valve ablation in the small infant whose anterior urethra cannot accommodate the infant cystoscope. This is rarely necessary. The tip of an infant sound provides a guide to incise onto the perineal urethra, which is exposed with traction sutures. After valve ablation, an indwelling urethral catheter allows the perineal urethrostomy site to close without any formal suture closure. I have not seen an increased incidence of strictures with perineal urethrostomy. Transvesical ablation of valves has been described for use in patients who have had temporary vesicostomy. 12 The leaflets are well visualized from above because the prostatic urethra is dilated; their anatomy is particularly evident. The sphincter cannot be seen, however, and great caution must be exercised when a transvesical approach is employed. Several methods for valve ablation have been suggested to avoid the use of the cystoscope in small infants. These have included an insulated crochet hook with an electrically active inner curve with which the valve leaflet is engaged and cut. 10 A fluoroscopically guided balloon avulsion technique has also been suggested.3 I have found that in virtually all cases a direct vision technique is safely applicable with its attendant precision of valve ablation.

OUTCOMES
Complications Three principal complications may result from primary valve oblation: failure to relieve obstruction, sphincteric damage, and urethral injury with subsequent stricture. Persistent obstruction after valve ablation becomes a serious problem when it remains unrecognized for a prolonged period. An adequate voiding cystourethrogram usually detects significant remaining obstruction, and a repeat valve fulguration is indicated. More subtle findings would include recurrent infections, incomplete emptying, diminished stream, worsening upper tract changes, or high-pressure voiding (if urodynamics has been performed). A thorough initial valve ablation followed by a careful assessment of the efficacy of the treatment should avoid this complication. It should be kept in mind, however, that it is better for the child to undergo repeat valve ablation than to require reconstructive surgery to repair a damaged sphincter. Sphincteric damage is a direct result of damage to the external sphincter and may result in incontinence. Although most of the incontinence in patients with PUV is secondary to bladder dysfunction and not sphincteric damage, this is a significant complication. Careful urodynamic assessment is essential in these cases. 1 Management of sphincteric injury is best avoided when the surgeon has a thorough understanding of the three-dimensional anatomy of valves and their relationship to the sphincter. Several brief applications of the electrode to the valve leaflets are much safer than one mighty cut. Furthermore, adequate vision is essential. If the field of view is obscured by blood, placing a catheter and returning another day is in the patient's best interest. In my experience, sphincter injury is rare. Urethral strictures due to instrumentation have become much less common with improved infant cystoscopes. I have not experienced the high incidence of strictures reported elsewhere. 7 It is imperative to use an appropriately sized, well-lubricated instrument, with great gentleness in all situations. Advancing the cystoscope under direct vision avoids false passages and urothelial damage. Minimizing movement of the instrument and keeping the procedure as brief as possible are also helpful. Use of the fine-wire or Bugbee electrode appears to be associated with fewer urethral strictures than the resectoscope loop. 2 For teaching purposes, I have routinely employed a video attachment to the cystoscope, permitting excellent viewing of the procedure by both operator and observers. A small, loosely fitting, nonreactive catheter is preferred for postoperative drainage. Valve ablation in the setting of a patient with urinary diversion can be performed, but a urethral stent must be left in place for 3 to 5 days to avoid coaptation of the urethral walls. Should there be evidence for persisting or recurring obstruction, careful evaluation is needed to differentiate between a stricture and persisting valves. Results The technical results of valve ablation are excellent. The overall results in regard to renal function and patient survival are directly related to the status of preexisting renal function. In addition, a small percentage of patients may develop “valve” bladders, which are poorly compliant, high-pressure organs that require intensive pharmacologic treatment and, in some instances, bladder augmentation. The surgical treatment of posterior urethral valves requires thorough knowledge of the three-dimensional anatomy of the valve, careful radiographic imaging, properly maintained and used infant instruments, good vision, and delicate technique. A satisfactory result can be anticipated, but a recognition of the potential complications is necessary.

CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Bauer SB, Dieppa RA, Labib KK, Retik AB. The bladder in boys with posterior urethral valves: a urodynamic assessment. J Urol 1979;121:769–773. Crooks K. Urethral strictures following transurethral resection. J Urol 1982;127:1153–1154. Diamond DA, Ransley PG. Fogarty balloon catheter ablation of neonatal posterior urethral valves. J Urol 1987;137:1209–1211. Gross RE. The surgery of infancy and childhood. Philadelphia: WB Saunders, 1953;730–739. Hendren WH III. Posterior urethral valves in boys: a broad clinical spectrum. J Urol 1971;106:298–307. Jorup S, Kjellberg SR. Congenital valvular formations in the urethra. Acta Radiol 1948;30:197–208. Myers DA, Walker RD III. Prevention of urethral strictures in the management of posterior urethral valves. J Urol 1981 26:655–657. Randall A. Congenital valves of the posterior urethra. Ann Surg 1921;73:477–485. Retik AB, Burke C. Urethral valves. In: Libertino JA, Zinman L, eds. Reconstructive urologic surgery. Baltimore: Williams and Wilkins, 1977;287–292. Williams DI, Whitaker RH, Barratt TM, Keeton JE. Urethral valves. Br J Urol 1973;45:200–210. Young HH, Frontz WA, Baldwin JC. Congenital obstruction of the posterior urethra. J Urol 1919;3:289–365. Zaontz M, Firlit C. Percutaneous antegrade ablation of posterior urethral valves in infants with small caliber urethras: alternative to urinary diversion. J Urol 1986;136:247–248.

Chapter 100 Megalourethra Glenn’s Urologic Surgery

Chapter 100 Megalourethra
Mark P. Cain

M. P. Cain: Department of Pediatric Urology, Indiana University School of Medicine, Indianapolis, Indiana 46202.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Reduction Urethroplasty Repair of Mip Outcomes Complications Results Chapter References

Megalourethra is an uncommon anomaly involving the penile urethra, with fewer than 50 cases documented in the English literature. This congenital dilation of the urethra was initially thought to be a subset of saccular urethral diverticulum, but megalourethra is clearly distinct in pathologic findings, as well as etiology. 1,2 There are two separate hypotheses regarding the etiology of megalourethra and the associated anomalies. One theory is an arrest or failure in the migration and fusion of the mesenchymal elements that form the vascular connective tissue of the corpora spongiosum, resulting in patulous enlargement of the penile urethra. 1,5,8 Alternatively, recent fetal studies have demonstrated the presence of an obstructing glanular urethral plug in association with megalourethra. 10 Variability in the degree and duration of the delay in canalization of the urethral plug could explain the wide range of pathologic changes in both corpora spongiosa and cavernosa. The primary defect identified in megalourethra is insufficient development of the corpora spongiosa. The absence of structural supporting tissue allows massive distention and elongation of the urethra. The dilation of the urethra increases with voiding, resulting in pooling of urine and postvoid dribbling. Distal urethral obstruction has not been identified, and despite the marked dilation of the anterior urethra, the glanular urethra is of normal caliber in most cases. Two categories of megalourethra have been described: scaphoid and fusiform ( Fig. 100-1). The more common scaphoid variant is associated with ventral ballooning and dorsal deviation of the penis during voiding due to absence/deficiency of spongiosal tissue. With fusiform megalourethra, one or both corpora cavernosa are also focally deficient, causing a circumferential ballooning of the urethra. Associated anomalies that are commonly seen with both forms of megalourethra include prune belly syndrome, hydronephrosis, vesicoureteral reflux, renal dysplasia, anorectal and cardiac anomalies, and cryptorchidism. 1,6,8 In most series, the fusiform variant is more commonly associated with morbid conditions. However, death and/or renal insufficiency is also reported in over 60% of patients with scaphoid megalourethra. 1,6,8

FIG. 100-1. A: Scaphoid megalourethra. Corpora spongiosum (CS) is deficient throughout ventral aspect of penile urethra. Corpora cavernosum (CC) is normal. B: Fusiform megalourethra. Both corpora spongiosum and cavernosum are deficient. Note that glans (gl) is normal in both variants.

A third category of megalourethra is the megameatus-intact prepuce (MIP) variant of hypospadias. The penile spongiosal tissue is normal in these patients, and there does not seem to be an increased incidence of associated congenital anomalies.

DIAGNOSIS
The majority of patients present with an obvious penile deformity shortly after birth, but patients with milder anomalies may present later in life with voiding symptoms (enuresis, incontinence, poor stream, urinary tract infection). 1 All patients should undergo radiographic evaluation including renal sonography and voiding cystourethrogram to screen for associated anomalies. Sonography should be obtained early to rule out obstruction of the upper urinary system and/or renal dysplasia. The voiding study should be delayed to avoid the unnecessary risk of urosepsis. Long-term prophylactic antibiotics as well as systemic antibiotics at the time of urethral instrumentation are recommended. 8 Patients with the MIP variant of hypospadias do not need radiographic imaging.

INDICATIONS FOR SURGERY
The primary indication for repair of megalourethra is to improve urinary drainage through the dilated urethra and to improve the cosmesis of the redundant phallic skin. Management of any associated anomalies, including hydronephrosis, often takes precedence over urethral reconstruction. Surgery is generally delayed until 6 to 12 months of age for correction of penile megalourethra. Consideration of gender reassignment has been suggested for surviving newborns with complete absence of both corpora cavernosa. 5 Hypospadias repair is recommended between 6 and 12 months of age for patients with MIP hypospadias.

ALTERNATIVE THERAPY
Due to the fact that this surgery is intended to reconstruct a congenital defect, there is no alternative therapy that is effective.

SURGICAL TECHNIQUE
Reduction Urethroplasty The classic reduction urethroplasty for repair of megalourethra was described by Nesbitt 7 (Fig. 100-2). A single dose of perioperative antibiotic is administered, usually a first-generation cephalosporin. A distal traction suture of 5-0 prolene is placed through the glans, and the glanular urethra is calibrated with sounds to ensure adequate size. A circumferential incision is then made 8 mm below the coronal sulcus. Using sharp dissection, the skin and subcutaneous tissue are carefully degloved from the shaft of the penis to the level of the scrotal fat. This dissection should begin by establishing the plane above Buck's fascia on the dorsal aspect of the phallus prior to dissecting the redundant skin covering the urethra, which will have poorly developed spongiosal tissue.

FIG. 100-2. Nesbitt reduction urethroplasty for megalourethra. The urethra is opened vertically in the midline, followed by excision of the lateral redundant tissue. Reconstruction is carried out over a 12-Fr catheter using two layers of running suture. The penile skin is reapproximated to the coronal margin circumferentially.

A 12-Fr red rubber catheter is placed and a midline urethrotomy is made for the entire length of the involved urethra. The redundant urethra is then excised on each lateral margin, leaving adequate tissue to allow a tension-free two-layer closure. The urethra is then reapproximated using a running 7-0 polyglactin or polyglycolic acid (PGA) suture, locking every three to four sutures to avoid shortening of the suture line. A second layer of spongiosal/subcutaneous tissue is reapproximated using a running 7-0 PGA suture. The urethral lumen can also be tapered by one of the infolding techniques described for the repair of megaureter. Kalicinski 4 described a method of ureteral folding that is easily applied to the megalourethra ( Fig. 100-3). After degloving the skin from the penile shaft, the lumen of the urethra is reduced by tailoring over a 12-Fr catheter by running a continuous 6-0 PGA horizontal mattress suture on the lateral aspect of the entire length of the redundant urethra. This excluded segment is then folded over the ventral surface of the urethra and secured in place with interrupted 6-0 PGA sutures. The Starr technique utilizes interrupted 6-0 PGA sutures in a Lembert fashion to imbricate the urethra over a 12-Fr catheter ( Fig. 100-4).9

FIG. 100-3. Kalicinski's technique for ureteral infolding applied to megalourethra.

FIG. 100-4. Starr procedure of ureteral imbrication applied to megalourethra.

A sleeve-type circumcision is then carried out, and the penile skin is reapproximated to the coronal collar using interrupted 6-0 chromic suture. The catheter is removed and 6-Fr silastic ventriculoperitoneal shunt tubing is placed just inside the bladder lumen to provide urinary diversion. This is secured in place to the glans with two 5-0 prolene sutures and allowed to drain freely into the diaper. The penis is coated with Mastisol, then dressed with two layers of OpSite (Smith and Nephew, Hull, United Kingdom). Repair of MIP Repair of the MIP variant of hypospadias requires both urethral reduction and advancement. The most widely used technique is the pyramid procedure described by Duckett and Keating (Fig. 100-5).3 A 5-0 prolene traction suture is placed through the distal glans, followed by three 6-0 PGA traction sutures marking the lateral margins of the urethra and the ventral midline. These sutures allow maximal exposure of the megameatus and help prevent inadvertent urethrotomies during mobilization of the distal urethra.

FIG. 100-5. Pyramid procedure for repair of megameatus intact prepuce variant of hypospadias. 3

The glans and distal margins of the urethra are infiltrated with 0.5% lidocaine, with 1:200,000 epinephrine for hemostasis. Lateral glanular incisions are then made and extended around the proximal urethral meatus. The urethra and lateral glans wings are then carefully mobilized using scissors, taking care to leave ample tissue on the lateral margins of the urethra. The redundant urethra is then tailored by excising a V-shaped wedge of tissue in the midline. Ventral urethral closure is carried out using two layers of continuous 7-0 PGA suture over a 12-Fr catheter. Glansplasty is performed in two layers, using horizontal 6-0 polyglyconate sutures to approximate the deep subepithelial layer, followed by interrupted horizontal mattress 7-0 PGA to close the epithelium. Interrupted 6-0 chromic suture is used to reapproximate the coronal collar and penile skin. Urethral drainage and dressings are identical to megalourethra repair. Postoperatively, patients are maintained on oral antibiotic prophylaxis (first-generation cephalosporin) and given anticholinergics for bladder spasms as needed. The dressing and silastic urethral stent are removed on postoperative day 5 or 6.

OUTCOMES
Complications There are few reported complications regarding megalourethra repair. The risk of urethrocutaneous fistula is minimized using any of the procedures described in this chapter. There is adequate tissue for multilayer closure. Overlying suture lines are avoided with the sleeve approximation of the penile skin. Late stricture and/or formation of focal diverticuli due to the long suture line are potential late complications. The pyramid procedure for correction of MIP has also been very successful, with minimal risk of fistula, stricture, or glans dehiscence associated with other hypospadias procedures. Results The results for both megalourethra and MIP repair have been uniformly excellent with regard to cosmetic appearance and function. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Appel RA, Kaplan GW, Brock WA, Streit D. Megalourethra. J Urol 1995;135:747–751. Dorairajan R. Defects of spongy tissue and congenital diverticuli of the penile urethra. Aust N Z J Surg 1963;32:209–214. Duckett JW, Keating MA. Technical challenge of the megameatus intact prepuce hypospadias variant: the pyramid procedure. J Urol 1989;141:1407–1409. Kalicinski ZH, Kansy J, Kotarbinska B, Joszt W. Surgery of megaureters: modification of Hendren's operation. J Pediatr Surg 1977;12:183–188. Lockhart JL, Reeve HR, Krueger RP, Glenn JF, Henry HH. Megalourethra. Urology 1978;12:51–54. Mortensen PHG, Johnson JW, Coleman GU, Lirenman DS, Taylor G, McLoughlin MG. Megalourethra. J Urol 1985;134:358–361. Nesbitt TE. Congenital megalourethra. J Urol 1955;73:839–842. Shrom SH, Cromie WJ, Duckett JW. Megalourethra. Urology 1981;27:152–156. Starr A. Ureteral plication, a new concept in ureteral tailoring for megaureter. Invest Urol 1979;17:153–158. Stephens FD, Fortune DW. Pathogenesis of megalourethra. J Urol 1993;149:1512–1516.

Chapter 101 Hypospadias Glenn’s Urologic Surgery

Chapter 101 Hypospadias
Warren Snodgrass

W. Snodgrass: Methodist Children's Hospital, Lubbock, Texas 79410.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Distal Hypospadias Proximal Hypospadias Failed Prior Hypospadias Repair Buccal Mucosa Graft Outcomes Complications Results Chapter References

Hypospadias is a congenital defect resulting from incomplete tubularization of the urethral plate. The cause for this developmental arrest is unknown, but it is probably related to inadequate androgenization before the 20th week of gestation, when urethral formation is completed. Normal urethral development begins with extension of the urethral plate from the urogenital sinus distally to the ventral surface of the genital tubercle. Proliferation of mesenchyme adjacent to the plate creates urethral folds, which ultimately fuse in the midline and thereby tubularize the plate to form the penile urethra. The glanular urethra develops from a combination of ectodermal ingrowth from the tip of the genital tubercle in contact with the urethral plate, and ventral fusion of glanular folds. 4 Failure of the urethral and glanular folds to close, therefore, results in a proximal meatus with the remaining urethral plate open as an epithelial strip to the tip of the glans. This anomaly occurs in approximately 3 per 1000 male births. The meatus may be found anywhere from the perineum to the midglans, although in over 80% of cases it is located distal to the midshaft.

DIAGNOSIS
Hypospadias is usually recognized during the initial physical examination in the newborn nursery. Typically, the ventral prepuce is deficient, creating the appearance of a dorsal hood that only partially covers the glans. The penile raphe is often displaced from the midline, and spontaneous erection may reveal a downward glanular tilt due to inadequate ventral skin. True chordee, which is usually the result of disproportionate development of the dorsal and ventral surfaces of the corpora cavernosa,2 occurs more often with proximal than distal hypospadias, and its severity is difficult to accurately assess preoperatively. Occasionally, a hypospadiac meatus is concealed by a normal prepuce. These boys are often circumcised before referral because the defect is not apparent.

INDICATIONS FOR SURGERY
Modern surgical techniques to correct hypospadias should result in a functionally and cosmetically normal penis that appears to have only been circumcised. Left untreated, the hooded prepuce and meatal anomaly become more obvious with penile growth and may be a source of embarrassment. Furthermore, even glanular hypospadias may be associated with a transverse web of tissue at the dorsal lip of the meatus that deflects the urinary stream. Proximal hypospadias may contribute later to infertility from abnormal sperm deposition. Repair of the urethral defect, straightening of chordee, and circumcision are usually achieved with a single outpatient operation. The optimal age for surgery in most boys is 6 months.

ALTERNATIVE THERAPY
There is no therapeutic alternative to surgical correction. In the past, boys with distal hypospadias, in whom the defect was largely cosmetic, often were only circumcised without repair of the lesion. Given modern techniques and their excellent results, correction of all but the most minor cases of glandular hypospadias is recommended.

SURGICAL TECHNIQUE
Distal Hypospadias Most surgeons performing hypospadias repairs in the past 20 years have relied on a few standard operations designed for specific meatal and glanular defects. The choice between meatal advancement and glanuloplasty (MAGPI), the glans approximation procedure (GAP), meatal-based flaps, and onlay island preputial flaps was made in a given case according to the designation of the meatus as small and mobile or patulous and fixed, and the glans as flat or deeply grooved. It seemed unlikely that one operation would be found to correct the spectrum of meatal and glanular configurations comprising distal hypospadias. However, the tubularized and incised plate (TIP) repair has demonstrated this versatility. 6 TIP Repair First, a circumscribing incision is made 2 mm proximal to the hypospadiac meatus and the penis is degloved to the penoscrotal junction. Artificial erection using normal saline tests for chordee, which persists after release of the ventral skin in approximately 10% to 15% of boys with distal hypospadias and is corrected in this situation by dorsal plication. The tunica albuginea is exposed bilaterally on the dorsum adjacent to the neurovascular bundle and opposite the point of curvature. It is superficially incised longitudinally for a distance of 5 mm without disturbing the erectile tissues. A 6-0 prolene suture closes the tunica albuginea transversely on each side, with the knot buried. Next, the ventral glans is infiltrated with 1:100,000 epinephrine along the borders of the urethral plate. Parallel longitudinal incisions are then made to separate the plate from the glans. Even when the ventral glans surface is flat and the urethral plate appears narrow, these incisions are no more than 6 to 8 mm apart and do not encroach upon glans tissues. The glanular wings are further mobilized laterally for subsequent tension-free closure. The edges of the urethral plate are now retracted laterally and the entire plate is incised with tenotomy scissors from the hypospadiac meatus distally. This incision extends into the submucosal tissues, dividing the urethral plate into two epithelial strips. This is the key step in the procedure, which widens and deepens the plate to enable tubularization without additional skin flaps. Interrupted 7-0 chromic sutures are used to close the urethral plate over a 6-Fr silicone stent with the knots placed outside. Care is taken to avoid making the neomeatus too snug, as usually only one or two stitches is necessary distal to the midglans region. The dorsal surface of the plate is not sutured and reepithelializes without scarring. The entire neourethra is covered with vascularized subcutaneous tissue dissected from the dorsal prepuce and shaft skin and rotated ventrally. Then the glanular wings are closed in the midline with 6-0 chromic vertical mattress sutures. Three stitches are normally required and, again, constriction of the meatus should be avoided. It is not necessary to suture the glans to the neomeatus.

Next, the ventral subcoronal prepuce is reapproximated to complete the mucosal collar. The dorsal prepuce is split longitudinally, which allows ventral midline skin closure to simulate the median raphe. Excess rugated skin is excised so that the final result is a normal-appearing, circumcised penis. A Tegaderm dressing is applied. The stent provides urinary drainage into diapers for 1 week ( Fig. 101-1).

FIG. 101-1. TIP: (A) Horizontal line indicates circumscribing incision to deglove penis. Vertical lines show junction of urethral plate and the ventral glans. (B) Parallel incisions separate urethral plate from glans. (C) Midline incision of urethral plate from meatus to granular tip. (D) Incision has widened and deepened the urethral plate. (E) Plate tubularized over 6-Fr stent. Dorsal subcutaneous tissues are rotated ventrally to cover the repair. (F) Midline closure of glans wings, mucosa collar, and ventral shaft skin.

Proximal Hypospadias TIP Repair The TIP procedure is also used to correct many cases of midshaft and penoscrotal hypospadias. 7 The operative technique is the same as described above, with care taken to preserve the longer urethral plate when degloving the penis. Nesbitt plication is more often required, especially with penoscrotal defects, although many cases of midshaft hypospadias have only skin tethering without true chordee. As with distal hypospadias, midline incision of the urethral plate opens it to a surprising width, allowing simple tubularization ( Fig. 101-2).

FIG. 101-2. Proximal TIP repair. (A) U- shaped circumscribing incision preserves the proximal aspect of urethral plate. (B) Distal urethral plate separated from glans. (C) Entire urethral plate divided in the midline to widen and deepen it. (D) Plate tubularized over catheter.

Preputial Pedicle Flaps In an occasional boy with proximal hypospadias, midline incision of the urethral plate does not widen it sufficiently for primary tubularization. If there is no chordee attributed to the urethral plate, it is preserved as the dorsal surface of the neourethra to which an only-island preputial flap is sewn. This flap can be used along the entire length of the neourethra as described by Hollowell et al., 5 or limited to the penile urethra and anastomosed at the coronal level to the incised and tubularized glanular plate. 7 The latter approach is feasible because the glanular segment of the urethral plate includes ectodermal tissues from the genital tubercle and is usually supple even when the proximal urethral plate is not. Advantages of a combination technique include reduced length of the proximal flap and improved cosmetic appearance of the meatus (Figure 101-3).

FIG. 101-3. Preputial pedicle flap and TIP repair. (A) U-shaped circumscribing incision made to deglove penis. (B) Distal urethral plate separated from glans. Rectangular flap outlined on dorsal inner preputial skin. (C) Urethral plate incised but proximal portion appears narrow. (D) Preputial flap laid onto the proximal urethral plate. Granular plate is primarily tubularized. Junction of the proximal and distal segments is at the corona.

A dysplastic urethral plate may contribute to ventral curvature and division of the plate is necessary in these unusual cases in order to straighten the penis. A tubularized preputial flap 3 is then used to construct the penile urethra in combination with the TIP procedure for the glanular urethra. Failed Prior Hypospadias Repair Boys with failed repairs present a spectrum of challenges. Any hypospadias procedure can result in urethral stricture, meatal regression, or wound dehiscence, and there is often insufficient skin remaining to mobilize flaps for urethral reconstruction. Again, the TIP procedure may be useful in these circumstances as the urethral plate is often supple despite failed MAGPI, GAP, and meatal-based flap repairs, and skin flaps are not required. 6

Buccal Mucosa Graft In severe cases, the urethra is so diseased that part or all of it must be discarded. Free grafts from a variety of sites have then been used as onlay or tubed flaps. The current preference is buccal mucosa, which is easily harvested and is non-hair-bearing. 1 First, the penis is degloved and any residual chordee straightened. The scarred urethra is excised to healthy tissues to prepare the recipient bed before the graft is harvested. As with pedicle flaps, it is preferable to reconstruct the urethra with an onlay patch, as opposed to a tube graft, if the dorsal urethral epithelium is still healthy. A dental retractor exposes the inner cheek and the endotracheal tube is positioned to the opposite side of the mouth. The cheek is painted with betadine solution. Since buccal mucosa grafts do not tend to shrink, the desired size is outlined as a rectangle on the inner cheek with a marking pen. Care is taken to avoid injury to Stenson's duct, which is visible near the third molar. The graft edges are infiltrated with 1:100,000 epinephrine for hemostasis, and then the graft is harvested with sharp dissection superficial to the buccinator muscle. The cheek wound is closed with a running 4-0 chromic suture. Any fat or muscle remaining on the graft is excised. It is then either tubularized or sewn to the dorsal urethral strip using 6-0 chromic suture. The glanular urethra is preferably repaired using the TIP procedure and the two segments of the urethra are joined at the level of the corona. The entire repair is then covered with a tunica vaginalis pedicle 8 mobilized from one testicle through a small scrotal incision. A 6-Fr stent is left indwelling for 2 weeks. It is important to immobilize the patient for the first 5 days after surgery when the critical phase of graft revascularization occurs ( Fig. 101-4).

FIG. 101-4. Buccal mucosa free graft. (A) Rectangular flap outlined on inner cheek surface. (B) Free graft laid onto healthy dorsal urethral tissues.

OUTCOMES
Complications Complications of hypospadias surgery include fistulas, strictures, meatal stenoses, diverticuli, and wound dehiscences. Overall, problems are reported in 7% of boys undergoing TIP urethroplasty for distal hypospadias 6 and in 12% with proximal repairs. 7 The most common complication after TIP repair has been fistulas. These occurred more frequently early in the experience with this technique when a barrier layer over the neourethra was not routinely used. All fistulas have been to the penile shaft without involving the glans and so have been easily closed. No strictures of the neourethra have been reported with this procedure. Cystoscopy in a few patients has shown the urethra to appear normal, without visible scarring from the dorsal incision. Furthermore, stenosis of the meatus is uncommon but could result from overzealous closure of the distal urethral plate or glans. Diverticuli may develop from meatal stenosis or use of too wide a skin flap for urethral reconstruction. Since the TIP repair creates an adequate meatus and does not require skin flaps, it is not surprising that no diverticuli have occurred. Partial or complete dehiscence may result from inadequate mobilization of the glanular wings, creating tension in the glans closure. Large stents can also disrupt the wound, and so use of a 6-Fr stent is routinely advised. The complication rate of onlay island preputial flaps is approximately 8%, also consisting primarily of fistulas, although diverticuli have occurred when too wide a flap is used. 2 In contrast, tubularized preputial flaps have at least a 15% complication rate, which includes stricture of the proximal anastomosis. 5 To date, limited experience combining preputial flaps with the TIP procedure has not led to stricture of the distal anastomosis between penile and glanular neourethral segments, 7 although this remains a possible complication. Problems associated with complex reoperations depend on a variety of factors, but buccal mucosa grafts appear to result in fewer strictures or diverticuli than do other free grafts. 1 Results The goal of modern hypospadias surgery is a functionally and cosmetically normal penis. The TIP procedure creates a neourethra greater than 10 Fr in diameter and a meatus that is vertically oriented and normally positioned. Since the operation essentially recapitulates normal urethral development, it is useful for most hypospadias repairs, both distal and proximal. It may also be combined with other techniques such as preputial flaps or buccal mucosal grafts for complex and redo constructions. With its low complication rate and superior cosmetic results, TIP repair has replaced most other techniques to correct hypospadias. CHAPTER REFERENCES
1. 2. 3. 4. 5. Baskin LS, Duckett JW. Mucosal grafts in hypospadias and stricture management. AUA update series, Lesson XIII, 1994;34:270. Baskin LS, Duckett JW, Ueoka K, Seibold J, Snyder HM III. Changing concepts of hypospadias curvature lead to more onlay island flap procedures. J Urol 1994;151:191. Duckett JW. The island flap technique for hypospadias repair. Urol Clin North Am 1981;8:503. Glenister JW. The origin and fate of the urethral plate in man. J Anat 1954;288:413. Hollowell JG, Keating MA, Snyder HM III, Duckett JW. Preservation of the urethral plate in hypospadias repair: extended applications and further experience with the onlay island flap urethroplasty. J Urol 1990;143:98. 6. Snodgrass W, Koyle M, Manzoni G, Hurwitz R, Caldamone A, Ehrlich R. Tubularized, incised plate (TIP) hypospadias repair: results of a multicenter experience. J Urol 1996;156:839. 7. Snodgrass W, Koyle M, Manzoni G, Hurwitz R, Caldamone A, Ehrlich R. Tubularized, incised plate urethroplasty for proximal hypospadias. Section on Urology, American Academy of Pediatrics, Boston, 1996. 8. Snow BW. Use of tunica vaginalis to prevent fistulas in hypospadias surgery. J Urol 1986;136:861.

Chapter 102 Exstrophy and Epispadias Glenn’s Urologic Surgery

Chapter 102 Exstrophy and Epispadias
Stephen A. Kramer

S. A. Kramer: Section of Pediatric Urology, Mayo Clinic, Rochester, Minnesota 55905.

Diagnosis Indications for Surgery Timing of Procedures Suitability for Closure Epispadias Alternative Therapy Surgical Technique Preoperative Assessment Closure of the Exstrophic Bladder Osteotomy Interval Treatment Bladder Neck Reconstruction Bladder Augmentation Epispadias Management of Urinary Incontinence Genital Reconstruction Outcomes Chapter References

Exstrophy and epispadias are both a part of the spectrum of developmental anomalies of the lower urinary tract and lower abdomen. Both anomalies are associated with failure of fusion of the pubic symphysis, defects in the abdominal wall, and a deficiency in the ventral aspect of the genitourinary system. The classic embryologic model of Muecke21 strongly suggests that bladder exstrophy is caused by overgrowth of the cloacal membrane, which prevents mesenchymal migration and midline fusion of embryonic mesoderm. The various other embryologic defects in this spectrum have been categorized as “split symphysis variants” and are discussed elsewhere. Exstrophy of the bladder is the most common of the midline fusion defects; its estimated incidence is 1 in 10,000 to 1 in 50,000 live births. The anomaly occurs 3 to 4 times as often in males, and there is an increased familial incidence among siblings (1 in 275) and offspring of index cases (1 in 70 live births). 7 This chapter deals with primary surgical closure of bladder exstrophy and correction of epispadias. The techniques for urinary diversion and ureterosigmoidostomy are described elsewhere in this text. The rarity of primary epispadias was documented by Dees, 2 who recorded only 56 patients in a survey of eight major medical centers having a combined admission index of more than 5 million patients. In 1982, we reported 94 patients with epispadias. 7 The male-to-female ratio varies between 3:1 3 and 5:1.11

DIAGNOSIS
Classic exstrophy involves pubic diastasis with outward rotation of the pelvic girdle, wide separation of the rectus abdominis muscles, vesicoureteral reflux, frequent inguinal hernias, anterior displacement of the anus with tendency to prolapse or stenosis, and unilateral or bilateral cryptorchidism. The male external genitalia demonstrate shortening and dorsal curvature of the penis due to wide separation of the crura. Females have a bifid clitoris, short urethra and vagina, stenotic and anteriorly displaced vagina, and divergence of the labia and mons pubis. It is important in evaluating a patient with exstrophy/epispadias to examine the genitalia, abdomen, and perineum thoroughly, carefully documenting the degree of exstrophy, the pliability and size of the patch of bladder mucosa, the position of the urethral meatus, and other associated anomalies.

INDICATIONS FOR SURGERY
The presence of the midline defect of exstrophy and/or epispadias is an indication for surgery. The surgery for exstrophy is staged and will require multiple procedures. In the male, the completed first stage creates incontinent penopubic epispadias. The second stage of repair involves vesical neck reconstruction, ureteral reimplantation to correct reflux, and bilateral inguinal herniorrhaphy. In males, a third and often a fourth procedure is necessary to establish satisfactory penile length and to perform definitive urethroplasty. Augmentation cystoplasty often is necessary in children with small bladders to increase size and improve detrusor compliance. The parents of affected children must be clearly aware of the multiple procedures and possible revisions that may be necessary over a period of years in order to effect complete urinary continence and to provide a satisfactory cosmetic appearance of the external genitalia. Rarely, these procedures are unsuccessful and continent urinary diversion may be necessary. Timing of Procedures Clearly, closure of bladder exstrophy is easiest and most appropriately accomplished at the newborn stage. Neonatal closure has the potential to preserve detrusor function and reduce the occurrence of mucosal metaplasia, which may have neoplastic implications. Also, substantial social benefit is afforded to the child and family when the obvious physical defect is corrected early in life. Approximately 95% of all patients with bladder exstrophy are amenable to closure as newborns. When suitable patients are seen within 48 hours of life, bladder and pelvic ring closure can be performed without osteotomy because of the malleability of the pelvic ring. Bladder closure subsequent to this time often requires osteotomies, and this technique has been associated with a lesser chance of wound dehiscence and improved incontinence.3 Suitability for Closure The initial consideration for the urologist who assesses the neonate with bladder exstrophy is to judge the suitability of the bladder for functional primary closure. The mucosa of the bladder may be smooth and healthy, but often it is polypoid and shows metaplastic changes ( Fig. 102-1). Edema and injury to the smooth and delicate bladder mucosa can be prevented by placing the child in an incubator without a diaper and directing a stream of saline mist onto the bladder to keep it moist. If this noncontact care cannot be arranged immediately or continued, the bladder should be protected by a nonadherent silastic membrane, sterile drape, or plastic wrap to prevent the mucosa from sticking to the clothing or diapers, which can damage the delicate epithelium. Vaseline gauze should not be used because it becomes dry and may lift off the delicate epithelium when removed. The small bladder with a capacity of 5 to 10 ml that demonstrates elasticity and contractility can be expected to develop adequate size and capacity subsequent to successful bladder closure. 7 A small fibrotic bladder patch that is stretched between the edges of a small triangular fascial defect without either elasticity or contractility should not be selected for the usual closure procedure. Bladder augmentation with small bowel, large bowel, or stomach may subsequently be necessary to achieve adequate bladder capacity.

FIG. 102-1. Male neonate with classic bladder exstrophy.

Epispadias Epispadias in males is classified according to the position of the dorsally displaced urethral meatus. The degree of penile deformity and the occurrence of urinary incontinence are related to the extent of the dorsally displaced urethral meatus. In males, epispadias may be penopubic, penile, or glandular. Essentially all patients with penopubic epispadias and two-thirds of those with penile epispadias are incontinent and require vesical neck reconstruction. 11 Patients with glandular epispadias are continent, and surgical reconstruction is confined to urethroplasty, glansplasty, and release of chordee. Most females have complete epispadias, in which the urethral cleft involves the entire length of the urethra and sphincter mechanism. The external genitalia are characterized by a bifid clitoris, flattening of the mons pubis, and separation of the labia. 12

ALTERNATIVE THERAPY
Alternative therapy would include cystectomy and urinary diversion. The defect is sufficiently severe with many long-term complications that nonsurgical therapy is not a viable option.

SURGICAL TECHNIQUE
Surgical attempts to close exstrophic bladders using flaps of abdominal skin, fascia, or both date back to at least the 1850s. 1 The first successful coverage of the bladder was performed in 1871 by Maury, 3 who used a technique in which one skin flap was taken from the abdominal wall and a second flap from the scrotum and perineum. Trendelenburg proposed sacroiliac osteotomies to allow full rotation of the pelvis, which enabled approximation of the pubic rami. This principle of “closing the open book” by use of iliac osteotomies was rediscovered and reported by Schultz. 1 Ansell1 recognized the advantages of early closure, preferably within the first 48 hours of life, because the tissues are elastic and the bony pelvis can be molded readily at this time. Jeffs and Lepor 7 must be credited with bringing together the techniques and modifications of a planned and staged approach, which is used most often today for closure of bladder exstrophy. This approach involves primary neonatal bladder closure, interval treatment, vesical neck reconstruction, and bilateral ureteral reimplantation at age 3 to 4 years and penile elongation and urethroplasty as a single- or two-stage procedure at age 4 to 5 years. Some authors have proposed formal penile elongation and urethroplasty at the time of initial neonatal bladder closure. Theoretical advantages of this lengthy and elaborate reconstruction include the ability to drop the bladder neck deep into the pelvis and obtain a more secure bladder closure. Early urethroplasty may provide increased outlet resistance and thereby enlarge bladder capacity. These advantages must be weighed against the risk of necrosis of the corpora cavernosa or damage to the neurovascular bundles. Duckett 3 described an innovative technique that involves use of the shiny paraexstrophy skin to create flaps to bridge the urethral defect after release of the epispadiac chordee. This procedure is applicable to both males and females and should enhance urinary control by allowing elongation of the urethra and permitting the bladder neck to migrate cephalad to a more normal intrapelvic position. The use of these flaps remains controversial. Some authors have reported an increased rate of urethral stricture when paraexstrophy flaps have been incorporated into the urethroplasty. Newer techniques of penile elongation and urethroplasty have produced superior results, and therefore it may not be necessary to use paraexstrophy flaps to achieve additional urethral length.19,20 Preoperative Assessment Abdominal ultrasonography should be done preoperatively to assess the upper urinary tracts and abdomen. Closure of the Exstrophic Bladder The goal of primary closure is to convert the exstrophic bladder to incontinent penopubic epispadias with free egress of urine while preserving renal function. The child is placed on the operating table with the thighs abducted, but the drapes should be arranged so that the legs can be brought together at the end of the operation. It is important to divert the urine initially to prevent continuous leakage intraoperatively, which results in progressive edema and can contribute to wound dehiscence postoperatively. Plastic feeding tubes of either 3- or 5-Fr caliber are passed into each proximal ureter and anchored to the trigonal mucosa with 5-0 chromic catgut sutures. A marking pencil is used to delineate the mucocutaneous junction of the bladder and urethra with the skin ( Fig. 102-2). A strip of mucosa 2 cm wide, extending from the distal trigone to below the verumontanum in the male and to the vaginal orifice in the female, is outlined for prostatic and posterior urethral reconstruction. In neonates undergoing penile elongation and urethroplasty, the incisions should continue distally around the urethral plate as described for formal penile elongation (see below). Incisions are made as outlined with a no. 15 blade and continued circumferentially around the bladder, and extreme care is taken to avoid the umbilical vessels. At the apex of the incision, a triangle of skin is removed, the umbilicus is ligated with 2-0 silk, and the umbilical hernia is repaired.

FIG. 102-2. Incisions are outlined to delineate mucocutaneous junction of bladder and urethra with skin. (A) Male. (B) Female. (© 1989, Mayo Foundation, Rochester, MN.)

Although controversial, I still prefer the use of paraexstrophy flaps to obtain more tissue for closure of the proximal urethra and bladder neck. The paraurethral incisions are deepened to develop thick paraexstrophy flaps and expose the periurethral muscles and underlying corpora ( Fig. 102-3). Extreme care should be taken

to avoid the pudendal nerves and vessels.

FIG. 102-3. Urethral plate is divided and shiny non-hair-bearing paraexstrophy skin flaps are developed. (© 1989, Mayo Foundation, Rochester, MN.)

The urethral plate is divided distal to the prostatic urethra and verumontanum, thereby separating the prostatic urethra, prostate, and bladder from their intracrural position. This procedure allows the bladder to migrate cephalad and creates a 2- to 3-cm gap between the verumontanum and distal urethral plate. This defect is filled in by the paraexstrophy flaps of shiny skin adjacent to the exstrophy ( Fig. 102-4 and Fig. 102-5). These pedicle flaps are brought together in the midline with fine Vicryl sutures and are anastomosed to the distal mucosa of the prostatic urethra ( Fig. 102-6).

FIG. 102-4. Paraexstrophy flaps are mobilized and moved distally. Corpora is approximated. (© 1989, Mayo Foundation, Rochester, MN.)

FIG. 102-5. Flaps are rotated to cover corpora and elongated penis. (© 1989, Mayo Foundation, Rochester, MN.)

FIG. 102-6. Midline approximation of skin flaps. Mucosa of prostatic urethra is tubularized over a 12-Fr silastic stent. The intersymphyseal band is divided. (© 1989 Mayo Foundation, Rochester, MN.)

Tubularization of the bladder neck is accomplished by suturing the tissue snugly over a 12-Fr silastic stent placed through the posterior urethra and trigone neck. The lateral edges of the flaps are brought together superiorly in the midline to close the roof of the neourethra, creating a penopubic epispadias. This maneuver allows the bladder and prostatic urethra to drop back beneath the bony pelvis. Paraexstrophic flaps also may be used in the female to achieve further lengthening of the urethra. Once the urethra has been tubularized, the 12-Fr stent is exchanged for a 10-Fr stent to remove tension from the suture line. I have found that use of a urethral stent decreases the risk of postoperative bladder neck contracture, as well as improving urinary drainage in the early postoperative period. The bladder edge is dissected sharply from the rectus muscle and freely mobilized from the peritoneum. Sharp scissor dissection is carried down to the subcutaneous fat at the bladder neck. The medial edge of either pubic rami then is identified because this landmark is the key to exposure of the medial borders of the rectus muscles and bladder neck or prostatic tissues. The fibromuscular bar that unites the pubic bone to the bladder base now is divided laterally ( Fig. 102-6). With this mobilization, it will be possible to invert the bladder and establish a functional bladder neck ( Fig. 102-7). It is important to bring these muscles of the urogenital diaphragm toward the midline to provide support for the urethra and bladder neck. Circumferential dissection of the bladder edge should now be complete, with the rectus muscles and rectus fascia identified and developed enough to allow easy placement of sutures. Bladder closure is then accomplished in two layers with running absorbable suture material.

FIG. 102-7. Closure of intersymphyseal band (urogenital diaphragm). Bladder closure is accomplished with two layers of running Vicryl sutures. (© 1989, Mayo Foundation, Rochester, MN.)

Closure of the symphyseal defect is an integral and important part of primary bladder closure. This maneuver tends to approximate the “opened ring” of the sphincteric mechanism. Furthermore, approximation of the symphysis allows easier approximation of the rectus muscles and abdominal fascia in the midline and decreases the risk of subsequent bladder or abdominal dehiscence. A single no. 1 nylon or prolene suture on a cutting needle is placed through the fibrocartilaginous tissues of the pubis in a horizontal mattress fashion to oppose the pubes ( Fig. 102-8). The knot is tied anteriorly to avoid erosion of the posterior urethra. During this maneuver (and often midline abdominal closure), it is important to have an assistant or circulating nurse rotate the newborn's trochanters inward to allow the pubic bones to be approximated in the midline. The suprapubic tube, ureteral stents, and silastic urethral-bladder neck stent exit through the bladder wall separately and are secured at the skin with fine nylon sutures and irrigated to confirm their patency. It is often necessary to widely mobilize and undermine the rectus fascia and subcutaneous tissue to allow a tension-free midline fascial anastomosis. If it is not possible to accomplish a tension-free midline closure, groin flaps should be outlined and used for coverage of the abdominal defect.

FIG. 102-8. Approximation of pubis with horizontal mattress suture tied anteriorly. (© 1989, Mayo Foundation, Rochester, MN.)

Osteotomy In children more than 72 hours old the procedure is started with bilateral anterior iliac osteotomies; both the outer and the inner tables of the ileum are broken. 4 Approximation of the two halves of the pubis by use of iliac osteotomies may improve abdominal wall closure and decrease the chance of wound dehiscence and fistula.7 The fibrocartilage of the pubic symphysis is united in front by horizontal mattress sutures of no. 2 nylon placed directly through the calcified portion of the pubis and tied anteriorly. After completion of the osteotomies, the infant is placed in the supine position and the bladder exstrophy is closed as described above. Regardless of whether bilateral iliac osteotomies are performed, I routinely place the child in modified Bryant's traction for 2 to 3 weeks postoperatively. The Ace bandages are rewrapped daily and the circulatory status of the lower extremities is monitored carefully by us and our colleagues in pediatric orthopedics. The use of this immobilization technique combined with liberal sedation with pediatric belladonna and opium suppositories and urinary diversion via ureteral stents provides the best opportunity to achieve a successful first-stage result without wound dehiscence. The success of this stage is critical and related directly to subsequent urinary continence. Interval Treatment Ureteral stents are removed between postoperative days 7 and 10, the silastic urethral-bladder neck stent is removed at 2 weeks, and the suprapubic tube is removed at 3 weeks. Urethral calibration is performed in the office at the initial 3-month visit and then again every 6 to 12 months. Between bladder closure and 3 years of age, children are followed up every 3 to 6 months with either excretory urography or ultrasonography to rule out hydronephrosis and occasionally with isotope renography to assess renal function. Every 6 months, urine is cultured and the serum creatinine value determined. Essentially all patients with bladder exstrophy have reflux and thus children are maintained on antibiotic suppression indefinitely until their bilateral ureteral reimplantation has been accomplished. Rarely, a child will become dry after neonatal closure alone; however, this has occurred in only a few patients in our series, most commonly females. Bladder Neck Reconstruction An adequate bladder capacity is the single most important factor related to successful operation for urinary continence. 23 Vesical neck reconstruction is deferred until the child is at least 3 years old. Most bladders will enlarge to a capacity of 60 to 75 ml and allow vesical neck reconstruction and bilateral ureteroneocystostomy at this time. In select children with small bladders, urethroplasty may be preformed prior to vesical neck reconstruction in an attempt to increase urethral resistance and enlarge bladder capacity. Young25 described the basic operative technique for reconstruction of the vesical neck for treatment of urinary incontinence. Dees 2 introduced the modification of bilateral triangular reduction of the vesical neck and prostate in the male and the vesical neck and urethra in the female. Leadbetter 16 further modified the Young-Dees operation by demonstrating that proximal reimplantation of the ureters would allow more generous resection and reconstruction in the trigonal region and thereby enhance surgical success. Mollard 20 described a modification of bladder neck reconstruction by incising one muscle flap longitudinally and the other transversely, rolling one flap over the urethra and using the longer muscle flap to encircle the bladder neck. This technique gives a kink to the urethrovesical junction and has contributed to urinary continence in 60% of patients. The patient is placed in the supine position with the head of the operating table lowered slightly. A Pfannenstiel incision affords ready access to the perivesical space. The vesical neck and urethra are mobilized anteriorly and laterally. It is important not to mobilize the posterior aspect of the bladder base and urethra in order to preserve the neurovascular supply to this area. The bladder is opened in the midline, and the incision is extended distally into the urethra, almost to the triangular ligament. Proximal reimplantation of the ureters allows more extensive tubularization of the trigone in reconstructing the bladder neck. Ureteroneocystostomy can be performed by various techniques, but it usually requires combined transvesical and extravesical dissection.

After ureteral reimplantation, parallel incisions are made 12 mm apart, starting on the proximal urethra and extending longitudinally and posteriorly through the base of the bladder (Fig. 102-9).9 Bilateral triangular flaps are made on each side of the urethra and bladder wall ( Fig. 102-9). The newly formed posterior strip of bladder should be approximately 4 to 5 cm in length and 1.5 to 2 cm in width. The mucosa on either side of the bladder wall is excised, except for a central strip of mucous membrane. The neourethra is reconstructed in three layers over a 10- or 12-Fr silastic tube ( Fig. 102-9). The mucosa is closed with interrupted absorbable sutures of chromic catgut or Vicryl, and the musculature is closed in an overlapping fashion. In boys undergoing vesical neck reconstruction, the prostatic urethra affords additional substance for closure and can increase bladder outlet resistance further. Vesicourethral suspension is created to supplement urethral resistance. The neourethra is suspended from the pubic bones or the intrasymphyseal fibrous bands between them if diastasis is present.

FIG. 102-9. Leadbetter procedure. (A) Incisions are started on proximal urethra and extended longitudinally and posteriorly through base of bladder. (B) Bilateral triangular flaps are made on each side of urethra and bladder wall. Mucosa on either side of bladder wall is excised; central strip of mucous membrane is left. (C) Neourethra is reconstructed in three layers over silastic tent. Musculature is closed in overlapping fashion. (D) Triangular flaps, or “dog ears,” should be preserved to ensure adequate bladder capacity. (E) Bladder closure is accomplished in two layers with running absorbable sutures. Suprapubic cystostomy tube and urethral and ureteral stents are left indwelling. (From Kramer SA. Urinary incontinence: surgical treatment of urinary incontinence. In: Kelalis PP, King LR, Belman AB, eds. Clinical pediatric urology. 2nd ed. Philadelphia: WB Saunders, 1985:326–345. © 1989, Mayo Foundation, Rochester, MN.)

The triangular flaps (“dog ears”) that remain after tubularization of the neourethra should be preserved to ensure adequate bladder capacity ( Fig. 102-9). The bladder is closed in two layers with running absorbable sutures, and a suprapubic cystostomy tube is brought out through the bladder dome ( Fig. 102-9). The urethral stent is removed in 10 to 14 days, at which time the suprapubic tube is clamped for a voiding trial. In children with urinary retention or elevated residual volumes, cystostomy drainage may be maintained temporarily or the patient may be started on intermittent self-catheterization. Bladder Augmentation Augmentation cystoplasty has enhanced urinary continence and provides an attractive alternative to urinary diversion in selected patients with small bladder capacities. 10 We have shown that bladder pressures produced after intestinocystoplasty may be related more to the configuration of the bowel than to its intrinsic properties. 5 Detubularization of the bowel with interruption of the strong circular muscle layer of the bowel wall is an important principle. This technique will create a cystoplasty of large volume that will dissipate the pressure waves created by the contractions. Epispadias The treatment of primary penopubic epispadias, or complete epispadias after exstrophy closure, remains a formidable challenge to the pediatric urologist. The surgical management is virtually identical to that of closed bladder exstrophy. The goals of surgical reconstruction in boys are to achieve complete urinary continence and to establish an adequately long, straight penis that is functional for normal sexual intercourse. In the male child with epispadias, there is lateral displacement of the crura and separation of the pubic rami, resulting in a short, broad penis that is tethered to the anterior abdominal wall. The dorsal curvature results from (a) shortage of dorsal penile skin, (b) fibrous bands between the penis and the pubic bones, (c) shortness of the urethral plate in relation to the corpora cavernosa, and (d) intrinsic chordee, implicating intrinsic deformity of the corpora themselves. 13 Female epispadias is a rare congenital anomaly characterized by incompetency of the urethra and vesical neck and total urinary incontinence. 12 In girls, the procedure is directed to producing urinary control and creating a satisfactory cosmetic appearance of the external genitalia. In most patients, the defect involves the external genitalia and manifests by anterior separation of the labia majora and labia minora and bifidity of the clitoris. The vagina may be anteriorly displaced and stenotic, requiring V-Y vaginoplasty to allow satisfactory intercourse. There also may be a deficiency in muscular pelvic support, which predisposes to prolapse of the uterus. Voiding cystourethrography demonstrates absence of the urethrovesical angle, wide open vesical neck, and a short patulous urethra. A thin strip of skin overlies the mons and extends inferiorly between the bifid clitoris. Plastic repair of the external genitalia involves excision of this skin strip, approximation of each medial half of the denuded clitoris, and closure of the mons. Hendren 6 described a single-stage procedure for the treatment of female epispadias with incontinence. In selected girls, it may be necessary to create further resistance to the distal urethra by detachment of the corpora and approximation of the bulbocavernosus and ischiocavernosus muscles of the introitus. Management of Urinary Incontinence In patients with epispadias and urinary incontinence, it is important to undertake endoscopic assessment of the external sphincter and vesical neck to determine the extent of repair necessary to establish urinary control. In boys with penopubic epispadias, or girls with complete epispadias, the external sphincter is deficient between the 10 and 2 o'clock positions and the bladder neck is incompetent. Surgical reconstruction in boys with epispadias and urinary incontinence requires vesical neck reconstruction. 15 Genital Reconstruction Historically, penile elongation was accomplished by release of dorsal skin chordee only or simple V-Y plasty of the urethral plate. It is now clear that excision of superficial scar tissue or dartos fascia or skin procedures alone are insufficient to produce a satisfactory cosmetic result with adequate genital function. 13,18 Even in patients who have undergone urinary diversion, reconstruction of the external genitalia and urethroplasty remain important considerations. 18 The urethra should be adequate for transport of semen, and penile elongation should provide for adequate sexual intercourse. Johnston 8 described a means of penile elongation that involved dissection through a V-Y incision and partial detachment of the crura of the penis from the ischiopubic rami. The mobilized crura were then advanced and approximated in the midline. Woodhouse and Kellett 24 demonstrated significant deformities of the corpora, possibly resulting from their detachment from their inferior attachments to the pubic rami. Therefore, it appears safest to limit penile lengthening to partial detachment of the crura from the ischial pubic rami. Although some patients with balanitic and distal penile epispadias can undergo correction in a single-stage procedure (release of dorsal chordee and urethroplasty using a flip-flap, reverse island flap, or free tube graft technique), most patients with epispadias have the penopubic variety with severe dorsal chordee. Correction of this deformity requires extensive proximal dissection and mobilization of the corpora cavernosa. The corpora must be widely mobilized and dissected proximally as they diverge and continue laterally to their attachments on the pubic rami ( Fig. 102-10). They are partially detached from the ischiopubic bones, and care is taken to preserve the neurovascular supply entering dorsally and laterally. Two recent descriptions of penile elongation and urethroplasty have significantly improved the cosmetic appearance of the penis in patients with bladder exstrophy and epispadias. The Cantwell-Ransley 22 technique involves proximal mobilization and rotation of the corpora such that the urethra is placed ventrally and the neurovascular bundles are moved into their proper position in the dorsal midline ( Fig. 102-11). The urethral plate is tubularized within the glans using a basic Thiersch technique, and irregular edges on the dorsum of the epispadiac glans are excised. A distal vertical incision with horizontal closure brings the neomeatus to the glans tip and allows the glans to be closed in a conical fashion. The Mitchell 19 modification involves complete penile disassembly with division of the glans penis ( Fig. 102-12). Both techniques

leave the urethral plate intact and create a normal ventrally placed urethra.

FIG. 102-10. Corpora are widely mobilized and dissected proximally to their insertion on ischiopubic rami. ( A, B, Kramer SA, Mesrobian HGJ, Kelalis PP. Long-term followup of cosmetic appearance and genital function in male epispadias; review of 70 patients. J Urol 1986;135:543. © 1984, Mayo Foundation, Rochester MN. C, Kramer SA, Jackson IT. Bilateral rhomboid flaps for reconstruction of the external genitalia in epispadias-exstrophy. Plast Reconstr Surg 1986;77:621. By permission of the American Society of Plastic and Reconstructive Surgery.)

FIG. 102-11. (A) Preliminary glansplasty. A longitudinal incision is made through the distal fusion of the lateral wings of the glans and distal urethra. This is closed transversely using 6-0 polydioxanone sutures. The maneuver makes a great difference to the final cosmetic appearance by displacing the terminal urethral meatus slightly ventrally. (B) Dissection of neurovascular bundles and corporal bodies. The neurovascular bundles are now well seen sweeping laterally and ventrally at the midpoint of the corpora. Artificial erection at this stage will show upward angulation of the corpora at this point. The corpora are dissected free of adhesion to the pubic rami for 1 to 2 cm. Further dissection is not advantageous. The neurovascular bundles are elevated from the bodies of the corpora and retracted. (C) Tubularization of urethral plate. The urethral plate is now tubularized over a 10-Fr silastic (Dow Corning, UK) urethral stent or catheter. It is advisable to place a suprapubic catheter in position by guided puncture before closing the urethra. Urethral closure stops at the proximal end of the glans at this stage. The urethra and dartos pedicle are displaced onto the ventral aspect of the penis. (D) Closure of glans. A broad strip of glans tissue is excised on each side and the glanular urethra closure completed. The raw surfaces of the glans now come together on the dorsal aspect using vertical mattress sutures of 6-0 polydioxanone. The corporal rotation and approximation is now reinforced with some additional sutures of 5-0 polydioxanone between the site of the cavernocavernostomy and the glans. (E) Skin closure. The base of the dorsum of the penis is covered by employing distally based triangular skin flaps from the margins of the original midline incision to prevent scarring causing bow stringing to the abdominal wall.

FIG. 102-12. (A) Initial circumscribing incision. (B) Outline of urethral plate, which should be completely marked with methylene blue. (C) Ventral aspect of penis after careful dissection of penile skin from shaft. Note lateral neurovascular bundles. (D) Urethral plate is carefully dissected from corpora. Note distal incision (dotted line) for division of urethral plate. (E) Corporeal bodies and two hemiglans are completely separated. (F) Urethral plate is brought to ventrum after corporeal bodies are rotated medially and sutured together in dorsum. Note that suture line in urethral plate now lies against groove between corporeal bodies. (G) Completion of urethral meatus. (H) Completion of shaft skin coverage. (Adapted with permission from Mitchell ME, Bagli D. Complete penile disassembly for epispadias repair: the Mitchell technique. J Urol 1996;155:300–304.)

The adequacy of release of the chordee is determined by injecting sterile saline into each corpora cavernosa separately and simultaneously because they do not communicate. Persistent dorsal chordee at the time of artificial erection may necessitate ventral dissection with plication or excision of transverse ellipses of tunica albuginea to achieve a straight penis. Rarely, extensive dorsal shortening may require excision of tunica albuginea and placement of a dermal skin graft. Penile lengthening and urethroplasty are most often combined as a single-stage procedure as described above. When penile elongation and urethroplasty are performed separately, it is important to release the dorsal chordee initially and defer urethroplasty for 6 to 12 months. After penile straightening, urethroplasty can be accomplished with either full-thickness grafts or preputial flaps. Meticulous attention to detail is necessary to avoid damage to the verumontanum, which could result in anejaculation. Patients with bladder exstrophy or epispadias have a midline scar and skin defect at the penoabdominal angle. In females, the mons pubis will be deficient. In both males and females, the hair-bearing skin will be displaced laterally, leaving non-hair-bearing skin in the midline. The cosmetic result can be corrected by the use of laterally based bilateral rhomboid flaps which are mobilized and rotated medially ( Fig. 102-13).14

FIG. 102-13. (A) Bilateral rhomboid flaps are outlined from lateral pubic area. (B) Transposition of laterally based rhomboid flaps medially brings hair-bearing skin into midline and reconstructs central dorsal gap in pubic area. (C) Completed repair brings abdominal wall skin onto dorsal aspect of penis and clearly delineates penopubic angle. (From Kramer SA, Jackson IT. Bilateral rhomboid flaps for reconstruction of the external genitalia in epispadias-exstrophy. Plast Reconstr Surg 1986;77:621–629. © permission of the American Society of Plastic and Reconstructive Surgeons.)

OUTCOMES
Recent reviews of functional bladder closure have shown marked improvement in the frequency of successful reconstructions ( Table 102-1). We noted a significant improvement in urinary continence in patients who underwent augmentation cystoplasty in addition to vesical neck reconstruction. 17

TABLE 102-1. Urinary continence after exstrophy

The factors that are critical to the establishment of urinary continence in patients with primary epispadias are similar to those with bladder exstrophy and include (a) deferring vesical neck reconstruction until the child is at least 3 years old, (b) a well-developed bladder with adequate capacity and musculature, and (c) the maturation of the prostate at puberty in boys. 15 The achievement of urinary continence in patients with epispadias is summarized in Table 102-2.

TABLE 102-2. Urinary continence after complete epispadias

CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Ansell JS. Exstrophy and epispadias In: Glenn JF, ed. Urologic Surgery. 3rd ed. Philadelphia: JB Lippincott, 1983;647–663. Dees JE. Congenital epispadias with incontinence. J Urol 1949;62:513–522. Duckett JW, Jr. Epispadias. Urol Clin North Am 1978;5:107–126. Gearhart JP, Forschner DC, Jeffs RD, Ben-Chaim J, Sponseller PD. A combined vertical and horizontal pelvic osteotomy approach for primary and secondary repair of bladder exstrophy. J Urol 1996;155:689–693. Goldwasser B, Barrett DM, Webster GD, Kramer SA. Cystometric properties of ileum and right colon after bladder augmentation, substitution or replacement. J Urol 1987;138:1007–1008. Hendren WH. Congenital Female epispadias with incontinence. J Urol 1981;125:558–564. Jeffs RD, Lepor H. Management of the exstrophy-epispadias complex and urachal anomalies. In: Welsh PC, Gittes RE, Perlmutter AD, Stamey TA, eds. Campbell's urology. Vol. 2. 5th ed. Philadelphia: WB Saunders, 1986;1882–1921. Johnston JH. Lengthening of the congenital or acquired short penis. Br J Urol 1974;46:685–687. Kramer SA. Urinary incontinence: Surgical treatment of urinary incontinence. In: Kedalis PP, King LR, Belman AB, eds. Clinical Pediatric Urology. 2nd ed. Philadelpia: WB Saunders, 1985; 326–345. Kramer SA. Augmentation cystoplasty in patients with exstrophy-epispadias. J Pediatr Surg 1989;24:1293–1296. Kramer SA, Kelalis PP. Assessment of urinary continence in epispadias: Review of 94 patients. J Urol 1982;128:290–293. Kramer SA, Kelalis PP. Surgical correction of female epispardias. Eur Urol 1982;8:321–324. Kramer SA, Mesrobian HGJ, Kelalis PP. Long-term follow-up of cosmetic appearance and genital function in male epispadias; review of 70 patients. J Urol 1986;135:543–547. Kramer SA, Jackson IT. Bilateral rhomboid flaps for reconstruction of the external genitalia in epispadias-exstrophy. Plast Reconstr Surg 1986;77:621–629. Kramer SA, Kelalis, PP. Correction of total in continence in male and female epispadia S. J Pediatr Surg 1981;16:812–816. Leadbetter GW, Jr. Surgical correction of total urinary incontinence. J Urol 1964;91:261–266. Mesrobian H-GJ, Kelalis PP, Kramer SA. Long-term follow-up of 103 patients with bladder exstrophy. J Urol 1988;139:719–722. Mesrobian H-GJ, Kelalis PP, Kramer SA. Long-term follow-up of cosmetic appearance and genital function in boys with exstrophy: review of 53 patients. J Urol 1986;136:256–258. Mitchell ME, Bagli D. Complete penile disassembly for epispadias repair: the Mitchell technique. J Urol 1996;155:300–304. Mollard P. Bladder reconstruction in exstrophy. J Urol 1980;124:525–529. Muecke EL. The role of the cloacal membrane in exstrophy: The first successful experimental study. J Urol 1964;92:659–667. Ransley PG, Duffy PG, and Wollin M. Bladder exstrophy closure and epispadias repair. In: Spitz EL, Nixon HH. Paediatric surgery. 4th ed. London: Butterworths, 1988;620–632. Ritchey ML, Kramer, SA, Kelalis PP. Vesical neck reconstruction in patients with epispadias-exstrophy. J Urol 1988;139:1278–1281. Woodhouse CRJ, Kellett MJ. Anatomy of the penis and its deformities in exstrophy and epispadias. J Urol 1984;132:1122– 1124. Young HH. An operation for the cure of incontinence associated with epispadias. J Urol 1922;7:1–32.

Chapter 103 Congenital Anomalies of the Scrotum Glenn’s Urologic Surgery

Chapter 103 Congenital Anomalies of the Scrotum
David R. Roth

D. R. Roth: Scott Department of Urology, Baylor College of Medicine, and Department of Pediatric Urology, Texas Children's Hospital, Houston, Texas 77030-2399.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Bifid Scrotum Penoscrotal Transposition Scrotal Hypoplasia Scrotal Ectopia Webbed Penis Scrotal Inclusion Cysts Chapter References

Isolated congenital anomalies of the scrotum, including inclusion cysts, the bifid or hypoplastic scrotum, penoscrotal transposition, and webbed penis, are unusual. Most often these anomalies are found in association with other abnormalities: penoscrotal transposition and bifid scrotum with hypospadias, bifid scrotum and scrotal ectopia with exstrophy, and scrotal hypoplasia with cryptorchidism. When these anomalies occur in conjunction with other genital abnormalities, the scrotum can be surgically repaired with excellent results at the time of the primary procedure. It is important, however, that neonatal circumcision be avoided in boys with congenital scrotal anomalies involving the penis because the prepuce often is utilized for the subsequent reconstruction.

DIAGNOSIS
The diagnosis of scrotal congenital anomalies is made by physical examination. Rarely do these anomalies occur as solitary lesions and the patient should be appropriately worked up for the other associated anomalies. The surgeon should examine the scrotum for its relative position to the penis, the rugations and median raphe, and the testes in each scrotal compartment as well as the penis, urethra, and rectum.

INDICATIONS FOR SURGERY
Timing of the surgical repair of urogenital anomalies is of importance from the aspects of feasibility of the surgery, safety of the surgery to the patient, and the psychological impact of the anomaly and surgery. Because congenital scrotal anomalies do not interfere with urinary function, repair can be scheduled at the time that is most appropriate for the infant and most convenient for parents and physicians. Technically, from the surgeon's point of view, there is very little to be gained by delaying surgery beyond the child's fourth to sixth month since by that time the genitalia have developed to sufficient size for easy reconstruction. With optical magnification and the standardization of fine sutures, excellent results can be expected even in the very young child. In addition to the degree of scrotal development, the safety of anesthetics has always been a concern in determining the minimum age at which an operation seems appropriate. However, as a result of the proliferation of specially trained and dedicated pediatric anesthesiologists, the anesthetic complication rate is now minimal in patients as young as 3 months of age. To delay repair beyond that time is no longer necessary in centers dedicated to pediatric surgery. Psychological concerns limit the opposite end of the time spectrum. 7 The child's anxiety concerning hospitalization, gender identity, and subsequent sexual development must be considered. If genital surgery is performed before the child is 18 months of age, he neither remembers the surgery nor associates the experience with an abnormality of his penis or scrotum. Therefore, a window (4 to 18 months) exists for surgery, limited on the younger side by both anesthetic and technical concerns and on the older side by memory and psychological concerns. Parents need to determine a time during that 14-month period that is best for their schedules. Because most of these surgical repairs require only external relocation of skin, they usually can be performed on an outpatient basis or, rarely, with a hospital stay of a single night. A further consideration is that it might be easier for the parents to care postoperatively for a boy who is younger and not yet walking. Certain technical points are relevant to all of these operations and warrant mentioning. For instance, optical magnification has proved to be very useful. Several companies now make loupes in powers of 2.5× to 4.5× that practitioners have found to be valuable when performing delicate surgery on the genitalia. Also helpful are the fine (6-0 and 7-0) absorbable (plain, chromic, polyglycolic acid) sutures that are excellent materials with which to repair a child's scrotum. They do not have to be removed and are absorbed quickly, so that skin tracks are unlikely to form. Prophylactic antibiotics are seldom necessary in uncomplicated cases.

ALTERNATIVE THERAPY
There are no alternatives to surgical correction of the congenital anomaly

SURGICAL TECHNIQUE
Bifid Scrotum Isolated bifid scrotum is rare. When it occurs, the corpus spongiosum appears to be continuous with the median raphe of the scrotum, which is a fibrous band that separates and divides the scrotum into two individual parts. Surgical reapproximation of the two hemiscrotums can be achieved after excision of the fibrous midline band and the underlying urethra is thus preserved. This requires mobilization of each hemiscrotum to the extent that it can be elevated and moved medially to the midline. Closure is accomplished in at least two layers. Deep absorbable sutures allow fixation and reconstruction of the midline. Fine absorbable sutures should be used to close the skin. Generally, interrupted simple sutures are used, but a running subcuticular closure may leave a less visible scar with neither suture tracks nor cross-hatching. Penoscrotal Transposition Various degrees of penoscrotal transposition exist, ranging from the complete form, in which the scrotum is actually anterior and cephalad to the base of the penis, to incomplete forms, in which the penis emerges from the center of the scrotum, and the milder forms, in which only the superior edges of the scrotum lie anterior to the penis. For correction of the anomaly, the two hemiscrotums are mobilized, swept interiorly and medially, and sutured together. It may be necessary to transpose the penis cephalad to the scrotum. This can be achieved by using a skin bridge ( Fig. 103-1) or by dividing the abnormal scrotum in its midline cephalad and caudal to the phallus and swinging both halves below the penis ( Fig. 103-2).2 Some authors suggest leaving a segment of skin intact cephalad to the penis to avoid jeopardizing the vascularity and lymphatic drainage of the penile shaft skin. For less severe cases of penoscrotal transposition, the wedge of ectopic scrotal skin is removed and the resultant defect closed, thereby eliminating the problem. 6 This approach is most appropriate when there is incomplete mild penoscrotal transposition ( Fig. 103-3).

FIG. 103-1. (A) In the moderately severe case of penoscrotal transposition, the phallus is circumscribed and freed of supporting tissue. An incision is made cephalad to the base of the phallus. (B) The penis can then be brought back through the new opening and the skin sutured about it. The original opening is then sewn closed.

FIG. 103-2. (A) In the more severe case of penoscrotal transposition, each hemiscrotum is circumscribed. (B) Once each side is freed enough to allow it to rotate caudally and medially the two portions are sewn together. Some authors suggest leaving a bridge of skin cephalad so as to decrease the possibility of a vascular insult.

FIG. 103-3. In the mild case of penoscrotal transformation, a V-shaped wedge of ectopic scrotal skin is excised (A, B). (C) The defect is then closed, thereby eliminating the ectopic scrotal skin.

Scrotal Hypoplasia Scrotal hypoplasia is almost always restricted to boys with cryptorchidism. A limited course of androgen stimulation (testosterone enthanate 2 mg/kg) will induce scrotal development and enlargement. 1 This allows easier surgical placement of either a testis or prosthesis in the poorly developed scrotum at the time of inguinal surgery for the undescended testis. The hormonal treatment should be undertaken only in conjunction with either an orchiopexy or placement of a testicular prosthesis because the effects of the testosterone are temporary; if the scrotum is not distended, it may revert to its initial appearance. Scrotal Ectopia Although less common than either scrotal transposition or a bifid scrotum and usually associated with cloacal exstrophy, scrotal ectopia occasionally is found in the otherwise normal child.4 The ectopic scrotal tissue can be found on the inner aspect of the thigh or caudal and inferior to the external inguinal ring. Often the ipsilateral testis can be found within the ectopic tissue. Correction is accomplished by relocating the ectopic tissue, by way of a flap or graft, or by utilizing the normally positioned contralateral hemi-scrotum as a reservoir for both gonads and discarding the ectopic tissue. 8 The latter can be accomplished by stretching local tissue either primarily or after pretreatment with parenteral testosterone enthanate (2 mg/kg). The ectopic scrotal tissue in that case can then be excised. Webbed Penis In boys with a webbed penis, scrotal skin is tethered to the ventrum of the penile shaft. This tethering produces a web of skin stretching from the distal penis to the scrotal base. The webbed penis causes no problem during childhood. However, as the scrotal skin is hair bearing, future intercourse could be difficult or uncomfortable. Therefore, a webbed penis should be corrected during infancy. A modified circumcision can often correct the defect. The circumcision incision is brought more distal than normal on the ventrum, thereby preserving all penile shaft skin possible in that location. After the prepucial skin is excised the additional length on the ventrum allows the scrotum to fall away from the glans and distal penis. If necessary skin from the dorsum can be mobilized and swept ventrally to provide additional shaft skin ( Fig. 103-4). A second type of repair can be performed by incising the web transversely and closing it longitudinally, thereby separating the penis from the median raphe of the scrotum. A circumcision should be considered at the same time because it assists in the approximation of the skin ( Fig. 103-5).

FIG. 103-4. In the mildly webbed penis, a modified circumcision may be all that is required. By making the ventral incision at the phimotic band, the ventral skin can be

repositioned to appropriately cover the penile shaft.

FIG. 103-5. (A) In the moderately webbed penis, the defect can be repaired by transversely incising the web (A) and closing it longitudinally (B). As in the mild cases, a modified circumcision incision with preservation of all the ventral skin can be helpful in recovering the penile shaft.

In more severe cases of webbed penis, a U-shaped incision is made about the phallus. 5 This releases the penis from the dependent scrotum. Flaps are developed to allow ventral closure of the penis with fine absorbable sutures. The scrotum is closed in a side-to-side manner ( Fig. 103-6).

FIG. 103-6. (A) In a severely webbed penis, an incision is made between the penis and scrotum. (B) Skin flaps are elevated in all directions so that the surgical defect can be closed. (C) A two-layer closure is used to stabilize and approximate the skin and underlying tissues.

Scrotal Inclusion Cysts Midline scrotal inclusion cysts are generally dermatoid in origin and can be managed by local excision. 3 However, care must be taken not to confuse these with the sinus associated with an imperforate anus. Both are located in the median raphe and can be multiple. Because cysts can lead to calculi formation or infection, local excision should be considered. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. Gearhart JP, Jeffs RD. Use of parenteral testosterone therapy in genital reconstructive surgery. J Urol 1987;138:1077–1078. Glenn JF, Anderson EE. Surgical correction of incomplete penoscrotal transposition. J Urol 1973;110:603–605. Hamada Y, Sakiyama H, Nakashima K, Oka M. Median raphe cysts and canal of the penis. Eur Urol 1982;8:312–313. Lamm DL, Kaplan GW. Accessory and ectopic scrota. Urology 1977;9:149–153. Perlmutter AD, Chamberlain JW. Webbed penis without chordee. J Urol 1972;107:320–321. Redman JF. The surgical correction of incomplete scrotal transposition associated with hypospadias. J Urol 1983;129:565–567. Schultz JP, Klykylo WM, Wacksman J. Timing of elective hypospadias repair in children. Pediatrics 1983;71:342–351. Spears T, Franco I, Reda EF, Hernandez-Graulau J, Levitt SB. Accessory and ectopic scrotum with VATER association. Urology 1992;40:343–345.

Chapter 104 Cryptorchidism and Pediatric Hydrocele/Hernia Glenn’s Urologic Surgery

Chapter 104 Cryptorchidism and Pediatric Hydrocele/Hernia
Stanley J. Kogan and Bhagwant Gill

S. J. Kogan: 311 North Street, Suite 310 White Plains, New York 10605. B. S. Gill: Eastchester Professional Center, Bronx, New York 10461-2330.

Diagnosis Cryptorchidism Hernia Hydrocele Indications for Surgery Hernia Hydrocele Alternative Therapy Hormonal Therapy of Cryptorchidism Surgical Technique Herniorraphy and Hydrocelectomy Orchiopexy Absent Testes Outcomes Complications of Orchiopexy Complications of Herniorraphy and Hydrocelectomy Results Chapter References

Childhood hernia repair and cryptorchidism surgery represent by far the most common surgical conditions encountered in a pediatric urology practice. Major changes have occurred in the contemporary management of cryptorchidism recently, as have significant refinements in the surgical treatment. First and foremost has been the growing body of data indicating that early treatment seems to offer a superior approach to the traditional treatment advocated at older ages. Second, advances in technology have brought about significant changes in diagnosis and surgical management. This chapter reviews this information and indicates the optimum present management of these conditions. Descent of the testis and epididymis is necessary for normal testis development and production of a fertile ejaculate, with experimental and clinical observations of testes secondarily dislocated into an extrascrotal location supporting this claim. The 2° to 3° temperature difference between the scrotum and abdomen is critical in this regard. Between the 12th and 17th week of gestation the testis normally completes a course of transabdominal migration, coming to occupy a location just within the internal inguinal ring. During the seventh month of gestation, after a period of lack of movement by the testis, a sudden swelling of the gubernaculum and extension of the processus vaginalis into the scrotum is followed by descent of the epididymis and testis. This phase of descent, termed transinguinal migration, is under hormonal control, and therefore disorders of gonadotropin production or androgen synthesis or action may affect this phase and are often associated with cryptorchidism. Gubernacular abnormalities, including insertions into locations other than the scrotum, i.e., the perineum and thigh have been observed. Fibrous obliteration of the inguinal canal occurs sometimes and both these latter conditions are incriminated as causes of testicular maldescent. Abnormal structure and function of the spinal nucleus of the genitofemoral nerve and its neurotransmitters, affecting gubernacular function, also have been suggested. The frequency of cryptorchidism relates directly to fetal maturity and size. Cryptorchidism occurs commonly in preterm infants, approximating 30%, with an ncreasing frequency associated with diminishing birth weight. The frequency at full term is 3% to 4%, confirming that most cryptorchid testes descend during the last trimester. Further descent occurs during the first several months after birth, so that by 1 year of age the frequency of cryptorchidism is 0.7% to 1.0%. Many studies show that spontaneous descent after the first year is uncommon and that the frequency later in life is the same as that noted at 1 year of age. King has classified the cryptorchid testes as to the underlying cause of maldescent:
6

End-organ failure: a smaller-than-normal testis with a paucity of germ cells and often abnormal ductal structures, noted even at birth. Ectopic testes: descent occurs through the inguinal canal but the testis is subsequently misdirected to an ultimate extrascrotal location. Inadequate gonadotropin stimulus: inferred when a high undescended testis descends in response to hormonal therapy or when cryptorchidism is noted in conjunction with an obvious endocrine deficiency syndrome (i.e., anencephaly, Kallman's syndrome, etc.). Mechanical obstruction, i.e., fibrous obliteration of the inguinal canal. Inadequate intraabdominal pressure to “push” the testis down, i.e., as seen in prune belly syndrome, omphalocele, and so forth. Whereas this classification is arbitrary and some categories overlap others, it serves a useful purpose indicating that the causes of cryptorchidism are varied and that the ultimate potential of these testes is not uniform. We personally recommend a functional classification based on whether the testis is palpable or impalpable at examination, as this classification provides a practical approach to therapy (Table 104-1).

TABLE 104-1. Classification of cryptorchidism

A large number of studies have characterized the location of the testis in cryptorchidism. Based on these as well as our personal experience, approximately 3% to 4% are absent, 10% are intraabdominal, 50% to 70% are high scrotal or ectopic in location, and the remainder are intracanalicular. In addition to true cryptorchidism, ectopic testes represent another form of maldescent. True undescended testes fail to reach the scrotum though they follow a normal line of descent. Diminished spermatic vessel length and fixation of the testis are the usual cause. Ectopic testes follow a usual course of descent until they emerge from the external inguinal ring, where they are then misdirected to an ectopic location ( Figure 104-1), the most common being the superficial inguinal pouch located between the overlying subcutaneous tissue and the underlying external oblique aponeurosis. Fibrous obstruction of the scrotal inlet and an abnormal gubernaculum are the usual causes of ectopia.

FIG. 104-1. Locations of true undescended testes and ectopic undescended testes. (From Kogan SJ, Saenger P. Cryptorchidism. In: Bardin CW, ed. Current therapy in endocrinology and metabolism. Vol. 3. Toronto: BC Decker, 1988;236–244.)

Impalpable testes constitute approximately 20% to 25% of cryptorchid testes. Reports indicate that between 20% to 55% of impalpable testes are located intraabdominally, and between 20% and 50% are absent. Impalpable testes pose unique management problems. Spontaneous descent occurs infrequently in this group, and a greater frequency of hernias and ductal system abnormalities occur. Special diagnostic tests are important in diagnosing whether the impalpable testis is absent and a well-thought-out specific surgical approach is necessary. Surgical treatment is more difficult and results in this group are poorer; therefore, this group represents a difficult management problem. Hernias and hydroceles in childhood result almost exclusively from failure of fusion and obliteration of the processus vaginalis ( Fig. 104-2), though testicular tumors and inflammatory conditions can also cause hydroceles to occur. Incarceration may occur when any intraabdominal structures protrude into the sac and get trapped. Strangulation occurs more commonly in infancy and requires immediate exploration to prevent necrosis of the sac contents as well as infarction of the testis, which can occur from increased pressure on the delicate spermatic vessels.

FIG. 104-2. Diagrammatic representation of persisting patent processus vaginalis, hydrocele, and hernia. (From Lewis JE Jr., Atlas of infant surgery. St Louis: CV Mosby, 1967;89.)

Hydroceles represent persistence of the processus vaginalis along its course with partial or complete fusion proximally. Virtually all hydroceles are communicating, i.e., a narrow canal exists proximally between the peritoneal cavity and hernia sac that allows intermittent filling and emptying of the hydrocele sac distally.

DIAGNOSIS
Cryptorchidism Patient relaxation is essential during the physical exam to document testis presence and position. Warm, gentle hands and a quiet examining room are helpful. Gentle milking action from the superior iliac crest toward the scrotum often delivers an initially impalpable testis to a palpable or even scrotal position. Obesity makes the examination more difficult. Placing the patient in a sitting, cross-legged position maximally relaxes the cremasteric muscles and lets the testis assume its most distal position without tension. Palpation of a nubbin of tissue within the scrotum does not necessarily indicate an atrophic testis, since sometimes a long-looped vas deferens or elongated epididymis can simulate a testis that lies above in an impalpable position. Scrotal size is not a reliable indicator of testis absence, as cryptorchid testes often are associated with scrotal hypoplasia. Undescended testes must be distinguished from retractile testes that may lie in an extrascrotal position, but with careful positioning and coaxing they can be made to lie within the scrotum without continued traction. The latter maneuver forms the basis of diagnosing this condition. Retractile testes are encountered at any level along the normal line of descent, though they are usually palpated in the groin. Testis growth, maturation, and ultimate fertility are reported as normal in this condition and its true importance is in distinguishing it from a cryptorchid testis. At times this distinction is difficult, and a diagnostic/therapeutic course of human chorionic gonadotropin (HCG) is helpful. Numerous and diverse methods have been utilized to image the impalpable testis. Pneumo- and direct contrast peritoneography and gonadal arteriography and venography are of historical interest only; they are no longer in use. Ultrasound has the obvious advantage of being radiation-free; however, its accuracy is limited in young children. False-positive results are uncommon, but false-negative results (failing to demonstrate a testis) occur frequently. Therefore the occurrence of a negative study does not exclude the presence of a testis. CT scanning requires adequate sedation in young children and represents a moderate radiation burden. The general lack of body fat limits its diagnostic accuracy to far less than the 100% needed in dealing with this problem. We use computed tomography (CT) infrequently in searching for impalpable testes, i.e., sometimes in older boys, adults, and in reoperative cases where an initial incomplete exploration has failed to demonstrate a testis. Magnetic resonance imaging (MRI) has the obvious advantage of being radiation-free, but it requires sedation in very young children and has a less-than-perfect accuracy at that age. Laparoscopy done to determine location and presence of an impalpable testis at the time of surgical exploration has enjoyed a surging popularity. Proponents claim that the incision is placed more appropriately and the surgery is simplified when the testis is located exactly. If an absent testis is confirmed, an extensive intraperitoneal search is avoided and all that is needed is a simple inguinal incision for placement of a testicular prosthesis, if desired. Laparoscopic findings of spermatic vessels entering the internal inguinal ring direct the incision to that area, and a testis located high in the abdomen would indicate placement of the incision at a suitable higher location. Despite the attractiveness of this approach, some do not utilize laparoscopy routinely in these circumstances. Our experience confirms previous reports indicating that 95% of impalpable testes may be localized through a standard transverse inguinal incision with a short intraperitoneal extension if necessary. Our surgical approach for orchiopexy utilizes the same inguinal incision used for exploration and thereby does not benefit from exact localization, since extension of the incision superiorly and laterally can allow exposure almost up to the lower pole of the kidney in young children. More important, because the testis is present 60% to 75% of the time when impalpable a large number of “unnecessary” laparoscopies must be done to identify a much smaller number of situations where laparoscopy is helpful. Routine laparoscopy, therefore, does not seem to be rewarding, though in older boys and adults where an intraperitoneal exploration has a higher morbidity, where a previously performed surgical exploration has not satisfactorily excluded an occult retained testis, or where the surgeon plans for a laparoscopic orchiopexy, routine laparoscopy as an initial step is logical. When both testes are impalpable, careful distinction must be made between bilateral impalpable cryptorchidism and bilateral anorchia. In this condition, measurement of plasma luteinizing hormone (LH) and follicle-stimulating hormone (FSH) levels and the testosterone response to HCG stimulation has been used to diagnose bilateral testicular absence. Raised gonadotropin levels and a lack of a testosterone rise from HCG (both criteria must be present) indicate bilateral absent testes, and

in this circumstance a formal surgical exploration is not necessary for confirmation of this diagnosis. When one or both components are lacking, however, a thorough surgical exploration must be made because a hidden testis or testes may be present. Further detailed description of this diagnostic test and its interpretations and limitations are discussed elsewhere in more detail. 5 Hernia The usual presentation of a childhood inguinal hernia is an intermittently palpable asymptomatic groin swelling. In some instances the swelling is never observed by the examining physician and diagnosis rests on obtaining an excellent history from parents to confirm the clinical suspicion. Hydrocele Hydrocele presents as a scrotal mass in which the testis is not palpable and transilluminates. Since most hydroceles in children are communicating, the diagnosis is suggested by the variability of the size of the mass. Other diagnostic modalities that may prove useful include transscrotal ultrasound. A loculated hydrocele may occur anywhere along the spermatic cord and can sometimes be quite firm, simulating an incarcerated inguinal hernia or spermatic cord tumor. Transillumination of the mass and ultrasonography usually clarify this important distinction well. Sometimes, especially in older children, an isolated hydrocele may occur in the absence of a patent processus vaginalis. The etiology is not often clear in these circumstances, and an underlying epididymitis or testis tumor may be the cause.

INDICATIONS FOR SURGERY
Current treatment indications include possible improvement and preservation of fertility; possible prevention of malignancy; correction of often-associated hernias; prevention of testis torsion, which occurs with an increased frequency in cryptorchidism; and improvement in body image with two testes in a normal-appearing scrotum. Of these, fertility enhancement is by far the most important. The detrimental effects of late treatment are well known with only about one-third of bilateral cases and 50% to 70% of unilateral cases having normal fertility. Also, though most cryptorchid testes have a normal complement of germ cells at birth, a progressive loss occurs with aging, so that 25% to 40% are already aspermatogenic by age 2 to 3 years. Leydig cell abnormalities, both structural and functional, resulting from hormonal understimulation occur as well. It has been suggested that this deterioration occurs early on (by 1 year of age) rather than resulting from an acquired process later; hence, treatment by the first birthday is indicated. The cryptorchid patient is at greater risk for tumor formation later in life, but few data indicate that treatment, even early, diminishes this risk. Statistical calculations have indicated an increased risk of 20 to 40 times normal, with an absolute risk of 50 per 100,000 population per year. Though intraabdominal testes constitute only about 10% of cryptorchid testes, they account for almost half of those developing tumors and may have a fourfold increased risk over other cryptorchid testes. Though there are few data presently indicating beneficial effects of early orchiopexy regarding malignancy prevention, the information regarding malignancy occurrence when treatment is provided later as well as the fertility data suggest that little is lost in providing early treatment from both standpoints. Hernia Treatment of the often-associated hernia accompanying the cryptorchid testis is an integral part of the orchiopexy procedure. Hernias occur commonly especially in association with intraabdominal testes. Infants presenting with hernias who also have a cryptorchid testis present should have both repaired at the time of presentation regardless of age in order to diminish the risk of incarceration. Similarly, torsion of the poorly fixed undescended testis that may have abnormalities of suspension from the epididymis occurs also with greater frequency. Clinicians dealing with cryptorchid boys find that an impaired body image can result, as can fears of sterility and personality changes, offering additional reasons for treatment. Children with incarcerated hernias should be sedated with “DPT” (meperidine 2 mg/kg body weight, promethazine I mg/kg, and chlorpromazine 1 mg/kg) and placed in Trendelenberg's position; most will reduce spontaneously in this way. However, elective hernia repair must be done shortly thereafter because many will reincarcerate within a short time. In children, controversy exists regarding the necessity to explore the contralateral inguinal canal when a unilateral hernia exists. A contralateral patent processus vaginalis is often found, especially in younger children; frank hernias similarly occur with a greater frequency in the younger aged. Contralateral hernias definitely present subsequent to a unilateral hernia repair when only one side is repaired, and the frequency differs as to the side of the initial presentation. When the left is repaired initially, a 41% contralateral occurrence was noted; if the right occurred initially, only 14% subsequently occurred on the left. Therefore, some advocate so-called routine exploration of the contralateral groin in this context. Others argue against this approach, citing the added anesthesia time necessary as well as the small but definite incidence of testis atrophy, vas injury, and secondary extrascrotal testis fixation that sometimes occurs after hernia surgery. We personally utilize an intraoperative diagnostic pneumoperitoneum to aid in this decision, creating a carbon dioxide pneumoperitoneum through the open hernia sac to a pressure of 12 to 15 cm H2O; others have utilized intraoperative peritoneoscopy through the open hernia sac to inspect the contralateral inguinal ring. Hydrocele Many hydroceles can be quite large. However, size should not in most circumstances influence a decision toward early surgical repair unless a clear-cut hernia is present as well. Most undergo spontaneous closure by 15 months of age, with disappearance of the swelling. Persistence beyond this age implies that the associated processus vaginalis remains patent and little chance of spontaneous closure exists. Surgical exploration and repair through an inguinal incision is indicated. If a hernia is diagnosed in association, at any age, repair of both is indicated at the time of presentation.

ALTERNATIVE THERAPY
Despite these indications, some have suggested that in some instances alternatives to orchiopexy should be considered. 4 Though cryptorchid testes almost always can be situated surgically in the scrotum, those that are more poorly developed and have ductal abnormalities of significance (i.e., many intraabdominal testes) might be better treated by orchiectomy, especially since fertility is reduced in these testes. Certainly in the untreated pubertal cryptorchid boy this alternative to orchiopexy should be considered as most of these testes are sterile. In other situations, such as severe mental retardation, ejaculatory failure, and in certain genetic disorders, omission of treatment could be considered. Hormonal Therapy of Cryptorchidism Though the majority of undescended testes will descend by 6 months of age, spontaneous descent will even occur up to the first birthday, with more than two-thirds of testes undescended at birth descending spontaneously by then. Treatment is best initiated at that time with HCG, since gonadotropin-releasing hormone (GNRH) is not yet available in the United States for treatment of cryptorchidism. In general bilateral cryptorchids respond more favorably to HCG, with between 30% and 50% of testes descending; the success rate is about half these figures for unilateral cryptorchids. Initial testis location affects results, with usually the more distal testes responding favorably to hormonal treatment. Retractile testes in question can be quite successfully descended with HCG. We recommend a “long” course of treatment rather than the abbreviated short course (three injections over a 10-day period). At ages up to 12 months, 500 IU should be administered twice weekly for up to 5 weeks; for children age 1 to 6 years, 1500 units with the same frequency and duration. Usually results can be seen within 4 weeks of treatment onset. Repeat courses of treatment offer little added advantage. Ten percent to 20% of patients will have some degree of testis retraction, and in these where complete descent has occurred a second subsequent course of treatment will be of advantage. GNRH, widely used outside the United States, has offered an additional form of hormonal treatment, avoiding the virilizing side effects seen sometimes with HCG. A wide range of success has been reported, with inclusion of retractile testes inadvertently within the treatment group cited as the major cause for this discrepancy. As with HCG, the best results are seen with the distal-most located testes, and GNRH seems ineffective in descending intraabdominal testes.

SURGICAL TECHNIQUE
The incision utilized for hernia and hydrocele surgery and treatment of uncomplicated cryptorchidism is identical. Intraabdominal testes require special approaches, and other incisions are utilized. The standard inguinal incision is used for entry into the inguinal canal. This is placed transversely in the lower abdominal skin crease, avoiding oblique incisions that do not parallel Langer's lines because they often leave broad scars ( Fig. 104-3). Scarpa's fascia is divided and the external oblique

aponeurosis opened parallel to the inguinal ligament. Special care is provided to identify and avoid transection of the ilioinguinal nerve, which hugs the underside of the external oblique fascia.

FIG. 104-3. Placement of standard inguinal incision.

It is often surprising how high the spermatic cord can be dissected at the time of orchiopexy using this incision, and we have sometimes reached the lower pole of the kidney by opening the internal inguinal ring and providing appropriate retraction. The limiting factor in this instance is not the skin incision but rather the internal ring, which should be opened generously. Various lower abdominal incisions have been used to gain access to intraabdominal testes when both are high or when separate inguinal incisions are to be avoided. A midline approach (horizontal or vertical skin incision) can be used to approach the testis either trans- or preperitoneally, the latter accomplished by dissecting laterally along the rectus sheath until the retroperitoneum is entered and the spermatic cord structures are encountered. When an extensive exposure of bilateral intraabdominal testes is necessary, both rectus muscles can be transected just above the pubis, resulting in a wide exposure. In each of these instances a new external inguinal ring is fashioned just lateral to the pubic tubercle, so that the spermatic cord makes the most direct route into the scrotum ( Fig. 104-4).

FIG. 104-4. Lower abdominal incision for transabdominal orchiopexy. (A) Intraabdominal view showing testis intraperitoneally at pelvic brim. (B) Testicular mobilization from posterior peritoneum and restraining bands. (C) New external ring created at level of lateral rectus muscle border and superior pubic ramus. (D) Testis brought through the opening and placed in the scrotum. (From Kogan SJ. Cryptorchidism. In: Kelalis PP, King LR, Belman AB, eds. Clinical pediatric urology. Philadelphia: WB Saunders, 1985; 864–887.)

Scrotal incisions are part of the orchiopexy procedure in children, as testis retraction may occur if the testis is simply pulled into the scrotum with a traction suture passed through the scrotal skin. The technique advocated by most child surgeons is the subcutaneous pouch method, fixing the testis between the skin and the underlying dartos muscle. Following transverse incision of the scrotal skin but not the underlying dartos, a plane is developed between the two with a hemostat that is adequate in size to accommodate the testis and paratesticular structures. The underlying dartos is incised adequately to accommodate the spermatic cord, and the testis is pulled through with a previously placed capsular suture, which is then passed through the dependent scrotal skin, pulling the testis into the pouch. This may be tied to itself, or passed through a gauze pledget for retention when a high testis is brought down, allowing for longer fixation. Thereafter the scrotal incision is closed with 5-0 chromic catgut. Herniorraphy and Hydrocelectomy Following exposure of the interior of the inguinal canal through the standard inguinal incision, the spermatic cord and hernia sac are circumscribed with a rubber drain. The edge of the hernia sac or patent processus vaginalis, present on the anteromedial aspect of the cord, is gently grasped with a hemostat, and the sac is gently teased off the underlying cord with a forceps pushing the cord structures away. Elevating the sac with an underlying finger is helpful. When the cord structures are completely free the sac is transacted, further dissected up to the internal inguinal ring, then doubly ligated with an absorbable suture. 4-0 is used in infants, 3-0 in most other prepubertal boys, and 2-0 thereafter. The distal sac or hydrocele, if present, is then delivered into the wound and carefully laid open. Though some surgeons feel that laying open the sac alone is sufficient, we prefer to sew the sac behind the cord or testis with 5-0 chromic sutures because reformation of the hydrocele sometimes occurs if the edges are not everted. Usually the hydrocele sac need not be trimmed. Extreme care is needed in the dissection of large or thick-walled hydrocele sacs as they may hide the vas deferens or insinuate themselves between the vessels and vas, displacing it and resulting in inadvertent vas transection. Scrotal incisions are usually never utilized for hydrocele repair in children because there is invariably an accompanying patent processus vaginalis that must be ligated within the groin. In adolescence, rarely a hydrocele may occur in the absence of a patent processus vaginalis. Exploration through a scrotal incision may be done although a thorough search for a patent processus vaginalis is necessary as well. Orchiopexy Despite the refinements occurring in orchiopexy surgery over the years, the basic principles originally defined by Bevan in 1899 still apply: adequate mobilization of the testis and spermatic vessels followed by repair of the associated hernia, followed by adequate fixation of the testis within the scrotum. Surgical ingenuity has resulted in a number of procedures to achieve satisfactory scrotal testis placement, so that nowadays it is possible to place virtually every testis within the scrotum. The procedure of choice is determined mostly by the testis location; therefore, procedures for treating palpable and impalpable testes will be considered separately. Palpable undescended testes are best approached by the standard inguinal incision described previously. Following opening the inguinal canal, the testis is grasped with a wet sponge, elevated, and the distal attachments of the gubernaculum to the scrotum are divided, taking care to avoid a long-looped vas deferens that sometimes extends distally. The testis and cord are then dissected up to the internal inguinal ring, dividing the binding cremasteric tissue ( Fig. 104-5A). No attempt should be made to dissect within the cord itself as this does not result in any significant cord lengthening and may actually damage the spermatic vessels or vas deferens. The hernia sac is then dissected free of the underlying cord structures. The sac is grasped carefully on its anteromedial aspect with a hemostat, avoiding underlying cord structures, is elevated, and then is carefully teased off the cord with a blunt forceps, serially rotating it with an underlying finger to circumscribe the sac completely. If the sac is inadvertently entered, a hemostat grasps each edge, and it is freed by looking through its opened wall from within while dissecting the cord structures off from without. The sac is then transected and ligated with a 3-0 absorbable suture. Often no additional dissection is necessary as adequate cord length has been achieved; however, if additional length is necessary one should not hesitate to open the internal inguinal ring with the cautery on its anterior aspect. Further division of the tethering lateral spermatic fascia ( Fig. 105-3C) allows for significant additional lengthening. Anatomic dissections have demonstrated that up to

three-quarters of the entire potential lengthening possible are gained by this maneuver, with only one-quarter from the dissection within the inguinal canal. Thereafter a pathway is made into the scrotum with the dissecting finger ensuring that there is no obstructing tissue, and scrotal testis fixation and groin closure are done as described previously.

FIG. 104-5. (A) The external oblique aponeurosis has been opened and the distal gubernacular attachments have been divided for orchiopexy. (B) Hernia sac has been dissected off the cord up to the internal inguinal ring and is ready for ligation. This may be done without actually opening the sac. (C) Hernia has been ligated and has dropped back into a retroperitoneal location. Inset shows the lateral spermatic fascia being divided.

Impalpable undescended testes pose special surgical problems: the incisions utilized may be different, the techniques of orchiopexy are often more extensive, and since in at least one-quarter of cases the testis will be found to be absent, a properly planned exploration is essential. Unilateral impalpable testes can be initially approached through the standard inguinal incision. If the testis is not found within the inguinal canal, often a small protruding tongue of peritoneum is present, the harbinger of an intraabdominal testis above. Opening the internal inguinal ring often will allow the testis to prolapse out into this sac. If the testis is still not present, a limited local exploration sometimes identifies the vas deferens, which can be traced to the testis, but if this exploration is unrewarding one should not delay in opening the peritoneum to search within for the testis. Though thought to be a retroperitoneal organ, the testis often is more easily visualized transperitoneally, usually in a position adjacent to the bladder, i.e., similar in position to the ovary in the female. When bilateral impalpable testes are present approach through one of the previously described lower abdominal incisions can be used. When a testis is identified, a gentle stretch with a traction suture through its lower pole allows for assessment of spermatic vessel length. If adequate length appears to be present, an extended high dissection of the cord, freeing it from its peritoneal and lateral spermatic fascial attachments, can achieve adequate scrotal placement. If this assessment indicates that the vessels are too short, specialized orchiopexy procedures are necessary. Orchiopexy by spermatic vessel transaction has undergone numerous refinements since its original description, with current success rates as high as 80% to 90% being cited. 8 In this procedure, the tethering spermatic vessels are intentionally divided, relying on the collateral blood supply along the vas deferens. In bilateral cases, following transabdominal exposure of the testes the internal spermatic vessels are freed through the posterior peritoneum to just below the kidneys. A Fowler-Stephens test is done by applying atraumatic bulldog clamps to the vessels high away from the testis, which is then incised and observed for bleeding over a several minute period. If adequate, the tethering short spermatic vessels are then ligated, and a broad-based medial pedicle of peritoneum and underlying vascular-bearing adventitia surrounding the vas deferens is developed and mobilized, placing downward traction on the testis at the same time. The pedicle is then brought lateral to the pubic tubercle with the testis making a direct course into the scrotum. In unilateral cases, after the peritoneum is opened through an extended standard inguinal incision, the same maneuvers are done. The hernia sac is not dissected off the cord because this may injure the collateral blood supply immediately adjacent to the testis. Careful attention to detail is critical in achieving success in performing this procedure. Failure to leave a broad medially based peritoneal strip attached to the vas deferens, doing this procedure as a salvage operation after a conventional orchiopexy dissection (i.e., after total mobilization of the testis and cord) is done, ligating the spermatic vessels too close to the testis, directly injuring the vasal artery during the dissection, and proceeding with the operation after a failed Fowler-Stephens test all may lead to failure. 2,8 A staged testicular vessel transaction orchiopexy has been described where the spermatic vessels are simply ligated at an initial stage without mobilization, and a several-month interval is allowed for the vasal collateral blood supply to hypertrophy. At a second stage the vessels are divided and the operation is conducted as indicated for the single-stage procedure. This is a safe and reliable procedure, and it may have special applicability in very small boys where the vessels are similarly very small. This procedure may also be performed laparoscopically, first clipping the vessels laparoscopically at the first stage, and by performing an open or laparoscopic second-stage repair several months later. Added experience with these procedures has led to a primary laparoscopic surgical approach to the intraabdominal testis. I personally have not utilized this approach primarily because I do not feel that insertion of four adult size laparoscopic ports (one 10-mm port and three 5-mm ports) in a small boy is any less traumatic than the previously outlined incisions. However, with continued miniaturization of the laparoscopic instruments this approach may be able to be performed as a truly minimally invasive procedure and may be much more advantageous. Microvascular reanastomosis of the transacted length-limiting internal spermatic vessels to the inferior epigastric vessels may offer additional safety beyond simple ligation alone. The epigastric vessels are exposed within the inguinal canal and on the posterior aspect of the rectus muscle, and then, following spatulation, the smaller spermatic artery is reanastomosed under the operating microscope using an approximating microsurgical clamp and 10-0 or 11-0 nylon interrupted sutures. Though desirable in theory, routine use of this procedure is not common because good results are obtainable by less extensive means, and because microvascular orchiopexy at the optimum age of 1 year is difficult because of vessel size. When neither of these procedures is possible in treating the intraabdominal testis, a staged orchiopexy may be utilized. Similar success rates (82% to 90%) to the previously mentioned procedures have been reported. Following maximal high mobilization of the vessels and testis through one of the previously described incisions, the testis is sutured to the most distal position obtainable without tension, i.e., the pubic tubercle or inguinal ligament, with a 3-0 or 4-0 nylon suture. A reexploration is done 6 to 12 months later. Interval lengthening of the spermatic cord usually allows for dependent scrotal placement at this time. This second procedure may be difficult because of interval scarring and testis fixation, and care must be taken to avoid vas or vessel injury. For these reasons orchiopexy in a single stage is preferred, when feasible. It can be seen that there are a plethora of available methods available to treat these difficult cases; the urologist should be knowledgeable of these and should be facile with at least some. Absent Testes In the course of exploring for an impalpable testis, in approximately 25% to 40% the testis will prove to be absent. During the exploration if structures resembling a blind-ending miniature spermatic cord are encountered in the inguinal canal they are traced within the internal ring to the point of divergence of the vas and vessels, positively identifying the latter as spermatic vessels as they course up the retroperitoneum. In this instance, no further exploration for a testis is indicated since blind-ending spermatic vessels are the absolute criteria for ending the exploration; in no instance has a testis been found elsewhere when these vessels are positively identified ending blindly below. If a vas deferens alone is encountered the exploration must be carried further because a wide separation of the vas and an intraabdominal testis may be present. Similarly, if no structures are found, the exploration must proceed further since an intraabdominal testis may be present in this instance as well. This is done by opening the internal ring and searching the local retroperitoneal area, and opening the peritoneum locally and searching from within if necessary. A key point of technique that bears emphasis in these explorations is the need to positively identify the internal spermatic vessels by opening the internal inguinal ring and tracing the vessels superiorly, since the gross appearance of structures encountered in these explorations is often confusing. A looped epididymis or vas deferens may mimic the appearance of a blind-ending spermatic cord or atrophic testis and may be accidentally excised. An absent testis may be misdiagnosed when the testis lies in an intraabdominal position above. Intraoperative examples of these perils are demonstrated in Fig. 104-6.

FIG. 104-6. Examples of misleading groin structures in the inguinal canal simulating a blind-ending spermatic cord and absent testis. (A) Long-loop vas with nubbin at end (arrows). Testis lies within internal ring. (B) Similar example with long-looped vas deferens and nubbin at end, extending down inguinal canal. Testis above was intraabdominal. (C) Older child with structure mimicking atrophic testis with collapsed tunica vaginalis at end (arrows), which had exited through the external inguinal ring. (D) Blind-ending gubernacular-like structure passing down through inginal canal (arrows). Testis lies above within the abdomen. (From Kogan SJ. Cryptorchidism. In: Kelalis PP, King LR, Belman AB, eds. Clinical pediatric urology. Philadelphia: WB Saunders, 1985; 864–887.)

In cases where monorchia or bilateral anorchia is diagnosed, the structures are excised and testicular prostheses placed through the inguinal incision. Prostheses should not be placed when a scrotal incision has been made because extrusion can occur postoperatively. Recent controversy regarding potential dangers of silicone prostheses has been largely resolved. Saline-filled silicone shell prostheses are available for implantation now through a follow-up study protocol. In children, repair of the dilated internal inguinal ring is often required in any of these circumstances. An approximation of transversalis fascia to itself (above or below the spermatic cord) at the internal ring may be done providing it is of adequate consistency; alternatively, the conjoined tendon may be sutured to the shelving edge of the inguinal ligament, always resulting in a strong repair. Care must be taken to allow adequate space for spermatic cord mobility as it exits the internal ring. A hemostat may be gently insinuated next to the cord to demonstrate the latter. The external oblique aponeurosis is then closed with a running suture, and then the subcutaneous tissue. In children, for this closure we usually use 4-0 chromic or absorbable synthetic sutures in infants, 3-0 in young children, and 2-0 in older children. The skin similarly is closed with a subcuticular absorbable suture (5-0 or 4-0 synthetic), as children protest vociferously the removal of nonabsorbable sutures.

OUTCOMES
Complications of Orchiopexy The complication rate for impalpable testes in older literature ranges from 20% to 50%, though recent experience is more favorable. Retraction is the most common complication, occurring in up to 10% in one series. Incomplete initial dissection is the main reason for inadequate testis location following orchiopexy, and a retroperitoneal dissection was necessary in 58% of patients to correct this problem. Atrophy occurs next most commonly, and vas deferens injury in 1% to 2%. Attention to surgical detail, gentle handling of tissues, and respect for the small vessels and vas deferens encountered in young children will minimize these complications. Complications of Herniorraphy and Hydrocelectomy Complications are rare following these procedures. Failure to satisfactorily ligate the hernia sac or patent processus vaginalis can result in persistence of these conditions. Injury to the underlying vessels and/or vas deferens should not ever occur, with gentle tissue handling. Occasionally, though, a small testis will be encountered in follow up after these repairs, indicating some compromise of the testicular vasculature. A more common but preventable complication is the occurrence of displacement of the testis to an extrascrotal location after surgery, necessitating a subsequent reoperative orchiopexy. Every hydrocele and hernia repair should be concluded by confirmation of the testis location in its normal dependent scrotal position at the conclusion of the procedure. RESULTS Various series have analyzed the surgical results following orchiopexy regarding testis size, consistency, location, and vas deferens injury, indicating an overall success rate exceeding 90%. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Docimo SG. Results of surgical therapy for cryptorchidism: literature review and analysis. J Urol 1995;154:1148–1150. Gibbons MD, Cromie WJ, Duckett JW Jr. Management of the abdominal undescended testicle. J Urol 1979;122:76–79. Hadziselimovic F. Cryptorchidism. In: Gillenwater JG, Grayhack JT, Howards SS, Duckett JD, eds. Adult and pediatric urology. Vol. 2. Chicago: Yearbook, 1987;1974–1985. Hinman F Jr. Unilateral abdominal cryptorchidism. J Urol 1979;122:71–75. Jarow JP, Berkovitz GD, Migeon CJ, et al. Elevation of serum gonadotropins establishes the diagnosis of anorchia in bilaterally cryptorchid boys. J Urol 1986;136:277–279. King LR. Overview: surgical treatment of cryptorchidism. In: Whitehead ED, ed. Current operative urology–1989. Philadelphia: JB Lippincott, 1989;368–369. Kogan SJ. Cryptorchidism. In: Kelalis PP, King LR, Belman AB, eds. Clinical pediatric urology. Philadelphia: WB Saunders, 1985; 864–887. Kogan SJ, Saenger P. Cryptorchidism. In: Bardin CW, ed. Current therapy in endocrinology and metabolism. Vol. 3. Toronto: BC Decker, 1988;236–244. Kogan SJ, Houman BZ, Reda EF, et al. Orchiopexy by testicular vessel transaction: a critical analysis of 35 vessel transactions. J Urol 1989;141:1416–1419. Maizels M, Gomez F, Firlit CF. Surgical correction of the failed orchiopexy. J Urol 1983;130:955–957.

Chapter 105 Imperforate Anus and Cloacal Malformations Glenn’s Urologic Surgery

Chapter 105 Imperforate Anus and Cloacal Malformations
W. Hardy Hendren III

W. H. Hendren III: Department of Surgery, Children's Hospital, Boston, Massachusetts 02115.

Diagnosis Indications for Surgery Surgical Technique Imperforate Anus Cloacal Malformations Cloacal Exstrophy Outcomes Complications Results Chapter References

Imperforate anus occurs with an incidence of about 1 in 5000 births. It is slightly more common in boys. The most serious cases, however, are those in girls with a cloacal malformation in which there is confluence of the urinary tract, the genital tract, and the bowel into the urogenital sinus. Imperforate anus, like all malformations, occurs in the broad spectrum, which can be classified by relating the level of the rectal pouch to the pelvic floor or puborectal muscle: (a) low or translevator cases, (b) intermediate cases, and (c) high supralevator cases. In the least severe form, the rectal pouch is low, communicating with the perineum through a fistula, usually anterior to the normal anal location. As severity increases, the fistula moves forward to the vaginal fourchette in girls or to the base of the scrotum in boys. When the pouch is still higher, there is usually a fistula from the rectum to the vagina in girls or the urethra in boys. Occasionally, the fistula ends in the bladder. Rarely, no fistula is present. Development of the lower gut is intimately associated with development of the lower genitourinary tract. It is not surprising, therefore, that if the bowel develops abnormally, the genitourinary tract is likely to do the same. The more severe rectal anomalies have a higher likelihood of having associated genitourologic defects than do minimal cases. The association of genitourinary malformations with imperforate anus has been documented in many series. 1,3,5,8,9 and 10 For this reason, it is important to get screening diagnostic studies of the urinary tract in all babies with imperforate anus, including those with a low anomaly. Magnetic resonance imaging (MRI) allows one to screen the lumbosacral spine in all of these babies to determine whether the spinal cord ends normally at the level of the first lumbar vertebra or is tethered downward. Neurosurgical release of a tethered spinal cord can prevent secondary neurologic deficit from stretching of lumbosacral nerve roots during growth of the patient. Detethering the spinal cord does not reverse already established nerve deficit. The most common urologic problems, aside from the rectal fistula usually present, are vesicoureteral reflux, renal agenesis, and cryptorchidism. Other problems include ureteropelvic junction (UPJ) stenosis, horseshoe kidney, ectopic ureter, ureterocele, posterior urethral valves, hypospadias, crossed renal ectopia, and vaginal duplication. Although cloacal malformations are the most severe end of the spectrum of imperforate anus in female infants, imperforate anus and cloacal malformations should be considered as two separate topics because they differ greatly in their physical appearance in the newborn, the diagnostic measures that should be performed, and their surgical management, both early and late. At the farthest end of the spectrum is cloacal exstrophy, which is also called “vesicointestinal fissure.” It occurs once in 200,000 to 400,000 births and is said to occur from premature rupture of the cloacal membrane before descent of the urorectal septum. Once uniformly fatal, the majority can be reconstructed today to provide a satisfactory functional result and quality of life.

DIAGNOSIS
In patients with an imperforate anus, an upside-down film has traditionally been obtained in neonates with no fistulas to determine the level of the rectal pouch, but this film can be misleading. A very high pouch can appear low if the infant is exhaling, and a low pouch can appear high if the infant is taking a breath when the picture is taken. I consider the invertogram to be of little practical value. I never explore the perineum to look for a low pouch; it is an invitation to disaster. If there is no fistula and no meconium staining of the perineum, I assume it is a high pouch and perform a colostomy. Cystourethroscopy will usually show the location of the rectourethral fistula in the male. Soon after birth, screening urologic studies are performed. They include a voiding cystourethrogram to rule out vesicoureteral reflux, posterior urethral valves, and other lower tract anomalies. The upper tracts are assessed by ultrasonography or intravenous pyelography. The rectal fistula can be visualized via instillation of radiopaque contrast medium in the distal limb of the colostomy. When imperforate anus occurs in the most severe form in female infants, the rectum, vagina, and bladder all converge to enter a common passage to the perineum, the urogenital sinus. Like imperforate anus itself, cloacal malformations vary in all degrees of severity. There can be low confluence of all of these structures, or the three organ systems can converge high, at the bladder neck, or even in the trigone of the bladder. It is vital, therefore, that these infants be studied completely both endoscopically and radiographically in order to plan their management. At birth, the infant often shows marked abdominal distention. Abdominal roentgenogram may show a large, rounded “cyst” with an air–fluid level. This is the vagina, distended in utero by urine, and filled with gas postnatally from the colon fistula. The chronically distended vagina displaces the bladder forward, angulating the bladder neck. This often causes hydronephrosis from chronic intrauterine obstruction of the lower urinary tract. The appearance of the perineum varies in infants with cloacae. Most commonly, the perineum is blank, showing no obvious opening of anus, vagina, or urethra. At the mons, there is a pseudophallus, which is smaller than a penis and larger than a normal neonatal clitoris. Rudimentary labia minora may be present. The urogenital sinus may open at the base of the phallic structure or may extend to its tip like a stenotic urethra. In others, the clitoris, labia, and vaginal introitus may look almost normal, but probing of the introitus discloses only one actual opening, the urogenital sinus. There may be a good gluteal cleft with an indentation where the anus normally would be located. If the sacrum is abnormal, however, the perineum is often rounded in appearance (“rocker bottom”). This usually denotes poor innervation. Some infants present with the forme fruste of a cloacal malformation in which there is a urogenital sinus into which the bladder and vagina empty, but the anus opens on the perineum more anteriorly than it should. Repair requires moving it further back. Radiographic evaluation should include MRI of the lumbosacral spine to rule out spinal cord tethering, intravenous pyelogram or ultrasound of the upper tracts, and injection of contrast into the urogenital sinus to visualize the bladder, vagina, and rectal fistula. After colostomy has been performed, the distal limb of the colostomy is filled to better delineate the confluence of bowel, vagina, and urogenital sinus. When I evaluate a patient who has undergone prior vesicostomy or vaginostomy, those orifices are injected also. Thorough endoscopic visualization of the anatomy is mandatory. In the distended neonate who cannot tolerate a long anesthesia, this may be an abbreviated procedure to accomplish a cutback of the urogenital sinus far enough to allow intermittent catheterization to decompress the distended vagina and, secondarily, the bladder. An indwelling catheter is not desirable. If it enters the vagina, it may empty the vagina but simultaneously obstruct the adjacent bladder neck. If it enters the bladder, it can empty the bladder but not the vagina. One can pass through the endoscope two small plastic catheters, one into the bladder and another into the vagina. When there are two vaginas separated by a midline septum, the septum can be incised with a cutting electrode to create one distended vagina to empty instead of two. In the 154 patients with cloacal malformations whom I have encountered, 20 had no vagina, 68 had two vaginas, and 66 had one vagina. The internal anatomy in cloaca cases can vary just as widely as the external appearance of these infants. The colon fistula usually ends in the posterior midline of the

vagina just inside the confluence of the vagina with the urogenital sinus. The vagina can enter the urogenital sinus at any level from just beneath the skin of the perineum up to the bladder neck. Indeed, the vagina can empty into the bladder itself. If there are two vaginas, there is usually a vertical septum between them, and the rectal fistula is at its base. Other locations where the rectal fistula can enter include (a) directly into the urogenital sinus caudad to the entry of the vagina, (b) into the apex of the vagina with a long fistula running down the back wall of the vagina, and (c) anterior to the vagina, just behind the urethra. All of the anatomic details of a cloacal case should be worked out preoperatively by radiographic and endoscopic study. In especially complex cases where there are additional orifices visible at endoscopy but their significance is not clear, the radiologist can come to the operating room to perform fluoroscopy while catheters are placed in the various openings to instill contrast medium. Cloacal exstrophy is shown in Figure 105-1. There is an open cecum between the two hemibladders and a blindly ending colon hangs down in the pelvis behind the cecum. There is no anus or rectum. Ampholocele is present. The terminal ileum may prolapse giving an “elephant trunk” appearance of the everted bowel.

FIG. 105-1. Newborn cloacal exstrophy demonstrating typical anatomy with the prolapsed ileum appearing like an elephant' trunk.

INDICATIONS FOR SURGERY
In patients with imperforate anus, if urologic evaluation studies show an obstructive uropathy such as UPJ obstruction or massive vesicoureteral reflux, the urologic problem should be corrected first. The anorectal repair is elective in its timing once colostomy has been performed. The urinary tract poses the greatest risk to life in cloaca cases. Therefore, if there is a urologic problem such as massive reflux, it should be corrected first. The rectal pull-through should never be done as a preliminary procedure. There is no need to hurry with the rectal repair, and a previous rectal pull-through can significantly complicate repair of the more difficult aspect of the anatomy, namely, the convergence of the genital tract with the urogenital sinus. In cloacal exstrophy, although two-thirds of these patients are genetically males, there is usually a small hemipenis on each side that makes penile reconstruction much less feasible than in simple exstrophy-epispadias cases. Most, but not all, surgeons think the female gender role is advisable in most of these babies because in later life an “incomplete” female can cope better than an “inadequate” male.

SURGICAL TECHNIQUE
Imperforate Anus The newborn with a low rectal pouch usually has a fistula to the perineum, or the posterior fourchette of the vagina in girls. When the opening is large enough that meconium is passing, it is possible to gently dilate the fistula to decompress the colon and perform an elective anoplasty several weeks or months later. This may be the wisest course in an infant with multiple anomalies like esophageal atresia, duodenal atresia, congenital heart disease, and imperforate anus. In full-term, newborn babies in good general condition, however, definitive anoplasty is usually done immediately for low anomalies. In some cases, a fistula is present but may escape detection unless high-powered loupes are used to see it. Green discoloration of the perineum or meconium in the scrotal raphe usually denotes a low pouch. Sometimes the fistula can track subcutaneously for several centimeters in the scrotal raphe. Perineal operation suffices to exteriorize a low-lying rectal pouch. When no fistula is visible, it is best to assume that the patient has a high pouch with a fistula to the urinary tract in boys or to the vagina in girls. Only rarely is there a low pouch without a fistula. In most high anomalies, it is best to proceed with a decompressing colostomy. I prefer a lower abdominal vertical incision, exteriorizing the upper sigmoid colon as a divided loop colostomy. The ends are brought out through a separate small incision in the left lower abdomen. It is important to leave enough colon distal to the colostomy to allow pull-through 6 months to a year later. Failure to do that may make a pull-through impossible without simultaneously taking down the colostomy, which is usually to be avoided. Because most of these infants have a fistula to the urinary tract, loop colostomy should be avoided. This decreases the likelihood of fecal spillover from the distal colon into the urinary tract. In some infants, urine passes through the fistula to the distal colon segment, causing significant absorption of urinary solute and hyperchloremic acidosis. That can be managed by early definitive pull-through surgery, with division of the fistula between the rectal pouch and the urinary tract, or by temporary drainage of the distal colon segment by placing a small, soft catheter in the inferior limb of the colostomy. Since deVries and Peña described the advantages of the posterior sagittal position for definitive correction of the majority of imperforate anus malformations, I have used that approach almost exclusively, except for female infants with closed exstrophy anomalies. 2,7 The posterior sagittal approach is shown in Fig. 105-2. With the infant in the prone, jack-knife position, the perineum is stimulated electrically, marking the point of maximum sphincteric contractility, where the anus should be positioned. An incision is made vertically from the coccyx through that point, dividing the pelvic muscles in the midline and exposing the rectal pouch. It can be helpful to have a Foley catheter in the distal colon segment by way of the colostomy. Air is insufflated into it through a syringe to distend the lower rectum as an aid in its identification. Furthermore, it is vital to position a metal sound in the urethra to locate it easily and avoid injury to the urethra. The lower rectal pouch is opened posteriorly, exposing the fistula. The fistula is isolated and divided. It is important to avoid narrowing the urethra when it is closed with interrupted sutures. The rectum is mobilized widely up to the cul-de-sac peritoneum. It is then tapered in caliber, so that it fits comfortably in the muscle complex, which is closed over it. Urethral drainage is maintained with a soft polymeric silicone (silastic) catheter or suprapubic tube for 10 days.

FIG. 105-2. Imperforate anus. (A) Repair through posterior sagittal approach. Anatomy showing muscle complex of levator ani and sphincters. (B) Vertical incision in prone position. Nerve stimulator used to locate center of sphincter muscle mass (marked with skin pencil). (C) Muscle complex has been opened to reveal rectal pouch. (D) Fistula to urethra isolated and closed. This closure must not compromise the urethra, nor must a “diverticulum” of rectum remain that can form stones. Rectal pouch mobilized widely to gain length for pull-through (a finger in the rectum is helpful). (E) End of rectal pouch trimmed and anastomosis to perineum is started. The figure shows side wall to be excised, thereby narrowing the rectum so that it will fit into the muscle complex that will be closed around it. (F) Completed tapering of rectal pouch. (G) Completed repair (sagittal view in prone position). New anorectal canal will be dilated for several weeks postoperatively before colostomy

closure.

A very high rectal pouch in the male is managed differently because mobilization from a posterior approach can be very difficult. In those cases, I open the posterior wound in the prone position but then turn supine to mobilize the rectal pouch. The fistula is ligated and divided from above, the colon is then pulled through the open perineal wound. The belly is then closed. The child is turned prone and the case proceeds as shown in Fig. 105-2. Dissecting a very high pouch from behind not only is tedious but also risks injury to the bladder neck, seminal vessels, and ureters. Conversely, mobilization of the terminal colon from above is very easy. Gentle dilation of the rectal passage is begun 10 days later to prevent postoperative stenosis. The colostomy can be closed 4 to 6 weeks later, sometimes sooner. Postoperative anorectal control is much better today in cases operated in this manner, as compared to operations used in former decades where the rectal pouch was not tapered and was not placed so accurately within the muscle complex. Cloacal Malformations These infants must be decompressed immediately, including both the colon and the urinary tract. I prefer performing a right upper quadrant divided loop transverse colostomy in these infants because the left colon may be needed in performing the definitive repair when the infant is older. End colostomy and left lower quadrant loop colostomy are to be avoided; they can complicate subsequent repair. Usually, the lower urinary tract can be decompressed by intermittent catheterization through the urogenital sinus opening. The catheter usually passes into the distended vagina, emptying it. Seldom does it pass forward into the bladder because the bladder is displaced anteriorly by the large vagina. Sometimes the urogenital sinus opening must be cut back a few millimeters to allow endoscopy; it may be necessary to decompress the vagina first by passing the endoscope through and draining it. This allows the bladder and bladder neck to fall posteriorly so that the endoscope can be passed into the bladder. If decompression by intermittent catheterization is not possible, a percutaneous cystostomy or vaginostomy tube can be placed, or vesicostomy can be performed. Open decompression, however, is not needed in most of these cases. A thorough endoscopy should always precede the definitive repair in these babies in order for the clinician to be completely familiar with their anatomy. The colon segment distal to the colostomy performed at birth generally remains filled with inspissated meconium or mucus. This should be cleaned out of the distal colon before embarking on a major reconstruction. In some babies, this is done by passing a ureteral catheter into the fistula, following it with the endoscope. By turning the scope upside down, the catheter lifts the upper margin of the fistula, allowing the beak of the endoscope to pass when it is at the 6 o'clock position. Irrigating saline, not water, through the scope can wash the mucus upward and out of the defunctionalized limb of the colostomy. In others with solid material present, a Foley catheter is passed through the fistula, filling the distal colon and removing pieces of solid material by probing the distal colon segment above with stone forceps. The preoperative endoscopy with emptying of the colon segment can be time consuming. Therefore, it is performed several days before the definitive operation, and not as a preliminary to the reconstruction. This is an important part of the preparation of these patients preoperatively. Retained fecal material in the defunctionalized bowel could result in disastrous contamination of the wound from intraoperative spillage. All patients have bowel preparation and antibiotics. Although endoscopy is done several days earlier, it should be repeated to rinse out the bladder, vagina(s), and rectum with antibiotic solution or dilute betadine. This reduces the likelihood of bacterial contamination. The patient is prepared and draped circumferentially as shown in Figure 105-3 so that the sterile field extends from the midthorax downward. This gives access to the perineum in the prone position, the belly in the supine position, or the perineum in the lithotomy position by holding upward the legs, which are wrapped in sterile drapes. Usually the operation commences from behind in the prone, jack-knife position.4 An exception would include a patient whose vagina is tethered upward by a previous vaginostomy, which would prevent bringing it down to the perineum. In my initial experience with cloacal malformations, I used abdominal perineal exposure. In recent years, I also have used the posterior sagittal approach because it facilitates separating the confluent structures from behind.

FIG. 105-3. Preparation of patient to allow prone, supine, and lithotomy positions during operation.

First, the perineum is stimulated electrically, marking the proper location for the anus. Skin marking dye can wash away. Therefore, two fine sutures are placed in the skin on each side as markers at the new anal site as well. Knife dissection is deepened in the posterior sagittal plane stimulating the muscles on either side as a guide to staying in the midline. One person can spread the tissues apart more evenly than two, who may pull with unequal force. It is important to place a urethral sound in the urogenital sinus. By elevating the sound firmly, the surgeon can cut into the posterior wall of the urogenital sinus, which is tented up by the sound. If the distal urogenital sinus is of normal caliber, there is no need to open it all the way to its meatus. If it is too wide and needs to be tapered in caliber, this incision should be brought to its distal end. Once the urogenital sinus is open, the incision in it can be extended more proximally until the structures that enter it are seen. The relationship of the rectal fistula, the vagina(s), and the bladder neck is determined endoscopically before the procedure. As each structure is entered, I place in it a small Foley catheter, filling the bag with air. The catheter can provide a means for gentle traction to facilitate separating the structures. Alternatively, multiple fine traction sutures can be placed in the end of the rectum and vagina to provide gentle traction during separation and mobilization. The surgeon should wear magnifying loupes to do this tedious separation. It is done by sharp dissection, not cautery, which can burn the delicate structures that are closely related one to another. When separating the rectum from the vagina, if the plane of dissection is difficult to define, it is better to err on the side of entering the vagina rather than the front wall of the rectum. This reduces the likelihood of having a colonic fistula develop postoperatively from a suture line in the front wall of the rectum. The colon is separated from the vagina all the way to the cul-de-sac peritoneum both in front and in back. Its lateral attachments must be divided to provide enough rectal length. The height of the pouch will determine how high this dissection must be pursued. If the dissection must be carried high, splitting the coccyx can help. The most tedious aspect of this dissection is separating the vagina from the urogenital sinus and bladder neck. The lower vagina usually encircles the upper urogenital sinus for almost half of its circumference, and so this plane is often difficult to establish. This is especially true when there are two vaginas and they diverge laterally away from each other. It is important to avoid inadvertent injury to the back wall of the bladder and bladder neck. Therefore, in performing this dissection, it is safest to err on the side of inadvertent entry to the vagina, not the urinary tract. To facilitate separating the structures, I usually open the back wall of the vagina to view it from the inside while doing that dissection. Once the lower vagina is separated from the urinary tract, freeing it further is easy. In an older patient, or in a secondary case where the rectum and vagina are large enough, it is a great help to have a finger in the vagina or rectum as a guide to its mobilization. This placement also can provide help in taking down a colostomy if the size of the bowel permits insertion of a finger. Another aid in dissecting the vagina from the back wall of the bladder is to distend the bladder with air or saline to facilitate keeping in the correct tissue plane. When the three organ systems are separated, the urogenital sinus must be closed to form the urethra. If it was initially too wide and has been opened all the way to its end, parallel incisions should be marked out with a marking pencil or brilliant green dye applied with a sterile toothpick. The edges are trimmed appropriately. At the vaginal introitus that tissue should be saved; it may provide flaps to join to the vagina. It is important to keep the urethral mucosal strip wide enough, so that stenosis is not created when it is closed in two layers to make a urethra. The first layer is a running inverting closure over a small straight plastic catheter from the bladder neck to the meatus. The suture used is generally a 3-0 or 4-0 monofilament polydioxanone (PDS) suture. A Foley catheter is not used because inadvertent traction on a Foley catheter postoperatively could pull an inflated bag through the repair and disrupt it. A second layer of interrupted sutures is placed. There is ample tissue at the introitus but seldom any excess tissue adjacent to the bladder neck or proximal urogenital sinus. Thus, it is vital to maintain the integrity of the tissue when dissecting

away the vagina. A two-layer closure of good tissue not traumatized by excessive handling or burning with cautery is the best insurance to avoid a postoperative urinary fistula. Next, the vagina is brought down. In a favorable case its mobilization will allow bringing it without tension to the perineum. It should be appreciated that when the wound is closed the perineal tissues reach higher than when they are being spread apart by a self-retaining retractor. In some cloacal cases, the vagina or vaginas are too small to reach the perineum. Interposition of a bowel segment is necessary. Even some relatively large vaginas will not reach the perineum because they are too high in the pelvis despite wide mobilization. Thus, bowel may be needed to extend downward a vagina that preoperatively looked as though it would reach the perineum. The best tissue in the vagina after wide mobilization is its lateral walls. This is especially true if the back wall was opened to view the front wall from the inside during separation from the bladder neck or if inadvertent openings are made in the anterior wall of the vagina during separation from the bladder neck. These openings are closed if possible. The vagina is then rotated 90 degrees to the right or left so as to have intact vaginal wall overlying the closure of the bladder neck and upper urogenital sinus. If a bowel extension is required to get the vagina down to the perineum, the suture line of the bowel to the vagina should be high enough to have intact bowel over the closure of the bladder neck and urogenital sinus rather than two adjacent suture lines. Following these basic principles, I have not had a postoperative vesicovaginal fistula occur. If a bowel segment is needed to extend the vagina to the perineum, the question will arise regarding what segment to use. When the rectum has been mobilized widely, so that it lacks any intrinsic local innervation, which might make it the preferred segment for rectum, it may be better to use it to extend the vagina. This is especially true when the patient has a left lower quadrant colostomy that can be taken down and pulled through as rectum. If, however, the true rectum is to be used as rectum, it may be better to take a segment of sigmoid colon to extend the vagina, or even a piece of small bowel. This is the reason to avoid disrupting the colon blood supply in the neonate by doing a left-sided colostomy in a cloaca case. When using small bowel, that is usually terminal ileum, the mesentery of which is long enough to get the bowel to the perineum. When making a small bowel vagina, it is better to reconfigure it as an upside-down “U,” sewing those limbs together to make it large enough in caliber. Anastomosis of the vagina to the bowel segment used to extend it has usually been done from above at laparotomy. In my early experience with cloaca, I made large flaps of vagina to rotate down to the perineum. This gave anatomic continuity, but a narrow lower vagina. Some of those children later needed to have the vagina enlarged. This can be done using an inlay flap of buttock skin and subcutaneous fat. I have not favored the method of making flaps of vaginal tissue in recent years. Flaps of introitus, or adjacent skin, can be fashioned to reach upward a short distance when not much extra length is needed. Excessive tension should be avoided lest the vagina retract. After the vagina has been joined to the perineum, the muscle complex is closed behind it to create a perineal body and the anterior part of the rectal sphincter. The colon is then brought downward and tacked meticulously to the surrounding tissues to prevent retraction and/or prolapse. Usually the rectal caliber is tapered, marking the amount to be removed first to avoid removing too much. Its mucosa is closed with a running, inverting suture. It is useful to place a Hegar dilator in the lumen to avoid making it too small. A second layer of interrupted sutures is placed in the muscular wall of the colon. There should be no problem with this two-layer closure, which is in back and not adjacent to the vaginal wall. The muscle complex is closed meticulously over this, with special care to reconstitute the sphincter at the perineum. The stent is removed from the urogenital sinus 10 days postoperatively. Contrast is instilled through a suprapubic tube placed at the time of surgery. The tube is placed percutaneously in cases repaired solely from behind, or a Malecot catheter is placed in the bladder from above if the patient required a laparotomy. Contrast medium instilled through the tube demonstrates absence of a leak and the configuration of the urogenital sinus. When the suprapubic tube is clamped, some children void without residual. Others do not. The tube should remain until the patient demonstrates ability to void without significant residual. If this does not take place 2 weeks postoperatively, a soft catheter is passed gently from below. If it goes into the bladder easily, the suprapubic tube can be removed and intermittent catheterization started. If there is difficulty passing a catheter gently, the trial of intermittent catheterization should be discontinued until the urogenital sinus has had a longer time to heal. If the child cannot void and a catheter is not passed easily, endoscopy should be performed on the new urethra with the child under anesthesia to determine what type of catheter is easiest to pass. In some that proves to be a soft coude-tipped catheter. Cloacal Exstrophy Some important principles have evolved. First, the bowel must be separated from the hemibladders, which are joined to form a small bladder. The cecum and microcolon should be spared for use as the distal gastrointestinal tract. The microcolon can dilate and elongate remarkably once used. Since gastroenteritis with diarrhea and dehydration is a major threat to these babies, it is important to retain all colon for its physiologic purpose of water resorption. It should not be used for bladder augmentation or as a urinary conduit as was commonly practiced in the past. After neonatal omphalocele closure, bladder closure, and separation of the bowel, much reconstructive surgery remains to be done. Iliac osteotomies can help approximate the wide apart pubic bones and the attached soft tissues. In some patients a bladder neck and urethra can be made. In others it cannot, and continence may rely on a reversed bowel nipple, a fascial sling, or another way to provide continence. In some, a “continent diversion” with a small catheterizable stoma is the best choice. In nearly all cases, bladder augmentation is needed. Most require ureteral reimplantation for reflux. All cloacal exstrophy patients should have MRI imaging of the spine to demonstrate downward tethering of the cord, which has been noted in all but one of our 43 patients with cloacal exstrophy. In some, neurosurgical release of the cord is required. Some of these patients have good perineal muscle and can undergo pull-through of the colon. In others with no contractile muscle, and sometimes a lipomeningocele of the buttocks, pull-through is not possible and permanent end colostomy will be required.

OUTCOMES
Complications Complications from cloacal surgery include perioperative as well as long-term sequelae. The perioperative complications include the risks of urinary leakage or obstruction as well as bowel leakage and obstruction. These are often long, complex operations that require expert anesthesia and constant monitoring of arterial blood pressure, venous pressure, serum electrolytes, hematocrit, and urinary output. Intraoperative fluid requirements may be as high as 20 to 25 ml/kg/hr if there is an extensive abdominal component to the reconstruction. Two children in this series died as infants prior to having reconstructive surgery—one from renal dysplasia and the second from severe congenital heart disease. Despite the complexity of many of the cases, there were no early or late deaths in those who underwent reconstruction. Long-term complications can include incontinence of urine or feces, recurrent urinary infections, inability to urinate, and due to the wide spectrum of presentation as well as the relative rarity of the condition, there is no single repair that is appropriate for all patients. As such the surgeon or team of surgeons must be completely familiar with all aspects of major intrapelvic and intraabdominal surgery in infants and children and must have expertise in surgery of the urinary tract, the genital tract, and the bowel; in addition, it should be done in those centers where major pediatric reconstructions are performed routinely. Technical misadventures in surgery of this magnitude can be disastrous. Results I have treated 154 cloacal cases; 94 were primary, either neonates with no prior surgery or older babies who had undergone only diverting colostomy. Sixty were secondary cases in which a major procedure had been performed previously, such as rectal pull-through, vaginostomy, or a urologic procedure. To emphasize the serious urologic aspects of these cases, 39 patients had undergone prior urinary diversion (8 ileal loop, 8 ureterostomy, and 22 vesicostomy). Vesicoureteral reflux needed surgical management in 96 of the patients. Some of our reconstructed patients with an abnormal sacrum have required continuing intermittent catheterization to empty. Therefore, all of these patients should know that this may be necessary. Others required catheterization initially but later were able to empty spontaneously. Most of the reconstructed patients do not require catheterization. Parents should be taught to pass daily a Hegar dilator of appropriate size into the vagina and into the anorectal passage. This averts stenosis, which can occur if dilation is not done. Initially it is done daily. After several weeks, it is done every other day, and with decreasing frequency until it is no longer needed. The size of the dilator varies with the patient's age. Fecal continence varies depending on whether the sacrum is normal, just as in other patients with less complicated imperforate anus. Some empty their bowel normally with excellent continence. Others require a daily enema washout to evacuate the colon and remain clean throughout the day. Long-term follow-up is mandatory in these children. An initially disappointing result with respect to continence often improves as the child grows older, especially if the repair was done at a young age. Many of these patients are too young to report sexual function. However, 17 who are adults have reported having coitus. Thirteen are married. Six have delivered

children, one vaginally, and the rest by cesarean section. An ideal cloacal case is one with no previous surgery except colostomy, good upper tracts, and a vagina and rectum that can be brought to the perineum through the posterior sagittal approach. Unfortunately, this ideal circumstance is often not present because the patient has had previous surgery or the anatomy is unusual. The following cases illustrate the spectrum in repair of cloacal malformations. Case 1: Primary Case Without Urologic Complications and Repairable Through Posterior Sagittal Approach An infant in whom a divided transverse colostomy had been performed at birth. Though an intravenous pyelogram initially showed moderate bilateral hydronephrosis, intermittent catheterization of the distended vagina was begun, the child thrived, and hydronephrosis disappeared. Endoscopy and radiography at age 16 months showed the anatomy in Figure 105-4A with a normal bladder and a rectal fistula entering the base of the septum dividing two vaginas. As reconstruction was performed, the vaginal septum was incised with a cutting electrode through the cystoscope. Using a posterior sagittal midline approach in the prone, jack-knife position, the urogenital sinus was opened. The rectum was dissected free from the vagina. The vaginas were separated from the urogenital sinus. The urogenital sinus was narrowed slightly, closing it in two layers and leaving a catheter through it into the bladder. The two vaginas, converted into one by opening the septum, were brought to the perineum. The back wall, made thin by separation of the rectum from it, was plicated. A perineal body was constructed. The bulbous rectum was tapered from behind (see Fig. 105-4B). The patient voided well postoperatively when her catheter was removed. Catheterization was possible but was not needed, and colostomy closure was performed soon afterward. The patient was in diapers until age 3 years but subsequently became trained for urinary and bowel control. MRI of the lumbosacral spine showed bulbous termination of the spinal cord at L1 but no evidence of tethering. Now age 14 years, she is a normal girl, a good athlete, with normal urinary and bowel function and menstrual periods. This case illustrates that with good upper tracts, normal spinal cord and sacrum, operable from behind, a normal functional result should be possible in most cases.

FIG. 105-4. Case 1. (A) Preoperative anatomy with moderately high confluence of vaginas into urogenital sinus. (B) After reconstruction, which was possible from posterior sagittal approach with pull-through of vagina and rectum without need to enter abdomen.

Case 2: Primary Case with More Complex Anatomy and Severe Vesicoureteral Reflux Requiring Correction Before Repair of Cloacal Malformation This is an infant in whom a right transverse colostomy had been performed on the first day of life. An intravenous pyelogram showed severe bilateral hydronephrosis and further workup showed high entry of two vaginas into the urogenital sinus near the bladder neck ( Fig. 105-5A). In addition, a long rectal fistula entered the base of the intravaginal septum, running up to the apex of the vaginas, where the rectal pouch was located. There were bilateral bladder diverticula which each contained a ureteral orifice, accounting for reflux.

FIG. 105-5. Case 2. (A) Preoperative anatomy. This infant required early attention to the problem of massive reflux causing pyelonephritis, deferring the rest of her reconstruction until later. (B) The operation is begun in the prone position. The vagina is disconnected. Drains are placed in passages for subsequent pull-through of bowel extension of vagina and rectum. (C) Completed repair in the supine position after five operations.

Initially, the midline vaginal septum was incised to make a single vagina and the urogenital sinus was cut back a few millimeters to permit intermittent catheterization of the vagina every 4 hours. The infant thrived but had recurrent urinary infection despite antibiotic therapy. Cystogram showed continuing massive vesicoureteral reflux. Therefore, at age 7 months, bilateral ureteral reimplantations were performed and the urinary infections stopped. When the child was 16 months old, the cloacal reconstruction was performed (see Fig. 105-5B, C). Through a posterior sagittal approach, the urogenital sinus was opened and the vaginas were separated from it. The urogenital sinus was repaired. Despite maximum mobilization, the vagina would not reach lower than the coccyx. The rectum was 4 cm above the level of the coccyx. Thus, total repair from behind, as done in case 1, was impossible. The child was turned into the supine position. The distal colon was used on the sigmoid arterial blood supply to extend the vagina to the perineum. The proximal sigmoid colon was pulled through to the perineum as rectum. She voided well postoperatively, allowing early removal of the suprapubic tube and discontinuance of intermittent catheterization, which had been done every 4 hours since age 2 weeks. The colostomy was closed 4 months later, at age 20 months. The patient is now 14 years old and has normal urinary control with stable upper tracts and no sign of either reflux or hydronephrosis. She uses an enema washout each day to empty the colon, which keeps her free from soiling, and she douches to remove vaginal mucus. Clearly, the abnormality of the urinary tract was the life-threatening problem in this child with massive reflux causing pyelonephritis. In retrospect, it would probably have been better to correct the reflux sooner than age 7 months because her reflux had little likelihood of spontaneous regression with the ureters located in diverticula. MRI of the lumbosacral spine showed no tethering of the spinal cord. In the past 9 years MRI studies were performed on 59 patients with cloacal malformations, of which 21 showed tethering of the spinal cord. Case 3: Secondary Case with Previous Rectal Pull-through and Unrepaired Genitourinary Malformations A 9-year-old girl with a cloacal malformation underwent a colostomy and tube vaginostomy at birth. At age 4 months, a loop ureterostomy was performed for left hydronephrosis and a rectal pull-through was done at age 2 years. This resulted in fecal incontinence, two indwelling tubes, and recurrent urinary infections for the next 7 years. Endoscopy and radiography were performed after injection of each orifice and tube. The girl's anatomy was found to be as shown in Figure 105-6A. At subsequent reconstruction (see Fig. 105-5B), beginning through the posterior sagittal approach, the vagina was separated from the urogenital sinus. It would not reach the

perineum, however. The sigmoid colon was used to extend the vagina to the perineum. The rectum was repositioned in the muscle complex, which did not surround it initially because the pull-through had been done too far anteriorly. The urologic drainage included (a) reimplantation of the right ureter, (b) closure of the cystostomy, (c) removal of both ureters which emptied ectopically into the vagina, using the better one for transureteroureterostomy, and (4) correction of UPJ obstruction of the poor lower pole of the kidney and drainage of the better upper pole of the kidney by pyeloureterostomy.

FIG. 105-6. Case 3. (A) Preoperative anatomy. Note tubes had been in bladder and vagina since infancy, and exteriorized upper pole of left kidney. (B) Scheme of repair starting in prone position and turning the patient to complete operation. (C) Completed repair.

Intermittent catheterization was necessary for 3 months after this reconstruction but was discontinued when the patient proved that she could empty the bladder without residual urine. Preoperatively, the patient had no rectal control, with the original pull-through anterior to the muscle complex. Early postoperative rectal control was described as much improved but not perfect. She was given a daily enema washout to prevent soiling for 4 months. Control improved further during the next year, and 4 years postoperatively her urine and bowel control were described as normal. This case underscores the important psychosocial aspects of incontinence of urine and stool. Like many of the secondary cases we have seen, rectal pull-through had been done during infancy, without tending to the more pressing genitourinary problems. In some children, the urinary tract should be done first. In others, all three tracts can be repaired simultaneously, as was done during her secondary reconstructive surgery. The anorectal pull-through almost never should be done as a preliminary procedure in an infant, putting off the other aspects of repair until later. Case 4: A Favorable Cloacal Exstrophy Case A 26-month-old infant was referred for further urologic management ( Fig. 105-7A), in whom an antenatal diagnosis had disclosed an abdominal wall defect. As a neonate the patient underwent a separation of the gastrointestinal tract from the bladder, closure of the bladder and omphalocele, and primary pull-through of the colon to the perineum. Testicles were removed and the patient was raised as a female. She had excellent bowel function with a daily enema washout.

FIG. 105-7. Case 4. (A) Preoperative anatomy showing a patulous bladder opening and normal kidneys with reflux. The colon had been pulled through in the newborn period. (B) After reconstruction showing narrowed bladder neck, cross-trigonal reimplants, and small bowel augmentation. The rectum was moved back to make room for a small bowel vagina.

The bladder capacity was small and there was vesicoureteral reflux to two normal kidneys. At age 34 months, iliac osteotomies were performed in order to bring together the pubic symphysis, the ureters were reimplanted by cross trigone technique, the bladder was augmented with ileum to increase its capacity, the rectum was moved more posteriorly, and a small bowel vagina was placed between the bladder and the rectum ( Fig. 105-7B). One month later, reoperation for gastric volvulus was performed. Seven months later, a repeat operation was done to narrow the bladder outlet further because the patient was still wet. At age 14, she is dry and intermittently catheterizes every 4 hours without difficulty. Her upper urinary tract is stable. She requires suppressive antibiotics to treat recurrent urinary infections, which seem confined to the bladder because the upper urinary tract is stable and she is without fever. The colon is emptied each day with a soap-suds enema, and the child is free from fecal soiling on this regime. These patients adapt well to this program and can lead fairly normal social lives as children, free from the burden of diapers or appliances. Our current preference for bladder augmentation is gastrocystoplasty. All infants with imperforate anus should be assessed for other malformations, especially in the urinary tract. The cloacal malformation, which represents the most severe type of imperforate anus in females, can be successfully managed if certain general principles are followed. The colon requires urgent decompression in the neonate. I prefer a right divided loop colostomy. The urinary tract must be assessed and often needs early intervention. Evaluation of these children should include MRI of the lumbosacral spine to identify those with tethering of the spinal cord, which can be released by neurosurgical intervention. Cloacal exstrophy patients present an especially difficult challenge. However, there is a satisfactory solution possible for most that will give a good quality of life. 6 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Belman BA, King LR. Urinary tract abnormalities associated with imperforate anus. J Urol 1972;108:823–824. DeVries PA, Peña A. Posterior sagittal anorectoplasty. J Pediatr Surg 1982;17:638–643. Hendren WH. Urological aspects of cloacal malformations. J Urol 1988;140:1207–1231. Hendren WH. Cloacal malformations. In: Walsh PC, Retik AB, Stamey TA, Vaughan DE Jr, eds. Campbell's urology. 7th ed., vol. 2. Philadelphia: WB Saunders 1998;1991–2018. Hendren WH, Oesch IL, Tschaeppeler H, Bettex MC. Repair of cloacal malformation using combined posterior sagittal and abdominal perineal approaches. Z Kinderchir 1987;42:115–119. Hendren, WH. CLOACA: The most severe degree of imperforate anus. Experience with 195 cases. Ann Surg 1998. In press. Peña A, deVries PA. Posterior sagittal anorectoplasty: Important technical considerations and new applications. J Pediatr Surg 1982;17:796–811. Rich MA, Brock WA, Peña A. Spectrum of genitourinary malformations in patients with imperforate anus. Pediatr Surg Int 1988;3:110–113. Santulli T, Kiesewetter WB, Bill AH Jr. Anorectal anomalies. A suggested international classification. J Pediatr Surg 1970;5:281– 287. Templeton JM, O'Neill JA Jr. Anorectal malformations. In: Welch KJ, Randolph JG, Ravitch MM, O'Neill JA Jr, Rowe MI, eds. Pediatric Surgery. 4th ed. Chicago: Year Book, 1986;1022.

Chapter 106 Ambiguous Genitalia Glenn’s Urologic Surgery

Chapter 106 Ambiguous Genitalia
Richard I. Silver and John P. Gearhart

R. I. Silver: Department of Surgery, Division of Urology, Alfred I. DuPont Hospital for Children, Wilmington, Delaware 19803. J. P. Gearhart: Department of Pediatric Urology, Johns Hopkins University School of Medicine, The James Buchanan Brady Urological Institute, Baltimore, Maryland 21287-2101.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Clitoroplasty Vaginoplasty Outcomes Complications Results Chapter References

The term “ambiguous genitalia” refers to genitalia in the spectrum of sexual development that are not clearly male or female. The genitalia may appear predominantly male but be ambiguous because of a small phallus, hypospadias, or a scrotum with fewer than two palpable gonads. Conversely, the genitalia may appear predominantly female, but with an enlarged clitoris or fusion of the labial folds. This situation is not just a dilemma but also a true medical emergency. Rapid determination of the genetic sex, gonadal sex, and genitourinary anatomy of a newborn with ambiguous genitalia is essential to enable gender assignment, surgical reconstruction, and acceptance of the child by its family. In addition to the serious psychosocial and surgical considerations, appropriate medical therapy may be necessary to prevent circulatory collapse due to inadequate adrenal steroid hormones. Therefore, an expeditious but careful approach to the diagnosis and management of this condition is vital to the life of the affected child. Four major categories of disorders disturb normal sexual differentiation and result in ambiguous genitalia: female pseudohermaphroditism, male pseudohermaphroditism, true hermaphroditism, and mixed gonadal dysgenesis ( Table 106-1). Of these, female pseudohermaphroditism is most common, and within this category the leading cause is congenital adrenal hyperplasia (CAH). CAH results from a deficiency in enzymes responsible for the synthesis of mineralocorticoids and glucocorticoids, resulting in overproduction of adrenal androgens. Female infants with CAH and ambiguous genitalia are said to have the adrenogenital syndrome, and 95% are due to a deficiency of 21-hydroxylase inherited in an autosomal recessive fashion. The result is a genetic female with masculinized genitalia, ranging from simple labial fusion to clitoromegaly, a glanular urethral meatus, and severe urogenital sinus anomalies.

TABLE 106-1. Causes of Ambiguous Genitalia

Because most patients with ambiguous genitalia, except genetic males with a phallus long enough to function as a penis, will undergo female sex assignment and reconstruction, the focus of this chapter will be on feminizing genitoplasty. Feminizing genitoplasty includes clitoroplasty, labioplasty, and vaginoplasty. Children with ambiguous genitalia who require feminizing genitoplasty can be divided into two groups: 1. Genetic females with extreme masculinization of the external genitalia, i.e., female pseudohermaphrodites, and 2. Genetic males with poor masculinization, as in male pseudohermaphrodites, those with mixed gonadal dysgenesis, and those with some forms of true hermaphroditism.

DIAGNOSIS
Following the birth of a child with ambiguous genitalia, clues to the cause are obtained by a thorough history, including questions regarding the pregnancy, previous births, and blood relatives. Maternal illness may have required the use of medications that affect steroid synthesis or metabolism, such as phenobarbital to treat maternal seizures. The mother is asked about exposure to other drugs that may cause sexual ambiguity, including progestational and androgenic agents. Recent virilization of the mother may suggest an adrenal or ovarian tumor that produces androgens. In cases of excess circulating androgens, either from exogenous or endogenous causes, the degree of virilization of the child will depend on the timing of the exposure. The presence of genital anomalies in other children or family members suggests a genetic mutation affecting normal sexual differentiation, including congenital adrenal hyperplasia, 5a-reductase deficiency, and androgen receptor insensitivity. Unexplained neonatal death in previous children is strong evidence for adrenal insufficiency and the diagnosis of adrenogenital syndrome. A thorough physical examination is performed, including careful determination of vital signs. Since some forms of CAH may be associated with hypertension, it is important to obtain accurate blood pressure measurements. Hydration status is assessed and careful examination of the abdomen, flank, back, perineum, groin, and genitalia performed. A diligent search is made for the presence of palpable gonads, a valuable clinical finding. Since ovaries almost never descend, the presence of a palpable gonad rules out the diagnosis of female pseudohermaphroditism. The morphology of the phallus is noted, with attention to its stretched length and curvature and the position of the urethral meatus. In cases of an absent phallus, the possibility of a genetic male with penile agenesis must be considered. The perineum is inspected for two separate urethral and vaginal openings. The vaginal orifice may be identified by the presence of opaque mucus. The labioscrotal folds are examined, which in the female should not be fused, although a minor degree of fusion is often seen posteriorly. Hyperpigmentation of fused labia majora may suggest a genetic female with adrenal insufficiency. The presence and position of the anus are noted; absence may indicate a cloacal anomaly rather than ambiguous genitalia. Digital rectal examination is performed to feel for a uterus. Over the next few days a series of tests are performed rapidly to establish the diagnosis. Genotypic sex is determined by sending blood for a karyotype as soon as possible. Although once popular, a buccal smear to detect Barr bodies is no longer recommended because it is inaccurate, does not allow assessment of mosaicism, and will not demonstrate other chromosome abnormalities. A biochemical evaluation is performed to assess adrenal and gonadal function. Serum electrolytes are monitored closely and adrenal steroid hormones and precursors, including 17-hydroxyprogesterone and pregnenolone, are determined to detect adrenal enzyme insufficiency. Serum gonadotropins, testosterone, and dihydrotestosterone are determined to detect the presence of functioning testicular tissue and the activity of 5a-reductase. If the results are inconclusive, a b-HCG stimulation test may be helpful to indicate the potential for gonadal hormone production, male phenotypic development, and male sexual function. Müllerian inhibitory substance (MIS), produced by testicular Sertoli cells, can also be measured to determine the presence of testicular tissue. An absent or low level of MIS indicates anorchia, insufficient testicular differentiation, or a defect in the MIS gene. Serum hormone levels may be normal and nondiagnostic in the first few days of life; if so, the tests are repeated. If a genetic mutation is suspected, blood can be obtained for analysis of a specific

DNA sequence. While the laboratory studies are performed, the internal anatomy is examined radiographically. Ultrasound and MRI may be used to look for the presence of Müllerian organs and gonads. The presence of a uterus and a fallopian tube excludes the diagnosis of male pseudohermaphroditism secondary to deficiencies in androgen production or tissue sensitivity. The surgical anatomy is defined by retrograde genitography, using contrast to opacify the urogenital sinus and demonstrate a vagina if one is present (Fig. 106-1). The cervix may be seen as an impression in the opacified vagina. An antegrade study, such as a voiding cystourethrogram, is inadequate because it may not illustrate the vaginal confluence with the urogenital sinus. For cases in which the internal genital anatomy and gonadal sex are indeterminate, laparoscopy and gonadal biopsy may be necessary to guide the gender assignment and surgical reconstruction.

FIG. 106-1. Retrograde genitogram demonstrating vagina entering the urogenital sinus in a female pseudohermaphrodite due to congenital adrenal hyperplasia (arrow).

INDICATIONS FOR SURGERY
The primary indication for surgery is to reconstruct the external genitalia to match the gender assignment of the child. Other indications are to remove tissue with a high potential for malignant degeneration or that may cause inappropriate sexual development at puberty. Before gender assignment, the chromosomal, gonadal, and phenotypic sex should be determined. In general, gender assignment is based on the potential for sexual function, not on the karyotype or the potential for fertility. Therefore, with rare exceptions, genotypic females should be reconstructed and raised as female. This is especially true in cases of CAH because, regardless of the degree of virilization, fertility is not compromised. Also with rare exceptions, genotypic males with ambiguous genitalia should be raised as female unless a phallus of adequate size is present or there are extenuating circumstances.

ALTERNATIVE THERAPY
Because of the nature of the conditions causing ambiguous genitalia, there are few therapeutic alternatives. Patients with congenital adrenal hyperplasia require appropriate glucocorticoid and mineralocorticoid steroid hormone replacement, with adequate therapy determined by following adrenocorticotropic hormone (ACTH) and plasma renin levels, respectively. Patients with isolated clitoromegaly, with otherwise normal genitourinary anatomy, may safely decline feminizing genitoplasty. However, these patients must be followed closely to ensure that the clitoral size becomes proportionate to the introitus over time. Those with more severe masculinization should be reconstructed in early childhood to allow proper psychosocial development and sexual function, with the timing of vaginoplasty depending on the anatomy (see below). Patients with a vagina entering the urogenital sinus should undergo vaginoplasty prior to menarche to allow free flow of menses and eliminate the risks of genitourinary tract obstruction and infection.

SURGICAL TECHNIQUE
All stages of feminizing genitoplasty are performed in the lithotomy position. The operative repair begins with endoscopic examination of the lower genitourinary tract to supplement the preoperative radiographs and confirm the level at which the vagina enters the urogenital sinus. This information will dictate whether or not the reconstruction will be performed in one or two stages. If the vagina enters the urogenital sinus distal to the external urinary sphincter, a cutback or flap vaginoplasty can usually be combined with clitoroplasty in a single procedure ( Fig. 106-2). However, if the vagina enters the urogenital sinus more proximally, requiring a free skin graft or pedicle skin flap, a vaginal pull-through, or a segmental bowel interposition for reconstruction, vaginoplasty is deferred to a later date.

FIG. 106-2. The anatomy of a female pseudohermaphrodite due to congenital adrenal hyperplasia prior to surgical reconstruction. (A) External view with the patient in lithotomy position. (B) Sagittal view of the corresponding internal anatomy.

Clitoroplasty Regardless of the internal anatomy, the first procedure is phallic reduction and recession, usually performed when the child is 6 to 12 months old and stable from an endocrine standpoint. The technique is designed to shorten the corpora cavernosa, taper the glans, and preserve the neurovascular bundles to maintain sensation. Reduction phalloplasty begins by marking a posteriorly based perineal skin flap in the shape of an inverted “U”, which is sharply incised and elevated, leaving it thick with the subcutaneous fat to ensure an adequate blood supply ( Fig. 106-3A). The flap is curved rather than pointed because a pointed flap may have a tenuous blood supply at its tip. Fine traction sutures and skin hooks are used to minimize tissue handling and pinpoint monopolar electrocautery with a needle tip electrode used to maintain hemostasis. The incision is carried along the ventral aspect of the phallus, sparing the ventral mucosal groove ( Fig. 106-3B). It continues circumferentially, just proximal to the corona (Fig. 106-3C). The skin of the phallus, including the prepuce, if present, is sharply elevated from the shaft along Buck's fascia, from the glans to the base. This skin is then split in the dorsal midline to create Fortunoff flaps, used later in the reconstruction of the labia. The suspensory ligament is divided (Fig. 106-4A). Then the urogenital sinus is sharply opened in the ventral midline from the meatus to what will become the posterior wall of the vagina ( Fig. 106-4B). It is important not to incise the urethra or the external urethral sphincter. The perineal skin flap is sutured to the posterior wall of the vagina to exteriorize the vaginal lumen, serving as the initial step in a flap vaginoplasty ( Fig. 106-4C).

FIG. 106-3. (A) The inverted U-shaped, posteriorly based, perineal skin flap is incised. (B) The subcutaneous dissection is performed to keep the flap thick and maintain a viable blood supply. (C) The skin incision is continued onto the ventral shaft of the phallus and circumferentially to leave a collar of skin attached to the glans. The skin on the phallus is sharply elevated, leaving the urogenital sinus intact.

FIG. 106-4. (A) The suspensory ligament of the phallus is divided sharply with scissors. (B) The urogenital sinus is opened sharply in the ventral midline with electrocautery to minimize bleeding. (C) Three interrupted sutures of 3-0 polyglactin are placed to fasten the Y-V perineal skin flap into the opened urogenital sinus to exteriorize the vagina.

After the phallus is completely exposed, a suture ligature of 2-0 polyglycolic acid is placed in each corporal body near its base, avoiding the neurovascular bundles. This is performed to occlude the cavernosal arteries and minimize bleeding with the resection of corporal tissue ( Fig. 106-5A). Reduction of the phallus is accomplished by making lateral incisions in each corpus cavernosum ( Fig. 106-5B). Use scissors to sharply excise the excess corporal tissue bilaterally ( Fig. 106-5C). Bleeding from the corporotomy incisions should be minimal and is controlled with careful pinpoint electrocautery; these incisions are not closed. If, after reduction of the corpora cavernosa, the glans is disproportionately large, it can easily be reduced by excision of one dorsal or two lateral wedges of tissue ( Fig. 106-6). The glans is then closed in one or two layers of interrupted 6-0 polyglactin suture. The final, but perhaps most critical, component of clitoroplasty is to reattach the glans to the pelvis (Fig. 106-7). The goal is to anchor it in a well-concealed, anatomically correct, and therefore cosmetically appealing, position. After exposing the pubic bone, two stitches of 3-0 silk are passed between the periosteum of the pubis and the tunica albuginea of the distal shaft just behind the glans; one stitch is placed on each side (Fig. 106-7A, B). When clitoroplasty is performed in this fashion, the neurovascular bundles are spared and sensation is preserved as demonstrated by pudendal evoked potentials. 3

FIG. 106-5. (A) The proximal corpora cavernosa are suture ligated with 2-0 polyglactin, avoiding the dorsal neurovascular bundles. (B) Incisions are made with a scalpel in the lateral aspect of the corporal bodies, and the corporal tissue exposed by passing a vessel loop between the incisions, using blunt dissection with the tip of a fine hemostat. (C) Excess corporal tissue is sharply excised with scissors.

FIG. 106-6. If the size of the glans clitoris is disproportionate after resection of the corporal bodies, wedge resection of the glans may improve the cosmetic appearance. This can be performed as a single dorsal wedge or two lateral wedges. Incisions are made sharply with a scalpel and are closed in one or two layers with 6-0 polyglactin suture.

FIG. 106-7. After the clitoris has been reduced to an appropriate size, the glans is secured in a concealed and anatomically correct position. (A) The fat overlying the pubic bone is cleared by sharp and blunt dissection. (B) Two sutures of 3-0 silk are passed through the periosteum of the pubic bone and the tunica albuginea of the distal shaft. (C) Reconstruction of labia is performed by draping the previously divided phallic skin around the sides of the clitoris. The medial edge of these flaps is sutured to the mucosal collar around the glans and to the edges of the dorsal strip of urogenital sinus mucosa and its underlying spongiosum tissue.

Labioplasty is then performed using the Fortunoff skin flaps created from the previously divided dorsal phallic skin. The medial edges of these flaps are sutured to the skin collar surrounding the glans and to the edges of the ventral phallic urethral strip using interrupted 5-0 polyglactin, incorporating the underlying corpus spongiosum in the sutures to reduce bleeding. The lateral aspect of the same dorsal skin flap is sutured to the labioscrotal folds, creating labia minora ( Fig. 106-7C). The labioscrotal flap is then extended into the angle formed by the perineal flap and the labia minora (a labioscrotal Y-V plasty) to construct the labia majora. Some of the labioscrotal skin may have to be excised before the Y-V plasty is completed to prevent this tissue from having a rugated appearance. A 10-Fr Foley catheter is placed in the bladder. The last step in a single-stage feminizing genitoplasty is to suture the edges of the perineal skin flap to the labioscrotal skin folds using interrupted 5-0 polyglactin suture. This maneuver completes the flap vaginoplasty (discussed below), resulting in a nicely feminized external genitalia ( Fig. 106-8).

FIG. 106-8. Appearance after reconstruction. (A) External appearance, with the labia separated to show the suture lines. (B) Sagittal view to show corresponding internal anatomy.

Vaginoplasty Not all cases of feminizing genitoplasty require a vaginoplasty, and not all cases requiring vaginoplasty should have it performed as part of a combined procedure. The timing and the proper type of vaginoplasty performed depends on the level at which the vagina enters the urogenital sinus as judged by the preoperative x-rays and endoscopy. 4,10 Four different types of vaginal reconstruction can be used, depending on the clinical findings: a cutback vaginoplasty, a “flap” vaginoplasty, a “pull-through” vaginoplasty, or a complete vaginal reconstruction using a free skin graft or pedicled skin flap or a segmental bowel interposition. For simple labial fusion, a cutback vaginoplasty is the procedure of choice. This is performed by placing the tips of fine forceps or a metal sound in the introitus behind the fused labia, and making a vertical, full-thickness incision in the midline raphé. Interrupted 5-0 polyglactin suture is used to close the incision transversely, in a Heineke-Mikulicz fashion, to widen the opening to the vagina ( Fig. 106-9). Alternatively, a small-flap vaginoplasty can be performed.

FIG. 106-9. For mild forms of masculinization, with only labial fusion, a cut-back vaginoplasty is performed. (A) A midline vertical incision is made and closed transversely, in a Heineke-Mikulicz fashion, to widen the vaginal introitus. (B) Sagittal view of the corresponding anatomy. (C) Result of reconstruction.

If the vagina joins the urogenital sinus in a low or only moderately high position, a flap vaginoplasty is performed. This is the most common variety of vaginoplasty used; when combined with a clitoroplasty in one procedure, it provides a superb cosmetic result in most cases. The perineal flap is developed as previously described, and a finger is placed in the rectum to guide the dissection and prevent inadvertent injury. The fibrofatty tissue in this area is divided to allow for a widely patent introitus. The urogenital sinus is opened in the ventral midline to expose the posterior wall of the vagina, and marking sutures are placed in the tissue edges as the incision advances cephalad. These marking sutures are important to provide exposure and control for the exact placement of the perineal flap into the opened vagina. It is extremely important that the perineal flap be advanced well into the vagina to create an introitus of adequate size and prevent postoperative stenosis. The perineal flap is advanced into this incision and anchored down with three interrupted sutures of 3-0 polyglactin to exteriorize the vagina ( Fig. 106-4C). Reverdin needles, sometimes referred to as Turner-Warwick needles, may be necessary to place the stitches properly into the posterior wall of the vagina. Intraoperative endoscopy is repeated at this point to ensure that the flap is in good position and that the vaginal orifice is widely patent. Once this is confirmed, the skin is approximated to the vaginal wall with interrupted 5-0 polyglactin sutures. Although some recommend routine dilation of this newly constructed vaginal orifice, in our experience this is usually not necessary. There is some controversy regarding the timing of vaginoplasty for the patient with a vagina that enters the urogenital sinus proximal to the external urethral sphincter. Although some advocate early reconstruction at the time of clitoroplasty, our preference is to delay vaginal reconstruction in this group until the child is at least 3 to 4 years old. By waiting for the child to grow, the pelvic anatomy becomes larger and better defined, making the surgery technically easier and safer. However, if the

Müllerian tract is obstructed or serves as a site for urinary stasis, reconstruction may be required earlier. In some cases, clean intermittent catheterization of the urogenital sinus may be necessary until proper reconstruction of the vagina can be performed. Although flap vaginoplasty works to exteriorize the low vagina, it is not an appropriate technique to exteriorize the high vagina entering the urogenital sinus more proximally. Not only would this technique be technically difficult, but it would require division of the external urinary sphincter and risk urinary incontinence. In addition, such a reconstruction would result in severe female hypospadias, with subsequent urinary stasis and infection. In this situation, the procedure of choice is to use local skin flaps combined with a perineal or abdominoperineal pull-through to exteriorize the vagina. 6,9 Our preference is the former procedure, as popularized by Hendren (Fig. 106-10, Fig. 106-11, Fig. 106-12 and Fig. 106-13). It is very important that when the vagina is divided a small cuff of vaginal tissue be left behind on the urethra. By closing this cuff transversely, the urethra can be closed without creating a stricture ( Fig. 106-11).

FIG. 106-10. Pull-through procedure for high vagina. (A) A posteriorly based, perineal skin flap is raised in an inverted U-shaped configuration, just as for the approach to the low vagina. A Fogarty balloon catheter is placed through the urogenital sinus, with the balloon placed and inflated in the vagina. (B) Sagittal view of the corresponding anatomy.

FIG. 106-11. Pull-through procedure for high vagina, continued. (A) The vagina is approached by dividing fibers of the urogenital diaphragm in the midline. Palpation of the Fogarty catheter balloon assists in identifying the vagina. (B) The posterior wall of the vagina is sharply divided in a transverse direction and traction sutures are placed in the free edge to maintain control.

FIG. 106-12. Pull-through procedure for high vagina, continued. (A) The anterior wall of the vagina is sharply divided, leaving a small cuff of tissue on the urethra (which is closed transversely), and additional traction sutures are placed to establish circumferential control. (B) Lateral perineal skin flaps are incised and raised to be advanced into the wound to meet the edges of the vagina.

FIG. 106-13. Pull-through procedure for high vagina, continued. (A) A small suction drain is placed in the perineal wound and the skin flaps are advanced and circumferentially sutured to the vagina. (B) The vaginal outlet as it appears postoperatively.

In some cases, a pull-through procedure is technically impossible because the vagina is very high and short, or the vagina is congenitally absent. In this case, a segmental bowel interposition works very well to reconstruct the vagina. 4,5,7 One advantage in using bowel for vaginal reconstruction is that it does not require dilation. Specific advantages of using sigmoid colon for this purpose include its redundancy, its proximity to the pelvis, the mobility of its vascular pedicle, and the similarity in caliber with that of the normal vagina. 5 The main disadvantages of this approach are vaginal mucus production, which can develop a foul smell, and the potential for vaginal prolapse. These problems are easily avoided by periodic douching and by anchoring the proximal vaginal cuff to the sacrum at the time of surgery ( Fig. 106-14 and Fig. 106-15).

FIG. 106-14. Sigmoid bowel interposition for vaginal substitution. (A) A segment of sigmoid colon is isolated with branches of the inferior mesenteric artery. (B) The bowel segment is divided and mobilized, and bowel continuity reestablished by performing a colocolostomy behind the isolated segment.

FIG. 106-15. (A) The pelvic peritoneum is incised, and sharp and blunt dissection are performed between the bladder and rectum to allow for placement of the bowel segment in the proper anatomic position. (B) The caudal end of the bowel segment is sutured to skin flaps in the perineum, and the cranial end is fastened to the presacral fascia to prevent prolapse.

The other option for reconstruction of the high, short, or absent vagina is to use skin, either as a free graft or as a vascularized pedicle flap. 9 When necessary, a tissue expander may be helpful to increase the locally available skin. The main disadvantage in using skin in the vagina is that it tends to become dry and contract, resulting in vaginal stenosis. Since this is more common with free skin grafts than with flaps, especially with split-thickness skin grafts, vascularized skin flaps are preferable. Although estrogen creams may ameliorate the problem of vaginal dryness, vaginal dilation is usually required to maintain an adequate vaginal orifice. Therefore, for those cases in which skin must be used for vaginal reconstruction, surgery is deferred until the patient is mature enough to dilate her own neovagina, but before the age of menarche. Again, this delay also allows additional time for development of the pelvis, making the surgery technically easier and safer. After surgery, a cross-shaped perineal pressure dressing is made using nonadherent gauze and an adhesive elastic tape, which is left in place for 48 hours. A Foley catheter is left in the bladder until the dressing is removed. Dilute betadine baths are then given twice a day and a hair dryer is used to dry the perineum atraumatically. If the wound is healing well, the Foley catheter is removed on the third or fourth postoperative day and, if necessary, the suture lines are cleaned daily by the parent with dilute hydrogen peroxide. After 5 days, betadine baths are switched to warm sitz baths and are continued twice a day. Intravenous antibiotics are administered until the Foley catheter is removed and then switched to oral prophylactic antibiotics for a period of 2 weeks.

OUTCOMES
Complications These procedures can be technically challenging and complications may develop despite the most careful approach. Prophylactic antibiotics are used to guard against wound and urinary tract infection, which threaten wound healing and flap viability. If the neurovascular bundles are not well identified during clitoroplasty, inadvertent injury to them may occur, with loss of sensation in the glans clitoris. Imprecise reconstruction of the urethra may result in stricture or diverticulum formation. Stenosis of the vaginal introitus may occur if the perineal Y-V skin flap is not adequately mobilized and secured to the posterior vaginal wall. Inadvertent injury to the rectum, which lies immediately behind the urogenital sinus, may occur during dissection in this area. Adrenal insufficiency may develop in patients with CAH if adequate steroid replacement is not provided. Results Despite these potential complications, reported results with these techniques have been excellent. 1,2,3,8 In addition, in our experience, surgical outcomes have been most satisfactory. The postoperative cosmetic appearance of the genitalia has been superb. The clitoris has normal sensation to touch and pinprick. No patient who has had a flap vaginoplasty performed as part of a combined perineal reconstruction has developed vaginal stenosis or required vaginal dilation. No episodes of glans sloughing, flap necrosis, wound infection, or urinary tract infection have occurred, and in all cases urinary continence has been maintained. With careful attention to details, meticulous surgical technique, and commitment to the principles of surgical reconstruction, consistently gratifying cosmetic and functional results such as these can be expected. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Azziz R, Mulaikal R, Migeon C, Jones HJ, Rock J. Congenital adrenal hyperplasia: long-term results following vaginal reconstruction. Fertil Steril 1986;46:1011. Bailez MM, Gearhart JP, Migeon C, Rock J. Vaginal reconstruction after initial construction of the external genitalia in girls with salt-wasting adrenal hyperplasia. J Urol 1992;148:680. Gearhart JP, Burnett A, Owen JH. Measurement of pudendal-evoked potentials during feminizing genitoplasty: technique and applications. J Urol 1995;153:486. Hendren W, Donahoe P. Correction of congenital abnormalities of the vagina and perineum. J Pediatr Surg 1980;15:751. Hendren WH, Atala A. Use of bowel for vaginal reconstruction. J Urol 1994;152:752. Hendren WH, Crawford JD. Adrenogenital syndrome: the anatomy of the anomaly and its repair. Some new concepts. J Pediatr Surg 1969;4:49. Hensle TW, Dean GE. Vaginal replacement in children. J Urol 1992;148:677. Newman K, Randolph J, Parson S. Functional results in young women having clitoral reconstruction as infants. J Pediatr Surg 1992;27:180. Passerini-Glazel G. A new one-stage procedure for clitorovaginoplasty in severely masculinized female pseudohermaphrodites. J Urol 1989;142:565. Powell D, Newman K, Randolph, J. A proposed classification of vaginal anomalies and their surgical correction. J Pediatr Surg 1995;30:271.

Chapter 107 Pediatric Vesical Diversion Glenn’s Urologic Surgery

Chapter 107 Pediatric Vesical Diversion
Steven J. Skoog

S. J. Skoog: Division of Urology, The Oregon Health Sciences University, Portland, Oregon 97201.

Indications for Surgery Alternative Therapy Surgical Technique Intubated Cystostomy Percutaneous Cystostomy Open Suprapubic Cystostomy Vesicostomy Blocksom Vesicostomy Lapides Vesicostomy Outcomes Complications Results Chapter References

Urinary diversion at the vesical level is particularly applicable to the pediatric patient. The abdominal location of the bladder facilitates the surgical procedure and the free drainage of urine into a diaper avoids the problems associated with urinary appliances. The bladder can be drained via tube (cystostomy) or by creation of a vesicocutaneous fistula (vesicostomy). Intubated cystostomy in the pediatric patient is always a temporary procedure. It is utilized in reconstructive surgery such as epispadias closure and complex hypospadias repair. The infant or child with urethral trauma or disruption is also a candidate for temporary intubated cystostomy. Long-term use of catheters in the bladders of children is associated with contracture, stones, infection, acquired vesicoureteral reflux, and renal damage. 3 Vesicostomy provides urinary diversion with unobstructed flow at the bladder level without the presence of a foreign body. Vesical diversion is usually performed as a temporary vesicocutaneous fistula that is closed at a later date when clean intermittent catheterization can be performed to empty the bladder. Continent vesical diversion can be performed by application of the Mitrofanoff principal. 10 This is a permanent form of diversion and is covered in Chapter 80 of this text.

INDICATIONS FOR SURGERY
The goal in the performance of vesical diversion is to provide unobstructed free drainage of urine. Decompression of the bladder effectively empties the upper urinary tract with relief of hydronephrosis. 8,9 Improved drainage reduces the incidence of pyelonephritis and stabilizes or improves renal function. 3 The temporary nature of this procedure implies that subsequent bladder function is not compromised and closure is facilitated. Recent studies of bladder capacity and emptying after temporary cutaneous diversion demonstrate minimal effect of diversion on these parameters. 5 Infants and children with neurogenic bladders, cloacal anomalies, patients with posterior urethral valves or prune belly syndrome, and, rarely, infants with severe vesicoureteral reflux and poor renal function are candidates for cutaneous vesicostomy.6 The largest experience with vesicostomy is in infants and children with myelomeningocele. 3

ALTERNATIVE THERAPY
There is no alternative therapy.

SURGICAL TECHNIQUE
Intubated Cystostomy The bladder can be effectively drained by insertion of a catheter into it by either a percutaneous or an open procedure. The open procedure is rarely indicated but its surgical approach is applicable to all vesical diversions. Prior surgery in the pelvis and a small contracted, thick-walled bladder are occurrences that may require an independent open approach. Percutaneous Cystostomy Percutaneous cystostomy provides the urologist with immediate access to the bladder. It is particularly useful in the infant or child with outlet obstruction (i.e., cloacal anomaly), urethral trauma, and as an adjunct to reconstructive surgery. A number of kits are available that rely on the Seldinger technique or a trocar for insertion. Insertion of the catheter requires a distended bladder. This should be accessed by palpation or ultrasound guidance. If the urethra is accessible, the bladder can be distended with saline after insertion of a Foley catheter. If the patient has experienced previous abdominal or pelvic surgery, then ultrasound is invaluable for localizing a safe site of entry and monitoring the location of the guidewire and fascial dilators. The infant or child is assessed, and conscious sedation is of value in the anxious, fearful child. The skin is prepped and 1% lidocaine is infiltrated down to the fascia 2 to 3 cm above the pubic symphysis. A 22-gauge spinal needle attached to a 10-cm 3 syringe is inserted perpendicular to the skin and gently advanced while aspiration takes place. Urine is obtained and a clamp is placed on the needle to mark the depth of the bladder. The needle is removed and a small transverse incision (5 to 10 mm) is made over the needle mark. It is extended down to the fascia, which is nicked in the midline with a no. 11 scalpel blade. The trocar is positioned inside the catheter. The assistant positions his or her hand above the bladder and stabilizes its position prior to insertion. The trocar is grasped with the left thumb and index finger at the level previously determined by the spinal needle. The right thumb and index finger are positioned near the connecting hub and the trocar is pushed with constant pressure into the bladder. Entry into the bladder is signaled by the presence of urine from the top of the trocar. The catheter is advanced 1 to 2 cm and the trocar removed. It is fixed to the skin with a 3-0 nylon suture ( Fig. 107-1).

FIG. 107-1. Percutaneous cystostomy. (A) A small 5- to 10-mm transverse incision is made over the previously made needle mark utilized to judge the depth of the bladder. It is extended down to the fascia. (B) The percutaneous cystostomy tube with obturator needle is carefully advanced into the bladder. (C) The suprapubic catheter is now sewn into place to the skin of the abdominal wall.

If the Seldinger technique is utilized, a guidewire is placed into the bladder via an 18-gauge needle. The needle is removed and serial fascial dilators are employed to dilate an appropriately sized tract over the guidewire into the bladder. A pigtailed drainage catheter is passed over the guidewire into the bladder and the guidewire is removed. The catheter can be secured to the skin with a suture or the disk supplied with the kit ( Fig. 107-2). An antibiotic ointment is applied to the entry site and the catheter is taped and connected to an appropriate collection device.

FIG. 107-2. A percutaneous pigtail suprapubic catheter set. Trocar, needle, guidewire, fascial dilators, and pigtail catheter with connecting tubing are essential components of percutaneous suprapubic access to the bladder.

Open Suprapubic Cystostomy Although this procedure is rarely performed inde-pendently, the principles are applicable to all open approaches to the bladder. The patient is positioned supine and the skin is prepped over the lower abdomen. A 3- to 5-cm transverse incision, usually in a skinfold, is made cephalad to the pubic symphysis. Electrocautery is used to incise the subcutaneous fat and rectus abdominous fascia. The fascia is mobilized off the underlying rectus muscle in a cephalad and caudal direction. The rectus muscles and transversalis fascia are separated in the midline and the bladder is visualized. The peritoneal reflection is mobilized superiorly proximal to the urachus. Allis clamps or 3-0 chromic sutures are placed lateral to the midline incision into the bladder. A 1-cm incision with electrocautery is made into the bladder. The distal end of the catheter is placed into the bladder. The catheter is secured to the bladder with the previously placed 3-0 chromic suture. An appropriate exit site for the cystostomy tube is selected. A 1-cm incision is made at that site and a Kelly clamp is passed through the skin and fascia. The distal end of the catheter is grasped with the Kelly clamp and pulled through to the skin. The rectus fascia is closed with a running 2-0 polydioxanone (PDS) suture. Scarpa's fascia is closed with 4-0 chromic and the skin is closed with a subcuticular 4-0 polyglactin (Vicryl) suture. The catheter is secured to the skin with a 3-0 nylon suture and taped with a gentle curve to the abdominal wall ( Fig. 107-3).

FIG. 107-3. Suprapubic cystostomy. (A) A small transverse incision 1 to 3 cm above the pubic symphysis is made into the skin. (B) The rectus fascia is incised along the length of the incision. (C) The rectus fascia is elevated with electrocautery. (D) The rectus muscles are bluntly separated in the midline. (E) The peritoneum is mobilized superiorly. (F) The bladder is secured with Allis clamps and a cystotomy incision is made with electrocautery. (G) Insertion of the Malecot catheter. (H) Placement of lateral closing sutures. (I) Watertight closure of cystotomy. (J) Placement of distal end of catheter through exit site. (K) Suprapubic tube sewn in place and secured to abdominal wall. (Modified from Smith MS. Cystostomy and vesicostomy. In: Glenn JM, ed. Urologic surgery. 4th ed. Philadelphia: JB Lippincott, 1991;1042–1049.)

Vesicostomy Two procedures have been utilized for pediatric vesical diversion: the Blocksom vesicostomy and the Lapides vesicostomy. The Blocksom vesicostomy is particularly applicable to infants and children as the stoma is located in a subumbilical position, which drains better into the diaper. 1 This technique avoids the use of skin flaps and mobilization of bladder flaps necessary with the Lapides vesicostomy. Subsequent takedown of the vesicostomy is technically easier with the Blocksom procedure. The Lapides vesicostomy was primarily performed in adults and little clinical experience with this procedure is available in pediatric patients. 7 Blocksom Vesicostomy A 2- to 3-cm transverse abdominal incision is made midway between the pubis and the umbilicus. The rectus fascia is opened transversely for 2 cm and 0.5 cm superiorly and inferiorly in the midline, creating a cruciate incision. No fascia is excised. If the bladder is not full, it is filled with saline via a urethral catheter. The fascia, skin, and rectus are retracted with small army/navy retractors and a 3-0 chromic holding suture is placed at the dome of the bladder. The peritoneum is bluntly swept superiorly until the urachus is visualized. The urachus is transected with electrocautery. It should easily protrude up into the incision with adequate mobilization of the dome of the bladder. Stay sutures of 3-0 chromic are placed lateral to the urachus. The urachus and a button of detrusor are excised with electrocautery. The size of the bladder defect depends on the patient's diagnosis. Thick-walled bladders due to posterior urethral valves or meningomyelocele are unlikely to prolapse and should calibrate to 24 Fr (using a bougie à boule). Prune belly syndrome requires a 28- to 30-Fr stoma as stenosis is more common, and patients with primary vesicoureteral reflux with normal bladders require a smaller stoma (18- to 20-Fr) to prevent prolapse. 3 The adventitia and detrusor muscle are secured to the tips of the flaps of the rectus fascia with 3-0 polyglactin (Vicryl). The stoma should rest without tension at skin level. The fascial defect is narrowed laterally with interrupted 3-0 polyglactin suture to an appropriate size as measured with bougies à boule. The bladder mucosa is everted by taking a larger bite of it with a small bite of submucosa and detrusor and fixing it to the skin edge with interrupted 4-0 polyglactin (Vicryl) sutures. The fat and Scarpa's fascia are approximated laterally to make an appropriately sized stoma. The skin is approximated laterally with interrupted sutures. The stoma is left intubated with a 12-Fr Foley catheter drained into a dual diaper system for 48 hours. Vaseline is applied around the stoma 3 times per day until the catheter is removed. The stoma is calibrated postoperatively and dilated as necessary with a soft catheter ( Fig. 107-4).

FIG. 107-4. Blocksom vesicostomy. (A) A 2- to 3-cm transverse incision midway between pubis and umbilicus. (B) Outline of cruciate incision in rectus fascia. (C) Stay suture in dome of bladder. (D) After mobilization of peritoneum, urachus is incised with electrocautery and mobilized into the incision. (E) The urachus is excised and an appropriate size opening made with electrocautery into the bladder. (F) The tips of the rectus fascial flaps from the cruciate incision are sewn to the detrusor an appropriate distance proximally to allow the creation of a stoma that is flush with the skin. (G) The rectus fascia is narrowed laterally with interrupted suture to an appropriate size as measured with bougies à boule. (H) The bladder epithelium is tacked to the skin and the lateral skin and subcutaneous tissue are approximated. (Modified from Belman AB, King LR. Vesicostomy: useful means of reversible urinary diversion in selected infant. Urology 1973;1:208–213.)

Lapides Vesicostomy The Lapides vesicostomy utilizes a cutaneous flap sewn to an anterior bladder flap to make a tube that is half skin and half bladder to prevent stomal stenosis. 7 The skin flap ideally is hairless to avoid stone formation and urinary infections. This procedure can be very difficult in obese patients and those with small, thick-walled bladders. Typically, it has been used in adults and a stoma appliance is applied. 7 The bladder is filled and a Pfannenstiel incision is made 2 to 3 cm above the pubis. The rectus fascia is divided transversely and mobilized off the rectus muscle superiorly to the umbilicus. The rectus muscle and posterior rectus fascia are divided in the midline to the level of the umbilicus. A rectangular skin flap measuring 3.5 cm at its base and 3.0 to 4.5 cm in length is outlined on the abdominal wall. The length of the skin flap is determined by the distance to the bladder, which is a function of abdominal wall thickness, bladder size, and bladder mobility. Prior to incising the skin flap, traction on the rectus fascia via the Pfannenstiel incision with Allis or Kocker clamps will ensure that the skin flap is appropriately aligned with the area excised from the rectus fascia. Care is taken to maintain a full thickness of skin and subcutaneous tissue as the flap is incised to ensure good blood supply. A generous portion of anterior rectus sheath is removed immediately underneath the skin flap. The peritoneum is reflected superiorly off the dome of the bladder. A bladder flap with its base located superiorly is outlined with a marking pen and 3-0 chromic stay sutures are positioned at the corners. Its base on a distended bladder measures 4 cm. A full- thickness bladder flap is incised and mobilized utilizing electrocautery. The stay sutures on the free end of the bladder flap are passed through the opening created by the skin flap. The bladder flap is sewn to the skin edge with 3-0 polyglactin (Vicryl) suture. The skin flap is secured with a Babcock clamp and passed beneath the skin bridge. The skin flap is brought down to the bladder and sewn to the bladder flap and bladder defect with interrupted 3-0 polyglactin suture. This is begun underneath the abdominal wall adjacent to the base of the skin flap. Sutures are alternated from side to side to facilitate closure of the tube. The remaining defect in the bladder is closed in the midline with interrupted sutures. The caliber of the tubed vesicostomy is tested with bougies à boule or index finger. If any abrupt edges are encountered, relaxing incisions into the rectus fascia are made. The perivesical space is drained with a Penrose drain brought out through a separate stab incision. The Pfannenstiel incision is closed in successive layers with interrupted absorbable suture. The vesicostomy is drained with a Foley catheter for 48 to 72 hours. A stomal appliance can be fitted when required ( Fig. 107-5).

FIG. 107-5. Lapides vesicostomy. (A) Outline of initial incisions and flap measurements. (B) The bladder flap is sewn in place as the abdominal skin flap is retracted. (C) Skin flap is secured to the bladder flap. (D) Closure of the bladder defect is completed. (E) Pfannenstiel's incision closed. (Modified from Parker CB. Alternative to urinary diversion: vesicostomy. AORN J 1984;39:968–972.)

OUTCOMES
Complications The overall complication rate for surgical vesicostomy is 15% to 17%. Major complications include stomal stenosis, bladder prolapse, and urinary infection requiring hospitalization. Failure to adequately drain the upper urinary tract is seen in only 1% to 8% of patients. Minor complications include asymptomatic bacteriuria, stone formation, and diaper dermatitis (Table 107-1).

TABLE 107-1. Complications of vesicostomy in children (Blocksom Technique)

TABLE 107-2. Clinical outcomes in children with Blocksom vesicostomy

Bladder prolapse is the most frequent and severe complication. It is particularly seen in patients with vesicostomy for high-grade vesicoureteral reflux. 3 It is usually due to a technical error with placement of the stoma below the urachus on the anterior bladder wall and a stomal opening that is too large. Placement of the vesicostomy cephalad to the urachus immobilizes the peritonealized portion of the bladder dome preventing prolapse. 2 Correction involves emergent return to the operating room if the prolapse cannot be manually reduced. The stoma is moved to a better position on the dome of the bladder and the stoma appropriately sized at 20 to 24 Fr. A prolonged period of intubation of the stoma is required until the edema subsides. Stomal stenosis is manifested by symptomatic infections, voiding from the urethra, and upper urinary tract dilation. The bladder may be palpable. Stenosis is due to a small fascial opening or excessive tension on the vesicocutaneous anastomosis. It has been reported to be more common in patients with prune belly syndrome. 3 Urodynamic evaluation to measure the pressure at which overflow through the vesicostomy occurs can be helpful in equivocal cases. 4 This complication is avoided by ensuring adequate stomal size and mobilizing the bladder dome for a tensionless vesicocutaneous closure. Bladder stones are very uncommon in the pediatric population. Prevention of infection, and urinary stasis and urinary acidification will protect against stone development. The stones are easily removed endoscopically via the vesicostomy. Peristomal dermatitis is a common problem following vesicostomy. Air drying of the skin around the stoma with the cool setting of a hair dryer is frequently all that is required. Topical treatment with Desitin or bag balm twice per day provides a protective barrier to irritated areas. Secondary fungal dermatitis responds to Mycostatin powder followed by United Skin Prep Spray, which holds the powder in place. Results Cutaneous vesicostomy is uniformly successful in decompressing the upper urinary tract, and stabilizing or improving renal function. Postvesicostomy hydronephrosis as assessed by intravenous pyelogram or renal sonography will show decreased dilation in 92% to 99% of patients following vesicostomy. 2,3,6,8 Resolution of vesicoureteral reflux is seen in 20% to 25% of patients. 3 A fourfold decrease in hospitalization for pyelonephritis after vesicostomy has been noted. 3 Persistent upper tract dilation in the absence of stomal stenosis implies that the obstruction is above the bladder level and supravesical diversion is required. dilated ureters may remain functionally obstructed, even with a vesicostomy, and another choice of diversion or reconstruction is necessary.
4

Marked

Bladder function is maintained in more than 75% of children after a period of defunctionalization due to vesicostomy. 5 The need for bladder augmentation, due to poor detrusor compliance, is related to the effect of the primary pathologic condition rather than the diversion itself. Bladder augmentation is most often required in children with myelodysplasia at the time of bladder closure or subsequently. 5 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Blocksom BH Jr. Bladder pouch for prolonged tubeless cystostomy. J Urol 1957;78:398. Bruce RR, Gonzales ET Jr. Cutaneous vesicostomy: a useful form of temporary diversion in children. J Urol 1980;123:927. Duckett JW, Zivlan O. Use and abuses of vesicostomy. AUA Update Series 1995;16:130. Hurwitz RS, Ehrlich RM. Complications of cutaneous vesicostomy in children. Urol Clin North Am 1983;10(3):503. Jayanthi VR, McLorie GA, Khoury AE, Churchill BM. The effect of temporary cutaneous diversion on ultimate bladder function. J Urol 1995;154:889. Krahn CG, Johnson HW. Cutaneous vesicostomy in the young child: indications and results. Urology 1993;41:558. Lapides J, Ajemian EP, Lichtwardt JR. Cutaneous vesicostomy. J Urol 1960;84:609. Noe HN, Jerkins GR. Cutaneous vesicostomy experience in infants and children. J Urol 1985;134:301. Smith MS. Cystostomy and vesicostomy. In: Glenn JR, ed. Urologic surgery. 4th ed. Philadelphia: JB Lippincott, 1991;1042–1049. Sumfest JM, Burns MW, Mitchell ME. The Mitrofanoff principle in urinary reconstruction. J Urol 1993;150:1875.

Chapter 108 Urinary Undiversion Glenn’s Urologic Surgery

Chapter 108 Urinary Undiversion
Andrew L. Freedman and Ricardo Gonzalez

A. L. Freedman: Pan-Pacific Pediatric Urologic Institute, Los Angeles, California 90095. R. Gonzalez: Department of Pediatric Urology, Children's Hospital of Michigan, Wayne State University School of Medicine, Detroit, Michigan 48201.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Evaluation Preparation Goals of Undiversion Techniques for the Upper Tract Techniques for Reconstruction of the Reservoir Procedures to Enhance Bladder Outlet Resistance Procedures for Urethral Substitution Postoperative Care Outcomes Complications Results Chapter References

The past two decades have witnessed a tremendous evolution in the use of urinary diversion for the management of children with congenital urinary anomalies. With the acceptance of clean intermittent catheterization and the development of innovative reconstructive techniques originally designed to reverse the wet ostomy, a permanent incontinent diversion is rarely necessary today. With a dwindling number of children remaining with wet ostomies, the techniques for undiversion have evolved from heroic salvage procedures into planned urinary reconstructions that obviate the need for permanent diversion. Along with this changing role of undiversion techniques has come a change in the philosophy of their application. Although still highly individualized, the goals of reconstruction can be clearly structured and the appropriate component techniques necessary to achieve those goals systematically selected. This approach can be applied equally to the reversal of a permanent or temporary diversion, as well as incorporated into the initial reconstruction. In this chapter, the indications and goals of undiversion and, by extension, urinary reconstruction, along with the most common techniques to achieve those goals are discussed. No text can anticipate all of the variations that may be confronted in clinical practice, but the individualized application of the appropriate reconstructive components can enable the successful reconstruction of almost any patient.

DIAGNOSIS
The diagnosis in this condition is the presence of any form of wet ostomy urinary diversion.

INDICATIONS FOR SURGERY
Current indications for undiversion in the diverted patient include (a) the development of upper tract deterioration related to the nature of the diversion, (b) correction of the underlying condition such that temporary diversion is no longer necessary, (c) intolerance for the incontinent urinary stoma and a desire to improve quality of life, independence, and body image, and (d) refunctionalization of the urinary reservoir in anticipation of renal transplantation.

ALTERNATIVE THERAPY
The alternative to undiversion is to maintain the current form of urinary diversion.

SURGICAL TECHNIQUE
Preoperative evaluation Preoperative evaluation includes not only the physical state of the patient but the psychological preparedness of the patient or the family in case of children. There must be an understanding that in reconstruction the need for additional procedures is not uncommon. Furthermore, reconstruction may potentially place the patient at an increased health risk if not carefully monitored and compliance with catheterization closely maintained. Patients and their families must understand that conscientious lifelong follow-up is critical to success. Medical evaluation begins with a careful review of the patient's history and medical records. A thorough understanding of the reasons for initial diversion as well as the procedures undertaken is critical. Original operative reports and x-rays can be vital to assessing what tissues are available for reconstruction as well as the most likely deficiencies. An understanding of the issues that lead to diversion may also add insight into expected bladder compliance, voiding and sphincter function, and likelihood of continence. Often, complete physiologic assessment will not be possible while diverted and treatment decisions will have to be made based on prediversion studies. Physiologic studies include assessment of each component of the urinary tract. Renal functional assessment is critical. Glomerular filtration rate and tubular function, especially concentrating ability, must be carefully evaluated. The relative contribution of each kidney to overall function may be assessed by nuclear scanning. Assessment of acidosis is of particular importance when bowel interposition is anticipated, as the use of stomach may be beneficial. Ureteral length, diameter, and tortuosity can be assessed radiographically. The presence of proximal obstruction can be evaluated by diuretic renography or antegrade pyelography. On occasion, retrograde pyelography may be necessary to assess the residual distal ureter. Screening for reflux is necessary. Bladder function is often difficult to assess due to long-term defunctionalization and its effects on capacity and compliance. Bladder neck function can be assessed by upright cystography and cystometrogram with needle electromyography to evaluate sphincter innervation. A competent bladder neck or an innervated sphincter will provide continence between catheterization in 80% of patients provided adequate bladder compliance. 5 When both the sphincter is deinnervated and the bladder neck is incompetent, an enhancement of bladder outlet resistance is required. In patients with a cutaneous vesicostomy, assessment of continence and bladder function can be aided by the placement of a temporary gastrostomy button into the vesicostomy. 3 This not only allows the measurement of bladder capacity and compliance but provides an easy access for catheterization. Bladder cycling has not been a productive procedure in our hands. Urethral evaluation is critical as many of these patients will be dependent on intermittent catheterization for bladder emptying. All patients should undergo a trial of intermittent catheterization prior to reconstruction. Urethral conditions that impair catheterization should be carefully elicited, such as strictures, false passages, intravaginal meatus, discomfort, or anatomic barriers that prevent patient independence with catheterization (such as obesity). The use of alternate conduits for catheterization (i.e., appendicovesicostomy) can often alleviate these difficulties. If an intestinal segment is to be used for bladder or ureteral reconstruction, its selection is often dictated by anatomic circumstances and the surgeon's preference.

Preoperative bowel imaging is rarely necessary in children. A detailed history of bowel habits and continence is important. In the neuropathic patient or child with a history of imperforate anus, fecal continence may be predicated on mild constipation. In these patients, loss of the ileocecal valve or excessive colonic length may lead to accelerated intestinal transit, diarrhea, and fecal incontinence. Consideration for bowel conservation by use of ureteral or gastric cystoplasty should be given in those patients with short gut (cloacal exstrophy, history of necrotizing enterocolitis). Lastly, it must be kept in mind that in children with a history of abdominal surgery the appendix has often been sacrificed. The evaluation of the patient with myelomeningocele should also include the input of physical and occupational therapists in order to assess what means of catheterization and use of assistive devices will serve to make the child as potentially independent as possible. Preparation Preoperative preparation for undiversion involves special requirements beyond those typical for major abdominal surgery. The urine must be sterilized with appropriate antibiotics in advance of surgery. If an intestinal segment is to be used, the bowel should undergo complete mechanical and antibiotic cleansing. Golytely in combination with erythromycin and neomycin base has been used predominately. In the child with myelomeningocele, the initiation of the bowel prep at home, including a clear liquid diet and enemas, greatly facilitates the in-hospital cleansing. Vitamin K supplementation should be given if the intestine is prepared to prevent coagulopathy, which may become clinically significant if early reoperation is necessary. Goals of Undiversion The goal of any urinary reconstruction is to create a unidirectional flow of urine into a low-pressure, continent reservoir of adequate capacity that can be emptied to completion either by volitional voiding or intermittent catheterization. Using this goal the urinary tract can be divided into component parts that are each responsible for achieving a separate goal. The upper tract is responsible for active urine transport and protection from reflux. The reservoir is required to be of adequate capacity and sufficient compliance. The outlet is responsible for continence and ease of voiding or catheterization. Although all parts must work together to achieve the desired results, this division provides a framework for assessing the reconstructive needs and available options for the individual child. Techniques for the Upper Tract Ureteroureterostomy Ureteroureterostomy is often the simplest and safest undiversion technique for patients with supravesical diversion and sufficient length of the distal ureters to allow a tension-free anastomosis. It can be performed to reverse a cutaneous ureterostomy or conduit diversion. The condition of the distal ureter, bladder, and bladder outlet must be assessed. If reflux is present it may correct itself following refunctionalization. If not, reimplantation may be performed at a later date. If the midureter has not been manipulated, on occasion a proximal anastomosis and distal reimplantation can be performed simultaneously. If only one ureter can be reanastomosed, then a transureteroureterostomy (TUU) can be combined in order to complete the reconstruction ( Fig. 108-1). If additional length is required to prevent tension on the anastomosis, the kidney can be mobilized caudally, although care must be taken to avoid injury to the blood supply to the upper ureter.

FIG. 108-1. (A) Technique for end-to-end ureteroureterostomy when the distal ureter is adequate and not expected to reflux. (B) Pyeloureterostomy and simultaneous ureteroneocystostomy can be used provided the midureter has never been operated on and is presumed to have an intact blood supply.

The ureteral ends are carefully mobilized and widely spatulated. The anastomosis is performed with 7-0 polydioxanone (PDS) in a running full-thickness suture. The suture is begun in the middle of the spatulated segment farthest from the surgeon to avoid the placement of a knot in the corner or the need to turn the anastomosis. The use of double-armed sutures greatly facilitates this. The closure should be watertight and a stent is left if the distal ureter has been manipulated, a 5- or 8-Fr infant feeding tube left for 7 to 10 days works quite well. A stentogram is often performed prior to stent removal. A Penrose drain is left in the vicinity of the anastomosis. Ureteroneocystostomy/Psoas Hitch If the proximal ureter is long enough to reach below the level of the iliac vessels and the bladder is sufficiently pliable to stretch above the iliac vessels, but the distal ureteral stump is inadequate for direct anastomosis or has massive reflux, implantation of the ureter into a long tunnel with fixation of the bladder to the psoas muscle is a safe and reliable way to achieve undiversion. If both ureters are present, the longest or less dilated ureter is reimplanted and a TUU performed proximal to the psoas hitch. After the diversion is taken down, the ureters are mobilized proximally; great care is taken to preserve the blood supply. If the ureter previously has been dissected extensively, inclusion of the gonadal vessels along with the mobilized ureter minimizes the risk of ischemia. The bladder seldom requires extensive mobilization or ligation of the contralateral vessels. Lysis of adhesions to adjacent structures caused by previous procedures is usually all that is required. The bladder is incised in midline of the anterior wall. Two fingers are introduced toward the dome and the bladder is stretched in a cephalad and lateral direction to ensure that it will reach the psoas muscle at a point above the iliac vessels ( Fig. 108-2). The cystotomy is extended to the tip of the stretching fingers. The psoas muscle is exposed by blunt dissection and the iliac vessels, the psoas tendon, and the genitofemoral nerve are identified. Three stitches of 2-0 PDS are placed around the psoas tendon, taking generous bites. However, care should be taken to not place the sutures so deep as to potentially damage the underlying femoral nerve. 9 The sutures are tied loosely to avoid muscle necrosis, and the needles are left attached. A submucosal tunnel is created at the apex of the cystotomy incision and carried downward such that the tunnel is 5 to 7 times the width of the ureter when flattened against a ruler. A wide tunnel free of angulations will not lead to obstruction despite a long length.

FIG. 108-2. Technique for psoas hitch and long-tunnel ureteral reimplantation.

The sutures previously placed in the psoas are then passed deeply through the bladder wall, though not through the urothelium, at a point lateral to the proximal end of the ureteral tunnel. The ureter is passed through the tunnel and the psoas fixation sutures are tied snugly. The distal end of the ureter is sutured to the bladder mucosa with interrupted 6-0 or 7-0 PDS sutures. The bladder mucosa is closed around the proximal ureteral hiatus with the mucosal stitch incorporating the ureteral adventitia. The muscularis closure stitch also incorporates the ureteral adventitia. These sutures act to prevent shortening of the ureteral tunnel. If the bladder is of adequate size and compliance, it may be closed in the midline in three layers. Alternatively, an enterocystoplasty can be performed by the cup-patch technique with anastomosis to the free cystotomy edges (Fig. 108-3). A TUU can be constructed between the contralateral ureter and the reimplanted ureter in an end-to-side fashion. The ureter or ureters are left intubated with a 5- or 8-Fr infant feeding tube. If a TUU is performed both ureters should be stented and the area of the anastomosis drained.

FIG. 108-3. Following long-tunnel ureteral reimplantation and psoas hitch (A) the bladder can be closed directly (B) or with the aid of a sigmoid or ileal cup patch enterocystoplasty (C).

Transureteroureterostomy Transureteroureterostomy is an extremely helpful and frequently used adjunct to major urinary tract reconstruction. It often greatly simplifies the establishment of a bilateral antireflux state by enabling the reimplantation of only one ureter or avoiding the need for reimplantation. The TUU is performed by mobilization of the donor ureter with great attention paid to preserving the blood supply. The donor ureter is transferred below the mesentery in a gentle sweeping arch. The course may be above or below the inferior mesenteric artery, but where the ureter crosses under the sigmoid mesocolon, angulation must be avoided. TUU can be safely performed between ureters of differing caliber, generally with the less dilated being reimplanted into the intestine or bladder and the larger ureter being the donor. The recipient ureter should not be widely mobilized and the surrounding tissues disturbed as little as possible. The anastomosis must be free of tension and watertight. The recipient ureter is incised on its medial aspect, never on the anterior surface as this will lead to twisting and potential kinking of the ureter. The donor ureter is incised on a bias in order to increase the circumference of the anastomosis ( Fig. 108-4). The anastomosis is performed with a continuous suture of 7-0 or 6-0 PDS, depending on the size of the ureter. The anastomosis is begun in the middle of the posterior wall and carried around to be tied on the anterior wall in order to prevent the placement of the knots in the corners. The anastomosis when completed should lie without tension and without the recipient ureter being deviated from its normal course. Although intubation is not usually necessary with this type of anastomosis, in undiversion cases there is commonly a distal anastomosis in the recipient ureter, and intubation with two polyethylene feeding tubes that exit through the common ureter is essential to ensure good urinary drainage. The use of a single stent is discouraged. When possible, the site of anastomosis should be drained with a Penrose drain in the region of the anastomosis.

FIG. 108-4. Technique for transureteroureterostomy. (From Gonzalez R, Sidi AA. The use of the bowel in the reconstruction of the urinary tract. In: Ehrlich RM, ed. Modern technics in surgery. Mt. Kisco, NY: Futura, 1988.)

Ileal Ureter On rare occasions where there is insufficient ureter available for a TUU, replacement of the ureter with bowel can be considered. When not performed in conjunction with bladder augmentation, isolated ileum may be used as a bridge between the renal pelvis or proximal ureter and the bladder. The ileum is brought retroperitoneally through a window in the mesocolon. If anastomosed to the renal pelvis the pelvis should be widely spatulated and the anastomosis created end to end with absorbable suture (Fig. 108-5). If anastomosed to the proximal ureter, the proximal ileal end is closed and an end-to-side anastomosis of the ureter is constructed as in an ileal conduit. Some authors recommend that the anastomosis be performed in a refluxing manner as the peristaltic properties of the bowel may ameliorate the effects of reflux, but we think that in children on intermittent catheterization this is unwise. The ileum can be reimplanted into the bladder in a nonrefluxing manner by the Le Duc technique or by creation of an ileal nipple. The use of ileal tapering and reimplantation in a long tunnel is rarely successful and should be avoided. Every attempt should be made to make the segment isoperistaltic and as short and straight as possible to improve drainage and limit dwell time. The ileal segment should be left intubated for 4 weeks to prevent kinking and improve drainage.

FIG. 108-5. Technique for ileal ureteral replacement with a refluxing anastomosis. The segment is retroperitonealized through a window in the mesocolon.

Other Ureteral Replacements When an augmentation is performed and the ureters are short but not excessively dilated, the use of an S-shaped sigmoid segment allows for a direct tunneled implantation of one ureter into the bowel. A 40-cm segment of descending and sigmoid colon allows the intestinal segment to reach the renal pelvis and at the same tine provides a nontubular reservoir with excellent compliance. The isolated intestinal segment is placed lateral to the reanastomosed colon to allow the tubular tail to reach the kidney along the left paracolic gutter. This allows the left ureter to be implanted into the tubular portion of the augmentation using the entire length of the ureter, if necessary, to obtain a satisfactory tunnel. The contralateral ureter can be brought across as high as the renal pelvis. The augmentation then proceeds as for a cup patch as discussed below ( Fig. 108-6).

FIG. 108-6. Technique for an S-shaped sigmoid cystoplasty. This technique is useful when the ureters are short of dilated and require a long tunnel. This technique combines the advantages of a nontubular reservoir with a tubular extension that can reach the renal pelvis on either side. (Modified from Gonzalez R, Sidi AA. The use of the bowel in the reconstruction of the urinary tract. In: Ehrlich RM, ed. Modern technics in surgery. Mt. Kisco, NY: Futura, 1988.)

Alternatively, the cecum may be used as the enlarging patch with the terminal ileum extended cephalad to replace the ureter. Unfortunately, this technique requires the use of a reinforced ileocecal valve in order to provide an antireflux mechanism. The removal of the ileocecal valve from the fecal stream may result in intractable diarrhea and loss of fecal continence in some children, particularly those with neurogenic bowel or a reconstructed anus. The cecum, following harvesting, is opened along the antimesenteric border and further detubularized by the addition of an ileal patch. The ileocecal valve is reinforced by intussuscepting it after detaching the mesentery of the last 8 cm of the terminal ileum. The intussuscepted ileal nipple is fixated with nonabsorbable sutures on the serosal side. The nipple can be stabilized with a row of absorbable staples. A rectangle of ileal and cecal mucosa is removed, and the nipple is sutured to the cecal wall. The inverted, detubularized ileocecal segment is anastomosed to the bladder ( Fig. 108-7). The ileal segment to which the ureters will be anastomosed should be long enough to permit a tension-free, end-to-end anastomosis. Excessive length should be avoided to minimize mucus production above the valve, which can lead to pyelonephritis and stone formation. The reinforced ileocecal valve is intubated with a 14- or 16-Fr catheter. A separate cystostomy tube is also used. Although this procedure can salvage a kidney with a severely compromised ureter, it should be undertaken only as a last resort due to the troubling complications that may occur with loss of the ileocecal valve. We use nipples only rarely now because in general we have been disappointed with the long-term results because of persistent reflux or obstruction.

FIG. 108-7. Technique for the creation of an ileocecal bladder augmentation with a reinforced, intussuscepted ileocecal valve as an antireflux mechanism. As this technique sacrifices the ileocecal valve it should be reserved as a last resort. (Modified from Gonzalez R, Sidi AA. The use of the bowel in the reconstruction of the urinary tract. In: Ehrlich RM, ed. Modern technics in surgery. Mt. Kisco, NY: Futura, 1988.)

Ureteroenterostomy When ureteral length is insufficient to allow implantation into the native bladder in a long tunnel, the ureter may safely be implanted into the augmenting segment directly or via a cephalad extension as described above. When placed into the augmenting segment directly a long submucosal tunnel has been used safely without an excessive incidence of stenosis. An alternative procedure is that of Le Duc whereby a mucosal groove is made and the ureter laid in it. Mucosal flaps are elevated and sutured to each side of the ureter (Fig. 108-8). The key to preventing stenosis is to preserve the ureteral blood supply without compromising on tunnel length.

FIG. 108-8. Technique for the Le Duc-type ureteroenterostomy.

Autotransplantation In rare cases, a renal unit with a severely shortened ureter can be salvaged by relocation of the kidney to the pelvis by autotransplantation and reimplantation of the pelvis or proximal ureter. Rarely, however, is an antireflux reimplantation possible and the risk of renal loss significant. On occasion, sufficient renal mobilization can be obtained by division and reanastomosis of the shorter renal vessel (i.e., the vein on the right, the artery on the left) with a decreased risk of vascular injury. However, the role of these measures in patients undergoing concomitant bladder reconstruction is severely limited. Techniques for Reconstruction of the Reservoir Augmentation Cystoplasty Bladder augmentation is often a necessary part of undiversion to achieve a suitable low-pressure, compliant, and adequate capacity reservoir. It is especially useful in the patient with a poorly compliant neurogenic bladder. Over the last several years much attention has been focused on techniques to ameliorate the most common long-term side effects of intestinal bladder augmentation, mucus production, and metabolic alterations. Although several techniques, including autoaugmentation, seromuscular colocystoplasty/gastrocystoplasty lined with urothelium, have been described clinically 2,8,13 for isolated bladder augmentation, thus far they have not established a significant role in undiversion surgery due to the added complexity of, or ancillary procedures required in, undiversion surgery. It is hoped that in the near future such techniques may be incorporated in even the most challenging reconstructions. Thus at present the most common and versatile method of bladder augmentation remains the enterocystoplasty. The fundamentals for a successful enterocystoplasty include the use of a detubularized bowel segment and an extended division of the bladder. Successful augmentation can be achieved with stomach, ileum, or colon, each with unique advantages and disadvantages. Primary considerations in choice of material include the patient's renal function and acid–base balance, previous bowel surgery, and the surgeon's preference. In our hands, when there are no specific contraindications, sigmoid colocystoplasty has proven to be the most reliable and well-tolerated procedure. In addition, colon has an advantage in that a non-refluxing ureteral anastomosis is possible if insufficient length is available to reach the native bladder ( Fig. 108-9). Stomach has been used in those patients with significant acidosis, particularly when the patient has an insensate urethra. Ileum is used when colon is absent or in short supply, in selected patients with cloacal exstrophy, in patients with a permanent colostomy, or when additional procedures using the ileum are planned, such as an ileal Mitrofanoff conduit. The cecum is rarely used in order to preserve the ileocecal valve as its loss may result in diarrhea and loss of fecal continence in children with neurogenic dysfunction.

FIG. 108-9. Technique for transcolonic ureteroenterostomy.

The basic steps in isolating an intestinal segment are similar regardless of the segment chosen. A bowel segment of desired length is mobilized and isolated, and Fogarty vascular clamps are placed on the bowel 10 cm proximal and distal to the proposed site of the incisions. The mesentery is separated from the bowel and divided between ligatures for 1.5 cm. The mesentery is divided toward the root only if necessary to mobilize the segment to its desired location. It is desirable to leave the mesentery to the isolated segment as broad as possible. The intestinal anastomosis is performed with an open technique, which is the same for the large or small intestine. The use of atraumatic clamps away from the area of the anastomosis allows greater freedom and mobility. The anastomosis is performed with a running 4-0 PDS suture in a simple, inverting fashion. Inversion of the mucosa is facilitated by taking larger bites of the seromuscular tissue than the mucosa ( Fig. 108-10). On the anterior wall several Connell stitches are required to completely invert the mucosa. The clamps are released, the anastomosis inspected, and the rent in the mesentery closed after final location of the isolated segment is determined.

FIG. 108-10. Technique for colocolostomy. (Modified from Gonzalez R, Sidi AA. The use of the bowel in the reconstruction of the urinary tract. In: Ehrlich RM, ed. Modern technics in surgery. Mt. Kisco, NY: Futura, 1988.)

Cup-Patch Enterocystoplasty Cup-patch enterocystoplasty is most commonly used when the bladder requires augmentation and the ureters can be implanted in the native bladder or in the augmenting bowel segment. Although this technique was originally described for the small bowel, it is easily adapted for the sigmoid colon which, due to its increased diameter, provides greater volume for a comparable length of segment used. Others have reported using the sigmoid without this detubularization, though with less satisfactory results. 11 A 20- to 25-cm segment of sigmoid is sufficient. In general, it is convenient to place this segment medial to the sigmoid and descending colon. The segment is incised along its antimesenteric border and folded into an inverted U configuration. The adjacent posterior edges are sutured together, beginning at the loose ends, with a running 3-0 polyglactin suture. The distal end of the suture line is sutured to the posterior apex of the cystotomy incision with two 3-0 polyglactin sutures, which are then run laterally to complete the posterior portion of the enterovesical anastomosis. The middle portion of the free intestinal edge is sutured to the anterior apex of the cystotomy with two 3-0 polyglactin sutures run laterally until closure is complete ( Fig. 108-11). The anterior suture line is reinforced by a second layer of 3-0 polyglactin in a Lembert fashion. Often closure will require suturing of bowel to bowel, and lateral dog ears of bowel may be trimmed. An extensive sagittal cystotomy is preferred in order to prevent a subsequent hour-glass deformity; however, it should stop 2 to 3 cm above the bladder neck if placement of an artificial sphincter is contemplated. Before closure, a 20- or 22-Fr Foley catheter is placed through the bowel wall as a cystotomy.

FIG. 108-11. Technique for cup-patch sigmoid or ileal cystoplasty. (From Gonzalez R. Urinary incontinence. In: Kelalis PP, King LR, Belman AB, eds. Clinical pediatric urology. 3rd ed. Philadelphia: WB Saunders, 1992.)

Gastrocystoplasty The use of stomach for bladder augmentation has the advantage of providing a source for the net secretion of acid. This may provide a great benefit in the patient with renal impairment as it not only prevents the development of metabolic acidosis from intestinal absorption but may reduce systemic acidosis if already present. In addition, the stomach produces significantly less visible mucus and provides a thick-walled patch well suited to nonrefluxing ureteral implantation. However, gastric augmentation can result in metabolic alkalosis or troublesome dysuria and hematuria. 12 Furthermore, its use is to be discouraged in the child likely to progress to end-stage renal failure, and transplantation as a defunctionalized bladder augmented with stomach is at great risk for ulceration and perforation. In our hands we have limited its use to those with systemic acidosis, short gut, and an insensate urethra, or in combination with intestinal segments in an attempt to prevent acidosis in patients at risk. The gastric segment is harvested from the body of the stomach. A triangular segment is outlined on the anterior surface and a hemostatic suture placed at the apex along the lesser curvature to control bleeding from the left gastric vessels ( Fig. 108-12A). The blood supply to the segment to be excised is usually based on the right gastroepiploic vessels. Those vessels along the greater curvature at the base of the segment are carefully preserved. The adjacent epiploic vessels to the greater omentum are divided. The gastroepiploic vessels along the greater curvature to the right of the segment are divided, although the corresponding epiploic vessels can be spared. This serves to create a reliable vascular pedicle to allow the segment to reach the bladder. The stomach and omentum are divided and a hand-sewn anastomosis completed. Some have advocated the dividing of the segment with the GIA stapler though we have found this unnecessary and the subsequent removal of the staples time consuming. The segment is then opened and rotated downward to meet the bivalved bladder ( Fig. 108-12B). The vascular pedicle is retroperitonealized by placement posterior and lateral to the mobilized ascending colon. The segment can well accept a tunneled ureteral reimplantation. Post-operatively a nasogastric tube is left until bowel function is well established and H2 blockade is maintained throughout the immediate postoperative period.

FIG. 108-12. Technique for gastrocystoplasty. (A) The patch is mobilized on a right gastroepiploic pedicle. (B) The segment is rotated downward and anastomosed to the opened bladder.

Ureterocystoplasty In an effort to avoid some of the undesirable side effects of augmentation with bowel, augmentation with native ureter has proven to be a useful technique in selected patients. Its use in undiversion is often limited by the absence of a still dilated distal ureter. Often the effect of undiversion has resulted in the loss of previous ureteral dilation. However, in some cases, the use of a TUU or removal of a nonfunctioning renal unit will provide sufficient ipsilateral distal ureter to allow meaningful augmentation. A critical component of this technique is the preservation of the ureteral blood supply. The ureter is harvested with a generous covering of periureteral tissue. An attempt should be made to preserve and include the gonadal vessels. The ureter, once freed, is opened along its anteromedial surface with care not to spiral the incision. The incision is carried through the intramural portion of the ureter. The ureter is then folded into a U configuration and the medial edges reapproximated to form a cup patch. The ureteral patch is then reanastomosed to the edges of the opened bladder wall identically to any other augment ( Fig. 108-13A, B). An alternate approach is to detach the ureter completely from the bladder but with preservation of the vascular supply from the vesicle artery. The ureter is then opened longitudinally, reconfigured, and anastomosed as described above ( Fig. 108-13C). We have not attempted to reimplant a contralateral ureter within this segment. In our experience, the ureteral segments have usually demonstrated excellent expansion, even if the initial patch appears limited. Patches incorporating the entire ureteral length, including the renal pelvis, have been successfully brought down to the bladder with acceptable vascular supply.

FIG. 108-13. Technique for ureterocystoplasty. (A) The harvested ureter is opened along the medial aspect and anastomosed in situ. (B) If the renal unit is salvageable, the ureterocystoplasty can be combined with a transureteroureterostomy. (C) Alternatively, the ureter can be detached from the bladder with preservation of the inferior vesical artery.

Continent Diversion When there is no usable native bladder, conversion from an incontinent diversion to a continent diversion should be considered. At present there are a wide variety of models for continent diversions from which to choose. A description of the various methods of continent diversion is beyond the scope of this chapter and can be found in Chapter 80, Chapter 81 and Chapter 82. Procedures to Enhance Bladder Outlet Resistance When preoperative evaluation indicates that the sphincter mechanism is incompetent there are several methods available to enhance outlet resistance. Options include bladder neck reconstruction, bladder neck fascial sling/cinch (BNFS/C), 1,16 the artificial urinary sphincter (AUS), urethral replacement by a perineal appendiceal conduit, and bladder neck closure with an abdominal continent catheterizable conduit. In patients with a neurogenic bladder and a virgin bladder neck we believe that an artificial sphincter or, in some females, a BNFS/C is preferable to bladder neck reconstruction. Although continence may be similar, AUS and BNFS/C are less likely to cause difficulty with intermittent catheterization. The AUS has the greatest success rate when it is the first operation done to increase outlet resistance. In someone with a previously operated on bladder neck, the placement of an artificial sphincter is less desirable due to an increased risk of failure from erosion.7 Bladder Neck Reconstruction A wide number of procedures have been devised to lengthen and tubularize the bladder neck including that sequentially described by Young, Dees, and Leadbetter (YDL), the Kropp, 10 and the Salle procedure, 15 along with numerous others. Currently, our use of these techniques is limited to the reconstruction of bladder exstrophy where we still rely on the YDL and have failed to achieve a significant advantage from the other methods. The procedure begins with a sagittal cystotomy extended anteriorly to the bladder neck. Two parallel incisions 5 cm long are made in the mucosa of the posterior bladder wall 15 mm apart, starting at the bladder neck. In most cases of undiversion, there is no reason to be concerned with the ureters in the trigone. Two triangles of mucosa lateral to the parallel incisions are excised. The mucosal strip is tubularized over a 12- or 14-Fr Foley catheter with a running polyglactin suture. The two lateral muscle flaps are wrapped around the mucosal tube, one overlapping the other like a double-breasted jacket. Bladder capacity may be reduced after this procedure, and augmentation may become necessary (Fig. 108-14).

FIG. 108-14. Modified technique for the Young-Dees-Leadbetter bladder neck tubularization in combination with a cup-patch enterocystoplasty. (Modified from Gonzalez R, Sidi AA. The use of the bowel in the reconstruction of the urinary tract. In: Ehrlich RM, ed. Modern technics in surgery. Mt. Kisco, NY: Futura, 1988.)

Artificial Urinary Sphincter When planning to place an artificial urinary sphincter, the dissection of the bladder and bladder neck should be done early in the procedure, preferably before opening of the bladder and peritoneum. The plane must be developed gently and by blunt dissection in order to avoid perforation of the posterior bladder wall. Palpation of a Foley catheter and balloon is an unreliable method to identify the bladder neck. Often after developing the plane on both sides there will seem to be only a thin band of tissue between one's fingers. The strong temptation to push through this tissue with a right angle clamp must be resisted as this tissue is usually composed of a fold of the posterior bladder wall and a perforation will surely result ( Fig. 108-15).

FIG. 108-15. When creating the dissection plane for placement of the artificial urinary sphincter cuff, care must be taken to avoid perforation of the posterior bladder neck.

In the male, the plane should be anterior to the seminal vesicles, which can be identified by the presence of a prominent venous plexus ( Fig. 108-16). In women, the dissection can be assisted by placement of a finger in the vagina ( Fig. 108-16). If dissection of the plane is difficult, the bladder dome can be incised and the dissection aided by inspection and palpation of the trigone and bladder neck from the inside. The sagittal cystotomy incision must end 2 to 3 cm from the site to be surrounded by the cuff to avoid placing the sphincter over a suture line.

FIG. 108-16. The proper planes of dissection for the AUS cuff. In males the cuff lies anterior to the seminal vesicles. In females a finger in the vagina may aid in dissection.

After completion of the dissection, the plane is marked by a vessel loop or umbilical tape and the bladder irrigated with methylene blue to rule out perforation. After completion of the reconstructive procedure, the operative field is irrigated with an antibiotic solution and the sphincter components are placed and connected. The cuff should be fitted such that it is very snug yet still able to be turned around the bladder neck. The reservoir is placed in an extraperitoneal pocket behind the abdominal wall. In the child with a small labia or scrotum the Securo-Tee model pump with a separate controller valve mechanism may provide greater comfort. The fascia is closed and the tubing connections made. Details regarding priming and connecting of the sphincter can be found in Chapter 55. Bladder Neck Fascial Sling/Cinch The most appropriate role for the bladder neck fascial sling or cinch (circumferential wrap of fascia around bladder neck) procedure remains poorly defined. Clearly in those with a failed artificial sphincter or bladder neck reconstruction, the sling/cinch may provide added continence without closing the bladder neck. We have also used it as a primary procedure in patients uneasy with the concept of a artificial sphincter or in those who have no chance of spontaneous voiding. Unfortunately, to date although results in females have been encouraging, the success rate in males has been unsatisfactory. 4 Furthermore, options for revision are extremely limited and often bladder neck closure is the only choice. Nonetheless there are clearly many women who may benefit from such a procedure without the added complexity of an artificial sphincter. The development of the bladder neck plane is identical to that required for the artificial sphincter. The fascia is harvested from the anterior rectus sheath leaving a medial strip adequate for fascial closure. A 2 × 10 cm segment is excised and passed behind the bladder neck. In the case of the sling, 2-0 prolene sutures tied over Teflon pledgets are used for suspension from the anterior abdominal wall ( Fig. 108-17). Modifications include encircling the urethra with the strip or anchoring the strip to Cooper's ligament. Beware of excessive tightening of the sutures as this may cause angulation and difficulty in catheterization. Ease of catheterization should be confirmed prior to closure.

FIG. 108-17. Bladder neck suspension with a fascial sling. (Modified from Gonzalez R, Sidi AA. The use of the bowel in the reconstruction of the urinary tract. In: Ehrlich RM, ed. Modern technics in surgery. Mt. Kisco, NY: Futura, 1988.)

In our modification of the cinch procedure 16, a buttonhole is made in one end of the strip and the other end is passed through, much as in measuring the urethra for the AUS, after passage around the bladder neck. The strip is tightened down snugly and pexed into position with interrupted 2-0 prolene sutures, between the two layers of strip, and to the bladder neck ( Fig. 108-18). The residual end can be suspended through the rectus sheath. Ease of catheterization should likewise be confirmed. Early results with this procedure in our hands has shown acceptable results in women but disappointing outcomes in men.

FIG. 108-18. Technique for the modified bladder neck fascial cinch procedure. Following wrapping around the bladder neck, the fascial strip is secured to itself and the bladder neck.

Procedures for Urethral Substitution When the urethra is absent or unsuitable for intermittent catheterization, creation of an alternate conduit becomes necessary. The procedure we have found to be most reliable is creation of a continent catheterizable stoma using the vermiform appendix based on the Mitrofanoff principle. 17 When the appendix is not available, the conduit can be constructed of tapered ileum, ureter, or fallopian tube. Although conduits of bladder wall and prepuce 14 have been described, we have had no experience with these techniques. The conduit can be brought to the umbilicus or abdominal wall, though we feel that the umbilicus is superior cosmetically as well as easier for the patient to locate by touch, especially if obese. Our results have shown the appendix to be superior to other materials though difficulty in catheterization due to stoma stenosis remains an occasional problem. In some female patients, especially those with a history of bladder exstrophy, the appendix can completely replace the urethra and be brought orthotopically to the perineum with excellent continence and cosmetic appearance. Abdominal Mitrofanoff The appendix is harvested from the cecum with a small cuff of cecum left attached. The base of cecum is oversewn with two layers of running 4-0 PDS suture. The mesoappendix is freed from the base of the cecum with care taken to avoid injury to the delicate blood supply. Little mobilization is needed when the appendix will be attached to the umbilicus. A submucosal tunnel is created in the native bladder or augmenting bowel segment and the appendix gently drawn through ( Fig. 108-19A). The tunnel must be wide enough to accommodate the broad mesoappendix without tension. A Le Duc anastomosis, i.e., placement of the appendix into a mucosal trough with fixation of the mucosal edges to the lateral walls of the appendix, may be necessary. The distal tip is excised, spatulated, and anastomosed with interrupted 4-0 Vicryl. The extravesical appendix is secured to the bladder serosa to prevent tunnel shortening. Ease of catheterization is assessed. Following bladder reconstruction, a V incision is made in the umbilicus and the proximal end of the appendix is brought through the abdominal wall fascia beneath the umbilicus and

secured. The end is spatulated cephalad to accept the V flap and the appendix is sutured to the umbilical skin using interrupted 4-0 PDS sutures ( Fig. 108-19B). The bladder and appendix are secured to the abdominal wall to prevent tension and prevent the “maypoling” of the intestine around the appendix. The extravesical portion of the appendix should be kept as short and straight as possible. When the appendix is too short an extension can be made from the base of the cecum. When the appendix is absent a tapered segment of ileum or a fallopian tube can be used in an identical manner. The conduit should be checked for continence and ease of catheterization with the bladder both full and empty.

FIG. 108-19. Technique for creation of a continent catheterizable appendicovesicostomy conduit to the abdominal wall (Mitrofanoff) (A). Technique for a V flap anastomosis between the appendix and the umbilicus (B).

Perineal Mitrofanoff In select patients the appendix can be brought out to the perineum. The appendix is harvested as described. The bladder is opened through the bladder neck. A submucosal tunnel is created in the floor of the bladder. The appendix is brought in through the posterior wall to lie in the tunnel with the tip extending out distally through the opened bladder neck and out to the perineal surface ( Fig. 108-20). The intravesical end is sutured into position and the bladder neck closed around the appendix. The perineal end is spatulated and sewn into position. A tongue of omentum can be wrapped around the extravesical appendix to fill the dead space and provide additional blood supply. We reserve this procedure almost exclusively for females with bladder exstrophy who wish to avoid an abdominal stoma.

FIG. 108-20. Technique for creation of a continent catheterizable appendicovesicostomy conduit to the perineum (perineal Mitrofanoff).

Postoperative Care The immediate postoperative care is the same as for any patient after major abdominal and intestinal surgery. After ureteral anastomosis, the ureteral tubes are left in for 7 to 10 days. A radiographic study is obtained to rule out leaks prior to removal of the tubes. The augmented bladder requires daily irrigation with saline to prevent occlusion of the bladder drainage tubes by mucus. The cystostomy catheter is left indwelling for a minimum of 3 weeks or until intermittent catheterization can be initiated. When a Mitrofanoff conduit has been constructed, indwelling catheterization may be necessary for up to 6 weeks. If bladder neck tubularization has been performed, urethral catheterization should be maintained for at least 4 weeks. An artificial sphincter, if placed, should remain deactivated for 6 weeks. Extravesical drains are typically not used; however, when placed they can usually be removed 5 to 7 days following surgery.

OUTCOMES
Complications Stenosis after ureteroureterostomy should only occur exceptionally. Preservation of the ureteral blood supply, a widely spatulated anastomosis, and careful technique aided by optical magnification and the use of fine, absorbable sutures are the keys to good results. Urinary leaks should also be rare if these guidelines are followed. If a leak is discovered on routine postoperative radiographs and the ureter is still intubated, prolonged intubation results in complete healing. If the leak occurs after the tubes are removed, a percutaneous nephrostomy is necessary. Complications associated with long-tunnel ureteroneocystostomy with a bladder hitch also are rare. Ureteral obstruction and vesicoureteral reflux are the most significant complications. Obstruction can result from ureteral ischemia or angulation, but not from a properly constructed long tunnel. Reflux, particularly after enterocystoplasty, is a most undesirable complication that must be corrected if persistent. Enterocystoplasty can be associated with both short- and long-term complications. Most early complications can be minimized or avoided by use of a careful surgical technique. Among the most serious early complication are leaks, or obstruction of the intestinal anastomosis, both of which require immediate surgical correction. The long-term complications include acid–base and electrolyte imbalances, bladder perforation, and stone formation. Chronic bacteriuria is inevitable after enterocystoplasty in patients who perform intermittent catheterization, but its long-term consequences in the absence of reflux are unknown even when the upper tracts are often colonized. Clinical pyelonephritis is not common. Its occurrence may indicate reflux, obstruction, or noncompliance with the prescribed program of intermittent catheterization. The potential risk of chronic diarrhea after exclusion of the ileocecal valve from the gastrointestinal tract has already been discussed. The reinforced ileocecal valve provides a poor antireflux mechanism regardless of technique used. Vigorous attempts to prevent reflux have led to partial obstruction. If the ileal segment proximal to an obstructed valve is long, the segment can become distended and the accumulated mucus can cause stone formation, intermittent obstruction, and pyelonephritis. Artificial sphincters are at risk for both infection and erosion. These complications are much greater in patients in whom the cuff is placed around the bulbous urethra or if the patient has undergone previous bladder neck injury or reconstruction. The use of bowel for augmentation at the same setting as sphincter implantation should not result in increased infections in a properly prepared patient. 6 All patients should have a sterile urine culture prior to sphincter placement. Patients should be advised of the possibility of mechanical failure requiring surgical revision due to long-term use of the device. The most troublesome complications of bladder neck reconstruction and bladder neck fascial slings are failure to attain continence and difficulty with catheterization. Endoscopic manipulation is the usual first step to correct catheterization difficulties. When it fails, creation of a Mitrofanoff conduit may become necessary to allow

reliable catheterization. Mitrofanoff conduits themselves can be plagued by catheterization difficulties, most often due to stomal stenosis. Dilation can on occasion be successful, although in most cases revision is necessary. Incontinence that is persistent several months following undiversion requires a thorough evaluation. Decreased bladder compliance or small capacity is treated either pharmacologically or surgically. Low outlet resistance may represent a failure of the original procedure or a misinterpreted preoperative evaluation. Appropriate measures as described above should be initiated. One should not hesitate to reoperate on these patients to obtain the desired result. Patient compliance with the surgeon's recommendations should not present a problem if a clear and realistic picture has been presented before undiversion was performed. Results The reconstruction of the previously diverted urinary tract is often desirable from the patient's psychological and social perspectives. Also, from a medical standpoint, it is desirable to eliminate an infected, freely refluxing system, which so often leads to progressive renal deterioration. Although renal transplantation in patients with intestinal conduits has produced acceptable results, undiversion before transplantation is often the better option. When considering undiversion or reconstruction to obviate cutaneous diversion, the surgeon must weigh the possible benefits against the known risk of complications and the potential need for reoperations to achieve the desired results. The decision must be made with a well-informed patient and family who have been presented with a realistic picture of the procedure and probable outcome. Performance of such a reconstruction on an ambivalent or poorly motivated patient can be disastrous. From a technical viewpoint, almost all patients are candidates for undiversion or a form of internal continent diversion. Anatomic factors such as ureteral length and function, bladder capacity and compliance, urethral continence and patency along with systemic metabolic status must be preoperatively evaluated. By the proper evaluation of each element of the urinary tract and the appropriate application of the individual measures to restore the necessary attributes, most patients can expect the creation of a continent, non-refluxing, and safe urinary system. It is through the thoughtful application of these reconstructive techniques that the classic permanent incontinent diversion can become for children truly a thing of the past. CHAPTER REFERENCES
1. Bauer SB, Peters CA, Colodny AH, Mandell J, Retik AB. The use of rectus fascia to manage urinary incontinence. J Urol 1989;142:516. 2. Cartwright PC, Snow BW. Bladder autoaugmentation: partial detrusor excision to augment the bladder without use of bowel. J Urol 1989;142:1050. 3. DeBadiola FIP, Ruiz E, Denes ED, Smith CA, Bukowski TP, Gonzalez R. New application of the gastrostomy button for clinical and urodynamic evaluation prior to vesicostomy closure. J Urol 1996;156:618. 4. Elder JS. Periurethral and puboprostatic sling repair for incontinence in patients with myelodysplasia. J Urol 1990;144:434. 5. Gonzalez R, Sidi AA. Preoperative prediction of continence after enterocystoplasty or undiversion in children with neurogenic bladder. J Urol 1986;134:705. 6. Gonzalez R, Nguyen DH, Koleilat N, Sidi AA. Compatibility of enterocystoplasty and the artificial urinary sphincter. J Urol 1989; 142:502. 7. Gonzalez R, Koleilat N, Austin C, Sidi AA. The artificial sphincter AS 800 in congenital urinary incontinence. J Urol 1989;142:512. 8. Gonzalez R, Buson H, Reid C, Reinberg Y. Seromuscular colocystoplasty lined with urothelium: experience with 16 patients. Urology 1995;45:124. 9. Kowalczyk JJ, Keating MA, Ehrlich RM. Femoral nerve neuropathy after the psoas hitch procedure. Urology 1996;47:563 10. Kropp KA, Angwafo FF. Urethral lengthening and reimplantation for neurogenic incontinence in children. J Urol 1986;135:533 11. Mitchell ME. Use of bowel in undiversion. Urol Clin North Am 1986;13:349 12. Nguyen DH, Bain MA, Salmonson KL, Ganesan GS, Burns MW, Mitchell ME. The syndrome of dysuria and hematuria in pediatric urinary reconstruction with stomach. J Urol 1993;150:707 13. Nguyen DH, Mitchell ME, Horowitz M, Bagli DJ, Carr MC. Demucosalized augmentation gastrocystoplasty with bladder autoaugmentation in pediatric patients. J Urol 1996;156:206 14. Perovic S. Continent urinary diversion using prenuptial penile or clitoral skin flap. J Urol 1996;155:1402 15. Salle JL, deFraga JC, Amarante A, et al. Urethral lengthening with anterior bladder wall flap for urinary incontinence: a new approach. J Urol 1994;152:803 16. Walker RD III, Flack CE, Hawkins-Lee B, Lim DJ, Parramore H, Hackett RL. Rectus fascial wrap: early results of a modification of the rectus fascial sling. J Urol 1995;154:771 17. Woodhouse CR, MacNeily AE. The Mitrofanoff principle: expanding upon a versatile technique. Br J Urol 1994;74:447.

Chapter 109 Circumcision Glenn’s Urologic Surgery

Chapter 109 Circumcision
Jay B. Levy and Stephen A. Kramer

J. B. Levy: Division of Urology, University of North Carolina School of Medicine, and Presbyterian Hospital, Carolinas Medical Center, Charlotte, North Carolina 28204. S. A. Kramer: Department of Pediatric Urology, Mayo Clinic, Rochester, Minnesota 55905.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique General Considerations Sleeve Resection Technique Dorsal Slit Procedure and Circumcision Gomco Clamp Procedure Outcomes Complications Chapter References

The medical indications for surgery on the foreskin include phimosis, paraphimosis, preputial neoplasms, balanitis, and posthitis unresponsive to conservative therapy. With the recent increase in the number of urinary tract anomalies detected with in utero ultrasonography, circumcision should be advised in order to decrease the risk of ascending urinary tract infections once a urinary tract anomaly has been identified. 3 In normal healthy infants, circumcision is completely a matter of family choice, and parents should be given objective information regarding the benefits and drawbacks of the procedure as well as the possible complications. Circumcision is contraindicated in boys with hypospadias, chordee without hypospadias, or a prominent dorsal preputial hood that may herald an underlying congenital deformity of the penis. Circumcision should also be deferred in children with significant penoscrotal tethering, buried penis with large fat pad, or small penis. Occasionally, as in the megameatus intact prepuce anomaly (MIP variant of hypospadias), there is a normal prepuce, despite an underlying penile abnormality. Therefore, it is mandatory that the prepuce be retracted prior to circumcision in all cases, to expose the glans and meatus before foreskin removal.

DIAGNOSIS
Phimosis is the condition in which fibrotic contraction of the preputial skin prevents retraction of the visceral surface of the foreskin back over the glans. This occurs in the young infant secondary to physiologic congenital adhesions between the glans and the foreskin. Phimosis may also result from the forcible retraction of the foreskin in young infants with resultant cicatrix formation that narrows the preputial aperture. Therefore, forcible retraction of the foreskin by the parents or the physician is ill advised. Posthitis is preputial inflammation that usually results from local irritation or infection due to a phimotic preputial opening. This local cellulitis is usually self-limiting, but it may progress to affect the glans (balanitis) or even the penile shaft or perineum. Paraphimosis is entrapment of the preputial skin proximal to the coronal sulcus. This occurs when the foreskin is forcibly retracted behind the glans penis and is not reduced promptly. Secondary swelling of the glans entraps the foreskin and can prevent reduction.

INDICATIONS FOR SURGERY
Circumcision, particularly in infants and children, is not a trivial procedure that can be performed within a few minutes by an inexperienced surgeon. To the contrary, circumcision should be performed by an experienced individual who can evaluate the patient preoperatively and identify contraindications to the procedure, manage possible complications, and evaluate the patient postoperatively. Indications include phimosis preventing the retraction of the prepuce, balanoposthitis, and paraphimosis.

ALTERNATIVE THERAPY
In patients with posthitis, treatment consists of a combination of both local and oral antibiotics, but ultimately circumcision is necessary, as phimosis is the underlying predisposing factor. If paraphimosis is identified early, reduction may be performed by gentle but firm glandular compression ( Fig. 109-1A). However, if the condition is not recognized early or is neglected, continued swelling of the glans may render manual reduction impossible. When this situation occurs, a longitudinal incision through the constricting ring may be required to allow reduction of the paraphimosis ( Fig. 109-1B). Significant edema and inflammatory changes make formal circumcision difficult, and one should postpone the procedure for several weeks.

FIG. 109-1. (A) Reduction of paraphimosis by firm compression against the glans penis. (B) Incision is made through the constricting ring when manual reduction is unsuccessful.

SURGICAL TECHNIQUE
General Considerations A properly performed circumcision requires meticulous technique and adequate anesthesia to ensure a safe and satisfactory cosmetic result. In adults, circumcision may be accomplished with local anesthesia using 1% lidocaine without epinephrine. The injection is performed through a cutaneous wheal at the base of the penis in the dorsal midline, with deeper infiltration subcutaneously around the circumference of the penile shaft. Particular attention is directed to generous infiltration of the dorsal midline and the angles between the corpus cavernosum and the corpus spongiosum on either side. In younger children and infants over 50 weeks gestation, general anesthesia is preferred. Neonatal circumcision is performed with anesthetic injection using 0.5% cm 3/kg of 1% lidocaine without epinephrine mixed in equal parts with 0.5% bupivacaine injected as a dorsal nerve block. The injection may be augmented using a “whiskey nipple.” Adequate hemostasis is mandatory to prevent postoperative bleeding or hematoma formation with breakdown of the suture line. Postoperative edema occurs

uniformly, as the penile shaft is devoid of fatty tissue; therefore, sutures must be tied securely but loosely enough to avoid cutting through the skin edges. Sleeve Resection Technique Although several techniques of circumcision are available, the sleeve resection technique allows uniformly good cosmesis with direct visualization of the glans throughout the procedure (Fig. 109-2). The foreskin is retracted and adhesions between the prepuce and the glans are broken manually. The visceral prepuce should be retracted until the deep purple coronal sulcus is visualized completely. With the prepuce in its usual position and without applying any tension, the coronal junction is outlined with a marking pencil and is incised circumferentially with care taken to follow the “V” of the frenulum on the ventral surface ( Fig. 109-2). A circumferential incision is then made 8 mm proximal to the coronal sulcus on the mucosal surface of the prepuce. The incision is straight across the base of the frenulum, which allows the frenulum to retract up into the V of the glans. The sleeve of skin between the two incisions is then excised in the avascular plane between dartos and Buck's fascia. Hemostasis is meticulously carried out using electrocautery with extreme care not to damage the neurovascular bundle. The skin edges are approximated with care using 4-0 or 5-0 chromic in an interrupted fashion. A compressive dressing of petroleum jelly and sterile gauze is applied.

FIG. 109-2. Sleeve resection procedure. (A) A circumferential incision is made in the outer preputial skin at the level of the coronal sulcus. (B) A second circumferential incision is made in the inner preputial skin just proximal to the coronal sulcus. (C) The sleeve of skin is excised. Hemostasis is obtained by ligature or electrocautery. (D) The skin edges are reapproximated with fine absorbable sutures in an interrupted fashion.

Dorsal Slit Procedure and Circumcision The dorsal slit procedure can be used with or without circumcision to promote healing of refractory balanitis, to allow removal of preputial calculi, to expose the glans for biopsy, and to reduce paraphimosis that persists after attempts at manual reduction. In the dorsal slit procedure, the edges of the prepuce are secured at 5 o'clock and 7 o'clock and at 12 o'clock. Both edges of the prepuce are crushed at 12 o'clock and an incision is made in the prepuce from its free end to within 8 mm of the corona of the glans. Bleeders are fulgurated or secured with absorbable sutures, and the severed edges of prepuce are approximated with interrupted 4-0 or 5-0 chromic catgut. Circumcision is deferred for a few weeks to allow the inflammation or infection to resolve. In small boys, the dorsal slit may be used as part of a circumcision. The skin is marked at the coronal margin as described previously for the sleeve technique. A dorsal slit is made to the mark at the coronal margin. Both layers f the prepuce are grasped with a hemostat and then divided along the marked line delineating the coronal margin. One must be careful on the ventral surface to provide a “V” for the frenulum. Hemostasis is obtained, and the skin edges are secured with 4-0 or 5-0 chromic catgut. A petroleum jelly dressing and sterile gauze are applied. Gomco Clamp Procedure Circumcision in the neonate or young infant can be performed with a Gomco clamp. The foreskin is retracted and the glans visualized to ensure that hypospadias does not exist. Adhesions between the foreskin and underlying glans penis are broken manually. A dorsal slit is made as described previously ( Fig. 109-3). An appropriately sized bell is placed over the glans and the prepuce drawn back over the bell. Careful inspection is done to align the midline raphe and prevent torsion, and to ensure that the appropriate amount of skin is being removed. The bell is then clamped in position. Appropriate tension is necessary in advancing the foreskin over the bell to prevent inadequate or overzealous circumcision. The clamp is applied as tightly as possible to obtain adequate hemostasis, and compression is maintained for 5 to 7 minutes. The skin distal to the clamp, at the junction of the clamp and the bell, is excised sharply with a knife (not by cautery, as this can lead to penile necrosis).

FIG. 109-3. Gomco clamp procedure. (A) A dorsal slit is made after carefully releasing all adhesions between the inner preputial skin and the glans. (B) An appropriately sized bell is placed over the glans, and the foreskin is drawn over the bell. (C) The clamp is placed over the bell and foreskin and locked in place for several minutes. (D) The excess foreskin is sharply excised and the clamp removed.

Although this technique can provide an expedient and satisfactory result, the surgeon must have experience and confidence in using this mechanical device to avoid serious sequelae. The most common surgical complication is bleeding due to inadequate vascular compression. In older boys, this technique is contraindicated because larger vessels are not adequately compressed and bleeding may occur. Inappropriate application of the Gomco clamp may result in amputation of all or part of the glans penis. Recently, we have developed a modification of the Gomco clamp that allows complete visualization of the glans penis at all times and will prevent this disastrous complication (Lauvetz, Levy, Kramer, unpublished data). If too much skin is removed, causing the shaft of the skin to be denuded, we recommend conservative local management, as regeneration of penile skin usually affords complete coverage of the shaft defect.

OUTCOMES
Complications Careful neonatal circumcision is associated with a very low complication rate. In two large series, this rate was 0.2%, with most problems being minor. 2,6 Complications include excessive skin removal, postoperative bleeding, penile adhesions, buried penis secondary to inadequate skin removal, meatal stenosis, urethrocutaneous fistula secondary to removal of excess ventral skin, necrosis of the glans, or penile amputation. 2,5 In most cases, excessive skin removal can be managed conservatively with local antimicrobial ointment and healing by secondary intention with good cosmesis. 1 Most postoperative bleeding ceases with manual pressure, which need not be excessive. Occasionally, suture ligation or a compressive dressing may be required. Urethrocutaneous fistulas can be avoided by careful

preoperative evaluation to detect minimal hypospadias or chordee without hypospadias in which there is deficient corpus spongiosum. Concealed penis due to a scarred preputial ring after circumcision usually results from inadequate skin removal and/or the presence of a large prepubic fat pad. 1 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Allen TD. Disorders of the male external genitalia. In: Kelalis PP, King LR, Belman AB, eds. Clinical pediatric urology. Vol. 2. Philadelphia: WB Saunders, 1976;638. Belman AB. Complications of circumcision. In: The penis. Urol Clin North Am 1978;5:25. Gee WF, Ansell JS. Neonatal circumcision: a ten-year overview. Pediatrics 1976;58:824. Nasrallah PF. Circumcision and concealed penis. Soc Pediatr Urol Newslett 1984;April 6, p. 25. Shulman J, Ben-Hur N, Neuman Z. Surgical complications of circumcision. Am J Dis Child 1964;107:149. Wiswell TE, Geschke DW. Risks from circumcision during the first month of life compared with those from uncircumcised boys. Pediatrics 1989;83:1011.

Chapter 110 Cystoscopic Stone Basket Extraction Glenn’s Urologic Surgery

Chapter 110 Cystoscopic Stone Basket Extraction
Stevan B. Streem

S. B. Streem: Section of Stone Disease and Endourology, Department of Urology, The Cleveland Clinic Foundation, Cleveland, Ohio 44195.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Recent estimates suggest that as many as 5% of American women and 12% of American men will develop at least one episode of symptomatic urolithiasis during their lifetimes, and that more than 1.2 million Americans will require treatment each year. While the social and economic impact of urolithiasis may be inestimable, one cost-demographic study suggests that more than $1 billion is now spent yearly on its evaluation and treatment. 2 The pathophysiology of ureteral calculi is essentially the same as that of renal calculi. Nearly all such stones form in the pyelocalyceal system and subsequently cause symptoms during passage through the ureter. The formation of a primary ureteral calculus is a rare event that almost always occurs in association with an anatomic or physiologic ureteral abnormality, such as a ureterocele or primary megaureter, that causes localized obstruction and stasis. Historically, up to 80% of ureteral calculi passed spontaneously. However, with newer, less invasive technology now available for ureteral stone management, fewer patients are treated expectantly, either at the initial presentation or during the subsequent course of follow-up. In fact, more recent studies suggest that intervention will ultimately be performed in at least 40% of such patients. 7

DIAGNOSIS
The diagnosis of a ureteral stone is usually suggested when a patient develops acute, colicky flank pain associated with gross or microscopic hematuria. An intravenous pyelogram is often diagnostic, but this requires visualization of contrast media to the level of a calcific density shown to lie within the urinary tract in at least two views (Fig. 110-1). Alternatively, ultrasonography or computed tomography scanning can allow visualization of a dilated ureter to the level of a stone. These studies are especially valuable when the use of intravenous contrast is contraindicated or when the stone is radiolucent ( Fig. 110-2). When these noninvasive studies are nondiagnostic, retrograde pyelography should prove definitive. There are rare cases in which retrograde instrumentation cannot be performed as in some diverted patients, or when a clinical setting such as urinary sepsis precludes the use of anesthesia that might be required for retrograde cystoscopic instrumentation. In those cases, percutaneous nephrostomy and antegrade pyelography will provide definitive information as well as acute relief of obstruction.

FIG. 110-1. (A) Plain x-ray reveals a 4-mm calcific density (arrow) overlying the area of the distal left ureter. (B) During intravenous pyelography, columnization of contrast is seen to the level of the stone. (C) The calcification is also seen in the course of the ureter on this oblique view, proving it to lie within the urinary tract.

FIG. 110-2. (A) Smooth, lucent defects (arrow) associated with localized obstruction in the left distal ureter during intravenous urography. (B) Thin-cut CT scan of the pelvis without intravenous contrast identifies this as a stone ( arrow) in the distal ureter. (Reprinted with permission from Coe, Favus, Pack, et al. Kidney stones: medical and surgical management. Philadelphia: Lippincott-Raven Publishers, 1996.)

INDICATIONS FOR SURGERY
Advances in technology have had a significant impact on ureteral stone management. However, while the manner in which we intervene may have changed, the absolute and relative indications for intervention have not. These indications, one or more of which may be present at any given time, include persistent or progressive hydronephrosis, failure of distal progression of the stone, pain requiring continued or recurrent use of narcotic analgesics, infection associated with the stone, socioeconomic factors, and stone size judged too large to pass spontaneously. By most estimates, stones measuring less than 4 mm in diameter have a 90% chance of passing spontaneously, whereas those measuring 4 to 7 mm have a 50% chance. Stones larger than 7 mm rarely pass spontaneously. 9,10 Of these indications for intervention, only associated infection is a relative contraindication to basket extraction. In such cases, obstruction should be relieved by simple retrograde passage of a catheter or stent, or, alternatively, institution of percutaneous nephrostomy drainage. Definitive stone extraction is then delayed until the infection has been treated and time given for any associated ureteral edema or friability to resolve. While there is no absolute stone size that precludes successful basket extraction, this procedure should generally be limited to stones located below the level of the iliac vessels that measure less than 6 mm in diameter. Cystoscopic basketing of stones located more proximally in the ureter is associated with a higher incidence of

significant complications including ureteral avulsion. When ancillary techniques such as balloon dilation of the ureter or ureteral meatotomy are used, even stones measuring up to 1 cm or greater can be approached this way. However, in addition to adhering to these general guidelines regarding size and location, safe and successful employment of cystoscopic stone basket extraction always requires evaluation and consideration of individual anatomy ( Fig. 110-3).

FIG. 110-3. (A) Plain film suggests a 6-mm calculus in the distal ureter that should be amenable to simple basket extraction. (B) Retrograde study in this patient reveals a relatively long segment of ureter distal to the stone that would not readily allow basket extraction of this stone intact. (C) Plain film suggests a larger (1–8 cm) stone, which seems too large for simple basket extraction. (D) Retrograde study in this patient, however, reveals that the stone is located very distally in a ureter of generous diameter. This stone is in fact amenable to basket extraction, though dilation of the intramural ureter or a ureteral meatotomy may be required.

ALTERNATIVE THERAPY
Currently, ureteral calculi lodged in the proximal ureter are best managed with extracorporeal shock wave lithotripsy (ESWL), whereas those in the midureter are managed by using ESWL with or without initial endoscopic retrograde manipulation. For larger stones in the distal ureter, there continues to be controversy regarding the respective roles of ESWL and ureteroscopy. It is clear, however, that for stones too large to safely be delivered intact through the distal and intramural ureter to the bladder, some form of lithotripsy, either extra- or intracorporeal, must first be used to fragment the stone. Yet there remains a relatively large number of patients in whom intervention will be required for a calculus small enough and distal enough that simple basket extraction, performed with fluoroscopic control, will prove definitive. In such cases, ureteroscopic stone extraction under direct visual control is always a viable alternative. However, ureteroscopic management can be associated with complications such as ureteral perforation or proximal stone migration. As such, in these selected patients, an initial attempt at simple basket extraction may be worthwhile as this procedure predictably allows safe and rapid resolution. Ureteroscopic extraction can then be reserved for those patients in whom fluoroscopic basketing fails.

SURGICAL TECHNIQUE
The procedure is most often performed with general, spinal, or epidural anesthesia. With the patient in a dorsal lithotomy position, cystoscopic examination is undertaken with attention to the appearance of the involved orifice. Unless the stone is seen bulging in the intramural ureter, the anatomy is further defined by gentle retrograde injection of contrast using a cone-tipped catheter. Basket extraction then can proceed. A number of instruments for extraction are available, though most are variations of a Dormia basket or Davis loop. In general, we prefer the use of a five-wire Dormia-type basket with a filiform tip. Compared to a blunt-tipped basket, the filiform tip allows less traumatic and more predictable passage of the basket beyond the stone. Under fluoroscopic control, the closed basket is passed proximal to the stone ( Fig. 110-4A). The basket is fully opened, then slowly withdrawn with the stone engaged (Fig. 110-4B). If the stone has not been delivered in the basket upon its withdrawal, the filiform tip is left in place through the orifice and alongside the stone while the basket is closed and subsequently passed proximally for a second attempt. In contrast to basket extraction under direct vision with a ureteroscope, during fluoroscopic basket extraction the basket is never closed down on the stone as this can result in entrapment of ureteral mucosa.

FIG. 110-4. (A) A closed Dormia basket with a filiform tip is passed proximal to the stone under fluoroscopic control. (B) The basket is fully opened and slowly withdrawn with the stone engaged. (Modified with permission from AC Novick, SB Streem, and JE Pontes (Eds.), Stewart's operative urology. Baltimore: Williams and Wilkins, 1989.)

If the stone cannot be bypassed with the basket because of impaction or ureteral angulation, an attempt can be made to pass a hydrophilic glidewire. This both straightens the ureter and opens up space between the stone and ureteral wall so that the procedure can proceed as planned. For stones of moderate size (6 to 8 mm), balloon dilation of the intramural ureter can be performed at the outset of the basket extraction. This may be accomplished by first passing a 4-mm or 5-mm balloon catheter over a guidewire, or by using a basket with an integrated balloon ( Fig. 110-5).4

FIG. 110-5. (A) Retrograde study reveals a moderate size stone which will require dilation of the intramural ureter. (B) A basket with an integrated dilating balloon is passed under fluoroscopic control. The balloon in inflated in the intramural ureter up to, but not adjacent to, the stone. (C) Dilation of the intramural ureter has allowed intact, atraumatic delivery of the stone. (D) Stone delivered with the integrated balloon-basket catheter.

Stones that become lodged in the intramural ureter during an attempt at basket extraction can be managed with ureteral meatotomy. 3 The meatotomy can be performed with endoscopic scissors, urethrotome, or Collings knife, though we routinely employ the latter. In these cases, control of the stone is maintained by gentle traction on the basket while the cystoscope is withdrawn and replaced with a 24-Fr resectoscope fitted with the Collings knife. Utilizing the cutting current only, the meatotomy is performed by incision of the intramural ureter beginning distally at the orifice and cutting proximally on the anteromedial aspect of the intramural ureter directly onto the stone. The incision is carried only as far proximally as is necessary to deliver the stone into the bladder ( Fig. 110-6) When the meatotomy has extended more than a few millimeters from the orifice, an internal stent should be left indwelling for at least 10 days.

FIG. 110-6. (A) For stones impacted in the intramural ureter during basket extraction, ureteral meatotomy is performed. Our preference is to use a Collings knife. Using a cutting current only, the meatotomy is performed by anteromedial incision of the intramural ureter directly over the stone. The incision is begun at the ureteral orifice and carried cephalad only as far as necessary to extract the stone. (B) Fluoroscopic control of ureteral meatotomy performed directly over a stone entrapped in a basket in the intramural ureter.

At times, it is impossible to engage an impacted intramural stone in a basket. In such cases, a meatotomy is performed without basket control as long as the stone is clearly visible bulging within the intramural ureter. A meatotomy is also utilized for management of stones in “adult-type” ureteroceles, though the incision should be oriented more laterally and horizontally from the orifice to prevent reflux as a sequela.

OUTCOMES
Complications Like any intervention, stone basket extraction can be associated with complications, the incidence of which is reported to range from 2% to 10%. frequency and severity of these problems can be minimized, and the success rate of basket extraction increased, with proper patient selection.
4,6

However, the

Perhaps the most common complication of basket extraction is ureteral perforation. This problem can be minimized by careful fluoroscopic monitoring during passage of the basket. Specifically, the course of the basket should be seen to follow that of the ureter. If there is any question that the basket has gone submucosally next to the stone or even perforated the ureter, the basket should not be opened. Rather, the closed basket should be withdrawn and contrast injected retrograde under fluoroscopic control. If, in fact, there is extravasation, it almost always results from perforation below or next to the stone, rather than above it. The safest approach is to attempt retrograde stent placement over a hydrophilic glidewire. If the stone cannot be bypassed in this manner, the patient can be carefully monitored assuming there is no associated infection or sepsis. If the clinical condition does warrant further emergent intervention, percutaneous nephrostomy drainage should be instituted. The most significant risk of basket extraction is ureteral entrapment in the extravesical ureter. 4 In the past, the only options for management of this problem were to allow the basket and stone to be left in place for up to 24 hours, hoping for spontaneous resolution, or to proceed with immediate open ureterolithotomy for removal of the stone and basket. More recently, two other techniques have become available for management of this problem. One of these is ESWL of the stone in the basket. 5 Extracorporeal fragmentation then allows withdrawal of the basket and the fragments. The other alternative is to pass a small-caliber ureteroscope directly into the basket to allow intracorporeal fragmentation of the stone under direct vision. Both of these modalities have been used successfully in our practice. Results When patients are properly selected, cystoscopic stone basket extraction performed under fluoroscopic control has proven to be a reliable procedure that provides rapid and safe resolution of distal ureteral calculi. Twenty years ago, Furlow and Bucchiere reviewed a 15-year Mayo Clinic experience with ureteral calculi. 6 Transurethral manipulation as the initial procedure proved successful in 588 (89%) of 661 patients. More recently, Morse and Resnick, reviewing their experience managing patients with ureteral calculi in the era of advanced technology, reported successful basket extraction of distal stones in 79% of patients. Ureteroscopy performed for failure of basket extraction proved an excellent salvage procedure that proved successful in 90% of patients in that setting. 7 At our center, where ESWL became available in 1986, 371 patients have since required intervention for distal ureteral calculi and a significant number of these patients were referred only after a primary attempt at intervention had already failed. During the last 10 years, 87 (23%) of these patients with distal calculi requiring intervention have been treated with ESWL. The indications for this have generally been large stone size (more than 1 cm), associated upper tract stones requiring ESWL (performed simultaneously), or failed attempts at retrograde access. Of 281 patients treated with retrograde manipulation, ureteroscopy has been used in 92 (33%), most of whom required intracorporeal lithotripsy. Basket extraction, with or without ureteral meatotomy, has been used in 165 (59%) patients. A meatotomy alone was used for 24 (8.5%) patients. Open surgical lithotomy has been performed in only 3 (0.8%) patients undergoing intervention for distal ureteral calculi during this time, and all of those patients were treated before 1988. As such, even in this era of advanced technology, stone basket extraction performed under fluoroscopic guidance can be a significant part of the urologic armamentarium for managing these patients. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Anderson KR, Keetch DW, Albala DM, et al. Optimal therapy for the distal ureteral stone: extracorporeal shockwave lithotripsy versus ureteroscopy. J Urol 1994;152:62. Clark JY, Thompson IM, Optenberg SA. Economic impact of urolithiasis in the United States. J Urol 1995;154:2020. Donovan MG, Heggarty J, Fitzpatrick JM, Butler M. Ureteral meatotomy by Sachse urethrotome in the management of lower ureteral stones. J Urol 1987;138:566. Drach GW. Stone manipulation: modern usage and occasional mishaps. Urology 1978;12:286. Durano AC Jr, Hanosh JJ. A new alternative treatment for entrapped stone basket in the distal ureter. J Urol 1988;139:116. Furlow WL, Bucchiere JJ. The surgical fate of ureteral calculi: review of Mayo Clinic experience. J Urol 1976;116:559. Morse RM, Resnick MI. Ureteral calculi: natural history and treatment in an era of advanced technology. J Urol 1991;145:263. Rutner AB. Ureteral balloon dilatation and stone basketing. Urology 1984;23:44. Sandegarde. Prognosis of stone in ureter. Acta Chir Scand 1956;(Suppl):219,1. Ueno A, Kawamura T, Ogawa A, Takayasu H. Relation of spontaneous passage of ureteral calculi to size. Urology 1977;10:544.

Chapter 111 Cystoscopic Treatment of Bladder Tumors Glenn’s Urologic Surgery

Chapter 111 Cystoscopic Treatment of Bladder Tumors
Keith A. Harmon and Michael J. Droller

K. A. Harmon and M. J. Droller: Department of Urology, The Mount Sinai Medical Center, New York, New York 10029.

Diagnosis Indication for Surgery Alternative Therapy Surgical Technique Special Cases of Transurethral Treatment Outcomes Complications Results Chapter References

Bladder cancer comprises a spectrum of diatheses ranging from superficial, mucosally confined tumors that pose minimal risk of progression to nonresectable cancers that are first detected when regional therapies are no longer likely to prove effective. 1,5,11 More than 50,000 new cases of bladder cancer are diagnosed annually in the United States, 75% of which are found to be superficial. Although recurrence is likely to occur after local resection in 50% to 70% of instances (notwithstanding the adjunctive use of intravesical therapies), the risk for progression appears to be no greater than 20% to 25%. On the other hand, the majority of patients diagnosed with muscle infiltrative cancer present with this stage of the disease as their initial clinical presentation. When patients present initially with superficial disease, those features of their disease that suggest an increased likelihood of progression include high grade, infiltration of the lamina propria, and concomitant carcinoma in situ. Although response rates for these various lesions, as the result of transurethral resection either alone or in combination with intravesical therapies, have been reported to be as high as 70%, the majority of these patients will experience recurrence and a substantial number will experience progression, albeit within the context of an indeterminate time course. 1 Bladder cancer has been associated with exposure to a variety of environmental factors. This association was first described for heterocyclic amines, with specific industries creating particularly high risk (dyestuffs, solvents, rubber manufacturing, furniture finishing, etc.). The most consistent and pervasive factor associated with the development of bladder cancer has been cigarette smoking. Host factors may be equally important in the development of bladder cancer. For example, individuals characterized as slow acetylators have been more frequently associated with the development of bladder cancer than have individuals characterized as fast acetylators. Presumably, acetylation is an important mechanism whereby carcinogens excreted in the urine may be detoxified such that their potential carcinogenic effect on the urothelium is mitigated or eliminated. Whether a genetic or familial disposition to the development of bladder cancer exists is unclear. Those features that suggest the development of bladder cancer in a familial clustering have been complicated by concomitant environmental carcinogenic exposures. Bladder cancer has been used as a paradigm for mechanisms of carcinogenesis traditionally comprised of events termed “initiation” and “promotion.” However, little is known about which specific events lead to mucosally confined tumors that do not present a life-threatening risk to the host and those that produce tumors with the ability to infiltrate the wall of the bladder and metastasize. Such distinctions have given rise to staging systems to characterize the nature of a particular disease and determine appropriate therapies. Transurethral resection of a bladder cancer is important not only to diagnose and assess the extent of disease, with the possibility of determining its prognosis on the basis of the stage elucidated, but also to ascertain what additional therapy may be necessary in its management. Moreover, transurethral resection may in itself be definitive treatment for a particular tumor diathesis. Although recurrence may be common, repeated transurethral resection may provide effective definitive therapy within the temporal context of the disease without the need for additional aggressive treatment. Effective transurethral resection requires careful preoperative and intraoperative planning. In addition, such treatment must not compromise subsequent therapies for management of the disease as it may recur in the same or more progressive forms.

DIAGNOSIS
Most patients with bladder cancer present with either gross or microscopic hematuria, which may not be accompanied by other symptoms. The initial urologic workup for hematuria should include microbiological culture of a midstream voided urine. If the culture proves positive for urinary tract infection, appropriate antibiotic therapy should be instituted followed by repeat urinalysis and culture. If gross or microscopic hematuria persists or if the culture shows no growth, radiologic evaluation of the upper tracts and cystoscopy is indicated. Radiologic evaluation of the upper tracts by intravenous urography (IVU) and cystoscopy should proceed. Although sonography of the upper tract has been adopted by many as the first imaging study in patients with hematuria, intravenous urography remains the most useful and definitive initial radiologic modality for evaluation for transitional cell neoplasia of the upper tract. Obstruction of the upper tract, suggesting either ureteral cancer or muscle-invasive transitional cell cancer of the bladder with involvement of the ureteral orifice, may then be assessed by computed tomographic (CT) scanning of the pelvis and bladder to define the extent of disease. CT scanning is also useful in determining multiplicity of disease and thickening of the bladder wall. In patients with suspected muscle-invasive disease as assessed cystoscopically (see below), the CT scan may be obtained prior to transurethral surgery to allow assessment of the extent of the cancer without misleading appearances created by the inflammatory response associated with resection and postoperative healing. Voided urine for cytopathologic analysis is also important in the preoperative diagnosis and assessment of a bladder cancer because it effectively samples the entire urothelium for any cellular abnormality. 10 Exfoliated cells from normal and neoplastic urothelium can be readily identified in voided urine samples. Although urine cytology is highly sensitive for the detection of high-grade tumors and carcinoma in situ (sensitivity 90% to 100%), urine cytology falls short in regard to the detection of low-grade tumors (sensitivity approximately 25%). 7 The voided urine specimen should be obtained before instrumentation to avoid introduction of artifact. Flow cytometry, which measures the increased DNA content of neoplastic cells, provides a modest improvement in sensitivity and specificity over conventional cytology. However, flow cytometry requires bladder washing specimens rather than merely voided urine specimens. 2 Qualitative fluorometric image analysis may enhance the diagnostic sensitivity and specificity of ploidy determination by selecting cells for analysis. Cystoscopy is the most important step in the diagnosis and preoperative assessment of bladder tumors. It is used not only to identify the presence of a bladder cancer but also to characterize its appearance (whether papillary or nodular), its location, multiplicity of lesions, and the character of the intervening urothelium. If the presence of transitional cell cancer of the bladder is established by radiologic or urinary studies, preoperative cystoscopy may not be necessary. In these instances, preoperative instrumentation may introduce infection or trauma that may compromise subsequent visualization and assessment of the bladder mucosa at the time of resection and obscure areas of possible carcinoma in situ or urothelial atypia ( Fig. 111-1).

FIG. 111-1. The relation of invasive infiltration and grade of malignancy is established jointly by the American Joint Committee on Cancer Staging (AJC) and the International Union Against Cancer (UICC).

INDICATION FOR SURGERY
Identification of a bladder cancer by initial radiologic evaluation, urinary cytology, and cystoscopy will indicate the need for transurethral resection. This will confirm the diagnosis histopathologically and provide an indication of the extent of the disease, in terms of both depth of penetration of the bladder wall and extent of involvement of adjacent and distant areas of the urothelium in the bladder and prostatic urethra.

ALTERNATIVE THERAPY
Endoscopic visualization of a bladder cancer in most instances will require transurethral resection for both diagnostic and staging purposes. If the condition of the patient is such that the risk of anesthesia is too great to permit formal transurethral resection and the cancer appears to be papillary and superficial, light electrocautery or laser treatment may be both appropriate and sufficient, with subsequent monitoring by repeat endoscopy and voided cytology done at appropriate intervals. Intravesical instillation of chemo- or immunotherapeutic agents is unlikely to be completely effective in eradicating a superficial tumor. However, such treatment may be effective prophylactically following the use of electrocautery or laser initially to eradicate the cancer. Tissue specimens for pathologic evaluation in these instances may be obtainable by cold cup biopsy. The absence of a grossly visible lesion but the existence of a positive urinary cytology may allow the presumptive diagnosis of carcinoma in situ to be made. Assuming that upper tract disease has been excluded, the use of bacillus Calmette-Guérin (BCG) by intravesical instillation may permit control of this form of cancer or lead to its eradication. If the cancer is observed to be nodular and appears to be deeply invasive, external radiation therapy may be provided to control the disease. This in itself, however, may introduce severe irritative symptoms and bleeding. The use of systemic chemotherapy would not be indicated in circumstances of deeply invasive bladder cancer in a patient who is medically unable to tolerate transurethral resection, simply because of the potential severity of toxicity that the systemic chemotherapy might create. In either case, most patients will experience recurrence, and a majority may ultimately develop metastatic disease.

SURGICAL TECHNIQUE
Antiembolism stockings or sequential compression devices can be placed on the lower extremities to decrease the risk of deep venous thrombosis. An intravenous antibiotic with broad-spectrum coverage of gram-positive and gram-negative bacteria can be administered for surgical prophylaxis. After satisfactory spinal or inhalation anesthesia is induced, the patient is placed in the lithotomy position with the perineum perpendicular to the plane of the table. Bimanual examination is then performed to evaluate the presence of a pelvic mass and its possible fixation. Palpation of a mass generally indicates deep muscle invasion, whereas fixation generally indicates extensive soft tissue involvement by the cancer with probable noncurability. The standard 23-Fr cystoscope with the 30 degree lens is inserted under direct vision through the urethra into the bladder. This permits assessment of the urethral mucosa as well as of the bladder mucosa before either has been traumatized by the cystoscope. As soon as the cystoscope has been introduced into the bladder, the inflow of irrigating fluid (generally sterile water) is discontinued and the bladder mucosa is inspected without further inflow and without draining the bladder. This permits examination of the mucosa and avoids potential trauma to the wall of the bladder by the beak of the cystoscope. The location of the primary lesion and any other lesions, the gross architecture of these lesions, and any areas of erythema should be noted. The appearance of the effluxing urine from each ureteral orifice should also be assessed. The dome of the bladder and that portion of the anterior wall adjacent to the bladder neck can be visualized with the 30 degree lens by manual compression on the lower abdomen with the bladder only partially full. The 30-degree lens is then replaced by the 70-degree lens, so that the mucosa can be examined further. The bladder can then be drained and a portion of the urine sent for cytologic evaluation. If indicated, barbotage can also be performed to secure a specimen for flow cytometry. Although barbotage may cause trauma to the bladder wall and result in artifactual areas of erythema that may be misinterpreted as carcinoma in situ, visual assessment before bladder drainage will permit the distinction between trauma and cancer to be made. A cold cup biopsy forceps is then introduced into the cystoscope. Using the 30-degree lens for visualization, cold cup biopsy specimens are obtained of any areas of erythema identified on initial mucosal evaluation as well as areas at the margin of any visible cancers. Although some have suggested taking biopsy specimens cephalad to each ureteral orifice and at the trigone, 8 this has not had a particularly high yield in detecting either epithelial atypia or carcinoma in situ at sites that otherwise appear normal. On the other hand, biopsy specimens of the prostatic urethra or at the margin of the visible tumor may be productive because these areas may be involved by carcinoma in situ. A Bugbee electrode can be used to cauterize the base of these biopsy specimens. Their locations should be indicated on the surgical pathology form for future reference. The 24-Fr resectoscope is then inserted into the bladder to resect any visible exophytic lesions. The pure cutting current should be used to avoid cautery artifact. For superficial papillary lesions, the loop should be placed initially just adjacent to the superficial portion of the lesion. The papillary fronds should then be resected with long, smooth strokes (Fig. 111-2). Fronds that have been resected may stick to the main body of the cancer. They need to be pushed away from the tumor with the cold loop so that assessment of the amount of tissue resected can be made and cauterization of centrally located blood vessels accomplished.

FIG. 111-2. Transurethral resection of bladder tumor. (A) Tumor of posterior bladder wall. (B) Resection initiated at the posterior margin of tumor. (C) Progressive

exposure of the underlying bladder muscle. (D) Complete removal of visible tumor and securing of portion of muscularis for staging. (E) Superficial resection and fulguration of the apparently normal surrounding mucosa. (F) Complete fulguration of the base of the tumor may be necessary.

Generally, cauterization is not necessary until the main blood vessel of the papillary tumor is encountered. It may be difficult to visualize this vessel until the bulk of the tumor has been resected. Therefore, resection of the bulk of the papillary lesion is rapidly pursued until the tumor has been resected flush with the surrounding mucosa. At this point, any bleeding can be controlled easily by cauterization of visible vessels. The resected specimen is then evacuated from the bladder with the Ellik evacuator and sent separately for pathologic evaluation. A deep resection specimen can then be obtained at the base of the primary tumor. Preoperative assessment by CT scanning or intraoperative assessment of the appearance of the tumor should provide some indication as to the thickness of the bladder wall and of the likelihood that a tumor has infiltrated the muscularis. Generally, superficial tumors have a thin bladder wall, especially if they are confined to the mucosa ( Fig. 111-3A). Therefore, deep resection of such tumors is unnecessary and should be avoided. On the other hand, it is important for both diagnostic and therapeutic purposes to document infiltration of the lamina propria. 1 Resection of the base of superficial tumors, therefore, should include a portion of the bladder wall, preferably with a portion of the superficial muscularis, and should be done with the cutting current using a light stroke to avoid perforation.

FIG. 111-3. CT scan of bladder tumors. (A) Superficial tumor. CT scan demonstrates thin bladder wall with normal configuration and a sharp angle between tumor and bladder wall. (B) Invasive tumor. CT scan generally reveals a thickened bladder wall beneath and adjacent to a more nodular-appearing tumor.

A nodular tumor, or one found on prior CT scanning to be associated with a thickened bladder wall, can be resected with less fear of perforating the bladder wall ( Fig. 111-3B). Deep resection is possible in these instances and is often accompanied by less bleeding than resection of a more superficial cancer. It is unlikely that resection by itself can cure a cancer that is deeply infiltrative. Removal of as much bulk as possible, however, may enhance the effect of any other therapeutic modalities used later (radiation therapy, systemic chemo-therapy). After the primary lesion has been resected, the margins of the tumor should be cauterized. Other sites of resection or of cold cup biopsy should be examined for hemostasis. A 22-Fr Foley catheter with a 5-ml balloon is then inserted for continuous bladder drainage. Generally, continuous bladder irrigation is unnecessary if hemostasis has been obtained by electrocautery. Special Cases of Transurethral Treatment Transitional Cell Cancer in the Prostatic Urethra Papillary transitional cell cancers can sometimes be seen in the prostatic urethra. In this instance, it is imperative to resect the papillary tumor with the understanding that such tumors are likely to be superficial, are not likely to involve the prostatic ducts, and are not likely to have infiltrated the prostatic stroma. These generally are seen in the setting of previous or simultaneous multifocal superficial tumors in the bladder. In patients who have not had previous prostatectomy, formal transurethral prostatectomy is neither necessary nor indicated. On the other hand, when multiplicity of disease or rapid recurrence with evidence of disease in the prostatic urethra indicates the possible need for intravesical therapy after bladder tumor excision, resection of the prostate to create an open prostatic fossa may be indicated. In these instances, it may be appropriate to resect first the lesions in the bladder and then superficially the lesions in the prostatic urethra, then to provide initial intravesical treatment to eradicate potentially viable cells in the bladder, and then to proceed with transurethral resection of the prostate with subsequent further intravesical instillation of therapeutic agents. By and large, however, superficial cancer involvement of the prostatic urethra will not require extensive transurethral resection of the prostate for further management. Transitional Cell Cancer in a Bladder Diverticulum Occasionally, papillary transitional cell cancer is found in a bladder diverticulum. The use of cold cup biopsy forceps for sampling of the base of such tumors is an appropriate conservative approach to take because formal resection of the base of the lesion may result in perforation of the diverticulum. CT scanning before resection is important in ascertaining the thickness of the diverticulum wall and determining whether the malignancy has penetrated that wall. In instances in which a solid tumor is seen to fill the diverticulum, it is likely that the tumor has already penetrated the diverticulum wall. Transurethral resection is unlikely to provide a cure. Confirmation of the diagnosis by cold cup biopsy is indicated in these instances. Deep resection is contraindicated because it is unlikely to cure the patient and may jeopardize the patient's status for subsequent treatment. As with superficial tumors, preoperative assessment by CT scanning is indicated in these situations. When there are areas of erythema within a diverticulum, cold cup biopsies of these areas are indicated to document the presence either of neoplasia (carcinoma in situ) or of nonneoplastic epithelial atypia. Extensive cauterization of the erythematous areas after biopsy specimens have been taken may assist in the eradication of tumor cells from the diverticulum and may also assist in the efficacy of subsequent intravesical instillations of therapeutic agents to eradicate persistent cells and prevent recurrence of disease. Cancer Involving the Ureteral Orifice When superficial cancer obscures a ureteral orifice, it should be resected superficially at first, as described earlier. As one approaches the base of the tumor, it becomes easier to determine whether the orifice can be visualized and whether the stalk of the tumor is any distance from the orifice or actually originates within it; the situation is generally the former. In these instances, the stalk can be resected without concern for the orifice. Occasionally, however, the orifice is occluded because the papillary tumor actually originates within the ureter and has extended outside the orifice to occupy an intravesical location. In these instances, the ureter may have been found to be obstructed on preliminary intravenous pyelography. Resection of the papillary lesion flush with the ureteral orifice can be accomplished, and deep resection of the orifice and intramural ureter may demonstrate muscle fibers of the bladder wall surrounding a protruding papillary lesion that occupies the lumen of the ureter. When the origin of the tumor is just within the ureteral orifice, further resection may eradicate disease completely. Resection should be done with the cutting current only because use of the cautery is likely to induce fibrosis and possible subjacent ureteral stricture formation. When the ureter is occluded by a muscle infiltrative transitional cell cancer, deep resection should be done in an attempt to remove the obstructive tissue and permit

recovery of renal function on that side. Transurethral resection is unlikely to cure the patient, however, and more definitive approaches are needed. Laser Ablation of Tumors Lasers have been advocated as a means of cauterizing superficial transitional cell cancers without the need for anesthesia. Because this approach does not permit a suitable specimen to be taken for pathologic evaluation, a biopsy specimen is generally taken with cup biopsy forceps before laser fulguration. This modality is particularly useful in the treatment of papillary superficial bladder cancers less than 1 cm in diameter. The tumor can then be cauterized superficially and its base also cauterized on the assumption that the depth of penetration of the laser energy eliminates any cancer cells that may have penetrated the lamina propria. Use of lasers is limited by the need for the probe to be positioned immediately adjacent to the visible lesion and the loss of energy that results when the probe is deflected in attempting to reach less accessible areas of the bladder. 12 No evidence suggests that lasers are more effective than electrocautery in preventing recurrence of superficial transitional cell cancers. Nor is there evidence that lasers are effective in the management of muscle infiltrative bladder cancer. In addition, because direct transmission of laser energy is needed for the satisfactory use of the laser in the fulguration of tumor cells, the use of the laser in the management of diffuse carcinoma in situ is limited. Because of the risk of deep transmission of the laser energy with potential cauterization of extravesical structures, the use of CT scan before laser treatment is advisable. Patients treated with laser should be monitored by the use of urinary cytology. Limitations in the use of laser therapy in dealing with diffuse disease has led to the search for light-activated agents that might sensitize cancer cells to the photodestructive effects of laser therapy. Hematoporphyrin derivatives and psoralens have been found to be effective in sensitizing tumor cells at specific wavelengths.3 The major side effect that has severely limited the clinical use of these agents is epithelial sensitization, which has made such patients vulnerable to burns from direct exposure to sunlight when the agents are given systemically. As such, a search for agents that are selectively taken up by cancer cells and that might be used in conjunction with laser phototherapy without these side effects has received strong impetus. Electrovaporization When the bulk of superficial cancer is too extensive to permit satisfactory removal by conventional trans-urethral resection alone, transurethral electrovaporization may enable the surgeon to remove the diffuse disease more rapidly and with minimal bleeding. The VaporTrode (Circon) vaporizes, desiccates, and coagulates the bladder tumor using a higher cutting current (200 W) than is used for traditional resection. Several caveats should be considered in using this approach in the transurethral management of transitional cell cancer. The first is that the VaporTrode has been formally studied only for the treatment of benign prostatic hyperplasia but not for the management of bladder cancer. 9 Therefore, successful use of this approach for bladder tumors is at present only anecdotal. The second is that electrovaporization provides no surgical specimen. Cold cup biopsy or resection should therefore be done in conjunction with this modality. Although transurethral electrovaporization may remove tumor more rapidly than conventional transurethral resection, all tumor cells may not be eradicated. The use of intravesical chemotherapy may be indicated in patients in whom the absence of muscle invasive disease has been confirmed pathologically but in whom rapidly recurrent or diffuse “superficial” disease is present. Adjuvant Intravesical Therapy An adjunct in the transurethral approach to the management of bladder cancer is the use of therapeutic agents by intravesical instillation. Objectives of such therapy are (a) treatment of a malignant diathesis when the bulk of superficial cancer is either too great or too extensive to permit satisfactory removal by transurethral resection alone, and (b) prophylaxis to prevent or delay recurrence and possibly thereby to prevent progression. Several caveats should be considered in using this approach in the transurethral management of transitional cell cancer. The first is that this approach is effective only therapeutically or prophylactically in the management of superficial disease. Thus, carcinoma in situ, stage TA, and stage T1 disease have been satisfactorily treated with a variety of intravesical agents. These agents have no effect on deeply invasive disease, notwithstanding rare anecdotal reports to the contrary. The second caveat is that intravesical treatments may be used in conjunction with transurethral resection in the management of superficial bladder cancer. In the setting of carcinoma in situ in the presence or absence of superficial papillary transitional cell cancers, for example, intravesical treatments may be used together with resection of any visible papillary disease to eradicate all malignant cells from the bladder. Similarly, intravesical treatments may be used in conjunction with transurethral resection of lamina propria invasive cancers with the rationale that transurethral resection may not have completely eliminated all malignant cells that may have penetrated the lamina propria and been left behind. The presence of diffuse papillary disease can be managed successfully both with extensive transurethral resection and with intravesical instillation of therapeutic agents when transurethral resection is unlikely to have removed all papillary tumors from the bladder or when the bulk of disease is such that transurethral resection is not feasible. The third caveat is that the use of these agents in a patient who may be at particular risk for the development of progressive disease requires careful surveillance for the persistence of positive cells during and after treatment. This permits more aggressive intervention in the event that conservative management by this approach proves unsuccessful. 6 The most common situation in which this may occur is when lamina propria invasive cancer is present in the setting of diffuse carcinoma in situ. lntravesical agents can be used immediately after transurethral resection of one or several superficial tumors to prevent tumor recurrence by implantation of tumor cells. The use of only a single dose of a particular agent has been shown to prevent rapid recurrence at sites (such as at the bladder dome) that may be less accessible to transurethral resection. 4 The use of intravesical agents to prevent rapid recurrence of superficial disease in the context of previous frequent recurrence is probably the most common application of this approach. The fourth caveat is that intravesical agents may ultimately not prevent recurrence or progression notwithstanding their apparent ability to delay such events. Therefore, rigorous adherence to careful surveillance by cystoscopy and urinary cytology is mandatory in following all patients through the years, and especially when high-grade lamina propria invasion or carcinoma in situ has been diagnosed. The most commonly used agents for intravesical instillation have been thiotepa, mitomycin C, adriamycin, and various strains of BCG. The first three agents have been found to exert their effects by direct cytotoxic action against bladder cancer cells. The mechanism of action of BCG appears to be immunologic. Generally, six to eight weekly instillations are used. All agents may have unique side effects that need to be taken into account in considering their use in the individual patient.

OUTCOMES
Complications The most common risk of transurethral resection of superficial bladder cancer is bladder perforation. This occurs with attempts to resect aggressively when the bladder wall is very thin. It may also occur when a tumor appears at the lateral wall and resection is done with the bladder distended such that the electric current stimulates the obturator nerve to produce an adductor reflex. This can then lead to penetration of the rigid resectoscope sheath through the wall of the bladder. Recognition of perforation can be made by the appearance of perivesical fat at the site of penetration and appreciation of a decreased return of irrigation fluid on attempted evacuation of fluid from the bladder. Perforation is generally not accompanied by excessive amounts of bleeding. Depending on the site of perforation, extravasation may occur retroperitoneally in the pelvis or intraperitoneally. If the latter, concern regarding the potential for trauma to the bowel may indicate the need for abdominal exploration. If extravasation has only occurred retroperitoneally, the amount of extravasation may indicate the need for stab wound placement of a perivesical drain. More often, extravasated fluid will be absorbed in due time and a urethral catheter is generally sufficient to decompress the bladder and allow spontaneous healing of the site of perforation as well as of the resection sites. Suprapubic catheter diversion is to be discouraged because of the possible persistence of tumor cells that may seed an extravesical tract. An additional complication is bleeding. This may occur either when adequate hemostasis has not been obtained or when the patient strains the lower abdomen

postoperatively on recovery from anesthesia to open up vessels that have been cauterized. Venous bleeding generally only requires evacuation of clots and institution of free drainage so that the bladder remains decompressed and spontaneous hemostasis can occur. If the bleeding is arterial, it may be necessary to cauterize the bleeding vessel. Critical to either approach is effective clot evacuation. Occasionally, continuous bladder irrigation can be useful in preventing clot formation and allowing satisfactory decompression of the bladder and healing. Results Transurethral resection of superficial bladder tumors (other than carcinoma in situ) is generally effective in eradicating the disease from the bladder. However, because the presence of bladder cancer often represents a clonogenic field change, other sites at which tumor cells were endoscopically invisible may account for recurrence in 50% to 75% of instances. Multiplicity of disease, carcinoma in situ, or lamina propria invasion should prompt the use of adjunctive intravesical agents to prevent recurrence and, in the case of the latter two instances, to possibly prevent progression. Several studies have suggested successful prevention of recurrence in the setting of superficial disease. Over time, however, the incidence of decreased recurrence in comparison to careful transurethral resection alone is at best 10% to 15%. Similar constraints have been reported in the apparent efficacy of intravesical chemotherapy or immunotherapy (BCG) in preventing progression. However, minimal if any benefit has been seen in the long term in patients with potentially aggressive disease (e.g., high-grade tumors with lamina propria invasion in the setting of diffuse carcinoma in situ). Muscle infiltrative disease is generally not cured by transurethral resection alone, and radical surgery is generally needed. The only occasional exception is when cancer in a papillary configuration infiltrates the muscularis only superficially. In patients who are not candidates for cystectomy, definitive radiation therapy in combination with further resection and systemic chemotherapy may delay recurrent disease. Ultimately, the majority of such patients will fail this form of more conservative treatment and will develop metastatic disease. At least 50% of patients found to have muscle infiltrative disease on initial clinical presentation will be found to develop metastatic disease within 2 years of initial diagnosis. Even prompt cystectomy with or without adjunctive radiation is not curative in these patients. The majority fail with distant metastatic disease although the risk of pelvic recurrence is also substantial. In those patients who are not candidates for cystectomy, palliation can be achieved with extensive transurethral resection in conjunction with radiation therapy. Ultimately, however, these patients will succumb to the cancer. The transurethral approach in the management of transitional cell cancer has a twofold objective: one is to confirm the diagnosis of bladder cancer and provide pathologists with sufficient material not only to diagnose the presence of disease but also to establish the stage of disease and the likelihood of recurrence or progression. The second is to eradicate the malignancy from the bladder. Careful preoperative assessment includes radiologic and cytologic methods. This may assist in establishing the extent of disease and planning an appropriate and safe approach to its management. CHAPTER REFERENCES
1. Anderstrom C, Johansson S, Nilsson S. The significance of lamina propria invasion on the prognosis of patients with bladder tumors. J Urol 1980;124:23. 2. Badlament RA, Fair WR, Whitmore WF Jr, Melamed MR. The relative value of cytometry and cytology in the management of bladder cancer: the Memorial Sloan-Kettering Cancer Center experience. Semin Urol 1988;6(1):22. 3. Benson RC Jr. Integral photoradiation therapy of multifocal bladder tumors. Eur J Urol 1986;12 (Suppl 1):47. 4. Burnand KG, Boyd PJR, Mayo MG, Shuttleworth KED, Lloyd-Davies RW. Single dose intravesical thiotepa as an adjuvant to cystodiathermy in the treatment of transitional cell bladder carcinoma. Br J Urol 1976;48:55. 5. Droller MJ. Approach to the management of minimally invasive transitional cell cancer of the bladder. In: Resnick MI, Kursh ED, eds. Current therapy in genitourinary surgery. Toronto: BC Decker, 1987;56–63. 6. Droller MJ, Walsh PC. Intensive intravesical chemotherapy in the treatment of flat carcinoma-in-situ: is it safe? J Urol 1985;134:1115. 7. Gamarra MC, Zein T. Cytologic spectrum of bladder cancer. Urology 1984;23(3):23. 8. Heney NM, Daly J, Prout GR Jr, Nieh PT, Heaney JA, Trebeck NE. Biopsy of apparently normal urothelium in patients with bladder carcinoma. J Urol 1978;120:559. 9. Kaplan SA, Te AE. Transurethral electrovaporization of the prostate: a novel method for treating men with beningn prostatic hyperplasia. Urology 1995;45(4):566. 10. Murphy WM, Crabtree WN, Jukkola AF, Soloway MS. Diagnostic value of urine versus bladder washing in patients with bladder cancer. J Urol 1981;126:320. 11. Prout GR Jr, Griffin PP, Shipley WU. Bladder carcinoma as a systemic disease. Cancer 1979;43:2532. 12. Smith JA Jr, Dixon JA. Tissue effects of lasers in the genitourinary system. In: Smith JA Jr, ed. Lasers in urologic surgery. Chicago: Year Book, 1985;16–27.

Chapter 112 Transurethral Resection, Incision, and Ablation of the Prostate Glenn’s Urologic Surgery

Chapter 112 Transurethral Resection, Incision, and Ablation of the Prostate
Richard P. Santarosa, Alexis E. Te, and Steven A. Kaplan

R. P. Santarosa, A. E. Te, and S. A. Kaplan: Department of Urology, Columbia Presbyterian Medical Center, New York, New York 10032.

Diagnosis Indications for Surgery Alternative Procedures Surgical Technique Preoperative Considerations Postoperative Care Outcomes Complications Results Chapter References

Transurethral resection of the prostate (TURP) is a common procedure performed by urologists for the treatment of obstructive uropathy secondary to benign prostatic hypertrophy (BPH). 4 It is a second generation procedure after the older open prostatectomy. The less invasive transurethral procedure has become the preferred approach with the advent of effective and efficient instrumentation to remove the obstructing prostatic adenoma. This chapter will describe the surgical indications, preoperative evaluation, and operative technique for the standard loop resection TURP, as well as the intraoperative problems which may be encountered. Recent modifications and alternative techniques to the standard TURP for the relief of prostatic obstruction, including transurethral incision of the prostate (TUIP), electrovaporization of the prostate and laser prostatectomy will be reviewed. Finally, the postoperative care and complications involving the standard TURP and the impact of newer technologies to these issues will be discussed.

DIAGNOSIS
The usual preoperative workup for TURP includes a complete blood count, chemistry profile, chest radiograph, and electrocardiogram. Imaging of the upper urinary tracts is necessary only in the presence of hematuria, infection, or stones. In a recent review, patients with renal insufficiency were noted to be at higher risk for complications after TURP. Therefore, it is very important to evaluate the cardiac, pulmonary, and renal function of patients prior to surgery. Patients should be evaluated for and free of urinary tract infection prior to any transurethral surgery. In the American Urological Association (AUA) cooperative study, the most common cause of death following TURP was sepsis. 10 Any patient in urinary retention with an indwelling Foley at the time of surgery should be treated preoperatively with antibiotics, even in the absence of evidence of acute infection. Coagulation parameters should be checked prior to surgery and aspirin should be discontinued for 7 to 10 days prior to surgery. Preoperative cystoscopy allows evaluation of size, length, and configuration of the prostate and to determine whether it can be adequately resected transurethrally. Our rule of thumb is that a prostate of approximately 80 to 100 g (or effectively 90 minutes resection time) can usually be resected with a standard loop resectoscope. A larger gland would be better treated in an open fashion. Similarly, a very small gland, or a gland that is anatomically configured such that the point of obstruction is primarily at the bladder neck, is probably best treated with transurethral incision rather than TURP. Cystoscopy is also useful for ruling out the presence of bladder stones and bladder tumors. A small tumor can be resected safely at the time of TURP, but an extensive tumor that is suspicious for being invasive should be biopsied and prostatectomy deferred until it is determined that the patient will not require a more extensive operation, such as cystoprostatectomy. 2

INDICATIONS FOR SURGERY
The most common indication for TURP is the symptomatic relief of obstructive voiding symptoms. Generally, patients with the worst voiding symptoms report the greatest postoperative success after TURP. 2 Following TURP, well over 80% of patients experience improvement in symptoms and urinary flow rates. 12 The decision to select surgical therapy for obstructive voiding symptoms, however, is a subjective decision based on the symptoms' impact on the patient's quality of life and impressions about what constitutes mild or severe symptoms. In an attempt to standardize therapy, the Agency for Health Care Policy and Research (AHCPR) has created clinical guidelines to help urologists and patients decide on the appropriate evaluation and treatment for voiding symptoms due to BPH (AHCPR/BPH guidelines). 9 In addition to these relative indications, there are also absolute indications for surgical therapy. Most urologists would agree that men with refractory urinary retention, recurrent infection, recurrent gross bleeding, bladder damage or calculi, hydronephrosis, or renal damage due to BPH are clear indications for surgical therapy in the appropriate surgical patient.

ALTERNATIVE PROCEDURES
Alternatives to surgical resection, incision, or ablation of the prostate include observation and medical therapy including alpha blockers and 5a-reductase inhibitors. Generally, the patients who have mild to moderate symptoms (AHCPR/BPH guidelines) are candidates for conservative or medical therapy, whereas surgery is best for patients failing conservative therapy and/or who have severe symptoms. Other alternatives that may be effective include open prostatectomy ( Chapter 31), thermotherapy (Chapter 136), or urethral stents.

SURGICAL TECHNIQUE
Preoperative Considerations The decision to utilize spinal or general anesthesia is up to the patient and the surgeon. Two studies that evaluated regional versus general anesthesia for TURP found no significant differences in comparative blood loss, complications, and operative mortality. 13 As with any surgical procedure, the patient must be aware of potential complications in order to make an informed decision on whether to undergo surgery. Complications inherent to the transurethral treatment of obstructing prostatic adenoma include bleeding, with the possible need for a transfusion; infection; incontinence; impotence; retrograde ejaculation; and, possibly, mortality. Transurethral Resection of the Prostate The patient is placed in the dorsal lithotomy position, with care taken care not to overextend the hip joints and not to place undue pressure at the calf. We prefer intermittent compression stockings on the lower extremities to prevent development of deep venous thrombosis and an O'Connor drape for manipulation of the prostate during the procedure. Initial placement of the resectoscope sheath can be hampered by a narrow urethra. Usually male urethral sounds are adequate to calibrate the urethra to 30 Fr so as to accommodate the most commonly utilized resectoscope (28 Fr). Occasionally, however, the distal urethra, especially at the fossa navicularis, cannot be made to accommodate this size resectoscope and so an internal urethrotomy must be performed. This can be done via a dorsal meatotomy with a curved blade, as described by Mebust,11 or an Otis urethrotome. Another option if the distal urethra is inadequate to accommodate a resectoscope sheath or if the patient has a penile prosthesis is to perform a perineal urethrostomy (Fig. 112-1).

FIG. 112-1. Perineal urethrotomy. (A) Conger clamp stabilizes skin and underlying tissues. (B) Incision through bulbar muscle. (C) Incision through urethral bulb, exposing sound. (D) Traction sutures exposing lumen. (E) Bulbar muscle layer closed. (F) Skin closed.

The basic resection technique has been performed and described in a number of ways. An important consideration common to all of these techniques is the need to proceed in an orderly, planned fashion. All techniques utilize the principle of resecting ventrally first so that the adenomatous tissue drops down, allowing the surgeon to resect from the top downward rather than from the floor upward. At times, the surgeon will prefer to remove the floor and part of the median lobe first to allow the prostatic chips to fall easily into the bladder and to improve irrigant flow before going on to the ventral portion of the adenoma. Before any significant resection is begun all key anatomic landmarks are identified: the ureteral orifices, the bladder neck, the length and configuration of the prostate, the location of the verumontanum, the location of the external urethral sphincter, and the relationship of these landmarks to the position of the resectoscope and the resection loop (Fig. 112-2). If a large median lobe or very elevated bladder neck inhibits movement of the resectoscope and prevents visualization of these landmarks, we have found it helpful to carefully resect the bladder neck region from the 5 o'clock to the 7 o'clock position. We are careful not to be overzealous in bladder neck resection and to avoid excessive cauterization in this region as this can lead to bladder neck contracture postoperatively.

FIG. 112-2. Ureteral relationships. (A) Frontal view relationship of ureteral orifices to intravesical adenomatous hyperplasia. (B) Lateral view relationship of middle lobe tissue resection with loop to ureteral orifice.

The technique we use has been summarized by Holtgrew in a recent review 4 that is a variation of the description by Nesbit ( Fig. 112-3). The first stage of the resection begins at the anterior region of the bladder neck at the 1 o'clock position. The resectoscope is positioned in the midprostatic fossa and the resection of the proximal portion of the prostate is carried out in a clockwise direction to the 6 o'clock position. The depth of the resection should be far enough down to expose the fibers of the capsule of the prostate around the bladder neck circumferentially. The resection is then continued on the other side from the 11 o'clock position counterclockwise to the 6 o'clock position and hemostasis is obtained. Care must be taken not to resect too deeply in the region of the posterior aspect of the vesical neck as the trigone may easily be undermined at this position, leading to occult extravasation of irrigation fluid outside the bladder. If at the conclusion of this stage of the resection the bladder neck appears to be partially obstructing, an incision of the bladder neck can be done with a Collings knife at the 6 o'clock position to reduce the incidence of postoperative bladder neck contracture, as recommended by Kulb. 7

FIG. 112-3. Transurethral resection of lateral lobes. (A) Resection begins at proximal anterior portion of lateral lobe at about 1 o'clock position. (B) Proximal lateral lobes are resected in a clockwise and counterclockwise manner to surgical capsule. (C) Distal lateral lobes resection is completed in a similar manner to the level of verumontanum.

The next stage is the resection of the bulk of the adenoma in quadrants. The resectoscope is placed just proximal to the verumontanum and the resection is begun at the 12 o'clock position and carried down to the 3 o'clock position taking long, deep swipes with the loop each time. Special care is taken so as not to carry any excursion of the loop distal to the verumontanum. The resection is taken down to the level of the fibrous capsule. This depth can be heralded by the presence of prostatic stones that characteristically form between the capsule of the prostate and the hypertrophied adenoma. As the level of the capsule is approached, one will often encounter pulsatile bleeding as arterial branches are exposed ( Fig. 112-4). These are distinct from the “oozing” type of bleeding one sees when a venous branch is transected. The venous bleeding does not have to be dealt with immediately, but the open arterial branches should be promptly coagulated as these will only lead to increasing visual difficulty as the procedure progresses. The resection is continued from the 3 o'clock to the 6 o'clock position to complete the bulk of the resection of the patient's left lateral prostatic lobe. When the left lateral lobe is completed, hemostasis is established on this side before proceeding to the opposite side. Beginning at the 12 o'clock position the resection is carried down to the 6 o'clock position to resect the right lateral lobe. This leaves the tissue at the apex of the prostate, which is handled last.

FIG. 112-4. Resection of apical adenoma. (A) Apical adenoma. (B) Sweeping lateral to medial movement of loop to remove apical adenoma.

The final stage of the resection removes the adenoma immediately proximal to the external sphincter ( Fig. 112-5). Care is taken not to cut or coagulate the verumontanum as this can lead to painful ejaculation. A small portion of prostatic tissue may extend distal to the veru. It is prudent to not try to resect this tissue as it will significantly increase the risk of postoperative incontinence. At the completion of the resection one should be able to place the resectoscope at the level of the veru and see easily into the bladder.

FIG. 112-5. Arterial supply to surgical capsule and adenoma.

Following the resection all chips are irrigated from the bladder with an Ellick evacuator. It is very important to evacuate all chips from the bladder as they can occlude the catheter postoperatively. The resectoscope is then reinserted and the bladder is inspected to ensure that no chips are being left behind in any bladder diverticula or the prostatic fossa. Finally, complete hemostasis is established. It is important not to irrigate vigorously after final hemostasis is achieved as this can disrupt clots and cause persistent bleeding after bleeding has been deemed satisfactory and the resectoscope removed. The bladder is left full and the resectoscope and sheath are removed. We prefer to place a 22- or 24-Fr three-way catheter with 30 to 60 cm 3 of fluid in the balloon and a slow rate of continuous irrigation for 12 to 24 hours. If persistent venous bleeding occurs that does not clear easily with continuous irrigation, then the catheter can be placed on gentle traction for a few hours. If the irrigant continues to be red and one suspects arterial bleeding, the resectoscope should be reinserted prior to leaving the operative suite and the bleeding vessel searched for and coagulated. Transurethral Incision of the Prostate Transurethral incision of the prostate (TUIP) was originally described by Keitzer et al. in 1961. However, Orandi published the first significant series on the procedure for the treatment of bladder outlet obstruction due to BPH. The indications for TUIP and TURP are identical, with the typical TUIP patient being a younger man with a small prostate. The procedure is generally faster and is associated with less blood loss and fewer complications than TURP. Patients in whom TUIP is most often recommended over TURP are those whose prostate has an estimated resectable weight of 30 g or less, patients who are at higher surgical risk due to associated comorbidity because operative time and blood loss are significantly lower with TUIP, and younger patients because of a lower incidence of retrograde ejaculation after TUIP. TURP, on the other hand, is recommended for patients with larger prostate glands, severe recurrent gross hematuria, or prostatitis when there is a need to remove the hemorrhagic or infected tissue. 8 The procedure can be performed with a variety of cutting knives, an electrosurgical loop, or laser. The goal of the procedure is to incise the prostate in one or two areas, through to the capsule or beyond. The incision is made from the inside of the bladder neck down to the verumontanum and should be deep enough to penetrate the prostate tissue down to and through the prostate capsule. The incision(s) can be unilateral or bilateral and can be made at a variety of locations around the bladder neck. A unilateral incision is usually performed at the 6 o'clock position, but great care should be taken to ensure that the rectum is not injured. Bilateral incisions are usually done at the 5 o'clock and 7 o'clock positions ( Fig. 112-6). Following the incision the bladder neck usually springs apart. Adequate depth of the incision is indicated by visualization of fibers from the prostatic capsule or even protrusion of periprostatic fat. At the completion of the procedure and with the scope in the more distal urethra, there should be no visual obstruction of the bladder. Since no prostatic tissue is obtained for pathologic examination as in TURP, separate prostate biopsies should be performed if indicated.

FIG. 112-6. Transurethral incision of the prostate. (A) Path of incision marked. (B) Six o'clock midline incision through prostatic capsule exposing fat. (C) Bilateral 5 o'clock and 7 o'clock incisions through prostatic capsule exposing fat.

Electrovaporization of the Prostate Transurethral electrovaporization of the prostate (TVP) is an electrosurgical modification of the classical loop resection. Recently, this technique has gained attention due to its ability to rapidly remove prostatic tissue with minimal blood loss. Most urologists find the technique very easy due to its similarity to the familiar loop resection technique. This technique has also been shown to require a shorter catheterization time and hospital stay. 6 TVP combines electrosurgical vaporization and desiccation of prostatic tissue. The energy source is the same as that used in the standard TURP with the energy

being transmitted to the tissue via a roller electrode instead of a wire loop. Electrovaporization, because of its immediate coagulation of transected vessels, lacks the significant fluid absorption and subsequent complications of the standard loop TURP. Electrovaporization has been shown to achieve a similar rate of symptomatic relief of prostatic obstruction as the standard TURP with minimal morbidity. The electrovaporization technique as well as standard electrosurgery involves electrosurgical concepts that are well known and that were described extensively by McLean in 1929. In the cutting mode, the TUR electrode loop rapidly heats the contacted tissue cells so that they explode into steam leaving a vaporized or cut space where the cells were present. Electrosurgical desiccation or coagulation occurs at either a coagulation current or low-power cutting current that causes the cells to dry out slowly instead of vaporizing rapidly ( Fig. 112-7).

FIG. 112-7. The electrosurgical tissue effects produced with the standard resection loop. (A) Desiccation. A low-power current generates heat in the tissue that slowly dries it out. (B) Fulguration. With a light touch, a coagulation current is applied with intermittent high-voltage sparks that fulgurate and char superficial tissue to a carbon texture and dries it out. (C) Vaporization (cutting). A high-energy cutting current rapidly heats cells on contact causing them to explode into steam. An incision or vaporized space is created in its path.

In electrovaporization of the prostate, the leading edge of the roller electrode vaporizes tissue away while the trailing edge causes desiccation of the underlying tissue (Fig. 112-8). The extent of desiccation is governed by the quality of the contact, the power level, and the time of contact by the electrode. Electrovaporization, therefore, removes unwanted tissue by vaporization and provides immediate hemostasis by sealing opened vessels (also preventing fluid absorption) via development of a zone of desiccation below the vaporized tissue.

FIG. 112-8. Roller electrode electrovaporization. Current travels through the point of least resistance. Current density is highest at the leading edge where fresh tissue has the least resistance. Current density is lowest at the trailing edge where desiccated tissue has the highest resistance. The leading edge vaporizes while the trailing edge desiccates and coagulates. Vaporization and desiccation are combined in one motion to remove tissue without bleeding.

The technique is performed in the same manner as that described for a standard loop resection, with the substitution of a roller electrode for the resection loop. This electrode fits the standard types of resectoscopes currently in use. The power source is the same as that used in the standard resection, with the power levels set approximately 50% higher. Numerous companies have developed and continue to develop new roller electrodes, with efficacy results still under study. The advantages of this technique over standard TURP are decreased blood loss and operating time, as well as shorter hospital stay. With TVP excess prostate tissue is vaporized with no residual sloughing and patients do not experience significant postoperative irritative symptoms. The postoperative complications with this technique are the same as that with a standard TURP, with the theoretical advantage of a lower probability of development of the TUR syndrome (see below). Laser Ablation of the Prostate One of the most widely used alternative therapies for the surgical removal of obstructing prostatic adenoma is laser ablation of the prostate or laser prostatectomy. Advances in laser delivery systems to remove prostatic tissue have been numerous. Most use a neodymium:yttrium-aluminum-garnet (YAG) laser as their power source. The most common delivery systems currently in use are transurethral ultrasound-guided laser prostatectomy (TULIP), free-beam side-firing fibers, contact tip lasers, and interstitial fibers. All use the concept of high energy delivered to the prostatic tissue, which results in some combination of immediate vaporization and delayed coagulative necrosis and slough of prostate tissue in order to reestablish a voiding channel through the prostatic urethra. The two basic types of lasers used in prostatectomy are termed contact and noncontact. Contact lasers (including contact tip lasers and interstitial fibers) work by vaporizing the prostatic tissue that actually touches the laser fiber. With a contact laser, the prostatectomy channel should be similar to that achieved with TURP. Contact tips are shaped to suit their individual purposes; round probes can vaporize tissue and create a channel, whereas chisel probes can make linear incisions. Coagulation occurs to about 1 mm beyond the vaporization zone. The technique for the contact tip probes calls for the fiber to be passed through the cystoscope; then, under endoscopic control, the contact tip probe is passed back and forth along the prostatic urethra from the bladder neck to the verumontanum, much like a standard loop resection TURP. A channel is created circumferentially as the prostatic tissue vaporizes. The postresection visual appearance of the prostatic urethra is the final assessment of adequacy of the voiding channel and the amount of energy delivered to the tissue during the resection is unimportant. The disadvantage of this technique is the relatively slow removal of prostatic tissue compared to the standard loop resection. Interstitial fibers work differently in that a number of fibers are placed through the urethra directly into the prostate and the laser energy is delivered through the tip, so that all of the energy is absorbed only by the prostate. The fibers also disperse the laser energy within the prostate so that much of the resulting defect, unlike in contact tip fibers, is the result of later tissue coagulative necrosis. The noncontact lasers (including TULIP devices and side-firing fibers) deliver energy to the prostatic tissue without actually making contact with the adenoma and depend on delayed sloughing of tissue following coagulative necrosis. The tip of the TULIP probe combines a laser transparent balloon to stabilize the device and maintain a fixed distance between the laser beam and the tissue. The device is positioned in the prostatic urethra and under continuous ultrasound monitoring the laser is directed at the quadrants of the prostate as the probe is moved from the bladder neck to the verumontanum, so that a circumferential treatment of the prostatic urethra is completed. The difficulty in this technique is having to depend on ultrasound guidance to localize the area of energy delivery as opposed to the more comfortable direct visualization possible with other techniques. The free-beam side-firing laser delivery system has been the most popular of laser systems used. This technique reflects the beam out of the fiber at 90 degrees to the axis of the cystoscope at roughly right angles to the prostatic adenoma. The fibers are designed to allow for various degrees of divergence of the beam once it is reflected, which is directly proportional to the spot size and power density. Fibers with a narrow angle of divergence have higher power densities and are more useful

for immediate vaporization of tissue. Larger angles of divergence spread the power over the tissue surface and are better for coagulating larger areas that later slough to produce a voiding channel. All fibers can provide tissue vaporization or coagulation depending on the technique, power, and time settings utilized. Thus, an understanding of the specific characteristics of the chosen fiber is important for the surgeon to be able to deliver the combination of immediate vaporization and later coagulative necrosis desired. The initial technique employed with side-firing lasers relied primarily on coagulation and the sloughing of the prostatic tissue over several weeks to produce the desired prostatic defect. More recently, a combination of various degrees of vaporization and coagulation has been used. Postoperative Care Following TURP we ambulate patients as soon as possible. Ambulation can be started on the night of surgery unless the patient has his catheter on traction. In this case, ambulation should begin the following morning. The patient is given a regular diet as soon as he has fully recovered from anesthesia. We always keep patients on stool softeners for a week postoperatively and forbid rectal examination, administration of rectal suppositories, or use of rectal temperatures in the immediate postoperative period. The Foley balloon, although overinflated, can slip back into the widely opened prostatic fossa after TURP. This can lead to discomfort and bleeding as this prevents the normal mechanism of hemostasis via contraction of the prostatic capsule. If this is suspected, the balloon can be deflated, advanced further, and reinflated with more fluid, and the bladder well irrigated. Additionally, an overinflated Foley balloon can actually sit over the ureteral orifices and may occlude them postoperatively. Once the irrigant is clear, the irrigation port is closed and the catheter allowed to drain by gravity for a few hours. If the drainage remains clear, the balloon is deflated and the catheter removed. The less invasive procedures for prostatic obstruction, such as lasers and incisions of the prostate, obviously entail more liberal postoperative care. Often after an uneventful TUIP, for example, the patient may be discharged on the day of surgery with the catheter being removed in the recovery room or early the next day. We also find it possible to routinely remove the catheter on the first postoperative day following TVP, even after resection of a fairly large prostate, with the patient being discharged from the hospital within 12 to 24 hours of his procedure. Depending on how much coagulative effect versus vaporization effect is employed in a laser prostatectomy, patients rarely require a catheter for more than 12 to 24 hours postoperatively, but should expect to pass small amounts of necrotic prostatic tissue out in the urine for a period of time. At times, this passage of tissue is accompanied by irritative voiding symptoms that eventually resolve.

OUTCOMES
Complications Arterial bleeding is characterized by jets of blood spurting in a pulsatile fashion. These should be controlled promptly with electrocoagulation, as described. All bleeding in a certain area of the resected fossa should be well controlled before moving onto the next subsequent area so that the bleeding does not get out of control. Loss of visualization through the resectoscope from uncontrolled hemorrhage can be disastrous. Postoperatively venous bleeding is characterized by initially clear irrigant after catheter insertion, followed by dark blood. This type of bleeding can be controlled by placing the catheter on mild traction. The modern mortality rate for TURP according to the AUA cooperative study is 0.2% and the morbidity rate was reported to be 18%. 10 These figures were similar to those of other reports. 11 Transfusion was required in 3.7% of patients. Intraoperative bleeding was related to two factors: size of the gland and operative time. Glands weighing more than 45 g and resection times of over 90 minutes both were associated with significantly higher bleeding rates. Absorption of irrigant leading to dilutional hyponatremia and symptoms of confusion, nausea, vomiting, hypertension, bradycardia, and visual disturbances is known as the TUR syndrome. This occurred in 2% of patients in the cooperative study and was, again, greater in patients with larger glands and with resection times of more than 90 minutes. This syndrome must always be carefully watched for by symptom assessment and measurement of serum electrolytes intra- and postoperatively. If it occurs, it should be treated promptly with intravenous saline and diuretics and termination of the procedure if appropriate. Intraoperative complications include irrigant extravasation and bladder perforation. Extravasation of fluid occurs when the prostatic capsule is perforated. The symptoms of extravasation include restlessness, nausea, vomiting, and abdominal pain localized to the lower abdomen and back. When one suspects extravasation the procedure should be immediately terminated once hemostasis is achieved. Extravasation can be simply managed with catheter drainage. Bladder perforation can occur from overdistention and/or injury to the inside of the bladder wall with the resecting instrument. One should become suspicious of this if a patient's lower abdomen becomes rigid or if a patient under spinal anesthesia experiences shoulder pain (due to irritation of the diaphragm from intraperitoneal irrigation fluid). Other postoperative complications include failure to void after the catheter is removed, clot retention, and infection. Failure to void may be secondary to a hypotonic bladder, which can be diagnosed via cystometrogram, or due to residual adenoma causing obstruction, which can be diagnosed with reinstrumentation. Residual obstructing adenoma must be resected to allow the patient to void. A hypotonic bladder may regain its tone with longer catheter drainage (usually 7 to 10 days). We do not feel that bethanechol chloride is useful in this scenario, although some urologists choose to try this in order to hasten return of bladder function. Delayed postoperative bleeding may result from increased activity or from straining to have a bowel movement. Usually this type of bleeding is self-limiting, but if prolonged it should be managed by irrigation of all clots from the bladder and prostatic fossa even if this means returning to the cystoscopy suite to do so. Often this is all that is required to stop further bleeding as a specific site of hemorrhage is rarely seen with cystoscopy. Late complications of prostatic resections include vesical neck contracture, urethral stricture, incontinence, and impotence. Vesical neck contracture typically occurs 4 to 6 weeks after surgery and is suspected when a patient initially has a good stream followed by a marked reduction. This complication can be remedied by vesical neck incision with a Collings knife, cold urethrotome, or loop resectoscope incision of the contracture. Urethral stricture rates were 2.7% in the AUA study and were most commonly due to trauma from the resectoscope. This most often occurs in the fossa navicularis. It can be treated with either sequential urethral dilation or urethrotomy. The most dreaded complication after TURP is incontinence resulting from damage to the external sphincter mechanism. However, other factors may often be the cause of incontinence after TURP. These include detrusor instability and residual adenoma blocking the external sphincter mechanism. Kegel exercises can be utilized to strengthen a damaged external sphincter mechanism, but results are variable depending on the extent of sphincteric damage. Collagen injections and artificial sphincters may be necessary in these patients with sphincteric damage. Anticholinergics may be helpful in controlling involuntary bladder contractions due to involuntary detrusor contractions. Impotence has been reported in the AUA study to be about 4%. This complication has been poorly studied, however, and rates of impotence after TURP have ranged from 4% to 40%. No clear mechanism has been established, but the proximity of the cavernous nerves to the lateral floor of the prostate should be noted. Results Results have shown that 90% of patients have significant benefit from TURP. The criteria to measure these outcomes are varied, with even more variability interjected by the broad spectrum for indications for surgery reported by investigators. Furthermore, many of the symptoms reported by patients are subjective and may exhibit a placebo effect of 40% or more. Using the BPH guidelines, 88% of patients undergoing TURP and 80% of patients undergoing TUIP reported symptomatic improvement. Additionally, the bothersomeness of the symptoms are directly correlated to the relative relief of the symptoms by the procedure. More objective measures such as flow rates show an average improvement in peak flow rates of 9.8 ml/sec in patients undergoing TURP and 7.3 ml/sec in patients undergoing TUIP. TURP remains the gold standard for the treatment of symptoms of bladder outlet obstruction due to BPH in appropriate surgical candidates, whereas TUIP should be considered in any patient with a smaller prostate (30 g or less), especially in younger patients who may be concerned with preservation of antegrade ejaculation. Most studies comparing TUIP with TURP, in patients with prostates of 30 g or less, found a longer operative time and a greater need for blood transfusions with 14,15 and 16 In these studies, TURP and TUIP achieved comparable efficacy for the relief of symptoms of bladder outlet obstruction. Additionally, retrograde ejaculation following TURP ranges from 50% to 95% and following TUIP is only from 0 to 37%. 1 It has been reported that the incidence of retrograde ejaculation with TUIP is significantly lower if one incision is used instead of two, with a unilateral incision achieving comparable efficacy to bilateral incisions. 3 Alternative procedures that are less invasive or result in shorter hospital stays with potentially less associated morbidity are certainly of great interest to patients and surgeons. Time and further studies will prove or disprove the utility of these newer alternatives. It must be kept in mind, however, that even the least invasive

procedure is nonetheless associated with potential surgical and anesthetic complications, and must be entered into with the same consideration for surgical indications as the more invasive surgical therapies. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Edwards LE, Bucknall TE, Pittarn MR, Richardson DR, Stanek J. Transurethral resection of the prostate and bladder neck incision: a review of 700 cases. Br J Urol 1985;57:168–171. Greene LF, Holcumb GR. Transurethral resection in special situations. In: Greene LF, Segura JW, eds. Transurethral surgery. Philadelphia: WB Saunders, 1979;216. Hedlund H, Ek A. Ejaculation and sexual function after endoscopic bladder neck incision. Br J Urol 1985;57:164–167. Holtgrew L. Transurethral prostatectomy. Urol Clin North Am 1995;22:357–368. Kaplan SA, Goluboff ET, Olsson CA, Deverka PA, Chimiel JJ. Effects of demographic factors, urinary peak flow rates, and Boyarsky symptom scores on patient treatment choice in benign prostatic hyperplasia. Urology 1995;45:398–405. Kaplan SA, Te AE. Transurethral electrovaporization of the prostate (TVP): a novel method for treating men with benign prostatic hyperplasia. Urology 1995;45:566–572. Kulb TP, Kainer M, Lingeman JE, et al. Prevention of post-prostatectomy vesical neck contracture by prophylactic vesical neck incision. J Urol 1987;137:230–231. Madsen FA, Bruskewitz RC. Transurethral incision of the prostate. Urol Clin North Am 1995;22:369–373. McConell J, Barry M, Bruskewitz R, et al. Benign prostatic hyperplasia: diagnosis and treatment. Clinical practice guidelines no. 8, AHCPR Publication 94-0582, Rockville MD, Agency for Health Service, U.S. Department of Health and Human Services, 1994. Mebust WK, Holtgrew L, Cockett ATK, et al. Transurethral prostatectomy immediate and postoperative complications: a cooperative study of 13 participating institutions evaluating 3,885 patients. J Urol 1989;141:243–247. Mebust W. Transurethral surgery. In: Walsh P, Retik A, Stamey T, et al., eds. Campbell's urology. 6th ed. Philadelphia: WB Saunders, 1992;2900. Melchior J, Valk WL, Foret JD, et al. Transurethral prostatectomy: computerized analysis of 2,223 consecutive cases. J Urol 1974;112:634–642. Neilson K, Anderson K, Asbjom J, et al. Blood loss in transurethral prostatectomy: epidural versus general anesthesia. Int Urol Nephrol 1987;19:287–291. Orandi A. Urological endoscopic surgery under local anesthesia: a cost reducing idea. J Urol 1984;132:1146–1150. Riehmann M, Knes JM, Madsen PO, et al. Transurethral resection (TURP) versus incision of the prostate: a prospective randomized study (Abstract). J Urol 1993;149:323A. Soonawalla PF, Pardanani DS. Transurethral incision versus transurethral resection of the prostate: a subjective and objective analysis. Br J Urol 1992;70:174–178.

Chapter 113 Endoscopic Laser Surgery Glenn’s Urologic Surgery

Chapter 113 Endoscopic Laser Surgery
Ken Koshiba and Toyoaki Uchida

K. Koshiba: Department of Urology, Kitasato University School of Medicine, Kitasoto, Sagamihara, Kangawa, 228 Japan, and Kitasato Institute Medical Center Hospital, Kitamoto, Saitama, 364 Japan. T. Uchida: Department of Urology, Kitasato University School of Medicine, Kitasoto, Sagamihara, Kangawa, 228 Japan.

Energy Effects On Tissue Characteristics of the Laser Beam Clinical Applications Indications for Surgery Procedures Alternative Therapy Operative Technique Preoperative Preparation Visual Laser Ablation of the Prostate Transurethral Evaporation of Prostate Outcomes Complications Results Other Uses of the Laser in Urology Bladder Tumors and Upper Urinary Tract Tumor Urethral and Ureteral Stricture Interstitial Cystitis Laser Treatment of Stones Chapter References

A laser is a novel form of energy with tissue effects differing from those possible by conventional surgical instruments. Small flexible fibers allow laser energy delivery through the endoscope and access to virtually all portions of the genitourinary system. Although the rationale for laser use in some areas of urologic surgery is improved therapeutic efficacy, the primary advantage often is reduced patient morbidity and decreased incidence of operative complications. A laser can be used to fragment stones, coagulate tissue, cut and vaporize, and weld tissue together. 12,16 The important variables allowing for this diversity of action are laser wavelength, pulse duration, and type of delivery system. This chapter discusses various applications of endoscopic lasers. Energy Effects on Tissue The word “laser” is an acronym for light amplification by stimulated emission of radiation. Laser energy is a type of electromagnetic radiation that requires no medium for transmission. Depending on the range of frequency, wavelength, and energy, many types of lasers exist in the universe and are depicted in the electromagnetic spectrum13 (Fig. 113-1). Laser energy interacts with tissue by being reflected, absorbed, refracted, or transmitted. Argon, carbon dioxide (CO 2), neodymium:yttrium-aluminum-garnet (Nd:YAG), potassium-titanyl phosphate (KTP), holmium (Ho), pulsed dye, and alexandrite laser systems have various medical applications. Their therapeutic values depend on the unique properties of each.

FIG. 113-1. Absorption spectrum in water and hemoglobin showing the wavelengths of different lasers. (Reprinted with permission from Stein BS, Kendall AR. Lasers in urology. I. Laser physics and safety. Urology 1984;23:405–410.)

Characteristics of the Laser Beam Nd:YAG (1,064 nm) is a solid-state laser produced by neodymium ions functioning in solution as the active medium dispersed in a YAG crystal. As seen in Figure 113-1, this laser is situated close to the infrared portion of the electromagnetic spectrum (thus, invisible) and has a wavelength that prevents its absorption by either water or hemoglobin. Therefore, it can be used in a fluid medium as in endoscopic surgery. The CO 2 (10,600 nm) laser is absorbed by water and argon (0.500 nm), absorbed more by pigmented tissue and hemoglobin. KTP (532 nm) is a frequency-doubled Nd:YAG laser whose effects on tissue are similar to those of the argon laser. Vaporization to some degree is possible. Ho (2,100 nm) does not penetrate as deeply as Nd:YAG and is used for incisions, vaporization of the tissue, and stone fragmentation. Pulsed dye (504 nm) and alexandrite (750 nm) lasers are transmitted through thin and flexible quartz fibers, cause minimal tissue injury, and are particularly useful for stone fragmentation. The power of a laser source is expressed in watts. The product of power in watts multiplied by time in seconds is expressed as energy in joules (watts × seconds). Power density is the area of target tissue to which power is delivered. Power density = power in the focal spot/area = watts/pr2 Power density is the most important parameter expressing the strength of a laser at a particular point and is directly proportional to the distance of the beam from the tissue. Fibers with narrowly divergent beams have much higher power density than those with widely divergent beams. Maximum power density is possible only if the tip is virtually in contact with tissue because at this point divergence is least. 6,16 In general, high power and short duration cause stone fragmentation, incision, and vaporization of tissue, whereas low power and long duration cause deeper penetration, coagulation necrosis, and minimal vaporization. Clinical Applications Various new devices and techniques are available as alternatives to conventional surgical treatment of benign prostatic hyperplasia (BPH). These are an efficient and

effective means for applying a large amount of energy precisely to a given target.

3,4,6,7,14

The Nd:YAG laser in particular is used for prostatic surgery and has several characteristics that make it useful in endoscopic urologic procedures ( Fig. 113-2). First, energy is transmitted fiberoptically through small quartz fibers that pass easily through a standard cystoscope. Second, a negligible amount of Nd:YAG laser energy is absorbed by water and hemoglobin (Fig. 113-1), thereby allowing the energy to be transmitted through irrigants often used in transurethral procedures.

FIG. 113-2. Endoscopic laser ablation therapy for benign prostatic hyperplasia.

INDICATIONS FOR SURGERY
In general, patients who are candidates for TURP may undergo laser ablation of the prostate. The relationship between prostate size and effectiveness of laser treatment has yet to be determined fully. Laser ablation of a smaller prostate is more easily conducted (less than 50 cm 3).3,4,6,7,14 Patients are adequately informed of the operative and postoperative course. Postoperative care and expected course, including catheterization requirement for a number of days, are discussed in detail. For patients undergoing laser coagulation prostatectomy, particular emphasis is placed on very gradual improvement in voiding symptoms that may be expected following surgery. This improvement is neither immediate nor initially dramatic; some symptoms may actually worsen until dissolution of the treated transition zone. Most or all of these potential risks and complications appear to diminish with laser coagulation of the prostate. 3,4,6,7,14 Procedures The advent of newer side-firing fibers has revived interest in the Nd:YAG laser for treatment of BPH. There are two major laser devices presently in use for this purpose: 1. A right angle, noncontact laser beam under endoscopic control (visual laser ablation of the prostate: VLAP). 2. A right angle, contact laser beam under endoscopic control (transurethral evaporation of the prostate: TEVP) Present-day laser application involves the use of many side-firing laser fibers in BPH treatment ( Table 113-1). The results differ according to the mechanism that causes side firing, power density, durability, size of the tip, angle of beam emergence, and arc of beam divergence. The mode of application may be contact or noncontact. Fiber tips are made of quartz, pure gold, or gold-plated alloy. The angle of deflection varies from 45 to 105 degrees and the arc of beam divergence from 7 to 38 degrees.

TABLE 113-1. Characteristics of various side-firing delivery systems

ALTERNATIVE THERAPY
Alternative therapies include observation, alpha blockers, 5a-reductase or other medical management, TURP, transurethral incision of the bladder neck(TUIBN), electrovaporization of the prostate, thermal therapy (transurethral or transrectal), urethral stents, balloon dilation of the prostatic urethra, or open prostatectomy.

OPERATIVE TECHNIQUE
Preoperative Preparation Preparation is the same as that for TURP. Low spinal or epidural anesthesia is generally used. The procedure may be conducted under local anesthesia for patients with smaller glands. 4 With the patient in the lithotomy position, a 21-Fr cystoscope equipped with a working port is introduced into the urethra. Several continuous flow cystoscopes designed for laser prostatectomy are available. Because little or no fluid absorption occurs during coagulation of the prostate with the Nd:YAG laser, special irrigation solutions are not required. Sterile water is relatively inexpensive compared to osmolar solutions and clear observation is possible. After carefully inspecting the prostatic urethra and urinary bladder, a laser fiber is passed through the working port of the cystoscope. The Nd:YAG laser is invisible to the human eye and thus the laser source is routinely provided with a low-energy helium:neon visible red aiming beam to facilitate directing it during surgery. In nearly all cases, the authors use a video display for laser prostatectomy. Preoperative preparation for laser surgery must always include careful attention to laser safety in the operating room. The Nd:YAG laser causes significant thermal injury to human tissue on contact and irreversible retinal damage occurs if the eye is struck. All operating room personnel, including the surgeon, nursing staff, and anesthesiologist, and the patient must take measures to ensure eye protection. Operating room doors should be appropriately marked to warn outside personnel that the Nd:YAG laser is in use and adequate protective eye equipment must be worn before entry. Special filters are available to cap the cystoscope lens to protect video endoscopy equipment from retrograde transmission of laser light. Visual Laser Ablation of the Prostate For noncontact-type urolase fibers with an arc of divergence of 30 degrees, power density is 2740 W/cm 2 at 2 mm from the surface of the tissue (Table 113-1 ). Continuous laser application is performed for at least 60 to 90 seconds at 40 to 60 W to maximize tissue coagulation. With noncontact-type urolase right angle laser fibers, laser power usually ablates to a depth of 2 to 3 mm and coagulates prostatic tissue to a depth and width of 10 to 15 mm, respectively, at each point at 60 W for

60 seconds irradiation ( Fig. 113-3).3,4,15

FIG. 113-3. Thermal effect with noncontact-type laser fiber. Arrows indicate back-scattering.

The laser is initially directed to the median or lateral lobe. This should first be done proximally and continued distally, working away from the bladder neck toward the verumontanum so that the tip of the laser fiber is cooled with irrigants and numerous bubbles in the bladder are eliminated. The 8- or 14-quadrant approach to the lateral lobes is typically used for these purposes. In the 8-quadrant method, irradiation is done at 1, 3, 5, 7, 9, and 11 o'clock positions. For the 6-quadrant, irradiation is done in most protruding areas of the prostate and at 3 o'clock and 9 o'clock positions in the middle of each of the first irradiated spots and verumontanum ( Fig. 113-4). For a 15 to 25 cm3 prostatic adenoma whose bladder neck is less than 2.5 cm from the verumontanum, single 6-quadrant set laser application generally leads to adequate lateral lobe coagulation. 3,4,7,14 When the bladder neck is more than 2.5 cm from the verumontanum, another 6-quadrant laser application is required at about every 1 to 2 cm along the length of the lateral lobes. 3,14 Spot laser application to obstructing median lobe tissue is conducted in essentially the same manner. From a single (6 o'clock) to a three times (5, 6, and 7 o'clock) application is required for adequate tissue coagulation ( Fig. 113-5). A spot irradiation is applied at 40 to 60 W for 30 to 60 seconds. For Nd:YAG laser coagulation prostatectomy, if total resectable prostate tissue is estimated, it is proposed that at least 1000 joules laser energy per gram be administrated. 3,4,14 For all fibers with an external metal mirror reflecting mechanism, routine cleaning of the mirror with a wet gauze sponge is recommended. The buildup of spattered particulates on the mirror surface during treatment may lead to superheating and consequent degradation of the metal, loss of efficiency in laser transmission, and eventual meltdown and destruction of the instrument.

FIG. 113-4. Operative approach for lateral lobes with noncontact-type laser fiber.

FIG. 113-5. Operative approach for median lobe with noncontact-type laser fiber.

When using noncontact-type laser fibers, transurethral incision of the bladder neck at 5 o'clock and 7 o'clock positions and/or channeling TURP to minimize postoperative urinary retention are oth conducted, with consequently better clinical outcome. 14 Transurethral Evaporation of Prostate The contact free-beam technique, or “TEVP,” is virtually the same as standard TURP. 6 The laser is initially directed to the bladder neck at 60 W until circular fibers of the neck become visible. Furrows are then made in the lateral lobe at 2, 4, 8, and 10 o'clock positions ( Fig. 113-6). Tissue is evaporated by holding the laser fiber in contact with the tissue and dragging at a rate of 1 cm for 20 to 30 seconds of energy delivery. The fiber tip is rotated 15 degrees left and right during dragging. This makes a wide furrow and prevents bleeding. A furrow 5 to 7 mm deep with a 2- to 3-mm rim of coagulated tissue is produced ( Fig. 113-7). Following completion of the first pass, a second or third is made in the same area wherever residual tissue is seen. Tissue between the furrows is evaporated in a similar manner. The dragging speed of the fiber is faster in charred areas (10 sec/cm) and slower in coagulated areas (35 to 45 sec/cm) owing to differences in laser beam absorption by charred (rapid) versus blanched (slow) tissue. 6 The speed is faster at the apex where there is less depth of prostatic tissue and the area is close to the external sphincter. The limits of laser ablation for BPH are marked 0.5 cm proximal to the most prominent portion of the verumontanum. This procedure is considered complete when a large channel is evident by visual inspection. Hemostasis is achieved by directing the laser at low power (20 to 40 W) and moving the fiber away from the tissue (2 to 3 mm). The median lobe is treated in the same manner when present.

FIG. 113-6. Operative technique with contact-type laser fiber.

FIG. 113-7. Thermal effect with contact-type laser fiber. Arrows indicate back-scattering.

OUTCOMES
Complications Postoperative impotence should of course be less than that of TURP, though not 0%. Antegrade ejaculation is usually preserved if the bladder neck has not been lased. Intraoperative and postoperative bleeding are minimal. Bleeding necessitating transfusion, even in patients taking aspirin or warfarin has not been noted to date, nor has fluid absorption, rectal injury, or fistula been reported. 3,4,6,7,14 The incidence of urethral stricture is less following laser prostatectomy. However, bladder neck contracture was noted in 6% of our patients, possibly due to overlasing of the bladder neck lesion. 14 Urinary tract infection has been reported in 1% to 20% of patients following laser prostatectomy and epididymitis in 5% to 7%. 3,4,6,7,14 The incidence of urinary incontinence is rare for all modes of laser prostatectomy. In most studies on laser prostatectomy, the incidence of postoperative urinary retention lasting from a few days to several weeks is high, varying from 20% and 32% and exceeding the 6.5% incidence following TURP 3,4,5,6 and 7,14 The prolonged postoperative retention following TEVP using contact-type laser fiber is 5%, this being essentially the same as for TURP. 6 The need for reoperation following VLAP is 9%, but only 2% every year with TURP and 0% to 4% in the first year after TEVP.3,4,6,7,14 A complication unique to most laser prostatectomies is the high incidence of irritative voiding symptoms during the first few weeks following surgery. These symptoms can be managed temporarily with anticholinergics, a 1-blockers, or narcotics and are due to coagulated necrotic tissue that has not yet sloughed, as well as raw and unepithelialized mucosa. Results Clinically significant improvement and objective voiding were noted following laser prostatectomy. Patient symptoms, assessed according to the American Urological Association symptom index, decreased on an average of approximately 60% from the preoperative baseline in several reports. 3,4,6,7,14 As an objective measure of voiding outcome, peak urinary flow rate in these patients improved on an average of more than 100% compared to preoperative baseline values. 3,4,6,7,14 Laser therapy has several advantages over standard TURP, such as technical simplicity and absence of complications such as intraoperative bleeding, fluid absorption, retrograde ejaculation, impotence, and incontinence. Patients require less hospital stay and recover more quickly. The results of laser treatment have been shown to persist for at least 2 to 3 years. 3,4,6,7,14 Longer follow-up should be conducted to ensure that laser therapy will be conducted for BPH.

OTHER USES OF THE LASER IN UROLOGY
Bladder Tumors and Upper Urinary Tract Tumor The Nd:YAG laser is used to coagulate bladder tumors as much as 2 cm in diameter. Forty watts for 4 seconds of lasing are generally applied for spots that overlap tumors.12 One drawback of the free-beam Nd:YAG is that it penetrates the bladder, with potential risk of damage to the small bowel within the peritoneum. The Ho laser vaporizes the tumor and when in contact with tissue, the laser fiber sweeps off small bladder tumors. 12 The laser is not appropriate for initial bladder tumor treatment because grade and stage must be established by standard electrical resection. Lasers appear promising for treatment of upper tract tumors but only in certain patients. 2,11 The Nd:YAG laser is the first laser for treating bladder tumors and ablation of renal pelvic and ureter neoplasms. In the endoscopic use of the Nd:YAG laser, small-diameter fibers should lead to better results (400 µm or less). At 20 to 30 W, and the tip is removed over the surface to coagulate tissue (as indicated by a color change to white), the procedure being essentially the same as that for bladder tumors.2,11 This laser penetrates to a depth of 5 to 6 mm, though tumor depth may be greater than this. Coagulated tissue is removed to expose portions of the tumor that are otherwise not visible ureteroscopically. There is little risk of perforation or forward scattering in ureter application of the laser. Ho:YAG and argon lasers have urologic application. The energy of these lasers is absorbed within less than 1 mm and there is virtually no risk of forward scattering. 16 Both coagulate tissue, ablate, and actually remove tissue. Urethral and Ureteral Stricture Lasers incise tissue with intensive or minimal coagulation and are thus useful for incising a urethral stricture, the urethral valve, or ureteral stricture. There are various methods for Nd:YAG laser treatment of urethral strictures. Circumferential application of laser energy to the entire stricture and urethrotomy using the laser are particularly useful in this regard. 12 Contact-type Nd:YAG laser fibers can be passed through the cystoscope and thus are suitable for this purpose at 12 W. The urethrotomy is performed at the 12 o'clock position by placing the fiber tip in direct contact with the stricture. For ureteral stricture, lasers have been used retrogradely via miniaturized ureteroscopes. Contact-type fibers with Nd:YAG or Ho lasers are effective for making incision of urethral and ureteral strictures. Deeper penetration can be avoided and the stricture lesion can be cut. Interstitial Cystitis Although the mechanism remains as obscure as the cause of the disease, the Nd:YAG laser has been shown in all cases to provide symptomatic relief to most women with interstitial cystitis. 12 Frequently, there is less bladder pain after treatment and a reduction in urinary frequency and urgency. Symptomatic relapse is common at 3

to 10 months after treatment, but secondary response may be possible with a second laser treatment. Treatment is performed under general or regional anesthesia and an Nd:YAG laser fiber is inserted through the cystoscope. Thirty to 35 W of energy is used. The bladder should be distended only appropriately, and Trendelenburg's position may be useful for removing loops of the small bowel from the dome of the bladder. 12 Laser Treatment of Stones Calculus destruction using a ruby laser was first attempted in the 1960s. 12,16 Ureteral stones not treatable by extracorporeal shock wave lithotripsy (ESWL) are managed best by laser treatment using a rigid or flexible ureteroscope. To lessen ureteroscopic damage to the ureter as much as possible, the smallest caliber ureteroscope that allows visualization of the stone should be used. Flush-lamp pulsed dye laser systems operate at 60 to 200 millijoules (mJ) for 5 Hz of energy delivered via the 200- to 600-µm fiber system. Higher power may be necessary for some harbor stones. 9,16 The fiber is placed in direct contact with the stone. The efficiency of fragmentation can be assessed through the ureteroscope. The stone is broken into fragments of sufficiently small size, which are retrieved with a basket or grasping forceps or are allowed to pass. The primary advantage of pulsed dye laser lithotripsy is the least ureteral or renal injury of all forms of intracorporeal lithotripsy. But it may not be possible to fragmentate harder calculi, such as those composed of calcium oxalate monohydrate or cystine. Psihramis reported laser lithotripsy with pulsed dye lasers to be successful in 107 (88%) of 122 patients with ureteral calculi that could not be treated by ESWL. 9 Ureteral perforation occurred in two cases and a ureteral stricture developed in one patient. The alexandrite laser is operated at a wavelength of 750 nm. Laser energy is delivered at a frequency of 1 to 10 Hz, pulse duration of 150 to 800 nsec, and pulse energy of 30 to 80 mJ. In a large number of patients, Pertusa et al. successfully fragmented 98 of 112 stones and there was no major ureteral wall injury or fiber destruction. 8 The Ho:YAG laser is the most recent laser lithotripsy. Preliminary results with this laser demonstrate its high capacity for fragmentation with few complications, provided procedure is properly conducted. 10 Although smaller, flexible probes such as electrohydraulic lithotripsy or a laser have been shown to be quite useful ureteroscopically, larger stones are more apt to be found in present-day percutaneous nephrolithotripsy populations and are treated more efficiently by a mechanical device provided with a larger, more rigid probe such as ultrasound or Lithoclast. CHAPTER REFERENCES
1. Dretler SP. Laser photofragmentation of ureteral calculi: Analysis of 75 cases. J Endourol 1987;1:9. 2. Gaboardi F, Bozzola A, Dotti E, Galli L. Conservative treatment of upper urinary tract tumors with Nd:YAG laser. J Endourol 1994;8:37. 3. Kabalin JN. Benign prostatic hyperplasia, laser coagulation. In: Smith AD, Badlani GH, Begley DH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996;1078–1097. 4. Leach GE, Sirls L, Ganabathi K, Roskamp D, Dmochowski R. Outpatient visual laser-assisted prostatectomy under local anesthesia. Urology 1994;43:149. 5. Mebust WK, Holtgrewe HL, Cockett ATK, Peters PC, and Writing Committee. Transurethral prostatectomy: immediate and postoperative complications. A cooperative study of 13 participating institutions evaluating 3,885 patients. J Urol 1989;141:243. 6. Narayan P, Tewari A. Benign prostatic hyperplasia, laser therapy of the prostate, laser evaporation. In: Smith AD, Badlani GH, Bagley DH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996;1054–1077. 7. Norris JP, Norris DM, Lee RD, Rubenstein MA. Visual laser ablation of the prostate: clinical experience in 108 patients. J Urol 1993;150:1612. 8. Pertusa C, Albisu A, Acha M, Blasco M, Llarena R, Tanago JG. In vitro lithotripsy with the alexandrite laser. Eur Urol 1992;22:62. 9. Psihramis KE. Laser lithotripsy of the difficult uretral calculus. Results in 122 patients. J Urol 1992;147:1010. 10. Razvi HA, Denstedt JD, Chun SS, Sales JL. Intracorporeal lithotripsy with the holmium:YAG laser. J Urol 1996;156:912. 11. Schmeller NT, Hofstetter AG. Laser treatment of ureteral tumors. J Urol 1989;141:840. 12. Smith JA, Jr. Urologic laser surgery. In: Glenn JF, ed. Urologic surgery. 4th ed. Philadelphia: JB Lippincott, 1991;422–431. 13. Stein BS, Kendall AR. Lasers in urology. I. Laser physics and safety. Urology 1984;23:405–410. 14. Uchida T, Egawa S, Iwamura M, et al. A non-randomized comparative study of visual laser ablation and transurethral resection of the prostate in benign prostatic hyperplasia. Int J Urol 1996;3:108. 15. Uchida T, Ohori M, Iwamura M, et al. Penetration depth of the neodymium:YAG laser in the human prostate. Jap J Endourol ESWL 1995;8:73. 16. Watson G. Lasers. In: Smith AD, Badlani GH, Bagley DH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996;78–86.

Chapter 114 Ureteroscopy Glenn’s Urologic Surgery

Chapter 114 Ureteroscopy
Anuar Ibrahim Mitre, Jose Luis Chambo, and Sami Arap

A. I. Mitre, J. L. Chambo, and S. Arap: Division of Urology, Hospital das Clínicas, University of Sao Paulo School of Medicine, CEP 05422-970, Sao Paulo, SP, Brazil.

Applied Anatomy Indications for Surgery Patient Selection and Preoperative Evaluation Alternative Therapy Surgical Technique Instrumentation Ureteroscopes Ureteroscopy Ureteral Calculus Manipulation Ureteroscopic Calculus Fragmentation Postureteroscopy Management Hematuria Ureteral Tumors Ureteral Strictures Outcomes Complications Results Chapter References

Since the first cystoscopy more than a century ago, the endoscopic examination of the upper urinary tract has become a goal in itself. The development of the first rigid ureteroscope is attributed to Enrique erez-Castro. 6 In the early 1980s, ureterorenoscopy came into use as a routine procedure due to technological advances related to optical fibers and other endoscopic accessories. More recently, further refinements in technology brought about sophisticated optical equipment, with adequate working channels, that made ureteroscopic procedures easier and less traumatic. There are two different generations of nonflexible ureteroscopes. The first generation corresponded to 12.5-F instruments that required ureteral dilation in many patients. The second generation corresponds to semirigid ureteroscopes (they have a certain flexibility without distorting the image) ranging in diameter from 6 to 9.5 F. They have working channels between 3 and 5 F and make instrumentation easier, gentler, faster, and less traumatic. In addition, they usually do not require ureteral dilation. The surgeon should have two ureteroscopes available. One should be 6 to 7.5 F, suitable for direct endoscopic examination, and, when possible, used to treat the condition in which ureteroscopy is indicated. When a large ureteroscope is necessary, the surgeon should use an 8.5- to 9-F instrument, which finds the ureter already dilated rendering its introduction easier. Besides offering a better quality of image, it also allows the passage of a 5-F working instrument, making possible the accomplishment of the programmed procedure. Applied Anatomy The ureters are retroperitoneal structures with a sinuous course and a length that usually ranges from 28 to 34 cm. They have three normal anatomic areas of narrowing: the ureterovesical junction; the portion of the ureter that crosses the iliac vessels; and the ureteropyelic junction. Some particular peculiarities of the ureteral anatomy may render ureteroscopy extremely easy in some patients and difficult, or even impossible, in others. Several factors, even nonpathologic, eventually represent limitations concerning the ureteroscopic examination: position of the ureteral meatus in relation to the bladder neck; the submucosal course; the ureteral hiatus; ureteral diameter and elasticity; its relation to the iliac vessels; and the access angle of the ureteroscope to the upper ureter may be limited by the pubis (Fig. 114-1).

FIG. 114-1. Regional anatomy of the kidney and ureter (sagittal section). From the kidney, the ureter ascends to the level of the pelvic vessels and then descends into the pelvis to the bladder.

INDICATIONS FOR SURGERY
The expanded utilization of ureteroscopy in urology practice was a consequence of ingenious new techniques, besides the simplification of these procedures due to more advanced equipment. The main indications for ureteroscopy are listed in Table 114-1.

TABLE 114-1. Indications for Ureteroscopy

PATIENT SELECTION AND PREOPERATIVE EVALUATION
Ureteroscopy must be carried out in patients without urinary tract infections. When present they must be treated prior to the procedure. Besides the individual physician's experience, some exceptional conditions constitute limitations to the performance of ureteroscopy. Ureteral fixation to the retroperitoneum may limit its mobility and passage of the ureteroscope. The presence of a large median prostatic lobe may render the insertion of the ureteroscope impossible. Urethral stricture or stricture of the bladder neck also limits the use of this procedure. Cystocele, previous radical pelvic surgery, radiation therapy, retroperitoneal fibrosis, neoplasm, or previous ureteral surgery may produce ureteral angulation or a limitation in ureteral mobility preventing ureteral straightening necessary to the passage of a rigid or even a flexible ureteroscope. Patients must be informed about the possibility of prolonged general anesthesia, about the complications that may occur, and that instrumentation will eventually require the use of catheters or percutaneous nephrostomy. They also must be aware that failure of the procedure or an injury to the ureter may determine the need of open surgery.

ALTERNATIVE THERAPY
Other diagnostic modalities available include antegrade endoscopy via a tract through the kidney, open exploration, and obtaining biopsies or ureteral cytologies via a cystoscope with radiologic guidance. Alternative treatments include open surgery and lithotripsy.

SURGICAL TECHNIQUE
Instrumentation When using ureteroscopy for the treatment of urinary calculi, a preoperative plain radiography of the abdomen is essential to ascertain the presence and location of the stone to be treated. Ureteroscopy must be done in an operating room or in an interventional radiology room. Essential equipment includes an operating table with a radiolucent top, and fluoroscopic equipment, preferably a C-arm. The table must have leg supporters allowing the patient to be positioned in a lithotomy position. The lower limbs must preferably be in an asymmetrical position, with the contralateral limb to the ureter to be treated in an accentuated abduction and flexion so that the mobility of the ureteroscope is not limited. A 0.025-in. teflon-coated straight floppy-topped guidewire, 150 cm long, must be available as well as an open-ended 6-F ureteral stent. Also needed are double-pigtail catheters made of different materials and consequently varying in hardness and diameters in order to allow choice of the most adequate one for each case. Radiopaque contrast must also be available for contrasting the ureter. There are three types of ureteral dilators: bougie, facial, and balloon. Most surgeons prefer balloon dilators because they are less traumatic. The balloon dilator is inserted over a guidewire and inflated to a diameter of 12 F for a few minutes. Ureteral dilation was a more commonly employed procedure at a time when small-caliber ureteroscopes were not produced. Ureteroscopes Currently, a large variety of ureteroscopes are available, each model having its own features and advantages. The ideal situation for an endourologist is to have available at least three ureteroscopes: a semirigid 7 F, with a working channel of 3 F; a semirigid 8.5 F, with a working channel of 5 F; and a flexible 7.5-F ureteroscope equipped with active movement and a working channel of 3.6 F. Ureteroscopy Ureteroscopies are preferably performed with semirigid ureteroscopes as these are easier to insert and offer a better vision. The procedure begins with the introduction of a guidewire up to the kidney. When the guidewire has been placed, 1. 2. 3. 4. 5. It directs the insertion of the balloon dilator as necessary. It indicates the location of the ureteral meatus because the vision field of the ureteroscope is more limited than that of the cystoscope. It elevates the upper portion of the ureteral meatus thus facilitating the insertion of the ureteroscope. It provides endoscopic direction assuring that the ureteroscope is progressing within the ureter. It makes possible the introduction of a catheter for ureteral drainage, in case of failure or adverse events, until some measure is taken or the ureteral complication is spontaneously corrected.

In the event of any difficulty regarding the progression of the guidewire, we recommend contrastation of the ureter to clarify why advancement has been hindered. Insertion of the ureteroscope is facilitated by infusion of saline solution under pressure. We use a pressure cuff for introduction of the irrigation fluid at a pressure of 150 mm Hg (which actually represents a lower pressure at the extremity of the ureteroscope in view of the resistance exerted by the instrument channel). The ureteroscope should be inserted under the guidewire. This helps in elevating the ureteral meatus and straightening the ureter along its submucous course ( Fig. 114-2).

FIG. 114-2. Introduction of the ureteroscope may be facilated by the guidewire and slight rotation movements.

Passage of the ureteroscope must be done very carefully, with attention to its correct advancement in the ureter. The surgeon is guided by endoscopic impression and by detailed observation of the ureter mucosa, as well as by the guidewire. When progression of the ureteroscope stops, the instrument should be rotated circumferentially to obtain a better presentation of its extremity or to carry out further balloon dilation of the constricted segment. The surgeon should never apply force to the ureteroscope. This could lead to perforation or avulsion of the ureter ( Fig. 114-3).

FIG. 114-3. The progression of the ureteroscope must be assured to avoid perforation or avulsion.

Flexible ureteroscopes are more difficult to pass but with a good dilation of the ureteral path and a guidewire the procedure becomes easier. However, the use of a sheath previously introduced over the semirigid ureteroscope greatly facilitates the insertion of the flexible ureteroscope through this sheath ( Fig. 114-4).

FIG. 114-4. Flexible ureteroscope inserted through a sheath allows the direct vision of renal calices.

Fluoroscopic control enables the surgeon to supervise the advancement of the ureteroscope, and to evaluate the distance between the instrument and the site of the ureter to be reached. Ureteral Calculus Manipulation Once a ureteral calculus has been visualized, if its diameter does not exceed that of the ureteroscope, it can be removed with grasping forceps or with a Dormia basket. The Dormia basket allows the calculus to be better grasped. However, in large calculi the use of a Dormia basket is contraindicated as it may be retained. The extremity of the Dormia basket is inserted closed beside the ureteral calculus. Then the basket is opened under fluoroscopic control and slowly retracted. It may require several to-and-fro movements and rotation until the calculus is trapped. There is no need to close the basket. When retracted, the stone tends to remain in the opposite conic portion of the basket. The calculus is retracted until it contacts the ureteroscope; both stone and instrument are removed together under endoscopic direct vision. Ureteroscopic Calculus Fragmentation When the calculus is larger in diameter than the ureteroscope, fragmentation must be considered. Several energy sources are available to fragment urinary calculi: ultrasonic, laser, electrohydraulic, and pneumatic. We prefer the pneumatic energy source because it is highly efficient and is equipped with probes of 0.8 and 1 mm that can be inserted through working channels in 7- and 8.5-F ureteroscopes, respectively. 1 Pneumatic energy sources traumatize the ureteral wall less frequently than the ultrasonic and electrohydraulic probes. They are also much less expensive than the laser. The larger fragments are removed as described above. Postureteroscopy Management A retrograde pyelography performed at the end of the procedure allows the diagnosis of any perforation and facilitates the eventual placement of a ureteral catheter. The need for urinary tract drainage, postureteroscopy, depends on the trauma caused by the procedure. Following a ureteroscopic procedure involving minimal trauma, a draining catheter is frequently unnecessary. In conditions of more significant traumas, with ureter perforation, mucosa laceration, large edemas caused by impacted calculi, excluded or largely dilated kidneys, we stent the ureter with a double-J catheter for about 4 weeks. In intermediary situations we recommend the maintenance of a ureteral catheter for 48 hours. Hematuria Hematuria continues to be a diagnostic challenge to urologists. Gross hematuria requires invasive investigation when its cause cannot be determined by imaging or cytologic tests. Ideally, ureteroscopy must be performed in the presence of bleeding. Once the hematuria is lateralized the ureteroscopic examination is carried out. Care must be taken not to damage the mucosa with the guidewire, which also must not exceed the limits of the renal pelvis. Any trauma to the kidney could generate misdiagnosis of the cause of the hematuria. The procedure is initiated with the semirigid ureteroscope coated with a sheath to facilitate insertion of the flexible ureteroscope. This allows examination of the renal pelvis and calices in an attempt to clarify what is causing the bleeding. The calices should be inspected according to the sequence upper, medial, and lower because their presentation is easier, and to avoid any injury due to ureteroscope flexion mimicking a pathologic finding. Biopsies of the mucosa of the ureter may lead to the histologic diagnosis of its cause. Cauterization of the eventual bleeding site may stop the hematuria. Ureteral Tumors Ureteral and pelvic neoplasms can be diagnosed and biopsied by ureterorenoscopy. When they do not fill the entire ureteral lumen they can be treated through electrocauterization using the Bugby catheter, by the Nd-YAG laser, or by resection of the tumor with a ureteroresectoscope. 5 Ureteral Strictures In ureteral strictures, ureteroscopy may permit the introduction of a guidewire, especially when no other means can be used. A balloon dilator is inserted over the guidewire and the ureteral dilation is carried out under fluoroscopic control. The procedure is then complemented by an incision with a scalpel through the ureteroresectoscope or even by insertion of the cutting device through the 5-F working channel of the ureteroscope. 3 Stenting with a double- J catheter must be maintained for at least 4 weeks.

OUTCOMES
Complications Complications of ureteroscopy can be divided into early and late complications regarding their onset, and into major or minor according to the extent to which they

compromise the procedure and the urinary tract (Table 114-2 ). The complication rate ranges from 5% to 14% 2,4,7,8 and depends on the equipment, the surgeon's experience, and the size and location of the calculus.

TABLE 114-2. Complications of ureteroscopy

Ureteral injury is the most common complication from ureteroscopy. It has been reported that perforation occurs in 28% of cases, whereas avulsion occurs in up to 8%. While most perforations can be managed by placement of a stent, major perforations may require percutaneous drainage and possible open repair, and avulsions will require open repair. Ureteral strictures may result from perforation or dilation, and have been reported in up to 5% of patients. When performing ureteroscopy for stones, the stone may be lost in the retroperitoneum. If the stone cannot be retrieved, it can be left in place and the perforation drained with a ureteral stent. Results When used as a diagnostic modality, ureteroscopy can visualize the entire collecting system in over 90% of patients. In patients with a filling defect, the most common findings at ureteroscopy are transitional cell carcinoma and stone, varying from 20% to 40% each. The most common reason that ureteroscopy is utilized and the highest therapeutic success rates are seen in ureteral stones. Stones of the distal ureter are successfully removed or destroyed ureteroscopically in over 90% of cases in the distal ureter and in over 60% of the time in the proximal ureter. CHAPTER REFERENCES
1. Chambo JL, Mitre AI, El Hayek OR, Brito AH, Arap S. Experiencia com um novo tipo de litotridor intracorporeo: litotridor balistico-pneumatico Swiss Lithoclast. J Bras Urol 1996;22:68. 2. Francesca F, Scattoni V, Nava L, Pompa P, Grasso M, Rigatti P. Failures and complications of transurethral ureteroscopy in 297 cases: conventional rigid instruments vs. small caliber semirigid ureteroscopes. Eur Urol 1995;28:112. 3. Levy JB, Van Arsdalen KN. Ureteral and ureteroenteral strictures. AUA Update Series 1994;13(29):230. 4. Lytton B, Weiss RM, Green DF. Complications of ureteral endoscopy. J Urol 1987;137:649. 5. Martinez-Pinero JA, Matres MJG, Martinez-Pinero L. Endourological treatment of upper tract urothelial carcinomas: analysis of a series of 59 tumors. J Urol 1996;156:377. 6. Perez-Castro Ellendt E, Martinez-Pinero JA. La ureteroscopia transuretral. Un actual proceder urologico. Arch Esp Urol 1980;5:1. 7. Shi-Chung Chang, Chak-Man Ho, Hann-Chorng Kuo. Ureteroscopic treatment of lower ureteral calculi in the era of extracorporeal shock wave lithotripsy: from a developing country point of view. J Urol 1993;150:1395. 8. Stackl W, Marberger M. Late sequelae of the management of ureteral calculi with the ureteroscope. J Urol 1986;136:386.

Chapter 115 Percutaneous Lithotomy Glenn’s Urologic Surgery

Chapter 115 Percutaneous Lithotomy
Joseph W. Segura

J. W. Segura: Department of Urology, Mayo Clinic, Rochester, Minnesota 55905.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Percutaneous Access Technique of Dilation Nephroscopic Examination of the Kidney Stone Removal Ultrasound Electrohydraulic Lithotripsy Lasers Mechanical Impactors Completing the Procedure Outcomes Complications Results Chapter References

Percutaneous stone surgery was introduced in the early 1980s. It represented a significant revolution in the management of stones and the first attempt at minimally invasive surgery. Though it has been largely supplanted by the introduction of extracorporeal shock wave lithotripsy (ESWL), it allows destruction or extraction of stones without the morbidity of open stone procedures.

DIAGNOSIS
The diagnosis of renal stone is made radiographically. The stone is frequently associated with microscopic (or gross) hematuria and flank pain. Other symptoms include persistent urinary infections. Since most stones contain calcium, KUB or intravenous pyelography (IVP) will frequently reveal the stone. Uric acid stones and other less radio opaque stones may require renal ultrasonography or computed tomographic (CT) scans. It is important, before embarking on any planned stone surgery, however, to perform a functional study of the kidney to assess both the contralateral renal unit and the relative function of the kidney in question. This information can be inferred during an IVP, but any questions should be resolved with a radionuclide scan.

INDICATIONS FOR SURGERY
Percutaneous methods for stone removal are suitable for any stone in the kidney or ureter, although today these procedures are used primarily for large stones, stones associated with an obstructive uropathy, as a salvage procedure for failed shock wave procedures and other in situations where ESWL is not indicated. 1 The only absolute contraindication to percutaneous stone removal (PL) is an uncontrolled bleeding diathesis. Some patients may have anatomic abnormalities that make access impossible or unsafe, such as splenomegaly or a greatly enlarged colon. When uncertainty exists, a CT scan can often clarify the situation.

ALTERNATIVE THERAPY
Alternatives to percutaneous stone surgery include open stone surgery, endoscopic stone manipulation, ESWL, and chemolysis.

SURGICAL TECHNIQUE
The procedure may be divided into two parts: (a) percutaneous access and (b) stone removal. Arguments as to whether urologists or radiologists should perform the access and where this should happen, as well as disagreements as to the temporal relation between them, are tiresome, futile, and of little significance. Solutions to these questions will arise from the realities of one's local situation. The procedures outlined below have been used at the Mayo Clinic in some 3000 procedures and have been found to work well. Percutaneous Access Because successful access requires identification of the “target,” i.e., the kidney, it must be possible to identify this either by seeing the stone on fluoroscopy or by the presence of contrast medium in the collecting system. If difficulty is anticipated, a ureteral catheter should be passed as a preliminary procedure through which contrast can be injected. The patient is positioned prone on the x-ray table. Generally, C-arm fluoroscopy is preferred because the beam may be angled to assist in identifying the proper approach into the kidney ( Fig. 115-1). If access only is the goal at this time, the procedure may be performed under local anesthesia supplemented by intravenous sedation as necessary.

FIG. 115-1. The patient is placed prone on the table. The C arm is preferred because it can be angled to demonstrate the relation of the collecting system to the access needle.

The patient is prepared and draped, at which point a decision must be made as to the most desirable approach to the stone or stones in question. These points, together with the intrarenal anatomy and whether obstruction is present, will determine the access point. 3 Access is easiest and probably somewhat safer through the lower pole calyx. In our experience, unless there is a reason to visualize the ureter directly with a rigid instrument, such as a concomitant UPJ obstruction, associated ureteral stone, or the anticipation of large volumes of debris going down the ureter, most stones can be dealt with adequately by approaching any posterior calyx in the lower half of the kidney. Tracts placed through the upper half of the kidney facilitate access to the ureter but increase the risk of pleural injury. Multiple tracts may be necessary, particularly if the stone is associated with intrarenal obstruction. Placement of tracts is usually easier at this time than after stone removal has started

and any tracts that are thought to be necessary should be placed. They can be removed later if not needed. Entry should be through a posterior calyx, as access through an anterior calyx is usually unsafe owing to overlying intraabdominal structures. Under fluoroscopic control, a point is identified on the patient's back approximately midway between the tip if the 12th rib and the paraspinous muscles ( Fig. 115-2).

FIG. 115-2. In general, access through the lower half of the kidney is sufficient for most stones, particularly if flexible nephroscopy is available to facilitate vision down the ureter.

The C-arm is angled so that the axis of view is oriented down the anticipated tract into the kidney. After a nick is made in the skin, a 22-gauge needle is placed directly into the collecting system. We have preferred the Cope introducing system (Cook Urologicals), which includes this needle, its obturator, and all of the accessories necessary to obtain access. The needle is passed through the flank, into the kidney and through the appropriate calyx. The obturator is removed and urine aspirated. If urine is returned, contrast is injected to confirm proper placement. When the operator is satisfied with the approach, a small wire is placed through the needle into the collecting system, a catheter then placed over the wire, the wire removed, and a 0.015-in. guidewire passed through the catheter and as far down the ureter as practical. Dilation continues until a larger guidewire may be placed to act as the access wire. Any reasonably stiff wire will do; we use 0.038 in LR torque wires (Cook). In our institution, the wire is then taped to the patient's flank and the patient moved to the operating suite, anesthetized, after which the tract is fully dilated and stone removal begun. Otherwise, the operator proceeds with dilation followed by stone removal. Technique of Dilation Several methods of tract dilation are available. We have preferred to use flexible fascial dilators (Cook) manufactured in even French sizes from 6 to 24 Fr ( Fig. 115-3). Less flexible dilators are used to 36 Fr. These dilators will tend to follow the course of the guidewire and are less likely to kink the wire. 2 Using the Amplatz system, an 8-Fr stent is passed over the guidewire and the Amplatz dilators passed sequentially until the desired tract size is reached, usually 30 to 36 Fr ( Fig. 115-4). A sheath is passed over the last dilator into the collecting system and this functions both to tamponade the tract and to permit access into the collecting system. The tract may be dilated with a balloon and the Amplatz sheath passed over the balloon. A disadvantage of this approach is the cost of the balloon; however, the kidney and tract are “traumatized” only once and this approach may be done slightly more quickly than the alternatives. Rigid steel dilators are used when scar tissue or other problems make dilation difficult. While they are the most cost-effective, we find them somewhat more traumatic and they tend to kink the guidewire.

FIG. 115-3. Flexible dilator inserts over the guidewire into the collecting system. As long as the wire is in the proper position the dilator will follow the correct tract.

FIG. 115-4. The set of Amplatz dilators includes an 8-Fr catheter that passes over the guidewire. Each guidewire is passed sequentially until the desired tract size is reached. This is most often 30 to 36 Fr. The sheath is placed in the collecting system under fluoroscopic guidance and the dilator itself is removed.

After dilation has proceeded to about 14 Fr, a 10-Fr triple-lumen introducer (Bard) is passed over the guidewire and the inner sheaths removed ( Fig. 115-5). A second, safety wire is passed through this sheath down the ureter, and the outer sheath then discarded. The reserve wire is draped off to the side while dilation proceeds over the original wire. The size to which the tract should be dilated depends primarily on the size of the nephroscope. We use 28-Fr nephroscopes and use the nephroscope sheath to tamponade the tract ( Fig. 115-6). If the Amplatz system is used, the minimum tract size should be 30 Fr with larger sizes being more common.

FIG. 115-5. (A) This introducing system is composed of three concentric sheaths, the largest being 10 Fr (Bard). (B) The sheaths are passed over the guidewire following the course of the guidewire down the ureter. (C D) The inner two cannulas are then removed leaving a 10-Fr outer cannula. There is enough space through this outer cannula to put a second safety wire in place.

FIG. 115-6. (A) The nephroscope sheath with the obturator in place is placed over the guidewire into the kidney and into the renal pelvis under fluoroscopic control. (B) The obturator is removed and the telescope passed over the guidewire into the kidney. At this point the operator determines the location by direct visualization.

While there is no overwhelming reason to choose one system over another, there are some pros and cons to consider. The Amplatz sheath permits irrigating fluid to exit through the sheath and around the nephroscope, which considerably reduces the risk of extravasation ( Fig. 115-7). However, because there is little hydrodistention of the collecting system, visualization may be compromised. It is also my impression that the absence of hydrodistention results in more bleeding than occurs in closed systems. There is no doubt that larger fragments can be plucked from the kidney through the sheath than through a nephroscope sheath alone.

FIG. 115-7. The Amplatz sheath is positioned in the renal pelvis and the dilator removed. The working guidewire is apparent as it goes through the sheath and the safety wire is draped out to the side. The telescope can be passed over the working wire; once the proper position has been identified the working wire is removed.

Use of the nephroscope sheath to tamponade the tract requires the irrigating fluid to go down the ureter into the bladder (or out the suction ports). This results in hydrodistention of the collecting system, which facilitates intrarenal inspection and seems to tamponade small bleeders. Furthermore, because the irrigating fluid is recovered only from the Foley catheter and the suction ports, measuring and keeping track of the input and output (I&O) is facilitated. There is, nevertheless, an increased incidence of extravasation that doubtless contributes to the number of secondary procedures. Nephroscopic Examination of The Kidney Rigid nephroscopes are available from several vendors. Wolf, Storz, Circon-ACMI, and Olympus make excellent instruments, and while there is little reason to prefer one over the other, we continue to use Wolf equipment for our percutaneous procedures. The nephroscope is passed over the working wire with the obturator in place (Fig. 115-6). The first move is to inspect the collecting system to be sure that in fact one is in the collecting system and to become oriented in relation to the stone and to the ureteropelvic junction (UPJ). The working wire should not be removed until the operator is certain that the nephroscope is properly positioned. Fluoroscopy should be used to address any uncertainties. There is usually blood clot in the renal pelvis that must be removed before proceeding. Fiddling with the irrigating fluid and using suction may clean out the clots, but we often resort to the ultrasonic probe as a quick and effective way to clean out the collecting system. Flexible nephroscopy is used primarily as an adjunctive procedure to rigid nephroscopy and to inspect the kidney for residual stone fragments. We use 15-Fr Olympus flexible nephroscopes (which are the same as flexible cystoscopes). It can be surprisingly difficult to be certain of one's location using flexible nephroscopy alone and fluoroscopy is essential to remain oriented as one inspects the collecting system. These instruments have relatively small working channels (up to 5 Fr), but these will accommodate a wide range of instruments including baskets, graspers, lasers, and so forth. Stone Removal Stones up to 7.0 to 8.0 mm in size can be grasped and removed through a 28-Fr nephroscope and stones of up to 1.0 cm can be extracted intact through a 36-Fr Amplatz sheath. Such stones are usually managed by ESWL, but PL may be preferred in certain situations The most common reason to grasp and extract stones is to remove fragments resulting from intrarenal lithotripsy. Many types of grabbers and extractors are available. I prefer reusable alligator forceps employed through the rigid nephroscope; baskets are usually used with the flexible nephroscope, especially to remove stones in the ureter. Ultrasound Most stones must be fragmented before they can be removed. Some form of intrarenal lithotripsy is necessary to reduce these large stones to an extractable size. Ultrasound is unquestionably the most efficacious technique available because the stone fragments are removed as they are created. All commercial ultrasonic generators work on the same principle. A piezo-ceramic crystal is fixed to a hollow steel probe and an electric current passed through the crystal. This causes the probe to vibrate at ultrasonic frequencies ( Fig. 115-8). With the probe placed in contact with the stone, there is a Jack-hammer effect with the stone being gradually broken up into small pieces which are aspirated by continuous suction up the hollow probe ( Fig. 115-9). This system is particularly useful for relatively fragile stones,

such as struvite or calcium oxalate dehydrate, and remains practical for large and less fragile stones, such as cystine, for the same reason. Some stones, especially large calcium oxalate monohydrate stones, do not readily fragment with ultrasound into pieces that are small enough to be aspirated up the probe. Also, because the probe is rigid, not every area of the kidney can be reached, at least through a single tract.

FIG. 115-8. (A) A 28-Fr Wolf rigid nephroscope with an ultrasonic probe in place. Wide-lens optics are used throughout allowing excellent visualization with a second available accessory port. With the ultrasonic probe in place there is usually insufficient room for another device. (B) The ultrasonic generator illustrating the end of two different probes; one is a rotating drill bit, the other a solid ended probe. My preference is to use the solid ended probe. The ultrasonic probe screws on to the piece of ceramic crystal and this must be as tight as possible using the available tightener. If the connection is not tight, much energy will be lost at the junction of the probe and the piece of surrounding crystal.

FIG. 115-9. (A) The nephroscope is in the tract over the guidewire. Often one can feel the stone at this point (B). The working wire is removed and the ultrasonic probe is placed directly on the stone. Positive contact must be present or the probe will not fragment the stone. (C) With the probe on the stone, the stone is jack hammered into small pieces, these being aspirated through the hollow channel in the center of the probe. Irrigating fluid is continuously flushed around the sheath to both cool the probe and to aid in irrigating/aspirating the stone fragments.

When the stone is large, the first move should be to create sufficient space so that a nephrostomy tube can be placed in the collecting system, if necessary ( Fig. 115-10). If the patient has not previously been operated, it may be possible to rotate the rigid nephroscope to look down the ureter or elsewhere in the collecting system (Fig. 115-11).

FIG. 115-10. When removing a large stone such as a stag horn, the first move should be to remove enough material to make room for a nephrostomy tube should it become necessary to terminate the procedure for any reason.

FIG. 115-11. (A) The rigid nephroscope has been placed through a lower pole calyx. (B) The nephroscope is rotated cephalad at the same time that the end is oriented down the ureter. (C) One is able to look directly down the ureter and remove stone debris or otherwise inspect the ureter.

Electrohydraulic Lithotripsy When stone fragmentation with ultrasound proceeds slowly or if the stone cannot be readily fragmented into small pieces, some other method of fragmentation must be used. Electrohydraulic lithotripsy (EHL) is an effective, inexpensive method that works well for most stones. These probes are bipolar electrodes with the conducting wires separated by insulation and exposed at the tip. An electric current is passed through the wires and a spark generated at the end. The probe is positioned about 1 mm from the stone surface, preferably near an irregularity on the surface and the generator activated. The spark vaporizes a small amount of water forming a gas bubble and subsequent shock wave which passes into the stone, fragmenting it much like SWL.

Probes are available in various sizes, from 1.9 Fr to 9.0 Fr. Large probes work well through rigid instruments to break smaller pieces off the “main” stone. These can then be removed piecemeal with forceps or by the use of ultrasound (our own preference). The smaller probes are commonly used with flexible nephroscopes to fragment calyceal stones or stones in locations not accessible with the rigid nephroscope ( Fig. 115-12).

FIG. 115-12. The flexible nephroscope has been flexed to visualize a stone in the adjacent calyx. Flexible nephroscopy is a very useful adjunct, allowing inspection of most areas of the collecting system.

Lasers Lasers are less useful in the kidney as a primary method of stone destruction because they do not fragment stones as rapidly as ultrasound or EHL, and one is still faced with the prospect of fragment removal. They are most useful with the flexible nephroscope as a lithotripter for inaccessible stones. Coumarin-dye lasers have been available for the past 10 years or so (Candela Corp., EDAP-Technomed). With the laser probe placed against the stone, the laser is fired in nanosecond bursts, achieving very high power over a short time frame. As the energy is absorbed the stone is fragmented. Coumarin-dye lasers are ineffective against cystine because cystine is the wrong color for optimal energy absorption. Recently, holmium lasers have become available for use in the United States (Coherent, Xintec). These lasers have several advantages in that warm-up time is short and they will fragment essentially any stone by a process of vaporization and fragmentation. Their biggest drawback is that destruction of large stones can be tedious. Mechanical Impactors A mechanical impactor (Lithoclast) has been available in Europe and Canada for some time and recently has been approved for use in the United States. The device uses compressed air to impact the end of a steel rod. The impact is transmitted to the end of the rod that fragments the stone. The device is very effective against large stones, although when employed to reduce smaller stones to still smaller fragments its efficacy is compromised by a propensity to knock the stones around. One is still faced with the problem of stone removal, and attachments to the Lithoclast for the purpose of removing fragments by suction irrigation have at this writing not been very successful. Completing the Procedure Lithotripsy and stone removal continues until the patient is stone-free or until the procedure must be stopped for some reason. A stone-free state is determined by direct inspection with the rigid nephroscope, supplemented by flexible nephroscopy if indicated. We often endoscope the ureter to see if any fragments are present and frequently pass a basket down the ureter to remove small pieces. Dense residual stones are usually obvious on fluoroscopy, but a plain film taken while the patient is still on the table is more reliable. It may be necessary to stop the procedure because of bleeding that has begun to obscure vision, a discrepancy in I&O that cannot be accounted for by spillage, or simply a perception that the procedure has gone on long enough and should be stopped. Whatever the reason, one should appreciate that this is one of the advantages of endourology, that the procedure can be stopped and continued another day. In our practices, repeat procedures are usually done 48 hours later. By then bleeding has stopped and holes made in the collecting system have sealed off. A second guidewire is left in the collecting system as the nephroscope is removed. Our practice is to place a 22-Fr Foley catheter with a hole cut in the end (essentially a Councill catheter) over this wire, positioned so that the end is in the renal pelvis, and then to inflate the balloon with 2 to 3 ml of contrast so that the position can be accurately determined. Once satisfied with tube placement, a straight open-ended 6- to 7-Fr catheter is placed over the safety wire down as far as the pelvic ureter. This prevents edema of the ureter from closing the ureter; it also helps keep the ureter dilated, facilitating passage of any clots or unrecognized small fragments. It also facilitates reentry if the nephrostomy tube falls out. The tubes are stitched to the patient's flank and 12.5 g of mannitol given as a diuretic. It has been suggested recently by Bellman that no nephrostomy tube be left in place in order to avoid the morbidity of the tube. 1 While this might be practical in a clinical practice with relatively straightforward cases, I am reluctant to consider this with the complicated patients that compose our practices today. Leal has suggested that an 8-Fr catheter be left in place, so that if reentry is necessary access could easily be achieved while avoiding the morbidity of the larger tube. 4 Plain films and nephrostograms are obtained 24 to 48 hours afterward. If all is well, the stent is removed and the nephrostomy tube clamped. If the patient tolerates this, the nephrostomy tube is removed and the patient discharged with a drainage bag over the tract site. This will seal within 24 hours, unless the ureter is blocked by a clot or a stone fragment.

OUTCOMES
Complications One can conveniently if somewhat artificially divide complications of PL into those related to access and those related to stone removal. Injury to other organs is the most common problem associated with access, excepting failure to achieve access at all. The pleural cavity is easily transgressed particularly with an upper pole approach. Once identified, this problem is readily managed with a chest tube. Despite the somewhat intimate relationship between the kidney and colon, injury to the colon is uncommon. We have had two cases, both of which had chronically enlarged colons after intestinal bypass. These cases were both diagnosed postoperatively at the time of the nephrostogram. CT scan revealed that the tract had entered the posterior lateral colon. In both patients, the nephrostomy tube was pulled back into the colon, forming a “colostomy” tube, and a double-pigtail stent placed from below. These problems then resolved without sequelae, but more serious trauma might require open repair. Injury to the spleen has been reported and required splenectomy, whereas duodenal trauma responded to nasogastric tube placement. The kidney itself may be traumatized if the access does not traverse the parenchyma of the kidney ( Fig. 115-13).

FIG. 115-13. The preferred tract is in general a posterior calyx that can be accessed through the thickest portion of the parenchyma. Access through a thinner area may cause tearing as it did in this case, making extravasation more likely.

While not strictly speaking a complication, failure to render the patient stone-free is the most common undesirable result of PL. These fragments may obstruct the ureter in the immediate postoperative period or later may become themselves surgical stones. The most significant risk to the patient is arterial bleeding either at the time of the PL or in the postoperative period. This results from injury to a branch artery either at the time of access or at the time of lithotripsy. Sometimes this is immediately obvious with large quantities of red arterial blood under pressure passing out the nephroscope or out the nephrostomy tube. More commonly, this is diagnosed in the immediate postoperative period. This patient should have an immediate arteriogram with embolization of the offending vessel. Surgery should be avoided if possible because nephrectomy will be the likely result. The overall risk of this problem in large series is about 0.5% to 0.6%. Venous bleeding during the procedure is suggested by large amounts of venous blood appearing in the pelvis or nephroscope when the irrigating fluid is not running and which abates with irrigation. Injection of contrast will often demonstrate a connection to the venous system. This is treated by nephrostomy tube placement, clamping the tube for 30 to 40 minutes in order to tamponade the bleeder with the concomitant administration of intravenous mannitol. We have performed this maneuver many times with success. The risk of transfusion is a function of many variables and in most recent series has been about 4% to 5%. One should make an attempt to keep track of the I&O by totaling the Foley catheter drainage and drainage from the suction and drapes, and then comparing this to the amount of irrigating fluid used. While there will be inevitable inaccuracies, one should consider stopping the procedure if more than 500 ml or so cannot be accounted for by spillage, and so forth. Perforation of the collecting system with an ultrasonic probe, dilator, or perhaps by pushing a sharp fragment through the pelvic wall can occasionally occur. If the procedure can be completed in a reasonable length of time, it should be; otherwise a nephrostomy tube should be placed. The hole will be sealed off 24 to 48 hours later when the procedure can be completed. Results Stone-free rates vary from 85% to 99% and are a function of the complexity of the problems being treated. 5,6 Percutaneous methods are appropriate at least theoretically for any stone anywhere in the kidney. In the current era, PL is used where SWL is less effective. These include patients with large stones, patients in whom the risk of multiple SWLs and ancillary treatments make SWL less desirable, as a salvage procedure for other modality failures, and for stone removal where concomitant management of associated obstruction is indicated. PL can also be employed as part of a sandwich therapy for complicated staghorn or struvite stones, where ESWL alone has been shown to be less effective as a monotherapy. 7,8,9 and 10 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Bellman GC, Candela J, Davidoff R, Gerspach J, Kurtz S. “Tubeless” percutaneous renal surgery (Abstract). J Urol 1997;157(4):45 (Suppl). Clayman RV, Casteneda-Zuniga WR, Hunter DW, Miller RP, Lange PH, Amplatz K. Rapid balloon dilatation of the nephrostomy tract for nephrostolithotomy radiology. 1983;147:884–885. Coleman CC, Kimura Y, Casteneda-Zuniga WR, et al. A systematic approach to puncture-site selection for percutaneous urinary tract stone removal. Semin Intervent Radiol 1984;1:42–58. Fernandez AZ, Durham NC, Leal JJ. The use of a modified 8F pediatric nasogastric tube as a nephroureterostomy tube in patients following percutaneous nephrolithotomy (Abstract). J Urol 1997;157(4):45 (Suppl). Lam HS, Lingeman JE, Mosbaugh PG, et al. Evolution of the technique of combination therapy for staghorn calculi: a decreasing role for extracorporeal shock wave lithotripsy. J Urol 1992;148:1058. Meretyk S, Gofril O, Gafni O, et al. Complete staghorn calculi: random prospective comparison between extracorporeal shock wave lithotripsy monotherapy and combined with percutaneous nephrostolithotomy. J Urol 1997;157:780–786. Patterson DE, Segura JW, Benson RC Jr, et al. Endoscopic evaluation and treatment of patients with idiopathic gross hematuria. J Urol 1984;132:1199–1200. Patterson DE, Segura JW, LeRoy AJ. Long-term follow-up of patients treated by percutaneous ultrasonic lithotripsy for struvite staghorn calculi. J Endourol 1987;1:177–180. Segura JW, Patterson DE, LeRoy AJ, et al. Percutaneous removal of kidney stones: review of 1000 cases. J Urol 1985;134:1077–1081. Segura JW, Patterson DE, LeRoy AJ, May GR, Smith LH. Percutaneous lithotripsy. J Urol 1983;130:1051–1054.

Chapter 116 Endopyelotomy for Ureteral Pelvic Junction Obstruction Glenn’s Urologic Surgery

Chapter 116 Endopyelotomy for Ureteral Pelvic Junction Obstruction
Culley C. Carson III

C. C. Carson III: Division of Urology, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599.

Diagnosis Indication for Surgery Alternative Therapy Surgical Technique General Principles Antegrade Endopyelotomy Retrograde Endopyelotomy Acucise Endopyelotomy Outcomes Complications Results Chapter References

The normal ureteral pelvic junction (UPJ) provides a dependent funnel-shaped exit for urine leaving the kidney and entering the ureter. Peristaltic activity that originates in the renal upper pole calyx is conducted through the renal pelvis and down the ureter. Narrowing of the UPJ can occur as a result of extrinsic, intrinsic, or vascular defects producing a restriction of urine flow and peristaltic activity. The resultant compensatory hypertrophy of the renal pelvic wall and subsequent renal pelvic dilation leads to the symptoms of UPJ obstruction and, in some cases, a diminution of tubular function with alteration in urinary concentration and progressive renal failure. Urinary stasis in the renal pelvis can produce calculi or result in urosepsis.

DIAGNOSIS
Selection of patients for treatment of UPJ obstruction is dependent on documentation of obstruction at the level of the UPJ and its association with symptoms and radiographic findings. Patients with primary UPJ obstruction requiring treatment vary in age, but most are younger than 20 years. The most common presenting symptoms include urinary tract infections, intermittent hematuria with or without history of flank trauma, and intermittent flank pain. Neonates may present with a large flank mass, vague abdominal discomfort to palpation, or vomiting or diarrhea that may be associated with urinary tract infections. Frequently the symptoms of flank pain and hematuria are associated with increased fluid intake. For patients in whom UPJ obstruction is suggested, excretory urography is usually the first procedure performed although renal ultrasound performed as a screening procedure may suggest hydronephrosis. Confirmation of UPJ obstruction is best performed using a furosemide-stimulated renogram to produce high urinary flow rates. This procedure will not only document significant obstruction but also may reproduce flank discomfort from increased pelvic dilation associated with UPJ obstruction. This diuretic renogram, usually performed using diethylenetriamine pentaacetic acid (DTPA) and followed by intravenous administration of 0.3 to 0.5 mg/kg of furosemide in a well-hydrated patient, will document an obstruction-appearing pattern on renogram with curves of isotope concentration that do not decline 20 minutes after diuretic administration. However, this renogram may not be diagnostically valid for patients with severely compromised renal function because the diuretic response will be inadequate. Evaluation of the lower urinary tract is also important before surgical intervention for UPJ obstruction. While it is rare that distal ureteral abnormalities occur, if one is suspected it is important to eliminate the possibility of ureteral strictures with appropriate imaging studies. In children it is important to eliminate the possibility of vesicoureteric reflux and to document a normal distal ureter. Voiding cystourethrography and retrograde or antegrade pyelography should be performed before definitive repairs are undertaken.

INDICATION FOR SURGERY
Surgery should only be suggested after (a) adequate imaging studies have documented UPJ obstruction and eliminated the possibility of other ureteral abnormalities and (b) urine cultures have proven sterile. If renal drainage is shown to be adequate, if calices are of normal caliber, and if the patient has no pain, pyuria, renal calculi, or other symptoms of UPJ obstruction, then UPJ repair is not appropriate. A period of observation and repeated radiographic studies may be necessary to confirm the necessity for surgical intervention. When one or both kidneys demonstrate obstruction associated with symptoms, however, definitive corrective surgery should be planned.

ALTERNATIVE THERAPY
The choice of surgical approach to UPJ obstruction must be made on the basis of a combination of history, physical examination, and careful review of preoperative imaging studies. When a renal scan demonstrates minimal renal function and a satisfactory contralateral kidney, consideration can be given to nephrectomy of a poorly or nonfunctioning hydronephrotic kidney. Similarly, a kidney with multiple cortical abscesses or longstanding pyonephrosis should be removed if the contralateral kidney appears normal. Classic open repair of the UPJ with a dismembered pyeloplasty or other modified pyeloplasty should be considered if a ureterovascular defect is identified radiographically, suggesting a lower pole vessel crossing the UPJ. This defect is poorly treated with endopyelotomy. We have performed endopyelotomy previously on two patients with lower pole crossing vessels. Radiographic studies failed to demonstrate a ureteral vascular defect. Both of these endopyelotomies failed but had no significant morbidity from bleeding. Subsequent open repair demonstrated the lower pole crossing vessel and pyeloplasty was successfully performed without difficulty.

SURGICAL TECHNIQUE
General Principles The principles of endopyelotomy irrespective of approach are modeled after the descriptions of Davis and Smart in the late 1940s. The Davis intubated ureterotomy was a method of open pyeloplasty that was first described in 1943 as a way to correct UPJ obstruction. The technique involved passing a silver probe down the ureter through a pyelostomy incision. The ureter and UPJ were then incised over this probe using fine scissors to incise stenosis. A large-caliber drainage catheter or stent was placed across the area of incision and passed to the bladder. Separate nephrostomy drainage was maintained for 6 to 8 weeks. Davis's results were acceptable but his surgical procedure was supplanted by flap and dismembered pyeloplasties. During the healing following Davis's procedures, however, smooth muscle and mucosa of the ureter were documented to regenerate over the large indwelling stent, providing excellent UPJ drainage. Using these principles endoscopically, endopyelotomy has been performed with great success and major open surgery avoided. Early results demonstrated that a cold knife incision is preferable to electrocautery in producing better healing and less scarring with less thermal injury. 1,2 and 3 Cold knife incision preserves ureteral blood flow with reduced tissue destruction. Endopyelotomy can be performed via several approaches, the choice of which is determined by the patient's clinical situation and, to a lesser extent, by the surgeon's preferences. Antegrade Endopyelotomy

For antegrade endopyelotomy, adequate percutaneous nephrostomy access is the most critical step in successful UPJ repair. This access must be placed so that vision and manipulation of the UPJ is optimal. 2,3 The best percutaneous access is through a lateral infundibulum or upper pole renal calyx. Lower pole access, while excellent for stone removal, is generally inadequate for UPJ visualization and endopyelotomy. Once optimal percutaneous access has been achieved, an angiographic guidewire and catheter must be placed across the UPJ to control the incision and act as a safety wire for subsequent balloon dilation and stent placement. In cases of severe renal pelvic dilation and stenosis, antegrade guidewire passage with percutaneous access placement may be impossible. In this situation, a percutaneous access is placed and an initial retrograde guidewire passed cystoscopically prior to renal tract dilation. During this procedure, retrograde pyelography may be performed if necessary. Once a retrograde guidewire has been positioned in the renal pelvis, it can be retrieved through the nephrostomy tract before UPJ incision to allow adequate postincision stent placement. After the guidewire has been adequately placed through the UPJ, the patient is positioned prone and the renal access tract dilated for endoscope placement in the standard fashion. Use of dilation sheaths such as the Amplatz sheath facilitates endoscope change. The universal nephroscope is inserted and the UPJ is visualized and examined. If a retrograde safety guidewire has been passed it is grasped using endoscopic forceps and retracted through the percutaneous access tract. This retrograde-placed angiographic guidewire is exchanged for an antegrade guidewire by passing an open-ended ureteral catheter or safety wire catheter and exchanging for a new stiff guidewire under fluoroscopic control. The guidewire is maintained in position for the entire procedure to guide incision, dilation, and ultimate stent placement ( Fig. 116-1A).

FIG. 116-1. (A) Stenotic ureteropelvic junction (UPJ) with traversing guidewire. (B) Endopyelotome knife incising UPJ at posterolateral position, which is at 12 o'clock with the patient prone. (C) Traversing wire marking UPJ with periureteral adipose tissue visible.

If antegrade or retrograde passage of guidewires cannot be accomplished as a result of severe UPJ stenosis, more heroic techniques may be necessary. Methylene blue may be infused into the upper ureter using an open-ended ureteral catheter passed to the level of the obstruction. This area of obstruction may be observed with the nephroscope as methylene blue passes through the stenotic UPJ. The incision can be carried out through the methylene blue–defined tract. Because a safety guidewire is not present, however, this technique produces significant risks and avulsion of the ureter or unsuccessful stenting are possibilities. If methylene blue is visualized, initial incision should be followed by immediate direct vision placement of a safety guidewire to maintain control of the incised ureter and UPJ. Other possible procedures include retrograde placement of a fiberoptic light source or ureteroscope to the level of the UPJ. If a thin obstructing membrane is present at the UPJ, the light may be visible in the renal pelvis with the nephroscope light extinguished. Under these circumstances, a short incision may be carried out over the area of visible light and connection established. Because this is also a poorly controlled technique without a safety guidewire present, ureteral avulsion and further stenosis are significant possibilities. Once the UPJ is adequately visualized, clotted blood, debris, and any renal calculi are removed using the universal nephroscope and ultrasound probe before the UPJ incision is made. Cold knife incision can be carried out using the direct vision endopyelotome, which is a specially designed direct vision urethrotome for endopyelotomy and produces excellent results ( Fig. 116-1B). The endopyelotome consists of a working element and a scimitar-shaped hook blade knife that allows atraumatic passage of the knife and UPJ incision with retraction toward the operator. However, because the endopyelotome is not available in many hospitals, excellent results can be obtained using a standard direct vision urethrotome. This instrument visualizes the UPJ, controls the incision, and can be useful if an endopyelotome is unavailable. The incision is made by visualizing the angiographic guidewire passing through the UPJ. The cold knife is passed beside the angiographic guidewire and the incision is made on the posterolateral aspect of the UPJ to avoid damage to any aberrant lower pole vessels, which arise on the anterior surface of the ureter at the UPJ in as many as one-third patients with UPJ obstruction. Since patients are placed in a prone position for antegrade access, incision is carried out just off the 12 o'clock position. The cold knife is advanced through the UPJ and withdrawn producing a single full-thickness incision through the ureteral mucosa and ureteral wall. Several passes of the cold knife may be necessary to accomplish incision adequately, especially if significant scarring or a long area of stenosis is present. A full-thickness incision is essential to ensure adequate treatment. The incision is continued until periureteral adipose is clearly visualized ( Fig. 116-1C). Commonly, the UPJ will spring open when the incision is complete, especially in secondary UPJ obstructions. The nephroscope or endopyelotome can than be passed antegrade beyond the stenotic UPJ into the normal ureter. The incision should be continued beyond the area of stenosis into the most proximal portion of the normal ureter to ensure satisfactory results. An incision approximately 2 cm in length to include 0.5 to 1 cm of normal ureter is generally necessary for adequate UPJ incision. In order to evaluate results and ensure dilation of periureteral scarring, especially in secondary procedures, a 4-cm high-pressure angiographic dilation balloon is passed through the UPJ over the stiff angiographic safety guidewire after incision is complete. Guidance of this balloon is maintained fluoroscopically and overdilation of the UPJ is carried out to 15–18 Fr with a high-pressure dilation balloon using an inflation syringe to approximately 10 to 15 atmospheres. Fluoroscopic visualization of the balloon will confirm adequate incision. If waisting at the UPJ is observed, adequate incision has not been achieved and further incision distally or proximally must be carried out. Balloon dilation should be maintained for 5 to 10 minutes. After the balloon is deflated and removed, the angiographic safety guidewire is used to pass a 12- to 14-Fr ureteral stent that is advanced to the bladder. Special stents are available that measure 6 to 8 Fr distally and 12 to 14 Fr at the UPJ with a retention ring positioned in the renal pelvis. Stents used after endopyelotomy must include fenestrations and a position that allows the renal pelvis to drain externally or to the bladder ( Fig. 116-2). These stents are best controlled externally through percutaneous nephrostomy accesses. If drainage through the renal pelvis fenestrations is inadequate, an additional guidewire may be placed in the renal pelvis to position a self-retaining nephrostomy tube for additional drainage. This second tube should remain for 24 to 48 hours to allow drainage of any blood or other debris from the patient prior to the patient's discharge from hospital. The nephrostomy tube is then removed if the ureteral stent drainage is adequate. For adequate healing, the stent should remain in place for 6 to 8 weeks during which the patient may resume normal activity. If there is no extravasation of contrast material on nephrostogram and if the UPJ appears of adequate diameter at follow-up nephrostogram, the ureteral stent is retracted into the renal pelvis and removed. If drainage is adequate and the patient remains asymptomatic, the stent nephrostomy may be removed completely. Post incision follow-up, excretory urography or renography is performed approximately 6 weeks following stent removal ( Fig. 116-3). During the healing stage patients should be maintained on antibiotic prophylaxis for approximately 72 hours following tube removal.

FIG. 116-2. Endopyelotomy stent in place after endopyelotomy.

FIG. 116-3. (A) Pre-endopyelotomy UPJ obstruction on intravenous urogram (IVU). (B) Poststent removal IVU. (C) Drainage film 10 minutes after 10 mg of furosemide given.

Retrograde Endopyelotomy Endopyelotomy can also be performed through a ureteroscope by a retrograde technique. 5 Prior to attempts at ureteroscopic endopyelotomy, a double-J ureteral stent must be maintained for at least 10 to 14 days for renal drainage and passive dilation of the ureter and ureteral pelvic junction facilitating the ureteroscope passage. After a double-J stent has been in position for at least 2 weeks, the patient is positioned in lithotomy position and the ureteral stent is removed after placement of a stiff guidewire. All endopyelotomies are facilitated by the use of a stiff guidewire to limit guidewire kinking. A 5-Fr open-ended catheter is then positioned in the renal pelvis to provide drainage during the procedure. The guidewire is then removed because electrocautery is usually used to perform retrograde endopyelotomy and electrocautery in close approximation with a steel guidewire may result in unanticipated ureteral injury. Using standard techniques, the ureteral resectoscope is passed to the UPJ. Irrigation fluid is then changed from normal saline to 1.5% glycine for electrocautery. The UPJ is visualized and pulsations suggesting a periureteral crossing vessel can be helpful although rarely seen. The area of incision is posterolateral as previously described. Because the ureteroscope is being passed antegrade, the posterolateral position must be clearly identified prior to incision. A right angle cautery electrode is then placed through the resectoscope and using pure cutting current of approximately 50 W incisions are made posterolaterally until periureteral adipose tissue is seen. Incisions should be carried out proximal and distal to the UPJ obstruction as previously described for the antegrade technique. Fluoroscopic guidance is essential for this technique to ensure positioning of the ureteroscope and placement of the retrograde open-ended catheter. Following incision, UPJ balloon dilation is carried out prior to positioning of a double-J ureteral stent. An endopyelotomy Retromax stent is then inserted over the superstiff guidewire and maintained in position for 6 to 8 weeks. Because of the size of the endopyelotomy stent, it must be passed through a cystoscope sheath without optical element. The endopyelotomy stent is 14 Fr at the UPJ and 7 Fr at the bladder. Position of the stent is confirmed fluoroscopically prior to guidewire removal. Acucise Endopyelotomy A fluoroscopic method for ureteral pelvic junction incision has been developed using a specially designed dilation balloon with a cutting element attached to its side. The Acucise (Applied Medical, Laguna Hills, CA) balloon is positioned fluoroscopically and inflated with contrast medium demonstrating waisting of the ureteral pelvic junction obstruction. The Acucise balloon is a combination of a 3-cm, 24-Fr, low-pressure balloon with an external electrocautery wire that is activated by cutting electrocautery current (Fig. 116-4). The proximal Acucise handle consists of three elements: one arm for irrigation, one arm for balloon inflation, and a third to position the cautery wire under fluoroscopic control in a known position ( Fig. 116-5). Thus, the wire can be positioned posterolaterally with some accuracy. However, because torque sometimes changes this position, one cannot be sure of the guidewire position in all cases. The electrocautery wire is 2.8 cm in length with an exposed cutting surface 150 µm wide through which pure cutting current is applied at 75 to 100 W. Fluoroscopic guidance confirms positioning, waisting, and adequate incision of the ureteral pelvic junction. Because most investigators have demonstrated a lower pole crossing vessel in 13% to 16% of patients, endopyelotomy using the Acucise balloon must be carried out posteriorly. 1 The positioning as previously described is especially important in this situation.

FIG. 116-4. Acucise balloon with balloon and cautery incision wire.

FIG. 116-5. External control unit for Acucise balloon.

After the patient has been placed in the lithotomy position, retrograde ureterogram is carried out. A stent may be left in place over the ureteropelvic junction for 7 to 14 days to facilitate Acucise endopyelotomy. This stent placement, while not essential, facilitates balloon positioning. A stiff guidewire is placed under fluoroscopic control and positioned into the renal pelvis, which has been previously filled with contrast media. The Acucise catheter is then advanced under fluoroscopic control to the level of the UPJ. Additional contrast can be placed through the Acucise catheter side port to better visualize the renal pelvis. The Acucise cutting wire is positioned at the posterior or lateral UPJ under fluoroscopic control. The electrosurgical wire from the Acucise unit is then connected to electrocautery at a pure

cutting current of 75 to 100 W. The balloon is inflated with dilute contrast to visualize its position. Electrocautery is then activated and waisting will disappear after ureteral incision. Continued balloon dilation for 5 to 10 minutes following incision ensures adequate dilation. The balloon is then deflated and pulled to an area distal to the UPJ incision and additional contrast injected. For adequate incision to have occurred, extravasation should be seen at the area of incision. If extravasation is absent, consideration should be given for further incision. Following adequate incision, the Acucise catheter is removed and an endopyelotomy Retromax stent is positioned as previously described. The stent remains in position for 4 to 8 weeks. Because these stents have large proximal calibers, outpatient removal should be carried out using intravenous sedation.

OUTCOMES
Complications The most common complication is restricture or repeat stenosis. Restricture can be demonstrated on the poststent nephrostogram or on follow-up urography. It is thus important to follow patients with excretory urography, radio isotope renography, or nephrostography to confirm adequate postoperative results. Restricture or inadequate endopyelotomy will necessitate a second procedure. In some cases of failure, open pyeloplasty may be necessary. Open pyeloplasty is especially appropriate in those patients who fail endopyelotomy and are identified as having a crossing lower pole vessel. 4 Postoperative infection can not only adversely affect renal function but can also lead to failure of endopyelotomy. Infection significantly increases the incidence of pyelitis, ureteritis, and foreign body inflammatory response. Leakage of infected urine through the area of incision may further result in failure by an increased incidence of periureteral cicatrization, prolonged incisional leakage, and possible retroperitoneal abscess formation. It is imperative, therefore, that patients undergoing endopyelotomy by any approach receive appropriate perioperative antibiotic prophylaxis as well as postoperative antibiotic coverage during the time of healing. 5 Antibiotics may be continued for 72 hours following removal of all stents. Postincision hemorrhage may occur but is more commonly associated with the presence of a nephrostomy tube. Reports of hemorrhage following Acucise balloon incision have been several, documenting the caveats necessary for patients undergoing Acucise dilation. 6 Those caveats include the less controlled incision without direct vision and the difficulty in the diagnosis of postincision bleeding. Since bleeding, hematoma, and persistent hemorrhage occur in 1% to 2% of patients undergoing percutaneous nephrostomy and have been reported to occur in 2% to 4% of patients undergoing Acucise balloon dilation, this complication can be expected to occur in those patients undergoing endopyelotomy from all approaches. 6,7 In our experience, one patient undergoing antegrade endopyelotomy and one patient undergoing Acucise endopyelotomy have had hemorrhage significant enough to require angiography and embolization as the result of an arteriovenous fistula. With an open incision through the urinary tract, urinoma formation around the area of incision might be expected to be a common complication. With adequate nephrostomy drainage or ureteral stenting, however, this complication has not been identified. Extravasation during the immediate postoperative period is often demonstrated but adequate drainage results in resolution within 72 hours of incision. Occasionally, longer extravasation can be observed but ultimate resolution can be expected to occur with persistent external drainage through nephrostomy tube or double-J stenting. Because open pyeloplasty is often accompanied by renal pelvis excision, endopyelotomy may result in poorer appearances on urography than open pyeloplasty. The importance of this extra pelvic, however, is minimal and the results of the procedures can be measured by the subsequent drainage on diuretic urography or diuretic renography. If hydronephrosis persists and there is evidence of continued obstruction and inadequate urine flow across the UPJ, a repeat endopyelotomy may be warranted. Results The results of endopyelotomy by all approaches have been reported to be successful in 82% to 88% of patients. 7,8 and 9 Van Canh et al. have examined reasons for success and failure in 102 patients undergoing endopyelotomy followed for an average of 15 years. 9 Long-term success was 75% with 13% late recurrences. Success in patients with crossing vessels at the UPJ was 42% with 31% failures. Success with hydronephrosis and a large renal pelvis was 81% with 4% failures. With hydronephrosis and a crossing vessel, success was only 39% with 39% failures, whereas absence of hydronephrosis and crossing vessel was associated with 95% success and 0% failure. Thus, patients with crossing vessels at the proximal ureter or redundant renal pelvis appear to be less likely to have successful long-term results from endopyelotomy. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Chandhoke PS, Clayman RV, Stone AM, et al. Endopyelotomy and endoureterotomy with Acucise ureteral cutting balloon device: preliminary experience. J Endourol 1993;7:45–51. Gerber GS, Lyon ES. Endopyelotomy: patient selection, results, and complications. Urology 1994;43:2–10. Ketches BA, Seguro JW, LeRoy AJ, Patterson DE. Percutaneous antegrade endopyelotomy: review of 50 consecutive cases. J Urol 1995;153:701–703. Mattola JA, Fried R, Badlani GH, Smith AD. Failed endopyelotomy: implications for future surgery on the ureteral pelvic junction. J Urol 1993;150:821–823. Meretyk I, Meretyk S, Clayman RV. Endopyelotomy: comparison of ureteroscopic retrograde and antegrade percutaneous techniques. J Urol 1991;148:775–782. Nadler RB, Rao GS, Pearle MS, et al. Acucise endopyelotomy: assessment of long term durability. J Urol 1996;156:1094–1098. Perez LM, Friedman RM, Carson CC. Endoureteropyelotomy in adults: review of procedure and results. Urology 1992;39:71–76. Thomas R. Endopyelotomy for ureteropelvic junction obstruction and ureteral stricture disease: a comparison of antegrade and retrograde techniques. Curr Opin Urol 1994;4:174—179. Van Canh PJ, Wilmart JF, Opsomer RJ, Abi-Aad A, Wese FX, Lorge F. Long term results and late recurrences after endoureteropyelotomy: a critical analysis of prognostic factors. J Urol 1994; 151:934–937.

Chapter 117 Endoscopic Ablation of Upper Urinary Tract Tumors Glenn’s Urologic Surgery

Chapter 117 Endoscopic Ablation of Upper Urinary Tract Tumors
Mantu Gupta and Arthur D. Smith

M. Gupta: Department of Urology, Columbia-Presbyterian Medical Center, New York, New York 10032. A. D. Smith: Albert Einstein College of Medicine, Bronx, New York 10461, and Department of Urology, Long Island Jewish Medical Center, New Hyde Park, New York 11040.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Ureteroscopic Approach Percutaneous Approach Outcomes Complications Results Chapter References

The success of endoscopic approaches to renal calculi has fostered attempts to ablate upper tract tumors in a minimally invasive fashion. Representing 5% to 10% of all urothelial malignancies, upper tract transitional cell carcinoma (TCC) occurs in the collecting system and ureter with equal frequency and is associated with a 30% to 50% incidence of bladder tumors. 1 Lesions of the distal ureter are much more common than those in the middle or upper ureter (7:2:1). 2 One-third of tumors are multifocal with simultaneous bilateral tumors of the upper tract occurring in 1% and metachronous lesions following nephroureterectomy in 2% to 9%. 3 The incidence is highest in the sixth and seventh decades of life and affects three times as many males Nas females. The natural history of upper tract TCC is different from that of bladder carcinoma in that up to 60% of upper tract tumors are invasive, compared to 15% of bladder tumors. Grade and stage are the two most important predictors of survival. The gold standard therapy for upper tract tumors remains nephroureterectomy with a bladder cuff because of the low incidence of synchronous and metachronous contralateral lesions, the high recurrence rate if the distal ureter is not excised, the high incidence of multifocality, and the higher incidence of invasive lesions. In situations where a nephroureterectomy would relegate a patient to dialysis (solitary kidney, renal insufficiency, bilateral disease), we have advocated an initial minimally invasive approach with tumor resection for accurate grading and staging. This chapter will outline our approach to the diagnosis and minimally invasive management of these lesions as well as our results.

DIAGNOSIS
Although the diagnosis of upper tract TCC has been described in detail elsewhere, there are a few points that deserve special attention. Given advances in the design, optics, size, and deflectability of current flexible ureteroscopes, the etiology of a radiolucent filling defect in the upper urinary tract can be determined in almost all cases by direct endoscopic inspection and biopsy. We believe that whenever possible the tumor should be visualized and biopsied prior to definitive treatment. Sometimes flexible instrumentation or the angle required for visualization precludes direct biopsy. In these situations we have found that fluoroscopically positioning a wire basket adjacent to the tumor through the ureteroscope and twirling it to obtain a tissue specimen is superior to a brush biopsy. The entire basket is withdrawn in the open position and sent for cytologic evaluation. Another important point is that the contralateral collecting system should be evaluated thoroughly prior to treatment of the primary lesion. This includes contrast studies and selective ureteral cytologies. If either of these is not completely normal, then flexible ureteropyeloscopy should be performed.

INDICATIONS FOR SURGERY
The following are the principle indications for the endoscopic management of upper urinary tract tumors: 1. 2. 3. 4. 5. Renal insufficiency. Solitary kidney. Bilateral tumors. Comorbidities that contraindicate open surgery. Patient preference

Patients must be willing and capable of undergoing vigilant and frequent endoscopic surveillance. Continued definitive endoscopic management is usually reserved for those patients who do not have high-grade (grade 3) lesions or evidence of muscle invasive disease in the initially resected specimen. If either of these two are found, nephroureterectomy is usually recommended.

ALTERNATIVE THERAPY
Once the decision for endoscopic management has been made, two alternative approaches are available: ureteroscopic and percutaneous. The choice is based on lesion size, location, and multifocality. Small, accessible lesions (less than 1 cm) are preferably treated ureteroscopically because the integrity of the urinary tract is maintained. These most often include lesions of the ureter, renal pelvis, and upper calyces. Although nephrostomy tract recurrence is a concern, series thus far have indicated that the incidence is very small. Therefore, larger lesions or lesions in less accessible calyces that cannot be adequately resected or fulgurated through a ureteroscope can be more effectively managed percutaneously without significant concern.

SURGICAL TECHNIQUE
Ureteroscopic Approach Cystourethroscopy is performed to assure that new bladder lesions have not developed in the interval since the prior evaluation. A retrograde ureteropyelogram is performed to confirm location of the lesion. A 0.038-in. floppy-tipped guidewire is placed into the upper urinary tract. Balloon dilation of the orifice is performed if necessary to allow introduction of ureteroscopes. A second safety guidewire is then placed. A rigid ureteroscope is advanced to the level of the lesion if possible. A standard ureteral resectoscope loop is used to resect the tumor to its base, and then the base is fulgurated. A problem with this approach is that bleeding can obscure visualization. A variety of irrigation devices now are available that allow better visualization during ureteroscopy. More recently, we have preferred cold-cup biopsies of the lesion followed by laser fulguration. Laser fulguration causes substantially less bleeding. Special attention is taken to assure that adequate deep biopsies are performed to allow evaluation of muscle invasion. Tumors that are not accessible by a rigid ureteroscope are approached with a flexible ureteroscope. We prefer to place a peel-away sheath (9–11 Fr or 12–14 Fr) into the upper ureter or renal pelvis in order to allow easy removal and reinsertion of the flexible ureteroscope during the procedure. In addition, the peel-away sheath acts as a continuous flow system that prevents high pressures from developing and that allows better flow of irrigant, which is very helpful given the diminutive inflow lumens in the smaller flexible ureteroscopes. The flexible ureteroscopes that are currently available accommodate laser fibers with ease. The placement of instruments or fibers through the working channel of a flexible ureteroscope significantly limits the amount of active and passive deflection that is available. Sometimes the tumor can be seen in a middle or lower calyx but when the laser fiber is passed the calyx is no longer accessible. First accessing the calyx and then passing the fiber may work. Alternatively, a substantial amount of laser fiber is advanced beyond the scope and directed to the calyx. The fiber then acts as a wire

guide, allowing the scope to be advanced over the excess fiber and into the calyx. In the past, we have used the neodymium:yttrium-aluminum-garnet (Nd:YAG) laser set at 20 to 25 W with excellent results. The only problem can be that the tissue becomes charred and it is often difficult to assess the depth of fulguration and the amount of tissue remaining. With the advent of the holmium (Ho):YAG laser, tissue is evaporated cleanly with minimal deep penetration, less bleeding, and minimal charring. We feel that it is easier to assess the amount of tissue remaining and the depth of fulguration. After resection, a ureteral stent is left indwelling. The patient can usually be discharged the next morning if bleeding remains minimal. Percutaneous Approach Our percutaneous approach to upper tract tumors is represented in Fig. 117-1. The first step is cystoscopic placement of a 6-Fr open-ended ureteral catheter into the collecting system. The catheter is secured to a 16-Fr urethral Foley balloon catheter. The patient is then placed in the prone position and pressure points are carefully padded. Retrograde instillation of contrast opacifies the collecting system and allows choice of a calyx for resection. Choice of the appropriate calyx is one of the most important aspects of the procedure ( Fig. 117-2). In general, a direct approach is used for single tumors. For multifocal tumors, an upper pole calyx is preferable. If the tumor is very peripherally located in the calyx, a direct puncture sometimes does not allow adequate visualization. In this particular situation, an indirect approach, achieving access from the calyx located furthest away from the involved calyx, may be necessary.

FIG. 117-1. Treatment protocol for percutaneous management of transitional cell carcinoma.

FIG. 117-2. Approach to transitional cell carcinoma is similar to that for stones. Tumors in calyces (a, b, c) can be approached directly. Tumors of the renal pelvis (d) or upper ureter (e) are better approached through upper or middle calyces to allow easier access and inspection of the ureteropelvic junction.

Puncture is made with a 15-cm 18-gauge diamond-tipped trocar needle. A 80-cm floppy-tipped J wire is passed into the collecting system and sequential dilation is performed with Amplatz dilators. The ureteral catheter is grasped and through-and-through access is procured with a 0.038-in. Lunderquist-Ring torque wire extending from the flank skin to the urethral meatus and clamped at both sites. The tumor is inspected and areas suspicious for carcinoma in situ as well as random sites are biopsied with cold-cup forceps. The bulk of the tumor is removed using the cold-cup biopsy forceps, including deep bites. Alternatively, deep and superficial bites are taken and the bulk of the tumor is vaporized using a grooved rollerball electrode or simultaneously vaporized and resected with a rollerball loop electrode. Ho:YAG or Nd:YAG laser fulguration may also be used ( Fig. 117-3).

FIG. 117-3. Laser fulguration of tumors may cause less bleeding.

At the end of the procedure a 24-Fr reentry nephrostomy tube is placed. The atient returns to the operating room 4 to 7 days later for second-look nephroscopy. The resection site is rebiopsied and any residual tumor fulgurated with a rollerball electrode, cutting loop, or laser. A 12- or 14-Fr Cope loop nephrostomy tube is placed, preferably in the renal pelvis or the involved calyx. This can be subsequently used for six weekly bacillus Calmette-Guerin (BCG) instillations. At any step, if high-grade or muscle invasive disease is present nephroureterectomy is recommended. A third look with repeat biopsies is performed after BCG instillation. BCG is usually administered 1 to 2 weeks after the second procedure if the hematuria has resolved. The patient is admitted 1 day prior to administration and given culture-specific intravenous antibiotics. Intracavitary pressures are continuously monitored and kept at less than 25 cm H 20 (Fig. 117-4). Initially, saline is infused at 10 ml/hr and gradually increased to 50 ml/hr. If tolerated, BCG is then instilled [50 ml of 1 × 10 8 colony-forming units (CFU)] over 1 hour. The patient is asked to void 1 hour after completion of the instillation and the tube is clamped. The patient is monitored overnight for fever or any adverse side affects prior to discharge. This is repeated for 5 more weeks, with chest radiographs and liver function tests being monitored every 2 weeks. Two weeks after the last instillation, the collecting system is reinspected endoscopically. If tumor is present at that time, a nephroureterectomy is recommended.

FIG. 117-4. Typical arrangement for BCG instillation with special attention to the monitoring of intracavitary pressure.

Follow-up, regardless of initial approach, is performed every 3 months for the first year, every 6 months for the next 4 years, and then yearly. Physical examination, urine cytology, and either intravenous pyelography or retrograde ureteropyelography is performed at each visit. Computed tomography (CT) and ureteroscopy are performed yearly. Higher grade lesions or multifocal tumors are followed more aggressively with more frequent flexible ureteropyeloscopy.

OUTCOMES
Complications Complications of percutaneous resection of TCC in 37 kidneys (35 patients) are presented in Table 117-1. Complications that occurred are similar to those attributable to percutaneous renal surgery for stone disease. The pleural complications were both treated with thoracostomy tubes. Bleeding from arteriovenous fistulas required angiographic embolization in 1 patient and nephroureterectomy in 2 others. The average fall in hematocrit in 37 procedures was 6.8% (range 0 to 16%). Eighteen patients required transfusion of at least 1 unit of packed red blood cells. The perirenal abscess was treated with CT-guided percutaneous drainage. The urethral stricture was managed by internal urethrotomy and the ureteropelvic junction stricture by percutaneous antegrade endopyelotomy.

TABLE 117-1. Complications of percutaneous renal surgery for TCC

Two patients had persistent postoperative fevers with a negative infectious workup, of which one was unrelated to BCG and spontaneously resolved. This patient was eventually treated with BCG without sequelae. The other patient's febrile course did start during BCG instillations. He continued with low-grade fevers, occasional chills, and malaise but a negative workup for BCG sepsis. This patient died in a motor vehicle accident and autopsy was refused. Nephrostomy tract recurrence was not noted in any of the above cases. One other patient not included in this series who had incidental discovery of TCC with squamous elements during percutaneous nephrolithotomy did have recurrence of squamous cell carcinoma 35 months later in the renal fossa and spleen along the nephrostomy tract. Results We recently evaluated results in 37 kidneys (35 patients). One of these patients had squamous cell carcinoma and is therefore excluded from analysis. Grade 1 tumors were found in 11 kidneys, grade 2 in 12, and grade 3 in 13. Six patients underwent immediate nephroureterectomy: 3 for muscle invasive disease, 2 for tumors that could not be completely resected, and 1 for iatrogenic bleeding. Nine patients underwent delayed nephrectomy: 8 for disease recurrence after initial endoscopic management and 1 for significant hemorrhage. Results of treatment stratified by grade are presented in Table 117-2 and Table 117-3. The recurrence rate was significantly higher in patients with grade 3 disease (50%) compared to grades 1 and 2 (18% and 33%, respectively). Endoscopic treatment of disease recurrence was not attempted for patients with higher grade lesions.

TABLE 117-2. Recurrence rates stratified by tumor grade

TABLE 117-3. Treatment outcome stratified by grade

The only deaths from disease occurred in patients with grade 3 tumors. Two deaths occurred in the 3 patients with grade 3 disease who had immediate nephrectomy for incomplete resection, and 4 deaths occurred among the 10 patients with grade 3 disease who had endoscopic treatment alone. In summary, endoscopic ablation of low- and moderate-grade noninvasive transitional cell tumors appears to be a safe and efficacious alternative to nephroureterectomy in carefully selected patients. CHAPTER REFERENCES
1. 2. 3. 4. Charbit L, Gendreau MC, Mee S, Cukier J. Tumors of the upper urinary tract: 10 years of experience. J Urol 1991;146:1243. Jarrett TW, Sweetser PM, Weiss GH, Smith AD. Percutaneous management of transitional cell carcinoma of the renal collecting system: 9-year experience. J Urol 1995;154:1629. McCarron JP, Mills C, Vaughn ED Jr. Tumors of the renal pelvis and ureter: current concepts and management. Semin Urol 1983;1:75. McDonald MW, Zincke H. Urothelial tumors of the upper urinary tract. In: deKernion JR, Paulson DF, eds. Genitourinary cancer management. Philadelphia: Lea and Febiger, 1987;1–39.

Chapter 118 Internal Urethrotomy Glenn’s Urologic Surgery

Chapter 118 Internal Urethrotomy
Joseph M. Khoury

J. M. Khoury: Division of Urology, University of North Carolina, Chapel Hill, North Carolina 27599.

Urethral Anatomy Etiology of Urethral Strictures Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Otis Urethrotomy Direct Vision Internal Urethrotomy Laser Urethrotomy Outcomes Complications Results Chapter References

The technique of internal urethrotomy was recorded in the sixth century BC in Hindu writings that described various surgical instruments, such as urethral dilators and catheters, to manage urethral obstruction. In 1817, Jean Civiale created a urethral sound with a protruding blade that was passed through a urethral stricture dividing it from within. In 1848, Maisonneuve constructed a urethrotome onto which a filiform guide could be connected, and 25 years later in the United States, Otis introduced various urethral dilators as well as the urethrotome that still bears his name. With the introduction of endoscopy, direct vision internal urethrotomy, as described by Sachse, has become an essential operative procedure to manage specific urethral strictures. 3 Urethral Anatomy There are four components of the urethra: the glans-meatus complex, and the penile, bulbar, and posterior urethra, which is sphincter-active ( Fig. 118-1). The glans meatus complex is the most distal component and is contained within the glans penis. The penile, or pendulous, urethra extends from the corona of the glans penis to the level of the suspensory ligament and from there continues to the urogenital diaphragm as the bulbar urethra. These three components, which comprise the anterior urethra, are enveloped by the corpus spongiosum, which eventually broadens between the crura of the corpora cavernosa to form a thick spongy vascular structure called the bulb, which is predominantly ventral and posterior to the urethral lumen. The posterior urethra has a membranous and prostatic portion; the membranous component is sphincter-active and terminates at the apex of the prostate gland.

FIG. 118-1. The entire male urethra visualized by an incision from the anterior bladder neck inferiorly to the urethral meatus through the dorsal aspect of the penis.

Etiology of Urethral Strictures The majority of urethral strictures are caused by trauma or inflammation. Internal urethral injury is usually iatrogenic caused by instrumentation such as the passage of urethral sounds, catheters, and endoscopes, or following procedures such as transurethral resection of the prostate. Such iatrogenic injury can be avoided by visually introducing the instrument rather than passing it blindly. The location of strictures following transurethral procedures or prolonged catheter drainage usually occurs at constrictive portions of the urethra such as the meatus, penoscrotal junction, and membranous urethra. External urethral injury is usually caused by blunt trauma such as a straddle injury to the perineum, penetrating injuries, and pelvic fractures. Usually these strictures are short, involving only the mucosa or superficial spongiosum. Inflammatory lesions of the urethra are usually associated with various forms of urethritis, such as Neisseria gonorrhea and Chlamydia organisms. Single infections usually resolve without significant injury to the urethral epithelium; however, multiple and incompletely treated infections may cause severe local inflammation resulting in significant fibrosis. These strictures are considerably longer, involving not only the urethral epithelium but the deep layers of the underlying spongy tissue resulting in dense scar formation ( Fig. 118-2).

FIG. 118-2. Common sites of stricture.

DIAGNOSIS
Urethral stricture disease usually presents with lower urinary tract symptoms, such as obstructed voiding. The slowing of the urinary stream may go unnoticed for years, until other manifestations occur, such as urinary frequency and urgency, recurrent urinary infections, prostatitis, epididymitis (particularly in young men), periurethral phlegmon, urethral cutaneous fistula, and urinary retention. In older men, there is occasionally some difficulty in deciding whether the symptoms are due to a urethral stricture or prostate gland enlargement.

On physical examination, there are usually no external manifestations of urethral stricture disease, except in those extreme cases in which urethral cutaneous fistula is noted or the classic water-pot perineum is found. Gently passing a urethral catheter to determine if the urethral lumen is patent can help determine whether urethral obstruction as a result of stricture disease is present. Urinalysis should be obtained, and it is not uncommon to find microhematuria or pyuria on microscopic review. Urine culture and sensitivity will document the presence of urinary infection. Preoperative uroflowmetry should be obtained as a baseline study to monitor treatment outcomes postoperatively. As a noninvasive urodynamic tool, it may suggest bladder outlet obstruction, remembering that the pressure-flow study can only determine whether bladder outlet obstruction exists. Retrograde urethrography is the mainstay of the radiographic investigation of urethral stricture disease. The urethrogram should not only demonstrate the stricture but should show the urethra proximal and distal to it. The length, caliber, location, multiplicity, and proximity of the stricture to the sphincter should all be identified. If the urethral stricture is not totally occluded, a small catheter may be negotiated past the stricture and the bladder filled so that a voiding cystourethrogram can be performed if the proximal margins of the stricture are not visualized on the retrograde urethrogram. The sonographic urethrogram has been reported to be equally efficacious or superior as a diagnostic modality compared to the standard radiographic evaluation. 5 The advantages include identifying the extent and thickness of periurethral fibrosis and spongiofibrosis, accurately measuring lumen size, and imaging the pendulous urethra to the fossa navicularis in a noninvasive fashion. Finally, urethroscopy may be helpful to compliment the findings on urethrography to further evaluate the length of stricture and gray urethra, which suggests further involvement of the underlying spongy or periurethral tissues.

INDICATIONS FOR SURGERY
Direct vision internal urethrotomy should be performed for short strictures that involve the epithelium of the urethra and underlying superficial spongiosum. This procedure should not be used only for its simplicity and because it is minimally invasive when compared to alternative treatment, i.e., open urethroplasty. Indeed, internal urethrotomy may be advantageous to manage anastomotic strictures that occur following open urethroplasty or postoperative strictures that result after transurethral surgery such as prostatectomy. Although Otis urethrotomy is rarely performed on a routine basis prior to transurethral resection of the prostate, some urologists find that it may prevent postoperative stricture formation. 8

ALTERNATIVE THERAPY
Alternative treatments to internal urethrotomy include urethral dilation provided that superficial mucosal injury of the urethra results only in strictures that are attenuated and soft and easily stretched by urethral bougienage with a low likelihood of recurrence. The strictures should be dilated gradually over several weeks until a 24-Fr dilator can be passed. It is appropriate to perform urethral dilation of such strictures every 6 to 12 months. The natural course of the stricture will proclaim itself during the weeks following the initial dilation. Urethral dilation should not be performed more than twice a year or for strictures that are multiple, long, or obliterative. Open urethroplasty is the best option for those strictures that are inappropriate for urethral dilation or direct vision internal urethrotomy, as it carries a higher functional long-term success when properly performed.

SURGICAL TECHNIQUE
Internal urethrotomy is a procedure in which a urethral stricture and its underlying scar is incised with the hope that the created cleft will epithelialize and remain open. The procedure can be performed blindly as described by Otis; however, direct vision internal urethrotomy is far more common and has superseded blind urethrotomy. The preoperative evaluation should include a urinalysis and urine culture, ensuring that the urine is sterile prior to the procedure. Urethral instrumentation should be avoided for 4 to 6 weeks to allow prior inflammatory processes to resolve. If the patient has incomplete bladder emptying or is in urinary retention, suprapubic cystostomy is preferred to urethral instrumentation, which has the potential to cause further urethral trauma and injury. Depending on the location and length of the stricture, the procedure may be carried out using local anesthesia with 2% lidocaine jelly alone, or with intravenous sedation, depending on the patient's tolerance to pain. If it is to be performed in conjunction with other transurethral endoscopic procedures, such as transurethral prostatectomy, then alternative types of anesthesia such as general or regional should be considered. Though it is controversial as to whether to give prophylactic antibiotics prior to transurethral surgery, it is the author's preference to give them 30 minutes prior to the procedure, followed by oral coverage for as long as the catheter remains indwelling. Otis Urethrotomy After the patient undergoes an adequate preoperative evaluation and is found fit for outpatient surgery, he is prepped and draped in the dorsal lithotomy position. If outpatient urethroscopy has not been performed, it is done so prior to any blind instrumentation. The urethra is then calibrated using the bougies à boule to determine the lumen size and location of the stricture. In the closed position, the Otis urethrotome is placed through the urethral meatus while holding the penis on stretch perpendicular to the pubic bone until it stops just proximal to the external sphincter. The calibrated dial on the urethrotome is turned to the desired setting, usually between 30 and 36 Fr, expanding the instrument. The operating surgeon withdraws the knife blade in one quick motion, thus incising the stricture(s) ( Fig. 118-3). The urethrotome is closed and removed, after which the urethra is recalibrated to 30 Fr using the bougies à boule. The procedure may be repeated if necessary, or direct vision internal urethrotomy can be performed if fibrotic bands have not been adequately incised.

FIG. 118-3. (A) Otis urethrotome used for urethrotomy. (B) Technique of internal urethrotomy. Position of Otis urethrotome in anterior urethra before opening and withdrawal of blade. The penis is held erect so that the tip of the instrument cannot enter the membranous urethra. Inset shows open urethrotome.

Direct Vision Internal Urethrotomy In a similar fashion to blind urethrotomy, the patient is prepped and draped in the dorsal lithotomy position. If urethroscopy has not been performed, then it is appropriate to inspect the urethra with a small-caliber (15- or 17-Fr) cystoscope. Adult (20-Fr) and pediatric (14-Fr) size urethrotomes are available and both use a forward-viewing telescope, such as a 0 degree lens ( Fig. 118-4). The pediatric urethrotome is superior to the adult instrument for incision of strictures of the adult male pendulous urethra because the action of the pediatric urethrotome is very controllable in this mobile region and is less uncomfortable for use in an outpatient setting with local anesthesia. Prior to beginning the procedure, one of a variety of cold surgical blades is inserted in the urethrotome and inspected to ensure that it is appropriately seated within the sheath and advances appropriately in the same fashion as a loop for transurethral prostatectomy. Saline or glycine are appropriate irrigating solutions, as these prevent hemolysis should extravasation into the tissues of the perineum occur.

FIG. 118-4. Sachse endoscopic urethrotome (A). Assembled Sachse optical urethrotome (B). Components include (top to bottom) working element, 0 degree lens, sheath, bridge assembly, and various knife blades. (Courtesy of Karl Storz, Endoscopy-America, Culver City, CA.)

The urethrotome sheath, with its rounded obturator, is placed through the meatus; the obturator is removed and then replaced with the resectoscope. With the instrument in the distal urethra and the urethra compressed around the sheath so that the inflow of fluid distends the urethra, the urethrotome is advanced until the stricture is seen. In certain situations, the stricture may be safely incised; however, if the urethral lumen is significantly narrow, a filiform, guide wire, or ureteral catheter can be negotiated along the working channel of the urethrotome, or alongside the instrument itself and through the stricture. With this guide, safe incision through the entire stricture length can be accomplished. The stricture should be incised at the 12 o'clock position to avoid the main vasculature of the urethra ( Fig. 118-5). Rather than moving the blade in a forward and backward motion, as with transurethral prostatectomy, the blade is extended and, with a short upward rocking motion, the entire urethrotome is advanced through the stricture in the direction of the surgeon ( Fig. 118-6). It is important not to cut too deeply in the dorsal position, particularly in the pendulous urethra, to avoid injury to the overlying corporal bodies. It is important not to cut too deeply in this position because fibrosis or inflammation could cause a plaque to occur, resulting in ventral penile chordee. Similarly, the surgeon should incise only the stricture and not the normal urethra distal and proximal to it, particularly when incising bulbar strictures that approximate the distal sphincter mechanism. After the urethrotome is removed, cystoscopy can be performed, if necessary, to inspect the bladder for any abnormalities. An 18-Fr silicone catheter is placed at the end of the procedure and left indwelling for 3 to 7 days, depending on the extent of surgery and the depth of incision. Superficial mucosal strictures usually require a short duration of catheter drainage, whereas more dense strictures with full-thickness spongiofibrosis may require stenting for up to 6 weeks. Prolonged catheterization should be avoided as it can potentiate abrasion and sloughing of the urethral mucosa.

FIG. 118-5. Urethroscopic view of cut stricture at the 12 o'clock position.

FIG. 118-6. Sagittal view of direct vision urethrotomy.

Laser Urethrotomy Laser urethrotomy has been advocated by some urologists as a better alternative to more traditional types of internal urethrotomy. 4,9,10 The Nd:YAG as well as the KTP 532 laser has been used with mixed outcomes. These differ in wavelength, depth of penetration, and energy scatter. The advantages include minimal bleeding and comparable success rates when compared to standard forms of internal urethrotomy. Blitz et al. performed laser urethrotomy in 58 patients including two men with artificial urinary sphincters. 4 They concluded that even though their results were comparable to standard forms of internal urethrotomy, the additional cost of the laser equipment and support staff did not justify its routine use. They do advocate laser urethrotomy as a safer procedure for men with an artificial urinary sphincter and concomitant urethral stricture disease.

OUTCOMES
Complications Although problems are usually uncommon following internal urethrotomy, the procedure is not complication-free. Immediate problems may include hemorrhage or extravasation of irrigating fluid into the tissues of the perineum should urethral injury or perforation occur. These untoward events are best managed with prolonged Foley catheter drainage. Urinary incontinence can occur only if the urethrotomy incision involves the distal sphincter mechanism of the membranous urethra in a patient who has previously undergone transurethral prostatectomy in which the bladder neck is resected. Erectile dysfunction and ventral chordee have been reported only rarely but should be discussed during the preoperative counseling session as a potential sequel to the procedure. Results Success rates for direct vision internal urethrotomy vary from 32% to 91%. 1,2,6,7 The vast difference in these results is based on various postoperative criteria that define a favorable or unfavorable outcome. Pansadoro and Emiliozzi reported on 224 men who were followed for a minimum of 5 years after direct vision internal urethrotomy. Their overall success was 32% with a recurrent stricture rate of 68% following one urethrotomy. 7 In this series, selected patients with a single, short (less than 1 cm) bulbar stricture enjoyed a success rate of 77% following internal urethrotomy. Those patients with multiple and/or long strictures, particularly located in the

penile or penile-bulbar urethra, did not fare as well. Direct vision internal urethrotomy continues to play an important role in the management of urethral stricture disease. It is a safe procedure that can be performed on an outpatient basis with relatively minimal complications. This operation is best reserved for those patients with a single, short (less than 1 cm) bulbar urethral stricture involving only the superficial spongiosum. In this select group, the success rate approaches 80%. This procedure should not be performed because of its simplicity for strictures that are long and multiple with significant periurethral and spongiofibrosis. In this latter group, open urethroplasty can provide more durable and long-lasting functional results. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Aagaard J, Andersen J, Jaszczak P. Direct vision internal urethrotomy. Br J Urol 1987;59:328–330. Abdel-Hakim A, Bernstein J, Hassouna M, Elhilali MM. Visual internal urethrotomy in the management of urethral strictures. Urology 1983;22:43. Atwater LL. The history of urethral stricture. Br J Urol 1943;15:39–51. Blitz BF, Palmer JS, Gerber GS, Levine LA, Rukstalis DB. Neodynium:YAG contact tip laser visual internal urethrotomy for the treatment of urethral strictures. J Urol 1996;155:212, Abstract #177. Gupta S, Majmudar B, Tiwari A, Gupta R, Kumar A, Gujral R. Sonourethrography in the evaluation of anterior urethral strictures: correlation with radiographic urethrography. J Clin Ultrasound 1993;21:231. Johnston SR, Bagshaw HA, Flynn JT, Kellett MJ, Blandy JP. Visual internal urethrotomy. Br J Urol 1980;52:542–545. Pansadoro V, Emiliozzi P. Internal urethrotomy in the management of anterior urethral strictures: long term follow up. J Urol 1996;156:73. Schultz A, Bay-Neilsen H, Bilde T, Christiansen L, Mikkelson AM, Steven K. Prevention of urethral stricture formation after transurethral resection of the prostate: a controlled randomized study of Otis urethrotomy versus the uninsulated metal sheath. J Urol 1989;141:73–75. Shanberg AM, Bagdassarian R, Tansley LA, Sawyer D. KTP 532, laser and treatment of urethral strictures. Urology 1988;32:517–520. Smith JA Jr, Dixon JA. Neodymium YAG laser treatment of benign urethral strictures. J Urol 1984;131:1080–1081.

Chapter 119 Transurethral Cystolitholapaxy Glenn’s Urologic Surgery

Chapter 119 Transurethral Cystolitholapaxy
Marshall L. Stoller and Donald L. Gentle

M. L. Stoller and D. L. Gentle: Department of Urology, University of California, San Francisco Urology Clinic, San Francisco, California 94143.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Transurethral Cystolitholapaxy Outcomes Complications Results Chapter References

Bladder stones have occurred in humans since early times. Twenty-three centuries ago, Hippocrates cautioned that “to cut through the bladder is lethal.” Lithotripsy for bladder calculi via perineal lithotomy was first referenced in documents from the 9th century. 8 By the 18th century, management of bladder stones included open perineal and suprapubic cystolithotomy. Jean Civiale developed his technique using a mechanical lithotrite that popularized transurethral fragmentation of bladder calculi in the 19th century. 11 The details of lithotripsy were often guarded by specialists to avoid their use by nonlithotripsy physicians. Endoscopic electrohydraulic lithotripsy was developed in the 1950s in Russia with successful fragmentation of bladder calculi in 1951. 3 The incidence of primary bladder calculi in the United States, which now is rare, has decreased as a result of improved infection control and nutrition. It is more common in boys from India, Indonesia, China, and the Middle East. Cultural differences such as diet, medical care access, and comorbidities may influence the formation of bladder calculi. The incidence of bladder stones is unknown; approximately 95% of the bladder stones are formed in males. Bladder stones likely develop from infected residual urine or from inflammation within the bladder. Urinary stasis from any source (i.e., benign prostatic hyperplasia, cystoceles, neurogenic bladders, bladder diverticuli, urethral strictures) promotes recurrent infections and the formation of bladder stones. Bladder calculi, for example, develop in 36% of patients with spinal cord injuries and associated neurogenic bladders within 8 years of injury. 4 Bladder inflammation from schistosomiasis or external beam radiation also promotes the formation of bladder stones as do foreign bodies/sutures, encrusted ureteral stents, indwelling Foley catheters, or prostatic urethral stents. Bladder calculi also occur in patients with continent urinary diversions, ileal urinary conduits, and augmented bladders.

DIAGNOSIS
Chronic urinary tract obstruction, stasis of urine, and recurrent infections from urea-splitting organisms are typical manifestations of bladder stones. Patients complain of an abrupt cessation of their urinary stream, and some have associated pain radiating to the penis/urethra. Voiding is frequently positional, and hematuria may be present. Physical examination may reveal suprapubic fullness or a palpable bladder if urinary retention is present. Associated physical findings may include neurologic deficits in patients with neurogenic bladders or cystoceles in women. Excessive postvoid residual urine can document urinary stasis and microscopic urinalysis can identify infected urine or hematuria. Plain radiographic images can diagnose radiopaque bladder calculi ( Fig. 119-1). Bladder ultrasonography can identify radiolucent or radiopaque calculi that shift with position and demonstrates classic posterior acoustic shadowing. Retrograde urethrograms can be utilized to demonstrate urethral stricture disease.

FIG. 119-1. KUB with multiple radiopaque bladder stones. Symptoms included positional voiding, abrupt cessation of urinary stream with pain radiating to the glans penis, and recurrent urinary tract infections.

Historically, bladder calculi were diagnosed by transurethral passage of van Buren sounds resulting in clicking noises or tactile vibrations from contact with the bladder stone. This is rarely the initial diagnostic procedure today. Presently, cystoscopy most commonly identifies bladder stones by direct visualization.

INDICATIONS FOR SURGERY
Bladder stones are abnormal and require urologic evaluation. Indications for intervention include recurrent infections, urinary retention, suprapubic pain, or hematuria. In patients with urinary diversions or bladder augmentations, recurrent infections or difficulty with catheterization may prompt evaluation and/or treatment for stones. The etiology of the underlying urinary stasis/bladder infections should be identified and treated in addition to the management of the bladder calculi. Predisposing factors to the formation of bladder calculi have been managed in a staged fashion after appropriate intervention for the stones has occurred. Increased complication rates of up to 21% have been reported in patients undergoing transurethral resection of the prostate (TURP) and cystolitholapaxy for bladder stones under the same anesthesia. 9 Upper tract urinary stones that successfully pass into the bladder usually pass uneventfully through the urethra with voiding. Bladder stones most often do not originate in the upper urinary tracts. Bladder and/or urethral pathology must be suspected when bladder calculi are present.

ALTERNATIVE THERAPY
Transurethral cystolitholapaxy is the most common contemporary treatment of bladder calculi. Alternative interventions to cystolitholapaxy depend on patient preference, stone size, comorbidities, urethral caliber, and the etiology of bladder calculi. Open cystolithotomy most commonly results in a stone-free outcome but requires an incision, suprapubic catheterization, longer hospital stays, and prolonged convalescence. Transurethral fragmentation with the Bigelow lithotrite provides good mechanical crushing of the stones but is associated with a high complication rate ( Fig. 119-2). Poor optical visualization within the bladder may lead to gross hematuria and/or bladder perforation when using this lithotrite. Percutaneous approaches for bladder stones have reported 89% stone-free success rates for suprapubic cystolithotripsy. 7 Percutaneous lithotripsy may now be the treatment of choice in children with bladder stones to avoid potential injury to the small-caliber urethra. 1,5 Indications for percutaneous approaches for bladder access have been described previously. Relative contraindications for percutaneous cystolitholapaxy

include prior lower abdominal and pelvic surgery or scarred and fibrotic bladders.

1

FIG. 119-2. Lithotrite for transurethral mechanical crushing of the bladder stones. Shown are the lithotrite with its mechanical crushing jaws, a 90 degree lens, and light cord.

Extracorporeal shock wave lithotripsy (ESWL) monotherapy for bladder stones recently has been advocated. Twenty-nine patients treated by ESWL for bladder calculi with an MFL 5000 lithotriptor used intravenous sedation; however, 25% of these patients required multiple treatments. 6 Incomplete fragmentation requiring multiple procedures make ESWL therapy for bladder stones less effective than transurethral cystolitholapaxy. Multiple procedures may be required to remove bladder stones regardless of the type of intervention.

SURGICAL TECHNIQUE
Transurethral Cystolitholapaxy Transurethral cystolitholapaxy can be performed with flexible or rigid cystoscopes based on the surgeon's preference. Large working channels optimize visualization and make the rigid cystoscope our preferred instrument. The patient is placed on the operating room table and is given general or regional anesthesia. The patient is repositioned in the dorsal lithotomy position. Pressure points are adequately padded. The patient receives culture-directed preoperative intravenous antibiotics. The lower abdomen and genitalia are prepped with betadine and draped in a sterile fashion. Cystoscopy is then performed with a standard 21-Fr rigid cystoscope with video magnification and a large-caliber working channel. Endoscopic cameras with video imaging decrease potential fluid contact, electrical mishaps, and cervical spine pain to the operator. We rarely perform any endoscopic procedures without video imaging. Bladder stones are confirmed under direct visualization. Inspection of the bladder wall, ureteral orifices, trigone, dome, and urethra is undertaken to assess for potential abnormalities. If concomitant transurethral resection of the prostate is indicated, we prefer performing the cystolitholapaxy first for improved visualization prior to TURP. Hematuria is less during the litholapaxy and there is a reduced likelihood of intravascular fluid absorption that can occur during TURP. Higher complication rates have been reported when cystolitholapaxy is combined with TURP; recommendations for staged procedures have been advocated. 9 Transurethral cystolitholapaxy requires an external energy source for stone fragmentation. Electrohydraulic lithotripsy (EHL) is most commonly used though other sources of energy include lasers (Nd:YAG, pulse dye, holmium), ultrasound, or the pneumatic jackhammer (Swiss lithoclast). EHL produces rapid comminution of bladder stones under direct visualization and is our preferred energy source. Normal saline is commonly used as the irrigant for transurethral cystolitholapaxy. Prior to using the EHL probe through the cystoscope, connection to the power generator and testing for correct electrical firing should be performed. After verification of correct function, the large EHL probe (5- to 9-Fr) is then passed through the cystoscope into the bladder. The probe is advanced approximately 0.5 cm beyond the tip of the lens to avoid damage of the optical system during discharge. The EHL probe is then placed next to (1 mm away from) the calculus and fragmentation is initiated. Initial settings for bladder stone fragmentation using a 3- to 6-kV EHL generator begin at 30% to 40% power with a single firing mode. If fragmentation does not occur rapidly, one can increase the voltage settings and/or change to multiple shock firings. We typically increase our voltage to 80% to 100% power with repetitive shocks to completely fragment the stones as needed. EHL probes should be discharged under direct vision at all times to avoid bladder injury or instrument damage. Fragmentation of bladder calculi should be performed as efficiently as possible after starting the litholapaxy. The goals for transurethral cystolitholapaxy include rapid comminution of the bladder stones along with minimization of gross hematuria and potential bladder injuries. Initial maneuvers include pinning the stone against the bladder wall with the EHL probe prior to initiating lithotripsy. The ureteral orifices should be avoided when pinning the calculi against the bladder wall. Bladder stones frequently migrate within the bladder during lithotripsy and require chasing the stones. Cracking the outer shell of the bladder calculus is performed first. It is frequently the most difficult task. The EHL probe is directed 90 degrees against the surface of the stone to provide exterior fragmentation. Newer EHL probes with directed nozzles at the tip may become useful for stone fragmentation because they control scattering of the delivered energy. After initial fragmentation of the outer shell, the EHL probe is placed inside the cracked surface and internal shocks are delivered. Larger pieces of bladder stones are fragmented from the inside out until the pieces are small enough to be removed. Treatment of the same stone should be performed until an optimal size is obtained prior to comminution of other stones within the bladder. Persistence is required because many stones are difficult to corner and crack. The goal of cystolitholapaxy requires stone fragments to easily pass through the endoscopic sheath. Stone dust is not required. Ellik evacuator flasks are the most efficient modality for bladder stone removal after fragmentation. A newly developed stone retrieval net can be utilized to entrap the stones within the bladder. The net is passed per urethra alongside the cystoscope into the bladder and snares the bladder stones. While entrapped within the retrieval net, the bladder stones are then fragmented with EHL. When stone fragments are small enough, they fall through the net into the bladder and subsequently pass through the sheath during bladder drainage. Stone fragments should not be grasped and pulled with the cystoscope through the urethra. Stones can become entrapped in the urethra and are difficult to remove endoscopically. Open urethrotomy and stone retrieval may be required following attempts to remove the retained large bladder stone per urethra. All fragments should pass through the cystoscope sheath. Treatment of bladder stones within orthotopic bladders is similar to the above procedures. Lithotripsy for bladder stones in continent urinary diversions, augmented bladders, or ileal urinary conduits are also successful with the above approaches. For ileal conduits, endoscopy through the urinary stoma gives easy access to the loop diversion. However, complete visualization of the loop and stones may be difficult with large, redundant conduits and flexible endoscopes may be required. For continent diversions, the catheterizable stoma gives endoscopic access for stone fragmentation without jeopardizing the continence mechanism. However, the smallest endoscope with a working channel should be used to avoid disruption of the continent tract. Flexible cystoscopes may be less traumatic in this situation. If this is unsuccessful, percutaneous access and lithotripsy to the continent pouch is recommended. Frequent drainage of the bladder is mandatory throughout the litholapaxy procedures. This will remove bladder stone fragments and avoid overfilling with potential bladder perforation. This is important whether treating stones in native bladders, augmented bladders, ileal urinary conduits, or continent diversions. Urinary perforation occurs more frequently with urinary diversions as opposed to native bladders. Special circumstances for treatment of bladder stones include the inability to fragment calculi with EHL, large stones that completely fill the bladder, and stones associated with ureteroceles. Some bladder stones will not fragment despite EHL lithotripsy. Options for treatment at this point include mechanical lithotrites, percutaneous lithotripsy, or open surgery. Care should be taken during transurethral passage of the mechanical lithotrite to avoid injury to the urethra/bladder or to minimize gross hematuria. Prior to crushing the stone, torque should be applied to the lithotrite to ensure that the lithotrite jaws have not entrapped the bladder wall. A partially filled bladder decreases potential bladder injuries. Large calculi that fill the entire bladder or that are inaccessible within bladder diverticuli are challenging for endoscopic approaches; open surgical management at this point is advised. Stones associated with ureteroceles are managed after incising the ureterocele. Patience

is required regardless of the type of endoscopic intervention for bladder stones; if in doubt, open surgical removal should be performed.

OUTCOMES
Complications One hundred twelve patients treated over 10 years with transurethral intervention for bladder stones reported minor complications (27%) with no mortalities ( Table 119-1)10 Four of five bladder perforations occurred during use of the mechanical lithotrite, and the fifth was associated with grasping forceps. No bladder perforation was associated with EHL treatment alone. The majority of the other complications during cystolithotripsy occurred with concurrent TURP.

TABLE 119-1. Transurethral management of bladder stone

Complications of fever and urinary tract infection are diagnosed clinically. Physical findings are febrile episodes, malaise, or dysuria, and laboratory studies including CBC/differential white cell count and urine/blood cultures are used to confirm the infection. Treatment of infections includes antibiotics and optimal bladder drainage. Postoperative bleeding may occur after litholapaxy and is diagnosed by gross hematuria or large clots from the Foley catheter. The hemorrhage is usually self-limiting and treatment consists of bed rest, intravenous hydration, and urinary bladder drainage. Serial hematocrits should be followed. Occasionally, continuous three-way bladder irrigation is needed. If this is unsuccessful, transurethral fulguration of bleeding points may be required. Blood transfusions are rarely needed, though if the hemorrhage is severe, one must rule out a bladder or urethral injury. Bladder perforations usually are diagnosed intraoperatively after an unexplained lack of return of irrigant during the procedure. Confirmation by intraoperative cystogram is required. Extraperitoneal bladder perforations can be managed with Foley catheter bladder drainage. Prior to removal of the Foley, a cystogram confirming an intact bladder is required. Intraperitoneal bladder perforations are treated with open exploration and surgical closure of the bladder injury. Urinary bladders are drained postoperatively with a urethral catheter and suprapubic drainage tube. Hyponatremia can result from concurrent TURP and cystolitholapaxy. Lethargy or mental status changes in addition to low serum sodium confirms a diagnosis. It is minimized by shortening operative times during the TURP and the routine use of glycine irrigant during the prostate resection. Severe cases may be managed with diuretics or slow intravenous infusion of sodium chloride solutions (3% NaCl). Results Transurethral cystolitholapaxy successfully treats bladder stones. A nonrandomized retrospective analysis of 100 patients with bladder stones revealed that the types of interventions were based on the size of the stone alone. In this review, 43% had cystoscopic extraction alone with stones less than 1.25 cm, 19% were managed with mechanical litholapaxy for stones from 1.25 to 2.5 cm, and 38% were successfully treated with EHL for stones greater than 2.5 cm. 2 Complications were minor in all three groups; hematuria was the most common complication, occurring in the mechanical litholapaxy group. 2 CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Badlani GH, Douenias R, Smith AD. Percutaneous bladder procedures. Urol Clin North Am 1990;17(1):67. Bapat SS. Endoscopic removal of bladder stones in adults. Br J Urol 1977;49:527. Dawson C, Whitfield HN. The long-term results of treatment of urinary stones. Br J Urol 1994;74:397. DeVivo MJ, Fine PR, Cutter GH, et al. Risk of bladder calculi in patients with spinal cord injuries. Arch Intern Med 1985;145:428. Gopalakrishnan G, Bhaskar P, Jehangir E. Suprapubic lithotripsy. Br J Urol 1988;62:389 Hotiana MZ, Khan LA, Talati J. Extracorporeal shock wave lithotripsy for bladder stones. Br J Urol 1993;71:692. Ikari O, Netto NR Jr, D'Ancona CAL, Palma PCR. Percutaneous treatment of bladder stones. J Urol 1993;149:1499. Marketos SG, Lascaratos J, Malakates S. The first record of lithotripsy, in the early Byzantine era. Br J Urol 1994;74:405. Nseyo UO, Rivard DJ, Garlick WB, Bennett AH. Management of bladder stones: should transurethral resection be performed in combination with cystolitholapaxy? Urology 1987;29(3):265. Quint HJ, Moran ME, Drach GW. Bladder calculi: comparison of modes of therapy. Unpublished data. Urquhart-Hay D. Sir Henry Thompson BT, the first English urologist. Br J Urol 1994;73:345.

Chapter 120 Extracorporeal Shock Wave Lithotripsy Glenn’s Urologic Surgery

Chapter 120 Extracorporeal Shock Wave Lithotripsy
James E. Lingeman and John W. Dushinski

J. E. Lingeman: Methodist Hospital of Indiana Institute for Kidney Stone Diseases, and Department of Urology, Indiana University School of Medicine, Indianapolis, Indiana 46202. J. W. Dushinski: Department of Surgery, University of Calgary, and Rockyview Profesional Center, Calgary, Alberta, T2V 4R6 Canada.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Preoperative Evaluation and Preparation Anesthesia Patient Positioning Stenting and Manipulating the Stone Shock Wave Treatment Special Circumstances Postoperative Care and Complications Outcomes Complications Results Future Developments Chapter References

The introduction of extracorporeal shock wave lithotripsy (ESWL) in the early 1980s revolutionized the treatment of urinary tract calculi. In conjunction with percutaneous nephrolithotomy and ureteroscopy, ESWL has relegated open stone surgery to historical status at most institutions. The noninvasive approach of ESWL has become the initial procedure of choice for most stones in the kidney and upper ureter. At many institutions, ESWL is also the procedure of choice for distal ureteral calculi; however, others prefer the higher stone-free rate and lower secondary procedure rate of ureteroscopy. The first lithotriptor for use on humans was the HM3, developed by the Dornier Aerospace Company in Germany. The HM3 remains the most powerful lithotriptor in use, despite the many changes in lithotriptor technology incorporated in subsequent generations of machines. Newer machines were developed in an effort to reduce anesthetic requirements and improve imaging systems for stone localization, yet power was sacrificed in doing so. The result has been machines that require more shocks and more procedures to obtain the same effect. Most of the experience at our institution and around the world has been with the HM3. In this chapter we will present the principles of ESWL as established on the HM3 and also mention other machines when technique is altered. We will provide a general approach to the ESWL patient, discuss problem areas and contraindications for ESWL, and briefly discuss the potential for future developments in ESWL technology.

DIAGNOSIS
Symptomatic patients will complain of flank pain that may radiate and is colicky in nature. Other common presenting symptoms include nausea and vomiting as well as anorexia. It is important to question the patient about fevers and/or rigors associated with the flank pain or any other symptom that may indicate a urinary infection. Physical examination may show flank tenderness or signs of intestinal ileus. Important laboratory examinations include a urinalysis and a complete blood count (CBC). The diagnosis of renal stone is usually made radiographically with a KUB and intravenous urography. In cases in which the stone is not well visualized, the ureter is not seen below the stone, or in patients with contrast allergy, it may be necessary to perform a retrograde pyelogram. Ultrasonography can be used to identify uric acid stones, which are generally radiolucent and, to a lesser degree, cystine stones. There has been recent interest in the radiologic literature in using spiral CT for the diagnosis of renal stone.

INDICATIONS FOR SURGERY
The precise indications for ESWL are evolving and vary from institution to institution and machine to machine. However, some general guidelines are useful both to evaluate the patient as a candidate and to discuss treatment options with the patient ( Table 120-1). The indications for surgical intervention for stone disease traditionally were obstruction, symptoms, infection, bleeding, loss of renal function, and failure of appropriate medical management. As with any less invasive procedure, the indications for intervention with ESWL have broadened to include stones that have a high likelihood of becoming symptomatic and those causing questionable symptomatology.

TABLE 120-1. Transurethral management of bladder stone

Having stated the general indications for ESWL, it is important to state that anyone can be treated with ESWL; the main issue is who should be treated with ESWL. Absolute contraindications now include only uncorrected bleeding disorder and pregnancy ( Table 120-2). Bleeding diathesis strongly predisposes to perinephric hematoma formation after ESWL.5 All anticoagulants, aspirin compounds, and antiinflammatories should be discontinued, if possible, 2 weeks prior to the procedure. Coumadin is stopped long enough to normalize the coagulation profile.

TABLE 120-2. Contraindications to extracorporeal shock wave lithotripsy

Animal studies have revealed devastating effects of shock waves focused directly on chick embryos. 11 Concern regarding the teratogenic effects of ESWL should be foremost in the mind of all urologists treating women of childbearing age, but as yet no reports of deleterious effects have been reported in human embryos. There has also been concern in the urologic community regarding the effect of ESWL on the ovary; however, the small amount of material published to date suggests that fertility is not affected by ESWL. 14 Relative contraindications include urinary tract infection, obstruction distal to the stone, cardiac pacemaker, severe orthopedic deformities, severe renal failure, cystine stones, and lower ureteral stones in women of childbearing age. Certainly, any patient with an acute febrile urinary tract infection should not be treated with ESWL. Those patients with chronic urinary tract infections associated with their stones should have a minimum of 24 to 48 hours of culture-specific antibiotics prior to therapy. Many minor anatomic obstructions in the ureter can be relieved with ureteral stenting. This allows passage of fragments around the stent and also temporarily dilates the ureter. Anatomic abnormalities that impair urinary drainage (e.g., ureteropelvic junction obstruction, horseshoe kidney, pelvic kidney, calyceal diverticulum) may not allow stone fragments to pass after treatment. ESWL may be attempted in these patients in the hope that more invasive surgery can be avoided; however, the patient must be informed of the decreased likelihood of a stone-free result. The presence of a cardiac pacemaker was an exclusion criterion in the initial U.S. FDA trials on ESWL. The main concerns were possible electrical interference of or damage to the pacemaker function by the shock waves. Further experience has shown that safe and effective ESWL can be performed in pacemaker patients, providing certain precautions are taken. The patient should have a preoperative cardiology consultation, and a cardiologist should be close by during the treatment should emergency temporary pacing be required. Rate- responsive pacemakers should be switched to the VVI mode (non-rate-responsive), and dual-chamber pacemakers should be switched to the single chamber mode. 1 Tubless and piezoelectric lithotriptors are less likely to produce arrhythmias and problems with cardiac pacemakers. Severe orthopedic deformities are a relative contraindication only when the deformities preclude positioning of the stone at the F2 focal point. Treatment of cystine stones with ESWL is problematic. Cystine stones are relatively radiolucent (therefore difficult to localize and to assess fragmentation), frequently large, and/or multiple, and are unusually resistant to fragmentation by ESWL. In addition, these stones may be dissolvable with oral therapy when symptoms permit. We will occasionally treat small solitary cystine stones with ESWL but retrograde flexible ureteronephroscopy and percutaneous nephrolithotomy are used more commonly.4

ALTERNATIVE THERAPY
Alternatives to ESWL include endoscopic manipulation (via percutaneous or transurethral/ureteral access), laparoscopic ureterolithotomy, open surgery, or medical management (stone dissolution). In patients with asymptomatic small stones in the kidney, observation is also an option.

SURGICAL TECHNIQUE
Preoperative Evaluation and Preparation All patients should routinely have a history and physical examination to determine their suitability for ESWL treatment. Technical factors such as patient size and various dysmorphic features may make the patient unsuitable for treatment. Special attention should be placed on obtaining a history of the use of any medication which could affect the blood-clotting mechanisms. Anticoagulants, nonsteroidal antiinflammatory drugs (NSAIDs), aspirin, and antiplatelet medications all predispose to renal hemorrhage following ESWL. We provide patients with a list of compounds to avoid. Hypertension also predisposes to hemorrhage, especially when not controlled perioperatively. Coagulation profiles are indicated in selected cases. Routine lab work includes a CBC, electrolytes, blood urea nitrogen (BUN), creatinine, and urinalysis. Urine cultures are obtained in patients with a history of infection, recent transurethral procedure, and in patients in whom the possibility of struvite stones is high. An electrocardiogram (ECG) is ordered in selected cases, based primarily on the age of the patient and the type of anesthesia to be used. Patients are advised to restrict diet to clear liquid and to take a laxative on the evening prior to the procedure. This practice facilitates localization of the stone by minimizing bowel gas. A KUB film is obtained on the day of the procedure. In our institution, patients with contralateral stones greater than 5 mm are considered for bilateral ESWL in the same session (approximately 15% of cases). Thomas et al. have stated that bilateral simultaneous ESWL should be performed with caution due to their observation of long-term decreases in ERPF. 13 These findings have yet to be corroborated; however, their assertion that casual overtreatment should be avoided is a sound principle. Anesthesia The form of anesthesia is generally dictated by the type of machine used, but the patient's medical status and age as well as the experience of the surgeon and the anesthesiologist should also be considered. The original HM3 required general or regional (spinal or epidural) anesthesia in virtually all patients. Subsequent advances in anesthetic techniques have allowed treatment on the unmodified HM3 under intravenous sedation in most cases. Treatment under intravenous sedation is routine on the modified HM3, ensuing generations of spark gap lithotriptors and electromagnetic lithotriptors. Patients are treated comfortably using various combinations of short-acting agents such as narcotics (alfentanil and fentanyl), sedative-hypnotics (midazolam and propafol), and topical agents. Most patients are treated without anesthesia when using the piezoelectric machines. Patient Positioning Positioning the patient for ESWL can often be challenging and in such cases an experienced designated ESWL technician is an invaluable asset. Patients with an unusual body habitus should undergo simulation on the lithotripter to ensure that the stone can be visualized and positioned at F2. Stones in the kidney and ureteral stones above the iliac crest are treated with the patient in the supine position. Ureteral stones overlying the sacrum must be treated with the patient in the prone position. Distal ureteral stones are rarely treated at our institution. The higher stone-free rate and lower secondary procedure and retreatment rates of ureteroscopy are generally preferred by the patient. ESWL is performed in selected distal stone cases, generally via the gluteal region with the patient in the supine position. Thin patients with distal stones can also be treated in the prone position. The Stryker frame modification makes prone positioning simple and comfortable for the patient (Fig. 120-1). The prone position is also useful in treating the pelvic kidney, horseshoe kidney (particularly inferior and medial calyces), transplant kidney, and patients with severe kyphoscoliosis. 8

FIG. 120-1. The Stryker frame modification for prone positioning on the Dornier HM3.

The Siemens Lithostar has separate shock heads for each side of the patient. This can be exploited in the case of a proximal ureteral stone located very close to the spine and in patients with a distal ureteral stone that cannot be localized to the F2. In these cases, the contralateral shock head is used to target the stone while a pad is placed under the contralateral flank. 2 This technique is depicted schematically in Figure 120-2.

FIG. 120-2. Modified technique for the treatment of difficult upper ureteral stones with the Siemens Lithostar. (A) Conventional positioning. (B) Modified technique for a right-sided stone using the contralateral shock head. (From Carey PO, Jenkins J. New Lithostar treatment technique for difficult upper ureteral stones. J Endourol 1995;9:233.)

Stenting and Manipulating the Stone The routine stenting of ureters prior to ESWL has now been shown to be contraindicated. 3 In certain cases, however, stenting may be desirable and even necessary. All patients with stones larger than 2 cm in diameter, solitary kidneys, and suspected struvite stones are stented to ensure urinary drainage intraoperatively and postoperatively. Placement of a ureteral stent significantly decreases postoperative colic, ureteral obstruction rate, and Steinstrasse in patients with large stone burdens. It does not, however, guarantee drainage of the collecting system, as the stent may become blocked by stone debris. Stents are routinely left indwelling for 7 to 10 days, at which time a KUB is performed to assess fragmentation. Patients with stones smaller than 1.5 cm are rarely stented. Those with stones in the 1.5- to 2-cm range are individualized. Factors favoring stenting stones in this size range include treatment with a lower powered lithotriptor, recent urinary tract infection, relative radiolucency, and a suspicion that the stone is relatively resistant to fragmentation by ESWL (i.e., cystine, calcium oxalate monohydrate, or brushite suspected). There has been debate in the literature with regard to treatment of upper ureteral stones. Some centers prefer to treat stones in situ (in the ureter, without manipulation or stenting), whereas others prefer the push-bang technique (pushing the stone back up into the renal pelvis prior to ESWL). Manipulating upper ureteral stones into the renal pelvis prior to ESWL will increase fragmentation rates compared to in situ ESWL. Table 120-3 compares the various management options for upper ureteral stones. An impacted ureteral stone is thought to lack an expansion space surrounding its outer layer. Stones that are free in the renal pelvis and ureteral stones that are not impacted are surrounded by fluid, which allows movement of fragments away from the central stone. Small fragments are believed to shield large fragments from further fragmentation during ESWL treatment. In addition, the lack of a fluid layer may limit the transmission of shock wave energy to the stone by inhibiting the process of cavitation. We routinely treat pediatric patients in situ to avoid ureteral and urethral manipulation.

TABLE 120-3. Management options for upper ureteral stones

A ureteral catheter is placed to aid stone localization in all patients with radiolucent stones or low-density stones. The catheter provides orientation and will often aid in localization of low-density stones prior to the injection of contrast. In this circumstance, contrast is not injected initially in order to avoid losing the stone in the contrast. Shock Wave Treatment Treatment may begin once the patient has been adequately anesthetized and the stone has been positioned at F2. Large solitary stones require a treatment strategy. The portion of the stone closest to the UPJ should be treated first; then the treatment proceeds to the periphery. We reserve 10% of the total shocks for areas that are judged to be least fragmented. The degree of fragmentation is often difficult to judge, particularly for calyceal and ureteral stones. Assessment is easiest when treating renal pelvic stones since the fragments tend to disperse. Quick pics are better than fluoroscopy for judging fragmentation; however, they expose the patient to considerably more radiation than fluoroscopy. Fragmentation is especially difficult to assess using lithotriptors that employ only ultrasound for localization. Other disadvantages of ultrasound localization include an inability to localize most ureteral stones, difficulty in locating some small renal pelvic stones, difficulty in locating stones in obese patients, and a generally prolonged localization time for all stones. Although ultrasound has its advantages, including visualization of radiolucent stones, lack of radiation exposure, continuous imaging,

and the option of treating biliary tract calculi, very few ultrasound-imaging-only lithotriptors have been popular for urologic applications. Stone fragmentation and localization are reassessed every 200 to 300 shocks. We give 200 to 300 additional shocks to the area of the stone when the stone appears to be adequately fragmented because the degree of fragmentation can be difficult to assess. While newer lithotriptors generally have more advanced fluoroscopic systems, with higher image quality, determination of the endpoint of ESWL remains problematic and the same rule should be applied. The maximum number of shock waves that can be safely administered in a single ESWL session is not known for any lithotripsy system. Because some trauma to renal parenchyma occurs with every lithotriptor during ESWL, casual overtreatment should be avoided. With the unmodified HM3 lithotriptor, most centers have limited treatment to 2000 to 3000 shocks per session. Fragmentation later in an ESWL treatment session is not as efficient as at the start of treatment. There are two main reasons for this. First, fragments created early in the session form a buffer around the central unfragmented portion of the stone, absorbing shock wave energy. Intravenous diuretics or ureteral catheter irrigation are occasionally used to augment fragment dispersal. Second, the pressure generated at F2 decreases with electrode wear when using machines with a spark-gap generator. The drop in pressure at F2 can be compensated for by increasing the voltage by 1 to 2 kV each time the electrode is heard to misfire. This second reason does not apply to lithotriptors employing electromagnetic or piezoelectric generators. Special Circumstances Spinal Cord Injury Patients Spinal cord injury (SCI) patients are at increased risk for urolithiasis due to urinary infection, neurogenic bladder, and immobility. The immobility of these patients also decreases the rate of fragment passage following ESWL. Several important points should be kept in mind when treating SCI patients. Symptoms of autonomic dysreflexia are seen commonly in SCI patients undergoing ESWL treatment. Hypertension may be treated with sublingual nifedipine, and bradycardia with intravenous atropine. Occasionally, during severe or prolonged episodes, the ESWL treatment may need to be temporarily interrupted. Involuntary muscle spasms are also common and can make localization of the stone at F2 difficult. Spasms can be treated with intravenous benzodiazepines or pretreated with oral benzodiazepines in anticipation of spasms. Mechanical bowel preps may be given prior to the procedure to facilitate stone localization and possibly decrease the incidence of autonomic dysreflexia. Lastly, as always with SCI patients, pressure points must be well protected to prevent pressure sores. Morbidly Obese Patients Morbidly obese patients may have to be treated with ureteroscopy or percutaneous nephrolithotomy due to the weight restriction of the lithotriptor. With certain machines, the weight restriction can be exceeded if the table is lowered to its limit and is not raised while the patient is on the table (check with manufacturer before doing this). Another problem that must be overcome is positioning the calculus at the F2 focal point. If the distance from the skin to the stone is too great, it may not be possible to place the stone at the F2. A compression strap placed around the patient's abdomen may be used to push the stone toward the F2. Another technique that can be used if the stone cannot be positioned at the F2 is treatment of the stone in the blast path. 15 Treating in the blast path involves placing the stone in the axis of the pressure wave but not in the center of the focal zone. Figure 120-3 outlines how to determine the direction of the blast path. Fragmentation of calculi using this method can occur; however, the degree of fragmentation drops off as the distance from F2 increases. The blast path technique has also been used for treatment of stones in horseshoe kidneys and transplant kidneys. The blast path technique is less effective using machines with a very small F2 (e.g., piezoelectric machines).

FIG. 120-3. Determination of the blast path axis. (A) Phantom provided for x-ray alignment provides the axis of the ellipsoid. (B) blast path (dotted line) is determined by noting direction of phantom on fluoroscopic monitors. (From Lingeman JE, Smith LH, Woods JR, et al. Urinary calculi: ESWL, endourology, and medical therapy. Philadelphia: Lea and Febiger, 1989.)

Pediatric Patients Age less than 18 years was an exclusion criterion for the initial U.S. FDA trial on ESWL. Experience with adults, however, was extremely encouraging and gradually the pediatric population came to be treated with ESWL. Thousands of pediatric patients have been treated in the intervening years; yet, surprisingly, ESWL is still not approved by the FDA for the treatment of pediatric patients. The parents of children undergoing ESWL should be informed that, although for many pediatric patients ESWL treatment is the clinical standard, in the eyes of the FDA it remains experimental. Older machines often require modification to the gantry or frame with special spacers to prevent the child from dropping through. Newer, under-table designs do not require this modification, but small, thin children may require augmentation of the water cushion to keep the stone in the F2 focal point. This can be easily accomplished by placing an IV bag between the water cushion and the skin and by eliminating the acoustic interfaces with the application of an acoustic gel. Children generally tend to require more anesthesia and analgesia than adults. The exception is the piezoelectric machines, which truly can perform anesthesia-free lithotripsy on most pediatric patients. There are two points to remember regarding complications in pediatric ESWL patients. Due to the smaller body size in pediatric patients, the lung is in the path of the shock wave more frequently. The large number of acoustic interfaces in the lung readily attenuate shock waves, which may be manifested clinically by pulmonary contusion and hemoptysis. The lungs of all patients should be shielded with styrofoam if the path of the shock wave will cross the pleural cavity. The second point relates to obstructing ureteral fragments. Because of the delicate nature of the pediatric ureter, ESWL on children should be done only in centers with the instruments and expertise necessary for dealing with the sequela of obstructing fragments. Overall, pediatric patients do remarkably well following ESWL. Stones break up more efficiently, with fewer shock waves and lower voltage. Patients also have less morbidity secondary to the passage of stone fragments—yet another reason why stenting or manipulating of pediatric patients should be avoided. Calyceal Diverticuli Treatment of calyceal diverticuli with ESWL is a controversial form of management. Most centers would consider percutaneous management to be the primary mode of therapy. Selected cases with a radiographically patent diverticular neck and small stone burden (less than 1.5 cm maximal diameter) have been treated with a moderate degree of success.12 Stone-free status in these cases appears to be independent of symptomatic relief. Calyceal diverticuli associated with stone and infection are unlikely to be rendered infection free after ESWL and should be treated percutaneously. This point illustrates the suboptimal nature of ESWL treatment of stones in calyceal diverticuli. The urinary stasis associated with the diverticulum is the likely primary abnormality, not the stones within them. Biliary and Pancreatic Lithotripsy

ESWL has not revolutionized the treatment of biliary tract calculi as it has the treatment of urinary tract calculi, due largely to the advent of laparoscopic cholecystectomy. In selected cases, however, ESWL may be a viable option. The prime example is the patient whose medical condition precludes general anesthesia. Patients with severe cardiorespiratory disease can be treated anesthesia-free on the piezoelectric machines. When treating these patients, it should be kept in mind that obstructing common bile duct fragments may necessitate interventions requiring anesthesia. The patient suspected of having severe intraperitoneal adhesions would also be a possible candidate. An ultrasound localization system is usually required for biliary stones since radiopacity is opposite that of urinary calculi (90% radiolucent versus 10% radiopaque). Chemical dissolution with oral biliary acids (ursodeoxycholic acid, chenodeoxycholic acid) aids in the clearing of fragments. One prerequisite for the clearing of gallbladder stones is a functioning gallbladder. ESWL is also a viable option for common bile duct stones and obstructing pancreatic duct stones that cannot be managed endoscopically. Fluoroscopy is the imaging modality of choice for pancreatic stones because most are radiopaque and are difficult to localize with ultrasound. Patients are positioned supine, much as if treating an upper ureteral stone. A right-sided position is used for the head of the pancreas and the common bile duct, whereas a left-sided position usually works best for stones in the body or tail of the pancreas. Patients have also been treated in the prone position. Postoperative Care and Complications Virtually all patients are discharged on the day of treatment. All patients receive a prescription for analgesics, a urine strainer for stone collection, and are advised to push oral fluids. Antibiotics are given only in selected cases. Further treatment of renal stones in asymptomatic patients is not considered until the follow-up appointment at 3 weeks, at which time a KUB and renal ultrasound are routinely performed. Patients are offered retreatment for fragments that are 5 mm or larger. Patients with no residual fragments or fragments smaller than 5 mm receive a KUB with or without renal tomograms at 3 months. Patients may continue passing fragments for several weeks and even months following ESWL. Follow-up that is too early may lead to unnecessary retreatment.

OUTCOMES
Complications Most of the postoperative complications associated with ESWL treatment are related to the passage of stone fragments. Roughly 5% to 10% of patients will develop renal colic and most of the remainder will have some minor degree of flank pain. The likelihood of developing pain is directly proportional to the size of the stone treated. Patients who experience pain lasting longer than a few days or pain that is out of proportion to the treatment should be evaluated with renal ultrasound (to rule out perinephric hematoma) and a KUB. Clinically apparent perirenal hemorrhage occurs in less than 1% of patients and should be detected by the ultrasound. The vast majority of perinephric hematomas can be managed conservatively. Occasionally, a patient will require transfusion or, very rarely, angiography with embolization. The KUB film is obtained to assess for ureteral fragments and Steinstrasse. Indications for intervention are the same as those for the routine patient presenting with a ureteral stone. Options for intervention in patients with no sign of infection and less than 5 cm of Steinstrasse include ureteroscopy, stone basketing, and ESWL to the lead fragment. Observation is also an option in patients with tolerable symptoms. Symptomatic patients with greater than 5 cm Steinstrasse or signs of infection should have a percutaneous nephrostomy tube inserted. This relieves renal colic, allows drainage of infected urine, avoids the complications of attempted retrograde manipulation, and in most cases allows the stones to pass. ESWL initially was thought to be without adverse consequences. Subsequent studies on bioeffects have suggested that ESWL is a form of renal trauma. Histologic changes occurring in the kidney after ESWL include hemorrhage, endothelial cell damage, glomerular atrophy and sclerosis, and interstitial fibrosis. The long-term functional significance of these changes is unknown. 6 Concern regarding ESWL induced hypertension remains, and although the initial report linking ESWL to hypertension may have overestimated the connection, it seems plausible that ESWL may contribute to blood pressure changes in the same way as other forms of renal trauma, via renin-mediated mechanisms. 10 It is likely that multiple mechanisms, both renin-mediated and renin-independent, are involved. Other complications of ESWL are quite rare ( Table 120-4) when patients are evaluated and treated in an appropriate manner. It is important to recognize the factors (i.e., stone burden, urinary tract infection, hypertension, etc.) predisposing to complications and to modify technique or approach when required.

TABLE 120-4. Complications of ESWL

Results Stone size, location, and composition greatly affect the stone-free rate, morbidity rate, and secondary procedure rate following ESWL. Overall stone-free rates for various stone sizes and locations are seen in Table 120-5. These data were obtained using an unmodified Dornier HM3, arguably the most powerful lithotriptor in use. Results with other machines, particularly those with piezoelectric generators, can be expected to give lower stone-free rates and higher retreatment and secondary procedure rates. 9

TABLE 120-5. Stone-free rate after ESWL

Stone composition has a marked effect on the degree of stone fragmentation. Uric acid, struvite, apatite, and calcium oxalate dihydrate stones usually fragment quite readily. Stones composed of calcium oxalate monohydrate and calcium phosphate dihydrate (brushite) are relatively resistant and those composed of cystine are especially resistant. Treatment of most ureteral stones will eventually result in adequate fragmentation and passage of fragments in 80% to 90% of all patients. As with renal stones, stone size and composition will affect these rates significantly. Future Developments The urologic community will likely never undergo another period of upheaval like that following the introduction of ESWL. Future improvements in ESWL technology are likely to be incremental in nature. These incremental improvements will come from a better understanding of the interactions between shock waves, tissues, and stones. Only by studying these interactions will it be possible to design future machines that minimize the tissue effects and maximize the stone fragmentation effects of shock waves. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Albers DD, Lybrard FE, Axter JC, et al. Shock wave lithotripsy and pacemakers: experience with 20 cases. J Endourol 1995;9:301. Carey PO, Jenkins J. New Lithostar treatment technique for difficult upper ureteral stones. J Endourol 1995;9:233. Cass AS. Nonstent or noncatheter extracorporeal shock wave lithotripsy for ureteral stones. Urology 1994;43:178. Hockley NM, Lingeman JE, Hutchinson CL. Relative efficacy of extracorporeal shock wave lithotripsy and percutaneous nephrostolithotomy in the management of cystine calculi. J Endourol 1989;3:273. Krishnamurthi V, Streem SB. Long-term radiographic and functional outcome of extracorporeal shock wave lithotripsy induced perirenal hematomas. J Urol 1995;154:1673. Lingeman JE, McAteer JA, Kempson SA, et al. Bioeffects of extracorporeal shock wave lithotripsy: strategy for research and treatment. Urol Clin North Am 1988;15:507. Lingeman JE, Kulb TB. Hypertension following extracorporeal shock wave lithotripsy (Abstract). J Urol 1987;137:142A. Lingeman JE, Smith LH, Woods JR, et al. Urinary calculi: ESWL, endourology, and medical therapy. Philadelphia: Lea and Febiger, 1989. Lingeman JE, Siegel YI, Steele B, et al. Management of lower pole nephrolithiasis: a critical analysis. J Urol 1993;151:663. Lingeman JE, Woods JR, Nelson DR. Commentary on ESWL and blood pressure. J Urol 1995;154:2. Moran ME, Sandock D, Drach GW. Effects of high energy shock waves on chick embryo development (Abstract). J Urol 1990;143:230A. Streem SB, Yost A. Treatment of caliceal diverticular calculi with extracorporeal shock wave lithotripsy: patient selection and extended follow up. J Urol 1992;148:1043. Thomas R, Roberts J, Sloane B, et al. Effect of extracorporeal shock wave lithotripsy on renal function. J Endourol 1988;2:141. Vieweg J, Weber HM, Miller K, et al. Female fertility following extracorporeal shock wave lithotripsy of distal ureteral calculi. J Urol 1992;148:1007. Whelan JP, Finlayson B, Welch J, et al. The blast path: theoretical basis, experimental data, and clinical application. J Urol 1988; 140:401.

Chapter 121 Ureteral Stents and Endoscopic Treatment of Ureteral Obstruction Glenn’s Urologic Surgery

Chapter 121 Ureteral Stents and Endoscopic Treatment of Ureteral Obstruction
Gerhard J. Fuchs, Kamil Noordin, and Anup Patel

G. J. Fuchs: Department of Urology, U.C.L.A. Medical Center, Los Angeles, California 90095-1738. K. Noordin: Universiti Kebangsaan Malaysia, Salangor Darul Ehsan, Malaysia. A. Patel: Department of Urology, U.C.L.A. School of Medicine, Los Angeles, California 90095, and London E14 9UR, United Kingdom.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Stents, Drains, Follow-Up Outcomes Complications Results Chapter References

Strictures are one of the basic causes for ureteral obstruction and development of hydronephrosis with impairment of renal function. The literature does not clearly distinguish between a stricture (secondary to narrowing of the ureteral lumen caused by damage to the ureteral blood supply and/or scarring) and ureteral obstruction of other causes, such as extrinsic compression. In order to be called a ureteral stricture, a histologic lesion of the ureteral wall must be present. Common causes for ureteral strictures include iatrogenic damage, endoscopic surgery, extrinsic trauma, post chemo- or radiation therapy for malignancy, inflammatory factors, or malignant invasion. The vast majority of iatrogenic ureteral strictures referred to a urologist are related to ureteral injury at open surgery by nonurologic surgeons. The incidence of ureteral stricture after common pelvic operations is 0.5% to 1%, but following radical pelvic tumor surgery the incidence is 10%. Other relatively common causes for traumatic ureteral strictures include diathermy injury during laparoscopic surgery or secondary to traumatic ureteroscopy, yet the overall risk of ureteral stricture following endoscopic surgery is less than with open surgery (approx. 0.5% to 3%). Possible causes for endoscopic ureteral strictures include mucosal tears from overzealous balloon dilation, as well as mucosal tears or perforation caused by the instrument, accessories, or energy sources. Gunshot wounds account for 95% of the ureteral injuries secondary to abdominal trauma, whereas stab wounds or blunt abdominal trauma rarely affect the ureter. In patients undergoing suprapubic urinary diversion, strictures of the ureteroenteric anastomosis may occur, particularly if the blood supply of the ureter is tenuous or a long antireflux tunnel is made. If the urinary diversion was performed for the management of malignant disease, these strictures have to be distinguished from tumor recurrence at the distal ureter, which is of paramount importance if the planned treatment is endoscopic.

DIAGNOSIS
Ureteral strictures often display an insidious onset with the renal parenchyma irreversibly destroyed at the time of the diagnosis due to the slow development of “silent” hydronephrosis. Common signs of symptomatic strictures are flank pain mimicking stone colic or a patient presenting with obstructive pyelonephritis or urosepsis. Other symptoms may include an elevated creatinine level or decreased urine output (usually in patients with transplant kidneys and those with solitary kidneys). Basic evaluation of the patient with presumed ureteral obstruction is geared to identification of the cause of obstruction (rule out nonstricture causes such as stones, fungus balls, ureteral tumors, sloughed papilla, blood clots), assessment of renal function, and the ruling out or diagnosis of concomitant infection. If a stricture is confirmed, the location and length of obstruction are important parameters for planning of the treatment approach. Furthermore, the age of the stricture may also have prognostic value. The diagnosis is usually radiographic with contrast imaging studies (intravenous pyelogram or CT urogram). If renal function is impaired, renal ultrasound and noncontrast computer tomogram studies are first-line examinations. It is essential to perform pyelography as part of the retrograde cystoscopic placement of ureteral stents or, if retrograde stenting is not feasible, during placement of a percutaneous nephrostomy tube. Nuclear imaging studies are usually employed after decompression of the obstructed upper urinary tract with compromised function to assess the return of renal function prior to treatment. In most instances, radiologic diagnosis is straightforward and only occasionally is a diuretic intravenous pyelogram or a Whitaker test needed to further assess the otherwise equivocal findings. Urine osmolality comparisons from the percutaneous nephrostomy at baseline and 6 weeks after decompression may also be useful in evaluating the return of good renal function.

INDICATIONS FOR SURGERY
Therapeutic principles in the management of ureteral strictures include relief of the obstruction, if necessary with concomitant antibiotic treatment, followed by repair of the stricture with endoscopic surgery or open surgery. At times, surgical correction is not possible or prudent, such as in a patient with terminal cancer, where management may consist of temporary urinary drainage, such as that afforded by indwelling ureteral stents, percutaneous renal drainage, or ureterostomy. In patients undergoing definitive repair of the stricture, the site, length, and duration of the stricture are the basic parameters determining treatment ( Table 121-1). In general, strictures of less than 1.0 to 1.5 cm are amenable to endoscopic management, whereas longer strictures do not respond well to incisional treatment and need dismembered repair by open or laparoscopic surgery. With increased duration of the stricture, usually more than 3 months, the stricture tends to become hypovascular and when dilated the resultant scar tends to be denser than its predecessor. Therefore, incisional therapy is required rather than balloon dilation.

TABLE 121-1. Ureteral stricture treatment approaches based on ureteral stricture sites and length of stricture

Preoperative stenting relieves the obstruction, may restore renal function, and allows for eradication of a concomitant infection (with antibiotic treatment). Furthermore, the ureteral stent slowly distends the course of the ureter and facilitates endoscopic access to the stricture and manipulation/incision of the stricture. In cases of severe stricture when a guidewire cannot be passed via a cystoscope (retrograde) or under fluoroscopic guidance (antegrade), flexible endoscopy and direct visualization of the stricture site will often allow placement of a guidewire across the narrow segment, thus allowing insertion of a stent. Depending on the promptness

of return of renal function, the stent will be left in place for 2 to 6 weeks prior to treatment of the strictured segment.

ALTERNATIVE THERAPY
Alternative treatment options to endoscopic surgery for ureteral strictures include catheter dilation, balloon dilation, endoureterotomy (hot/cold knife intraluminal incision), Acucise balloon wire incision, and open surgical or laparoscopic excision of the strictured ureteral segment and reanastomosis. 2,3,4,5 and 6. Catheter dilation has been performed in the past with graduated ureteral catheters or coaxial catheters. Due to poor long-term patency results of less than 20%, this technique is now only useful for gaining access to a narrow ureteral segment for ureteroscopic incision. Likewise, balloon dilation up to 21 Fr employing high-pressure balloons has been rarely successful for longstanding hypovascular strictures, with reported success rates of 20% to 40%. Only strictures shorter than 1.0 cm and of very short duration have been reported to be successfully treated with balloon dilation (up to 70%). Presently, we use balloons only for fresh iatrogenic strictures secondary to pelvic surgery (e.g., partial ureteral obstruction by a periureteric suture). A recent modification of balloon dilation, the endoburst technique, utilizes a high-pressure balloon to disrupt the stricture and dilates the strictured area to 30 Fr. Although initial results have been encouraging, no long-term results are available and it is hard to believe that an uncoordinated bursting procedure could have better results than a carefully executed incision. A combination of a low-pressure balloon with an electrosurgical cutting wire is the Acucise device. 2,4 Long-term success with this technique for ureteropelvic junction (UPJ) stenosis of up to 75% can be expected. We feel that an endoscopically executed incision is always preferable when technically feasible because this gives total control over the completion of the incision. When we use the Acucise device we always control the depth of the cut with a flexible endoscope. In 12 of 16 cases the Acucise maneuver had to be performed repeatedly (at the same session) despite the radiologic appearance of a sufficient cut (loss of waist of balloon, extravasation of contrast). We therefore reserve the use of the Acucise for ureteral strictures that cannot be reached with a rigid endoscope, such as ureteroenteric anastomosis strictures. Open surgical techniques and, in experienced hands, laparoscopic repair of ureteral and UPJ strictures have become the rare exceptional indications. At our institution about 3% of all patients referred for UPJ stenosis repair require a dismembered technique, usually in patients with a large baggy renal pelvis draped over a crossing vessel, or crossing vessel compromising the lumen of the ureter. In these cases we usually employ the laparoscopic approach but open surgery is certainly still a valid option when there is not sufficient experience in laparoscopic surgery available. Likewise, excision of a longer strictured segment and a spatulated end-to-end anastomosis or ureteroneocystostomy usually require an open surgical approach, as does ureteral replacement with bowel segments or ureteral reimplantation.

SURGICAL TECHNIQUE
Endoscopically controlled incision (endoureterotomy/pyelotomy) is our procedure of choice for strictures not longer than 1.0 to 1.5 cm that can be accessed with an endoscope.2,6,7 In most female patients and whenever technically feasible in male patients the approach is by retrograde rigid ureteroscopy. We have previously discussed the benefits of preoperative stenting and although this technique is generally helpful it is not always necessary, especially with the miniaturization of the instruments and in female patients. A sterile urine is of utmost importance. Also, if there is a radiologically non-functioning renal unit, this should be drained first and treatment decisions held until a return of differential function to at least 25% is seen. General or epidural anesthesia is usually employed and perioperative antibiotics started at anesthesia induction. The patient is positioned in lithotomy position on an endoscopy table with fluoroscopy capabilities. Cystoscopy is performed using a 17-Fr cystoscope under video control; a retrograde pyelogram is then performed to assess the anatomy of the upper urinary tract and to determine the extent of the pathology. For incisional therapy we feel that it is imperative that a 0.038-in. floppy-tipped safety guidewire be positioned across the narrow ureteral segment and safely curled in the renal pelvis. Very rarely and only in the hands of the very experienced endourologist is there an indication to perform an incision without the benefit of a safety wire, such as in a combined antegrade/retrograde approach to connect an avulsed ureter or treat a completely occluded ureter. After placement of the safety wire, we routinely administer a diuretic (20 mg of Lasix) to induce diuresis and a positive pressure from the renal parenchyma to reduce the risk of pyelorenal reflux and infectious complications. The stricture is accessed usually with a 9.5-Fr semirigid ureteroscope, which rarely will require dilation of the ureteral orifice. In cases of narrowed ureters and/or previous pelvic surgery or radiation therapy, it is advisable to place an indwelling stent first for a period of 2 weeks to allow for slow distension of the entire ureter. This will more readily allow advancement of the ureteroscope and facilitate instrumentation in the ureter or at the UPJ. The rigid ureteroscope allows for precise positioning of the ureterotomy and is therefore the ideal instrument for incisional therapy. Flexible ureteroscopes, on the other hand, are reserved for cases where access with a rigid scope cannot be gained (some tall, male patients and in antegrade management of ureteral strictures or strictures of the ureteroenteric anastomosis or transplant kidney ureters). For most of our patients over the past 10 years we have used a cold knife whenever good endoscopic exposure could be gained, such as with percutaneous antegrade pyelotomy or in the distal ureter when a retrograde approach is used. However, when employing the retrograde ureteroscopic approach for repair of the UPJ, we prefer an angled electrode (“hot” knife, 70 W; pure cut) because this allows a better controlled deep ureterotomy than a cold knife. Likewise, we employed an electrode in the mid- and upper ureter, whereas in the distal ureter either technique can be used. Using the hot knife also affords the possibility of coagulation of small bleeders and forward cutting. It is advisable, however, to use electrocoagulation sparingly to avoid deep thermal injury to the ureteral blood supply and subsequent scar (stricture) formation. More recently (in the past 3 years), we have used the holmium laser technology for incisional therapy that allows vaporization of the fibrous scar tissue comprising the stricture. Due to the shallow depth of penetration (0.5 mm) the strictured ureteral segment can be incised layer by layer under direct vision with less inflammatory reaction around the incision than with other energy sources. Usually, a 365-µm fiber is used (0.5 J at 10 Hz) with rigid instruments and a 200-µm fiber (same energy settings) is available for use with flexible instruments. The smaller fibers now allow incisions in areas where rigid instruments do not reach (some male patients, intrarenal strictures, urinary diversion strictures). It is most effective to execute one deep cut through all wall layers until periureteral fat is visualized and the lumen of the ureter is wide open. We have found that using contrast extravasation as an indicator for a sufficient incision is not reliable in more than half of cases and therefore we always visualize the completeness of an incision endoscopically. Small bleeders of the vascular envelope of the ureter do not need to be coagulated and very rarely should intraoperative bleeding become a problem. The use of intraluminal ultrasound for evaluation of the periureteral vasculature is optional. 1 We have routinely used this 6.2-Fr catheter which is fitted with a 12.5-MHz to 20-MHz 360 degree rotating ultrasound transducer for the past 3 years. This allows the surgeon to study the topographic anatomy of the 1- to 2-cm area around the ureter through cross-sectional imaging with identification of potentially hazardous vessels in the vicinity of the scar to be incised. We have found this device helpful in the planning of an electro- or laser incision of ureteral strictures (especially in patients who have undergone previous surgery nd in patients with strictures of ureteroenteric anastomosis). While use of this device increases the comfort level of the surgeon in making the incision into an area that has been visualized and vascular hazards have been ruled out, it is by no means necessary to have this aid when engaging in incisional therapy. Stents, Drains, Follow-up All endoureterotomies and endopyelotomies share one final pathway, i.e., the placement of a ureteral stent to minimize leakage of urine into the retroperitoneum and to facilitate healing of the ureteral wall. Animal research has shown that the longitudinally incised, partially resected ureter heals in a patent fashion around a stent with regeneration of both the urothelial and muscular layers in about 6 weeks. Our present routine is to leave a 7-Fr indwelling ureteral stent and drain the bladder via Foley catheter (after incisions of the distal and midureter) ( Table 121-2). On postoperative day 3, a voiding cystourethrogram (VCUG) is obtained. If there is no evidence for continued extravasation, the Foley catheter is removed. The indwelling stent is removed 6 weeks postoperatively and at that time we obtain a renal ultrasound as a baseline examination.

TABLE 121-2. Ureteral stents and endoureterotomy

Follow-up of the asymptomatic patient includes a return visit at 6 weeks for a contrast imaging study (Lasix IVP/scintigram) depending on which studies were obtained preoperatively. If patency of the ureter and unimpaired drainage are confirmed, this imaging study is then matched with a renal ultrasound and further follow-up is via ultrasound at 3, 6, and 12 months. Also at 1 year we obtain a diuretic imaging study, and if this confirms patency the patient is followed with yearly ultrasound to capture the occasional late failure. In the patient with a ureteroenteric anastomosis stricture we usually approach the stricture in an antegrade fashion and the stenting protocol is the same as for an antegrade UPJ incision. 7,8 A 7-Fr antegrade stent in a 22-Fr renal drainage tube is placed all the way down into the bladder, which avoids the distal ureteral stricture formation that was seen with stents ending in the ureter. After 1 week a nephrostogram is obtained; if patency is confirmed and there is no extravasation at the incisional site, the nephrostomy tube is clamped. We then remove the ureteral stent after 6 weeks and obtain a nephrostogram through the indwelling PCN tube. If unimpaired drainage down to the bladder is demonstrated, the nephrostomy tube is again clamped, and it is well tolerated (no pain or fever), the nephrostomy is removed within 48 hours. If the nephrostogram shows obstruction at the UPJ, which is usually secondary to edema from the tube, we start intermittent clamping for several days and repeat the nephrostogram after 1 week. Ureteral strictures in patients with transplant kidneys mostly require a percutaneous antegrade or combined antegrade/retrograde technique and the stenting protocol usually is the same as for the patient with an antegrade UPJ incision.

OUTCOMES
Complications Potential complications of treatment of ureteral strictures include bleeding, damage to adjacent organs, infection, extravasation with urinoma and restricture, and loss of renal function. When the anatomic relations of the ureter and surrounding structures are observed, bleeding and damage to adjacent organs can be kept to a minimum. Of major concern is the presence of aberrant vessels, especially for UPJ incisions where the incidence of vascular anomalies is up to 40%. However, in the vast majority of cases the presence of a crossing vessel does not preclude incisional therapy because only few crossing vessels (5%) do indeed impact on the lumen of the ureter.1,5,8,9 and 10 A spiral CT with contrast (CT urogram) of the area of interest allows noninvasive preoperative diagnosis of the vasculature. Intraluminal ultrasound is also an accurate, but not widely available, intraoperative modality for diagnosis of periureteral vessels and other anatomic structures. 1 Infectious complications can be kept to a minimum when the urine is checked at certain critical junctures such as 5 days prior to any treatment and prior to any postoperative manipulation, such as VCUG, nephrostogram, and removal of any drainage tubes. When appropriate antibiotic therapy is initiated prior to any manipulation, infectious complications are extremely rare. Proper drainage is the number one precaution against urine extravasation and complications thereof. Extended follow-up will help prevent the complication of “silent” failure and subsequent loss of renal failure. Results Success rates of 80% to 90% can be expected for ureteroscopic incision of UPJ stenosis and ureteral strictures. 3,6,7 More recently the holmium laser has been used in 24 cases of dense strictures and UPJ stenosis, and preliminary results (up to 24 months) confirm patency in all cases (two cases needed two sessions). Although endoscopic and open surgical stricture repairs enjoy success rates of 80%, the individual response cannot be predicted and extended follow-up is of importance. Failure of a procedure will usually be noted within the first 3 to 6 months; late failure is only seen in about 10% of patients. 8,10 We usually obtain a renal ultrasound at the time of removal of the last stent as a baseline. Six weeks later an ultrasound is obtained. If this ultrasound is normal, the next ultrasound is obtained 6 weeks later. If there is evidence of increasing hydronephrosis the kidney is checked in 2-week intervals and contrast imaging studies are obtained as needed. For the first 6 months the patient is followed with ultrasounds at 4-week intervals and at month 6 an IVP is obtained. If there is no indication for failure, follow-up ultrasounds are obtained at 9 and 12 months, then every 6 months for a year, and from then on an as-needed basis. Management of ureteral strictures can to a large extent be achieved endoscopically in an effective, minimally invasive fashion. Proper patient selection, careful endourologic technique, and extended follow-up will guarantee success in 80% to 90% of patients. Open surgery is reserved for complex stricture problems and endoscopic treatment failures. New technology for diagnosis and treatment has shown some promising results; however, further follow-up is needed to determine the role of this costly technology. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Bagley DH, Liu JB, Grasso VR. Endoluminal sonography in evaluation of the obstructed ureteropelvic junction. J Endourol 1994;8:287. Chandhoke PS, Clayman RV, Stone AM, et al. Endopyelotomy and endoureterotomy with the Acucise ureteral cutting balloon device: preliminary experience. J Endourol 1993;7:45. Clayman RV, Basler JW, Kavoussi L. Ureteronephroscopic endopyelotomy. J Urol 1988;144:246. Faerber GJ, Richardson TD, Farah N. Retrograde treatment of ureteropelvic junction obstruction using the ureteral cutting balloon catheter. J Urol 1997;157:454. Gupta M, Smith AD. Crossing vessels at the ureteropelvic junction: do they influence endopyelotomy outcomes? J Endourol 1996;10:183. Inglis JA, Tolley DA. Ureteroscopic pyelolysis for pelviureteric junction obstruction. Br J Urol 1986;58:250. Meretyk I, Meretyk S, Clayman RV. Endopyelotomy: comparison of ureteroscopic retrograde and antegrade percutaneous techniques. J Urol 1992;148:775. Motola JA, Badlani GH, Smith AD. Results of 212 consecutive endopyelotomies: an 8-year follow-up. J Urol 1993;149:453. Sampaio FJ, Favorito LA. Ureteropelvic junction stenosis: vascular anatomical background for endopyelotomy. J Urol 1993;150:1787. Van Cangh PJ, Wilmart JF, Opsomer RJ, Abi-Aad A, Wese FX, Lorge F. Long-term results and late recurrence after endoureteropyelotomy: a critical analysis of prognostic factors. J Urol 1994;151:934.

Chapter 122 Basic Laparoscopy: Transperitoneal and Extraperitoneal Approaches Glenn’s Urologic Surgery

Chapter 122 Basic Laparoscopy: Transperitoneal and Extraperitoneal Approaches
Leonard G. Gomella and David M. Albala

L. G. Gomella: Department of Urology, Thomas Jefferson University School of Medicine, Philadelphia, Pennsylvania 19107. D. M. Albala: Department of Urology, Loyola University School of Medicine, Maywood, Illinois 60153, and Hines Veterans Hospital, Hines, Illinois 60141-5142.

Indications for Surgery Alternative Therapy Surgical Technique Transperitoneal Laparoscopy Establishing the Pneumoperitoneum Primary Trocar Placement Hasson Technique (Open Laparoscopy) Secondary Trocar Placement Trocar Removal Extraperitoneal Laparoscopy Developing the Extraperitoneal Space Pelvic Extraperitoneal Laparoscopy Retroperitoneal Laparoscopy Pediatric Laparoscopy Postoperative Management Outcomes Complications Results Chapter References

Traditional open urologic surgical procedures can be performed by the transperitoneal or extraperitoneal approach. Modern laparoscopic techniques now permit many of these traditionally open procedures to be performed using minimally invasive surgery. 6 Chapter 123, Chapter 124, Chapter 125, Chapter 126, Chapter 127, Chapter 128, Chapter 129, Chapter 130, Chapter 131, Chapter 132 and Chapter 133 include laparoscopic procedures commonly performed by urologists in the United States. This chapter provides the basis for any laparoscopic procedure and describes techniques for entering the intra- or extraperitoneal space, insufflate, place viewing and access ports, and the steps necessary to complete the procedure.

INDICATIONS FOR SURGERY
Laparoscopic surgery is indicated in lieu of many open procedures in which the goal is to minimize the recovery period for the patient. Contraindications to transperitoneal laparoscopy include inability to tolerate general anesthesia or pneumoperitoneum (i.e., severe cardiac or pulmonary disease), extreme obesity, intestinal obstruction and/or substantial distention, massive hemoperitoneum, generalized peritonitis, extensive prior abdominal surgery, abdominal wall infection, uncorrectable coagulopathy, large abdominal hernias, or advanced intraabdominal malignancy. Contraindications to extraperitoneal laparoscopy include most contraindications as listed above, plus prior surgery or inflammation in the extraperitoneal space. Patients should be informed of the alternative approaches available, including open surgery. The patient also must be informed of the risk that the procedure may be terminated or converted to an open approach due to inability to complete the procedure or complications. Patients often view laparoscopic surgery so favorably that they may expect to have no discomfort and return to full activities immediately. The patient must understand that interventional laparoscopy is still a surgical procedure with some of the inconveniences associated with any operation.

ALTERNATIVE THERAPY
Alternatives to laparoscopic surgery include open surgery, cystoscopic or other endoscopic procedure, or percutaneous approaches under endoscopic or x-ray control.

SURGICAL TECHNIQUE
Transperitoneal Laparoscopy The majority of laparoscopic procedures are performed through the transperitoneal approach. The peritoneal cavity can be reliably entered and distended. It provides a large working space. Patient preparation including a routine bowel preparation, such as milk of magnesia, the day before and an enema given on the day of surgery will help decompress the lower intestine (desirable during procedures such as pelvic lymphadenectomy). General endotracheal anesthesia is essential for interventional laparoscopy. Patients should be positioned with adequate padding and secured to the table to allow table repositioning during the procedure. The arms should be tucked at the sides after large-bore intravenous access is completed. A vac-pack is useful for procedures such as transperitoneal nephrectomy that require the patient to be rolled to a lateral position. Nasogastric and urinary catheters lessen the risk of visceral injury and facilitate visualization. Broad-spectrum antibiotics are used in the event of inadvertent bowel injury. The field should be prepped widely in case there is a need to convert to an open procedure. For pelvic laparoscopy, such as lymphadenectomy in males, the genitalia should be prepped into the field so any pneumoscrotum that may develop can be decompressed. For procedures such as bladder neck suspension in females, the patient should be in the lithotomy position with the vagina prepped. Necessary equipment should be reviewed by the surgeon in charge of the procedure. Detailed information on laparoscopic instrumentation is available elsewhere. 5 The positioning of monitors should be determined based on the procedure. For pelvic surgery, a single monitor at the foot is usually sufficient; for procedures such as nephrectomy, having monitors on either side of the bed is recommended. Establishing the Pneumoperitoneum The most commonly used site for introduction of the Veress needle for insufflation is at the inferior or superior margin of the umbilicus. The umbilicus is the central point of the peritoneal cavity, making it an ideal observation site. Here the abdominal wall is only two layers thick (fascia and peritoneum) and is easy to traverse percutaneously (Fig. 122-1). Placement of the needle in the abdominal cavity is a blind procedure with potential for injury to the underlying structures. If the patient has undergone prior surgery, open laparoscopy using the Hasson technique is considered the safest technique (see below).

FIG. 122-1. The Veress needle is most often placed at the inferior or superior aspect of the umbilicus. The needle is directed into the hollow of the pelvis, below the bifurcation of the great vessels.

The patient should be placed in 10 to 20 degree Trendelenberg position to help lift the intestines out of the pelvis. The Veress needle type has a spring-loaded, blunt-tipped obturator to help to prevent injury. A small incision is made at the inferior or superior border of the umbilicus. The abdominal wall can be elevated with a sponge or by grasping the periumbilical area with towel clamps. These maneuvers may raise the umbilicus up and away from the intestines but, more importantly, stabilize the abdominal wall. The needle is grasped like a dart along the shaft to limit its excursion into the abdominal cavity with the angle of entry directed into the pelvis. Two distinct “pops” are felt as the needle passes through the fascia and peritoneum. As an alternate technique, the patient may be left flat and the needle is directed at a 60 degree angle to the sacral promontory, below the bifurcation of the great vessels. The umbilicus lies directly over the aortic bifurcation at L4–L5 ( Fig. 122-1). Deep penetration with the needle directly posteriorly could produce vascular injury. Once the needle is in the peritoneal cavity, confirmatory tests are performed prior to insufflation, although none are foolproof. These include: 1. “Color test.” Aspiration of colored (red, yellow, green, brown) or malodorous fluid suggests improper placement. 2. Drop test. Apply a drop of saline inside the hub of the needle and lift the abdominal wall. If in proper position, the drop will enter the abdomen due to the negative intraperitoneal pressure. 3. Modified Palmer test. Inject 10 ml of saline into the needle and attempt to aspirate. Inability to aspirate the fluid suggests that the fluid has dispersed into the abdomen and the needle is in correct position. 4. Initial pressure reading less than 8 mm Hg. The insufflator is turned on with no flow to obtain a pressure reading. 5. A decrease in pressure with elevation of the abdominal wall. If perforation of a viscus occurs, the needle should be removed and discarded. A new needle may then be inserted at another location or the surgeon may choose to perform an open laparoscopy using the Hasson technique (see below). Complete the insufflation and insertion of the primary trocar. The injury can be examined with the laparoscope and a decision made as to the appropriate management. If blood appears, the Veress needle should be withdrawn slightly. Its positioning should be retested, the pneumoperitoneum established, and the injury inspected for possible repair. Because of the small size of the Veress needle, the majority of injuries do not require open operative intervention. Begin CO2 insufflation at low flow rate; this should initially maintain an intraabdominal pressure of less than 8 mm Hg. A high initial intraabdominal pressure with a low flow suggests improper needle placement. Once insufflation is underway, the abdomen should be observed to assure that a symmetric pneumoperitoneum is developing. Monitor the insufflation by percussion over the liver, noting the characteristic dull echo tone that indicates proper insufflation. If the distention appears correct, the rate of flow of CO 2 may be increased to “high flow” (usually more than 6 L/min). Insufflate until an intraabdominal pressure of 15 to 20 mm Hg is reached in the adult; a total volume of 5 to 6 L in an average adult should generate an adequate pneumoperitoneum. The pressure should be decreased to 10 to 12 mm Hg after all trocars are placed to prevent complications due to barotrauma. If preperitoneal insufflation occurs, an attempt can be made to open the peritoneum with laparoscopic scissors and guide the tip of the trocar beneath it. CO 2 is then insufflated into the true peritoneal cavity, compressing the preperitoneal gas. Alternatively, evacuation of the preperitoneal space with a needle and syringe, and reinsertion of the needle at the same or another site (i.e., superior umbilical position) may be attempted. Open laparoscopy can be used if these techniques are unsuccessful. Primary Trocar Placement When the pneumoperitoneum is established, the Veress needle is removed and the primary (laparoscope) trocar is inserted. A 10- or 10/11-mm trocar is used in adults to allow passage of the laparoscope. ( Note: The trocar designation refers to the size of the instrument that can be inserted into the trocar and not the overall diameter of the trocar.) The incision site used for the Veress needle should be enlarged and the subcutaneous tissue spread to the fascia using a hemostat. The skin incision should be sufficient to allow the trocar to pass without resistance. It is helpful to press the end of the trocar on the skin to create an impression that serves as a guide to the size of the skin incision. The proper technique to insert a trocar is to apply firm, steady pressure with a gentle, twisting movement. A finger can be held along the shaft to serve as a brake from pushing the trocar in too far ( Fig. 122-2). Pressure should be applied using the arm and elbow and not the weight of the shoulder and upper body. The trocar should be directed into the pelvis below the bifurcation of the great vessels in the midline ( Fig. 122-2). The abdominal wall should be lifted and stabilized with hemostats ( Fig. 122-3). Entry into the peritoneum is usually indicated by a sudden decrease in resistance and the characteristic clicks made by the trocar safety shield locking in place. Most disposable trocars have a safety shield that covers the sharp tip after peritoneal entry to limit injury. The sharp obturator is removed and a quick check will determine if the trocar has entered the peritoneal cavity. Briefly opening the desufflate valve or the stopcock on the trocar will cause a rush of gas to escape (“whoosh test”), suggesting intraperitoneal placement.

FIG. 122-2. Primary trocar is held with a finger along the shaft to limit the depth of entry of the trocar.

FIG. 122-3. The abdominal wall is stabilized with towel clips placed on either side of the umbilicus during primary trocar insertion.

The chance of a serious injury at the time of the insertion of the primary trocar is much greater than with the Veress needle because of its larger size and the greater force required to penetrate the abdominal wall. Although the ideal working pressure of the pneumoperitoneum is 10 to 12 mm Hg, an initial pressure of 15 to 20 mm Hg is needed until the all trocars are placed to ensure a tense pneumoperitoneum. The CO2 insufflator is connected to the stopcock on the trocar and insufflation is resumed. The laparoscope, attached to a high-resolution video camera, should be white-balanced and introduced for exploration of the abdomen. Visual inspection of the abdomen is necessary to assess possible injury by the Veress needle or trocar. Lens fogging is caused by passage of the room temperature laparoscope into the warm, humid abdomen. As the laparoscope warms during the procedure, fogging becomes less problematic. To deal with fogging, the following techniques are useful: heating the laparoscope before insertion in sterile heating blocks or in warmed sterile saline; applying antifogging solutions; and using self- cleaning/heating laparoscopes (Hydrolaparoscope, ACMI, Stamford, CT). Limit the amount of time that the scope is removed during the procedure to prevent cooling of the lens. If fogging or debris covers the lens, it can be gently wiped on a clean piece of bowel; touching fat or a blood-tinged surface may leave a film on the end of the laparoscope. Hasson Technique (Open Laparoscopy) Many surgeons use this as the primary access for all patients and do not use the Veress needle pneumoperitoneum and initial blind primary trocar placement. The open laparoscopy (Hasson technique) is also useful both for correcting preperitoneal insufflation with the Veress needle and for high-risk patients with multiple adhesions. The advantage is that entry into the peritoneal cavity is under direct vision, minimizing the risk of injury. An infraumbilical incision is made and two stay sutures (such as 0 prolene) are placed through the fascia on either side. The fascia and peritoneum are directly visualized and opened ( Fig. 122-4A). The Hasson-style trocar has a blunt tip and different features based on the manufacturer. Some trocars have a tapered end that seals in the pneumoperitoneum, whereas others rely on retention balloons or grips to maintain the trocar in the peritoneal cavity. The stay sutures are either attached to the cannula to hold it in position, if needed, or secured with hemostats for use during closure of the fascia ( Fig. 122-4B). The insufflation tubing is attached to the Hasson-style trocar and immediately high flow can be selected. The pneumoperitoneum should develop rapidly.

FIG. 122-4. (A) Open laparoscopy using the Hasson technique. The peritoneum is entered under direct vision. (B) A blunt tipped Hasson style is secured using the stay sutures.

Secondary Trocar Placement The placement of secondary or working trocars can be carefully monitored externally and internally to reduce the risk of injury. In addition to reusable trocars with safety shields, checking that there is full pneumoperitoneum before trocar insertion, spreading subcutaneous tissue with a hemostat, using an adequate skin incision, verifying anatomic landmarks, and directing the trocar into the pelvis will limit trocar injuries. The room is darkened and the light from the laparoscope is used to transilluminate the anterior abdominal wall to identify vessels such as the superficial inferior epigastric. Trocar sizes and sites are selected according to the procedure to be performed, the patient's anatomy, and the surgeon's preference. At least one larger secondary trocar (10-mm or 10/11-mm) is usually placed to allow the passage of larger instruments such as clip appliers and to allow removal of specimens. Linear stapling devices are passed through 12-mm trocars. Extended length trocars are available if patients have particularly thick abdominal walls. Trocar placement for specific procedures is noted in the appropriate chapter that follows. The selected site for the secondary trocar is gently pushed with the index finger while the site is observed through the laparoscope. The site is evaluated to confirm that there are no underlying vessels or bowel. A skin incision of appropriate diameter to accommodate the trocar is made. Skin incisions should be sufficient to remove the skin as a point of resistance during trocar insertion but not large enough to allow gas leakage. Spread the subcutaneous tissue down to fascia with a clamp until the impression of the clamp tip is visible on the peritoneum. The hemostat is removed and the port is introduced with the same technique described for the primary trocar except that the progress into the abdomen is followed on the monitor. Stability threads, if used, should be screwed into position so that one thread is visible through the peritoneum. Suturing trocars to the skin using a heavy silk tether to prevent accidental removal during the case may be used in place of the stability threads. Orient secondary trocars to the surgical site. This configuration minimizes the side-to-side excursion required of the ports and the crossing swords problem is avoided. A general rule is to place secondary trocars in the midline or at least 8 cm from the midline in adults to avoid the rectus sheath ( Fig. 122-5). If the rectus muscle is penetrated by the trocar, there is risk of bleeding from the muscle or epigastric vessels.

FIG. 122-5. Secondary trocars should be placed in the midline (linea alba) or at least 8 cm from the midline to avoid injury to the rectus or the epigastric vessels. (Shaded area is generally safe.)

Trocar Removal At the end of the procedure, the pressure should be lowered to 8 mm Hg or less to observe for any bleeding that may be tamponaded by the pneumoperitoneum. A brief survey of the abdomen should verify that there was no injury outside the operative field. Trocar sites should be examined prior to and after removal of the sheaths for bleeding or herniation. During trocar removal and fascial closure, bowel can become trapped. Trocar removal and suturing should be observed laparoscopically. Ports that are 10 mm or larger may require fascial suturing in adults. If the Hasson technique was used, the umbilical port can usually be closed by tying the preplaced sutures together. Before the final trocar is removed, as much CO 2 as possible should be expelled by compressing the abbdomen with the flapper vale or stopcock opened on the last port to reduce postoperative shoulder pain due to diaphragmatic irritation from CO 2. Pneumoscrotum if present should also be manually decompressed bat this time. If standard placement of the initial 10-mm port was used, the following procedure is useful: All secondary 10-mm sites are sutured while the umbilical camera is in position. A cystoscope or 5-mm laparoscope is placed in a 5-mm trocar site and the closure of the umbilical site is observed intraabdominally. The pneumoperitoneum is evacuated by holding down the flapper valve and manually compressing the abdomen, and the final 5-mm port is removed. Fascial closure can be accomplished by a variety of techniques. Typically 0 or 2-0 absorbable suture (i.e., Vicryl) on a small-curved needle (i.e., CT-3 or UR-5) is used. Fascial edges are grasped with toothed forceps or Scanlon clamps. Army/navy retractors facilitate exposure. Laparoscopic closure needles (i.e., Endoclose, U.S. Surgical, Norwalk, CT) are also available that are similar to a Stamey-style needle. A free 0 or 2-0 absorbable suture is engaged and passed percutaneously along the side of the trocar, through the fascia and into the abdominal cavity using laparoscopic guidance. ( Note: if stability threads were used to secure the trocar, they may interfere with passage of the closure device. Release the thread and slide it out of the incision while maintaining the trocar in the abdominal cavity.) The end of the suture is held with a laparoscopic grasper while the spring-loaded end is depressed to release the suture (Fig. 122-6). The closure device is then passed on the opposite side of the trocar, with the free end of the suture placed in the end of the needle using laparoscopic guidance. Once engaged, the device is withdrawn with the attached end of the suture. The trocar is removed and the ends of the suture tied from the outside using laparoscopic control.

FIG. 122-6. One technique to close the trocar site uses a device such as the Endoclose fascial closure needle. Here the 2-0 absorbable suture has been passed into the abdomen alongside the trocar. The suture is grasped, the needle is withdrawn and passed on the opposite side of the trocar.

The skin site should be thoroughly irrigated prior to closure since herniation is associated with trocars larger than 10 mm. Often, when it does occur, it is due to wound infection rather than improper fascial closure. Skin sites are closed with staples or subcutaneous suture reinforced by Steri-strips.

EXTRAPERITONEAL LAPAROSCOPY
Traditional laparoscopy approaches the target organs through the peritoneum, in contrast to open urologic surgery, which is frequently in the extraperitoneal (pre- and retroperitoneal) space. The use of transperitoneal laparoscopic techniques for a variety of procedures is widely accepted. However, significant complications, such as bowel or vascular injuries, can occur when utilizing the transperitoneal approach. The extraperitoneal approach has been investigated for a variety of procedures, including lymphadenectomy and bladder neck suspension as well as ureteral and renal procedures. There are several potential advantages to the extraperitoneal laparoscopic approach. By avoiding the peritoneal cavity, the risk of visceral and vascular injury may be reduced and preperitoneal landmarks, such as Cooper's ligament and the iliac vessels, can be visualized directly. Intestinal retraction may be easier because the peritoneal envelope surrounds the intestines and individual bowel loops need not be retracted. Prolonged ileus may be less common and fluid may be more easily contained. Herniation through trocar sites, postoperative ileus, and adhesions may be less than that of the transperitoneal approach. The extraperitoneal approach may be limited due to prior surgical procedures or inflammatory processes that may obliterate this potential space. Excessive fat in the extraperitoneal space may obscure the anatomy, particularly in the retroperitoneum. However, the more obese patient may be better suited for an extraperitoneal pelvic lymphadenectomy than the transperitoneal approach. The limited retroperitoneal working area may make placement of trocars more difficult, as well as limiting the removal of large masses, such as renal tumors. There are data to suggest that there may be more CO 2 absorption in the extraperitoneal space. This can be generally managed by aggressive ventilation. 8 Developing the Extraperitoneal Space Simple extraperitoneal insufflation will cause the gas to track along fascial planes and not develop the space. Balloon dissection of the extraperitoneal space is key to performing any procedure in this area. Gaur was the first to describe balloon distention of the extraperitoneal space using a simple device consisting of a surgical glove finger mounted on a red rubber catheter secured with a silk tie. 3 The catheter is connected to a blood pressure bulb insufflator and the balloon is inflated to 110 mm Hg intermittently until a visible bulge is seen in the abdomen. Workers at the University of Iowa have replaced the glove finger with a more durable finger cot from an O'Connor-style drape. Alternatively, they recommend using two glove fingers tied over each other to limit rupture. Instead of using air, saline could be used to

distend these self-made balloons. 10 Several commercially available trocar-mounted balloons are available. The PDB (Origin Medsystem, Menlo Park, CA) is a clear silicone balloon mounted on a 10-mm trocar with a capacity of 700 to 1000 ml of air. An advantage of this device is that the inflation and distention of the space can be laparoscopically monitored through the clear balloon. The Spacemaker (General Surgical Innovations, Portola Valley, CA) is trocar-mounted and designed to be filled with saline. Pelvic Extraperitoneal Laparoscopy Bladder neck suspension, hernia repair, and pelvic lymph node dissection can be approached in the pelvic extraperitoneal space. Patient preparation and positioning are identical to that of transperitoneal laparoscopy. A vertical skin incision is made 1 to 2 cm below the inferior umbilical crease in order to avoid the confluence of the anterior and posterior rectus sheaths at the umbilicus. The skin incision should be large enough to accept a 10/11- mm trocar. The tissues are spread with a clamp to the anterior rectus sheath. Two absorbable O (i.e., Vicryl) stay sutures are placed in each side of the midline. Next, the anterior rectus sheath is incised along the linea alba between the sutures. The two bellies of the rectus muscle are separated by blunt dissection and a finger is passed behind the rectus and above the posterior fascia. Blunt finger dissection is used to create the access space between the rectus and the posterior sheath. A balloon device of choice is lubricated with sterile jelly and passed behind the rectus but above the posterior sheath, and it enters the preperitoneal space at the level of the arcuate line ( Fig. 122-7). This is a reliable means of avoiding entry into the peritoneum. The balloon can be passed inferiorly with external manual guidance down to the pubis.

FIG. 122-7. Technique of trocar-mounted balloon dissection of the preperitoneal space.

The balloon is inflated to approximately 700 to 1000 ml with room air using an inflation bulb or with saline, depending on the device chosen or the surgeon's preference. Using a device such as the PDB, it is possible to introduce a laparoscope inside the balloon cavity to ensure adequate dissection and expansion of the preperitoneal space. The balloon is left inflated for several minutes, deflated, and inflated a second time. The balloon is removed and a 10-mm Hasson-style cannula placed. The CO 2 insufflator is attached and set on high flow at 8 to 10 mm Hg. Inspection of the preperitoneal space usually confirms adequate dissection of the prevesical space. The right side typically dissects more completely than the left. However, a minimal amount of blunt dissection will expose the external iliac vessels on the left side if needed. Additional trocars are placed under direct vision. The pressure is briefly increased to 15 mm Hg to facilitate secondary trocar placement. Prolonged pressure at this level may cause tracking of the gas into the subcutaneous tissues. Trocar insertion is identical to the transperitoneal approach (transillumination, visual inspection for vessels, etc.). It is critical to ensure that the lateral ports do not traverse the peritoneal cavity and that they are beyond the peritoneal reflection. After the trocars are placed, the pressure can be decreased to 10 mm Hg, which is maintained throughout the procedure. As an alternative approach, especially for procedures such as bladder neck suspension, where lateral exposure at the level of the iliac vessels is not needed, a small vertical incision is made in the midline about one-third of the distance below the umbilicus, between the umbilicus and pubis. Under vision, the rectus fascia is incised and stay sutures are placed on the fascia. The extraperitoneal space is developed with an index finger. The balloon device is inserted and inflated as above. Retroperitoneal Laparoscopy This form of laparoscopy, also known as retroperitoneoscopy, is used most often for renal and ureteral procedures. Studies have demonstrated that the peritoneum is never more posterior than the posterior axillary line and that placing the patient in the lateral position further moves the peritoneum anteriorly. 2 Finger dissection can then free up the anterior lip of the peritoneal reflection and allow more anterior trocar placement. Stenting of the ipsilateral ureter preoperatively is often helpful. Following the usual preparation for transperitoneal laparoscopy, the patient is placed in standard flank position. A 2-cm transverse skin incision is made just anterior to the tip of the 12th rib. An alternate site is Petit's triangle approximately two fingerbreadths above the anterior superior iliac spine ( Fig. 122-8). The posterior layer of the thoracolumbar fascia is identified and two stay sutures are positioned. The flank muscles are split to the anterior thoracolumbar fascia; two additional stay sutures are positioned. The anterior layer of thoracolumbar fascia is incised and the retroperitoneal space entered. To enter the retroperitoneum, the index finger should be used to develop the space initially and sweep the peritoneum away anteriorly.

FIG. 122-8. For retroperitoneal procedures, the initial trocar is placed 1 to 2 cm from the tip of the 12th rib (A) or in Petit's triangle (B), 1 to 2 cm above the anterior superior iliac spine. Additional ports can be placed as needed between the posterior and anterior axillary lines.

If a renal procedure (i.e., cyst excision, renal biopsy) is planned, it is helpful to place the balloon directly into Gerota's fascia to facilitate the procedure. The index finger is introduced into the retroperitoneum in a cephalad direction to palpate the lower pole of the kidney. In a thin patient, Gerota's fascia can be pierced with the finger and the perirenal space entered. Finger dissection within Gerota's fascia creates an adequate space for placement of the balloon adjacent to the renal parenchyma. In obese patients, the abundant retroperitoneal fat may preclude localization of the lower pole with finger dissection. Here Gerota's fascia can be grasped with two Scanlon or Allis clamps and delivered into the skin incision. Gerota's fascia is incised under vision and the balloon dilator is visually placed in Gerota's fascia. The retroperitoneal approach for laparoscopic nephrectomy in obese patients is considered to be contraindicated.

A commercially manufactured or self-made balloon, as noted above, is inserted in the retroperitoneum. The passage of the self-made balloon can be facilitated by backloading into a large Amplatz dilator sheath. The degree of balloon dilation varies with the patient's body habitus: 1 to 1.2 L of normal saline or room air in the average adult (Fig. 122-8). The balloon is kept inflated for 5 minutes to facilitate hemostasis. Following balloon removal, a Hasson blunt-tipped trocar is inserted in the retroperitoneum and secured with the preplaced stay sutures or with the trocar's retention mechanism (Fig. 122-9). Visualization of the psoas confirms proper balloon dilation of the retroperitoneum (or the lower pole of the kidney if the dilator was placed inside Gerota's fascia). The peritoneal envelope may be further mobilized with sweeping motions of the laparoscope. If difficulty is encountered in mobilizing the peritoneum, an operating laparoscope can be used to pass a blunt-tipped dissector into the retroperitoneum.

FIG. 122-9. Balloon dissection of the retroperitoneal space has been accomplished and the first port has been placed.

The insertion of secondary ports under manual, rather than laparoscopic, control to minimize trocar injury to the peritoneum has been described. 4 The laparoscope and the Hasson cannula are removed and the index finger of the left hand (for a right-handed surgeon) is introduced into the retroperitoneum. An S-shaped retractor is inserted into the retroperitoneum in such a manner that it lies immediately in front of the finger (i.e., the retractor is cradled by the finger). The finger tip mobilizes and retracts the peritoneum away from the abdominal wall; the S retractor prevents inadvertent trocar injury to the surgeon's finger. With the surgeon's right hand, the secondary ports are inserted under bimanual control. The aim is to introduce the trocar onto the S-shaped retractor. Secondary ports are placed based on the procedure to be performed. Typical placements for retroperitoneal procedures are demonstrated in Figure 122-8. No ports are positioned behind the posterior axillary line.

PEDIATRIC LAPAROSCOPY
Most principles of adult laparoscopy can be applied to the pediatric population with the following exceptions: 1. 2. 3. 4. In younger children, the bladder is an intraabdominal structure requiring more care with lower abdominal trocar placement. The total CO2 to insufflate is generally 2 to 3 L. Pressures should be kept lower than in adults (8 to 10 mm Hg). The 5-mm trocars are commonly used in children and must be sutured at the end of the case. Many standard laparoscopic instruments will be too large or long for pediatric laparoscopy; thus, specialized equipment is required.

Postoperative Management
Depending on the procedure, clear liquids can be given after the effects of anesthesia have cleared with diet advanced as tolerated. For procedures such as pelvic lymphadenectomy, discharge is typically within 24 hours. For more extensive procedures (i.e., nephrectomy), it may be necessary to advance the diet more slowly. The requirement for postoperative analgesics is usually minimal; excessive pain should raise the suspicion of a complication. Patients should be advised that delayed postoperative bruising is occasionally encountered. Return to full activities is more rapid than with the equivalent open procedure.

OUTCOMES
Complications Open laparoscopy using the blunt-tipped Hasson trocar offers the opportunity to minimize the risk of trocar injury. Significant perforation of a major blood vessel, bladder, or gastrointestinal tract with a trocar should be managed by immediate laparotomy and repair. The trocar and sheath should be left in place while the surgeon opens the abdomen to minimize further contamination or bleeding and to identify the site of injury. In selected instances, laparoscopic repair may be appropriate. Vascular injury within the abdominal wall, especially the inferior epigastric vessels, is sometimes seen. If there is a vascular injury to the anterior abdominal wall, a through-and-through suture can be placed on a bolster using a Stamey-style needle or laparoscopic closure device, then removed after 24 to 48 hours. Anesthetic problems can be caused by the absorption of CO 2 or the physiologic effects of the pneumoperitoneum. Both the intra- and extraperitoneal surfaces can readily absorb CO 2 and may cause hypercarbia. End-tidal CO 2 monitoring by the anesthesia team can often identify this problem before it becomes clinically significant. Increasing the minute ventilation can usually keep the blood CO 2 levels in a safe range. Barotrauma can be minimized by keeping the insufflation pressures generally below 12 mm Hg. High intraabdominal pressure (prolonged periods above 15 to 20 mm Hg) can interfere with cardiac output and respiration. High pressures can also result in subcutaneous emphysema that can exacerbate hypercarbia and occasionally cause pneumomediastinum and pneumothorax. Life-threatening gas embolism is rare and most often caused by direct insufflation of CO 2 into a vessel by the Veress needle. Subtle collection of gas in the venous system through an open vessel can rarely occur. The use of the esophageal Doppler monitoring for the characteristic changes that can be detected in the right atrium as free intravascular gas builds up is commonly employed at some centers. Detailed information of the management of laparoscopic complications is available. 1 Results In a recent review of over 580 diverse laparoscopic procedures, the current major complication rate was 0.92% and the minor complication rate was 7%. Transfusion was required in less than 1% whereas open laparotomy was necessary in only 2%. 7 Proper patient selection, formal training, the surgeon's experience, and working with others who are involved with laparoscopic surgery appear to be the best determinants for a successful outcome in urologic laparoscopy. 9 Laparoscopic surgery can be more technically demanding and time consuming than open surgical intervention. However, it is the patient who ultimately benefits from this minimally invasive technique with shorter hospital stay, less postoperative pain, a better cosmetic result, and more rapid return to full activities. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. Abdel-Meguid TA, Gomella LG. Complications of laparoscopy: prevention and management. In: Smith's textbook of endourology, St Louis, MO: Quality Medical Publishers, 1996. Capelouto CC, Moore RG, Silverman SG, Kavoussi LR. Retroperitoneoscopy: anatomical rationale for direct retroperitoneal access. J Urol 1994;152:2008. Gaur DD. Laparoscopic operative retroperitoneoscopy: use of a new device. J Urol 1992;148:1137. Gill IS, Grune MT, Munch LC. Access technique for retroperitoneoscopy. J Urol 1996;156:1120. Goldstein DS, Winfield H. Laparoscopic instrumentation. In: Gomella LG, Kozminski M, Winfield H, eds. Laparoscopic urologic surgery. New York: Raven Press, 1994;21. Gomella LG, Albala DM. Laparoscopic urological surgery, 1994. Br J Urol 1994;74(3):267–273. Gomella LG, Meguid TA, Hirsch IH, et al. Laparoscopic urologic surgery outcome assessment. J Laparoendosc Surg. Submitted.

8. Mullet CE, Viale JP, Sagnard PE, et al. Pulmonary CO 2 elimination during surgical procedure using intra- or extraperitoneal CO 2 insufflation. Anesth Analg 1993;76:622. 9. See WA, Cooper CS, Fisher RJ. Predictors of laparoscopic complications after formal training in laparoscopic surgery. JAMA 1993;270:2689. 10. Winfield H, Lund GO. Extraperitoneal laparoscopic surgery: creating a working space. Contemp Urol 1995.

Chapter 123 Laparoscopic Pelvic Lymph Node Dissection: Transperitoneal and Extraperitoneal Techniques Glenn’s Urologic Surgery

Chapter 123 Laparoscopic Pelvic Lymph Node Dissection: Transperitoneal and Extraperitoneal Techniques
Blake D. Hamilton and Howard N. Winfield

B. D. Hamilton: Division of Urology, University of Utah, Salt Lake City, Utah 84132. H. N. Winfield: Department of Urology, VA Palo Alto Health Care System, Palto Alto, California 94304-1290, and Department of Urology, Stanford University School of Medicine, Stanford University Medical Center, Stanford, California 94305-5118.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Patient Preparation Operative Technique Postoperative Care Outcomes Complications Results Chapter References

Laparoscopic surgery has made tremendous strides in many surgical fields during this decade. In this time period, the laparoscopic pelvic lymph node dissection (LPLND), first described by Schuessler 7 and then by Winfield, 8 has been the most commonly performed laparoscopic urologic procedure. Although the frequency of the LPLND has decreased somewhat due to improvements in the clinical staging of prostate cancer, it remains one of the standard applications of laparoscopy in our field. In addition, it provides an opportunity for the urologist to master the laparoscopic skills necessary for more advanced procedures. In this chapter, we will discuss the indications for this procedure, the technique of both the transperitoneal and extraperitoneal approaches, as well as the results and complications that have been described.

DIAGNOSIS
Prostate cancer is normally diagnosed by transrectal ultrasound– directed biopsy. Less frequently, the diagnosis is made after transurethral resection of the prostate for presumed benign disease. Other malignancies, such as bladder, penile, and urethral cancer, are typically diagnosed by biopsy of the primary lesion.

INDICATIONS FOR SURGERY
The most common indication for LPLND is the need to assess the pelvic lymph nodes in patients with prostate cancer who have a significant risk of metastatic disease. While not all authors agree on the exact indications, there are some general guidelines that may be used. Assessment of pelvic nodal status is warranted in any patient with substantial risk of positive nodes and in whom the diagnosis of positive nodes will change the treatment plan. Prognostic factors that place prostate cancer patients in this higher risk category include: 1. 2. 3. 4. Stage T2B or T3, where nodes will be positive in as many as 50%, depending on prostate-specific antigen (PSA) and Gleason score. Stage T1A, T1B, or T2A with accompanying Gleason score greater than 6. Patients with PSA greater than 20 ng/ml, especially if Gleason grade greater than 6. Elevated serum enzymatic prostatic acid phosphatase with negative nuclear bone scan (stage D0).

In addition, LPLND may be performed prior to radical perineal prostatectomy to accurately assess nodal status. In these cases, laparoscopic mobilization of the seminal vesicles can accomplish the most difficult part of the perineal dissection. Other urologic malignancies may also be adequately staged with LPLND, including bladder, penile, and urethral carcinomas. Patients with transitional cell cancer of the bladder, with radiologic evidence of pelvic lymphadenopathy that is inaccessible to CT-guided biopsy, may have a minimally invasive procedure (LPLND) to correctly stage their disease. Patients with adenocarcinoma or squamous cell carcinoma of the bladder are not likely to be helped by radical cystectomy if the pelvic nodes are positive. LPLND can simply and correctly assess these nodes. In some instances, an extended LPLND should be performed to sample nodal tissue superior and lateral to the bifurcation of the iliac vessels. Squamous cell carcinoma of the penis generally metastasizes to the superficial and deep inguinal lymph nodes, although additional metastases to the iliac lymph nodes suggest a much worse prognosis. Some authors recommend bilateral pelvic lymphadenectomy as the initial staging procedure for all patients with invasive penile cancer and palpably enlarged inguinal lymph nodes. Positive nodes from a LPLND would then eliminate the need for an inguinal dissection with its attendant morbidity (10% to 40%) and mortality (3%). Isolated urethral cancer is a rare urologic neoplasm that generally presents in advanced stages, more commonly in females. The proximal urethral lymphatics drain to the external and internal iliac lymph nodes, as well as the obturator nodes, which can be sampled with LPLND when indicated prior to radical exenteration.

ALTERNATIVE THERAPY
Open surgical lymphadenectomy is the standard alternate approach to laparoscopic lymphadenectomy. If enlarged nodes are visible on an imaging study such as CT scan, then percutaneous needle biopsy may also be considered.

SURGICAL TECHNIQUE
Patient Preparation Each patient must be fully informed of the risks and benefits of LPLND, as well as the alternatives. This discussion should include the potential injuries, the possibility of conversion to open PLND, and disclosure of the surgical team's experience with LPLND. The preparations are identical to that described in Chapter 122. The patient undergoes routine preoperative evaluation, including a type and screen for blood products. On the day before surgery, he receives a mechanical bowel preparation and fasts overnight. This cleanses the bowel somewhat in case of a bowel injury. More importantly, it also serves to decompress the bowel loops, which improves the laparoscopic exposure. The patient receives a parenteral broad-spectrum antibiotic 1 hour preoperatively and is admitted to the hospital following surgery. Operative Technique Transperitoneal Approach In the supine position, the patient is administered a general anesthetic. A Foley catheter is placed and an orogastric tube is generally inserted. Wide adhesive tape is used across the thighs and chest to secure the patient to the operating table during the procedure. Pneumatic compression boots are placed and activated prior to surgery. The abdomen is prepared and draped as for a lower midline laparotomy. Pneumoperitoneum is established by using a Veress needle or by the open laparoscopy (Hasson) technique, according to the surgeon's preference and the patient's risk factors. These techniques are discussed in Chapter 122. A 10-mm laparoscopic port is placed through the initial incision and a 10-mm, 0 or 30 degree laparoscope is inserted. The abdominal cavity is examined for injury or gross

metastatic disease. Adhesions may need to be taken down prior to placement of the other ports. For most patients, three additional ports are sufficient, placed in a diamond configuration (Fig. 123-1A). The 10-mm suprapubic port is placed 3 to 5 cm above the symphysis pubis and the two 5-mm lateral ports are placed midway between the umbilicus and the anterior superior iliac spine off the lateral border of the rectus abdominis muscle. For more obese patients, five (rather than four) ports are used, resulting in a horseshoe configuration ( Fig. 123-1B). The additional port is used to retract fat-laden loops of bowel or fatty urachal tissue draping from the anterior wall. All ports are secured to the skin with a 2-0 silk suture.

FIG. 123-1. (A) The diamond configuration for trocar placement in transperitoneal laparoscopy. (B) The horseshoe configuration is useful for patients who are obese.

The patient is placed in a 30 degree Trendelenburg position and the table is rolled laterally 15 to 30 degrees so that the operative side is elevated. This allows the bowel to fall by gravity toward the head and the contralateral side, exposing the region of interest. The sigmoid colon and cecum are sometimes adherent to the pelvic side wall and must be freed before the exposure is suitable. This is accomplished by incising along the white line of Toldt and pushing the bowel medially. The key laparoscopic landmarks are the obliterated umbilical artery (OUA), the internal inguinal ring, and the vas deferens. The external iliac vessels are often identified coursing beneath the posterior peritoneum. The peritoneal incision begins high up over the pubis midway between the OUA and the internal inguinal ring and extends cephalad, medial to the external iliac vein ( Fig. 123-2). The vas deferens is divided with electrocautery. (The vas can be clipped but these clips may become dislodged into the operative field.)

FIG. 123-2. The initial peritoneal incision is made between the obliterated umbilical artery and the internal inguinal ring. (Modified from Gomella LG, Kozminski M, Winfield M, eds. Laparoscopic urologic surgery. New York: Raven Press, 1994.)

The lateral aspect of the nodal package is developed by bluntly sweeping the fibroadipose tissue from the medial and anterior surfaces of the external iliac vein, beginning at the pubis and working toward the bifurcation of the iliac vessels. In approximately 40% of patients there is an accessory obturator vein branching from the external iliac vein near the pubis. Frequently, it can be spared and the node package teased out from behind it. However, division of this vein is of little consequence. Remember to avoid the ureter as it crosses over the iliac vessels at the common iliac bifurcation. The medial aspect of the nodal package is developed by gentle medial retraction of the obliterated umbilical artery and blunt dissection of the underlying tissue toward the iliac vessels. The obturator nerve may come into view and can be bluntly cleared of fibrolymphatic tissue. As the medial aspect is freed, the distal nodal tissue can be thinned and lifted upward. Near the pubis there frequently are venous variations that must be worked around. Judicious use of cautery and hemoclips allows the node package to be liberated, while obliterating the transected lymphatics. If not previously seen, the obturator nerve must be identified at this point. The nodal package is grasped at its distal extent and retracted up and cephalad. The dissection is then completed to the level of the iliac vessel bifurcation, where the proximal lymphatic vessels are clipped or cauterized ( Fig. 123-3). Once completely free, the nodal tissue is grasped with laparoscopic Russian or spoon forceps and delivered through a 10-mm port with a gentle twisting motion. The port valve must be manually opened to avoid shredding the tissue.

FIG. 123-3. A completed lymph node dissection.

The obturator fossa is again inspected for residual lymphatic tissue and persistent bleeding, which are addressed accordingly. The table is rolled laterally in the opposite direction and the procedure repeated on the contralateral side. Frozen section evaluation may be useful depending on the intraoperative findings and the indications for surgery. When both sides have been completed, the entire abdominal cavity is again inspected for inadvertent injury and persistent bleeding. The intraabdominal pressure is lowered to 5 mm Hg to identify any bleeding previously tamponaded by the pneumoperitoneum. The ports are removed and the incisions closed according to the surgeon's preference. We use the Carter-Thomason (Inlet Medical, Eden Prairie, MN) port closure device to pass a 2-0 (PDS) suture through the abdominal wall fascia on each side of a 10-mm port opening, quickly securing each closure. The closure is completed with a subcuticular 4-0 (Vicryl) suture and each incision covered with a small transparent, permeable dressing. Extraperitoneal Approach The laparoscopic pelvic lymph node dissection may also be performed without entering the peritoneal cavity. As with most alternative approaches, extraperitoneal LPLND (eLPLND) has advantages and disadvantages when compared with the transperitoneal approach. By staying out of the peritoneal cavity, time-consuming

mobilization of adherent bowel is avoided. The intact peritoneum serves as a natural retractor of the abdominal contents, which may be of significant benefit in the obese patient. This may lessen the amount of extreme Trendelenburg position necessary for exposure. The risk of postoperative intraperitoneal adhesions may be decreased or avoided when there is no direct manipulation of the intraperitoneal contents. Finally, the theoretical risk of spilling malignant cells into the peritoneal space is avoided. On the other hand, there are several disadvantages to this approach. The working space created by extraperitoneal insufflation is generally more confined and tends to have a tattered appearance, caused by the shearing forces created by balloon dilation. This is contrasted with the large, smooth working space created by pneumoperitoneum. Furthermore, the vas deferens is displaced cephalad and the usual laparoscopic landmarks are somewhat distorted, creating a confusing picture for the laparoscopist. Previous lower abdominal, intra- or extraperitoneal surgery in this region may further limit the expansion of the preperitoneal space. These difficulties may be overcome with some experience. Preliminary studies indicate that there is increased CO 2 absorption during extraperitoneal LPLND. While this can generally be managed with increased respiratory rate and tidal volume by the anesthesiologist, it may become a more serious problem in those patients with poor cardiopulmonary reserves. Compromised cardiopulmonary function should serve as a relative contraindication to the extraperitoneal approach. 2 Pelvic extraperitoneal approach is described in Chapter 122. Briefly, a 3-cm vertical incision is made just inferior to the umbilicus and deepened down to the rectus fascia, which is carefully opened without entering the peritoneum. The preperitoneal space is then bluntly created by sweeping a finger along the undersurface of the rectus muscles down toward the pubis. This allows for the introduction of a balloon dilator as described by Gaur et al. 1 While commercial models are available, we prefer to make our own by securing the finger cot of a transurethral resection (TUR) drape to a 20-Fr red rubber catheter with free silk ties. This device is placed in the preperitoneal space and inflated with 800 to 1000 cm 3 of saline, expanding the space of Retzius and creating sufficient working space for the node dissection. With the extraperitoneal space dilated, the pubis and external iliac vessels come clearly into view. A Hasson-type cannula can now be secured within the periumbilical incision and the workspace insufflated with CO 2 to a pressure of 12 to 15 mm Hg. Additional ports are placed in an extraperitoneal position under laparoscopic guidance in a configuration as described for the transperitoneal LPLND. The dissection proceeds in a similar fashion, but without the need for a peritoneotomy. If the peritoneal membrane should be disrupted, the extraperitoneal space will likely become obliterated due to CO 2 filling the peritoneal cavity and compressing the extraperitoneal space. The procedure will most likely be converted to a standard transperitoneal LPLND. Postoperative Care All patients are admitted to the hospital for a brief (less than 24 hours) observation and receive two more doses of prophylactic antibiotics. The orogastric tube is removed prior to extubation, and the diet is advanced as tolerated. The Foley catheter is removed once the patient is alert and oriented. Routine postoperative nursing care is important for detecting any complications (e.g., delayed bleeding). The patients are generally discharged within 24 hours and return to normal activity within a week.

OUTCOMES
Complications Complications of LPLND include all of the problems encountered in general laparoscopic surgery, including vascular and visceral mishaps. In addition, there are certain problems that are more specific to this procedure. In working near the bladder, there is increased risk of injury if the dissection is carried medial to the OUA. The ureter is at risk for injury (especially a thermal injury) while dividing the lymphatic branches near the bifurcation of the iliac vessels. Troublesome bleeding may occur from the accessory obturator veins, the obturator vessels (if dissecting below the obturator nerve), or the iliac vein itself. Obviously, a significant laceration here may require emergent laparotomy. Postoperatively, there is the expected risk of lymphocele formation, although this does not seem to be higher than in open PLND. In the largest reported series, the rate of conversion to open PLND ranges from 3% to 5%, with factors such as obesity and history of previous surgery or radiation contributing to the need for laparotomy. In a large multicenter study of 372 patients undergoing LPLND, 4 there were 55 complications (15%). Laparotomy was required in 13 of these patients, 7 at the initial operation and 6 at a later date. All authors agree that there is a steep learning curve and the majority of problems and conversions occur early in each surgeon's experience. With increased skill in laparoscopic suturing techniques, some repairs (e.g., closure of a small enterotomy) may be undertaken without laparotomy. Results There are now several large series published that demonstrate that LPLND is a safe and effective procedure. 3,5,6 Winfield and associates have reported on LPLND in over 200 patients, most with the transperitoneal approach. They reported a mean laparoscopic retrieval of 4.6 and 4.5 nodes from the right and left sides, respectively. Of the first 26 patients who underwent immediate open exploration following LPLND, only 1 patient had a microscopic focus of metastatic prostate cancer near the bifurcation of the iliac vessels that was missed during the laparoscopic dissection, for a false-negative rate of 4%. Overall, 18% of their patients had metastases, but this number may increase as many patients with low risk for metastasis are now staged without pelvic node dissection. They also reported a mean operating room time of 163 minutes and inpatient stay of 2.0 days, although we currently perform the procedure routinely in approximately 90 minutes and patients are generally discharged within 24 hours. Blood loss and postoperative analgesic requirements are minimal. Laparoscopic pelvic lymph node dissection has become a standard procedure for many urologists. It requires a modest level of laparoscopic skill and has a fairly brief, though steep, learning curve. It has been demonstrated to be a safe and effective procedure in the hands of many urologists at many institutions. While the indications for LPLND have declined somewhat, we believe that this procedure will continue to play a significant role in the staging of urologic malignancies. CHAPTER REFERENCES
1. Gaur DD, Agarwal DK, Purohit KC. Retroperitoneal laparoscopic nephrectomy: initial case report. J Urol 1993;149:103. 2. Glascock JM, Winfield HN, Lund GO, Donovan JF, Ping STS, Griffiths DL. Carbon dioxide homeostasis during transperitoneal or extraperitoneal laparoscopic pelvic lymphadenectomy: a real-time intraoperative comparison. J Endourol 1996;10:319. 3. Glascock JM, Winfield HN. Pelvic lymphadenectomy: intra- and extra-peritoneal access. In: Smith AD, et al., eds. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996;870. 4. Kavoussi LR, Sosa E, Chandhoke PS, et al. Complications of laparoscopic pelvic lymph node dissection. J Urol 1993;149:322. 5. Kerbl K, Clayman RV, Petros JA, Chandhoke PS, Gill IS. Staging pelvic lymphadenectomy for prostate cancer: a comparison of laparoscopic and open techniques. J Urol 1993;150:396. 6. Rukstalis DB, Gerber GS, Vogelzang NJ, Haraf DJ, Straus FH, Chodak GW. Laparoscopic pelvic lymph node dissection: a review of 103 consecutive cases. J Urol 1994;151:670. 7. Schuessler WW, Vancaillie TG, Reich H, Griffith DP. Transperitoneal endosurgical lymphadenectomy in patients with localized prostate cancer. J Urol 1991;145:988. 8. Winfield HN, Donovan JF, See WA, et al. Laparoscopic pelvic lymph node dissection for genitourinary malignancies: indications, techniques, and results. J Endourol 1992;6:103.

Chapter 124 Laparoscopic Varix Ligation Glenn’s Urologic Surgery

Chapter 124 Laparoscopic Varix Ligation
James F. Donovan, Jr., and Eduardo Sanchez de Badajoz

J. F. Donovan, Jr.: Department of Urology, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73104. E. Sanchez de Badajoz: Department of Urology, Universidad de Malaga, Malaga, Spain.

Diagnosis Indications Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

The laparoscopic varix ligation was first reported in 1988 using a single-trocar technique. equipment, including video technology and instruments.

14

The procedure has evolved with the recent advances in laparoscopic

DIAGNOSIS
The physical exam is the primary diagnostic tool, with smaller varicoceles often becoming apparent with the Valsalva maneuver. Other diagnostic modalities include thermography, sonography, Doppler and color Doppler sonography, and venography, which are more sensitive, yet may yield more false positives, and are therefore not widely used.

INDICATIONS
Whether by laparoscopy or traditional approach, varix ligation is indicated in patients who demonstrate a palpable varix and at least one of the following: 1. Male subfertility, as demonstrated by abnormal seminal fluid analysis (SFA) and failure to impregnate a partner who is free of demonstrable female factor infertility 2. Adolescent testicular growth retardation that persists for 6 months, and 3. Pain that is not attributable to other intrascrotal pathology We advocate neither ligation of subclinical varices nor prophylactic varicocelectomy in the adolescent.
12

Practical considerations to the use of laparoscopy in varix ligation exclude patients who have undergone a retroperitoneal varix ablation. Since both the Palomo and the laparoscopic approaches isolate the spermatic vessels in the retroperitoneum immediately above the internal ring, laparoscopic mobilization of the spermatic vascular bundle would be compromised in those patients who demonstrate a persistent or recurrent varix following open retroperitoneal exposure. We feel that these patients are best treated by an alternative procedure such as transvenous ablation. 12 We have, however, successfully performed salvage varicocelectomy in patients who have failed transvenous or inguinal varix ablation. Other conditions may present relative contraindications, i.e., obesity, previous abdominal surgery, and umbilical hernia. In these situations, with experience and appropriate equipment selection (i.e., Hasson cannula in place of Veress needle in patients with previous abdominal surgery), one may proceed with laparoscopic surgery using appropriate care. We have safely performed laparoscopic varix ligation in patients with relative contraindications to laparoscopy (obesity, previous abdominal surgery, etc.). 9,10 We now routinely use the Hasson cannula during laparoscopic varix ligation.

ALTERNATIVE THERAPY
The laparoscopic approach for varix ligation offers an alternative to the Palomo (retroperitoneal), Ivanissevitch (inguinal), and Marmar-Goldstein (subinguinal) surgical procedures. 4,6,9,13 Non-surgical options include transvenous embolization with coils or balloons or sclerosis using a variety of sclerosing agents. 12

SURGICAL TECHNIQUE
Laparoscopy is typically performed under general anesthesia. Matsuda et al. have reported the use of local anesthesia during laparoscopic varix ligation when no attempt was made to preserve the testicular artery. 10 However, we have found that preservation of the spermatic artery is technically demanding and prolongs the operative time required to achieve complete venous ablation without damage to the artery. The use of local or regional anesthetics in our procedure would be associated with increased patient discomfort; thus we perform laparoscopic varix ligation only on patients receiving general anesthesia. Following induction of general anesthesia, a nasogastric tube is placed and the abdomen prepped widely to include the external genitalia. To avoid contamination of the surgical site while maintaining access to the scrotum and genitals, the bladder is drained by straight catheter after the patient is draped. Manual retraction of the testes during laparoscopy assists in identifying the spermatic vessels and collateral veins that traverse the internal ring, thereby defining the necessary extent of venous ablation. A thorough discussion of pneumoperitoneum and trocar placement is presented in Chapter 122. However, choice of technique used to achieve pneumoperitoneum and selection of trocar insertion sites in varix ligation deserve brief mention. In our earlier protocols for laparoscopic varix ligation, we used the Veress needle to insufflate the peritoneal cavity with CO 2. Although no complications arose due to the use of the Veress needle, we sought to diminish the likelihood of inadvertent injury to intraabdominal organs during establishment of the pneumoperitoneum. Therefore, we converted to the Hasson cannula in lieu of the Veress needle. Although insertion of the Hasson cannula requires careful dissection, which is somewhat more time consuming when compared to Veress needle insertion, insufflation through the 10-mm Hasson trocar, at a rate of 6 to 8 L/min, proceeds more quickly than through the Veress needle, which offers a maximum flow rate of 1.5 L/min. Comparison of time lapse from skin incision to laparoscope insertion for either Hasson cannula open laparoscopy or Veress needle showed no statistically significant difference. The laparoscope is inserted through the subumbilical port and the peritoneal contents inspected. Optimal sites for insertion of the additional operative ports are then determined. The site of varix ligation is at a point 3 to 5 cm cephalad to the internal ring. This site can be identified on the body surface by transilluminating the abdominal wall at the internal ring from within the peritoneal cavity. Once located, this site serves as a point of reference for port placement. Instrument port sites are selected in each lower quadrant so that instruments will easily reach the internal ring. Placement of sheaths too far from the internal ring and the site of dissection results in the instrument fulcrum (the point of abdominal wall traverse) being far from the instrument tip, exaggerating movements and making delicate dissection more difficult. Placement of sheaths too close to the proposed site of venous ligation results in problems with proper operation of scissors or the curved dissector when the jaws of these instruments are restricted in the distal lumen of the trocar sheath. The optimal position of the port is at a level just caudal to the umbilicus and lateral to the rectus muscle ( Fig. 124-1). These sites are more cephalad than the sites we originally recommended (McBurney's point). 2,3 Due to the advent of 5-mm reusable clip appliers (Ethicon-Germany) and experience with intracorporeal suture ligation, we use 5-mm ports at each of the two instrument insertion sites.

Figure 124-1. Appropriate position of trocars for varix ligation. The array of trocars is used in either left or bilateral varix ligation. The subumbilical trocar is either a Hasson cannula for open laparoscopy or a standard trocar inserted following Veress needle insufflation. The operating trocars are 5.5 mm when a 5-mm clip applier is available or 10 mm when one must rely on a 10.5-mm hemoclip applier.

The principle instruments that we use during laparoscopic varix ligation are curved scissors and a curved dissector. We routinely employ a 5-mm laparoscopic Doppler probe (Meadox, Medsonics) to assist in identification and isolation of the spermatic artery or arteries. When the bulk of large or multiple veins exceeds the capacity of the 5-mm hemoclip, alternative measures are needed to achieve ligation. Rather than separating the venous branches into several small bundles, we use a silk suture to ligate those vessels too large to be accommodated by the 5-mm clip. Typically, an instrument tie is accomplished using two curved dissectors; however, one may choose to utilize 5-mm reusable instruments specifically designed for laparoscopic instrument tie (Storz, Ethicon). After the laparoscope and instrument trocars are in place, attention is directed to the identification and dissection of the spermatic veins. The patient is placed in the Trendelenburg position. If the sigmoid colon is fixed over the spermatic veins cephalad to the internal ring (the proposed site of surgery), the colon is mobilized to expose the underlying spermatic vessels. A 3- to 5-cm incision through the peritoneum is made parallel and lateral to the spermatic vascular bundle with caudal limit 3 to 5 cm above the internal ring in order to minimize scrotal insufflation ( Fig. 124-2 and Fig. 124-3). The medial flap of the peritoneum is grasped and the underlying spermatic vessels bluntly swept from the underside of this flap. From the midpoint of this first incision, we make a second perpendicular incision through the medial peritoneal flap overlying the spermatic vessels to the lateral aspect of the iliac artery ( Fig. 124-4). This resulting T incision provides ample exposure for access to the spermatic veins at a point just cephalad to the internal ring. External traction on the testis, while the surgeon is observing vascular structures at the internal ring, helps to identify all veins that may contribute to the varicocele and thus require ligation or, in the case of small vessels, electrocoagulation.

FIG. 124-2. (A) A 3- to 5-cm incision through the peritoneum is made lateral to the spermatic vascular bundle and extends cephalad from a point 3 to 5 cm above the internal ring. (B) When a coincident laparoscopic inguinal hernia repair is planned, the peritoneal incision extends anterior and lateral to the internal ring and continues anterior and medial to reach the lateral edge of the rectus muscle.

FIG. 124-3. Incision through the peritoneum lateral to the spermatic vascular bundle. Traction on the testis will define collateral vessels traversing the internal ring that should be ablated. The extent of planned varix ligation should be determined prior to performing the initial peritoneal incision.

FIG. 124-4. The medial peritoneal flap is lifted and the spermatic vessels swept from the underlying surface. A medial incision bisects the peritoneal flap and provides exposure of the underlying spermatic vascular bundle.

We free the entire spermatic vascular bundle from the underlying psoas muscle ( Fig. 124-5). While primarily designed to facilitate thorough dissection of spermatic vessels, this maneuver makes possible the expeditious application of hemoclips or ligature around the entire spermatic vascular bundle in the event of significant bleeding. Loose adventitial tissue is stripped from the spermatic vessels. The spermatic vascular bundle is then bluntly separated into medial and lateral bundles. Typically, there are three to eight spermatic veins. In contrast, the spermatic artery is usually single and located posterior and medial to the veins. We attempt to identify the spermatic artery by inspection for pulsation in either the medial or lateral bundle. Extreme care must be taken to differentiate between true spermatic

arterial pulsation and transmitted pulsation from the iliac artery. When used to assist in identification of the spermatic artery, the laparoscopic probe significantly reduces operative time.7 After the location of the spermatic artery is confirmed, the vascular bundle not containing the artery is divided between ligatures ( Fig. 124-6). The process of arterial identification and preservation, venous ligation, and division is repeated serially with the remaining vascular bundle. When a single vein is isolated or when several veins are of appropriate dimension, the 5-mm hemoclip applier speeds ligation prior to division with curved scissors. The process is repeated until only the spermatic artery remains (Fig. 124-7).

FIG. 124-5. The entire spermatic vascular bundle is mobilized from the underlying psoas muscle. A combination of sharp and blunt dissection facilitates separation of the spermatic vessels from surrounding tissues. One should avoid deep dissection in order to spare the underlying genitofemoral nerve crossing anterior to the psoas muscle.

FIG. 124-6. (A) The vascular bundle that does not contain spermatic artery is ligated, then divided. For small venous aggregates or single spermatic veins, the 5-mm reusable clip applier is utilized. Otherwise, we prefer suture ligature. A 5- to 10-cm ligature is passed into the abdominal cavity. One end is passed from right to left under the vessels to be ligated extending only 1 cm beyond the vessels. The other end is long to facilitate instrument tie. (B) The long end is grasped by the left-hand instrument allowing sufficient slack to permit easy wrapping around the tip of the right-hand instrument. The instrument tie can be accomplished with two curved dissectors with tips rotated up. Alternatively, instruments specifically designed for laparoscopic suture and ligation may be used. (C) The nonarterial vascular bundle is ligated. (D) The spermatic veins are divided between ligatures.

FIG. 124-7. Upon completion of the procedure, all spermatic veins have been ligated and divided by either clip or suture. Only the spermatic artery(ies) remains and arterial patency is documented by Doppler.

On occasion, a patient who presents for varix ligation may have a coincident inguinal hernia. The inguinal hernia may be repaired using a laparoscopic approach. Our approach to varix ligation in combination with laparoscopic hernia repair requires an extension of the lateral peritoneal incision anterolateral and then medial to the internal ring, providing exposure to the abdominal wall circumscribed by Hasselbach's triangle. We feel that the presence of a clinically palpable hernia warrants repair and we recommend repair, especially in the young male. Other laparoscopic surgeries that could be performed in combination with elective varix ligation are difficult to envision (e.g., incidental laparoscopic appendectomy is unwarranted). After completion of varix ligation, perform an orderly and systematic departure from the abdomen as described in Chapter 122. The peritoneal cavity is passively desufflated and the subumbilical fascia then closed with 0 PDS suture. When the Hasson trocar has been used, it is released from the stay sutures. These sutures are then tied across the wound to approximate the fascia. Frequently, impressive scrotal emphysema occurs due to extravasation of CO 2 through the internal ring along the spermatic vessels. This scrotal gas can be expressed back into the peritoneal cavity and vented through an open trocar site. Some gas inevitably remains in the subcutaneous tissues where it is later resorbed without consequence.

OUTCOMES
Complications Due to the age group undergoing varix ligation, the risk of complications is minimal. In a recent series, the average operating room time was 43 minutes, the hospital stay was 12 hours, and there were no complications noted in the transperitoneal varix ligation group. 5 Results The use of laparoscopy in performing varix ligation remains controversial. 1,7 The procedure may be performed safely and a prospective randomized study in progress will address the possible benefits of the laparoscopic approach in comparison to the inguinal and subinguinal techniques. CHAPTER REFERENCES
1. Donovan JF Jr. Laparoscopic varix ligation editorial. Urology 1994;44(4):467–469. 2. Donovan JF, Winfield HN. Laparoscopic varix ligation. J Urol 1992;147:77.

3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Donovan JF Jr. Laparoscopic varix ligation with the Nd:YAG laser. J Endourol 1992;6:165. Goldstein M, Gilbert BR, Dicker AP, Dwosh J, Gnecco C. Microsurgical inguinal varicocelectomy with delivery of the testis: an artery and lymphatic sparing technique. J Urol 1992;148:1808. Gomella LG, Abdel-Meguid TA, Hirsch IH, et al. Laparoscopic urologic surgery outcome assessment. J Laparoendosc Surg. 1997;7:77–86. Ivanissevich O. Left varicocele due to reflux: experience with 4,470 operative cases in forty-two years. J Int Coll Surgeons 1960;34(6):742. Jarrow JP. Varicocele repair: low ligation (editorial). Urology 1994;44:470–472. Loughlin KR, Brooks DC. The use of a Doppler probe to facilitate laparoscopic varicocele ligation. Surg Gynecol Obstet 1992;174:326. Marmar JL, DeBenedictis TJ, Praiss D. The management of varicoceles by microdissection of the spermatic cord at the external inguinal ring. Fertil Steril 1985;43(4):583. Matsuda T, Horii Y, Takeuchi H, Yoshida O. Laparoscopic varicocelectomy. J Urol 1991;145:325A. McClure RD, Khoo D, Jarvi K, Hricak H. Subclinical varicocele: the effectiveness of varicocelectomy. J Urol 1991;145: 789. Murray RR Jr, Mitchell SE, Kadir S, et al. Comparison of recurrent varicocele anatomy following surgery and percutaneous balloon occlusion. J Urol 1986;135:286. Palomo A. Radical cure of varicocele by a new technique: preliminary report. J Urol 1994;61:604. Sanchez de Badajoz E, Diaz Ramirez F, Marin-Martin J. Tratamiento endoscopico del varicocele. Arch Esp Urol 1988;41:15.

Chapter 125 Transperitoneal Laparoscopic Nephrectomy and Nephroureterectomy Glenn’s Urologic Surgery

Chapter 125 Transperitoneal Laparoscopic Nephrectomy and Nephroureterectomy
Inderbir S. Gill and Sakti P. Das

I. S. Gill: Section of Laparoscopy and Minimally Invasive Surgery, Department of Urology, The Cleveland Clinic Foundation, Cleveland, Ohio 44195. S. P. Das: Department of Urology, Kaiser Permanente Medical Center, Walnut Creek, California 94596.

Diagnosis Transperitoneal Nephrectomy Indications Alternative Therapy Surgical Technique Transperitoneal Nephroureterectomy Indications Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Laparoscopic nephrectomy has become an increasingly feasible alternative to open nephrectomy for benign renal disease since the pioneering report by Clayman and associates 1 in 1991. Building on this concept, laparoscopic radical nephrectomy for renal cancer and lapa-roscopic nephroureterectomy have recently been performed at some centers. Laparoscopic nephrectomy can be performed by either the transperitoneal or retroperitoneal technique. This chapter describes the indications, technique, postoperative management, results, and complications of transperitoneal laparoscopic simple nephrectomy and nephroureterectomy. The extraperitoneal approach is discussed in Chapter 128.

DIAGNOSIS
Standard imaging studies such as computed tomography (CT) scan, magnetic resonance imaging (MRI), ultrasound, retrograde pyelography and excretory urography, and nuclear scans are used in the diagnosis of conditions such as renal cell carcinoma or nonfunctioning kidney.

TRANSPERITONEAL NEPHRECTOMY
Indications Laparoscopic simple nephrectomy has been performed for virtually all benign diseases that can result in a symptomatic poorly functioning or nonfunctioning kidney. 2,6 The one relative contraindication may be xanthogranulomatous pyelonephritis, which results in an intense perirenal inflammatory fibrotic reaction. In one series of 60 patients undergoing laparoscopic nephrectomy, two of three patients with xanthogranulomatous pyelonephritis had to be converted to open surgery for this reason. 3 Indications for laparoscopic radical nephrectomy include T1 and T2 (less than 8 cm diameter) renal carcinomas. Alternative Therapy Open transabdominal or retroperitoneal nephrectomy is the main alternative approach to laparoscopic nephrectomy. The kidney can also be approached using extraperitoneal laparoscopy. Surgical Technique Standard preoperative preparations are noted in Chapter 122. If a patient is unfit for open nephrectomy, he is not likely to be a candidate for laparoscopic nephrectomy either. A skilled anesthesiologist, familiar with prolonged laparoscopic procedures, is a vital member of the team. We recommend a limited outpatient mechanical bowel preparation: one bottle of magnesium citrate at 12 noon on the day before surgery with clear liquids permitted until midnight. A Foley catheter and nasogastric tube are placed and the patient turned in a lateral decubitus position, as for open nephrectomy. In order to minimize postoperative neuromuscular complications and paresthesias, it is important to protect all bony prominences with soft padding and place extremities in a neutral position. Transperitoneal laparoscopic nephrectomy involves the following nine steps: (a) peritoneal insufflation in the lateral position; (b) placement of five ports; (c) colonic mobilization; (d) ureteral dissection; (e) lateral retraction of the superior and inferior renal poles; (f) renal hilar control; (g) completion of nephrectomy; (h) organ entrapment and removal; and (i) exit. Peritoneal Insufflation The patient is placed in the flank position. 4 Peritoneal insufflation is performed by inserting a Veress needle in the mid-clavicular line at the level of the umbilicus. CO is insufflated to maintain a pneumoperitoneum of 15 mm Hg pressure. Port Placement A 12 mm-port (primary port, PP) is placed at the level of the umbilicus in the midclavicular line. Four secondary ports are placed under laparoscopic visualization (see Fig. 125-1 for a detailed description). After the two midclavicular line ports are inserted, the colon is mobilized laparoscopically, and the kidney, enclosed in the perirenal fat, is visualized. It is at this juncture that the two anterior axillary line ports are inserted, one each in the area of the upper and lower renal pole, respectively. For the majority of the laparoscopic procedure, the video laparoscope is positioned in the primary port, the surgeon works through the 12-mm upper midclavicular line (UMCL) and the 12-mm lower midclavicular line (LMCL) ports, and the assistant works through the 5-mm upper anterior axillary line (UAAL) and 5-mm lower anterior axillary line (LAAL) ports.
2

FIG. 125-1. Port placement for transperitoneal laparoscopic nephrectomy. Following lateral insufflation, a 12-mm port is placed in the midclavicular line at the level of the umbilicus. Under laparoscopic visualization, two additional ports are now placed in the midclavicular line: an upper port (12 mm) in a subcostal location (UMCL port) and a lower port (5 mm) approximately 3 cm caudad to the umbilicus (LMCL port). The laparoscopic procedure is now initiated and the line of Toldt is reflected in order to mobilize the color and bring the kidney into view. At this time two ports are placed in the anterior axillary line: an upper port (5 mm) at the tip of the 12th rib (UAAL port) line, and a lower (5 mm) port at the level of the umbilicus (LAAL port). In general, the surgeon works through the two MCL ports and the assistant utilizes the two AAL ports.

Mobilization of the Colon The colon is mobilized by incising the ipsilateral line of Toldt. During a right laparoscopic nephrectomy, the peritoneal incision extends from the right common iliac artery, coursing lateral to the cecum and ascending colon and around the hepatic flexure. At its cephalad end, the incision is carried medially in a horizontal manner between the liver and transverse colon, thus exposing the hepatorenal space of Morrison. The ascending colon and hepatic flexure are bluntly rolled medially. This brings the medially located second part of the duodenum into view. The duodenum is dissected medially until the anterior surface of the inferior vena cava is clearly visualized (endpoint of dissection). Note that the duodenum and inferior vena cava may look disconcertingly similar through the laparoscope. In case of uncertainty, lowering the pneumoperitoneum to 5 mm Hg allows the inferior vena cava to distend, thus facilitating its identification. During a left laparoscopic nephrectomy, the line of Toldt is incised from the left common iliac artery up to the splenic flexure where the lienorenal and phrenicocolic ligaments are divided. This detaches the splenic flexure from the spleen and tail of the pancreas. The left colon and splenic flexure are bluntly mobilized medially until the anterior surface of the aorta is visualized (endpoint of dissection). Dissection of the Ureter The ureter is identified in the retroperitoneal fat just medial to the psoas muscle. In general, the ureter is located somewhat more medially than one would anticipate. In case of difficulty in locating the ureter, the following maneuvers may be helpful: (a) identify the gonadal vessels that course anterior and parallel to the midureter; (b) gently stroke the retroperitoneal fat in a horizontal manner (at right angles to the longitudinal axis of the ureter) with an atraumatic grasper, and look for ureteral peristalsis; (c) look for the ureter where it crosses the common iliac vessels. The ureter is mobilized and retracted laterally. To free up the lower AAL port, which would otherwise be engaged with housing locking traumatic forceps for lateral retraction of the ureter, one can employ a percutaneously placed suture to loop the ureter. 8 A 0 silk suture is inserted through a percutaneously placed 14-gauge angiocathether, laparoscopically looped around the ureter, and retrieved through the angiocatheter. Traction on this suture retracts the ureter laterally. Alternatively, the ureter can be clipped and divided at this point. Lateral traction of the intact or divided ureter facilitates subsequent renal hilar dissection. Renal Polar Dissection and Retraction Gerota's fascia is entered and the lower renal pole is identified and mobilized circumferentially. Similarly, the upper renal pole is detached from the undersurface of the adrenal gland and mobilized. The adrenal gland is carefully preserved. Hemostasis along its inferior surface must be confirmed because this can be a source for troublesome postoperative bleeding. The assistant retracts the upper and lower renal poles laterally by an atraumatic grasper, one each inserted through the UAAL and LAAL ports. Such lateral traction places the renal hilum on gentle stretch, facilitating its dissection. Renal Hilar Control Hilar dissection aims to circumferentially mobilize the main renal artery and vein individually, creating a 360 degree window around each vessel. The anteriorly located renal vein is usually identified and skeletonized first. The more posteriorly located renal artery, often coursing along the posterosuperior border of the renal vein is mobilized subsequently. A hook electrode can facilitate hilar dissection. While skeletonizing the main renal vessels, one must be aware of the adrenal, lumbar, or gonadal venous branches in this area. A 10-mm multifire endoclip applier is introduced through the 12-mm UMCL or LMCL port to occlude the renal artery. Five 9-mm titanium clips are applied and the artery is divided leaving three clips on the side of the aorta ( Fig. 125-2). The renal vein is usually too wide for the 9-mm clips, and a 12-mm endo-GIA vascular stapler is used to simultaneously occlude and divide the renal vein. En masse ligation of the entire renal hilum with the endo-GIA stapler may result in an arteriovenous fistula and should be avoided.

FIG. 125-2. Renal hilar control. The renal artery and vein are individually mobilized and securely occluded (three clips on proximal stump of each blood vessel). Lateral traction on the upper and lower renal poles and the ureter places the renal vessels on gentle stretch, facilitating their dissection.

Completion of Nephrectomy Dissection Nephrectomy is now completed. Minor adhesions over the lateral and posterior aspects of the kidney are divided. If intact, the ureter is double clipped and transected. Using locking traumatic forceps through the UAAL port, the ureter is securely grasped adjacent to the renal pelvis, and the kidney is temporarily “parked” on the liver or spleen. Entrapment and Removal of Specimen Intraabdominal entrapment of the excised specimen is performed. For intact removal, we routinely employ the Endocatch device (Autosuture Company, USSC, Norwalk, CT). This device consists of a polyurethane pouch attached to a self-opening, spring-loaded, flexible metal ring ( Fig. 125-3). It is available in two sizes: (a) the 10-mm Endocatch with a 2.5-in. diameter metal ring and a 6-in.-deep pouch, and (b) the 15-mm Endocatch II with a 5-in. diameter metal ring and a 9-in.-deep pouch. The Endocatch is inserted through the 12-mm LMCL port. If the 15-mm Endocatch II device is used, the 12-mm UMCL port is removed and the 15-mm shaft of the Endocatch II is inserted under laparoscopic visualization. This may require enlargement of the skin and fascial incision followed by finger dilation of the tract. The Endocatch has several advantages. It is easy to deploy, and the mouth of the sack opens automatically without the aid of any additional instruments. It is easily maneuverable and allows very rapid entrapment of the excised specimen. However, the pliable polyurethane pouch of the Endocatch is not strong enough to withstand the stresses of intraabdominal morcellation. For intact removal, the LAAL port site skin incision is extended as necessary in order to remove the specimen. The skin incision can be safely made over the prongs of a Kelly clamp inserted through the LAAL port site into the abdomen, alongside the sac. Under laparoscopic guidance, the prongs of the Kelly clamp are used to tent up the anterior abdominal wall, thereby protecting the underlying viscera. The skin incision can now be safely enlarged, and the entrapped kidney is delivered intact.

FIG. 125-3. Endocatch device. The polyurethane pouch is attached to the flexible, self-opening metal ring. The Endocatch is available in two sizes (see text for details).

If morcellation is desired, the sturdier, impermeable LapSac (Cook Urology, Spencer, IN) is employed. This plastic and nylon reinforced sac is also available in two sizes: 5 × 8 in. or 8 × 10 in. After first being mounted on a metal introducer in a clockwise manner, the LapSac is inserted into the abdomen through the 12-mm UMCL port (Fig. 125-4). The sac is pulled into the peritoneal cavity with atraumatic forceps and disengaged from the introducer by twisting in a counterclockwise direction. The laparoscope is now removed from the primary port and reinserted through the 12-mm UMCL port. The mouth of the sac is triangulated open by three grasping forceps, one each passed through the LAAL, LMCL, and primary ports ( Fig. 125-5). The tip of the laparoscope is placed within the sac and rotated in ever-widening circles with the aim of opening up the sac completely. The kidney is then maneuvered into the sac ( Fig. 125-5). The grasper located in the primary port now delivers the drawstrings of the LapSac to the grasper located in the LAAL port, and the mouth of the sac is tightened. Pneumoperitoneum is lowered to 5 mm Hg and hemo-stasis is confirmed. Once again, the adrenal bed and the under surface of the liver or spleen must be inspected thoroughly.

FIG. 125-4. (A) LapSac being delivered into the abdomen by laparoscopic grasping forceps. The LapSac is optimally inserted into the abdomen through the 12-mm UMCL port. (B) Note: The excised kidney, secured by a locking traumatic forceps grasping the proximal ureter, is parked on the liver, until the LapSac has been opened.

FIG. 125-5. The mouth of the LapSac is triangulated open by three graspers. The kidney is now inserted in the sac.

The LAAL port is removed and the drawstrings and the mouth of the sack are delivered outside the abdomen. With strong traction, the redundant portion of the sack is retrieved through the port site incision such that the entrapped kidney is held taut against the undersurface of the anterior abdominal wall. Intraabdominal morcellation is performed under careful and continuous laparoscopic monitoring to avoid inadvertent bowel injury using a morcellator or ring forceps working inside the impermeable LapSac. Exit Ports are removed sequentially under laparoscopic view after reconfirming hemostasis. Any residual pneumoperitoneum is vented out. Fascial and skin closure are completed in standard fashion.

TRANSPERITONEAL NEPHROURETERECTOMY
Indications Laparoscopic nephroureterectomy is primarily indicated for symptomatic end-stage reflux nephropathy or for transitional cell carcinoma of the upper urinary tract. Alternative Therapy Open or laparascopic extraperitoneal nephroureterectomy can be employed. Surgical Technique With the patient in the lithotomy position, rigid cystoscopy is performed. A superstiff 0.035-in. guidewire is inserted into the targeted ureter. A 5-mm ureteral balloon dilator is threaded over the guidewire. The balloon is distended in the intravesical part of the ureter. Using an Orandi knife, the entire extent of the submucosal ureteral tunnel is unroofed. Alternatively, if ablation of the adjacent bladder cuff is desired along with the nephroureterectomy, the submucosal ureter and adjacent bladder wall can be resected using a standard resectoscope. The balloon dilator is removed, a Foley catheter inserted per urethra, and the ureteral guidewire secured to the Foley catheter. The patient is then placed in the lateral decubitus position, and five ports are placed as outlined in the preceding section for nephrectomy. For performing a
4

laparoscopic nephroureterectomy, an additional sixth port (12 mm) is placed in the midline, midway between the umbilicus and symphysis pubis ( Fig. 125-6). Using electrocautery scissors, the pelvic peritoneum immediately medial to the ipsilateral medial umbilical ligament (obliterated umbilical artery) is incised. This incision is then continued cephalad and laterally along the line of Toldt as described above. To gain access to the pelvic juxtavesical ureter, three structures must be sequentially clip-occluded and divided: (a) vas deferens or round ligament; (b) the medial umbilical ligament near its take-off from the internal iliac artery (Note that at this location the umbilical artery is not occluded and can result in serious hemorrhage. Therefore, the artery must be carefully dissected and double-clipped prior to division.); and (c) the superior vesical vessels.

FIG. 125-6. Port placement for laparoscopic transperitoneal nephroureterectomy is similar to that for laparoscopic transperitoneal nephrectomy, except for an additional suprapubic midline port (12 mm).

The ureter is now freed and dissected distally to the ureterovesical junction. Dissection of the surrounding perivesical fatty tissue allows the entire area of the ureterovesical junction to be mobilized anteriorly. If the ureter and bladder cuff were previously completely resected and cystoscopically detached from the bladder, cephalad traction on the juxtavesical ureter should allow the ureter to be plucked from the bladder. In such cases, formal bladder closure has not been necessary in our experience. Ten to 14 days of Foley catheter drainage usually allows complete healing of the rent in the bladder. Alternately, if the submucosal ureter was merely unroofed, it now needs to be laparoscopically detached from the bladder. An Endo- GIA 30 tissue stapler (U.S. Surgical), inserted through the 12-mm suprapubic port, is positioned across the ureterovesical junction and the adjacent 2-cm bladder cuff. The ureteral guide wire is removed. Firing the stapler simultaneously places three staggered rows of staples on the bladder cuff and detaches the ureter ( Fig. 125-7). Transperitoneal laparo-scopic nephrectomy is now completed as described above, and the renal and ureteral unit is retrieved en bloc.

FIG. 125-7. (A) Cephalad traction, exerted on the distal ureter, places the ureterovesical junction on stretch. (B) The Endo-GIA tissue stapler is now positioned caudad to the ureterovesical junction and fired. (C) This excises a 2-cm cuff of the bladder en bloc with the ureter, while simultaneously closing the bladder with three staggered rows of staples.

The nasogastric tube is removed in the operating room. Ambulation is resumed on the evening of surgery following which the Foley catheter and pneumatic compression stockings are discontinued. If a nephroureterectomy has been performed, the Foley catheter is left for a period of 10 to 14 days. A postoperative cystogram confirms the absence of extravasation before removing the Foley catheter. Oral liquid intake is often resumed on the evening of surgery. Diet is advanced during the first day as tolerated. Usually oral analgesics confer adequate pain control by the first or second postoperative day.

OUTCOMES
Complications In a recent multi-institutional review of 153 patients undergoing laparoscopic simple nephrectomy for benign disease, complications occurred in 19 patients (12%) (Table 125-1).5 There was a significant learning curve associated with this procedure: 67% of the complications occurred in the initial 20 cases at each institution.

TABLE 125-1. Complications of laparoscopic nephrectomy for benign disease in 153 patients

While embarking upon a laparoscopic nephrectomy, the urologist must pay attention to 7 factors in the recognition and avoidance of complications: patient selection, patient positioning, planned access, intraoperative fluid balance, precise vascular dissection, controlled exit from the abdomen, and undue postoperative abdominal pain. Patient selection must take into account the significantly longer operative and anesthesia times associated with laparoscopic nephrectomy, and its increased potential for CO2 and fluid retention. This admonition is most pertinent in patients with compromised cardiopulmonary status. Proper patient positioning, critical in preventing postoperative neurological and musculoskeletal sequela, remains the surgeon's responsibility, and should not be delegated. Careful padding, avoidance of positional stresses, and antiembolic stockings are routine. Details of the access and exit techniques are discussed in the

chapter on general laparoscopy. It is important to remember that a prolonged, tense pneumoperitoneum induces transient oliguria. Accordingly, intraoperative intravenous fluid administration should be aimed at maintaining hemodynamic stability and not at inducing diuresis. This is particularly important in patients with limited cardiopulmonary reserve, since overhydration may precipitate pulmonary edema and cardiac failure. Vascular injury usually requires emergent conversion to open surgery. Control is optimally achieved by a definite, preplanned set of maneuvers. A general laparotomy set must be available routinely. A blunt laparoscopic instrument is placed at the site of vascular injury to tamponade hemorrhage and serve as a guide to the site of injury once the abdomen is opened. The laparoscope is now torqued toward the body wall which is incised directly over the laparoscope. The sheath of the laparoscope safeguards against injury to the abdominal viscera during such emergent entry. Following open laparotomy, vascular control is secured. Postoperative abdominal pain, disproportionate to the size of the skin incisions, should be treated as an acute abdomen. Intraabdominal hemorrhage, bowel perforation, and acute pancreatitis must be considered. 5,7,10 Results The principal author recently evaluated data available in the literature on 172 patients undergoing laparoscopic simple nephrectomy by the transperitoneal approach (Table 125-2). Mean operating time was 258 minutes (range 210 to 355 minutes). Mean estimated blood loss was 293 cm 3 (range 200 to 454 cm 3). Hospital stay averaged 4.9 days (range 3 to 10 days) and convalescence required 1.5 weeks (range less than 1 to 1.8 weeks). Twenty patients (12%) were converted to open surgery.
6

TABLE 125-2. Laparoscopic nephrectomy: current experience

In the initial series of six patients undergoing laparoscopic nephroureterectomy for upper tract transitional cell carcinoma at Washington University, mean operating time was 7.3 hours (range 5.5 to 9.4 hours), estimated blood loss was 180 ml (range 100 to 250 ml), hospital stay averaged 4.6 days (range 4 to 7 days), and convalescence was 5 weeks (range 1 to 11 weeks). Major complications occurred in 16% of patients. 8 The feasibility and benefits of laparoscopic nephrectomy and nephroureterectomy for benign diseases are now established. Except for special contraindications discussed earlier, in the vast majority of instances, it behooves us to offer the choice of laparoscopic nephrectomy to our patients with benign renal pathology. Experience with laparoscopic radical nephrectomy for malignant conditions is growing steadily, albeit in select centers of excellence. 9 The procedure is feasible and the data of pathologic containment are encouraging. However, the technical difficulties of dissecting large tumors with adequate margins of clearance, coupled with proper entrapment and organ retrieval, have limited the application of laparoscopic nephrectomy to selected patients with localized renal cancer. Overall, further experience, advances in instrumentation, and longer follow-up are necessary before laparoscopic radical nephrectomy can advance from the investigational to the established status. Nevertheless, at present we are encouraged by its technical feasibility and remain cautiously optimistic about the potential of laparoscopic radical nephrectomy in selected patients. CHAPTER REFERENCES
Clayman RV, Kavoussi LR, Soper NJ, et al. Laparoscopic nephrectomy: initial case report. J Urol 1991;146:278. Coptcoat MJ, Joyce AJ, Popert R, Eden C, Rassweiler J. Laparoscopic nephrectomy—the King's experience. Min Invas Ther (Suppl 1) 1992;1:67. Eraky L, el-Kappany H, Shamaa MA, Ghoneim MA. Laparoscopic nephrectomy: an established routine procedure. J Endourol 1994;8:275. Gill IBS, Clayman RV. Laparoscopic nephrectomy. Atlas of the Urol Clin North Am 1993;1(2):65–72. Gill IBS, Kavoussi LR, Clayman RV, et al. Complications of laparoscopic nephrectomy in 185 patients: a multi-institutional review. J Urol 1995;154:479–483. Gill IBS, Clayman RV, McDougall EM. State-of-the-art: advances in urologic laparoscopy. J Urol 1995;154:1275. Kerbl K, Clayman RV, McDougall EM, Gill IS, Wilson BS, Chandhoke PS, Transperitoneal nephrectomy for benign disease of the kidney: a comparison of laparoscopic and open surgical techniques. Urology 1994;43:607. 8. Kerbl K, Clayman RV, McDougall EM, Urban DA, Gill IBS, Kavoussi LR. Laparoscopic nephroureterectomy: evaluation of first clinical series. Eur Urol 1993;23:431. 9. Nakada SY, McDougall EM, Clayman RV. Laparoscopic extirpation of renal cell cancer: feasibility, questions, and concerns. Semin Surg Oncol 1996;12:100. 10. Rassweiler JJ, Henkel TO, Stock C, Potempa DM, Greschner M, Aiken P. Transperitoneal and retroperitoneal laparoscopic nephrectomy. Indications and results. J Endourol 1993;7:S175. 1. 2. 3. 4. 5. 6. 7.

Chapter 126 Retroperitoneoscopic Nephrectomy and Nephroureterectomy Glenn’s Urologic Surgery

Chapter 126 Retroperitoneoscopic Nephrectomy and Nephroureterectomy
Inderbir S. Gill and Sakti Das

I. S. Gill: Section of Laparoscopic and Minimally Invasive Surgery, Department of Urology, The Cleveland Clinic Foundation, Cleveland, Ohio 44195. S. Das: Department of Urology, University of California School of Medicine, Walnut Creek, California 94596.

Nephrectomy Indications for Surgery Alternative Therapy Surgical Technique Nephroureterectomy Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

NEPHRECTOMY
Since the kidneys are located in the retroperitoneum, a direct retroperitoneal approach during urologic laparoscopic surgery would appear prudent. Early attempts at retroperitoneoscopic surgery (lumbar sympathectomy, ureterolithotomy, nephrectomy) were hampered by the inadequate working space and suboptimal pneumoretroperitoneum. 8 Due to these limitations, the transperitoneal route has been the preferred approach for the majority of urologic laparoscopic surgery. In 1992, Gaur described an atraumatic balloon dissection technique that has revived interest in retroperitoneoscopy and pelvic extraperitoneoscopy. 2 A current international survey of 24 centers (3988 laparoscopic procedures) revealed that while only 28% of urologic laparoscopic surgery in 1993 was being performed by the retroperitoneal approach, this increased to 51% by 1996. 3 This chapter describes the indications, contraindications, special considerations, technique, and results of retroperitoneoscopic nephrectomy and nephroureterectomy. Since the optimal technique of retroperitoneoscopic surgery is still evolving, this chapter necessarily reflects our personal bias on this topic. For a more comprehensive understanding of the technique of retroperitoneoscopic nephrectomy, the reader is well advised to first peruse the preceding chapter on transperitoneal laparoscopic nephrectomy.

INDICATIONS FOR SURGERY
Retroperitoneoscopic nephrectomy is optimally indicated in adults with end-stage, symptomatic, benign renal pathology, where the kidney is either normal or atrophic in size. For an enlarged kidney, the restricted retroperitoneal space may hamper adequate visualization and dissection during retroperitoneoscopy. For a large hydronephrotic kidney or large renal cyst, we partially aspirate the specimen first under direct camera vision, thereby rendering the size manageable for retroperitoneoscopic extirpation. Retroperitoneoscopic nephrectomy is also ideally suited to the pediatric population. The minimal retroperitoneal fat in the child makes retroperitoneoscopy an expeditious approach to the kidney. We have employed this approach in children with end-stage hydronephrosis, multicystic dysplasia, chronic pyelonephritis, and reflux nephropathy. The general contraindications to retroperitoneoscopic surgery are noted in Chapter 122. In radical nephrectomy for renal cancer, the completely retroperitoneoscopic approach presents significant technical challenges in dissecting the often enlarged kidney and adrenal, including Gerota's fascia, and the subsequent entrapment and extraction of the specimen. In these kidneys, we have used a combined retro- and transperitoneal approach to radical nephrectomy, whereby the renal artery is approached retroperitoneoscopically; the remainder of the dissection and extraction of the specimen is performed by converting to the transperitoneal laparoscopic approach.

ALTERNATIVE THERAPY
Nephrectomy can also be performed by the open surgical technique or the transperitoneal laparoscopic approach.

SURGICAL TECHNIQUE
The essential differences between retroperitoneoscopic and transperitoneal laparoscopic nephrectomy lie in the technique of access, presentation of renal anatomy, and special technical considerations. Accordingly, these three aspects are discussed in detail. Thereafter, important technical aspects of retroperitoneoscopic nephrectomy and nephroureterectomy are highlighted. Access Technique The patient is secured in the flank position with the affected side up, and kidney rest elevated. Both the surgeon and the first assistant/camera person stand facing the patient's back. In contrast, during transperitoneal laparoscopic nephrectomy, as mentioned in Chapter 125, the surgeon faces the patient's abdomen, and the assistant faces the patient's back. We employ the open Hasson technique to obtain retroperitoneoscopic access. 4 The basics of retroperitoneal access are detailed in Chapter 122. The initial horizontal skin incision (1.5 to 2 cm) is made anterior to the tip of the 12th rib for the initial trocar access ( Fig. 126-1). The posterior and anterior thoracolumbar fascia is incised and an index finger is inserted into the retroperitoneal space. Finger dissection is performed in a cephalad direction, attempting to create a space inside Gerota's fascia for placement of the balloon dilator ( Fig. 126-2). In about 40% to 50% of patients, despite the above maneuvers, we are unable to position the balloon dilator inside Gerota's fascia. In this circumstance, finger dissection is performed between the psoas muscle posteriorly and Gerota's fascia anteriorly. Thus the balloon is distended in the pararenal retroperitoneal fat posterior to Gerota's fascia ( Fig. 126-3). We currently employ the Origin (Origin Medsystems, Menlo Park, CA) trocar-mounted balloon distention device for creating a working space in the retroperitoneum, which is discussed in detail in Chapter 122 (Fig. 126-4).

FIG. 126-1. Site of primary trocar placement at the tip of the 12th rib (black arrow). Patient is placed in the flank position (right side up); photograph is taken from the patient's back. H denotes the head end of the patient. 11" and 12" denote the 11th and 12th ribs. Q, quadratus lumborum muscle; I, iliac crest. (Adapted from Gill IS, Grune MT, Munch LC. Access technique for retroperitoneoscopy. J Urol 1996;156:1120.)

FIG. 126-2. Digital dissection of the retroperitoneal space. This dissection is aimed at creating a space inside Gerota's fascia for subsequent placement of balloon dilator. However, such subfascial dissection is possible in only 50% to 60% of cases. In the remainder of the cases, we perform finger dissection posterior to Gerota's fascia. (From Gill IS, Grune MT, Munch C. Access technique for retroperitoneoscopy. J Urol 1996;156:1120.)

FIG. 126-3. The dissecting balloon is optimally positioned within Gerota's fascia in a posterior direction. Thus the balloon is placed between the psoas muscle (posteriorly) and the kidney or Gerota's fascia (anteriorly) as shown in this line diagram. K, kidney.

FIG. 126-4. The trocar-mounted balloon device consists of a transparent silicone balloon mounted on a 10-mm laparoscopic trocar. This allows laparoscopic monitoring of the balloon distention process. Two designs of the balloon are depicted (Preperitoneal Distension Balloon System, Courtesy of Origin Medsystems, Menlo Park, CA.)

Upon balloon deflation and removal, the primary port (PP) is placed. We prefer the 10-mm Bluntport (Hasson-type cannula) with an internal fascial retention balloon and an external adjustable foam cuff (Origin Medsystems, Inc., Menlo Park, CA). We favor this trocar because it provides an air-tight seal at the primary port site. Such an air-tight seal is more difficult to achieve with a standard Hasson canula. CO 2 is instilled to establish a pneumoretroperitoneum of 15 mm Hg pressure and the laparoscope is inserted. Visualization of the lower renal pole and/or ureter confirms proper balloon dissection of the retroperitoneum. During retroperitoneoscopy we insert most, but not all, of the secondary ports under bimanual control. This bimanual technique is employed for inserting only those secondary ports that are in close proximity to the primary port. For secondary ports that are located at a comfortable distance from the primary port and can therefore be adequately monitored endoscopically, laparoscopic guidance during port placement is preferred. In comparison, during transperitoneal laparoscopy, all secondary trocars are placed under endoscopic control. To safely insert a secondary port under endoscopic control, its point of entry into the retroperitoneum must be clearly visualized laparoscopically. This may be difficult in the restricted retroperitoneal space because the laparoscopic trocars are located closer together during retroperitoneoscopic nephrectomy (6 to 8 cm apart) as compared to transperitoneal laparoscopic nephrectomy (12 to 14 cm apart); consequently, the adjacent undersurface of the flank abdominal wall cannot be adequately visualized endoscopically, despite torquing of the laparoscope. Accordingly, endoscopically guided secondary port placement may be imprecise, with greater potential for inadvertent, and potentially unrecognized, peritoneotomy and visceral injury. Bimanually guided placement of secondary ports minimizes the chances of peritoneal transgression. The technique of insertion of secondary ports under bimanual control is presented in Chapter 122 (Fig. 126-5). During retroperitoneoscopic nephrectomy, four trocars are placed in a diamond configuration ( Fig. 126-6). In addition to the PP, three secondary trocars are placed:

FIG. 126-5. The bimanual technique of secondary trocar ( curved arrow) placement during retroperitoneoscopy. The surgeon's finger retracts the peritoneum away from the undersurface of the abdominal wall; the S retractor ( straight arrow) prevents trocar injury to the surgeon's finger.

FIG. 126-6. Diamond configuration of port placement during right retroperitoneoscopic nephrectomy. The primary port (PP) is located at the tip of the 12th rib and is inserted by the open (Hasson) technique. The 12-mm lower midaxillary line (LMAL) port is placed 2 to 3 cm cephalad to the iliac crest. The 5-mm anterior axillary line (AAL) port is placed at the level of the PP. The 5-mm upper midaxillary line (UMAL) port is placed near the tip of the 11th rib; care must be taken to guard against transpleural placement of this trocar. H, head end of the patient.

A 5-mm upper midaxillary line (UMAL) port located at the tip of the 11th rib A 12-mm lower midaxillary line (LMAL) port located 2 cm cephalad to the iliac crest A 5-mm anterior axillary line (AAL) port, located at the level of the primary port During retroperitoneoscopic nephrectomy, the laparoscope is positioned in the primary port. The surgeon works through the LMAL and AAL ports. The assistant retracts the kidney anteriorly through the UMAL port. Presentation of Renal Anatomy It is important to realize that during retroperitoneoscopy, the kidney presents in an end-on, caudad-to-cephalad orientation. The inferior pole is identified initially, and further dissection reveals the midportion and superior pole of the kidney ( Fig. 126-7). This contrasts with the side-on approach during transperitoneal laparoscopy, where the anterior surface of the kidney and the renal hilum are visualized initially. The transperitoneal approach more closely simulates the situation encountered during open surgical nephrectomy.

FIG. 126-7. Diagrammatic representation of the end-on anatomic orientation of the kidney during retroperitoneoscopy. Note that the kidney is located above the 12th rib, whereas the laparoscopic trocars are placed below the 12th rib. The dark arrow denotes the caudad-to-cephalad direction of retroperitoneoscopic dissection while exposing the kidney. The assistant, working through the UMAL port, retracts the kidney anteriorly (small arrow). The surgeon works through the LMAL and AAL ports. The video laparoscope is located in the primary port (PP).

Retroperitoneoscopy allows early visualization of the posterior aspect of the renal hilum. Since retroperitoneoscopic dissection proceeds in a caudad-to-cephalad direction, the renal vascular structures can be identified by tracing the ureter cephalad to the renal hilum. During a left nephrectomy, the aorta can be used as a guide to locate the origin of the left renal artery. Special Technical Considerations Certain caveats bear emphasis as regards retroperitoneoscopic port placement: Posteriorly located ports can limit instrument maneuverability because the surgeon has to work over the hump of the psoas and erector spinae muscles. The LMAL port must be positioned at a reasonable distance (more than 3 cm) from the iliac crest. Port placement too close to the bone will limit instrument maneuverability and torque capability. Secondary ports are placed with a judicious combination of bimanual and laparoscopic monitoring. Try to separate out the ports as much as possible. Clashing of swords occurs if the ports, and therefore instruments, are located close together. In this regard, the anterior ports can often be placed even more anterior to the anterior axillary line. However, it is critical to clearly visualize the peritoneal line of reflection endoscopically before inserting the secondary port. Cadaveric and live human studies have demonstrated that the peritoneal reflection is consistently located anterior to the posterior axillary line. When a patient is placed in the flank position, the anteroposterior extent of the potential retroperitoneal space is increased twofold. 1 By blunt dissection, with the finger or swab stick inserted through the incision at the tip of the 12th rib, the line of peritoneal reflection can be mobilized even further anteromedially, thus creating a larger retroperitoneal space and allowing greater distance between port sites. If the initial balloon dilation is successfully performed inside Gerota's fascia, the proper plane of dissection on the renal surface gets developed automatically to a large extent. The lower renal pole and ureter are readily visualized. If subfascial balloon placement within Gerota's fascia is unsuccessful, the balloon is manually directed, posterior to the kidney. Balloon inflation between the psoas and the kidney displaces the kidney anteriorly and cephalad in the retroperitoneum. Gerota's fascia must then be incised laparoscopically and the perirenal fat dissected to identify the inferior renal pole. In our experience, it may often be easier to locate the lower renal pole than the ureter in this circumstance. As mentioned, we are able to successfully position the balloon dilator within Gerota's fascia only 50% to 60% of the time. However, in our pediatric experience of retroperitoneoscopic nephrectomy in 10 patients, positioning the balloon within Gerota's fascia does not appear to be as relevant for obtaining good exposure. The minimal retroperitoneal fat in the pediatric age group allows ready identification of the renal surface even when balloon dissection has been performed outside Gerota's fascia.

An inadvertent peritoneal rent during retroperitoneal nephrectomy does not require routine conversion to the transperitoneal approach. 4 In our experience, equilibration of pressure occurs rapidly across the peritoneotomy and usually does not preclude retroperitoneoscopic completion of the procedure. However, the peritoneal cavity must be decompressed of CO 2 at the completion of the procedure. Mobilization of the upper renal pole from the undersurface of the adrenal gland must be performed under good visualization, using electrosurgical scissors. A 30 degree laparoscope provides superior visualization of this area. Meticulous hemostasis is important because the adrenal bed can be a source of significant postoperative hemorrhage. After securing the trocars, retroperitoneoscopic nephrectomy involves the following steps: 1. Mobilize lower pole and posterior surface of the kidney. Retroperitoneoscopic nephrectomy begins with identification of the ureter and/or the lower renal pole. Once the kidney is identified, its posterior surface is mobilized completely. The anterior renal surface is left attached to the peritoneum at this stage of the procedure, thereby preventing the kidney from flopping over. 2. Control renal artery and vein. Retroperitoneoscopy facilitates early access to the posteriorly located renal artery. The renal vessels are dissected in a posterior-to-anterior direction. Vascular control is achieved in a manner similar to that described for transperitoneal laparoscopic nephrectomy. To control the renal vein, the Endo-GIA stapler (Autosuture Endo-GIA, U.S. Surgical Corp.) is inserted through the 12-mm LMAL port. 3. Mobilize upper pole and anterior surface of the kidney. Before addressing its upper pole, the remainder of the kidney must be mobilized. The lateral and anterior renal surfaces are dissected sharply. Caudad traction on the mobilized kidney places the fatty attachments around the upper pole on stretch. The kidney is then carefully detached from the undersurface of the adrenal gland using electrosurgical instruments. 4. Organ extraction. We employ the Endocatch device to entrap and extract the kidney (see previous chapter on transperitoneal nephrectomy). The Endocatch is inserted through the primary port at the tip of the 12th rib. 5. Exit. As described in the chapter on basic laparoscopy ( Chapter 122).

NEPHROURETERECTOMY
Indications for Surgery Nephroureterectomy is indicated for end-stage renal disease due to chronic reflux or obstructive uropathy and for transitional cell carcinoma of the renal pelvis and ureter.

ALTERNATIVE THERAPY
Standard open surgery and transperitoneal laparoscopy are alternative approaches.

SURGICAL TECHNIQUE
Port placement is similar to retroperitoneoscopic nephrectomy as described above, except that the AAL port is 12 mm (instead of 5 mm) ( Fig. 126-6). A double-balloon technique is employed to dissect the ureter distally into the true pelvis. 5 The distending balloon is laparoscopically directed into the pelvis and positioned alongside the ureter. The balloon is inflated with 400 to 500 ml of saline to mobilize the pelvic peritoneum medially. After removal of the balloon, the ureter can be followed distally to the ureterovesical junction. An Endo-GIA tissue stapler is introduced through the 12-mm AAL port to transect and secure the juxtavesical ureter as close to the bladder as possible. Alternatively, a transfixation suture (2-0 Vicryl) is laparoscopically passed through the bladder and tied. The ureter is transected above the suture, and the distal stump can be further reinforced by a catgut endoloop ligature. As mentioned in the preceding chapter, in order to facilitate excision of the distal ureter, the intramural ureter may be initially resected cystoscopically, thus completely detaching the ureter from the bladder. 5 Retroperitoneo-scopic nephrectomy is then performed as described above.

OUTCOMES
Complications The complications are generally the same as for transperitoneal laparoscopic nephrectomy (see Chapter 125). Results A current analysis reveals that more than 200 retroperitoneoscopic nephrectomies have been reported in the literature through 1996 ( Table 126-1). In this collective experience, the mean operative time was 191 minutes; estimated blood loss was 107 cm 3; mean hospital stay was 3.7 days; and convalescence averaged 1.9 weeks. Overall, minor and major complications occurred in 7 (3.5%) and 5 (2.5%) patients, respectively. Conversion to open surgery was required in 13 (6.5%) cases. 6,7,8 and 9

TABLE 126-1. Retroperitoneoscopic nephrectomy: international experience—December 1996 a

In our limited experience with retroperitoneoscopic nephroureterectomy for symptomatic end-stage vesicoureteral reflux in 2 patients, operating time was 6 and 5.5 hours; blood loss was 200 ml; hospital stay was 4 days; and convalescence required 3 weeks. One patient developed a pelvic hematoma, postoperatively, requiring transfusion with 2 units of packed cells. A postoperative cystogram in one patient revealed a 3- to 4-cm refluxing stump of ureter. It was concluded that improved access to the distal ureter is needed during retroperitoneoscopic surgery. 5 Both retroperitoneoscopy and transperitoneal laparoscopy have inherent disadvantages and advantages. Current disadvantages of retroperitoneoscopy include a constrained working space leading to proximity of trocar sites and limited maneuverability of instruments. From a technical standpoint, at least in the learning curve phase, retroperitoneoscopy may be technically more demanding than transperitoneal laparoscopy. The balloon dilator may be distended improperly within the flank muscle layers. Unrecognized peritoneal transgression during secondary port placement may cause injury to intraperitoneal viscerae. Subcutaneous emphysema during retroperitoneoscopic surgery may result in elevated levels of systemic CO 2.10 However, to our knowledge, no clinically significant sequelae have been reported from such hypercarbia to date. Use of the Bluntport canula as the primary port, as described herein, has eliminated problems with air leak and subcutaneous emphysema in our experience. In comparison to transperitoneal laparoscopy, retroperitoneoscopy offers certain distinct advantages. It allows direct access to the retroperitoneally located

genitourinary organs. It precludes the need to enter the peritoneal cavity and mobilize the colon, thereby minimizing postoperative ileus. The chances of intraperitoneal organ injury are minimized considerably, although not eliminated. The potential for malignant seeding or bacterial contamination of the peritoneal cavity is removed. Postoperative fluid collections (urinoma, hematoma) are restricted to the retroperitoneum and can be drained adequately. During laparoscopic nephrectomy, retroperitoneoscopy allows initial control of the renal artery. Although total operative times may be somewhat decreased by the retroperitoneoscopic approach, postoperative discomfort, hospital stay, and convalescence are currently unaltered when compared to the transperitoneal laparoscopic approach. During retroperitoneoscopy, attention must be directed to four steps: 1. 2. 3. 4. Location of primary port Optimal positioning of the balloon dilator Placement of secondary ports with a combination of bimanual and laparoscopic control Prevention of port crowding

At our institutions, the vast majority of laparoscopic nephrectomies for benign disease are now performed with the retroperitoneoscopic approach. With further experience, retroperitoneoscopy may emerge as the preferred approach for laparoscopic surgery of the upper urinary tract. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Capelouto CC, Moore RG, Silverman SG, Kavoussi LR. Retro-peritoneoscopy: anatomical rationale for direct retroperitoneal access. J Urol 1994;152:2008. Gaur DD. Laparoscopic operative retroperitoneoscopy. J Urol 1992;148:1137. Gill IS. Retroperitoneal and pelvic extraperitoneal laparoscopy: an international perspective (Abstract). P17 5. J Endourol 1996;17;504. Gill IS, Grune MT, Munch LC. Access technique for retroperitoneoscopy. J Urol 1996;156:1120. Gill IS, Munch LC, Lucas BA, Das S. Initial experience with retroperitoneoscopic nephroureterectomy: use of a double-balloon technique. Urology 1995;46:747. Gill IS, Das S, Munch LC, Grune MT. Retroperitoneoscopy and pelvic extraperitoneoscopy: 124 cases (Abstract). J Endourol 1996;17:503. Mandressi A, Buizza C, Antonelli D, et al. Retro extraperitoneal laparoscopic approach to excise retroperitoneal organs: kidneys and adrenal gland. Min Inv Ther 1993;2:213. McDougall EM, Clayman RV, Fadden PA. Retroperitoneoscopy: the Washington University Medical School experience. Urology 1994;43:446. Rassweiler JJ, Henkel TO, Stock C, Frede T, Alken P. Retroperitoneoscopic surgery: technique, indications and first experience. Min Inv Ther 1994;3:1. Wolf JS Jr, Monk TG, McDougall EM, McClennan BL, Clayman RV. The extraperitoneal approach and subcutaneous emphysema are associated with greater absorption of carbon dioxide during laparoscopic renal surgery. J Urol 1995;154:959.

Chapter 127 Laparoscopic Retroperitoneal Renal Procedures Glenn’s Urologic Surgery

Chapter 127 Laparoscopic Retroperitoneal Renal Procedures
John B. Adams II

J. B. Adams II: Department of Surgery, Section of Urology, Medical College of Georgia, Augusta, Georgia 30912-4050.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Procedures Renal Cystectomy Nephropexy Ureterolithotomy Chapter References

Renal failure can develop as a result of a myriad of disorders. Renal biopsy and subsequent analysis by a variety of microscopic techniques provide important diagnostic information to determine the cause of renal disease responsible for the renal failure.

DIAGNOSIS
The etiology of intrinsic renal insufficiency is most often sought by nephrologists. Biopsy findings can help direct the medical management of renal disease. Preoperatively, a KUB and renal ultrasound or noncontrast CT can be helpful in locating and characterizing the kidney to be biopsied.

INDICATIONS FOR SURGERY
Nephrotic syndrome due to a glomerular lesion is one of the most common indications for renal biopsy. There is a subgroup of patients in need of a renal biopsy that can be well served by a laparoscopic renal approach. Patients with a solitary kidney, severe hypertension, bleeding disorders, morbid obesity, and patients who are uncooperative are candidates for alternatives to percutaneous needle biopsy. 7 Direct vision for obtaining adequate tissue for diagnosis and to ensure hemostasis is the advantage of the laparoscopic retroperitoneal approach.

ALTERNATIVE THERAPY
A percutaneous needle biopsy under ultrasonographic guidance is the most common method of obtaining renal tissue for diagnostic purposes. When this technique is not successful, an open biopsy through a subcostal or dorsal lumbotomy incision can be performed, but this method is invasive.

SURGICAL TECHNIQUE
Patient preparation and positioning are identical for standard retroperitoneal laparoscopic surgery (see Chapter 122). A 1.5-cm incision is made in the posterior axillary line just lateral to the sacrospinalis muscle 2 cm above the iliac crest. The surgeon is positioned behind the patient, as most of the dissection is easily performed from this position. Finger dissection is carried down to the lumbodorsal fascia where sharp incision is performed to enter the retroperitoneal space. A 10-mm port is placed and a 0 degrees laparoscope introduced to verify correct anatomic position. Balloon dilation is carried out to dissect the retroperitoneum and create a functional working space (see Chapter 124). The pneumoretroperitoneum is then maintained at 15 to 20 mm Hg and the retroperitoneal space is inspected visually. A second 10-mm port is placed under direct vision in the posterior axillary line just below the 12th rib ( Fig. 127-1).

FIG. 127-1. Patient shown in the flank position from the back. Port sites shown for retroperitoneal laparoscopic procedures. 0, 10-mm port; F, 5-mm port (optional in renal biopsy procedure).

Further dissection can be carried out using the laparoscope or a 5mm grasper through the other 10-mm port to identify the lower pole of the kidney, which is usually positioned medially. Cup biopsy forceps or spoon forceps are used to obtain several samples of kidney tissue to be sent to the pathologist for routine and special studies (Fig. 127-2). A large-core Tru-cut type of biopsy needle can be passed percutaneously to obtain the sample. It is helpful to have the pathologist advise if special solutions are needed to process the tissue and to verify that enough tissue has been obtained before terminating the procedure.

FIG. 127-2. Retroperitoneal view of lower pole of kidney prepared for renal biopsy.

Hemostasis can be achieved with electrocautery or argon beam coagulator. The retroperitoneal pressure should be reduced to 8 mm Hg and the biopsy site and dissected space examined to ensure adequate hemostasis. The port site fascial defects can either be closed under direct vision with a 2-0 Vicryl suture or left open because the procedure is exclusively retroperitoneal. The skin edges are closed with a running subcuticular 4-0 Vicryl suture. Sterile strips are used to reapproximate the skin edges. The Foley catheter is removed in the recovery room. A 24-hour observation period is recommended while the patient resumes a regular diet and ambulates since delayed bleeding has been reported. Oral pain medications are utilized depending on the level of renal function.

OUTCOMES
Complications There are limited series of laparoscopic retroperitoneal biopsies, making the frequency of complications difficult to estimate. Gaur, in a large series of 17 patients undergoing routine biopsy, had two patients who had problems with hemostasis. 2 One patient needed an incision enlarged to control the bleeding, whereas another had gross hematuria that resolved spontaneously. No patients required blood products. If electrocautery is not adequate to control bleeding, then an argon beam coagulator should be available. Results In Gaur's series, 100% of patients had an adequate amount of renal tissue obtained for diagnosis. 2 Because of the direct visualization using this technique, an adequate biopsy may be ensured. Also, the surgeon may be confident of adequate hemostasis during the course of the biopsy. The laparoscopic renal biopsy through a retroperitoneal approach is an excellent method for obtaining renal tissue for diagnostic purposes. This technique can be used in cases where a percutaneous biopsy is not the preferred method and can be accomplished with minimal patient morbidity.

PROCEDURES
Renal Cystectomy Introduction Simple cysts are the most common lesions of the kidney. The incidence of cysts increases with age; cysts are usually found on routine imaging studies performed for other reasons. Symptoms, although rare, are attributable to the size of the cyst or to its impingement on adjacent structures such as the ureter or renal parenchyma. Diagnosis Pain from renal cysts is a diagnosis of exclusion. Renal cysts can be identified incidentally on intravenous pyelogram (IVP). CT scan is the most accurate method of characterizing the renal cyst wall. If there are any questions about the cyst wall or intraluminal material, a renal ultrasound may be helpful to differentiate between a solid or cystic mass. Magnetic resonance imaging (MRI) is a relatively new modality but is gaining experience in characterizing renal cysts accurately. Indications for Surgery Symptomatic, solitary renal cysts can be managed using percutaneous aspiration and injection of sclerosing agents. For patients with recurrent symptomatic cysts, or cysts that are not easily accessed through the percutaneous approach, a laparoscopic retroperitoneal technique could be a minimally invasive alternative. If a cyst is suspicious for malignancy (thick wall, thickened septations, hyperdense cyst fluid), aspiration and biopsy of the cyst wall are indicated. Alternative Therapy Aspiration and injection of sclerosing agents is considered to be the initial course of management for symptomatic renal cysts. Other methods to treat these cysts are retrograde ureterorenoscopy and antegrade percutaneous resection. Open surgical marsupialization can be performed but carries the morbidity associated with a flank incision. If malignancy is suspected, cyst fluid can be aspirated and submitted for cytologic evaluation. Surgical Technique Standard positioning and preparation are carried out and the retroperitoneal space is developed with a balloon dissector as previously described (see Chapter 122). Renal parenchyma and the cyst can be easily identified in the retroperitoneum. Radiographs should be present in the operating room to direct the laparoscopic dissection anteriorly or posteriorly depending on the location of the renal cyst. Laparoscopic ultrasound may also be used in some cases where cysts are difficult to locate. A second instrument may be utilized for grasping by placing a 5-mm trocar midway between the iliac crest and the 12th rib in the posterior axillary line ( Fig. 127-1). This provides an additional instrument for clearing away perirenal fat to expose the cyst and for countertraction when unroofing the cyst. The cyst wall is incised with dissecting scissors or a hook blade and sent for pathologic analysis ( Fig. 127-3). The fluid is aspirated and sent for cytology. The base of the cyst is examined for evidence of neoplastic change. If suspicious areas are identified, cup biopsy forceps are used to perform frozen section analysis. The edges of the cyst are cauterized and hemostasis is verified after the pneumoretroperitoneum has been lowered to 8 mm Hg. The argon beam coagulator (ConMed Corporation, Utica, NY) can be used to quickly paint over any raw bleeding surfaces. If the argon beam coagulator is used, the retroperitoneum must be vented frequently to prevent buildup of pressure.

FIG. 127-3. Renal cyst ready to be excised with dissecting scissors.

If the cyst is large, then a window can be fashioned using dissecting scissors. This decompresses the cyst without necessitating the removal of the entire cyst wall.

Retroperitoneal fat can be packed into the defect to help maintain window patency. An alternate approach is to excise the entire cyst wall. If there is ever any concern that the collecting system has been entered, indigo carmine can be administered. The procedure is terminated in the standard fashion for retroperitoneal surgery (see Chapter 122). Outcomes Complications Bleeding is always a possible complication of marsupialization of a renal cyst. However, this has not been noted in the several series using a laparoscopic approach. 10,11,12 and 13 Reaccumulation of cyst fluid is uncommon but can occur when the window fashioned during marsupialization becomes occluded. Neih et al. demonstrated that the use of fat was helpful in maintaining the patency of the marsupialization window. 11 When excising cysts in the hilum or medially situated cysts, care should be taken to know where adjacent structures lie (e.g., vasculature and ureter). Problems can be avoided using careful dissection techniques. Results A number of investigators have successfully managed renal cysts with a laparoscopic approach. 5,10,11,12 and 13 The largest series of 20 patients treated by Guazzoni et al. showed resolution of the cysts and associated symptoms at a 1-year follow-up. 5 There is a variety of methods including unroofing the cyst and marsupialization of multiple cysts. 11,12 Munch et al. described a retroperitoneal approach that approximates the one above. 10 The laparoscopic retroperitoneal approach is an excellent and minimally invasive method of treating symptomatic renal cysts. It avoids the potential complications of transperitoneal laparoscopic approach and can be performed with minimal patient morbidity. Nephropexy Introduction Renal descent of more than two vertebral bodies or more than 5 cm when the patient moves to the erect position is the definition of nephroptosis. The condition is more common in thin women and is demonstrated preferentially on the right side. Nephroptosis is usually an incidental finding and is rarely symptomatic. Diagnosis An IVP with the patient in the supine and standing position is the initial test obtained after a careful history suggests symptoms caused by a hypermobile kidney. The flank pain, lower abdominal pain, and nausea associated with nephroptosis are thought to be secondary to renal ischemia or acute obstruction. The kidney might become hydronephrotic when it descends into the pelvic position. Palpation of the kidney in the lying and standing positions can confirm radiographic test results. A nuclear renal scan is useful in detecting an obstructive pattern after renal descent, as well as diminished blood flow to the kidney when the patient is erect. In addition, the pain associated with obstruction should be relieved in these patients by the placement of a ureteral stent. Alternative Therapy The surgical management of nephroptosis is varied and more than 170 different procedures have been recorded since the late 1800s. Laparoscopic nephropexy adds another dimension to the less invasive options. The retroperitoneal approach provides a direct route to the kidney and may be helpful in patients who have had previous abdominal surgery. Surgical Technique Preparation, positioning, and balloon dissection are used as described in Chapter 122. Direct vision using a 0 degrees laparoscope is used to identify the retroperitoneal structures. A second 10-mm trocar is placed in the posterior axillary line just below the 12th rib. Dissection is carried laterally until the psoas muscle is identified and the kidney visualized medially. The entire posterior and lateral surface of the kidney should be cleared of perirenal fat to allow for accurate placement of tacking sutures. A 5-mm port can be placed midway between the two initial port sites, just lateral to the sacrospinalis muscle ( Fig. 127-1). The operative table is then moved to a head-down position. This causes the hypermobile kidney to be positioned as superiorly as possible. Sutures are then placed in the renal capsule, which is of adequate strength. Care should be taken not to place the sutures too deep, as laceration of the renal parenchyma may cause troublesome bleeding. The sutures placed in the retroperitoneum can be successfully placed in the fascia overlying the quadratus and psoas muscle. Nonabsorbable 3-0 sutures are placed superiorly, laterally, and inferiorly for at least a three-point fixation ( Fig. 127-4). Intracorporeal knot tying is difficult in the enclosed retroperitoneal space. However, several devices now can help with laparoscopic suturing. The Endo-Stitch (Auto Suture, USS, Norwalk, CT) can make suture placement and knot tying less cumbersome. The Lapara-Ty (Ethicon Endosurgery, Cincinnati, OH) allows clips to be placed on the ends of the suture, making knot tying unnecessary. The procedure is terminated in a fashion identical to that of other retroperitoneal laparoscopic interventions. A regular diet can be written for the patient on the evening of surgery and the patient released after a 24-hour observation period.

FIG. 127-4. Kidney pexed to the fascia overlying the posterior retroperitoneum. Sutures have been placed to fix the kidney superiorly, laterally, and inferiorly.

Outcomes Complications There are very few complications noted in the literature that covers laparoscopic nephropexies. 1,8 One problem occurring postoperatively may be the recurrence of the nephroptosis. Careful dissection around the kidney and meticulous placement of the sutures should provide adequate fixation of the kidney. Movement of the table to the head-up position after sutures have been placed can also verify the stability of the kidney within the retroperitoneum. Bleeding due to sutures that are placed too deeply in the parenchyma can be avoided by careful suturing technique. Results A number of methods of laparoscopic nephropexy have been reported. 1,8 Hubner et al. used a polyglactin net to suspend the kidney in a cephalad position in 10 patients, with success in all cases. 8 In six patients, Elashry et al. used capsular tacking sutures to fix the kidney with good results. 9 The retroperitoneal laparoscopic

technique provides a direct approach to the ptotic kidney and gives the surgeon another minimally invasive option in the management of nephroptosis. Ureterolithotomy Introduction The explosion of technologic developments in the field of endoscopy has enabled the urologist to choose from an ever-widening variety of techniques to treat ureteral stones. Ureteroscopy allows the extraction or fragmentation under direct vision of calculi in the ureter. Patients can also be treated with extracorporeal shock wave lithotripsy (ESWL) with good success rates. However, these techniques are not always successful due to anatomic considerations and/or stone size. Diagnosis In patients with ureteral calculi, the diagnosis is confirmed with a KUB and IVP. A retrograde pyelogram (RPG) is helpful in delineating the course and patency of the ureter below the obstructing stone and should be an integral part of the evaluation. This can be performed at the time of the laparoscopic procedure when a stent is to be placed. Indications for Surgery Laparoscopic retroperitoneal ureterolithotomy can be employed to remove impacted ureteral stones that have failed endoscopic methods or lithotripsy. Patients who have anatomic anomalies or are morbidly obese, which may preclude the use of more traditional methods of endoscopic stone removal, are also potential candidates for laparoscopic ureterolithotomy. Ureteral stones that are large, hard, and impacted may be treated initially with laparoscopic ureterolithotomy as an alternative to multiple endoscopic procedures. Alternative Therapy Endoscopic techniques and ESWL are successful in most stone cases. Percutaneous approaches are also possible with endoscopic removal from above in an antegrade fashion. If these techniques fail, open ureterolithotomy through a subcostal or dorsal lumbotomy approach can be used. The morbidity of open surgery, however, can be avoided by utilization of a laparoscopic ureterolithotomy technique. Surgical Technique After appropriate intravenous antibiotics have been administered and general anesthesia achieved, the patient is placed in the lithotomy position. A ureteral stent is then maneuvered past the obstructing ureteral stone using fluoroscopic guidance. If the stone cannot be bypassed, then a 0.038-in. floppy-tip guidewire is positioned to the level of the stone through an open-ended 5-Fr stent. The stent and wire are secured to the Foley catheter and kept in the sterile field. The use of a lighted ureteral stent in order to help locate the affected area of the ureter has been reported. The patient is then moved to the appropriate lateral decubitus position corresponding to the side with the ureteral stone. A standard retroperitoneal approach is made, as described in Chapter 122. After balloon dissection, a 0 degrees laparoscope is used to inspect the extraperitoneal space. A second 10-mm trocar is placed under direct vision just below the 12th rib in the midaxillary line. A 5-mm trocar can be placed in the posterior axillary line, midway between the 12th rib and the iliac crest, for additional instrument utilization ( Fig. 127-1). Dissection through the two upper ports is utilized to identify the midureter, while the two lower ports are used to manipulate the proximal ureter. The ureter should be visible medially, just below the posterior inferior surface of the kidney. Identification of the lower pole of the kidney and psoas muscle should give the laparoscopist a good anatomic starting point to begin looking for the ureter. Once the ureter is identified, the impacted stone can usually be identified as a bulge in the ureter, but fluoroscopy should be available to assist in stone localization if needed. After the stone has been located, an incision in the ureter can be made with an endoscopic knife or fine-tipped 5-mm scissors. The ureteral defect is extended to a length that affords comfortable removal of the stone or stone fragments, with care taken not to spiral the incision ( Fig. 127-5). The stone is then removed through a 10-mm port under direct vision. If the stone is too large to remove through the port, a small organ entrapment sack can be used. The stone is broken up in the sack and removed in pieces through a 10-mm trocar site.

FIG. 127-5. Incision in ureter with impending removal of stone with grasping forceps.

The ureteral stent should be clearly visible, and stone fragments can be irrigated free of the ureteral lumen. If the wire is in place up to the area of obstruction, it can now be advanced into the proximal ureter under direct laparoscopic vision. Using fluoroscopy, a double-J stent can be placed in the renal pelvis over the wire and the wire removed. If the incision in the ureter is less than 1 cm, it can be left open or a single suture can reapproximate the ureteral edges. A small, flat suction drain is introduced through the lateral 5-mm port and sutured in place. The 10-mm port sites are closed in the standard fashion. Postoperatively, a Foley catheter is left in place for 2 days. When the catheter is removed, if the suction drain output does not increase, the drain is also removed. If an increase in drainage occurs, the Foley is reinserted for an additional 24 hours and the procedure is repeated. The patient can be discharged in 2 to 3 days. The stent is removed in 4 weeks. A follow-up IVP is obtained 6 to 8 weeks after the stent is removed. Outcomes Complications To avoid bleeding, there should be careful dissection in the retroperitoneum around the ureters. Micali's series had three complications in a series of 17 patients who had transperitoneal laparoscopic removal of ureteral and renal pelvic stones. 9 Two patients had a prolonged ileus and one patient had a retroperitoneal urinoma that was drained secondarily. This should not occur if there is adequate drainage of the retroperitoneal space and a stent is left to drain the renal collecting system. After trauma from the stone and stone removal it is possible to have ureteral stricture disease. This has not been commented on in the literature. However, a follow-up intravenous pyelogram 6 to 8 weeks following removal of the ureteral stent is advisable. If the ureteral incision is closed, there should be a loose approximation of the mucosa. Otherwise, an incision less than 1 cm in length may be left open in a stented system. Failure to locate the ureteral calculus necessitates use of fluoroscopy

during the case. In one series, stone removal was successful in 15 of 17 cases. 9 One stone was not located laparoscopically during the initial use of an investigational three-dimensional system. The other stone was a branched calculus that had to be broken with a lithotrite through a small incision. Most complications can be avoided by utilizing careful laparoscopic dissection techniques. Results A majority of stones can be taken care of using endoscopic or percutaneous techniques. Because of the success of endoscopic and lithotripsy techniques, the number of ureterolithotomies performed with laparoscopic techniques is small. However, some patients become candidates for laparoscopic removal when more conventional minimally invasive techniques are not possible or have failed. Harwood has been successful with nine cases in which laparoscopic techniques were used to remove impacted ureteral stones.6 Three of these ureterolithotomies were performed via a retroperitoneal route. One was successful whereas two had to be converted to an intraperitoneal approach secondary to a limited working space. Gaur et al. presented a retroperitoneal laparoscopic approach to midureter and upper ureter calculi. 4 He successfully removed a number of large calculi using this technique. In a subsequent series of eight patients, removal of stones from the renal pelvis was successful in five of the eight. 3 The procedure was completed using open conversion in the remaining three patients. The conversions were due to inability to dissect the retroperitoneal space in one and a peritoneal tear due to balloon inflation in another. A third patient had dense peripelvic adhesions that made laparoscopic dissection very difficult. However, these reports indicate that a retroperitoneal laparoscopic approach to the upper and midureter is technically feasible and should be considered as an addition to the minimally invasive techniques utilized in removing ureteral stones. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. Elashry OM, Nakada SY, McDougall EM, Clayman RV. Laparoscopic nephropexy: Washington University Experience. J Urol 1995;154:1655–1659. Gaur DD, Agarwal DK, Khochikar MV, Purhoit KC. Laparoscopic renal biopsy via retroperitoneal approach. J Urol 1994;151(4):925–926. Gaur DD, Agarwal DK, Purhoit KC, Darshane AS. Retroperitoneal laparoscopic pyelolithotomy. J Urol 1994;151:927–929. Gaur DD, Agarwal DK, Purhoit KC, Darshane, AS, Shah BC. Retroperitoneal laparoscopic ureterolithotomy for multiple mid-ureteral calculi. J Urol 1994;151(4):1001–1002. Guazzoni G, Montorsi F, Berganaschi F, et al. Laparoscopic unroofing of simple renal cysts. Urology 1994;43(2):154. Harewood LM, Webb DR, Pope AJ. Laparoscopic ureterolithotomy: the results of an initial series and an evaluation of its role in the management of ureteric calculi. Br J Urol 1994;74(2):170–176. Hinman F Jr. Open renal biopsy. In: Lamsback W, ed. Atlas of urologic surgery. Philadelphia: WB Saunders, 1988;839–890. Hubner WA, Schramek P, Pfluger H. Laparoscopic nephropexy. J Urol 1994;152:1184. Micali S, Moore RG, Averch TD, Adams JB, Kavoussi LR. The role of laparoscopy in the treatment of renal and ureteral calculi. J Urol 1997;157:463–466. Munch LC, Gill IS, McRoberts JW. Laparoscopic retroperitoneal renal cystectomy. J Urol 1994;151:135–138. Nieh PT, Bihrle W. Laparoscopic marsupialization of massive renal cyst. J Urol 1993:150:171–173. Stoller ML, Irby PB, Osman M, Carroll PR. Laparoscopic marsupialization of a simple renal cyst. J Urol 1992;150:1486–1488. Teichman JMH, Hulbert JC. Laparoscopic marsupialization of the painful polycystic kidney. J Urol 1995;153:1105–1107.

Chapter 128 Laparoscopic Pyeloplasty Glenn’s Urologic Surgery

Chapter 128 Laparoscopic Pyeloplasty
Robert G. Moore and Jeffrey A. Cadeddu

R. G. Moore: Department of Surgery, St. Louis University, St. Louis, Missouri 63110–0250. J. A. Cadeddu: The James Buchanan Brady Urological Institute, Johns Hopkins Bayview Medical Center, Baltimore, Maryland 21224.

Diagnosis Indications for Surgery Alternative Treatment Surgical Technique Outcomes Complications Results Chapter References

Ureteropelvic junction (UPJ) obstruction is an impairment of urinary flow between the renal pelvis and the ureter. The etiologies of the obstructed UPJ can be divided into two categories: congenital and acquired. Congenital conditions consist of intrinsic narrowing of the UPJ, a high insertion of the ureter into the renal pelvis, or an extrinsic compression of the UPJ by a segmental renal vessel or tissue band. Acquired etiologies include stone disease, urothelial tumors, and inflammatory or postoperative strictures. An obstructed UPJ can be diagnosed at any age, ranging from the neonatal period to the geriatric age group, and occurs with an equal male-to-female distribution in adults. This chapter will review the diagnosis of the obstructed UPJ, indications for surgery, operative techniques, outcome of laparoscopic pyeloplasty, and potential complications.

DIAGNOSIS
The classic presentation of an obstructed UPJ in the adult patient is episodic flank or abdominal pain after a fluid challenge. Other potential signs and symptoms include hematuria associated with minor trauma and urinary tract infection with flank pain. An obstructed UPJ is diagnosed and evaluated by three separate radiographic studies: intravenous pyelogram (IVP), renal scan with furosemide washout, and retrograde pyelogram. Intravenous pyelogram is the initial radiographic examination chosen when an obstructed UPJ is suspected. This study delineates the anatomy of the renal parenchyma and collecting system and allows a gross overall assessment of the function of the affected renal unit. A diuretic nuclear renal scan can measure the percentage of renal function and the degree of obstruction of each kidney. Renal units with high-grade obstruction should be reassessed after drainage. If the distal ureter and length of the narrowed UPJ is not demonstrated on IVP, a retrograde pyelogram is performed. The optimal timing of this procedure is just prior to definitive repair of the UPJ obstruction. Recently, both spiral computed tomography and endoluminal ultrasound have been utilized to diagnose the presence of a UPJ crossing vessel.

INDICATIONS FOR SURGERY
The four indications for surgical intervention in the obstructed UPJ are as follows: 1. 2. 3. 4. Flank or abdominal pain secondary to obstruction Deteriorating renal function Urinary tract infection Development of renal calculi

Patients ideally suited for laparoscopic pyeloplasty include those with a large redundant renal pelvis and/or a segmental crossing vessel associated with an obstructed UPJ. Relative contraindications for laparoscopic pyeloplasty include prior open renal surgery, history of pyelonephritis, or history of prior trauma.

ALTERNATIVE TREATMENT
Open pyeloplasty is the gold standard for treatment of the obstructed UPJ, with long-term success rates of 90% to 100% in both the adult and pediatric populations. 1 Endourologic techniques have been utilized to treat the obstructed UPJ since 1983. Antegrade, retrograde (ureteroscopic), and fluoroscopic (Acucise) endopyelotomy techniques yield success rates of 66% to 86%. 1 Risk factors for failure of these incisional techniques include a large redundant renal pelvis and/or a segmental crossing renal vessel. Patients with these characteristics are ideally suited for reconstruction of the UPJ. Laparoscopic pyeloplasty was conceived to combine the high success rate of open pyeloplasty with the decreased morbidity of laparoscopic surgery.

SURGICAL TECHNIQUE
The specific equipment needed for the procedure is noted in Table 128-1. The patient is placed under general endotracheal anesthesia. An internal ureteral stent is placed with the aid of a flexible cystoscope under fluoroscopic guidance. Ureteral stent selection should be of sufficient length to extend from the bladder to an upper pole calyx and thus prevent the end of the stent from lying in direct contact with the newly reconstructed anastomosis. Both a Foley catheter and orogastric tubes are placed to decompress the bladder and stomach, respectively.

TABLE 128-1. Specific instrumentation for laparoscopic pyeloplasty

The patient is placed in a 60 degree lateral decubitus position, with the loin positioned over the kidney rest. An axillary roll is placed beneath the contralateral axilla. Both arms are flexed toward the patient's head and padded with a small pillow. A pillow is placed between both legs with the top knee flexed and the contralateral leg straight (Fig. 128-1). The patient is secured to the table with 2-in. cloth tape across the shoulders and hips. The two video monitors are placed at the head of the table, one on each side ( Fig. 128-2). The surface of the flank and abdomen is shaved and prepared.

FIG. 128-1. The 60 degree lateral decubitus position for the laparoscopic approach to the upper ureter.

FIG. 128-2. Positioning of operating room equipment and personnel. S, surgeon; A, assistant surgeon; N, nurse.

The procedure is performed via the transperitoneal approach. 3,4,6 Basic transperitoneal laparoscopic techniques are discussed in Chapter 122. A total of four trocars are utilized in a diamond configuration ( Fig. 128-3). The first trocar is placed under direct vision with a Visiport (U.S. Surgical, Norwalk, CT) in the midclavicular line 2 cm below the costochondral angle. Two additional 10/12-mm trocars are placed under direct vision; one in the midclavicular line in the ipsilateral lower quadrant and one periumbilical. The respective colonic flexure is mobilized medially by incising the lateral peritoneal reflection, and an additional 5-mm trocar is placed in the midaxillary line. All trocars are secured to the skin with O polyglycolic acid suture.

FIG. 128-3. Trocar sites for laparoscopic pyeloplasty. 12 mm = ; 5 mm = .

If the affected kidney has a large redundant renal pelvis, it can often be seen through the peritoneal surface lateral to the colon. 3,4,6 After reflection of the colon, the proximal ureter is identified in the retroperitoneum as it crosses the psoas muscle. The recently placed ureteral stent is easily palpated and the renal pelvis is identified by following the ureter cephalad. This can be facilitated by a sweeping motion that is parallel to the ureter. A crossing vessel is encountered 50% of the time. All arterial vessels should be conserved to preserve the most renal parenchyma. The renal pelvis, UPJ, and proximal 3 cm of ureter are freed from adjacent structures by a combination of blunt and sharp dissection. Extensive dissection of the proximal ureter should be avoided to preserve collateral vascular supply. In performing a dismembered pyeloplasty (Fig. 128-4) the renal pelvis above the obstructed UPJ is circumferentially incised. After completion of the pyelotomy incision, the proximal end of the stent is removed from the collecting system. Care must be taken to avoid cutting the internal stent. The ureter below the obstruction is incised circumferentially and the resected UPJ is removed from the end of the stent. The proximal ureter is then spatulated laterally with laparoscopic scissors for 1 cm. If a redundant renal pelvis is present, it can be excised at this time. Renal stones may be extracted through the pyelotomy incision by passing a flexible cystoscope via a 10/12-mm cannula.

TABLE cellSpacing=0 cellPadding=0 align=left border=0 hspace="10" vspace="5"> FIG. 128-4. Dismembered pyeloplasty for a ureteropelvic junction obstruction by a lower pole crossing vessel. (A) Outline of incisions. (B) The apex of the spatulated ureter is anastomosed to the dependent cut renal pelvis anterior to the lower pole vessel. (C) The posterior anastomosis is completed followed by the anterior row. (D) The remaining pyelotomy incision is closed last.

A Foley Y-V pyeloplasty is indicated when an obstructed UPJ is associated with a high ureteral insertion or a small renal pelvis. After exposure of the UPJ and renal pelvis, a wide-based V-shaped flap is constructed from the anterior surface of the renal pelvis ( Fig. 128-5). The apex of the V flap is placed at the UPJ, while the base of the triangular flap is located on the anterior medial surface of the renal pelvis. From the apex of the triangular flap an incision is then extended caudad traversing the UPJ obstruction along the lateral aspect of the ureter.

FIG. 128-5. Foley Y-V plasty for a ureteropelvic junction obstruction associated with a high insertion of the ureter. (A) Outline of incisions. (B) The apex of the spatulated ureter is anastomosed to the dependent cut renal pelvis. (C) The posterior wall is completed first. (D) Completed approximation of anterior wall.

Polyglycolic acid suture (4–0) is utilized for all suturing. All suturing and stitch placement at the ureteral-pelvic anastomosis is performed with the Endo-Stitch (U.S. Surgical). 5 When a ventral segmental crossing renal vessel is encountered, it is repositioned dorsal to the anastomosis. The initial stitch in a dismembered pyeloplasty is placed from the apex of the spatulated proximal ureter to the lowest position of the renal pelvis (see Fig. 128-4). Sutures are placed such that the tied square knots are external to the urinary tract. All knots are tied intracorporally with the Endo-Stitch ( Fig. 128-6 and Fig. 128-7) because these have proven stronger than extracorporally secured knots.

FIG. 128-6. Endo-Stitch manipulation of suture to form a square knot.

FIG. 128-7. Endo-Stitch completion of square knot.

The posterior row of interrupted sutures is placed next, with care taken to ensure that a mucosal-to-mucosal tension-free anastomosis is created (see Fig. 128-4). The ureteral stent is placed back into the upper pole collecting system, and the anterior row of interrupted sutures are placed. In all, 6 to 10 interrupted sutures are utilized to reconstruct the ureteropelvic anastomosis. The remaining pyelotomy incision is closed with a running 4-0 polyglycolic acid suture. Suturing of the Foley Y-V pyeloplasty is similar to the dismembered pyeloplasty closure. All suture material and knot tying techniques are the same, with the exception that all sutures placed are of the interrupted variety. The apex of the flap is sutured down to the inferior portion of the ureteral incision (see Fig. 128-5). The posterior portion of the flap is closed first. The ureteral stent is then placed back in the renal pelvis and the anterior part of the anastomosis is closed. A closed suction drain is placed dorsal to the newly reconstructed UPJ and brought out through the lateral 5-mm cannula. This drain is secured to the skin with a 2-0 silk suture. The colon is replaced over the operative field and all 10/12-mm trocar sites are closed with 0 polyglycolic acid suture under direct vision. The orogastric tube is removed in the operative suite. A clear liquid diet is started on the night of surgery and advanced as tolerated to a regular diet. The Foley catheter is removed on postoperative day 2. The closed suction drain is discontinued on postoperative day 3 if drainage is less than 20 cm 3. The patient is discharged from the hospital when a regular diet is tolerated (usually on day 3). The ureteral stent is removed in the office 3 to 6 weeks postoperatively. An IVP is obtained 3 months postoperatively to assess the patency of the newly repaired UPJ. Renal scan (MAG-3) with Lasix washout is performed 6 months postoperatively to quantitate function and to rule out obstruction of the repaired renal unit. Thereafter, an IVP or renal scan is obtained at yearly intervals.

OUTCOMES
Complications Potential complications of laparoscopic pyeloplasties are similar to those of open pyeloplasties. The complications common to both techniques consist of bleeding, infection, recurrence of the UPJ obstruction, and persistent urinary leak (urinoma). A urinoma can be treated with percutaneous drainage of the collection and a percutaneous nephrostomy. A complication unique to laparoscopic surgery is injury to adjacent intraperitoneal organs. If recognized intraoperatively, the respective organ is repaired using either laparoscopic or open techniques. Patients experiencing abdominal pain during the early postoperative period should be thoroughly evaluated for surgical etiologies. Five complications (10%, four minor and one major) occurred in the first 50 laparoscopic pyeloplasties performed at Johns Hopkins University (Robert G. Moore, personal data). One patient had to be readmitted 1 day after discharge for an adynamic ileus. She responded to conservative therapy and was discharged after 2 days. Of the four minor complications, one occurred intraoperatively and three occurred in the postoperative period. A right-side colonic diverticulum was inadvertently

clipped. This was recognized intraoperatively and the diverticulum excised with a GIA stapler. This patient recovered without sequelae and was discharged from the hospital on postoperative day 3. In another patient, thrombophlebitis occurred at an antecubital intravenous site. He was treated conservatively and was discharged home on postoperative day 6. Two patients developed severe flank pain secondary to obstruction after ureteral stent removal. The obstruction was due to ureteral edema induced by balloon calibration of the UPJ at the time of ureteral stent removal. This maneuver is no longer performed. Results The results of thirty consecutive laparoscopic pyeloplasties performed at Johns Hopkins University were recently published. 4 With more than 1 year follow-up the success rate was 97% (29 of 30). The average operative time was 4.4 hours. Estimated blood loss was minimum (mean = 102.5 cm 3), with no patient requiring a transfusion. Parenteral narcotic requirement was 37.7 mg morphine sulfate, and the hospital stay averaged 3.4 days. Early laparoscopic pyeloplasty results have demonstrated equivalent success rates to open pyeloplasty while significantly decreasing the postoperative morbidity. 2 Despite this encouraging short-term success, long-term follow-up needs to be assessed. With advancement in laparoscopic suturing techniques and increased operator experience, the operative time has decreased. Despite these advances, a high level of expertise is required in both laparoscopy and intracorporeal suturing and may limit the use of this procedure in general urology centers. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. Adams JB, Schulam PG, Moore RG, Partin AW, Kavoussi LR. New laparoscopic suturing device: initial clinical experience. Urology 1995;46:242. Brooks JD, Kavoussi LR, Preminger GM, Schuessler WW. Comparison of open and endourologic approaches to the obstructed ureteropelvic junction. Urology 1995;46:791. Chen RN, Moore RG, Kavoussi LR. Laparoscopic pyeloplasty. J Endourology 1996;10:59. Moore RG, Averch TD, Adams JB, Schulam PG, Chen RN, Kavoussi LR. Laparoscopic pyeloplasty: experience with the initial 30 cases. J Urol. In press. Moore RG, Brooks JD. Ureteropelvic junction obstruction: assessment of minimally invasive therapies. Contemp Urol 1995;7:47. Nakada JY, McDougall GM, Clayman RV. Laparoscopic pyeloplasty for secondary ureteropelvic junction obstruction: preliminary experience. Urology 1995;46:257.

Chapter 129 Laparoscopic Bladder Neck Suspension Glenn’s Urologic Surgery

Chapter 129 Laparoscopic Bladder Neck Suspension
J. Stuart Wolf, Jr. and Elspeth M. McDougall

J. S. Wolf, Jr.: Section of Urology, Ann Arbor Veterans Affairs Medical Center, and University of Michigan Hospital, Ann Arbor, Michigan 48109–0330. E. M. McDougall: Washington University Medical School, St. Louis, Missouri 63110.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Dilation of Retropubic (Extraperitoneal) Space Port Placement and Creation of Pneumoretroperitoneum Exposure of the Endopelvic Fascia and Cooper's Ligament Bilaterally Endopelvic Fascia Stitch Cooper's Ligament Stitch Securing the Suture at Cooper's Ligament Cystoscopy Exiting the Operative Site Postoperative Care Outcomes Complications Results Chapter References

The controversies regarding the management of anatomic stress urinary incontinence (stress urinary incontinence types I and II) are reflected in the large number of surgical interventions that have been devised. These procedures generally have as their goal the replacement of the bladder neck and proximal urethra to their proper and well-supported retropubic position. Until recently, operations for anatomic stress urinary incontinence could be divided into the open retropubic repairs (e.g., Marshall-Marchetti-Krantz and Burch) and the vaginal repairs (e.g., Pereyra with modification by Raz, Stamey, and Gittes). With the introduction of laparoscopic bladder neck suspension in 1991 by Vancaillie and Schuessler, a third group of operations has emerged. 10 The laparoscopic approach to stress urinary incontinence attempts to combine the superior effectiveness of the retropubic procedures with the relatively lower morbidity and higher degree of patient acceptance associated with the vaginal procedures. 1,2,4

DIAGNOSIS
Incontinence in women can be due to several causes. As bladder neck suspension is intended only for the treatment of patients with uncomplicated urethral hypermobility associated with stress urinary incontinence (types I and II), proper diagnosis is paramount. Physical examination is performed with attention to reproducing the incontinence, evaluating the pelvic musculature, and examining for evidence of neurologic disease. The primary conditions that need to be ruled out are intrinsic sphincter dysfunction (stress urinary incontinence type III) and a detrusor cause for the incontinence. Intrinsic sphincter dysfunction is suggested by an abdominal leak point pressure less than 60 mm Hg. Values between 60 and 100 mm Hg are equivocal. Values greater than 100 mm Hg indicate that intrinsic sphincter dysfunction is not present. 6 The most commonly used test to evaluate for a detrusor component to the incontinence is a cystometrogram, although a video urodynamic study appears to be more accurate.6

INDICATIONS FOR SURGERY
Patients with urethral hypermobility associated with stress urinary incontinence, but without intrinsic sphincter dysfunction, are candidates for laparoscopic bladder neck suspension. Early in a surgeon's experience, it would be prudent to select only patients with primary incontinence who have not undergone a previous pelvic operation. Although a prior vaginal urethropexy does not significantly increase the difficulty of a laparoscopic repair, prior retropubic bladder neck suspension can produce marked scarring. Such patients should be avoided early in one's experience. The issues of when to perform laparoscopic bladder neck suspension with concomitant vaginal procedures or the efficacy of addressing other pelvic prolapse problems (i.e., rectocele and enterocele) laparoscopically are controversial.

ALTERNATIVE THERAPY
Alternative treatments for stress urinary incontinence types I and II include the following: 1. 2. 3. 4. 5. 6. Pelvic strengthening exercises Sympathomimetic drug therapy Nonsurgical urethral or bladder neck prostheses Vaginal urethropexy Vaginal sling procedure Open bladder neck suspension

SURGICAL TECHNIQUE
General anesthesia with endotracheal intubation is recommended because good relaxation of the tissues facilitates the operative manipulations. The head-down tilt position predisposes regurgitation of gastric contents. Pulse oximetry and carbon dioxide capnography are routinely used, but invasive hemodynamic monitoring is generally not necessary. Antithrombotic stockings are applied and the patient is securely affixed to the operating table in a low lithotomy position. A naso- or orogastric tube is inserted (this will be removed immediately postoperatively) and a Foley catheter placed into the bladder. Figure 129-1 illustrates the operating room setup for a right-handed surgeon.

FIG. 129-1. Operating room setup for laparoscopic bladder neck suspension by a right-handed surgeon.

Although the first report of laparoscopic bladder neck suspension described a Marshall-Marchetti-Krantz procedure, most urologists performing laparoscopic bladder neck suspension have reported using laparoscopic variations on a Burch colposuspension. 10 An extraperitoneal version of this technique is used at Washington University in St. Louis, as described in this section. Dilation of Retropubic (Extraperitoneal) Space Details on the extraperitoneal approach are provided in Chapter 122. A 1-in. transverse incision is made in the midline, halfway between the umbilicus and symphysis pubis. The subcutaneous tissue spreads down to the anterior rectus fascia with a clamp. Absorbable 0 Vicryl sutures are placed transversely at the superior and inferior aspects of the exposed fascia, and the fascia is incised transversely between the sutures with a no. 12 blade. A finger is then slid under the anterior fascia, down to the symphysis pubis, and a space is digitally dissected behind the symphysis pubis. A dilating balloon is created by tying the middle finger of a size 8 sterile surgical glove onto the tip of a 16-Fr straight red rubber catheter with two strands of 0 silk suture. After generously lubricating the balloon with surgical lubricant, the balloon catheter is backloaded into a 30-Fr Amplatz sheath. The sheath/balloon catheter assembly is directed through the fascial incision and behind the pubis. Once the tip is in position the sheath is retracted, leaving the balloon in the retropubic space. The balloon is inflated with 1 L of normal saline dispensed from a 60-cm 3 syringe (Fig. 129-2). This is simplified by placing 1 L of fluid into a bowl and injecting the fluid until the bowl is empty. The fluid is aspirated with a laparoscopic irrigator/aspirator instrument and the balloon catheter removed.

FIG. 129-2. Dilating balloon filled with 1 L of saline in the retropubic space. (Redrawn from Smith AD, Bagley DH, Badlani GH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996.)

Port Placement and Creation of Pneumoretroperitoneum A Hasson-type cannula is placed into the retropubic space and secured with the two previously placed fascial sutures. Carbon dioxide is insufflated to a pressure of 10 to 15 mm Hg and the retropubic space is visualized with a 10-mm, 30 degrees laparoscope. There is usually some degree of blood staining but no active bleeding should be seen. The posterior symphysis pubis and bladder, with the Foley catheter balloon at the bladder neck, can be readily identified at this point. Two secondary ports are then inserted—a 12-mm port at the level of the primary port and a 5-mm port 2 cm above the symphysis— both on the left lateral rectus border ( Fig. 129-3). It is important to identify the left inferior epigastric vessels and position the 5-mm port medial to these vessels. (N.B. This description is for a right-handed surgeon. During suture placement, the dominant hand operates the laparoscopic instrument and a finger of the non-dominant hand can be placed in the vagina.)

FIG. 129-3. Port placement for laparoscopic bladder neck suspension by a right-handed surgeon (see text for description). (Redrawn from Smith AD, Bagley DH, Badlani GH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996.)

Exposure of the Endopelvic Fascia and Cooper's Ligament Bilaterally With the laparoscope in the midline port, the laparoscopic scissors attached to the electrocautery cord are inserted through the lateral 12-mm port. A dissecting grasper goes through the 5-mm port. Using a combination of blunt and sharp dissection, with judicious use of electrocautery, the shiny white endopelvic fascia is exposed on either side of the junction of the proximal urethra and bladder neck ( Fig. 129-4). Care should be taken not to disturb tissues in the midline because this is unnecessary and risks bleeding from damage to the dorsal venous complex. The index finger of the nondominant hand is placed in the vagina. The tissues lateral to the bladder neck are elevated, identified by the balloon of the Foley catheter, to assist in exposing the endopelvic fascia for suture placement. Additionally, dissecting down onto the tension provided by the surgeon's own finger greatly facilitates the displacement of tissue. An area of endopelvic fascia approximately 2 × 2 cm is clearly defined.

FIG. 129-4. Exposure of the endopelvic fascia lateral to the bladder neck. (Redrawm from Smith AD, Bagley DH, Badlani GH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996.)

Cooper's ligament is then exposed on both sides. This is facilitated by turning the 30 degrees laparoscope upside down. The landmark is the pubic tubercle, which is

determined with gentle probing. The firm ligaments coursing laterally from this landmark lead to Cooper's ligament and the surgeon should define an area 1 cm high × 2 cm wide. Endopelvic Fascia Stitch On the free end of a 0 braided nonabsorbable suture on a rounded needle [we use 0 Ethibond polyester coated with polybutilate (Ethicon, Cincinnati, OH) on an SH needle], a small loop is tied 6 in. from the needle. This is most easily performed by tying a series of square knots to fix a loop on the tip of a mosquito clamp. The diameter of the loop should be approximately twice that of the needle. After cutting off the excess suture to leave a 1-cm tail beyond the loop, the suture is grasped 2 cm behind the needle with a laparoscopic needle holder that has been placed through a suture introducer. The needle holder is pulled back into the introducer, so that the needle is contained within the introducer, and the entire assembly is passed down the lateral 12-mm port. The needle and suture are advanced and the needle is passed to grasping forceps placed through the 5-mm port. The needle holder and suture introducer are removed and the needle holder reinserted through the same port without the introducer. The needle is grasped at the midpoint with the needle holder and locked into place such that it is at right angles to the needle holder in both the vertical and horizontal planes. With a finger of the non-dominant hand in the vagina to elevate the paravaginal tissues just lateral to the bladder neck, a generous bite of endopelvic fascia is taken lateral to the junction of the proximal urethra and bladder neck ( Fig. 129-5). The orientation of the stitch is not critical, but a full thickness of tissue (excluding vaginal mucosa) is necessary. Once the needle tip is visible, the nondominant hand is removed from the vagina and grasping forceps or second needle holder inserted through the 5-mm port. The tip of the needle is grasped, the first needle holder is moved back to the rear end of the needle, and the needle is rotated forward. This should expose enough of the needle to allow it to be rotated out of the tissue to complete the stitch. The suture is pulled through until the end loop is only a few centimeters from the tissue. Then, the suture is anchored by passing the needle through the loop. The endopelvic stitch is completed by repeating the placement of the needle and suture through the endopelvic fascia lateral to the bladder neck ( Fig. 129-6).

FIG. 129-5. With a finger from the surgeon's non-dominant hand elevating the paravaginal tissues, a curved needle is rotated through a full thickness of this tissue (excluding vaginal mucosa). (Redrawn from Smith AD, Bagley DH, Badlani GH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996.)

FIG. 129-6. A second suture is placed in the paravaginal tissue. (Redrawn from Smith AD, Bagley DH, Badlani GH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996.)

Cooper's Ligament Stitch Again, it is important to have the needle secured at a right angle to the needle holder for placement of the suture in the ipsilateral Cooper's ligament. The needle is passed just under the surface of Cooper's ligament by rotating the needle carefully. Torquing against this structure will cause the needle to bend. The needle is rotated through the tissue until the tip can be secured in the grasping forceps and rotated out of the tissue. Once the needle has been retrieved, it is dropped into the pelvic basin. The procedure for suturing the endopelvic fascia to Cooper's ligament is repeated on the contralateral side. Securing the Suture at Cooper's Ligament With the sutured portion of the paravaginal tissues elevated by the assistant with a finger in the vagina to relieve tension, the suture is pulled through Cooper's ligament until taut. The paravaginal tissue will not be touching Cooper's ligament, but it will be fixed in this position by the firm attachments of the suture. Laparoscopic knot tying onto a rigid structure such as Cooper's ligament is difficult. Because of this, we employ a polydiaxanone suture clip (Lapra-Ty, Ethicon, Cincinnati, OH) to secure the suture. The Lapra-Ty suture clip locks onto the suture tightly, obviating the need for intracorporeal knot tying. With the suture held snugly by grasping forceps placed through the 5-mm port, the 10-mm Lapra-Ty clip applier is introduced with loaded clip through the lateral 12-mm port. The suture is positioned in the back V of the clip and is slid down to the point where the suture exits Cooper's ligament. The handle of the clip applier is squeezed to securely lock the clip on the suture (Fig. 129-7). At this point, a second bite of Cooper's ligament with the needle and suture is obtained. After passing the needle for a second throw through Cooper's ligament, the end is secured with a Lapra-Ty clip. The contralateral suture is closed in the same manner to complete the bladder neck suspension ( Fig. 129-8).

FIG. 129-7. A suture incorporates paravaginal tissue and Cooper's ligament on both sides. With the assistant elevating the bladder neck, a Lapra-Ty clip is applied to fix the suture at Cooper's ligament. (Redrawn from Smith AD, Bagley DH, Badlani GH, et al. Smith's textbook of endourology. St. Louis: Quality Medical Publishing, 1996.)

P>FIG. 129-8. Completed repair with well-supported bladder neck.

Cystoscopy Indigo carmine is administered intravenously as the suture placement is being completed. The Foley catheter is removed and a flexible cystoscope inserted in the bladder. After inspection of the bladder to confirm no transgression of the bladder or urethra by suture material, the cystoscope is withdrawn to the bladder neck to confirm that it has been well supported (the light from the cystoscope is easily visible laparoscopically, even with the laparoscopic light source on maximal illumination). Both ureteral orifices should be observed to confirm efflux of blue urine. The cystoscope is removed and the Foley catheter replaced. Exiting the Operative Site After reducing the insufflation pressure to 5 mm Hg and inspecting the operative site for hemostasis, the two lateral ports are removed under direct vision. With the assistant holding fingers over these port sites to maintain the pneumoretroperitoneum, the laparoscope is withdrawn to the tip of the midline cannula and the fixation sutures released. The cannula and laparoscope are slowly pulled out, as a unit, inspecting the tract for hemostasis as the equipment is removed. The midline fascial defect is closed by tying together of the fixation sutures. Being extraperitoneal, neither of the lateral port sites require fascial closure. The skin incisions are reapproximated with 4-0 absorbable subcuticular stitches and sterile tape. Postoperative care The Foley catheter is removed the following morning and a postvoid residual is obtained after the first voiding. If the residual is less than 90 cm 3, then the patient is discharged without the catheter. If, however, the postvoid residual is more than 90 cm 3, then the patient is taught self-administered intermittent catheterization and discharged. Catheterization must be performed 4 times daily until the postvoid residual is less than 90 cm 3. Patients usually leave the hospital the morning after surgery, although in elderly patients an extra day in the hospital may be necessary if intermittent self-catheterization is required. Surgery on an outpatient basis is considered in young, healthy patients because the Foley catheter can either be removed by the patient at home or by a nurse in the physician's office. We have found postoperative analgesics with oral medication to be sufficient for most patients. Oral antibiotics are continued for 5 to 7 days postoperatively.

OUTCOMES
Complications Potential complications of laparoscopic bladder neck suspension include: 1. 2. 3. 4. 5. Physiologic complications associated with the anesthetic or use of gas insufflation (cardiac, pulmonary, venous gas embolism, etc.) Complications due to laparoscopic/surgical manipulation (bleeding, nerve injury, visceral injury including bladder injury, etc.) Wound complications (infection, skin separation, dehiscence, hernia formation) Prolonged urinary retention Failure of the procedure to correct incontinence

In the series illustrated in Table 129-1, complications occurred in 12% of the laparoscopically treated patients, 20% of the open group, and 3% of the Raz group (not including prolonged urinary retention and failure of the procedure). 3,5,7 In the series of McDougall and associates, the only two complications were a bladder laceration that occurred during the balloon dilation of the retropubic space and a pelvic hemorrhage requiring a 2-unit blood transfusion. 5

TABLE 129-1. Three studies comparing laparoscopic Burch colposuspension to contemporaneous Raz vaginal needle suspension and/or open Burch colposuspension

Results In terms of surgical efficacy, the short-term (less than 1 year) success rates of laparoscopic bladder neck suspension are promising (82% to 100%). Only three reports to date describe patients with a mean follow-up in excess of one year (all three utilizing a laparoscopic Burch colposuspension). The cure rates range from 71% to 85%, with mean follow-ups from 17 to 24 months.5,7,8 Since approximately one-quarter of treatment failures occur more than 2 years after anti-incontinence surgery, longer follow-up is mandatory to assess the effectiveness of this newly introduced procedure. 9 Until such long-term data become available, the most useful studies are three reports comparing laparoscopic bladder neck suspension to vaginal and/or open retropubic procedures (Table 129-1).3,5,7 Although the groups were similar within each study with regard to patient characteristics and date of operation, these investigations were nonrandomized. Thus, they may be subject to significant bias. Several differences between the groups are striking nonetheless. Compared to the open or vaginal procedures, the requirement for pain medication and hospital stay following the laparoscopic procedures was less. Although the time required to resume spontaneous voiding did not differ greatly between the laparoscopic and open groups, patients in the former group voided much sooner than did patients undergoing the Raz procedure. Not surprisingly, operative times were longest in the laparoscopic group. With limited follow-up, the success rates within each study did not differ greatly. In conclusion, the technique outlined above for laparoscopic Burch colposuspension employs principles and anatomy that are familiar to most urologists. Suturing is the most difficult aspect of the procedure. Alternatives to suturing during laparoscopic Burch colposuspension include staple fixation of mesh and the use of fibrin

tissue adhesive. In our opinion, the potential for problems associated with staples in the vaginal wall outweighs any technical expediency of this approach. Fibrin tissue adhesive for this application has been assessed only in short-term studies. For the time being, then, suturing is required for optimal results with laparoscopic Burch colposuspension. It appears that the laparoscopic approach to stress urinary incontinence produces less pain than either the open or the vaginal approach. The laparoscopic procedure offers a much shorter hospital stay than the open approach and a much shorter duration of catheterization than the vaginal procedure. Problems with operating time and complication rates will hopefully be reduced with further experience. This has been the case with most other laparoscopic procedures. The final key factor—the long-term success rate—is yet to be determined but preliminary findings are promising. Prospective randomized trials are presently being instituted. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Bergman A, Ballard CA, Koonings PP. Comparison of three different surgical procedures for genuine stress incontinence: prospective randomized study. Am J Obstet Gynecol 1989;160:1102. Bergman A, Koonings PP, Ballard CA. Primary stress incontinence and pelvic relaxation: prospective randomized comparison of three different operations. Am J Obstet Gynecol 1989;161:97. Das S, Palmer JK. Laparoscopic colpo-suspension. J Urol 1995;154:1119. Jarvis GL. Surgery for genuine stress incontinence. Br J Obstet Gynaecol 1994;101:371. McDougall EM, Klutke CG, Cornell T. Comparison of transvaginal versus laparoscopic bladder neck suspension for stress urinary incontinence. Urology 1995;45:641. McGuire EJ, Cespedes RD. Proper diagnosis: a must before surgery for stress incontinence. J Endourol 1996;3:201. Polascik TJ, Moore RG, Rosenberg MT, Rosenberg LRK. Comparison of laparoscopic and open retropubic urethropexy for treatment of stress urinary incontinence. Urology 1995;45:647. Radomski SB, Herschorn S. Laparoscopic Burch bladder neck suspension: early results. J Urol 1996;155:515. Rodriguez R, Partin AW, Mostwin JL, Kavoussi LR. Long term follow-up of surgically treated stress urinary incontinence (Abstract 810). J Urol 1995;153 (Suppl):431. Vancaillie TG, Schuessler W. Laparascopic bladderneck suspension. J Laparoendosc Surg 1991;1:169.

Chapter 130 Laparoscopic Management of Lymphoceles Glenn’s Urologic Surgery

Chapter 130 Laparoscopic Management of Lymphoceles
Blake D. Hamilton and Howard N. Winfield

B. D. Hamilton: Division of Urology, University of Utah, Salt Lake City, Utah 84132. H. N. Winfield: Department of Urology, VA Palo Alto Health Care System, Palo Alto, California 94304–1290, and Department of Urology, Stanford University School of Medicine, Stanford University Medical Center, Stanford, California 94305-5118.

Diagnosis Indications for Surgery Alternative Therapy Surgical Procedure Preoperative Preparation Operative Technique Outcomes Complications Results Chapter References

P>Pelvic lymphocele is a known complication of renal transplantation with an incidence reported between 1% and 18%. 2,5 It may also occur after 1% to 2% of pelvic lymph node dissections for staging of urologic malignancies. Many pelvic lymphoceles are small and asymptomatic, but lymphocele size and location may result in clinical sequelae ranging from minor lower extremity and genital lymphedema to loss of a transplanted kidney to sepsis and death. The laparoscopic treatment of pelvic lymphoceles was first reported by McCullough et al. in 1991. 7 This was a natural application of new surgical technology, providing the patient with the most effective procedure without the attendant morbidity of a laparotomy. In the ensuing 5 years, the laparoscopic technique has been reported by numerous surgeons with minor variations. It is now widely accepted as a first-line treatment for pelvic lymphoceles. It requires a modest level of laparoscopic skill and is well tolerated by the patient, generally requiring a brief (less than 24 hours) hospital stay and accompanied by a return to baseline activity within a week.

DIAGNOSIS
Lymphocele is most frequently diagnosed by ultrasound or CT scan. Occasionally, ultrasound-directed aspiration may be needed to differentiate a true lymphocele from urinoma, hematoma, seroma, or abscess.

INDICATIONS FOR SURGERY
The vast majority of lymphoceles are asymptomatic and in the absence of any specific symptoms do not require intervention. In the posttransplant patient, the lymphocele may cause oliguria, hydronephrosis, renal insufficiency, lower extremity swelling, or abdominal discomfort. After radical prostatectomy, extremity swelling or abdominal discomfort may be the presenting complaint.

ALTERNATIVE THERAPY
Traditional techniques for treating lymphoceles have distinct disadvantages. Simple needle aspiration carries minimal risk, but an unacceptable recurrence rate of 50% to 100%. Sclerosing agents (e.g., tetracycline, povidone-iodine, ethanol) improve these success rates but result in dense fibrosis around the transplanted kidney and ureter that may prove hazardous in the long term 8. External drainage requires prolonged treatment with the risk of infection, especially in the immunosuppressed patient. The most effective treatment has been open surgical exploration and marsupialization of the lymphocele into the peritoneal cavity through the transplant incision or a low midline incision. This treatment, however, is not free of morbidity, particularly in the immunosuppressed patient.

SURGICAL PROCEDURE
Preoperative Preparation With transplant-related lymphoceles, it is imperative to know the location of the ureter relative to the lymphocele in order to avoid ureteral injury during the dissection. If the serum creatinine is less than 2.0 mg/dl, a computed tomography (CT) scan with intravenous contrast should demonstrate the position of the ureter above or below the lymphocele (Fig. 130-1).

FIG. 130-1. CT scan demonstrating the close proximity of the transplant ureter (with stent in place), anterior to the lymphocele.

If the serum creatinine is greater than 2.0 mg/dl, then a 5-Fr end-hole angiographic catheter is placed in the transplant ureter with a flexible cystoscope in the office and a CT scan is obtained. The catheter demonstrates the position of the ureter and is removed after the CT scan. In addition, the patient is given a preoperative mechanical bowel preparation on the day prior to surgery, and the patient's blood is typed and screened. Operative Technique Standard transperitoneal laparoscopy preparation is used ( Chapter 122). Pneumoperitoneum is established as previously described, with a Veress needle or with the open laparoscopy technique (Hasson, Chapter 124). We prefer the Hasson technique because of the location of the transplant kidney ( Fig. 130-2). In addition, many transplant patients have had peritoneal dialysis catheters with some degree of peritonitis resulting in adhesions that increase the risk of injury with blind needle placement. Once the primary port is secured, a 10-mm, 0 or 30 degrees laparoscope is used to survey the intraperitoneal structures. Two additional ports are placed in the midline or contralateral quadrant with the exact position determined by the size and location of the lymphocele.

FIG. 130-2. Suggested port placement for laparoscopic drainage of posttransplant lymphocele.

Localization of the lymphocele is clearly the most critical and often the most difficult part of the operation. If the lymphocele is large enough, it can be readily seen bulging into the peritoneum. Smaller lymphoceles, or those hidden deep in the pelvis, may be more difficult to locate. Several maneuvers are useful to assist with identification of the lymphocele. If the Foley catheter is kept on the sterile field, the bladder may be filled and emptied repeatedly so that the bladder and an adjacent lymphocele can be identified laparoscopically. Laparoscopic ultrasonography with a 5- to 7.5-MHz probe (if available) may identify the lymphocele. A previously placed pigtail catheter, or a percutaneous needle placed into the lymphocele intraoperatively with external ultrasound guidance, may be used to empty and fill the lymphocele with dilute indigo carmine. These techniques, alone or in combination, should facilitate identification of the lymphocele while minimizing inadvertent injury to adjacent structures. Once the lymphocele is identified, a 5-mm laparoscopic needle is used to aspirate the fluid and confirm its identity. This fluid should be typical, straw-colored lymph, although it may be blood-tinged. The wall of the lymphocele is now grasped and incised with a cautery hook or laparoscopic scissors. A window is created by excising a substantial portion of the lymphocele wall that abuts the peritoneum. The size of this window is dictated by the size and position of the lymphocele, but should include most of the visible wall. The edges of the cut lymphocele wall are meticulously cauterized, and internal loculations are thoroughly disrupted. As the window is extended, great care must be taken to avoid injury to the ureter and bladder. In general, the peritoneal window should be extended laterally. With video magnification, the ureter may actually be more easily avoided. If the course of the ureter seems problematic, as judged by preoperative images, a temporary stent placed at the beginning of the case may help preserve the ureter. An omental flap may be brought through the peritoneal window to reduce the likelihood of postoperative recurrence and to minimize the risk of internal herniation of a bowel loop. The lower free edge of the omentum is gently drawn down to the lymphocele site. Any adhesions are divided for full mobilization. The omental flap is pushed through the peritoneal window and fixed to the free peritoneal edges with clips. If no omentum is available for this purpose, an alternative method of fixation is to place 3 to 5 interrupted sutures through the cut peritoneal edges and lymphocele wall. External drains are not normally placed. Free fluid is evacuated and the abdomen is exited in the standard fashion. We prefer to close the fascia of any sites accommodating 10-mm or larger ports. The Carter-Thomason port closure device (Inlet Medical, Minneapolis, MN) can achieve a rapid and secure closure with 0 polydioxanone sutures. The skin is approximated with an absorbable subcuticular suture and the incision is covered with a transparent, permeable dressing. The patient is admitted for a brief (usually less than 24 hours) hospital stay.

OUTCOMES
Complications There are scattered reports of ureteral, bladder, and other injuries. In Gill's series, there was one complication—a bladder injury—that was repaired laparoscopically without further sequelae. 3 Winfield et al. reported two complications that required open surgical repair: a bladder injury and a transplant ureter transection. 9 Gruessner et al. attempted 14 procedures and converted to open surgery in 5 patients. 4 Two patients with initially successful laparoscopic drainage required open laparotomy 3 and 12 weeks later, respectively. Gruessner et al. had no inadvertent injuries. Therefore, they concluded that laparoscopic drainage of lymphoceles is safe but technically more difficult if the lymphocele is posterior and inferior to a transplanted kidney. Results Many groups have reported their results with small numbers of patients. 1,5,6 There are few larger series. Gill et al. reported on three cohorts of patients with lymphoceles treated by laparoscopic or open marsupialization. 3 The latter was group divided into a contemporary and an historical group ( n = 12, 12, and 13, respectively). There were no recurrences in the laparoscopic group (4 and 3 in the other two groups). These patients experienced the expected quick discharge and return to activity. In only 1 of the 12 laparoscopic patients was an omentoplasty performed. Winfield et al. reported on 11 patients with laparoscopic marsupialization. One patient had recurrence at 3 months and required a repeat laparoscopic drainage procedure, which was successful. All of their patients were recurrence-free at 6 to 27 months follow-up. Laparoscopic drainage of a pelvic lymphocele is an excellent application of new technology, providing the patient with the most effective treatment while causing minimal morbidity. The procedure can be performed reliably and safely by most laparoscopists, with the reminder that localization of the lymphocele is the crux of the operation. Lymphoceles that are deep in the pelvis, or posterior and inferior to a transplant kidney, are a greater laparoscopic challenge. While some la-paroscopic procedures continue to require scrutiny, we believe that this procedure has become a first-line approach for the definitive treatment of pelvic lymphoceles. CHAPTER REFERENCES
1. Ancona E, Rigotti P, Zaninotto G, Comandella MG, Morpugo E, Costantini M. Treatment of lymphocele following renal transplantation by laparoscopic surgery. Int Surg 1991;76:261–263. 2. Braun WE, Banowsky LH, Straffon RA, et al. Lymphoceles associated with renal transplantation: a report of fifteen cases and review of the literature. Am J Med 1974;57:714. 3. Gill IS, Hodge EE, Munch LC, Goldfarb DA, Novick AC, Lucas BA. Transperitoneal marsupialization of lymphoceles: a comparison of laparoscopic and open techniques. J Urol 1995;153:706–711. 4. Gruessner RW, Fasola C, Benedetti E, et al. Laparoscopic drainage of lymphoceles after kidney transplantation: indications and limitations. Surgery 1995;117:288–295. 5. Kay R, Fuchs E, Barry JM. Management of postoperative pelvic lymphoceles. Urology 1980;15:345. 6. Khauli RB, Stoff JS, Lovewell T, Ghavamian R, Baker S. Post-transplant lymphoceles: a critical look into the risk factors, pathophysiology and management. J Urol 1993;150:22–26. 7. McCullough CS, Soper NJ, Clayman RV, So SSK, Jendrisak MD, Hanto DW. Laparoscopic drainage of a post-transplant lymphocele. Transplantation 1991;51:725–727. 8. Teruel JL, Escobar ME, Quereda C, Mayayo T, Ortuno J. A simple and safe method for management of lymphoceles after renal transplantation. J Urol 1983;130:1058. 9. Winfield HN, Hochstetler J, Terrell RB, Brown BP. Laparoscopic marsupialization of post-renal transplant lymphoceles (Abstract). J Urol 1996;155:657A.

9

Chapter 131 Laparoscopic Management of the Impalpable Undescended Testicle Glenn’s Urologic Surgery

Chapter 131 Laparoscopic Management of the Impalpable Undescended Testicle
Gerald H. Jordan

G. H. Jordan: Department of Urology, Eastern Virginia Medical School, Norfolk, Virginia 23510.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Diagnostic Laparoscopy Laparoscopic Orchiopexy Outcomes Complications Results Chapter References

An undescended testis is defined as a testis that resides in an extrascrotal position. Those testicles can be either truly maldescended or ectopic. Maldescended testicles would imply that the testicle is arrested some place along a normal path from its origination at the nephrogonadal ridge to its target, the scrotum. An ectopic testicle implies that somewhere along the path of descent the testicle is led to a position that is not truly in the line of normal descent. Undescended testicles may be palpa-ble or impalpable. A palpable undescended testicle can either be retractile or truly undescended within the canal or in an extraabdominal ectopic position. Impalpable testicles are truly undescended (high cannicular or intraabdominal location), ectopic, (abdominal), or absent. 8,10,11

DIAGNOSIS
The diagnosis of undescended testicle can be made by physical examination, imaging techniques, provocative hormonal tests, and laparoscopy. By definition, physical examination fails to show any testicle in the impalpable testis, but may be of help in diagnosing ectopic, retractile, or testes within the inguinal canal. Venography shows one the location of the pampiniform plexus only. The determination, of whether there is associated gonadal tissue or not, is not made. Ultrasound in some cases images the undescended testicle; however, many an orchiopexy has been undertaken when an ultrasound has shown an “inguinal gonad” that turns out to be a lymph node. Although magnetic resonance imaging (MRI) is better, it is not as accurate as laparoscopy. There are some who in the face of the findings of bilateral impalpable testicles would proceed with a challenge of b-human chorionic gonadotropin (HCG) in order to see if a response in the serum testosterone is noted. A major problem has been the reports from a number of centers advising varying challenge protocols, the varying protocols inevitably arising from those cases in which a given protocol was utilized, “no response noted,” and yet undescended gonadal tissue was eventually found. Thus the question remains, Does the HCG challenge diagnose with complete accuracy all cases of bilateral testicular loss? Many, myself included, do not feel that an absent response with HCG stimulation obviates the need for further investigation in these cases. Laparoscopy has been found useful for both the diagnosis of the impalpable undescended gonad and the management of the impalpable undescended gonad, especially in children. 2,4,6,7 and 8 Diagnostic laparoscopy has been found to consistently yield a definitive diagnosis, to be useful in the assessment of the testicular mobility, and has been found by many to be useful in the planning of the surgical approach. While laparoscopy is invasive and requires general anesthesia, the advantages by most are felt to far outweigh the disadvantages, particularly now that the laparoscopy can be part of the management, as opposed to just the preparation. The goal of diagnostic laparoscopy is to determine if there is gonadal tissue in the clinical setting of an impalpable testis and whether the gonad is suitable for orchiopexy or better removed. Any diagnostic test that is utilized for the diagnosis of the impalpable undescended testicle must uniformly fulfill these goals; and, to my knowledge, only laparoscopy uniformly accomplishes these goals. 4,12

INDICATIONS FOR SURGERY
The indication for diagnostic laparoscopy is the finding on physical examination of an impalpable undescended gonad, either unilateral or bilateral. The goal of therapeutic laparoscopy for the undescended testicle is either removal of the undescended gonad or permanent fixation of the testis in the scrotum. Hormonal therapy has promoted testicular descent in some cases. To my knowledge, the use of hormonal therapy in an attempt to promote descent in no way complicates either open orchiopexy or laparoscopic orchiopexy. The indications for orchiopexy are as follows: (a) possible improved fertility; (b) relocation of the testis to a site amenable for examination; (c) correction of associated hernias; (d) prevention of testicular torsion; and (e) alleviation of psychological trauma resulting from an empty hemiscrotum. Some of these goals are aimed at the removal or arrest of the histologic abnormalities encountered in undescended gonads. 5,10 The indications for laparoscopic orchiopexy are the same as for orchiopexy in general. As already mentioned, the procedure has been found effective, in one modification or another, for the management of all testicles from high cannicular, associated with hernias, to high abdominal. However, the higher the testicle, the more profound the anatomic aberrancies. The issue then surrounds the advocacy of performing orchiopexy for the severely dysmorphic abdominal testicle. That issue has been extensively argued and will continue to be; however, the development of sperm aspiration techniques associated with various applications of assisted fertilization seems to favor a try at orchiopexy. Timing of orchiopexy is important. It is now clear that spontaneous descent of testicles can occur during the first year of life. However, after the first year of life, it would appear that permanent histologic changes can occur and, optimally, orchiopexy should be performed at or just before the first birthday. This approach maximizes the opportunity for those few testicles that will descend during the first year of life, while preventing the histologic changes that occur in those testicles that remain maldescended beyond the first year of life. We have adopted the following algorithm, which illustrates the approach to management of the impalpable undescended testicle ( Fig. 131-1).

FIG. 131-1. Working protocol for management of the impalpable undescended testicle incorporating laparoscopy.

A variant of the abdominal testicle has been found and has been designated the medial ectopic abdominal testicle. These testicles during descent come to rest medial to their respective medial umbilical ligament. Associated with these testicles are readily apparent vascular gubernacular structures that extend to the respective location of a normal internal ring, usually closed, i.e., no patent processus vaginalis. The vas deferens is quite short; most of these testicles have not been noted to

have looping vas or dissociation of the paratesticular tubular structures; and, by definition, the spermatic vessel leash is short. These testicles appear to be quite “ovarian.” The prominent gubernacular vessels are very similar in appearance to a round ligament. There is often a very prominent peritoneal fold, reminiscent of the broad ligament. The testicles, as opposed to having a vertical orientation, appear to have a horizontal orientation. While these testicles can be placed in the scrotum using laparoscopic techniques, such testicles are very difficult to correctly place in the scrotum. Because they are already medial to the obliterated umbilical artery, the advantages of medial transposition are negated. Because the testicles are not associated with looping of the vas, the advantages of spermatic vessel division, either primary or staged, are likewise negated. If unilateral, I would consider orchiectomy as opposed to orchiopexy. If associated with bilateral maldescensus, aggressive mobilization can be attempted. Another option to the case of bilaterality would be the microscopic free transfer of these testicles.

ALTERNATIVE THERAPY
A number of surgical approaches have been used for the impalpable undescended testicle: inguinal (extended inguinal), primary open abdominal, staged laparoscopic first stage and open second stage, staged laparoscopic (both stages), and primary laparoscopic. Laparoscopic orchiopexy, either primary or staged, should be viewed as a minimally invasive alternative to primary abdominal orchiopexy or extended inguinal orchiopexy. Intraabdominal testicles occur anywhere in the retroperitoneum; however, they are most commonly found just inside the inguinal ring. For the higher testicle, the staged approach has been employed. Division of the spermatic vessels to “aid” with orchiopexy was advocated as early as 1903 by Bevan. 1 The staged approach based on collateralization along the long loop of the vas deferens was an extension of the technique described by Fowler and Stephens. Originally, a long looping vas was felt to be a prerequisite for the Fowler-Stephens procedure; however, intraabdominal testicles with non-long looping vas deferens have been successfully addressed. The staged approach has been likened to other forms of delay, a term in the reconstructive literature that implies the transposition of tissue at a first stage, with division of the axial blood supply at a second stage, thus allowing the transfer tissue to survive on random collateralization occurring between the first and second stages. Clearly, the staged orchiopexy does not accomplish this. Instead, the axial blood supply or at least part of it is divided at the first stage. There remains a second axial blood supply, but one is led to believe that the second blood supply enhances during the period between first stage and second stage. Whether this enhancement truly occurs has recently been called to question by Koff and Associates. 9 Clinically, most would agree that the paravasal blood supply, after division of the spermatic vessel leash and a waiting period, does appear to be more prominent. After a 6-month wait following the initial ligation of the spermatic vessels, using either laparoscopic techniques or an open technique, the testicle is brought to the scrotum based on the paravasal vascular supply. Some authors have suggested that testicles that have been managed with staged division of the spermatic vessel leash be evaluated with color Doppler in an effort to evaluate the viability of the testicle prior to undertaking the second stage. Realistically, however, even if by Doppler the testicle does not appear to be viable, the remaining gonadal tissue should be removed as atrophy by ultrasound criteria cannot be felt to be pathognomonic of complete atrophy of the gonadal tubular structures. Ostensibly, these structures could continue to carry the risk of malignant transformation.

SURGICAL TECHNIQUE
Diagnostic Laparoscopy A number of approaches have been used with regard to diagnostic laparoscopy for the undescended testicle. The most commonly employed approach is to place a cannula at a supra- or an infraumbilical position, and to visualize the pelvis looking for the spermatic vessels and the vas deferens. In most cases, diagnostic laparoscopy per se requires the placement of a single cannula only. This cannula can be quite small with 4- or 5-mm cannulas readily available, and cannulas as small as 2.5 mm available from some manufacturers. The size of the initial access cannula depends on the intent of the surgeon (i.e., diagnostic laparoscopy followed by open surgery versus followed by laparoscopic surgery to correct the defect). In some cases, it is also useful to have a manipulating instrument such as a Veress needle; other small probes have been used for this purpose that do not require a formal secondary cannula placement. The umbilical cannula position allows for complete examination of the abdomen, in particular detailed examination of the pelvis. In the case of the unilateral impalpable undescended testicle, the exam can begin on the side of the normally descended testicle. Those vessels should be readily visualized. If there is difficulty locating them, traction on the descended testicle causes the cord structures to move and very easily defines their location. Attention is then switched to an identical location on the side of the impalpable testicle. Potential findings include the following: 1. A normal-appearing vas and vessels merging at the position of a closed internal ring ( Fig. 131-2)

FIG. 131-2. The laparoscopic appearance of a normal right groin; the spermatic vessel leash can be seen joined by the vas deferens passing through a closed internal ring. Traction is on the testicle, emphasizing the location of the ring.

2. Normal vas and vessels merging at an open internal ring, i.e., patent processes or hernia ( Fig. 131-3)

FIG. 131-3. The laparoscopic appearance of the right groin in which there is a patent processus vaginalis (hernia).

3. Normal vas and vessels merging at an open internal ring which, with palpation of the groin, reveals an emerging or peeping high cannicular testicle ( Fig. 131-4)

FIG. 131-4. (A) Laparoscopic appearance of the right groin. The spermatic vessel leash joined by the vas can be seen passing adjacent to the open internal ring (patent processus vaginalis). (B) With gentle pressure on the groin, the testicle can be seen delivered to the abdomen. Notice the bubbles created by the insufflation mixed with peritoneal fluid.

4. A true low abdominal testicle (Fig. 131-5)

FIG. 131-5. The laparoscopic appearance of the low right abdominal testicle. The laparoscopic grasper is elevating the testicle.

5. Blind-ending spermatic vessels, often within proximity of a blind-ending vas deferens ( Fig. 131-6)

FIG. 131-6. The laparoscopic appearance of the left groin: classic blind-ending vas and blind-ending vessels in proximity to each other.

In the case of the impalpable undescended gonad with normal vessels merging at either an open or a closed ring, these findings warrant further inguinal exploration. In virtually all cases, that inguinal exploration can be accomplished from above (i.e., laparoscopically). In the case of a high cannicular testicle, laparoscopic orchiopexy has been shown to be extremely effective. For higher testicles, many would consider staged orchiopexy. 3,13 In the vast majority of cases of true low abdominal testicles, primary laparoscopic orchiopexy is most effective. Laparoscopic Orchiopexy The technique of laparoscopic orchiopexy begins with careful patient positioning on the operating table. The patient must be fixed to the table to allow the table to be manipulated through the extremes of Trendelenburg, reverse Trendelenburg, and rolled positions. The arms should be tucked at the side. Preparation and draping must be suitable for an open abdominal procedure, be it planned or necessary. The skin is prepared nipple to midthigh, table side to table side. A urethral catheter and an oral gastric tube are inserted. Children can readily lose body heat during laparoscopic procedures due to the fact that the insufflation gases are not heated and warming devices must be used. With the availability of high-intensity illumination and three-chip camera technology, virtually all pelvic surgery in children can be accomplished using 5-mm or smaller optics. The cannula placement schema for laparoscopic orchiopexy is illustrated ( Fig. 131-7).

FIG. 131-7. Illustration of the cannula placement schema for right laparoscopic orchiopexy.

I do not favor closed Veress needle access and thus initially a small transverse incision is created in the infraumbilical crease. The incision is carried sharply down to the fascia, which is identified and tagged with Vicryl sutures. These sutures allow the fascia to be elevated. The fascia is then opened and, with the use of skin hooks, the deep abdominal wall structures are elevated, eventually allowing a small peritoneotomy to be made. Through that small incision, a 5-mm cannula is placed with a screw fixation device that achieves excellent gas seal. Specialized open access cannulas are not needed. For the diagnostic portion of the procedure, an examination of the pelvis is begun. Once the inspection of the pelvis is complete, a decision is made concerning the therapeutic intervention and secondary cannulas are placed.

For an impalpable testicle that is found to be in the high inguinal region (“emerging” or “peeping”), or a testicle that is found immediately proximate to the internal ring with vessels that allow mobilization, single-stage laparoscopic orchiopexy is undertaken. In addition, primary laparoscopic orchiopexy has recently been advocated for the management of the high palpable inguinal testicle. The procedure begins with the patient in the Trendelenburg position, the table tilted away from the undescended gonad. In the case of a true abdominal testicle, a peritoneal incision is made lateral to the spermatic cord at the upper pole of the testicle and continued in a rostral direction. The spermatic cord is rolled medially and elevated from the retroperitoneal tissues using blunt dissection. Adjacent to the patent processus vaginalis, if it is present, the gubernaculum is divided ( Fig. 131-8). Care must be taken with this maneuver as a loop of the vas deferens can be seen in these structures ( Fig. 131-9), making it imperative to define the caudal extent of looping of the vas prior to dividing the gubernaculum. These gubernaculum attachments are vascular and must be adequately cauterized before being divided. These inferior attachments retract caudally when divided, and significant bleeding in the groin is possible. In the high inguinal testicle, the initial incision is made adjacent to the patent processus, thus allowing the hernia sac to be dissected from the spermatic vessels. This allows for delivery of the testicle to the abdomen and for division of the inferior attachments.

FIG. 131-8. Laparoscopic appearance and accompanying diagram of right testicle—gubernacular attachments being divided.

FIG. 131-9. Laparoscopic appearance of the left groin; testicle is retracted into the abdomen and a looping vas is seen extending adjacent to the inferior (gubernacular) attachment.

Dissection is then carried to the peritoneum overlying the vas deferens, medial to the testicle. That peritoneum is opened, and the vas is carefully elevated, taking care to avoid injury to the paravasal structures ( Fig. 131-10). The vas is mobilized sufficiently to allow placement of the testicle in the scrotum, but vigorous mobilization of the vas deferens is purposefully avoided as the incidence of testicular atrophy has been attributed by some to be associated with vigorous mobilization of the vas. The vas, however, must be sufficiently mobilized so that when the testicle is brought into the scrotum there is not tenting of the paravasal attachments next to the ureter creating ureteral obstruction.

FIG. 131-10. (A) Photograph and accompanying diagram of a right testicle, having been freed from the adjacent hernia sac, being retracted into the abdomen. (B) Diagram showing the right testicle, retracted into the abdomen, the gubernacular attachments being divided using a combination of sharp and electrocautery dissection.

The two peritoneal incisions are connected, creating a triangular shaped flap of peritoneum. This flap of peritoneum ostensibly may preserve some of the collateral vessels between distal spermatic vessels and the paravasal vasculature. Realistically, however, this is simply the easiest way to accomplish the dissection. Likewise, including this peritoneal flap does allow one the opportunity to fall back on division of the spermatic vessel leash and primary Stephens-Fowler orchiopexy. After the spermatic cord, testicle, and vas have been sufficiently mobilized ( Fig. 131-11), the surgeon's attention is directed to the respective hemiscrotum. Fixation of the testicle and scrotum can be accomplished by whatever maneuver is the surgeon's preference; I generally create a subdartos pouch. A small transverse skin incision is made and carried to the dartos fascia. The dartos pouch is created; tacking sutures of 6-0 prolene are placed on the fascia; the fascia is then opened and the canal between the scrotum and the pubic tubercle created. At that point, the fascia is penetrated with a clamp, the tips of the clamp being readily detected from within. This allows for dilation of the “fascial window,” at which point a 5-mm rod is introduced via the developed canal. As the indentation of the rod is observed, the rod is guided into the peritoneal cavity under direct vision just medial to the obliterated umbilical ligament. The rod should literally “melt its way” into the abdominal cavity. If significant pressure is required to pass the rod, one must fear that the rod could be passing through the medial structures (i.e., the bladder) or perhaps passing through the vascular structures of the umbilical ligament. With the rod in the abdominal cavity, a threaded dilator is passed, dilating the canal to 10 mm, allowing for placement of a 10-mm cannula (Fig. 131-12).

FIG. 131-11. (A) Photograph with (B) accompanying diagram of a right testicle totally mobilized from its attachments to the right groin suspended in the intraabdominal cavity. Note the peritoneal reflection between the spermatic vessels and the vas deferens.

FIG. 131-12. (A) Diagram illustrat-bing the passage of a 5-mm rod medial to the medial umbilical ligament. The canal has then been dilated with a threaded 10-mm dilator. Laparoscopic cannula is passed over the dilator. (B) The appearance of laparoscopic cannula passed through the right hemiscrotum into the abdomen.

Testicular grasping forceps are introduced. The testicle is positioned into the forceps, and the testicle is then delivered into the scrotum via the cannula ( Fig. 131-13). As the vessels or vas come under tension, the attachments can be further freed. Lucent cannulas have been found to be very useful here as these cannulas allow one to visualize the testicle as it descends to what will be its eventual position, well placed in the scrotum. When there, the grasper is removed from the testicle; the testicle is secured with the previously placed prolene sutures and placed in the dartos pouch.

FIG. 131-13. (A) Photograph showing an intraabdominal testicle being pulled into the laparoscopic cannula using testicular grasping forceps. The testicle is then delivered to the right hemiscrotum. (B) Photograph illustrating the outside appearance with the testicle delivered to the level of the scrotum within the lucent cannula.

The scrotal skin is closed and insufflation pressures are dropped; the areas of dissection are checked for hemostasis. Early in my series, I attempted to reoppose the parietal peritoneum; however, it has been found that with dutiful dissection of the hernia sac, closure of the parietal peritoneum is not necessary; the area readily reperitonealizes without recurrences of hernia found to date. 6,7 With regard to the application of laparoscopic techniques for the Fowler-Stephens procedure, Bloom 2 described the first stage occlusion of the spermatic vessels using pelvioscopic clipping. In his series, the testicle was later located using an open technique. Vascular clips can be used for ligating the cord structures. Some centers have merely coagulated the spermatic vessels. With the advent of better operative laparoscopic procedures, this center, along with others, has performed the second stage laparoscopically. In terms of technique, the second-stage staged laparoscopic orchiopexy is not unlike the procedure already described for primary orchiopexy. Obviously, the extended dissection of the spermatic vessels is not required. It cannot be overemphasized that the undescended testicle is found proximate to the blind-ending vessels, not necessarily the blind-ending vas. If one finds a blind-ending vas deferens, only in the pelvis, one cannot declare the testicle “vanished” without further exploration and the findings of blind-ending vessels. The author recently encountered a case that caused him to question the certainty of the previous statement; however, in that case, a patient was found to have a normal-appearing vas deferens passing through an open internal ring. Merging with that vas deferens were structures that appeared to be normal spermatic vessels. On dissecting these structures during an attempt at orchiopexy, the hernia sac was dissected, the vessels and vas deferens were dissected, but a true undescended gonad was not found. Instead there was found only a prominence associated with the vas deferens/epididymis that was felt to possibly represent viable gonadal tissue. Fortunately, the decision was made to continue orchiopexy; and as these structures were freed up, beneath the cecum a testis was found. It is clear that the true spermatic vessels went to the testis. However, prominence of the vessels extending to the open ring (i.e., gubernacular attachments) were such that one could easily have been led astray by the findings of these vessels ending proximate to what ended up being only an epididymis. Had one explored via an extended inguinal approach, I believe a misdiagnosis would have been made. Likewise, had I not decided to attempt orchiopexy, the true high abdominal testicle would not have been found, at least not at that procedure. In children, all of the cannula sites, no matter how small, must be closed. At this center, closure of the cannula site is with 3-0 Vicryl suture on a strong circular needle. These children do profit from adjuvant caudal anesthesia using bupivacaine (Marcaine). Additionally, injection of the cannula sites with bupivacaine has also been found advantageous. The children are awakened, recover from anesthesia, and are discharged. In most cases, the diet is advanced rapidly. With the exception of instructions to keep the child from playing on straddle implements, virtually no physical restrictions are imposed. Most parents report that the child is “a little bit slow” on the day of surgery. If surgery is performed in the afternoon, on the following morning the child might remain a little slow. However, within 24 hours of surgery the vast majority of parents report that the child “suddenly awakens” and from that point on is seemingly unimpeded by the laparoscopic procedure.

OUTCOMES
Complications Complications have almost all been associated with blind cannula placement used in association with Veress needle insufflation. To the author's knowledge, the only

other complication encountered using diagnostic laparoscopy was a single case of gas embolism, and the details of this case are not known. The complications that have been reported with laparoscopic orchiopexy have been acute testicular atrophy, a case of hernia with bowel incarceration at the site of the closure of the parietal peritoneum, a case of bladder injury occurring at the time of passing the testicle medial to the obliterated umbilical artery, and a case of avulsion of the testicle during the orchiopexy and with that case the testicle was salvaged by converting to microorchiopexy. Results Diagnostic laparoscopy is regarded as a highly effective and safe procedure to localize and diagnose the nature of the impalpable testicle. In most series, a large number of procedures have been done with virtually 100% accuracy and, in most series, no complications. Studies by both Body and Koyle 4,11,12 have shown the superior accuracy of laparoscopic “exploration” over open exploration. The literature supports diagnostic laparoscopy as the single most accurate modality for the diagnosis and localization of the impalpable undescended gonad. Our experience and the experiences of others have shown laparoscopic orchiopexy to be a valid and effective alternative to open orchiopexy ( Fig. 131-14). At this center, approximately 45 testicles have been brought to the scrotum using laparoscopic techniques, with only one case of possible acute atrophy. In that case the testicle was extremely small when brought down; however, with longer follow-up, the testicle has not shown much, if any, growth. In another case, the child developed a small hematoma in the “neoinguinal” canal that resolved without sequelae.

FIG. 131-14. Postoperative appearance at 6 months showing a patient following laparoscopic right orchiopexy for a right intraabdominal testicle. Note the complete normal appearance of the right hemiscrotum. A penny is placed on the child's abdomen for reference. Note that the laparoscopic cannula sites are virtually invisible. Drawn on the child's abdomen are what could have been the abdominal incisions for an open abdominal orchiopexy.

CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Bevan A. The surgical treatment of undescended testicle: a further contribution. JAMA 1903;41:718. Bloom DA. Two-step orchiopexy with pelvioscopic clip ligation of the spermatic vessels. J Urol 1991;145:1030. Bloom DA, Ritchey ML, Jordan GH. Pediatric peritoneoscopy (laparoscopy). Clin Pediatrics 1993;32(2):100–104. Boddy SA, Corkery JJ, Gornall P. The place of laparoscopy in the management of the impalpable testis. Br J Surg 1985;72(11):918–919. Giarola A, Agostini G. Undescended testis and male fertility. In: Bierich JR, Giarola A, eds. Cryptorchidism. London: Academic Press, 1979. Jordan GH, Bloom DA. Laparoscopic genitourinary surgery in children. In: Gomella LG, Kozminski M, Winfield HN, eds. Laparoscopic urologic surgery. New York: Raven Press, 1994;223. Jordan GH, Robey EL, Winslow BH. Laparoendoscopic surgical management of the abdominal/transinguinal undescended testicle. J Endourol 1992;6:157. Jordan GH, Winslow BH. Laparoscopic single stage and staged orchiopexy. J Urol 1994;152:1249–1252. Koff SA, Sephic PS. Treatment of high undescended testes by low spermatic vessel ligation. J Urol 1992;156(2):799–803 Kogan SJ. Cryptorchidism. In: Kelalis PP, King LR, Belman B, eds. Clinical pediatric urology. 2nd ed. Vol. 2. Philadelphia: WB Saunders, 1985;86 Koyle MA, Rajfer J, Ehrlich RM. Undescended testis. Pediatr Ann 1988;17(1):39, 42–46 Koyle MA, Pfister RR, Jordan GH, Winslow BH, Ehrlich RM. The role of laparoscopy in the patient with previous negative inguinal exploration for impalpable testis (Abstract 35). J Urol 1994;151:236A 13. Ransley PG, Vordermark JS, Caldamone AA, Bellinger MF. Preliminary ligation of the gonadal vessels prior to orchiopexy for the intra-abdominal testicle: a staged Fowler-Stephens procedure. World J Urol 1984;2:266.

Chapter 132 Laparoscopic Adrenalectomy Glenn’s Urologic Surgery

Chapter 132 Laparoscopic Adrenalectomy
H. Tazaki

H. Tazaki: Department of Urology, New York Medical College, Valhalla, New York 10595.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Instruments Preparation Anterior Transperitoneal Approach Posterior Retroperitoneal Approach Postoperative Management Outcomes Complications Results Chapter References

In urologic surgery, laparoscopic surgery is most commonly utilized for pelvic lymphadenectomy as a staging diagnostic technique in patients with prostate cancer. Indications for laparoscopy in other malignant diseases are still controversial. Though most adrenal masses are benign, surgical intervention is generally indicated whether the mass is functioning 1 or nonfunctioning. On the other hand, location of adrenal glands is deep in retroperitoneal space and surgical approach is often difficult, even by open surgery. While the laparoscopic approach to the kidney is becoming common by either intra- or extraperitoneal routes, it is now possible to use this approach to the adrenal. The advantages of laparoscopic adrenal surgery are the reduced morbidity and shortened hospital stay that has been confirmed by the number of successful cases reported since 1993. 2,3,4,5

DIAGNOSIS
The diagnosis of adrenal lesions is discussed in detail in Chapter 1, Chapter 2, Chapter 3 and Chapter 4.

INDICATIONS FOR SURGERY
The indications for surgery are any functioning adrenal mass or any adrenal mass that is larger than 2 cm.

ALTERNATIVE THERAPY
Alternative therapy is open surgery.

SURGICAL TECHNIQUE
Instruments Standard instruments for urologic laparoscopic surgery can be used for laparoscopic adrenalectomy. These instruments include trocars, clamps, scissors, and hemoclips. Laser coagulators and ultrasound aspirators are helpful for a bloodless approach to the adrenal. 8 Rigid and flexible endoscopes can be used. We prefer to remove the adrenal with an endopouch, which accommodates the adrenal following its excision and allows the tissue to be taken out through a port at the final stage of surgery. The advantage of laparoscopic adrenalectomy is that it is less invasive with lower postoperative morbidity and quicker recovery. This is balanced by the increased time required in the operating room compared to open surgery and higher cost per procedure partly due to the increased operative time. However, recent reports show a remarkable improvement in operating time by introducing new instruments designed for smooth dissection and controlling hemorrhage from small vessels. An example of the new technology is the coagulation shears (Harmonic Scalpel, Ultra Cision), which reduces the use of hemoclips resulting in shortened operating time. Preparation Laparoscopic adrenalectomy is performed under intubated general anesthesia. A combination of general anesthesia with epidural anesthesia is recommended especially for obese patients due to the prolonged anesthesia time. Positioning of patients on operating table is extremely important. The operating table must be rotatable from flat to oblique positions as well as from flat to head-up positions. Positioning the patient for surgery varies according to the approach chosen for the adre-nalectomy. These approaches are transperitoneal an-terior, extra- or intraperitoneal flank, and posterior. Except for the gasless technique to explore the operating field, a pneumoperitoneum with CO 2 gas is used to obtain the adequate operating space for handling the instruments. Monitoring of end-tidal CO 2 and arterial blood gas throughout surgery is extremely important. Laparoscopic approaches to the adrenal gland are classified as those of open surgery. Anterior and flank approaches are more common than the posterior approach. Even in these two common approaches, there are many technical variations described. Therefore, the most basic techniques are described in the following paragraphs. Anterior Transperitoneal Approach The patient is placed in supine position under general anesthesia. A blind laparotomy using Veress needle is not recommended, especially in the obese patient ( Fig. 132-1 and Fig. 132-2). An open laparotomy is made at the umbilicus by the Hasson method and the pneumoperitoneum is established with the insufflation of CO 2, as described in Chapter 122. This pneumoperitoneum creates a space for additional trocar insertion and subsequent port placement for the laparoscope and other instruments. At this time, position of the patient is changed from supine to hemilateral position with the diseased side up. Care must be taken when a trocar is inserted along the midaxillary line or its posterior side because the ascending or descending colon may be injured. It is therefore important to have the preinserted camera watching these puncture sites during trocar placement.

FIG. 132-1. Positions of trocar ports for anterior transperitoneal adrenalectomy (left side). MCL, midclavicular line; MAL, middle axillary line; (1) Umbilicus (for open laparoscopy); (2, 3) Ports for 10/12-mm trocars; (4) port for 5-mm trocar.

FIG. 132-2. Positions of trocar ports for anterior transperitoneal adrenalectomy (right side): (1) Umbilicus (for open laparoscopy. (2, 3) Ports for 10/12-mm trocars on MCL. (4) Port for 5-mm trocar on MAL.

The right side allows better exploration of the adrenal than the left. The dissection begins by pushing the fatty tissue of the omentum laterally. The line between the lower lateral end of the liver and the colon is confirmed and an incision is made at the hepatic flexure. The incision is extended from the lateral colon to the lower surface of the liver along the duodenum. Gerota's fascia is opened and the upper pole of the right kidney is exposed. As the kidney is pushed down the adrenal is visualized. The adipose tissues around the adrenal are bluntly divided from lateral to medial until the vena cava and right renal vein are in sight. The inferior adrenal vein and other small vessels are clipped and divided until the right adrenal vein is identified attached to the vena cava ( Fig. 132-3). The right adrenal vein is usually very short and it is necessary to double-clip the vein prior to division. The excised mass is placed in the endopouch and removed through a trocar port.

FIG. 132-3. Laparoscopic view of anterior transperitoneal adrenalectomy (right side). Right adrenal vein is short, draining directly into vena cava. Sweep the fatty tissues and expose the adrenal vein, which is then double-clipped.

The left adrenal is located higher and in a narrow space between the descending colon and the spleen. The anterior surface of the left adrenal is covered by the stomach and omentum and the left colonic flexure, which is higher than the right side. Therefore, selection of port positions for left adrenalectomy should be higher than for the right side. In the left lateral position, along the lateral line of the descending colon, the peritoneum is incised and the dissection is carried to the upper pole of the kidney without entering Gerota's fascia. Care should be taken not to encounter bleeding when the splenocolonic ligament is incised and divided. Thus the splenic flexure of the colon can be replaced medially as the peritoneal wall along lateral of the descending colon is incised to the level of the lower pole of the kidney. Using a Babcock clamp inserted through an 11-mm port near the midline, the colon should be kept outside of sight. Gerota's fascia is incised at the level of upper pole of the left kidney as the lower surface of the spleen is pushed cranially. The phrenic branch vessels must be carefully clipped or fulgurated. Then anterolateral surface of the adrenal is divided bluntly by pushing the tail of the pancreas in a cranioventral direction. The renal hilum is exposed and the central vein and the left renal vein are identified. The central vein is doubly clipped and divided ( Fig. 132-4). As the dissection proceeds along the cranial side of the adrenal, the left inferior phrenic vein is identified, clipped, and divided. The adrenal is placed in a pouch and removed through a trocar port. After assuring complete hemostasis, all instruments are removed. It is recommended that a Penrose drain be left in the retroperitoneal space.

FIG. 132-4. Laparoscopic view of anterior transperitoneal adrenalectomy (left side). Left adrenal vein drains into the left renal vein. The adrenal vein is double clipped and divided.

Posterior Retroperitoneal Approach

Balloon dilation of the retroperitoneal space using techniques described by Gaur and subsequent pneumo-retroperitoneum can allow an approach to the adrenal without entering the peritoneal cavity and may reduce incidence of intraperitoneal complications. 4 However, some disadvantages to endoscopic retroperitoneal surgery are less visibility due to a darker operating field and a lack of landmarks due to scattering of light against the adipose tissue in the retroperitoneal space. The patient is placed in a flank position on the table ( Fig. 132-5). A 1.5-cm skin incision is made near tip of the 12th rib at the midaxillary line and the muscle layers are bluntly separated ( Fig. 132-6). A balloon dilation trocar (Preperitoneal Distention Balloon System, Origin Medsystems, Inc., Menlo Park, CA) is inserted into the hole and the balloon is inflated with 800 ml of air, creating a space for endoscopic manipulation. The Origin balloon trocar is changed to an 11-mm trocar and the laparoscope is introduced for observation. Two more operating trocars are introduced extraperitoneally on the line along the 12th rib.

FIG. 132-5. Patient's positioning for posterior retroperitoneal adrenalectomy by the flank approach. Special care should be taken for protection of arms and legs to avoid complications.

FIG. 132-6. Positions of trocar's for posterior retroperitoneal adrenalectomy. (right side AAL, anterior axillary line; PAL, posterior axillary line. (1) Trocar for balloon dilatation in retroperitoneal space right under the 12th rib. (2, 3) On AAL and PAL for 10/12-mm trocars. (4) Port for 5-mm trocar.

Another technique for the posterior approach is to place the patient in an extended abdominal position ( Fig. 132-7) and make a small incision at the midscapular line 2 cm below the 12th rib. The fascia of quadratus lumborum muscle is bluntly opened and the balloon dilation technique is applied. Two more ports are then created, avoiding damage to the iliolumbar muscle, for a direct approach to the upper pole of the kidney and the adrenal. The pressure in the retroperitoneal space is controlled at 12 to 15 mm Hg throughout the surgery.

FIG. 132-7. Patient positioning for posterior retroperitoneal adrenalectomy via the extended abdominal position. Balloon dilation of retroperitoneal space through a port on middle scapular line 2 cm below the 12th rib.

For a right adrenalectomy, Gerota's fascia is opened at the level of the upper pole of the kidney, allowing a direct approach to the right adrenal. The advantages are an operating field not disturbed by the liver or the intestines but the trade-off is that only the upper pole of the kidney can be used as a landmark. Clipping vessels from this approach is relatively easier ( Fig. 132-8).

FIG. 132-8. Laparoscopic view of posterior retroperitoneal approach. Upper pole of the kidney is a landmark for the adrenal. Laser coagulation shears (LCS) are a useful tool.

Left adrenalectomy, direct approach to the adrenal in the retroperitoneal space, is similar to adrenalectomy of the right side, though identifying the tail of the pancreas

is often difficult. Clipping the adrenal vein is the final stage of a left adrenalectomy from the retroperitoneal approach. Postoperative Management A chest x-ray is obtained on postoperative day 1 if there are any signs of suspected pulmonary complications. Patients are discharged on postoperative day 2.

OUTCOMES
Complications Complications from laparoscopic adrenalectomy include disease-specific and technique-specific adverse events. Disease-specific events would include adrenal insufficiency in patients with Cushing's disease and are addressed in Chapter 1, Chapter 2, Chapter 3 and Chapter 4. Technique-specific events include hemorrhage, hypercarbia, pneumothorax, and pneumomediastinum, which can be avoided by careful attention to detail and close cooperation and monitoring by the anesthesiologist. Results As this is a relatively new approach to the adrenal, most reports are anecdotal. 7,8 However, results indicate that the procedure is technically feasible, offers quicker recovery and lower morbidity, and with more experience will become the most common approach to adrenal lesions. CHAPTER REFERENCES
1. Gagner M, Lacroix A, Bolte E. Laparoscopic adrenalectomy in Cushing's syndrome and pheochromocytoma (letter to the editor). N Engl J Med 1992;327:1033. 2. Gaur DD. Laparoscopic operative retroperitoneoscopy. Use of new device. J Urol 1992;148:1137–1139. 3. Guazzoni G, Momtorsi F, Bocciardi A, et al. Transperitoneal laparoscopic versus open adrenalectomy for benign hyperfunctioning adrenal tumors. A comparative study. J Urol 1995;153:1597–1600. 4. Kelly M, Jorgensen J, Magarey C, Delbridge L. Extraperitoneal laparoscopic adrenalectomy. Aust NZ J Surg 1994;64:498–500. 5. Nakagawa K, Murai M, Deguchi N, et al. Laparoscopic adrenalectomy: clinical results in 25 patients. J Endourol 1995;9:265–267. 6. Rassweiler JJ, Henkel TO, Potempa DM, Copcoat M, Aiken P. The technique of transperitoneal laparoscopic nephrectomy, adrenalectomy and nephroureterectomy. Eur Urol 1993;23:425–430. 7. Suzuki K, Kageyama S, Ueda D, et al. Laparoscopic adrenalectomy: the initial 3 cases. J Urol 1993;149:973–976. 8. Tazaki H, Baba S, Murai M. Technical improvement in laparoscopic adrenalectomy. Tech Urol 1995;1:222–226.

Chapter 133 Renal Cysts Glenn’s Urologic Surgery

Chapter 133 Renal Cysts
Michael P. O'Leary

M. P. O'Leary: Department of Surgery, Division of Urology, Harvard University Medical School, and Division of Urology, Brigham and Women's Hospital, Boston, Massachusetts 02115.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Cystic disease of the kidney may be divided into genetic and nongenetic categories. Genetically determined disorders include autosomal recessive (infantile) polycystic kidneys, autosomal dominant (adult) polycystic kidneys, and juvenile nephronophthisis, which is also autosomal recessive. Nongenetically determined cystic malformations include multicystic kidneys, multilocular cystic kidneys, and simple cysts. This chapter focuses on simple renal cysts. Renal cysts are believed to develop in tubular segments and Bowman's capsule, with simple cysts developing from the convoluted tubule, whereas the acquired and genetic forms may develop anywhere along the nephron and collecting duct. 3 Regardless of their origin, most are lined with a single epithelial layer. Simple cysts are extremely common, occurring in up to 50% of individuals over the age of 50. 9 The majority are asymptomatic and are found incidentally at the time of radiographic imaging for nonrenal conditions. Most are unilocular and, because they arise from the cortex, they distort the normal renal contour. They may enlarge somewhat with time, although most probably remain stable in size and appearance. Multicystic dysplastic kidneys may represent an extreme form of hydronephrosis and are generally encountered in infants, whereas multilocular cysts occur in young children and are believed to be part of a spectrum of neoplasia, associated with Wilms' tumor. 5 Cysts may also develop in end-stage kidneys and are present in 50% of patients on chronic dialysis. 2

DIAGNOSIS
The diagnosis of renal cysts used to be made during intravenous urography with tomography. Today ultrasound (US) and computed tomography (CT) are the principal diagnostic tools, and most cysts are found incidentally. Occasionally a patient with a very large simple cyst may present with vague abdominal or flank discomfort and fullness on physical exam, though most patients are asymptomatic. The sonographic criteria for simple cysts are absence of internal echoes, increased through transmission of sound, and a sharply defined wall. On CT, cysts are sharply marginated, with attenuation values in the range of water (–10 to –20 Hounsfield units) ( Fig. 133-1). They do not enhance after administration of contrast material. Further elucidation may be obtained using magnetic resonance imaging (MRI) in which cysts appear as round, homogeneous, and low intensity (dark) on T1-weighted images and increased intensity (lighter) on T2-weighted images.

FIG. 133-1. CT scan of a renal cyst (arrow).

Complex renal cysts present a greater challenge both diagnostically and therapeutically because 15% of renal tumors may present this way. classification system using CT to aid in diagnosis ( Table 133-1).1

4

Bosniak has proposed a

TABLE 133-1.

INDICATIONS FOR SURGERY
Surgery for the Bosniak type I simple renal cyst is rarely warranted. Occasionally, patients who present with flank pain are discovered to have no other etiology on radiographic evaluation for their complaint other than a large simple cyst, and may be candidates for therapy.

ALTERNATIVE THERAPY
These patients may be candidates for cyst aspiration under CT or ultrasound guidance. Although such cysts will almost certainly recur, it is reasonable to conclude that if the patient's pain resolves after cyst aspiration, then some definitive form of therapy may be warranted and justifiable. The use of percutaneously injected sclerosing solutions has limited value in our experience. Open surgical approaches have been favored in the past 9 but the potential morbidity of the procedure must

be weighed against the benefits. A minimally invasive approach offers a reasonable alternative.

SURGICAL TECHNIQUE
Our favored approach for treating renal cysts, either simple or multiple, has been laparoscopic marsupialization. This technique has also been reported to treat symptomatic polycystic kidneys.7 This is done under general endotracheal anesthesia with the patient placed in a 45 degree lateral flank position. An infraumbilical minilaparotomy incision is made and the blunt-tipped Hasson trochar is inserted in the peritoneal cavity. A purse string fascial suture of 2-0 Vicryl is used to secure the trochar and later to close the fascial defect. Carbon dioxide is insufflated to create the pneumoperitoneum, which is maintained at a pressure of 10 to 15 mm Hg. A 10-mm port is placed under laparoscopic guidance approximately 5 cm lateral and inferior to the umbilicus ( Fig. 133-2). A third 5-mm trochar is placed below the costal margin in the anterior axillary line. A fourth port is advocated by some but is rarely necessary in our experience. The white line of Toldt is incised and the colon is reflected medially. Gerota's fascia is then incised. The cysts are then generally readily visualized. Using an Orandi-type needle the cyst can be punctured and cyst fluid sent for cytologic diagnosis. The cyst can then be unroofed with laparoscopic shears and its walls carefully inspected to rule out malignancy, which, while extremely rare, has been reported. 8 No attempt is made to fulgurate the cyst wall, but generally the cyst wall is at least partially resected. Then, using endoscopic clip appliers, we have clipped a portion of the peritoneum to the cyst wall to ensure communication with the peritoneal cavity and decrease the likelihood of recurrence. Bleeding is minimal. No drains are necessary. The ports are removed under direct vision, the pneumoperitoneum is evacuated, and the trochar sites are closed with 2-0 polyglactin suture for the fascia and 4-0 for the skin in a subcuticular fashion.

FIG. 133-2. Laparoscopic port sites.

OUTCOMES
Complications The overall complication rate for laparoscopic renal surgery has been reported to be in the range of 15% to 20%, with the majority of these being minor and occurring in the postoperative period. 6 These procedures have generally been nephrectomies. Intraoperative complications are similar to those of any laparoscopic procedure. The surgeon must be prepared (and likewise the patient well informed preoperatively) to convert to an open procedure in the rare event it becomes necessary. Results Immediate relief of flank discomfort is usually noted if indeed the cyst was the true etiology of the pain. It is therefore imperative to clearly inform the patient preoperatively of the possibility of continued discomfort postoperatively. It is always wise to remember that the surgeon who operates for pain alone embarks on a perilous journey. CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. Bosniak MA. The small (less than 3.0 cm) renal parenchymal tumor: detection, diagnosis, and controversies. Radiology 1991; 179(2):307–318. Glassberg KI. Campbell's urology. 6th ed. Philadelphia: WB Saunders, 1992;1443. Grantham JJ. Pathogenesis of renal cyst expansion: opportunities for therapy. Am J Kidney Dis 1994;23(2):210–218. Hartman DS, Aronson S, Frazer H. Current status of imaging indeterminate renal masses. Radiol Clin North Am 1991;29(3):475–96. Kissaine JM, Smith MG. Pathology of infancy and childhood. 2nd ed. St. Louis: CV Mosby, 1975;587. Moore RG, Kavoussi LR. Laparoscopic nephrectomy. In: Loughlin KR, Brooks DC (Eds.), Principles of endosurgery. Cambridge: Blackwell Scientific, 1996;127–139. Segura JW, Brown JA, Torres VE, King BF. Laparoscopic marsupialization of symptomatic polycystic kidney disease. J Urol 1996;156:22–27. Stenling R, Ljungberg B, Holmberg G, Sjodin J, Hietala S. Renal cell carcinoma in a renal cyst: a case report and review of the literature. J Urol 1990;143:797–799. Walton KN. Urologic surgery. 4th ed. Philadelphia: JB Lippincott, 1991;138.

Chapter 134 Robotics, Telepresence, and Virtual Reality in Urologic Surgery Glenn’s Urologic Surgery

Chapter 134 Robotics, Telepresence, and Virtual Reality in Urologic Surgery
Roland N. Chen and Louis R. Kavoussi

R. N. Chen: Department of Urology, The Cleveland Clinic Florida, Fort Lauderdale, Florida 33309. L. R. Kavoussi: Department of Urology, The James Buchanan Brady Urological Institute, and Department of Urology, Johns Hopkins Bayview Medical Center, Baltimore, Maryland 21224.

Robotics in Urologic Surgery Telepresence Surgery Virtual Reality Conclusion Chapter References

Rapid advances in telecommunication and computer technology have thrust mankind headlong into the information age, forever altering the way we communicate and do business. The early effects of such changes are now becoming apparent on the surgical landscape. Robotics, telepresence, and virtual reality are making their way from manufacturing and entertainment to the operating room. These new tools will soon enhance our ability to perform surgery quickly and skillfully. Because of their expense and dependence on the rapid processing and transfer of data, these disciplines remain in their infancy. However, continued improvements in optical fiber, silicon microchip, superconductor, and supercomputer technology will soon allow these barriers to be overcome. Future challenges lie not only in developing useful technologies but in making them practical and cost-effective. Robotics in Urologic Surgery The term “robot” is derived from the Czechoslovakian word robota, which denotes obligatory servitude. Although robots have been used extensively in industrial manufacturing, undersea applications, and space exploration over the past four decades, they have only recently been introduced in the operating room. A robot consists of a computer control system and a mechanical component which, for a surgical robot, is usually an arm that conveys motion to an attached instrument ( Fig. 134-1). A robotic arm can have multiple joints that translate or rotate. A robot's dexterity is determined by the total number of motions that the joints allow, described as degrees of freedom. For example, the human elbow has 1 rotational degree of freedom, whereas the shoulder and wrist each have 3 rotational degrees of freedom. Therefore, the human palm has 7 degrees of freedom, or a sum of the degrees of freedom provided by the shoulder, elbow, and wrist. A minimum of 6 degrees of freedom is necessary for an attached instrument to have complete freedom of motion. The computer and accompanying software that control the robot's motions can coordinate complex movements of multiple joints. Each software package is uniquely dedicated to a specific application.

FIG. 134-1. Illustration of a surgical robotic arm.

There are several advantages of utilizing robots in surgery. Robots are capable of performing tasks with great precision and without fatigue, thereby improving the efficacy, reducing the potential morbidity, and decreasing the length of a procedure. They also may potentially replace costly personnel by alleviating the need for a surgical assistant. Several important factors must be considered in the design of a robot for medical applications. Surgical robots must be cost-effective, intuitive to use, and ergonomic with respect to the patient, as well as the operating room equipment and personnel. In addition, redundant built-in safeguards are necessary to minimize the risk of inadvertent patient injury by the robot. Because of technological limitations and safety issues, the application of robots in surgery remains relatively limited. Their role in surgery will undoubtedly expand as improvements and innovations in software and hardware are achieved. Robots have initially been employed as assistants in procedures performed on tissues that provide fixed landmarks as reference points. Neurosurgeons, for example, have utilized robots that provide a three-dimensional road map based on computed tomography (CT) and magnetic resonance imaging (MRI) data as localizers and navigators for performing stereotactic neurosurgery. Orthopedic surgeons have employed robots to drill the femoral shaft in preparation for uncemented hip prostheses. Using similar concepts, a group of investigators at the Imperial College in London 6 developed a robotic system that is capable of performing rapid and accurate transurethral resection of the prostate ( Fig. 134-2). This system was first introduced in 1989 and continues to evolve. The most recent design employs a robotic arm that controls a 24-Fr resectoscope that is mounted on a constraining safety frame. Prior to initiation of the procedure, cystoscopy is performed and the resectoscope loop placed at the verumontanum. The prostatic size and shape based on transurethral ultrasound measurements are then input into the computer, allowing a three-dimensional model of the prostate to be constructed. Once the cutting limits are established, the robot performs a series of programmed cuts, removing one or more cones of tissue from within the prostatic fossa. The procedure is monitored by the surgeon via the attached endoscopic camera. Because the cutting sequence is preprogrammed, the resection process can be completed in less than 5 minutes. The significantly shortened procedure time allows hemostasis to be achieved following completion of the procedure, reduces the anesthetic risk, and minimizes intraoperative bleeding.

FIG. 134-2. Imperial College robotic system for transurethral resection of the prostate. (A) Pattern of resection performed by the robot. (B) Robotic safety frame.

In a trial of 30 patients in whom manual resection was performed using the safety frame, these authors have reported results comparable to standard transurethral resection of the prostate at 1 year following treatment. Five patients were treated successfully in an initial clinical trial using the unaided robotic system. This trial demonstrated that transrectal ultrasonography is inaccurate for measuring the prostatic dimensions. A modified system using transurethral ultrasonography as the imaging modality is currently being developed. The Imperial College group in London has also investigated the use of a robot for acquiring percutaneous renal access. 8 This group developed a system that employs a passive robot mounted to the operating table. Two-dimensional images of the intrarenal collecting system are obtained intraoperatively via a mobile C-arm fluoroscopic unit. These images are processed by a microcomputer, which in turn provides three-dimensional coordinates to the robot. The system allows the surgeon to select the trajectory of the access needle. Experiments evaluating system performance have demonstrated a targeting accuracy of within 1.5 mm. No in vitro or in vivo experiments have been performed to date. A group at Johns Hopkins have also developed a robotic system for acquiring percutaneous renal access. 1 Two-dimensional images are obtained via biplanar fluoroscopy and processed by a microcomputer. Once the needle trajectory is selected by the surgeon, the active robot positions, orients, and drives the access needle into the desired calyx. Experiments evaluating the performance of this system have demonstrated a targeting accuracy of within 1.0 mm. Ex vivo experiments in porcine kidneys demonstrated entry into the selected calyx in 10 of 12 attempts (83% accuracy) using this system. Robots designed as assistants for general laparoscopic surgery have also found use in urology. The AESOP (Automated Endoscopic System for Optimal Positioning, Computer Motion, Inc, Goleta, CA) robot, an FDA-approved device, has been used with considerable success as an assistant in laparoscopic surgery. 7 The robotic arm is controlled by a foot pedal or hand controller and is used to hold the laparoscope and/or a laparoscopic retracting device. Partin et al. evaluated the use of the AESOP robot in a series of 17 laparoscopic urologic procedures including nephrectomy, pyeloplasty, retroperitoneal lymph node dissection, and bladder neck suspension.7 The laparoscopic procedures were performed successfully in all cases and independently of additional human assistance in 14 of the 17 cases. These authors also found that laparoscopic cases performed with or without the robot did not differ significantly in terms of setup or breakdown time. In a blind study comparing the AESOP robot to human laparoscope holders, Kavoussi et al. found that the AESOP robot made significantly fewer inadvertent movements and rotations and came into contact with other instruments or adjacent organs a significantly fewer number of times than human assistants. 4 The next generation of robots to hold laparoscopic instruments will be more intelligent. They will be designed with voice recognition and will respond to the surgeon's vocal commands. Such robots may also be designed to recognize specific images such as the tip of a laparoscopic instrument. Using such a system, the surgeon holds the tip of the instrument in the field of interest and instructs the robot to center the image around the tip, bringing the field of interest into center view. Future designs may incorporate retinal motion detectors that track the movements of the surgeon's eyes, obviating the need for the surgeon to provide verbal commands. Telepresence Surgery As the term implies, telepresence brings the surgeon via global telecommunications from a remote location to the operative site to participate in a surgical procedure. The implementation of telepresence systems would improve patient access to specialty care and consultation for difficult cases by obviating the need for travel by the patient to the surgeon. Such systems would also facilitate the introduction of novel procedures into the community and allow physicians to provide remote specialty health care to patients who live overseas or in rural areas, and to patients in outer space or on the battlefield. Ideally, complete operative information is transmitted to the surgeon at the remote site in real time, providing the surgeon with the perception of actually being present at the operative site. At the most basic level, the task of the surgeon at the remote site may be to perform a consultation with a patient or to advise a less experienced surgeon during a procedure (telementoring) by giving instructions or recommendations via the telepresence system. The surgeons at the remote and surgical sites may communicate by verbal or written means. For example, the advisor at the remote site can speak directly to the surgeon and simultaneously draw an arrow on the video screen in order to indicate where the surgeon at the operative site should dissect or cut (telestration). At the next level of complexity for telepresence surgery, the surgeon (master) works at a computer workstation at the remote site and controls one or more robotic manipulators (slaves) at the operative site that perform or assist in the procedure. The potential applications for such systems include not only telementoring but also performing surgery on patients in very distant or potentially dangerous locations such as outer space or on the battlefield. Kavoussi et al. evaluated the use of a telepresence operative system in laparoscopic urologic surgery 3 (Fig. 134-3). In this study, a remote surgeon, who operated the robotically controlled laparoscope via a control pad, telementored an inexperienced primary surgeon and assistant as they performed a laparoscopic nephrectomy, cholecystectomy, and splenectomy in a swine model. A similar arrangement was then utilized to perform laparoscopic procedures in three human patients including a laparoscopic cholecystectomy, varix ligation, and bladder neck suspension. All of the procedures were performed successfully with the aid of the telementoring system.

FIG. 134-3. A laparoscopic telementoring system.

Moore et al. evaluated the feasibility of telementoring in a series of 23 laparoscopic urologic procedures including nephrectomy, bladder neck suspension, and pelvic lymphadenectomy.5 Experienced remote surgeons telementored inexperienced laparoscopic surgeons using real-time video images, a two-way audio communication system, a remotely controlled AESOP robot laparoscope holder, and a telestration system. The remote site and the operating room were in adjacent buildings connected via a hard-wired system. These authors found that telementoring allowed the procedures to be performed successfully in 22 of the 23 cases (96%). When the telementored cases were compared to matched procedures in which the experienced surgeon worked side by side with the inexperienced surgeon, telementoring resulted in longer operative times for complex cases. However, there were no differences in terms of complication rate, postoperative analgesic requirement, length of hospital stay, or number of days to full activity. These authors concluded that telementoring is safe, feasible, and may be useful in surgical training. Schulam et al. employed a similar telementoring system linking two sites separated by approximately 3.5 miles via an existing telephone line. 11 All seven laparoscopic urologic procedures were performed successfully with this system. This study demonstrated that, using current technology such as personal computers and a single high-bandwidth telecommunications link, an effective and safe telementoring system is feasible even today. There are currently highly complex prototype telepresence surgery systems that have been used successfully in ex vivo and animal models. Green et al. have designed a telepresence surgery system that allows open surgery to be performed from a remote site. 2 The patient and robots are positioned and the operative field is prepared by a technician at the surgical site. The surgeon sits at a remote workstation and manipulates instruments that are interfaced with a computer system that controls the robots at the operative site. These dexterous robots, with 6 degrees of freedom, hold and manipulate the instruments that actually perform the surgery. A pair of CCD video cameras at the operative site provides the surgeon with three-dimensional vision that is displayed in real time at the remote site on a 120-frame-per-second stereographic video monitor with a liquid crystal shutter. The instrument hand controllers are designed with responsive, force-reflecting servomotors that provide the surgeon with force and tactile feedback. With these features, for example, the surgeon can feel resistance as a needle is driven through tissues or as a vessel is cut with scissors. The surgeon communicates verbally with the surgical assistant and nurses at the operative site via two-way audio channels. Experiments have demonstrated that this telepresence system provides sufficient dexterity to slice a grape into 1-mm sections. Simple ex vivo and in vivo animal

experiments have been performed successfully using this system. The Zeus Telepresence Surgery System (Computer Motion, Inc., Goletas, CA) is another working prototype system that was designed specifically to allow surgeons to perform remote laparoscopic surgery 12 (Fig. 134-4). Again, the operative field is prepared by a technician at the operative site. The technician places the laparoscopic ports and positions the robots, laparoscope, and instruments. The surgeon operates from the remote workstation and is provided with a two-dimensional laparoscopic view of the operative field. The human–machine interface consists of instrument handles manipulated by the surgeon that control dexterous robotic manipulators at the operative site. This system has been used successfully in ex vivo laparoscopic surgery on animal viscera.

FIG. 134-4. A laparoscopic telepresence surgery system. Flexible ureteroscopy simulator. (Courtesy of HT Medical, Rockville, MD.)

Telepresence surgery has the potential to vastly improve surgical techniques in the future. Because the robot mimics the surgeon's movements and can potentially be programmed to reduce large, tremulous movements into fine, smooth motions, telesurgery could become invaluable for performing microsurgical procedures. Indeed, NASA's Jet Propulsion Laboratory has developed a robot that can scale down movements by a factor of up to 100:1. 10 These advances are a prelude to the possibility of someday performing complex surgery at the cellular or nuclear level. Virtual Reality The concept of virtual reality involves the creation of an artificial site in cyberspace that appears via human–computer interfaces to be so convincing that it approaches reality. Participants enter an imaginary three-dimensional world in which they can move around and interact with the environment. Although most current virtual reality systems have been designed primarily for entertainment, one of the most successful applications of such technology is the flight simulator. Pilots and astronauts practice taking off, flying, and landing on such simulators and, perhaps more importantly, they learn how to react to adverse situations before they are encountered in actual flight. Current virtual reality surgical systems are simulators intended to be used as training devices as well. Because of technological limitations, however, these systems often provide cartoon-like video images that often do not appear to be realistic. High Techsplanations (HT Medical, Inc., Rockville, MD) has been at the forefront of virtual reality medical and surgical simulation technology. This company employs supercomputers to simulate real-time procedures such as subclavian line placement, cardiac catheterization, and bronchoscopy. HT Medical has developed a real-time laparoscopic pelvic lymph node dissection simulator. A pair of laparoscopic instrument handles are used to perform the procedure in which deformable soft tissue structures such as blood vessels can be grasped, manipulated, or cut. HT Medical has also recently developed a virtual reality system that allows urologists to practice flexible ureteroscopy ( Fig. 134-5). The human–computer interface is a small, flexible ureteroscope that can be advanced into a black box. The surgeon watches a video image of the view from the tip of the ureteroscope as it is advanced through the ureter and into the intrarenal collecting system. A deflection mechanism provides the surgeon the ability to manipulate the tip of the ureteroscope into each of the calyces where stones, intrarenal tumors, and air bubbles may be seen.

FIG. 134-5. Prototype of virtual reality technology that can be used with a ureteroscope (courtesy of HT Medical, Inc., Rockville, MD).

The next generation of virtual reality systems will not only provide more accurate detail, but they will also become significantly more interactive. Very high-resolution three-dimensional images will be displayed with smooth and life-like motion. Pulsating vessels will bleed when cut and tissue planes can be bluntly dissected. The surgeon will experience tactile sensation and force feedback. The purpose of virtual reality surgical simulators is to provide surgical residents with practical operative experience, just as pilots and astronauts learn to fly on sophisticated flight simulators today. They will also learn how to cope with adverse situations such as intraabdominal bleeding and renal lacerations in cyberspace prior to confronting them in actual situations. Future surgeons may also be tested on such virtual reality systems in order to demonstrate a certain level of technical proficiency for certification purposes.

CONCLUSION
Robots, telepresence, and virtual reality are evolving technologies that will continue to become integrated into surgical practice. Continued improvements in telecommunications and computer technology are vital for the continued development of these disciplines. With increasing speed and volume of data transmission, computer graphics for virtual reality systems will improve dramatically. Not only will images be significantly more life-like, but movements will be rapid and smooth. More sophisticated human–machine interfaces will allow robots to follow surgeons' commands more faithfully and to dampen or filter out unintentional movements such as tremors. The cost of such systems will fall dramatically as their use becomes widespread and as the necessary technology becomes widely available. With increasing sophistication and resolution of non-invasive imaging modalities, three-dimensional digital images from CT, MRI, or ultrasound may be integrated onto the surgeon's real-time video image in order to provide the surgeon with a greater intraoperative understanding of anatomic and pathologic detail. This would essentially be a detailed, superimposed road map for the surgeon and robot to follow. With increasing sophistication in both software and hardware, robots will gain increasing independence until perhaps someday much of a procedure will be performed by robots independently, with the role of the surgeon focusing on surgical indications and monitoring the progress and safety of the procedure. Indeed, through continued development of nanotechnology and micromachines, robots may one day be miniaturized to the point where automated microrobots with built-in end effectors can be placed into the body and directed to the site of pathology where surgery is performed at the cellular level. CHAPTER REFERENCES
1. Bzostek A, Schreiner S, Barnes A, et al. An automated system for precise percutaneous access of the renal collecting system. Unpublished data, 1996.

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Green P, Satava R, Hill J, Simon I. Telepresence: advanced teleoperator technology for minimally invasive surgery (Abstract). Surg Endosc 1992;6:91. Kavoussi LR, Moore RG, Partin AW, Bender JS, Zenilman ME, Satava RM. Telerobotic assisted laparoscopic surgery: initial laboratory and clinical experience. Urology 1994;44:15. Kavoussi LR, Moore RG, Adams JB, Partin AW. Comparison of robotic versus human laparoscopic camera control. J Urol 1995;154:2134. Moore RG, Adams JB, Partin AW, Docimo SG, Kavoussi LR. Telementoring of laparoscopic procedures: initial clinical experience. Surg Endosc 1996;10:107. Nathan MS, Davies B, Hibberd R, Wickham JEA. Devices for automated resection of the prostate. Proc First Int Symp MRCAS. Pittsburgh, 1994;342–344. Partin AW, Adams JB, Moore RG, Kavoussi LR. Complete robot-assisted laparoscopic urologic surgery: a preliminary report. J Am Coll Surg 1995;181:552. Potamianos P, Davies BL, Hibberd RD. Intraoperative registration for percutaneous surgery. Proc 2nd Int Symp Med Rob Comput Assist Surg 1995;156. Sackier JM, Wang Y. Robotically assisted laparoscopic surgery. Surg Endosc 1994;8:63. Satava RM. Robotics, telepresence, and virtual reality: a critical analysis of the future of surgery. Min Inv Ther 1992;1:357. Schulam PG, Docimo SG, Saleh W, Breitenbach C, Moore RG, Kavoussi LR. Telesurgical mentoring: Initial clinical experience. Surg. Endosc 1997;11:1001–1005. Wang Y. Zeus Telepresence Surgery System. Computer Motion, Inc., Goletas, CA, personal communication, 1996.

Chapter 135 Cryosurgical Ablation of the Prostate Glenn’s Urologic Surgery

Chapter 135 Cryosurgical Ablation of the Prostate
Harry S. Clarke

H. S. Clarke: Division of Urology, Emory University School of Medicine, Atlanta, Georgia 30322.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Cryosurgical ablation of the prostate was first investigated more than 30 years ago. The initial approach was transurethral freezing reported by Gonder, Soans, and Smith in which the indications for cryoablation was prostatic obstruction either from benign prostatic hyperplasia or adenocarcinoma. 3 From these initial studies, Flocks et al. developed a treatment of prostate cancer, utilizing a perineal approach. 2 Once the prostate was isolated in the surgical field, the cryoprobes were inserted and freezing was accomplished by the circulation of liquid nitrogen through the hollow probes. The procedure was open and the extent of freezing required was determined by direct visual inspection. While the 10-year survival was similar stage for stage with both radiation therapy and radical prostatectomy, the procedure was largely abandoned due to the technical difficulties and the complications encountered. 1 There was no accurate means of assessing the extent of freezing and patients suffered the consequences of thermal injury to the bladder, ureters, and rectum. Further reports on outflow obstruction due to sloughing of necrotic debris, urinary tract infection, hemorrhagic cystitis, and urethrocutaneous fistulas severely dampened the initial enthusiasm for this therapy as a means of treating the elderly and poor-risk surgical candidate. Cryosurgery for the treatment of prostate cancer has currently received renewed enthusiasm due in large part to the technological advances that have allowed circumvention of the morbidity previously described. The significant innovations included a new liquid nitrogen delivery system, which allows the delivery of super cooled liquid nitrogen or argon gas through small (3-mm) probes; percutaneous access technique; and the ability to evaluate the extent of freezing by transrectal ultrasound imaging.4

DIAGNOSIS
The diagnosis of prostate cancer is made by direct biopsy of the prostate, usually under ultrasound guidance. This is covered more extensively in Chapter 36.

INDICATIONS FOR SURGERY
The ideal candidate for cryosurgical ablation of the prostate is an individual with local recurrence after treatment with radiation therapy because there is no other successful definitive therapy, other than palliation, for these patients. Other patients without prior treatment for T1 to T3 prostate cancer may also benefit from this therapy; however, these patients should be well informed of the investigational nature of this procedure and it should be done only in the setting of a properly controlled study.

ALTERNATIVE THERAPY
Alternatives to cryotherapy include observation, hormonal deprivation, radiation therapy (in patients who have not had prior radiation), and radical prostatectomy. In patients who have had prior radiation therapy, neither additional radiation nor radical prostatectomy is generally indicated. Salvage prostatectomy in our institution was associated with no patient having an undetectable prostate-specific antigen (PSA) beyond 18 months and a 75% rate of incontinence.

SURGICAL TECHNIQUE
The technique of the procedure as initially described by Onik et al. begins with the establishment of suprapubic drainage via a punch using a nephrostomy tube with a coil on one end, guided by flexible cystoscopy. 4 A guidewire is inserted through the flexible cystoscope and the cystoscope is removed. A urethral warming catheter is then inserted over the guidewire, and the guidewire is removed. The warming catheter is connected to inflow and outflow tubing from a fluid pump that circulates normal saline for urethral warming at 44°C. A Bookwalter retractor is attached to the table at the patient's right side. The large ring is used to hold the flexible cystoscope and for support of loose guidewire ends and the cryoprobes, so that they do not have to be hand-held during application of the energy. The scrotum is elevated utilizing a gauze sling stapled in position and then secured to the retractor as well. Attention is then directed to the ultrasound machine. This begins with testing of the alignment to assure coordination between the needle guide and the screen guideline in a pan of water. The sonographic probe is inserted into the rectum, and the gland is examined in transaxial and longitudinal views. Five needles are inserted sequentially percutaneously in the perineum, two anterolaterally, two posterolaterally, and one posterior to the urethra under sonographic guidance (Fig. 135-1). Each needle is visualized first in the transaxial plane until it penetrates the scan plane about midprostate. The view is then switched to longitudinal, and the needle is advanced to the cephalad margin of the gland. A guidewire is then advanced through the needle until its floppy end can be visualized at the cephalad tip of the needle. The needle is then withdrawn and the skin is incised with a fascial incisor needle that slides over the guidewire.

FIG. 135-1. Placement of the needles for cryoablation of the prostate.

Once the skin is incised, a fascial dilator with the proximal end split for about 4 cm is then inserted over the guidewire to dilate the channel. A small amount of saline is then injected with a syringe and a spinal needle into the dilator channel in order to facilitate its visibility with ultrasound. The guidewire is removed, and the cryoprobe is inserted to the cephalad most part of the dilator, and the dilator is then pulled back using the split in its proximal end, leaving the cryoprobe in direct contact with the tissue. The cryoprobe is then turned on at –70°C in order to “stick” the probe to the tissue and the remaining four probes are inserted in similar fashion, taking care to

insert the anterior probes first, in order not to have shadowing from the posterior probes. Figure 135-2 is a transaxied view of the prostate with the fine cryoprobes and urethral warmer in position. Each cryoprobe is supported from the retractor ring by a rubber strap (a urinary leg bag strap), and after all have been inserted, they are tied together with an opened 4 × 4 sponge to keep them from moving. Figure 135-3 is a photograph of a patient with the probes in position secured to the buckwatter retractor.

FIG. 135-2. Ultrasonographic image of prostate with cryoprobes (p) in position, two anterior and three posterior. The urethra is shown in the center (u).

FIG. 135-3. A patient with probes in position secured to the buckwatter retractor.

The anterior cryoprobes are turned on first, taking the temperature down to about –190°C, and formation of the ice ball is monitored easily as a very hyperechoic acoustic interface line that develops at the junction of the ice ball and the unfrozen tissue. As the line progresses posteriorly, the two posterolateral probes are turned on, and further progress of the ice ball toward the rectum can be monitored with precision. Last, the probe posterior to the urethra is turned on. When the freezing process is complete, with care taken to monitor progression of the ice ball toward the rectum very carefully, the cryoprobes are turned off and the tissue is allowed to thaw. Once thawed a second freeze is performed in the same fashion as the first. The active portion of the cryoprobe extends from the tip up the probe for 4 cm. If the prostate length is greater than this, then the cryoprobes once thawed are pulled back the appropriate length to cover the untreated area and the freezing sequence is repeated to cover this tissue at the apex. Similarly the diameter of the active freeze zone for each cryoprobe is 4 cm, and care must be taken to ensure that there is adequate overlap of the freeze zones of the five probes to guarantee complete freezing of the entire prostate gland. The seminal vesicles can be treated by insertion of cryoprobes directly into those structures. Care must be taken to avoid freezing the distal ureter when this is done. Additional probes can be placed as needed for larger prostate glands. Thermocouples may be placed at the level of the neurovascular bundle on either side and at the level of Demonulli's fuscia in the suburethral area to monitor the temperature as the ice ball advances. After thawing is complete, the cryoprobes are removed along with the sheaths. Gentle pressure is applied to the perineum for 5 minutes to prevent hematoma formation. Each perineal puncture is closed with a stitch of 3-0 chromic suture. The urethral warming catheter is then removed as the last step in the procedure. The procedure is carried out under general or regional anesthesia. The patient goes home on the first postoperative day with the suprapubic tube in place. He is instructed to clamp the suprapubic tube after 1 week for a trial of voiding, and the tube is removed when the residual urine falls below 100 cm 3. There is almost always some scrotal and perineal edema and ecchymosis, but this resolves spontaneously in all patients.

OUTCOMES
Complications In addition to the edema and ecchymosis encountered by all patients undergoing cryoablation of the prostate, many will experience some sloughing of urethral tissue. This comes out as a rather fine and homogeneous material, and in most instances with the use of the urethral warming device it does not cause obstruction. If significant obstruction does occur it can usually be managed with a period of intermittent catheterization, but in some instances it may require transurethral resection of necrotic material. If resection is necessary, then care must be taken to avoid resection of distal viable tissue as incontinence may ensue. Incontinence has been reported in as many as 5% of patients after cryoablation. The majority of this resolves without intervention in most patients. Urinary tract infection, both with typical urinary tract pathogens as well as some less typical organisms, such as Staphylococcus epidermidis and Candida albicans, have been reported, especially in patients who are experiencing significant amounts of sloughing of urethral tissue. Treatment with the appropriate antibiotics rapidly resolves these infections and the accompanying symptoms. The occurrence of urethrorectal fistula is less than 1%. The first symptom of this may be some mild watery diarrhea. This may be managed conservatively with several weeks of Foley catheter drainage being all that is required. More significant fistulas may require laparoscopic diversion and primary repair. Impotence has been reported in patients undergoing cryoablation; however, this has not been well established or studied in a series of men with documented potency prior to undergoing the procedure. Results When cryoablation has been successfully performed, the PSA in patients drops rapidly to the undetectable level (less than 0.1) and remains there. Those patients whose PSA does not nadir to the undetectable range often will have positive biopsies, or the presence of persistent benign glands in their biopsy specimens, and will have subsequent rises in their PSA and positive biopsies at a later date. Should this be the case, the patient may have the procedure repeated, undergo teletherapy, or undergo radical prostatectomy in some instances. The results of cryoablation of the prostate, while short term, have been very promising. The majority of centers performing the procedure report similar results, with negative biopsy results of 70% to 80% 6 months after the procedure. The results have held up with similar findings on subsequent biopsies at 2 and 3 years. The

longest current follow-up is just now approaching 5 years, and further studies will be pivotal in determining the long-term utility of this procedure. CHAPTER REFERENCES
1. 2. 3. 4. Bonney WW, Fallon B, Gerber WI. Cryosurgery in prostatic cancer: survival. Urology 1982;19:37–42. Flocks R, Nelson C, Boatman D. Perineal cryosurgery for prostatic carcinoma. J Urol 1972;108:933–935. Gonder MJ, Soanes WA, Smith V. Experimental prostate cryosurgery. Invest Urol 1964;1:610–619. Onik GM, Cohen JK, Reyes GD, Rubinsky B, Chang ZH, Baust J. Transrectal ultrasound–guided percutaneous radical cryosurgical ablation of the prostate. Cancer 1993;72:1291–1299.

Chapter 136 Transurethral Microwave Thermotherapy Glenn’s Urologic Surgery

Chapter 136 Transurethral Microwave Thermotherapy
Aaron P. Perlmutter

A. P. Perlmutter: Brady Prostate Center, Department of Urology, New York Hospital–Cornell Medical Center, New York, New York 10021.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Pretreatment Considerations Treatment Posttreatment Outcomes Complications Results Chapter References

Transurethral microwave thermotherapy is a treatment for symptomatic benign prostatic hyperplasia (BPH). This noncancerous enlargement of the prostate gland can begin as early as 30 years of age, is histologically identifiable in half of men age 60, and approaches 90% by age 85. However, the prevalence of clinical symptomatic BPH has been estimated at approximately 40% over age 60. The risk factors for development of symptomatic BPH include increasing age, the presence of androgens, and family history. Simplistically, the enlargement of the periurethral transition zone tissue creates a physical obstruction to urine flow, and this increase in outflow resistance results in changes in bladder function that are noticed by the patient. However, the pathophysiologic interactions are not well understood and are most likely multifactorial because there is poor correlation between prostate size, the severity of symptoms, and the urodynamic determination of obstruction. The sequelae of the natural history of untreated BPH include the following: Urinary retention Upper tract deterioration Bladder calculi Recurrent infections Gross hematuria However, the cumulative occurrence of these events during 3 to 5 years of observation is at most 10%. 2,10 Treatment of BPH is usually initiated because of the bothersome obstructive and irritative symptoms that interrupt normal daytime lifestyle patterns and reduce sleep. It is in the area of improvement of symptoms caused by prostatic outflow obstruction that transurethral microwave thermotherapy has had its greatest impact.

DIAGNOSIS
BPH describes a histologic process, and this process can create bladder outflow obstruction (BOO). The patient usually seeks treatment for lower urinary tract symptoms (LUTS). The diagnosis of BPH is the means that connects this growth process to the symptoms of voiding dysfunction experienced by the patient. The diagnosis of BPH is made by history, physical examination, and symptom score assessment. The examination allows for palpation of a distended bladder and for detection of an enlarged or abnormal prostate. Interestingly, patients with BPH by exam and symptoms only have urodynamic proven BOO 60% to 70% of the time. 1 Urine flowmetry, residual urine volume, and imaging are sometimes used as adjunctive tests. Bladder pressure–flow studies provide the most definitive diagnosis. Although urethrocystoscopy is not necessary for the diagnosis of BPH, it identifies patients with intravesical middle lobes, high median bars, bladder neck stenosis, and very asymmetric prostates. Urethrocystoscopy is indicated in the pretreatment evaluation for transurethral thermotherapy since the above subgroups are unlikely to respond to treatment. Not all patients empirically treated with symptomatic prostatism have BOO from BPH. There are patients with the symptoms of “prostatism” without bladder outlet obstruction from BPH (i.e., neurologic diseases, primary bladder diseases, or urethral stricture), and there is a subset of patients with histologic BPH and obstruction who are without symptoms (“silent prostatism”). In order to maximize the likelihood that a patient will respond to transurethral microwave thermotherapy, nonprostatic causes of LUTS should be excluded.

INDICATIONS FOR SURGERY
Transurethral thermotherapy is indicated for the treatment of symptomatic prostatic outflow obstruction. The goal of the treatment is heat-mediated alteration of the adenoma, which results in a decrease in symptomatology and, depending on the temperatures achieved, a decrease in outflow obstruction through tissue necrosis. 6,8 Patients with urinary retention, upper tract deterioration, bladder calculi, and recurrent infections or gross hematuria are better treated with surgical prostatectomy. The most commonly presenting patient for transurethral thermotherapy is a man with symptomatic prostatism who wishes to avoid prostatectomy and has had either no response or an adverse response to medical therapy, or does not wish to take medication on a regular basis. Because the microwave antenna must be safely located in the prostatic fossa, there are certain prostatic size and shape criteria for safe treatment. A intravesical middle lobe or high median bar has the tendency to displace the catheter balloon away from the bladder neck, thus moving the antenna away from proper placement in the middle of the fossa. In addition, the prostatic urethral length must be sufficient to safely accommodate the antenna and the lateral lobes must be of sufficient thickness to safely contain the heat. Manufacturer recommendations vary, but in general the bladder neck to verumontanum length should be at least 3 cm and the total prostatic volume greater than 25 cm 3. Most studies have excluded patients with prostatic volumes greater than 100 cm 3; therefore outcome data in this size range are limited and many investigators consider this to be a relative contraindication. Although bladder neck treatment is generally avoided and thus antegrade ejaculation usually maintained, there can be coagulative necrosis of the ejaculatory ducts and patients should be counseled accordingly. Anticoagulated patients have been safely treated. There is little experience treating patients with previous pelvic irradiation. The radiation-induced cellular changes will likely alter the response to the microwave treatment, and the safety and efficacy are unknown. Exclusion criteria related to the treatment manipulation and microwave and heat safety are listed in Table 136-1.

TABLE 136-1. Patient selection for transurethral microwave thermotherapy

ALTERNATIVE THERAPY
Transurethral microwave thermotherapy is a new, minimally invasive therapy for the treatment of BPH. The goal of such a therapy is to provide symptom improvement without the hospitalization and side effects from transurethral prostatectomy (TURP). Transurethral thermotherapy occupies a unique niche because of its ability to treat symptomatic BPH in an office setting without the need for general or regional anesthesia. Using local anesthesia, the procedure can be performed in approximately 1 hour requiring only flexible instrumentation. Other new minimally invasive therapies that require rigid endoscopy and slightly more patient discomfort include prostatic stents and radiofrequency transurethral needle ablation. These procedures generally require more anesthesia than transurethral thermotherapy. The role for prostatic stenting is more limited than transurethral thermotherapy, and the outcome data for transurethral needle ablation are not as mature as those for transurethral thermotherapy. Two surgical approaches that require at least regional anesthesia but have much less morbidity than TURP are laser prostatectomy and transurethral incision of the prostate (TUIP). Laser prostatectomy retains the efficacy of TURP, although the onset of symptom improvement is not as rapid. TUIP is limited to prostate volumes less than 30 cm3 and may cause bleeding.

SURGICAL TECHNIQUE
The major transurethral thermotherapy devices currently in use include: The Prostatron (EDAP Technomed, Cambridge, MA) The T3 (Urologix Inc, Minneapolis, MN) The Urowave (Dornier MedTech, Marietta, GA) The Prostatron obtained FDA clearance for use in the United States in October 1996; FDA approval for the T3 and Urowave is currently pending. The devices differ in microwave antenna design, irradiation wavelength, wattage generated, and target temperatures. Like lithotriptors, since these different devices vary, one should be chosen after the details of the specific strengths and weaknesses are appreciated. A schematic of a thermotherapy unit and the treatment catheter is shown in Fig. 136-1. The sophisticated nature of this system is mandated by safety considerations and differences in prostate size and configuration, tissue characteristics, and tissue vascularity from patient to patient. Varying amounts of energy may be needed in different patients to generate and maintain the same temperatures.

FIG. 136-1. (A) Each device used for microwave treatment of the prostate has several basic components. A power generator is responsible for microwave emission. A coaxial cable within the treatment applicator serves as the microwave antenna, and an impedance matching network allows for efficient transfer of power from generator to antenna. In addition, a thermometry system allows for the monitoring of temperatures attained during treatment. These components are coordinated by a central computer control system that allows for the maintenance of an integrated system in which power delivered to the prostate may be regulated by temperature feedback. (B) An enlarged view of the intraprostatic placement of the microwave antenna and an idealized heating pattern.

Transurethral thermotherapy is applied by a microwave antenna located in the intraprostatic portion of a modified urethral catheter. The microwave power delivered, maximum urethral and rectal temperatures allowed, and treatment time are determined by the software provided with the device by the manufacturer. When the software program is initiated after proper catheter placement, the heating occurs automatically. The heating results from computer-controlled stepwise increases in power until a maximum power limit is reached or the machine resets because a safety temperature limit is reached in the urethra or rectum. Thermotherapy requires temperatures of greater than 45°C, and all software programs are designed to reach this temperature threshold. Some devices and softwares are able to reach the 50° to 60°C range; this is typical of Prostasoft 2.5 for the Prostatron and is referred to as higher temperature software. These higher temperatures are aimed to create actual cavitation in the prostatic urethra. The responsibilities of the surgeon include patient selection, pretreatment patient preparation, the treatment itself including catheter placement monitoring, and posttreatment management of voiding. Pretreatment Considerations In addition to the above-mentioned selection criteria, other patient characteristics have been identified which allow the selection of a patient most likely to benefit from transurethral microwave thermotherapy. 7 Once a patient elects transurethral thermotherapy, sterile urine is verified prior to treatment. The patient is instructed to use an enema at home before coming for the therapy because the rectal thermosensor can be uncomfortable if the rectal vault contains stool. In addition, the patient is asked to restrict fluid intake the day of the therapy because many of the thermotherapy catheters do not drain, and if the bladder becomes uncomfortably distended during treatment, the treatment might need to be aborted to drain the bladder. Treatment When the patient arrives for the office treatment, an oral antibiotic is administered. The use of pain medication and sedation depends on the temperature limit of the software being used and the temperament of the patient. Even with the lower temperature softwares, the patient perceives the need to void often with a sensation of urge. Oral Valium, Ativan, or Restoril are commonly used to reduce patient anxiety and discomfort. Since many patients experience bladder spasms during the treatment, a rapidly absorbed sublingual anticholinergic can be routinely given before therapy or be available for postprandial dosing. If a higher temperature software program is used, then more anesthesia and a monitored setting may by required. Most patients find that the 1-hour treatment time is easier if a headset for music or a television screen is available for distraction. The patient is placed in a dorsal lithotomy position and sterilely prepped in the same manner as for cystoscopy. This allows use of transrectal sonography to verify catheter placement and the placement of transperineal thermosensors in the research setting. Depending on the device being used, the supine treatment can be used. Lidocaine jelly, 2%, is instilled per urethra and left indwelling for 5 to 10 minutes. A small catheter is passed to drain the bladder completely. The treatment catheter is then passed into the bladder and the anchor balloon is inflated. Since some of these catheters can be stiff, it is helpful to observe the catheter pass through the prostate using transrectal ultrasound. This also allows visualization when the balloon is inflated, and can verify that the catheter is properly positioned and the balloon is at the bladder neck. Balloon placement can also be confirmed by transvesical ultrasound if preferred. The catheter, with the anchor balloon against the bladder neck, is taped in place to prevent migration during the procedure.

The rectal thermosensors are then placed and secured if necessary to the drapes, and the device tested and connected as per the manufacturer instructions. The therapy software is then initiated. The software program will then test all components and automatically perform the treatment. Monitoring during the procedure is important for safety. Patient movements can cause catheter migration, and the catheter can be visually inspected and a transvesical ultrasound can verify balloon placement. Although the devices have many internal safety features, the temperature profiles should be monitored. Sudden patient pain should raise the possibility of a problem and the surgical field should be inspected. Some bleeding around the catheter is normal for some of the devices. At the termination of the procedure, the treatment catheter and other monitoring equipment are removed. If a software program has been used that will likely result in urinary retention, then a Foley catheter is left and the patient discharged with a leg bag. Depending on the software, most patients are ready to void in 3 to 5 days. This seems to be dependent on prostate size, and patients with prostate volumes greater the 50 cm 3 may require 5 to 7 days. This is dependent on the software and temperatures achieved, and each surgeon eventually develops a workable algorithm. Patients discharged with indwelling urethral catheters often have bladder spasms that respond to oral anticholinergic medication. Posttreatment Patients discharged without a catheter usually take oral antibiotics for 3 more days. Since there is edema in the prostatic fossa, the use of alpha blockade, if tolerated by the patients, can be beneficial for the first few postprocedure weeks. Acetaminophen is sufficient for the small number of patients who complain of discomfort. Many patients notice a worsening of symptoms for several weeks until there is improvement. Persistently worse symptoms should prompt an office visit with a postvoid residual determination and urine culture. Unexplained symptoms at 3 to 6 months that have not resolved should prompt a cystoscopy. Even at this time period after treatment, depending on the temperatures achieved, necrotic tissue may be present in the prostatic fossa.

OUTCOMES
Complications One of the great advantages of transurethral microwave thermotherapy is the low incidence of complications. Some events are dependent on the energy dose delivered. Whereas urinary retention is expected in up to 25% of those treated with lower temperature software, it is universal in those treated with high-temperature software.6 Table 136-2 summarizes the range of post treatment events reported in sham controlled studies ( n = 335) using the Prostasoft 2.0 (lower temperature) for the Prostatron, Urowave, and the T3. 3,4 and 5,9 These procedures can be performed in the office using local anesthesia. Common TURP events including transfusion, bladder neck contracture, and retrograde ejaculation were absent.

TABLE 136-2. Transurethral thermotherapy outcomes

Results The results of new therapies are usually measured by changes in voiding symptoms and peak urinary flow rate. These values are reported means; therefore, some patients have minimal response, whereas others are above the mean. It is difficult to obtain an accurate indication, but reported studies indicate that between 6% and 12% of patients treated with the Prostasoft 2.0, T3, or the Urowave find no improvement in symptom score, and between 16% and 23% have no flow rate improvement. Therefore, there is a group of patients who will undergo the treatment without improvement and may be worse. Blute has reported that 1527 patients, 3 months after treatment with the Prostasoft 2.0, found a 61% (13.2 to 5.2) improvement in Madsen symptom score and a 41% (9.2 to 12.9 cm 3/sec) improvement in flow rate.4 Although the response is durable for many patients, there is a decrease of effect with time. For up to 4 years after treatment, 52% of patients required no further treatment; 36% had added medical therapy; and 11% required TURP. 4 The end of the 1990s will bring new technological improvements to transurethral microwave thermotherapy. Catheter design changes will allow better energy direction and deposition. Means for temperature monitoring will allow better dosimetry. This office-based therapy for BPH, first used clinically by Devonec and his colleagues in 1989, offers a very good chance of durable symptom relief for the patient with prostatism with a low chance of complications. CHAPTER REFERENCES
1. Abrams P. Objective evaluation of bladder outlet obstruction. Br J Urol 1995;76:11–16. 2. Ball AJ, Feneley RCL, Abrams PH. The natural history of untreated prostatism. Br J Urol 1981;53:613–616. 3. Blute ML, Bruskewitz R, Larsen TR, Mayer R, Utz W. U.S. multi-center randomized trial of a new high temperature office-based microwave system (T3) for the treatment of BPH. J Urol 1996;155:708A. 4. Blute ML, de Wildt M. Transurethral microwave thermotherapy for BPH. Contemp Urol 1996;Oct:66–80. 5. Blute ML, Patterson DE, Segura JW, Tomera KM, Hellerstein DK. Transurethral microwave thermotherapy v sham treatment: double-blind randomized study. J Endourol 1996;10:565573. 6. de la Rosette JJMCH, de Wildt MJAM, Hofner K, et al. Pressure-flow study analysis in patients treated with high energy thermotherapy. J Urol 1996;156:1428–1433. 7. de Wildt MJAM, Tubaro A, Hofner K, Carter SStC, de la Rosette JJMCH, Devonec M. Responders and nonresponders to transurethral microwave thermotherapy: a multicenter retrospective analysis. J Urol 1995;154:1775–1778. 8. Larsen TR, Bostwick DG, Corica A. Temperature-controlled correlated histopathologic changes following microwave thermoablation of obstructive tissue in patients with BPH. Urology 1996;47:463–469. 9. Ogden CW, Reddy P, Johnson H, Ramsey JWA, Carter SStC. Sham versus transurethral microwave thermotherapy in patients with symptoms of benign prostatic bladder outflow obstruction. Lancet 1993;341:14–17. 10. Wasson JH, Read DJ, Bruskewitz RC, et al. A comparison of transurethral surgery and watchful waiting for moderate symptoms of benign prostatic hyperplasia. N Engl J Med 1995;332:75–79.

Chapter 137 Interstitial Laser Therapy of Benign Prostatic Hyperplasia Glenn’s Urologic Surgery

Chapter 137 Interstitial Laser Therapy of Benign Prostatic Hyperplasia
Rolf Muschter

R. Muschter: Department of Urology, Grosshadern Hospital of Ludwig-Maximilians University of Munich, D-83177 Munich, Germany.

Diagnosis Indications for Surgery Alternative Therapy Surgical Technique Outcomes Complications Results Chapter References

Interstitial laser coagulation (ILC) of benign prostatic hyperplasia (BPH) achieved with fibers specifically designed for this purpose was first mentioned by Hofstetter in 1991,3 with the basic experiments and initial clinical results published by Muschter et al. in 1992. 9 Since then, several variations and technical and procedural devel-opments have been introduced and tested in clinical trials 2,5,7,8,12,14 (Fig. 137-1).

FIG. 137-1. Interstitial laser coagulation of the prostate.

The objective of ILC of BPH is to achieve a marked volume reduction, and to decrease urethral obstruction and both obstructive and irritative symptoms. Coagulation necrosis is generated well inside the adenoma, rather than at its urethral surface. Because the applicator can be inserted as deeply and as often as necessary, it is possible to coagulate any amount of tissue at any desired location. Post-procedurally, the intraprostatic lesions result in secondary atrophy and regression of the prostate lobes, rather than sloughing of necrotic tissue 6,7,8 and 9 (Fig. 137-1).

DIAGNOSIS
The common diagnostic procedures usually required for transurethral resection of the prostate (TURP) are sufficient for ILC. Special radiologic or endoscopic examinations are not necessary. For monitoring the success of ILC during follow-up, American Urological Association (AUA) symptom score, and quality of life index, urinary flow rate, residual urine volume, and prostate volume should be measured before therapy. Pressure–flow studies can avoid treating patients with, for example, predominantly neurogenic bladder problems. ILC does not provide tissue for pathologic examination. Therefore, evaluation with digital rectal exam-ination and serum prostate-specific antigen level is essential. Patients presenting with active urinary tract infection, acute epididymitis, or acute prostatitis must be treated with appropriate antibiotics until complete recovery is achieved before having ILC.

INDICATIONS FOR SURGERY
Candidates for ILC are all patients with moderate and severe voiding symptoms due to mechanical obstruction of the bladder outlet due to BPH who are otherwise candidates for surgical intervention, either transurethral or open. In principle, there is no limit to the prostate volume; the middle lobe can be treated too. Concomitant diseases, such as strictures of the urethra or bladder calculi, can be treated in the same session. Because ILC has essentially no major morbidity and can be performed with local anesthesia, even high-risk patients who are not candidates for TURP need not be excluded from treatment. Recurrent or residual BPH after previous prostate surgery, laser treatment, or microwave treatment can also be treated by ILC. Patients presenting with bladder cancer close to or at the bladder neck, or with possible invasion of the prostate, should not receive ILC. It is also contraindicated in cases of acute prostatitis or epididymitis and in the case of prostate abscess. Chronic prostatitis, however, which is frequently present in BPH patients, is no contraindication to ILC. Patients with symptomatic BPH who are suspected to have prostate cancer but not apt for curative treatment can be treated with ILC. If a curative therapy would be possible, however, screening biopsies should be done and examined before the planned laser treatment. If positive, these patients should not receive interstitial laser therapy.

ALTERNATIVE THERAPY
ILC was initially designed as a treatment option for patients with BPH of a severe degree unfit for surgery as an alternative to life-long catheterization. 3,9 Due to the very promising results of the first series of such patients, the indication was expanded. 7,8 Today, ILC should be seen as an alternative to any surgical or minimally invasive procedure to treat BPH.

SURGICAL TECHNIQUE
Because of their relatively deep penetration in water, efficient volumetric heating permitting necrotic temperatures deep into tissues, and the ability to be delivered with flexible optical fibers, neodymium/yttrium-aluminum-garnet (Nd:YAG) lasers (1064 nm) or diode lasers (approximately 800–1000 nm) are used for interstitial laser coagulation.9,10 and 11 Recent preliminary experiments demonstrated that the holmium-YAG (Ho:YAG) laser (2120 nm) is also suitable to generate interstitial lesions. Fibers employed for ILC must emit laser radiation at a relatively low power density. The currently most commonly used fiber emits the laser radiation circumferentially forward directed with a ring/cone-shaped beam profile (e.g., ITT Light Guide, Dornier, Germering, Germany and Kennesaw, GA; Fig. 137-2). Another type of fiber is a cylindrical diffuser tip emitting in all directions from the whole length of the applicator (e.g., Diffusor-Tip, Indigo, Cincinnati, OH; Fig. 137-3).

FIG. 137-2. Radiation pattern of the ITT Light Guide (Dornier, Germering, Germany and Kennesaw, GA) in water, laser fiber 600 microns, applicator 1.9 mm in diameter, 2 cm in length.

FIG. 137-3. Diffusor-Tip (Indigo, Cincinnati, OH); laser fiber 400 microns, applicator 2 mm in diameter, 2 cm in length, homogeneous diffuse radiation over 1 cm.

The optimal radiation parameters vary for different laser wavelength and applicator combinations. 10,11 Using constant laser power in the 5- to 7-W range, the maximal coagulation volume can be expected within approximately 10 minutes irradiation time without the risk of carbonization. 9 More rapid heating by using higher laser power reduces irradiation time but increases the risk of carbonization. 10 For short irradiation times, however, high powers are tolerable. In this type of laser energy application, irradiation starts with a relatively high power to rapidly heat the tissue and coagulate the blood vessels (e.g., 20 W for 30 seconds or 50 W for 10 seconds).7,12 The laser power is then reduced repeatedly to maintain the temperature in the center of the lesion at a high level just below the carbonization threshold and to allow further lesion growth ( Fig. 137-4, top). On-line temperature monitoring by use of an integrated thermocouple allows further optimization by use of a feedback control (Fig. 137-4, bottom).2 Optical feedback systems can also detect any carbonization. If this occurs at the applicator, laser irradiation is automatically terminated. This prevents overheating and therefore potential thermal fiber damage.

FIG. 137-4. (top) Standard irradiation program for ILC (for Nd:YAG laser and ITT Light Guide. (bottom) Various examples of possible laser power patterns in different individual applicator positions (for 830-mm diode laser and Diffusor-Tip.)

ILC can be performed with a local (e.g., periprostatic block), regional (e.g., spinal), or systemic (e.g., analgosedation) anesthetic. The procedure is suitable for the treatment of outpatients. ILC can be performed using the transurethral approach, in which the laser fiber is introduced from a cystoscope within the urethra, or the percutaneous (perineal) approach, in which the laser applicator is introduced through hollow needles in the perineum, guided by transrectal ultrasound. For the latter approach, an aiming device, as is used for the perineal placement of seeds for local radiation therapy of prostate cancer, is helpful. For penetrating the skin and guiding the fiber into position into the prostate, a special cannula large enough to accommodate the applicator is used. With an outer diameter of approximately 2.1 mm and sharpened tip, this cannula can be inserted without dilation. The commonly used transurethral ILC requires a rigid cystoscope with a working channel large enough to accommodate the fiber, which is usually between 5 and 6 Fr. The viewing angle of the telescope can be any in the range from 0 to 30 degrees. An ideal instrument has a small separate working channel that ends at the level of the telescope for optimal stabilization of the fiber during puncture ( Fig. 137-5). For cooling, continuous irrigation is not required throughout the procedure, but it will optimize visualization of the prostate and laser fiber.

FIG. 137-5. Standard cystoscope with special working element for interstitial laser coagulation for optimal fiber stabilization.

The goal of ILC is to treat a sufficient volume in each lateral lobe and in the median lobe (if needed) to produce therapeutic results while preserving healthy tissues. The total number of fiber placements is dictated by the total prostate volume and configuration. Any number of placements can be used. Generally it is better to treat one extra site than one too few. A general guideline would be a minimum of one to two applicator placements for each estimated 5 to 10 cm 3 of BPH tissue. Individual placements of the laser fiber can be spaced by about 0.5 to 1 cm and/or be performed at different angles and depths ( Fig. 137-6) to minimize substantial overlap of treatment volumes, which is not harmful but wastes time, but should be close enough that there are no gaps of untreated tissue. Fiber placement is basically limited by proximity of the produced treatment volume to the prostatic urethra, prostatic capsule, and points along the prostatic urethra distal to the bladder neck and proximal to the external sphincter.

FIG. 137-6. Scheme of possible applicator placements.

In general, the sites for fiber placement should be chosen where the mass or bulk of hyperplastic tissue is found. In the sagittal plane, fiber penetration should be in the direction of the urethra. With the patient in the lithotomy position, the direction of the prostatic urethra in most cases is approximately parallel to the operation table at the apex, turning ventrally near the bladder ( Fig. 137-7). Therefore in the apical zone it is best to puncture the lateral lobes at the 3 o'clock and 9 o'clock positions and penetrate parallel to the operation table. When closer to the bladder neck, it is suitable to penetrate more ventrally at the 2 o'clock and 10 o'clock positions. In larger prostates, multiple punctures at different fan-shaped angles are required.

FIG. 137-7. Applicator penetration orientations along the prostatic urethra.

In ILC, it is desired to preserve the urethra. If it is coagulated, however, ordinary tissue sloughing may occur and is not harmful (the treatment effect is then much like that of deep transurethral free beam coagulation). Avoiding coagulation of the urethra requires both a minimum depth and angle of fiber penetration. The applicator should be inserted at least to a depth of 0.5 cm beyond the irradiating part. In most types of applicators this corresponds to their proximal end or to a fiber depth marker. The fiber penetration angle in relation to the longitudinal axis of the urethra should be the maximum achievable. In the apical zone, this is approximately 30 to 40 degrees. Closer to the bladder neck it is often less because of the convex shape of the lobe. For effective treatment of this part of the lobe it is also possible to insert the fiber at a point closer to the apex at a lower angle and then carefully penetrate deeper toward the bladder neck ( Fig. 137-6). Fiber penetration of the capsule is almost impossible because of the limited penetration angle and depth achievable. In addition, the highly vascular nature of the prostate capsule acts as a heat sink and contributes to preventing potential coagulation of the capsule itself and adjacent structures. However, one should avoid advancing the applicator dorsally because there is little or no BPH tissue to treat and yet there is a potential risk of affecting structures adjacent to the prostate (neurovascular bundles and rectum) or inducing subtrigonal lesions with subsequent bladder neck strictures. All puncture points must be within the length of the prostatic urethra between the external sphincter and the bladder neck. The most apical puncture should be in the prostatic urethra just in front of the proximal end of the external sphincter so that no apical tissue goes untreated ( Fig. 137-8). The puncture closest to the bladder neck should not penetrate into the bladder. Accidental penetration of the applicator through the prostate into the bladder can be detected by a feeling of resistance to fiber penetration. If this occurs the fiber can simply be pulled back into the prostate and that site can be treated. When the median lobe is being treated, the fiber should always be advanced in the direction of the bladder to prevent subtrigonal penetration. A large median lobe should be treated with more than one puncture ( Fig. 137-7). Depending on its size and shape, punctures can be made at different levels and angles.

FIG. 137-8. Endoscopy during interstitial laser coagulation of the right apex, applicator in situ.

Postoperative (transurethral and/or suprapubic) catheterization is recommended. Bladder irrigation is usually not required but can be useful in individual cases to prevent clot retention. Sufficient voiding can be expected within 1 to 2 weeks in patients with normal detrusor function.

OUTCOMES

Complications No study reported any occurrence of impotence or sustained incontinence. Retrograde ejaculation occurred occasionally, with reported rates varying from 0 to 11.9%. The need for repeat BPH treatment because of treatment failure occurred at varying rates between 0% and 15.4% ( Table 137-1). Transient irritative symptoms, such as urgency, occurred in 0 to 12.6% of cases. Transient urge or stress incontinence was rare (less than 1%). Urethral strictures or bladder neck strictures were reported in a frequency of approximately 5% for the first series of patients but were not observed in subsequent series ( Table 137-2 ). Uncomplicated urinary tract infections occurred relatively frequently.

TABLE 137-1. Failure rate of ILC

Most complications, in particular the more serious ones, became less common or absent with increasing experience ( Table 137-2).

TABLE 137-2. Complications of ILC

Results Several studies indicated the effectiveness of interstitial laser coagulation of BPH regarding all of the three characteristics of the disease: symptoms, obstruction, and enlargement. All studies reported marked improvements in AUA score, peak flow rate, residual urine volume, and prostate volume 1,2,4,5,7,8,12,13 and 14 (Table 137-3 and Table 137-4). The latter was also demonstrated in studies using magnetic resonance imaging for volume measurements during follow-up. 1,6 Multivariate analysis was performed on the largest of the studies. 8 It showed no factors predicting final success or failure, such as initial symptoms, flow rates, residual volumes or prostate volumes, endoscopic or perineal access, etc., except the number of previous cases performed (effect of the learning curve). The latter can also be concluded from the results of consecutive multicentric studies of the same group of authors. 2,15 Some authors demonstrated in their multivariate analysis that the results correlated with the number of punctures per prostate volume, with less than one application per 5 to 7 ml of prostate volume the results were less good. 1

TABLE 137-3. Clinical results of ILC

TABLE 137-4. Decrease of prostate volume after ILC

Studies were also performed to compare the results with ILC to those of other laser techniques 4 and primarily TURP. 6,15 In one series, 48 patients were interstitially treated with standardized instrumentation (Nd:YAG laser, ITT Light Guide), application technique (transurethral application with an integrated cystoscope), and laser

parameters (power stepwise reduced from 20 to 7 W, 3 minutes total irradiation time per single-fiber placement), and compared prospectively with 49 TURP patients. 6 On average, within 12 months AUA score improved from 31.0 to 2.3 points (TURP: 31.1 to 3.5 points), life quality index from 4.7 to 0.6 points (TURP: 4.7 to 1.3 points), peak flow rate ( Fig. 137-9) from 9.4 ml/sec to 19.7 ml/sec (TURP: 8.9 ml/sec to 25.6 ml/sec), residual urine volume from 128 ml to 17 ml (TURP: 167 ml to 7 ml), and prostate volumes from 47.1 ml to 27.5 ml (TURP: 40.2 ml to 21.2 ml). Four laser patients were not completely satisfied and retreated by transurethral resection. The persistence of obstruction in these patients was caused by small tissue residues in unfavorable locations such as the apex or the bladder neck. In a prospectively randomized multicenter study a 10-W diode laser system was used. 15 Six months follow-up in 166 patients showed marked improvements in both groups; TURP, however, was significantly better (AUA score improving from 22.4 to 6.5 versus ILC from 21.5 to 9.7; peak flow rate improving from 8.3 ml/sec to 20.3 ml/sec versus ILC from 8.3 ml/sec to 14 ml/sec). The number of applications, however, was only 6.7 in average in an average prostate volume of approximately 50 ml.

FIG. 137-9. TURP versus interstitial laser coagulation. Urinary peak flow before and 12 months after treatment.

Urodynamic measurements were done in a limited number of patients before and after treatment. Pressure–flow studies demonstrated a sufficient decrease of intravesical/detrusor pressure, urethral opening pressure, and urethral resistance 4 (Table 137-5).

TABLE 137-5. Decrease of prostate volume after ILC

CHAPTER REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Arai Y, Ishitoya S, Okubo K, Suzuki Y. Transurethral interstitial laser coagulation for benign prostatic hyperplasia: treatment outcome and quality of life. Br J Urol 1996;77:93–98. De La Rosette JJMCH, Muschter R, Lopez MA. Interstitial laser coagulation in the treatment of BPH using a tissue adaptive laser system. J Endourol 1996;10(Suppl 1):S93. Hofstetter A. Interstitielle Thermokoagulation (ITK) von Prostatatumoren. Lasermedizin 1991;7:179. Horninger W, Janetschek G, Pointner J, Watson G, Bartsch G. Are TULIP, interstitial laser and contact laser superior to TURP? J Urol 1995;153:413A. McNicholas T, Alsudani M. Interstitial laser coagulation therapy for benign prostatic hyperplasia. SPIE Proc 1996;2671:300–308. Muschter R. Interstitial laser therapy. Curr Opin Urol 1996;6:33–38. Muschter R, Hofstetter A. Technique and results of interstitial laser coagulation. World J Urol 1995;13:109–114. Muschter R, Hofstetter A. Interstitial laser therapy outcomes in benign prostatic hyperplasia. J Endourol 1995;9:129–135. Muschter R, Hofstetter A, Hessel S, Keiditsch E, Rothenberger K-H, Schneede P, Frank F. Hi-tech of the prostate: interstitial laser coagulation of benign prostatic hypertrophy. In: Anderson RR, ed. Laser surgery: advanced characterization, therapeutics, and systems. SPIE Proc 1992;1643(111):25–34. Muschter R, Perlmutter AP. The optimization of laser prostatectomy. II. Other lasing techniques. Urology 1994;44:856–861. Muschter R, Perlmutter AP, Anson K, et al. Diode lasers for interstitial laser coagulation of the prostate. In: Anderson RR, ed. Lasers in surgery: advanced characterization, therapeutics, and systems. SPIE Proc 1995;2395(5):77–82. Muschter R, Sroka R, Perlmutter AP, Schneede P, Hofstetter A. High power interstitial laser coagulation of benign prostatic hyperplasia. J Endourol 1996;10 (Suppl 1):S197. Orovan WL, Whelan JP. Neodynium YAG laser treatment of BPH using interstitial thermotherapy: a transurethral approach. J Urol 1994;151:230A. Roggan A, Handke A, Miller K, Moller G. Laser induced interstitial thermotherapy of benign prostatic hyperplasia: basic investigations and first clinical results. Min Invas Medizin 1994;5:55–63. Whitfield N. A randomized prospective multicenter study eval-uating the efficacy of interstitial laser coagulation. J Urol 1996;155:318A.

Color Plate Glenn’s Urologic Surgery

Color Plate

COLOR PLATE 1. Normal appearance of the bladder urothelium before hydrodistention in a patient with symptoms consistent with interstitial cystitis. (B) Same patient following hydrodistention. The urothelium is abnormal, revealing minimal to moderate glomerulation. (C) Cystoscopic appearance of a patient with moderate glomerulations and submucosal hemorrhage. (D) Hunner's ulcer with marked hemorrhage surrounding the ulcer. This patient was successfully treated with focal Nd:YAG laser ablation therapy. (see Fig. 30-1.)

COLOR PLATE 2. Gangrene of the right testicle in a boy 16 years old who had undergone a 720-degree twist of the cord. (See Fig. 64-1.)

COLOR PLATE 3. Color Doppler sonography: avascular pattern of the testis 10 hours after torsion in a 16-year-old boy. (See Fig. 64-5.)