Surgical Management of Hepatobiliary and Pancreatic Disorders

Published on December 2016 | Categories: Documents | Downloads: 121 | Comments: 0 | Views: 2218
of 630
Download PDF   Embed   Report

Comments

Content

Hepatobiliary and Pancreatic Disorders
Second Edition

Surgical Management of

Edited by

Graeme J. Poston Michael D’Angelica René Adam

Surgical Management of Hepatobiliary and Pancreatic Disorders

Surgical Management of Hepatobiliary and Pancreatic Disorders
Second Edition Edited by Graeme J. Poston MS, FRCS (ENG), FRCS (ED)
Centre for Digestive Diseases University Hospital Aintree and Department of Surgery The Royal Liverpool University Hospitals Liverpool, UK

Michael D’Angelica MD
Weill Medical College of Cornell University and Memorial Sloan-Kettering Cancer Center New York, New York, USA and

René Adam MD, PHD
AP-HP Hôpital Paul Brousse Centre Hépato-Biliaire Villejuif, France

First published in 2003 by M. Dunitz Ltd, United Kingdom This edition published in 2010 by Informa Healthcare, Telephone House, 69-77 Paul Street, London EC2A 4LQ, UK. Simultaneously published in the USA by Informa Healthcare, 52 Vanderbilt Avenue, 7th floor, New York, NY 10017, USA. © 2011 Informa UK Ltd, except as otherwise indicated. No claim to original U.S. Government works. Reprinted material is quoted with permission. Although every effort has been made to ensure that all owners of copyright material have been acknowledged in this publication, we would be glad to acknowledge in subsequent reprints or editions any omissions brought to our attention. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, unless with the prior written permission of the publisher or in accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP, UK, or the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA (http://www.copyright.com/ or telephone 978750-8400). Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. This book contains information from reputable sources and although reasonable efforts have been made to publish accurate information, the publisher makes no warranties (either express or implied) as to the accuracy or fitness for a particular purpose of the information or advice contained herein. The publisher wishes to make it clear that any views or opinions expressed in this book by individual authors or contributors are their personal views and opinions and do not necessarily reflect the views/opinions of the publisher. Any information or guidance contained in this book is intended for use solely by medical professionals strictly as a supplement to the medical professional’s own judgement, knowledge of the patient’s medical history, relevant manufacturer’s instructions and the appropriate best practice guidelines. Because of the rapid advances in medical science, any information or advice on dosages, procedures, or diagnoses should be independently verified. This book does not indicate whether a particular treatment is appropriate or suitable for a particular individual. Ultimately it is the sole responsibility of the medical professional to make his or her own professional judgements, so as appropriately to advise and treat patients. Save for death or personal injury caused by the publisher’s negligence and to the fullest extent otherwise permitted by law, neither the publisher nor any person engaged or employed by the publisher shall be responsible or liable for any loss, injury or damage caused to any person or property arising in any way from the use of this book. A CIP record for this book is available from the British Library. ISBN-13: 978-1-84184-693-4 Orders may be sent to: Informa Healthcare, Sheepen Place, Colchester, Essex CO3 3LP, UK Telephone: +44 (0)20 7017 5540 Email: [email protected] Website: http://informahealthcarebooks.com/ For corporate sales please contact: [email protected] For foreign rights please contact: [email protected] For reprint permissions please contact: [email protected]

Typeset by Exeter Premedia Services Printed and bound in the United Kingdom

Contents
List of contributors Foreword Preface I ANATOMY/IMAGING/SURGICAL TECHNIQUE 1 Surgical anatomy of the liver and bile ducts
Robert Jones and Graeme J. Poston

vii x xi

16 Management of neuroendocrine tumor hepatic metastasis
Kaori Ito

154 166 173

17 Noncolorectal, nonneuroendocrine metastases
C. Kahlert, R. DeMatteo, and J. Weitz

1 17 24 36 46 53 63 73 81 89

18 Chemotherapy-associated hepatotoxicity
Martin Palavecino, Daria Zorzi, and Jean-Nicolas Vauthey

2 Anatomy of the pancreas
Margo Shoup and Jason W. Smith

19 Thermal ablation of liver metastases
Samir Pathak and Graeme J. Poston

180

3 Hepatic resection
Ajay V. Maker and Michael D’Angelica

4 Ultrasound for HPB disorders
Duan Li and Lucy Hann

ii. Primary 20 Resection for hepatocellular carcinoma
Rajesh Satchidanand, Stephen W. Fenwick, and Hassan Z. Malik

192

5 Liver surgery in elderly patients
Gerardo Sarno and Graeme J. Poston

6 Small solitary hepatic metastases: when and how?
David L. Bartlett and Yuman Fong

21 Treatment of laparoscopically discovered gallbladder cancer
Jason K. Sicklick, David L. Bartlett, and Yuman Fong

197 208

7 Managing complications of hepatectomy
Fenella K. S. Welsh, Timothy G. John, and Myrddin Rees

22 Liver transplantation for HCC: Asian perspectives
Shin Hwang, Sung-Gyu Lee, Vanessa de Villa, and Chung Mao Lo

8 Pancreatic resection
Thilo Hackert, Moritz Wente, and Markus W. Büchler

9 Surgical complications of pancreatectomy
Steven C. Katz and Murray F. Brennan

23 Non-surgical treatment of hepatocellular carcinoma
Ghassan K. Abou-Alfa and Karen T. Brown

216 223 229

10 Laparoscopy in HPB surgery
Nicholas O’Rourke and Richard Bryant

24 Resection of intrahepatic cholangiocarcinoma
Junichi Arita, Norihiro Kokudo, and Masatoshi Makuuchi

11 Cross-sectional imaging for HPB disorders (MRI and CT)
Lawrence H. Schwartz

25 Transplantation for hilar cholangiocarcinoma 100
Julie K. Heimbach, Charles B. Rosen, and David M. Nagorney

26 Rare vascular liver tumors
Jan P. Lerut, Eliano Bonaccorsi-Riani, Giuseppe Orlando, Vincent Karam, René Adam, and the ELITA-ELTR Registry

233

II LIVER A. Malignant i. Metastases 12 Liver metastases: detection and imaging
Valérie Vilgrain, Ludovic Trinquart, and Bernard Van Beers

109 118 135

B. Benign 27 Management of recurrent pyogenic cholangitis
W. Y. Lau and C. K. Leow

242 253 261 271

13 Surgery for metastatic colorectal cancer
René Adam and E. Hoti

28 Liver abscess: amebic, pyogenic, and fungal
Purvi Y. Parikh and Henry A. Pitt

14 Chemotherapy for metastatic colorectal cancer
Derek G. Power and Nancy E. Kemeny

29 Benign solid tumors of the adult liver
Mark Duxbury and O. James Garden

15 Multimodal approaches to the management of colorectal liver metastases
Gerardo Sarno and Graeme J. Poston

148

30 Liver trauma
Timothy G. John, Myrddin Rees, and Fenella K. Welsh

v

CONTENTS
31 Portal hypertension
Michael D. Johnson and J. Michael Henderson

280

IV PANCREAS A. Malignant

32 Liver transplantation for acute and chronic liver failure
Vincent Kah Hume Wong and J. Peter A. Lodge

288 301 308

42 Adenocarcinoma of the pancreas
André L. Mihaljevic, Jörg Kleeff, and Helmut Friess

380 401 407 414 432

33 Benign cystic disease of the liver
Stephen W. Fenwick and Dowmitra Dasgupta

43 Palliation of pancreas cancer
Michael G. House and Keith D. Lillemoe

34 Management of hydatid disease of the liver
Adriano Tocchi

44 Cystic tumors of the pancreas
Peter J. Allen and Murray F. Brennan

35 Surgical management of primary sclerosing cholangitis
Jason A. Breaux and Steven A. Ahrendt

45 Neuroendocrine pancreatic tumors 324
Stephen N. Hochwald and Kevin Conlon

46 Rare tumors of the pancreas
Jooyeun Chung, Lisa J. Harris, Hamid Abdollahi, and Charles J. Yeo

III BILE DUCTS AND GALLBLADDER A. Malignant 36 Management of advanced gallbladder cancer
Hiromichi Ito and William R. Jarnagin

B. Benign 329 333 47 Acute pancreatitis
C. Ross Carter, A. Peter Wysocki, and Colin J. McKay

439 451

37 Extrahepatic cholangiocarcinoma
Yuji Nimura

48 Chronic pancreatitis
Jakob R. Izbicki, Oliver Mann, Asad Kutup, and Kai A. Bachmann

38 Endoscopic management of malignant biliary obstruction
Nick Stern and Richard Sturgess

343

49 Pancreatic injury
Demetrios Demetriades, Beat Schnüriger, and Galinos Barmparas

463

B. Benign 39 Choledochal cyst detected in adulthood
Bilal Al-Sarireh and Hassan Malik

50 Pancreas transplantation 354 360
Khalid Khawaja

470

40 Bile duct injuries and benign biliary strictures
Steven M. Strasberg

V PEDIATRIC HPB DISORDERS 51 Pediatric HPB disorders 478 489
Maureen McEvoy and Michael P. La Quaglia

41 Gallstones and common bile duct stones—surgical and non-surgical approaches
Matthew P. Dearing and Michael Rhodes

373

Index

vi

List of contributors
Ghassan K. Abou-Alfa MD Assistant Attending, Memorial Sloan-Kettering Cancer Center, and Assistant Professor, Weill Medical College at Cornell University, New York, New York, USA Hamid Abdollahi MD Senior Resident (General Surgery), Department of Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania, USA René Adam MD, PhD AP-HP Hôpital Paul Brousse, Centre Hépato-Biliaire, Inserm, Unité 785, and Université Paris-Sud, UMR-S 785, Villejuif, France Steven A. Ahrendt MD Associate Professor of Surgery, University of Pittsburgh Medical Center, UPMC Passavant Cancer Center, Pittsburgh, Pennsylvania, USA Peter J. Allen MD Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Bilal Al-Sarireh MBBCh, FRS, PhD Consultant Hepatopancreatobiliary and Laparoscopic Surgeon, Swansea University, and Department of Surgery, Morristown Hospital, Swansea, UK Junichi Arita MD, PhD Hepato-Biliary-Pancreatic Surgery Division, Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Bunkyo-ku, Tokyo, Japan Kai A. Bachmann Department of General, Visceral and Thoracic Surgery, University Medical Center Hamburg-Eppendorf, Hamburg, Germany Galinos Barmparas Division of Trauma and Surgical Critical Care, University of Southern California, Los Angeles, California, USA David L. Bartlett Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, and National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA Eliano Bonaccorsi-Riani Th. STARZL Abdominal Transplant Unit, Cliniques Universitaires St Luc Université catholique de Louvain, Department of Abdominal and Transplantation Surgery, Brussels, Belgium Jason A. Breaux MD Surgical Oncology Fellow, University of Pittsburgh Medical Center, UPMC Cancer Pavilion, Pittsburgh, Pennsylvania, USA Murray F. Brennan Benno C. Schmidt Clinical Chair in Oncology, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Karen T. Brown MD Attending Radiologist, Memorial Sloan-Kettering Cancer Center, and Professor of Clinical Radiology, Weill Medical College at Cornell University, New York, New York, USA Richard Bryant MBBS, FRACS Royal Brisbane Hospital, Brisbane, Queensland, Australia Markus W. Büchler Department of General Surgery, University of Heidelberg, Heidelberg, Germany C. Ross Carter West of Scotland Pancreatic Unit, Glasgow Royal Infirmary, Glasgow, Scotland, UK Jooyeun Chung MD Department of Surgery, The Methodist Hospital, Houston, Texas, USA Kevin Conlon Professor of Surgery, University of Dublin, Trinity College Dublin, and Professorial Surgical Unit, Education Centre, AMNCH, Dublin, Ireland Michael D’Angelica MD Weill Medical College of Cornell University and Memorial Sloan-Kettering Cancer Center, New York, New York, USA Dowmitra Dasgupta MD, FRCS Consultant Hepato-Pancreatico-Biliary Surgeon, Department of Upper GI Surgery, Castle Hill Hospital, Cottingham, UK Matthew P. Dearing Department of Surgery, Norfolk & Norwich University Hospital, Norwich, UK R. DeMatteo Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Demetrios Demetriades Division of Trauma and Surgical Critical Care, University of Southern California, Los Angeles, California, USA Mark Duxbury Clinical Surgery, University of Edinburgh Royal Infirmary, Edinburgh, UK Stephen W. Fenwick MD, FRCS Consultant Hepatobiliary Surgeon, North Western Hepatobiliary Unit, University Hospital Aintree, Lower Lane, Liverpool, UK Yuman Fong MD Hepatobiliary Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Helmut Friess Chirurgische Klinik und Poliklinik, Klinikum rechts der Isar, Technische Universität München, Munich, Germany O. James Garden Regius Professor of Clinical Surgery, Clinical and Surgical Sciences (Surgery), University of Edinburgh, Royal Infirmary, Edinburgh, UK Thilo Hackert Department of Surgery, University of Heidelberg, Heidelberg, Germany Lisa J. Harris MD Senior Resident (General Surgery), Department of Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania, USA J. Michael Henderson Chief Quality Officer, Cleveland Clinic, Cleveland, Ohio, USA Stephen N. Hochwald MD Chief, Division of Surgical Oncology, University of Florida, Gainesville, Florida, USA Michael G. House MD Assistant Professor, Department of Surgery, Indiana University School of Medicine, Indianapolis, Indiana, USA

vii

LIST OF CONTRIBUTORS
Lucy Hann MD Professor of Radiology, Weill Cornell Medical Center, and Director of Ultrasound Memorial Sloan-Kettering Cancer Center, New York, New York, USA Julie K. Heimbach Mayo Clinic, Rochester, Minnesota, USA Steven N. Hochwald University of Florida Medical School, Box 100286, Gainesville, FL 32610–0286, USA E. Hoti AP-HP Hôpital Paul Brousse, Centre Hépato-Biliaire, Villejuif, France, and Liver Transplant Unit, Saint Vincent’s University Hospital, Dublin, Ireland Shin Hwang Professor, Division of Hepatobiliary Surgery and Liver Transplantation, Department of Surgery, University of Ulsan College of Medicine, Seoul, Korea Hiromichi Ito MD Department of Surgery, Michigan State University, Lansing, Michigan, USA Kaori Ito MD Department of Surgery, Michigan State University, Lansing, Michigan, USA Jakob R. Izbicki FACS Department of General, Visceral and Thoracic Surgery, University Medical Center Hamburg-Eppendorf, Hamburg, Germany William R. Jarnagin MD Hepatobiliary Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Timothy G. John MD, FRCSEd (Gen) Hepatobiliary Unit, Basingstoke and North Hampshire Hospitals NHS Foundation Trust, Basingstoke, UK Michael D. Johnson MD Digestive Disease Institute, Cleveland Clinic, Cleveland, Ohio, USA Robert Jones MB, ChB, MRCS Clinical Fellow, North Western Hepatobiliary Centre, Aintree University Hospital, Liverpool, UK C. Kahlert Department of Surgery, University of Heidelberg, Heidelberg, Germany Vincent Karam Centre Hépatobiliaire, Hôpital Paul Brousse, Villejuif, France Steven C. Katz MD Director of Surgical Immunotherapy, Roger Williams Medical Center, Providence, Rhode Island, USA Khalid Khwaja MD Director of Kidney and Pancreas Transplantation, Senior Staff Surgeon, Lahey Clinic, Burlington, Massachusetts, USA Nancy E. Kemeny MD Memorial Sloan-Kettering Cancer Center, New York, New York, USA Jörg Kleeff Department of Surgery, Klinikum rechts der Isar, Technische Universität München, Munich, Germany Norihiro Kokudo MD, PhD Hepato-Biliary-Pancreatic Surgery Division, Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Bunkyo-ku, Tokyo, Japan Asad Kutup Department of General, Visceral and Thoracic Surgery, University Medical Center Hamburg-Eppendorf, Hamburg, Germany W. Y. Lau Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong, SAR C. K. Leow Mount Elizabeth Medical Centre, Singapore, Singapore Keith D. Lillemoe MD Jay L. Grosfeld Professor and Chairman, Department of Surgery, Indiana University School of Medicine, Indianapolis, Indiana, USA Sung-Gyu Lee Professor, Division of Hepatobiliary Surgery and Liver Transplantation, Department of Surgery, University of Ulsan College of Medicine, Seoul, Korea Michael P. La Quaglia MD Department of Surgery, Pediatric Surgery Service, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Jan P. Lerut MD, PhD, FACS Th. STARZL Abdominal Transplant Unit, Cliniques Universitaires St Luc Université catholique de Louvain, Department of Abdominal and Transplantation Surgery, Brussels, Belgium Duan Li MD Assistant Attending Radiologist, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Chung Mao Lo Professor, Department of Surgery, Queen Mary Hospital, The University of Hong Kong, Hong Kong, China J. Peter A. Lodge MD, FRCS Professor and Clinical Director, HPB & Transplant Unit, St. James’ University Hospital, Leeds, UK Ajay V. Maker MD Director of Surgical Oncology, Creticos Cancer Center–Advocate Illinois Masonic Medical Center; Departments of Surgery and Microbiology/Immunology, University of Illinois at Chicago, Chicago, Illinois, USA Masatoshi Makuuchi MD, PhD Hepato-Biliary-Pancreatic Surgery Division, Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Bunkyo-ku, Tokyo, Japan Hassan Malik MD, FRCS Hepatobiliary Unit, Department of Surgery, University Hospital Aintree, Liverpool, UK Oliver Mann Department of General, Visceral and Thoracic Surgery, University Medical Center Hamburg-Eppendorf, Hamburg, Germany Maureen McEvoy MD Department of Surgery, Pediatric Surgery Service, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Colin J. McKay West of Scotland Pancreatic Unit, Glasgow Royal Infirmary, Glasgow, Scotland, UK André L. Mihaljevic Department of Surgery, Klinikum rechts der Isar, Technische Universität München, Munich, Germany David M. Nagorney Mayo Clinic, Rochester, Minnesota, USA Yuji Nimura MD President, Aichi Cancer Center, Chikusaku, Nagoya, Japan Giuseppe Orlando Th. STARZL Abdominal Transplant Unit, Cliniques Universitaires St Luc Université catholique de Louvain, Department of Abdominal and Transplantation Surgery, Brussels, Belgium

viii

LIST OF CONTRIBUTORS
Nicholas O’Rourke MBBS, FRACS Royal Brisbane Hospital, Brisbane, Queensland, Australia Martin Palavecino MD Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas, USA Purvi Y. Parikh MD Department of Surgery, Albany Medical College, Albany, New York, USA Samir Pathak MD, ChB, MSC, MRCS Clinical Fellow, North Western Hepatobiliary Centre, Aintree University Hospital, Liverpool, UK Henry A. Pitt MD Indiana University, Indianapolis, Indiana, USA Graeme J. Poston MS, FRCS (Eng), FRCS (Ed) Centre for Digestive Diseases, University Hospital Aintree, and Department of Surgery, The Royal Liverpool University Hospitals, Liverpool, UK Derek G. Power MD Memorial Sloan-Kettering Cancer Center, New York, New York, USA Myrddin Rees MS, FRCS, FRCS (Ed) Hepatobiliary Unit, Basingstoke and North Hampshire Hospitals NHS Foundation Trust, Basingstoke, UK Michael Rhodes Department of Surgery, Norfolk & Norwich University Hospital, Norwich, UK Charles B. Rosen Mayo Clinic, Rochester, Minnesota, USA Gerardo Sarno MD Clinical Fellow, North Western Hepatobiliary Centre, Aintree University Hospital, Liverpool, UK Rajesh Satchidanand MD, FRCS Clinical Fellow, North Western Hepatobiliary Centre, Aintree University Hospital, Liverpool, UK Beat Schnüriger Division of Trauma and Surgical Critical Care, University of Southern California, Los Angeles, California, USA Lawrence H. Schwartz Department of Radiology, Columbia University College of Physicians and Surgeons, and Radiologist-in-Chief, New York–Presbyterian Hospital/ Columbia University Medical Center, New York, New York, USA Margo Shoup MD, FACS Chief, Division of Surgical Oncology, Department of Surgery, Loyola University Medical Center, Maywood, Illinois, USA Jason K. Sicklick Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, USA Steven M. Strasberg MD, FRCS(C), FACS, FRCS (Ed) Pruett Professor of Surgery and Head Hepato-Pancreato-Biliary and Gastrointestinal Surgery, Washington University in Saint Louis and Barnes-Jewish Hospital, Saint Louis, Missouri, USA Jason W. Smith MD Chief Resident, Department of Surgery, Loyola University Medical Center, Maywood, Illinois, USA Nick Stern Consultant Gastroenterologist, Digestive Diseases Department, University Hospital Aintree, Liverpool, UK Richard Sturgess Consultant Gastroenterologist and Clinical Director, Digestive Diseases Department, University Hospital Aintree, Liverpool, UK Adriano Tocchi Head of 1st Department of Surgery and Chief of the Gastro-intestinal and Hepato-biliary Surgical Service, University of Rome Sapienza Medical School, Rome, Italy Ludovic Trinquart Department of Radiology, Assistance-Publique Hôpitaux de Paris, Hôpital Beaujon, Clichy, France Bernard Van Beers Department of Radiology, Assistance-Publique Hôpitaux de Paris, Hôpital Beaujon, Clichy; Université Paris; and Centre de recherche biomédicale Bichat-Beaujon, Paris, France Jean-Nicolas Vauthey MD Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas, USA Valérie Vilgrain Department of Radiology, Assistance-Publique Hôpitaux de Paris, Hôpital Beaujon, Clichy; Université Paris; and Centre de recherche biomédicale Bichat-Beaujon, Paris, France Vanessa de Villa Assistant Professor, Department of Surgery, Queen Mary Hospital, The University of Hong Kong, Hong Kong, China J. Weitz MD Department of Surgery, University of Heidelberg, Heidelberg, Germany Fenella K. S. Welsh MA, MD, FRCS (Gen Surg) Hepatobiliary Unit, Basingstoke and North Hampshire Hospitals NHS Foundation Trust, Basingstoke, UK Moritz Wente Department of Surgery, University of Heidelberg, Heidelberg, Germany Vincent Kah Hume Wong MBCB, MRCS Research Fellow in Hepatopancreatobiliary & Transplant Surgery, HPB & Transplant Unit, St. James’ University Hospital, Leeds, UK A. Peter Wysocki Department of Surgery, Logan Hospital, Meadowbrook, Queensland, Australia Charles J. Yeo MD The Samuel D. Gross Professor and Chair, Department of Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania, USA Daria Zorzi MD Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas, USA

ix

Foreword
As recent progress in hepato-pancreato-biliary (HPB) surgery has been evident since the first edition of this book was published eight years ago, Dr. Graeme Poston, Dr. Mike D’Angelica, and Dr. René Adam, internationally recognized authorities in HPB surgery, have attempted to rewrite the second edition, joined by selected numerous worldwide specialists renowned as expert authors in each field to present a current view of the surgical and non-surgical management of benign and malignant HPB disorders. This book demonstrates the wisdom of the new knowledge and technical skills of these diverse disciplines where cooperative efforts contribute toward the benefit of the patients with HPB disorders. The general surgeon will find this volume to be a useful source of current thoughts on how to manage the diverse HPB diseases. Yuji Nimura MD President, Aichi Cancer Center Professor Emeritus, Nagoya University Graduate School of Medicine Past President, International Hepato-Pancreato-Biliary Association (IHPBA)

x

Preface
Hepato-pancreato-biliary (HPB) surgery is now firmly established within the repertoire of modern general surgery. Indeed, in many major tertiary centers there are now specific teams for both pancreatic and liver surgery. However, in most hospitals outside these major centers the day-to-day management and decision-making for patients with these disorders remains the remit of the general surgeon. Following the launch of the highly successful first edition of this book eight years ago there have been considerable advances in the surgical management of HPB disorders. Many of these relate to related specialties (radiology, oncology, gastroenterology, and anesthesia) and also directly to surgery (liver transplantation, caval bypass and replacement, laparoscopic surgery to name but a few). As such the second edition has been completely rewritten from scratch. As with the first edition, the purpose of this edition is twofold. First, it is intended to cover the spectrum of common HPB diseases that will confront the general surgeon in his or her regular practice. Second, we hope that this work will be sufficiently comprehensive to cover the broad spectrum of HPB surgery for candidates coming to examinations at the completion of surgical training. We are indebted to the many international contributors for their perseverance and patience over the gestation of this project, which is greatly appreciated. Lastly, we are grateful to our publishers, Informa Healthcare, for their help during the preparation of this project. Graeme J. Poston Michael D’Angelica René Adam September 2010

xi

1

Surgical anatomy of the liver and bile ducts Robert Jones and Graeme J. Poston
lobar anatomy (2). The first successful elective liver resection was performed two years later by von Langenbuch, who excised a portion of the left lobe of the liver containing an adenoma in 1888 (9). He had to reopen the abdomen several hours after the operation because of reactionary hemorrhage, but was able to ligate the bleeding vessels and return the oversewn liver to the abdomen. Two years later in 1890, the Baltimore surgeon McLane Tiffany reported the successful removal of a benign liver tumor (10), and the following year Lucke described the successful resection of a cancerous growth of the liver (11). Surgery was now becoming a recognized treatment for liver pathology. Advances in surgery closely mirrored increased understanding of the functional anatomy of the liver (12–14). The first attempt to define the functional anatomy of the liver, which could possibly guide current surgical practice, was made by Cantlie in 1898, while working in Hong Kong. He dissected the livers of executed prisoners (15) and making vascular casts, he demonstrated that the main division between the right and left lobe in fact extended from approximately the gallbladder fossa, to the right side of the IVC, posterosuperiorly. Cantlie’s line, therefore, follow a line drawn from the gallbladder fossa, along the middle hepatic vein, to the IVC (Figs. 1.2 and 1.3) (3). In 1911, Wendel reported the first case of right lobectomy for a primary tumor (16), however this procedure did not follow the precise anatomical plane described by Cantlie. In 1939, while working in Paris, the Vietnamese surgeon Ton That Tung described the venous drainage of the liver in relation to the true lobar anatomy (Fig. 1.4) (17). The first anatomically correct description of a left lateral segmentectomy was made by Raven in 1948 while resecting metastatic colon cancer (18). Four years later, Lortat-Jacob and Robert finally described a similar approach to the true right hepatic lobectomy, based on the anatomical principles described by Cantlie (Fig. 1.6) (19). Healey and Schroy were the first to demonstrate in 1953 that the right lobe was further divided into an anterior and a posterior sector (20). They also showed that the left lobe was divided into a medial and lateral sector by the line of the falciform ligament and umbilical vein (Fig. 1.5). Understanding of the functional anatomy of the liver continued to develop, and in 1957, Goldsmith and Woodburne described a number of anatomical planes through the liver parenchyma that followed this functional anatomy. Their paper finally defined true right lobectomy (right hepatectomy), left lobectomy (left hepatectomy), and left lateral segmentectomy (Fig. 1.6) (21).

The success of any surgical intervention on the liver and bile ducts is totally dependent on a thorough working knowledge of their anatomy. As the number of patients undergoing hepatobiliary surgery is increasing, good understanding of the anatomy of this area is increasingly important for any surgeon with an interest in the gastrointestinal tract. Command of this anatomy is also essential for the successful interpretation of functional imaging of hepatobiliary anatomy. When operating on the liver and biliary tree, the surgeon has to obey three basic tenets.
● ●



Remove all pathologically involved tissue. Preserve the maximal amount of functioning nonpathological liver tissue. Perform safe resection, while ensuring adequate blood supply to the remaining hepatic parenchyma.

Historically, the liver was described according to its morphological appearance (1,2). However, these three tenets have altered the approach to surgery, and the liver is now considered from a functional and therefore surgical perspective.

morphological anatomy
Historically, when viewed at laparotomy, the liver appears divided into a larger “right” lobe, and a smaller “left” lobe by the umbilical fissure and falciform ligament (Figs. 1.1 and 1.2) (3). Situated on the inferior surface of the right lobe is the transverse hilar fissure, which constitutes the posterior limit of the right lobe. The “quadrate” lobe was defined as the portion of the right lobe lying anterior to this transverse hilar fissure and to the right of the umbilical fissure, its other margin being defined by the gallbladder fossa. The “caudate” lobe, which is anatomically and functionally separate from the rest of the liver, lies posterior to the hilum, between the portal vein and the inferior vena cava (IVC) (4). This historical anatomical approach does not consider the vasculature or biliary drainage of the liver and is of only limited use when planning surgical resection.

early application of the functional anatomy
Isolated liver wounds, usually as a result of military action, had been successfully treated since the early seventeenth century (5,7), but the first attempt at resection of a liver tumor was not made until 1886, when the French surgeon Luis excised a solid liver tumor by ligating and cutting through a pedunculated left lobe “adenoma.” Attempts to suture the severed pedicle were unsuccessful, and the stump was returned to the peritoneal cavity. Not surprisingly, the patient succumbed some six hours later (8). In 1888, Rex reported a “new” arrangement of the right and left lobes of the liver and further refined our understanding of

appreciation of segmental anatomy
Probably the most important anatomical contribution to modern liver surgery comes from the work of the late Claude

1

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
IVC Middle hepatic vein lying among Cantlie's line IVC

Left lobe Right lobe

IVC Right free border of lesser omentum Figure 1.1 Morphological anatomy.

IVC Cantlie's line Gallbladder Figure 1.3 Cantlie’s line.

Gallbladder Cantlie's line

Umbilical fissure

Quadrate lobe

supply (inflow and outflow), and therefore viability, to the remaining hepatic parenchyma. The description of Couinaud is the most complete and exact, and also the most useful for the operating surgeon, and therefore it is this description that will be used throughout this book.

segmental anatomy of the liver
Transverse hilar fissure Gastrohepatic omentum

Common bile duct, hepatic artery and portal vein

IVC Caudate lobe

Figure 1.2 Anatomical features.

Couinaud, who in 1957 produced a huge number of vasculobiliary casts of the liver (23,24). Couinaud was able to demonstrate that the liver appeared to consist of eight anatomical segments, each of which could potentially be separately resected without affecting the physiological viability of the other segments. Couinaud redefined the caudate lobe as segment 1 and Goldsmith and Woodburne’s left lobe as segments 2 and 3. The quadrate lobe was termed segment 4, and more recently has been subdivided by further studies of its portal blood supply into 4A (superiorly) and 4B (inferiorly). The right liver consists of segments 5 (anteroinferiorly), 6 (posteroinferiorly), 7 (posterosuperiorly), and 8 (anterosuperiorly) (Fig. 1.7). Couinaud later suggested a further clarification, in which the caudate lobe to the left of the IVC remained segment 1, with that to the right being redefined as segment 9 (25). Resections based on these anatomical segments enable the surgeon to safely operate following the three central tenets described above; remove all pathologically involved tissue, preserve the maximal amount of nonpathological liver tissue, and perform safe resection, while ensuring an adequate blood

These anatomical studies of the functional anatomy of the liver allow us to define hepatic segments based upon both the distribution of the portal pedicles and the drainage of the hepatic veins (Fig. 1.5). The three main hepatic veins (right, middle, and left) divide the liver into four sectors, each of which receives a portal pedicle containing branches of the hepatic artery, hepatic duct, and portal vein; thus producing an alternation between hepatic veins and portal pedicles. These four sectors, demarcated by the hepatic veins, are the portal sectors, each sector therefore receiving an independent portal supply. For the same reason, the scissurae containing the hepatic veins are termed the portal scissurae while the scissurae containing portal pedicles are the hepatic scissurae (Fig. 1.5). Thus, the liver is divided by the main portal scissura along the line of the middle hepatic vein into two discrete hemilivers, along the line previously described by Cantlie (15). We therefore refer to these hemilivers as right and left livers, rather than right and left lobes, to avoid confusion with the anatomical lobes, particularly since there is no visible surface marking that permits individualization of the “true” lobes. As described by Cantlie, the main portal scissura runs posteriorly from the middle of the gallbladder fossa to the right side of the IVC (Fig. 1.5). Therefore, the right and left livers, demarcated by the main portal scissura, are independent in terms of their portal and arterial vascularization and their biliary drainage. These right and left livers are both further divided into two by the other two portal scissurae, delineated by the right and left hepatic veins. Goldsmith and Woodburne refer to these further divisions as “segments” (21), but for the rest of this book, we will use the more generally accepted nomenclature of Couinaud, which refers to these divisions as “sectors” (23). The

2

SURGICAL ANATOMY OF THE LIVER AND BILE DUCTS
Right liver Left liver

Middle hepatic vein (usually enters left vein before IVC)

IVC Left heptic vein

Right hepatic vein

Caudate hepatic veins (variable)

Right inferior hepatic vein (variable) IVC Gallbladder, note that the middle vein may lie superficially in the gallbladder fossa Figure 1.4 Venous drainage of the liver.

IVC

Middle hepatic vein in main portal scissura following Cantlie's line Left hepatic vein in left portal scissura Lateral segment of left lobe

7 Right hepatic vein in right portal scissura

2 8 1 4

3

Right posterior sector

Falciform ligament 6 5 Medial segment of left lobe Right anterior sector Right liver Portal vein Left liver

Figure 1.5 Functional sectoral anatomy and relationship to hepatic scissurae.

right liver is divided by the right portal scissura (right portal vein) into an anteromedial (or anterior) sector containing segments 5 inferiorly and 8 superiorly, and a posterolateral (or posterior) sector containing segments 6 inferiorly and 7 superiorly (Fig. 1.5). When the liver lies in its normal position within the upper abdominal cavity, the right posterolateral sector lies directly behind the right anteromedial sector, and this scissura is therefore almost in the coronal plane. Therefore in the clinical setting (particularly when imaging the liver), it is better to speak of these anterior and posterior sectors (Fig. 1.5). The exact location of the right portal scissura is imprecise, because it has no external landmarks. According to Couinaud (23), it extends from the edge of the liver at the middle point between the back of the liver and the right side of the

gallbladder bed along the right hepatic vein posteriorly to the confluence of the right hepatic vein and the IVC (26–28). The venous drainage of the right liver is variable in that, in addition to the right and middle hepatic veins, there are often a number of smaller hepatic veins draining directly into the IVC from segments 6 and 7. Not infrequently (63–68%) segment 6 drains directly into the IVC through a distinct inferior right hepatic vein, larger than these other venous tributaries to the IVC, which can be a significant bonus in the preservation of residual hepatic function when undertaking extended left hepatectomies (Fig. 1.4) (29,30). The left portal scissura, along the left hepatic vein, divides the left liver into two sectors: an anterior sector containing segments 3 and 4 and a posterior sector containing segment 2

3

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C)

(D)

(E) Figure 1.6 Formal hepatectomies: (A) right hepatectomy; (B) left hepatectomy; (C) left lateral segmentectomy; (D) extended left hepatectomy; (E) extended right hepatectomy.

8 8 7 2 1 5 6 6 4 3 5 4 7 1

2

3

(A)

(B)

Figure 1.7 Functional division of the liver and of the liver segments according to Couinaud’s nomenclature (A) as seen in the patient and (B) in the ex vivo position.

4

SURGICAL ANATOMY OF THE LIVER AND BILE DUCTS
(Fig. 1.5). It is important to note that the left portal scissura does not follow the umbilical fissure; this portal scissura contains a hepatic vein and the umbilical fissure contains a portal pedicle. Therefore the left portal scissura lies posterior to the ligamentum teres, inside the left lobe of the liver (Fig. 1.5). The middle hepatic vein (defining the main portal scissura) usually enters the left hepatic vein some 1 to 2 cm before the left hepatic vein joins the IVC (Fig. 1.4) (30). Occasionally the middle and left hepatic veins enter the IVC separately, and in 2 out of 34 of Couinaud’s casts, the middle vein and left veins joined at more than 2.5 cm from the IVC (30). Such an anomaly must be detected and excluded during isolated resection of segment 4, since if it is not seen, and the last 2 cm of the left vein is damaged, segments 2 and 3 will be needlessly sacrificed (and in the case of extended right hepatectomy, threaten future remnant liver viability). The caudate lobe (segments 1 and 9) is the dorsal portion of the liver, lying posteriorly and surrounding the retrohepatic IVC. It lies directly between the portal vein (anteriorly) and the IVC (posteriorly). The main bulk of the caudate lobe lies to the left of the IVC, with its left and inferior margins being free in the lesser omental bursa (Fig. 1.2). The gastrohepatic (lesser) omentum separates the caudate from segments 2 and 3 of the left liver. The left portion of the caudate lobe lies inferior to the right between the left portal vein and the IVC, as the caudate process. This process then fuses inferiorly with segment 6 of the right liver. The amount of caudate lobe that lies on the right side is variable, but usually small. The anterior surface of the caudate lobe lies within the hepatic parenchyma against the posterior intrahepatic surface of segment 4, demarcated by an oblique plane slanting from the left portal vein to the left hepatic vein. The caudate lobe must be considered functionally as an isolated autonomous segment, since its vascularization is independent of the portal division and of the three main hepatic veins. It receives a variable arterial and portal blood supply from both the right and left portal structures, although the right caudate lobe consistently receives an arterial supply from the right posterior artery. Biliary drainage is likewise into both the right and left hepatic ducts. However, the left dorsal duct can also join the segment 2 duct. The small hepatic veins of the caudate lobe drain directly into the IVC. This independent functional isolation of the caudate lobe is clinically important in Budd–Chiari syndrome; if all three main hepatic veins are obliterated, the only functioning hepatic venous drainage is through the caudate lobe, which therefore undergoes compensatory hyperplasia. as they will leave behind devascularized residual liver and will also probably not adequately excise all the pathologically involved parenchyma. The usual anatomical hepatectomies can be considered in two groups: right and left hepatectomies in which the line of transection is the main portal scissura separating the right and left livers along the middle hepatic vein, and right and left hepatectomies in which the line of transection commences in the umbilical fissure. For some time the latter definition, initially proposed by Goldsmith and Woodburne (21), has been the accepted convention. We would encourage the use of the former definition, as segment 4 (quadrate lobe) is anatomically part of the left liver (Fig. 1.9), and this convention was adopted universally at the 2000 Brisbane Congress of the IHPBA (Brisbane Convention), and will be used hereafter in this book. Using this functional approach to liver anatomy, we can define numerous potential liver resections based upon the order (first, second, third) of the hepatic divisions (main portal scissura, anterior and posterior right portal scissurae, left portal scissura) (28). With regard to the first order division, right hepatectomy or hemihepatectomy (removal of the right liver/hemiliver) therefore consists of the resection of segments 5 to 8 (stipulating ± segment 1). Left hepatectomy or hemihepatectomy (removal of the left hemiliver or liver) is the removal of segments 2–4 (stipulating ± segment 1) (Fig. 1.6). In certain pathologies (multiple liver metastases or large tumors transgressing the main portal scissura) hepatectomies can be extended to include adjacent segments and sectors of the other liver. Therefore extended right hepatectomy (right trisegmentectomy or extended right hemihepatectomy) will also include resection of segment 4 (stipulating ±segment 1), taking portal structures to the right of the falciform ligament (Fig. 1.6). Similarly, extended left hepatectomy (left trisegmentectomy or extended left hemihepatectomy) would include resection of segments 5 and 8 en bloc with segments 2 to 4 (stipulating ± segment 1) (Fig. 1.6). When discussing second order divisions, individual sectors can be resected in isolation or in adjacent pairs depending upon the distribution of pathology. Therefore right anterior sectionectomy refers to the en bloc resection of segments 5 and 8 (between the main portal scissura (middle hepatic vein) and right portal scissura (right portal vein) on their pedicle of the anterior division of the right portal vein). Right posterior sectionectomy (previously referred to as right posterior or lateral sectorectomy) is the contiguous resection of segments 6 and 7, posterior to the right portal scissura (on the pedicle of the posterior division of the right portal vein) (Fig. 1.8). On the left side, isolated excision of segment 4 can be described as left median sectionectomy, although it is also legitimate to refer to it as resection segment 4 or segmentectomy 4. One area of confusion in these definitions of hepatectomies comes in the simultaneous resection of segments 2 and 3 (Fig. 1.10). Goldsmith and Woodburne originally described this procedure as a left hepatic lobectomy (21). Describing this as left lateral segmentectomy is technically wrong since the true left lateral segment (and sector) comprises no more than segment 2 (excision of which in isolation can therefore be

anatomical classification of hepatectomies
Hepatic resections can be classified as “anatomical” and “nonanatomical.” Anatomical hepatectomies (hepatectomies reglees) are defined by resection of a portion of liver parenchyma defined by the functional anatomy. These resections are called left or right hepatectomies, sectorectomies, and segmentectomies. Nonanatomical hepatectomies involve resection of a portion of hepatic parenchyma not limited by anatomical scissurae. Such resections are usually inappropriate,

5

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
described as left lateral or posterior sectorectomy). It is now accepted convention that resection of segments 2 and 3 is regarded as a left lateral sectionectomy (but can also legitimately be referred to as bisegmentectomy 2–3). With regard to the third order divisions, resection is now at the level of the individual hepatic segment(s). Therefore these resections are referred to as segmentectomy (classified according to the segment being removed: 1–9). Similarly, segments 5 and 6 can be resected en bloc (and this used to be described as a right inferior hepatectomy) and this should now be described as bisegmentectomy 5–6. If there is a significant right inferior hepatic vein draining segments 5 and 6, then segments 7 and 8 can be resected with the right hepatic vein (bisegmentectomy 7–8) (Fig. 1.8). lesser omentum being incised close to the liver. Opening the left coronary ligament allows ligation of the inferior phrenic vein. The caudate veins running directly to the IVC are now exposed and can be divided between ligatures as they run up the back of the caudate lobe. After the hilar plate is lowered to expose the right and left portal pedicles, the portal inflow to both the right and left caudate segments can be identified, ligated, and divided. The caudate lobe is now isolated and the main portal fissure is divided to separate segments 4, 7, and 8. Note that the caudate segment 1s not defined macroscopically from segment 6.

the biliary tract
Accurate biliary exposure and precise dissection are the two most important steps in any biliary operative procedure and are both totally dependent on a thorough anatomical understanding of these structures. Several authors have described the anatomy of the biliary tract (17,22,23), but unfortunately the surgical implications have been incompletely described and continue to be misunderstood by many surgeons.

surgical approach to the caudate lobe
This resection (segmentectomy 1 or 9, or 1 and 9 en bloc) is initially achieved by dissection of the coronary ligament up to the right of the IVC, being careful to avoid the right hepatic vein. The falciform ligament is then dissected to the IVC, the

(A)

(B)

(C)

(D)

(E) Figure 1.8 Other hepatic sectorectomies: (A) right posterior sectorectomy; (B) right anterior sectorectomy; (C) left medial sectorectomy (segments 4A and 4B); (D) right inferior hepatectomy; (E) right superior hepatectomy.

6

SURGICAL ANATOMY OF THE LIVER AND BILE DUCTS

intrahepatic biliary anatomy
The right liver and left liver are respectively drained by the right and the left hepatic ducts. The caudate lobe (segments 1 and 9) is drained by several ducts joining both the right and left hepatic ducts (20). The intrahepatic ducts are tributaries of the corresponding hepatic ducts, which form part of the major portal tracts invaginating Glisson’s capsule at the hilus and penetrating the liver parenchyma (Fig. 1.11). There is variation in the anatomy of all three components of the portal triad structures (hepatic ducts, hepatic arteries, and portal vein), but it is the portal vein that shows the least anatomical variability. In particular, the left portal vein tends to be consistent in location (23). Bile ducts are usually located above the portal vein whereas the corresponding artery will lie below. Each branch of the intrahepatic portal vein corresponds to one or two intrahepatic bile ducts, which converge outside the liver to form the right and left hepatic ducts, in turn joining to form the common hepatic duct. The left hepatic duct drains segments 2, 3, and 4, which constitute the left liver. The duct draining segment 3 is found a little behind the left horn of the umbilical recess, from where it passes directly posteriorly to join the segment 2 duct to the left

of the main portal branch to segment 2. At this point, the left branch of the portal vein turns forward and caudally in the recessus of Rex (23) (Figs. 1.12 and 1.13). As the duct draining segment 3 begins its posterior course it lies superficially in the umbilical fissure, often immediately under Glisson’s capsule. As such it is usually easily accessible at surgery to allow a biliary– enteric (segment 3 hepaticojejunostomy) anastomosis for biliary drainage if such access is not possible at the porta hepatis. The left hepatic duct then passes beneath the left liver at the posterior base of segment 4, lying just above and behind the left branch of the portal vein. After the left duct crosses the anterior edge of that vein it joins the right hepatic duct to form the common duct at the hepatic ductal confluence. In this transverse portion, where it lies below the liver parenchyma, it receives one to three small branches from segment 4 (23). The right hepatic duct (Fig. 1.14) drains segments 5 to 8 and arises from the convergence of the two main sectoral (anterior 5 and 8, and posterior 6 and 7) tributaries. The right posterior sectoral duct runs almost horizontally (26) and comprises the confluence of the ducts from segments 6 and 7 (Fig. 1.15). The right posterior duct joins the right anterior sectoral duct (formed by the confluence of the ducts from segments 5 and 8)

Figure 1.9 Completion of segment 4 resection with portal bifurcation lying inferiorly in front of the inferior vena cava.

Figure 1.11 Exposing the hilar plate by raising the inferior surface of segment 4B, thus demonstrating the condensation of Glisson’s capsule, which will cover the extra hepatic confluence of the right and left hepatic ducts.

Figure 1.10 Left lateral segmentectomy immediately prior to division of the portal structure lying inferiorly and the left hepatic vein lying superiorly.

Figure 1.12 Exposing the recessus of Rex by distraction of the falciform ligament to demonstrate the bifurcation of segment 3 and segment 4 bile ducts.

7

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
4 RPV 2

RHD RHA

4 (ant.) CHD 3

PV

HA

Recessus of Rex

Figure 1.13 Biliary and vascular anatomy of the left liver. Note the position of segment 3 duct above the corresponding vein and its relationship to the recessus of Rex.

as it descends vertically (26). This right anterior sectoral duct lies to the left of the right anterior sectoral branch of the intrahepatic portal vein as it ascends within the parenchyma (Fig. 1.15). The junction of the two main right biliary ducts usually occurs immediately above the right branch of the portal vein (23). The right hepatic duct is considerably shorter than its counterpart on the left, which it joins to form the common hepatic duct in front of the right portal vein (Fig. 1.15). The caudate lobe (segments 1 and 9) has its own separate biliary drainage. This segment comprises two anatomically and functionally distinct portions, a caudate lobe proper (which consists of a right and left part) located at the posterior aspect of the liver, and a caudate process passing behind the portal structures to fuse with segment 6 of the right liver. In nearly half of individuals, three separate bile ducts drain these distinct parts, while in a quarter of individuals, there is a common biliary duct between the right portion of the caudate lobe proper and the caudate process, while the left part of the caudate lobe is drained by an independent duct. However, the site of drainage of these ducts is variable. Most authors advocate en bloc resection of the caudate lobe during resection of hilar cholangiocarcinoma (31), since the tumor usually infiltrates these ducts draining the caudate lobe. Certainly these authors have demonstrated that in 88% of cases of hilar cholangiocarcinoma coming to resection there is histological evidence of tumor infiltration of the caudate lobe along these ducts.

extrahepatic biliary anatomy
The detail of this section will be confined to the upper part of the extrahepatic biliary tree, above the common bile duct, since the common bile duct is also covered in chapter 2. The right and left hepatic ducts converge at the right of the hilum of the liver, anterior to the portal venous bifurcation and overlying the origin of the right portal vein. The biliary confluence

Figure 1.14 Demonstration of the right hepatic duct lying within the gallbladder fossa.

Anterior sectoral duct

8

5 7 Posterior sectoral duct

LHD LPV 6 LHA CHD PV

HA

Figure 1.15 Biliary and vascular anatomy of the right liver. Note the horizontal course of the posterior sectoral duct and the vertical course of the anterior sectoral duct.

8

SURGICAL ANATOMY OF THE LIVER AND BILE DUCTS
is separated from the posterior aspect of the base of segment 4 by a fusion of connective tissue investing from Glisson’s capsule to form the fibrous hilar plate. This hilar plate has no vascular interposition and, when opened behind the posterior aspect of the base of segment 4, will display the extrahepatic confluence of the right and left hepatic ducts (Fig. 1.16). The main bile duct is divided into its upper part, the common hepatic duct, and lower part, the common bile duct, by the entry of the cystic duct from the gallbladder. This point of confluence of hepatic and cystic ducts to form the common bile duct is widely variable, and any surgeon performing the operation of cholecystectomy has a duty of care to their patient to be fully aware of this anatomic variability (lest they mistake the common bile duct, or less frequently the common or right hepatic ducts for the cystic duct, resulting in catastrophic consequences for the patient). The main bile duct normally has a diameter of up to 6 mm and passes downward anterior to the portal vein in the right free border of the lesser omentum. The bile duct is closely related to the hepatic artery as it runs upwards on its left side before dividing into its left and right branches, the right hepatic artery usually passing posteriorly to the bile duct. The cystic artery, which usually arises from the right hepatic artery, crosses the common hepatic duct as frequently anteriorly as it does posteriorly (Figs. 1.17 and 1.18). Calot’s triangle was originally defined by the common hepatic duct lying medially, inferiorly by the cystic duct and superiorly by the cystic artery (32). However, the usually accepted surgical definition of this triangle has been modified to that of the “cholecystectomy” triangle, which defines the upper border as the inferior surface of the liver (and therefore contains the cystic artery) (33). The junction of the cystic duct and common hepatic duct varies widely and may even occur behind the pancreas. The retropancreatic portion of the bile duct approaches the duodenum obliquely, accompanied by the terminal part of the duct of Wirsung (see chap. 2). These two ducts join to enter the duodenum through the sphincter of Oddi at the papilla of Vater (34,35).

gallbladder and cystic duct
The gallbladder lies within the cystic fossa on the underside of the liver in the main liver scissura, thereby defining the junction between the right and left hemilivers. It is separated from the hepatic parenchyma by the cystic plate, which is an extension of connective tissue from the hilar plate (described previously). The anatomical relationship of the gallbladder to the liver ranges from hanging by a loose peritoneal reflection to being deeply embedded within the liver parenchyma. The gallbladder varies in size and consists of a neck, body, and fundus, which usually reaches the free edge of the liver, still closely applied to the cystic plate. Large gallstones impacting within the neck of the gallbladder may create a Hartmann’s pouch (33), and inflammation secondary to this can obscure the anatomical plane between the gallbladder and the common hepatic duct (thus obliterating the cholecystectomy triangle). This degree of inflammation can make dissection during cholecystectomy difficult, increasing the risk of damage to the common hepatic duct (36). Other structures similarly threatened during this dissection as part of cholecystectomy for

Segment 4 Glisson's capsule Lig.teres RHD RHA RPV Cystic artery Cystic duct Gallbladder CHD Umbilical fissure Line of incision of hilar plate to expose left hepatic duct CBD HA Retroduodenal artery Gastroduodenal artery Splenic vein LPV LHD LHA

Cystic plate

Hilar plate Superior mesenteric artery and vein Figure 1.17 Anterior aspect of biliary anatomy. Note the hepatic duct confluence anterior to the right hepatic artery and origin of the right portal vein. Note also the course of the cystic artery, arising from the right hepatic artery and passing posteriorly to the common hepatic duct.

Figure 1.16 Demonstration of the relationship between the posterior aspect of the base of segment 4 and the biliary confluence. Note the extension of Glisson’s capsule to invest the portal structures at the hilum (hilar plate) and extending over the hepatic surface of the gallbladder (cystic plate). Exposure of the extrahepatic left hepatic duct is achieved by incising the hilar plate at the base of segment 4 medially as far as the umbilical fissure.

9

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
chronic cholecystitis include the right hepatic artery (in up to 50% of cholecystectomy bile duct injuries, so rendering the upper bile duct ischemic with ramifications for the timing of bile duct repair), the right hepatic duct, and in exceptional circumstances, a low-lying middle hepatic vein lying superficially just below the gallbladder fossa. The cystic duct arises from the neck of the gallbladder and in 80% of people descends to join the common hepatic duct in its supraduodenal course. Its length varies widely but its luminal diameter is usually between 1 and 3 mm. The mucosa of the cystic duct is arranged in spiral folds (valves of Heister) (33). In a small number of cases, the cystic duct joins the right hepatic duct or occasionally a right hepatic sectoral duct. The gallbladder receives its blood supply by the cystic artery, the anatomy of which varies widely (Fig. 1.18). The most common variant arises directly from the right hepatic artery, then dividing into an anterior and posterior branch. The venous drainage of the gallbladder is directly through the gallbladder fossa to the portal vein in segment 5 (Fig. 1.19).

(A)

(B)

(C)

biliary anomalies
The biliary anatomy described above, comprising a right and left hepatic duct joining to form a common hepatic duct occurs in between 57% (23) and 72% (8) of cases. This variance may be explained by Couinaud’s (23) description of a triple confluence of right posterior sectoral duct, right anterior sectoral duct, and left hepatic duct in 12% of cases, which Healey and Schroy do not describe. There are many other abnormalities in biliary anatomy. Couinaud described a right sectoral duct joining the main bile duct in 20% of individuals (right anterior sectoral in 16%, right posterior sectoral in 4%). In addition, a right sectoral duct (posterior in 5%, anterior in 1%) may join the left hepatic duct in 6% of cases. In 3% of cases, there is an absence of a defined hepatic duct confluence with all the sectoral ducts joining separately and in 2% the right posterior sectoral duct may join the neck of the gallbladder or be entered by the cystic duct (23) (Fig. 1.20). Similarly, there are common variations of the intrahepatic biliary anatomy. Healey and Schroy (20) describe the classical intrahepatic biliary arrangement outlined above in 67% of

(D)

(E)

(F)

(G)

(H)

Figure 1.18 The eight most common variations in the anatomy of the arterial supply (cystic artery) to the gallbladder.

(A)

(B) Figure 1.19 (A) Venous drainage of the gallbladder. (B) The lymphatic drainage of the gallbladder towards the coeliac axis.

10

SURGICAL ANATOMY OF THE LIVER AND BILE DUCTS
cases, with ectopic drainage of segment 5 in 9%, segment 6 in 14%, and segment 8 in 20% of the cases. In addition, they describe a subvesical duct in 20% to 50% of the cases (8,37). This subvesical duct may lie deeply embedded in the cystic plate and can join either the common or right hepatic ducts. This duct does not drain any specific area of the liver and never communicates with the gallbladder, but may be damaged during cholecystectomy and therefore contribute to postoperative biliary leak. On the left side, the commonest anomaly is a common union of ducts of segments 3 and 4 (25% of cases), and in only 2% does the segment 4 duct independently join the common hepatic duct (Fig. 1.21). Gross described a number of anomalies of the accessory biliary apparatus in 1936 (38). These include bilobed and duplicated gallbladder (39,40), septum and diverticulum of the gallbladder, and variations in cystic duct anatomy including a double cystic duct (41). More rare is agenesis of the gallbladder (42,43) (Fig. 1.22). Furthermore, the gallbladder may be abnormally positioned, either lying deep within the liver parenchyma or lying under the left liver (44). The union of the cystic duct with the common hepatic duct may be angular, parallel, or spiral. The most frequent union is angular (75%) (45), while the cystic duct may run parallel with the hepatic duct in 20%, both encased in connective tissue. In 5% of cases, the cystic duct may approach the hepatic duct in a spiral fashion, usually passing posteriorly to the common hepatic duct before entering on its left side (Fig. 1.23).

the arterial blood supply of the bile ducts
The hepatic artery usually arises as one of the three named branches of the coeliac trunk, along with the left gastric and splenic arteries (Fig. 1.24). The first named branch of the hepatic artery is the gastroduodenal artery and either of these arteries may then give rise to the right gastric and retroduodenal arteries (Fig. 1.24). The hepatic artery then divides into right (giving rise to the cystic artery) and left hepatic arteries. This arrangement holds true for 50% of cases. In nearly 25% of cases, the right hepatic artery arises separately from the superior mesenteric artery, indicative of the joint fore- and mid-gut origin of the liver (Fig. 1.25). In the remaining 25% of cases, the left hepatic artery arises from the left gastric artery. Occasionally, other variations will occur. These variations will be readily apparent to an experienced surgeon at operation. The authors do not advocate preoperative angiography to delineate these anomalies prior to routine hepatectomy. The extrahepatic biliary system receives a rich arterial blood supply (46), which is divided into three sections. The hilar section receives arterioles directly from their related hepatic arteries and these form a rich plexus with arterioles from the supraduodenal section. The blood supply of the supraduodenal section is predominantly axial. Most vessels to this section arise from the retroduodenal, right hepatic, cystic, gastroduodenal, and retroportal artery. Usually, eight small arteries, each 0.3 mm in diameter, supply the supraduodenal section. The most important of these vessels run along the lateral borders of the duct and are referred to as the 3 o’clock and 9 o’clock arteries. Of the arteries supplying the supraduodenal section, 60% run upward from the major inferior vessels while 38% run downward from the right hepatic artery. Only 2% are nonaxial, arising directly from the main trunk of the hepatic artery as it runs parallel to the bile duct. The retropancreatic section of the bile duct receives its blood supply from the retroduodenal artery. The veins draining the bile duct mirror the arteries and also drain the gallbladder. This venous drainage does not enter the portal vein directly but seems to have its own portal venous pathway to the liver parenchyma (47). It has been proposed that arterial damage during cholecystectomy may result in ischemia leading to postoperative stricture of the bile duct (47), although it seems unlikely that ischemia is the major mechanism in the causation of bile duct stricture after cholecystectomy.

ra rp Ih rp

ra Ih

57% (A) ra rp Ih

12% (B) ra Ih

rp 20% C1 (C) ra rp rp Ih Ih ra 16% C2 4%

6% D1 (D) ra rp

5% D2

1% 4 4 3 ra rp 1 1% E2 ra rp Ih 2

3

2 1 3% E1 (E) 2%

2% (F) Figure 1.20 Main variations of the hepatic duct confluence.

11

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
7 6 91% 7 8 7 8 7 8 6 5% (A) seg V 7 6 6 5 80% (C) seg VIII 3 c 1% 2 3 d 25% e 1% 2 2 5 20% a 67% 6 8 4% 5 10% (B) seg VI 2 3 b 1% 5 2% 5 2% 8 7 8 5 8 7

86% 7

7 3

2

2

3

3 f 1% 2

3 g 4% (D) seg IV Figure 1.21 Variations of the intrahepatic biliary anatomy.

the anatomy of biliary exposure
Although intraoperative ultrasound has made easier the location of dilated intrahepatic biliary radicals, surgical exposure of the extrahepatic biliary confluence and the segment 3 duct demands knowledge of precise anatomical landmarks. Biliary– enteric anastomosis necessitates precise bile duct exposure to facilitate the construction of a mucosa to mucosa apposition (36,48–50). To expose the extrahepatic biliary confluence, the base of the quadrate lobe (segment 4) is lifted upward and Glisson’s capsule is incised at its base (see Fig. 1.16) (51). This technique is also sometimes referred to as “lowering the hilar plate.” In only 1% of cases is this made difficult by any vascular imposition between the hilar plate and the inferior aspect of the liver. This maneuver will expose considerably more of the left hepatic duct than the right, which runs a shorter extrahepatic course.

Contraindications to this approach include patients with a very deep hilum, which is displaced upward and rotated laterally (36), and those patients who have undergone removal or atrophy of either the right or left livers resulting in hilar rotation. In this situation, the bile duct may come to lie behind the portal vein. When approaching the segment 3 duct (segment 3 hepaticojejunostomy), follow the round ligament (in which runs the remnant of the obliterated umbilical veins) through the umbilical fissure to the point where it connects with the left branch of the portal vein within the recessus of Rex. This junction may sometimes be deeply embedded within the parenchyma of the fissure. The bile ducts of the left liver are located above the left branch of the portal vein, whereas the corresponding arteries lie below the portal vein. Dissection of the round ligament on its left side allows exposure of either

12

SURGICAL ANATOMY OF THE LIVER AND BILE DUCTS
liver split to the left of the umbilical fissure in order to widen the fissure to achieve adequate access to the biliary system. Access to the right liver system is less readily achieved than to the left as the anatomy is more imprecise. However, intraoperative ultrasonography greatly enhances the ability of the surgeon to locat e these ducts at surgery. The ideal approach on the right side is to the segment 5 duct (52), which runs on the left side of its corresponding portal vein (23). The duct is exposed by splitting the liver over a short distance to the right of the gallbladder fossa, commencing at the right side of the porta hepatis. The segment 5 duct should lie relatively superficially on the left aspect of the portal vein to that segment.
1

(A)

1

2

3

(B)

1

2

radiological anatomy of the liver
Accurate preoperative localization of liver pathology using radiological techniques is of increasing importance, as any potential resection depends largely on the segmental localization. Imaging is generally performed using ultrasound, computed tomography (CT), and magnetic resonance (MR). Ultrasound is excellent for imaging bile ducts, cysts, abscesses, and tumors. Hepatic circulation can also be accurately assessed using a Doppler technique. Ultrasound is also the imaging modality of choice for the biliary tree. However, the accuracy of ultrasound imaging is very operator dependent, and fine detail can be limited. Examination is limited by body habitus, and can be restricted by overlying bowel gas. CT scanning is an excellent method of assessing the liver parenchyma. It is able to identify a variety of different pathologies, and CT with IV contrast is the most commonly used method of imaging liver metastases. MR is excellent for the imaging and characterizing primary liver tumors, and is useful for the identification of hemangiomas, which can resemble metastases on CT scanning. Methods for defining segmental anatomy on ultrasound, CT, and MR images follow the anatomical landmarks previously described (53). These methods generally involve using three vertical planes along the lines of the main hepatic veins to divide the liver into its four sectors, with a transverse scissura along the portal vein further subdividing these four sectors to give the eight Couinaud segments. These anatomical landmarks are generally easily identifiable on standard imaging. The middle hepatic vein, left hepatic vein, and ligamentum teres provide good landmarks for dividing the left liver into its four segments. The right hepatic vein can usually be clearly seen dividing the right liver into its two sectors.

(C)

(D)

1

2

Figure 1.22 Main variations in gallbladder and cystic duct anatomy: (A) bilobed gallbladder; (B) septum of gallbladder; (C) diverticulum of gallbladder; (D) variations in cystic duct anatomy.

(A) 75%

(B) 20%

(C) 5%

Figure 1.23 Different types of union of the cystic duct and common hepatic duct: (A) angular (75%); (B) parallel (20%); (C) spiral (5%).

the pedicle or anterior branch of the duct from segment 3. This dissection is achieved by mobilizing the round ligament and pulling it downwards, thereby freeing it from the depths of the umbilical fissure. This procedure usually requires the preliminary division of the bridge of liver tissue that runs between the inferior parts of segments 3 and 4. The umbilical fissure is then opened and with downward traction of the ligamentum teres an anterior branch of the segment 3 duct is exposed on its left side. Sometimes it may be necessary to perform a superficial liver split to gain access to this duct. In the usual situation of chronic biliary obstruction with dilatation of the intrahepatic bile ducts, the segment 3 duct is generally easily located above the left branch of the portal vein. However, in the situation of left liver hypertrophy, it may be necessary to perform a more extensive

hepatic veins
In an oblique ultrasonic view, the three hepatic veins join the IVC to form a characteristic W, with its base on the IVC. A similar view can be seen on CT scan. These veins are usually easily seen: the left hepatic vein separating segment 2 from segments 3 and 4, the middle hepatic vein separating segment 4 from 5 and 8, and the right hepatic vein separating 5 and 8 from 6 and 7.

portal system
The portal supply to the left lobe, when viewed obliquely, can be seen as a side-on “H,” with the left portal vein giving its

13

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Left branch of the hepatic artery Right branch of the hepatic artery Hepatic artery

3 o'clock artery 9 o'clock artery Common hepatic artery Retroduodenal artery

Gastroduodenal artery

(A)

M.H. artery L.H. artery R.H. artery Cystic Proper hepatic Right gastric Supraduodenal

Left gastric Aorta Celiac trunk

Splenic

Gastroduodenal (B)

Common hepatic

Figure 1.24 (A) The biliary duct blood supply; (B) conventional arterial anatomy of the liver (50%).

branch to segment 2, before dividing into the terminal branches to 3 and 4. The portal supply to the right lobe also demonstrates a sideon “H” in the oblique view. The right branch of the portal vein forms the cross bar of the H, with the branches to segment 5 to 8 forming the arms.

identified using the landmarks outlined above. The scans were then reviewed, with the lesion being attributed to the nearest portal branch. Sixteen percent of lesions had a different segmental location if the portal branch was used instead of the conventional technique (Fig. 1.29) (54).

gallbladder, ligamentum venosum, and falciform ligament
Radiological landmarks of these structures are fallible (Figs. 1.26–1.28). Significant variations in intrahepatic vascular anatomy may result in incorrect identification of lesion location. A study by Rieker et al. looked at CT scans of patients who underwent liver resection. The location of the lesion was

key points




A full understanding of the lobar, sectoral, and segmental anatomy of the liver and biliary system is an essential prerequisite for successful liver surgery. The surgeon must appreciate the wide variation in extrahepatic biliary anatomy.

14

SURGICAL ANATOMY OF THE LIVER AND BILE DUCTS

4a

2

8 (A) (B)

IVC

7

(C)

(D)

Figure 1.27 CT scan of upper liver in venous phase showing the left, middle and right hepatic veins draining into the inferior vena cava (IVC).

(E)

(F)

Figure 1.25 Variations in anatomy of hepatic arterial supply.

Figure 1.28 CT scan of the liver in portal phase showing the left portal vein passing anteriorly between segments 3 and 4 within the recessus of Rex.

RAPV LPV

RPPV

MPV

Figure 1.26 Portal phase CT scan through porta hepatis showing the left portal vein (L) lying centrally and the anterior (RA) and posterior (RP) divisions of the right portal vein (R).

Figure 1.29 Percutaneous direct portogram showing the relationships of the anterior (RAPV) and posterior (RPPV) to the main (MPV) and left (LPV) portal veins.

15

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

references
1. Glisson F. Anatomia Hepatis. London: Typ. Du-Gardianis, 1654. 2. Rex 1888. Cited in Hobsley M. The anatomical basis of partial hepatectomy. Proc R Soc Med Engl 1964; 57: 550–4. 3. Schwartz SI. Historical Background. In: McDermott WV Jr, ed. Surgery of the liver. Boston, MA: Blackwell Scientific, 1989: 3–12. 4. McIndoe AH, Counsellor VX. A report on the bilaterality of the liver. Arch Surg 1927; 15: 589. 5. Lau WY. The history of liver surgery. J R Coll Surg Edin 1997; 42: 303–9. 6. Mikesky WE, Howard JM, DeBakey ME. Injuries of the liver in three hundred consecutive cases. Int Abstr Surg 1956; 103: 323–4. 7. Dalton HC. Gunshot wound of the stomach and liver treated by laparotomy and suture of the visceral wounds. Ann Surg 1888; 8: 81–100. 8. Luis A. Di un adenoma del fegato. Centralblatt fur chirg 1887; 5: 99. Abstract from Ganzy, delle cliniche 1886, 23, No 15. 9. Langenbuch C. Ein Fall von Resektion eines linksseitigen Schnurlappens der Leber. Berl Klin Wosch 1888; 25: 37–8. 10. Tiffany L. The removal of a solid tumor from the liver by laparotomy. Maryland Med J 1890; 23: 531. 11. Lucke F. Entfernung der linken Krebsiten Leber Lappens. Cantrallbl Chir 1891: 6: 115. 12. Cattell RB. Successful removal of liver metastasis from carcinoma of the rectum. Lehey Clin Bull 1940; 2: 7–11. 13. Wangensteen OH. The surgical resection of gastric cancer with special reference to: (1) the closed method of gastric resection; (2) coincidental hepatic resection; and (3) preoperative and postoperative management. Arch Surg 1943; 46: 879–906. 14. Keen WW. Report of a case of resection of the liver for the removal of a neoplasm with a table of seventy six cases of resection of the liver for hepatic tumor. Ann Surg 1899; 30: 267–83. 15. Cantlie J. On a new arrangement of the right and left lobes of the liver. J Anat Physiol (Lond) 1898; 32:4–9. 16. Wendel W. Beitrage zur Chirurgie der Leber. Arch Klin Chir Berlin 1911; 95: 887–94. 17. Ton That Tung. La vascularisation veineuse du foie et ses applications aux resections hepatiques. These, Hanoi, 1939. 18. Raven RW. Partial hepatectomy. Br J Surg 1948; 36: 397–401. 19. Lortat-Jacob JL, Robert HG. Hepatectomie droite regle. Presse Med 1952; 60: 549–50. 20. Healey JE Jr, Schroy PC. Anatomy of the biliary ducts within the human liver. Arch Surg 1953; 66: 599–616. 21. Goldsmith NA, Woodburne RT. Surgical anatomy pertaining to liver resection. Surg Gynaecol Obstet 1957; 195: 310–18. 22. Hjortsjo CH. The topography of the intrahepatic duct systems. Acta Anat 1951; 11: 599–615. 23. Couinaud C. Le foie. Etudes anatomiques et chirurgicales. Paris: Masson, 1957. 24. Couinaud C. Lobes et segments hepatiques. Note sur l’architecture anatomiques et chirurgicales du foie. Presse Med 1952; 62: 709–12. 25. Couinaud C. Anatomy of the dorsal sector of the liver. In: Couinaud C, ed. New Considerations on Liver Anatomy. Paris: Couinaud, 1998: 39–61. 26. Ton That Tung. Les Resections Majeures et Mineures Du Foie. Paris: Masson, 1979. 27. Caprio G. Un caso de extirpacion die lobulo izquierdo die hegado. Bull Soc Cir Urag Montevideo 1931; 2: 159. 28. Bismuth H, Houssin D, Castaing D. Major and minor segmentectomies “reglees” in liver surgery. World J Surg 1982; 6: 10–24. 29. Mancuso M, Nataline E, Del Grande G. Contributo alla conoscenza della struttura segmentaria del fegato in rapportto al problema della resezione epatica. Policlinico, Sez Chir 1955; 62: 259–93.

30. Couinaud C. Surgical anatomy of the liver revisited. C Couinaud, 15 rue Spontini, Paris, 1989. 31. Mizumoto R, Kawarada Y, Suzuki H. Surgical treatment of hilar carcinoma of the bile duct. Surg Gynecol Obstet 1986; 162: 153–8. 32. Rocko JM, Swan KG, Di Gioia JM. Calot’s triangle revisited. Surg Gynecol Obstet 1981; 153: 410–14. 33. Wood D. Eponyms in biliary tract surgery. Am J Surg 1979; 138: 746–54. 34. Byden EA. The anatomy of the choledochaoduodenal junction in man. Surg Gynecol Obstet 1957; 104: 641–52. 35. Delmont J. Le sphincter d’Oddi: anatomie traditionelle et fonctionelle. Gastroenterol Clin Biol 1979; 3: 157–65. 36. Bismuth H, Lazorthes F. Les Traumatismes Operatoires de la Voie Biliaire Principale. Paris: Masson, Vol 1, 1981. 37. Champetier J, Davin JL, Yver R, Vigneau B, Letoublon C. Aberrant biliary ducts (vasa aberrantia): surgical implications. Anat Clin 1982; 4: 137–45. 38. Gross RE. Congenital anomalies of the gallbladder. A review of a hundred and forty-eight cases with report of a double gallbladder. Arch Surg 1936; 32: 131–62. 39. Hobby JAE. Bilobed gallbladder. Br J Surg 1979; 57: 870–2. 40. Rachad-Mohassel MA, Baghieri F, Maghsoudi H, Nik Akhtar B. Duplication de la vesicule biliaire. Arch Francais des Maladies de l’Appareil Digestif 1973; 62: 679–83. 41. Perelman H. Cystic duct duplication. J Am Med Assoc 1961; 175: 710–11. 42. Boyden EA. The accessory gallbladder. An embryological and comparative study of aberrant biliary vesicles occurring in man and the domestic mammals. Am J Anat 1926; 38: 177–231. 43. Rogers HI, Crews RD, Kalser MH. Congenital absence of the gallbladder with choledocholithiasis. Literature review and discussion of mechanisms. Gastroenterology 1975; 48: 524–9. 44. Newcombe JF, Henley FA. Left sided gallbladder. A review of the literature and a report of a case associated with hepatic duct carcinoma. Arch Surg 1964; 88: 494–7. 45. Kune GA. The influence of structure and function in the surgery of the biliary tract. Ann R Coll Surg Engl 1970; 47: 78–91. 46. Northover JMA, Terblanche J. A new look at the arterial blood supply of the bile duct in man and its surgical implications. Br J Surg 1979; 66: 379–84. 47. Northover JMA, Terblanche J. Applied surgical anatomy of the biliary tree. In: Blumgart LH, ed. Biliary Tract, Vol 5. Edinburgh: Churchill Livingstone, 1982. 48. Bismuth H, Franco D, Corlette NB, Hepp J. Long term results of Roux-enY hepaticojejunostomy. Surg Gynecol Obstet 1978; 146: 161–7. 49. Voyles CR, Blumgart LH. A technique for construction of high biliary enteric anastomoses. Surg Gynecol Obstet 1982; 154: 885–7. 50. Blumgart LH, Kelley CJ. Hepaticojejunostomy in benign and malignant bile duct stricture: approaches to the left hepatic ducts. Br J Surg 1984; 71: 257–61. 51. Hepp J, Couinaud C, L’abord et L’utilisation du canal hepatique gauche dans le reparations de la voie biliaire principale. Presse Med 1956; 64: 947–8. 52. Smadja C, Blumgart LH. The biliary tract and the anatomy of biliary exposure. In: Blumgart LH, ed. Surgery of the Liver and Biliary Tract, 2nd edn. Edinburgh: Churchill Livingstone, 1994: 11–24. 53. Strunck H, Stuckmann G, Textor J et al. Limitations and pitfalls of Couinauds segmentation of the liver in transaxial imaging. Eur Radiol 2003; 13: 2472–82. 54. Rieker O, Mildenberger P, Hintze C et al. Segmentanatomie der Leber in der Computertomographie: Lokalisieren wir die Lasionen richtig. Rofo 2000; 171: 147–52.

16

2

Anatomy of the pancreas Margo Shoup and Jason W. Smith
tributary ducts coming off at near right angles and that this duct opened into the duodenum, and he saw that there were occasionally two ducts in the gland (1). It was Santorini who finally concluded that, in the normal condition, there existed two ducts with the smaller of the two emptying into the duodenum by way of a small papilla approximately 2 cm nearer to the stomach than the major duct and this smaller duct bears his name (5). The smaller duct is patent all the way to the duodenum in only 60% of specimens and the duct of Wirsung represents the larger of the two; however, in about 10% of specimens, the duct of Santorini is the main drainage for the pancreas. Also in about 10% of cases, the two ducts are not in communication with each other (1) (Fig. 2.2). The parenchyma of the pancreas consists of small lobules divided by connective tissue. These lobules are centered around the main tributary ducts that run to the main pancreatic duct. Smaller branches off of these tributaries define further septated regions within the lobules of pancreatic tissue. The main branches of the pancreatic duct tend to meet the main duct on its superior and inferior aspect. The diameter of the main pancreatic duct is reported to be between 2.6 and 4.8 mm in the head, 2.0 and 4.0 mm in the body, and 0.9 and 2.4 mm in the tail (3). The duct runs in a relatively superficial position in the tail and after traversing the neck of the pancreas it dives deep into the parenchyma as it crosses the head and is near the dorsal surface of the pancreas as it nears the confluence with the common bile duct (CBD) and the duodenum (1). The lower portion of the CBD lies in contact with the head of the pancreas for between 2 and 7 cm and 40% of the time it lies in a groove between the surface of the pancreas and the duodenum. In the remainder of cases, it lies within the parenchyma of the pancreas (7). During embryological development, the lower duct of Wirsung arises in the ventral pancreatic bud adjacent to the early hepatic duct. Therefore, the association of the duct of Wirsung with the CBD is a consistent feature of the ductal anatomy of the pancreas (1). The duct of Wirsung and the CBD unite 6 to 8 mm within the papilla and form a common channel, which is slightly dilated and referred to as the ampulla of Vater. In just over 10% of cases, the two ducts do not form a short common channel and instead enter the duodenum independently on the papilla (5).

topography of the pancreas
The shape and size of the pancreas are highly variable but in general it has a roughly trapezoidal shape and lies in the retroperitoneum of the upper abdomen (1). It is a finely lobular structure with a tan to dull yellow color that reaches from the medial concavity of the duodenum up and to the left terminating at the hilum of the spleen. The volume of the pancreas increases rapidly during childhood, plateaus from 20 to 60 years, and then steadily decreases; however, the percentage of parenchyma versus fat in the pancreas continues to increase during life slowly replacing functional tissue (2) (Fig. 2.1). The pancreas is divided into three major regions, the head and uncinate, the neck, and the body and tail (3). The head is the most medial portion of the gland. It is the widest and thickest part, having the most globular ultrastructure and is cradled in the concavity of the duodenum lying just to the right of the second lumbar vertebra (1). There is an inferior projection to the head of the pancreas that lies posterior to the superior mesenteric vessels, which makes up the uncinate process. The head and uncinate are intimately associated with the duodenum, sharing an abundant network of anastomosing vessels. The posterior surface of the head of the pancreas is in apposition to the inferior vena cava, aorta, right spermatic and ovarian vessels, and right renal vessels and separated from them by the avascular fusion fascia of Treitz (4). The anterior surface is covered by the transverse colon and its mesentery (5,6). The neck of the pancreas is 2 to 3 cm in length and overlies the confluence of the superior mesenteric vein (SMV) and splenic vein by which it is grooved. It is related superiorly to the pylorus and first portion of the duodenum (3,4). The body of the pancreas extends from body of the second lumbar vertebra over the left kidney and begins to taper into the tail as it reaches the hilum of the spleen. The prismatic shape of the pancreas flattens in the tail. The splenic vein runs the length of the pancreas on the posterior surface, while the artery courses along the superior edge of the body. The body of the pancreas lies over the aorta and the left renal pedicle and kidney and is separated from these structures by the fusion fascia of Toldt (4). Inferiorly, it abuts the mesentery of the transverse mesocolon, while superiorly and anteriorly it abuts the stomach but is separated from it by the posterior parietal peritoneum (7).

arterial anatomy of the pancreas
The pancreas enjoys an abundant arterial blood supply that draws from both the celiac axis and the superior mesenteric artery (SMA). The pancreas is supplied from the celiac axis by the superior pancreaticoduodenal artery from the gastroduodenal artery (GDA), and the dorsal pancreatic and pancreatica magna arteries from the splenic artery (Fig. 2.3). The distal and inferior borders of the pancreas are supplied by the caudal and inferior pancreatic arteries, which are formed by

ductal anatomy of the pancreas
There are numerous configurations of the ducts of the pancreas and their relationships to each other, the duodenum and the common bile duct. The significance of the pancreas became understood only after the discovery of the main pancreatic duct by Wirsung in 1643. He noted that there was a duct that traversed the length of the organ with numerous

17

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 2.1 Overview of the relationship of the pancreas to other important structures in the upper abdomen. Plate 1098, From Anatomy of the Human Body, Henry Gray 1918.

(A)

(B)

(C)

Figure 2.2 (A) Duct of Santorini is patent all the way to the duodenum. (B) Duct of Santorini is the main drainage. (C) The two ducts are not in communication with each other.

ramifications with the dorsal pancreatic, pancreatica magna, and splenic arteries. The SMA gives rise to the more variable inferior pancreaticoduodenal (IPD) artery, which divides into branches to form both an anterior and posterior anastomotic arcade with branches from the superior pancreaticoduodenal artery (8). Superior Pancreaticoduodenal Artery The superior pancreaticoduodenal is a short branch of the GDA that arises after the takeoff of the right gastroepiploic artery (Fig. 2.4). It is angiographically identifiable in about 10% of specimens and is generally about 8 mm in length (9). Although rare, it is reported to occasionally arise from the left hepatic artery. When present, the superior pancreaticoduodenal artery divides into anterior and posterior branches, which anastomose with the inferior branches from the SMA. In the remaining cases, the posterior superior pancreaticoduodenal

(PSPD) artery is seen arising from the GDA prior to the right gastroepiploic takeoff. The anterior superior pancreaticoduodenal (ASPD) artery has a caliber between 1 and 3 mm and is considered the most important blood supply to the head of the pancreas. In the majority of cases, it is a terminal branch of the GDA after it has given off the PSPD and the right gastroepiploic arteries. The ASPD can be duplicated in up to 7% of cases and rarely is absent. Case reports of extremely rare anomalies exist, reporting the origin of this artery from almost all of the major branches of the celiac and SMAs (9). Posterior Superior Pancreaticoduodenal Artery This artery forms the superior portion of the posterior arcade that forms anastomoses with the posterior branch of the IPD artery. The PSPD artery is most commonly found as a branch of the GDA 1 to 2 cm after the takeoff of the hepatic artery (10). Up to 10% of cases may see the PSPD arise from the superior

18

ANATOMY OF THE PANCREAS

Cystic artery

Figure 2.3 Arterial anatomy of the pancreas, the celiac axis and its major branches. Plate 532, From Anatomy of the Human Body, Henry Gray 1918.

pancreaticoduodenal and in rare instances may arise from any of the hepatic arteries. The most common course of the PSPD after it leaves the GDA posteriorly is it runs over the portal vein (PV) and the anterior edge of the top of the pancreas where it enters the gland and finds the common bile duct and makes a right-handed spiral around the duct passing posterior to it just above the ampulla. It then runs deep in the parenchyma of the pancreas to find its connection with the posterior inferior artery. The PSPD gives off collateral branches to form the blood supply to the intrapancreatic portion of the common bile duct, it generally gives off the supraduodenal artery and occasionally the retroduodenal artery, rarely it may give a branch to the gallbladder or an accessory right hepatic artery (10). Inferior Pancreaticoduodenal Artery The IPD artery is present in about 70% of cases and is the common trunk that gives rise to the anterior and posterior inferior pancreaticoduodenal (AIPD and PIPD) arteries that form the anastomotic arcades supplying the head of the pancreas (11). In the remaining 30% of cases, the AIPD and PIPD arise directly from the SMA. The IPD may arise directly from the SMA as the first collateral branch from 2 to 5 cm distal to the origin and take a short course from its posterior takeoff into the inferior edge of the pancreatic parenchyma, or alternatively, it may arise as a common trunk with the first jejunal

Probe passed through epiploie foramen

a S t o m

c

h

C r e a t o r

O x e n t e

branch, the pancreaticoduodenaljejunal (PDJ) trunk in which case it takes a longer course to the pancreas. The IPD crosses posterior to the SMV and the posterior surface of the pancreas and does not give off any branches prior to dividing into its anterior and posterior termini (11). Anterior and Posterior Inferior Pancreaticoduodenal Arteries These arteries supply the inferior part of the anastomotic arches that supply the head of the pancreas. They arise most often from a common IPD artery. They may also originate directly from the SMA or less commonly directly from the first jejunal artery or from a replaced hepatic artery. The main course of the AIPD is to follow the inferior curve of the pancreas and find its partner the ASPD (12). It may give off a branch to the duodenal–jejunal flexure or to form a transverse pancreatic artery. The PIPD runs more posterior and cephalad than the AIPD and ultimately finds the PSPD or alternatively terminates as small end arteries. It may supply a collateral branch to the transverse pancreatic artery when present (12). Dorsal Pancreatic Artery The main blood supply to the neck and body of the pancreas is the dorsal pancreatic (DP) artery. It most commonly arises from the splenic artery near its origin at the celiac axis (13). It may also take its origin from the celiac trunk itself, the

19

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

S t o m a c h

Figure 2.4 Arterial anatomy of the pancreas, demonstrating the gastroduodenal and its branches of the anterior and posterior pancreaticoduodenal arteries forming the anastomotic arcades with the branches from the superior mesenteric artery. Plate 533, From Anatomy of the Human Body, Henry Gray 1918.

common hepatic or the GDA. Alternatively, the DP may arise from the SMA. The course of the DP artery is usually in the form of an inverted “T” with a right and left branch that form after a short 1 to 3 cm course. When the artery arises from the splenic artery, it tends to angle back to the right, if it takes off from the celiac, hepatic, or GDA, then it transverses the neck in a leftward direction. When coming from the SMA it comes up from the bottom of the pancreas. The right branch of the DP forms an anastomosis with left anastomotic pancreatic artery from the ASPD. The left branch becomes the transverse pancreatic artery (13). Caudal and Great Pancreatic Arteries The great pancreatic artery is often present and is given off from the splenic artery at the junction of the body and tail. It collateralizes with the transverse pancreatic artery. The caudal pancreatic artery takes its origin from the left gastroepiploic, the distal splenic artery or a branch from the splenic hilum and forms anastomotic connections with the great pancreatic and transverse pancreatic arteries (3). The arterial blood supply to the pancreas is rich and complex. Most of the primary arterial conduits form some

anastomotic connection and this shared blood supply is one of the challenges of pancreatic surgery. When operating in the deepest recesses of the abdomen, having an intimate knowledge of the standard arterial anatomy as well as the most common alternatives will allow the pancreatic surgeon to maximize patient safety. That same surgeon must keep in mind that the arterial anatomy in this area is subject to wide variation and that one must always be prepared to address the aberrant anatomy. To that end, having good preoperative imaging to establish before the operation what the arterial anatomy is can be a valuable aid whether by angiography or by computed tomography (CT) angiography. Venous Drainage of the Pancreas The veins of the pancreas follow the course of the corresponding arteries in most cases. They are generally more superficially located than the arteries and depending on the location in the pancreas drain into the PV, SMV, the inferior mesenteric vein, or the splenic vein. In the head of the pancreas, there is a venous arcade that mirrors the arterial anastomoses and of the four main veins all, but the PSPD vein, which empties directly into the PV, find their way to the SMV. In addition, there are

20

ANATOMY OF THE PANCREAS

12 12b1 1 12a2 3 5

12p1 12a1 12p2 9 7 11 10

8a 16 8p 10

2 6 4 13a 17a 14b

14a

14c 17b 14V

14d

18

13b

15

14d

Figure 2.5 Lymphatic drainage of the pancreas.

numerous small bridging veins between the head of the pancreas and the SMV and PV as they course behind the pancreas, which must be carefully ligated during a resection. The fact that there are rarely venous branches that enter the SMV or PV on their anterior surfaces makes the dissection along the plane anterior to these vessels possible during pancreaticduodenectomy. Two large veins drain the body and tail of the pancreas, the splenic vein, which courses along the superior edge of the pancreas and the transverse pancreatic vein along the inferior margin. The portal vein is formed on the posterior surface of the neck of the pancreas by the confluence of the splenic vein and the SMV. The inferior mesenteric vein may join at this point as well, but more commonly joins the splenic vein or SMV proximal to the confluence (Fig. 2.4).

lympatic drainage of the pancreas
The lymphatic drainage of the pancreas is rich and drains each lobular division with frequent anastomotic connections and the ultrastructure is similar to that in other solid organs of the abdomen(14) (Fig. 2.5). These lobular lymphatics coalesce to form several trunks that empty into the primary lymph node basins for the pancreas before quickly reaching the thoracic duct (15). The drainage of the pancreas can be roughly divided into right and left side based on the ventral and dorsal anlage of the primordial pancreas. The left side of the system drains the upper portion of the head, the neck, and body and tail,

while the right side drains the lower portion of the head, which developed from the ventral bud and constitutes the retroportal lymphatics (15,16). The superior pancreatic nodes drain the upper half of the neck, body and tail of the pancreas, and a portion of the head. They primarily lie along the superior border of the gland or in the gastropancreatic fold and gastrohepatic ligament (17). The inferior pancreatic nodes similarly drain the inferior half of the gland and lie along the inferior border as well as draining into the superior mesenteric nodes or the periaortic nodes. The anterior nodes are located along the surface of the pancreas that lies adjacent to the duodenum and are called the infrapyloric lymph nodes and the pancreaticoduodenal nodes. These anterior nodes may also drain into nodes along the root of the transverse colonic mesentery that is adjacent to the head of the pancreas. The posterior nodes run along the posterior pancreaticoduodenal border and include the nodes along the lower portion of the common bile duct, portal vein and nodes at the origin of the SMA. The tail of the pancreas forms several lymphatic trunks that reach out into the hilum of the spleen and form the superior and inferior lymph nodes (3,16). This simplified lymphatic mapping system is that adapted by the International Union against Cancer (UICC). A more comprehensive and clinically useful system was developed by the Japanese Research System, which divides lymph node stations into 18 different designations and rates them according to the likelihood of metastatic spread. Nodal stations 13 and 17 are the most likely to harbor disease with

21

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Right gastroepiploic vein Portal vein Splenic vein

PSPD-V Gastrocolic trunk

Superior mesenteric vein AIPD-V PIPD-V ASPD-V Figure 2.6 Major venous drainage for the pancreas. First jejunal tributary

The pancreas lies in the recesses of the upper abdomen and remains one of the most challenging organs to manage from a clinical or operative standpoint. Its rich blood supply, close associations with major vascular structures, intimate relation to the common bile duct, and the attachments to the duodenum and spleen all contribute to the complexity of surgical intervention in both malignant and benign disease (7). A thorough understanding of the three dimensional relationship of the arterial blood supply and major veins in proximity to the pancreas make approaching pancreatic resection possible. As we move into an era of minimally invasive surgery, being able to recognize the anatomy and its variations with minimal cues from adjacent structures will become increasingly important and continued study of these complex relationships allows the mind to know, so that the eye may see.

references
1. Opie EL. Anatomy of the Pancreas and its Variations. Disease of the Pancreas: Its Cause and Nature, 1st edn. Philadelphia, PA: J.B. Lippincott Company, 1903: 359. 2. Saisho Y, Butler AE, Meier JJ, et al. Pancreas volumes in humans from birth to age one hundred taking into account sex, obesity, and presence of type-2 diabetes. Clin Anat 2007; 20: 933–42. 3. Skandalakis LJ, Colborn GL, Skandalakis JE. Surgical anatomy of the pancreas. In: Baker RJ, Fischer JE, eds. Mastery of Surgery, Vol. 2, 4th edn. Philadelphia, PA: Lippincott Williams & Wilkins, 2001: 2448. 4. Kuroda A, Nagai H. Surgical anatomy of the pancreas. In: Howard J, Idezuki Y, Ihse I, Prinz R, eds. Surgical Diseases of the Pancreas, 3rd edn. Baltimore, MD: Lippincott Williams & Wilkins, 1998: 869. 5. Cattell RB, Warren KW. The anatomy and physiology of the pancreas. In: Cattell RB, Warren KW, eds. Surgery of the Pancreas. Philadelphia, PA: Saunders, 1953. 6. Hollinshead WH. The thorax, abdomen and pelvis. In: Hollinshead WH, ed. Anatomy for Surgeons. Vol. 2. New York: Medical Department, Harper and Row Publishers, 1971: 430. 7. Anson BJ, McVay CB, Callander CL. The Abdomen. Surgical Anatomy. Philadelphia, PA: Saunders, 1971. 8. Woodburne RT, Olsen LL. The arteries of the pancreas. Anat Rec 1951; 111: 255–70. 9. Bertelli E, Di Gregorio F, Bertelli L, Mosca S. The arterial blood supply of the pancreas: A review. I. The superior pancreaticoduodenal and the anterior superior pancreaticoduodenal arteries. An anatomical and radiological study. Surg Radiol Anat 1995; 17: 97–106, 101–3. 10. Bertelli E, Di Gregorio F, Bertelli L, Civeli L, Mosca S. The arterial blood supply of the pancreas: A review. II. The posterior superior pancreaticoduodenal artery. An anatomical and radiological study. Surg Radiol Anat 1996; 18: 1–9. 11. Bertelli E, Di Gregorio F, Bertelli L, Civeli L, Mosca S. The arterial blood supply of the pancreas: A review. III. The inferior pancreaticoduodenal artery. An anatomical review and a radiological study. Surg Radiol Anat 1996; 18: 67–74. 12. Bertelli E, Di Gregorio F, Bertelli L, Orazioli D, Bastianini A. The arterial blood supply of the pancreas: A review. IV. The anterior inferior and posterior pancreaticoduodenal aa., and minor sources of blood supply for the head of the pancreas. An anatomical review and radiologic study. Surg Radiol Anat 1997; 19: 203–12. 13. Bertelli E, Di Gregorio F, Mosca S, Bastianini A. The arterial blood supply of the pancreas: A review. V. The dorsal pancreatic artery. An anatomic review and a radiologic study. Surg Radiol Anat 1998; 20: 445–52. 14. Navas V, O’Morchoe PJ, O’Morchoe CC. Lymphatic system of the rat pancreas. Lymphology 1995; 28: 4–20. 15. Pissas A. Anatomoclinical and anatomosurgical essay on the lymphatic circulation of the pancreas. Anat Clin 1984; 6: 255–80. 16. Donatini B, Hidden G. Routes of lymphatic drainage from the pancreas: A suggested segmentation. Surg Radiol Anat. 1992; 14: 35–42.

47% and 29%, respectively (18,19) (Fig. 2.6). Classification of lymphatic involvement will become increasingly important as increasing numbers of targeted therapies become available in pancreatic cancer.

innervation of the pancreas
The pancreas receives fibers from both the sympathetic and parasympathetic nervous systems. The sympathetic innervation is via the splanchnic nerves, which carry both afferent fibers and efferent fibers, while the parasympathetic innervation is via the vagus nerve, which also has afferent and efferent supply to the pancreas. Parasympathetic innervation provides stimulatory signals to the islet cells to increase insulin secretion in response to food intake, while increased sympathetic tone suppresses insulin secretion and stimulates the secretion of glucagon (20,21). Efferent pain fibers are found in both the splanchnic and vagal nerves and localization of these fibers has been a difficult clinical problem in the management of pain in both inflammatory and malignant diseases of the pancreas. The right, and more prominently the left, celiac ganglion provide the majority of the direct innervation to the posterior head, body, and tail of the pancreas via fibers that course along the splenic artery (16). Neural ganglia around the common hepatic artery also provide fibers that course along the GDA to the head and uncinate process of the pancreas (22). Recently, it has been shown that the celiac ganglion bearing the splanchnic efferent fibers can be identified by endoscopic ultrasound and precise localization of neurolytic therapies can be applied to improve the success of these approaches (23,24). Enteropancreatic nervous connections have also been demonstrated from both the stomach and proximal duodenum to the pancreas (24–26). These connections suggest that there is crosstalk directly from the gastrointestinal tract to the pancreas coordinating exocrine and/or endocrine secretions with gut function.

22

ANATOMY OF THE PANCREAS
17. Hartley M, Finch-Jones M. Anatomy of the pancreas. In: Poston G, Blumgart L, eds. Surgical Management of Hepatobiliary and Pancreatic Disorders, 1st edn. London: Martin Dunitz, 2002: 19–28. 18. Bogoevski D, Yekebas EF, Schurr P, et al. Mode of spread in the early phase of lymphatic metastasis in pancreatic ductal adenocarcinoma: Prognostic significance of nodal microinvolvement. Ann Surg 2004; 240: 993–1000, discussion 1000–1. 19. Sakai M, Nakao A, Kaneko T, et al. Para-aortic lymph node metastasis in carcinoma of the head of the pancreas. Surgery 2005; 137: 606–11. 20. Benthem L, Mundinger TO, Taborsky GJ, Jr. Parasympathetic inhibition of sympathetic neural activity to the pancreas. Am J Physiol Endocrinol Metab 2001; 280: E378–81. 21. Jarhult J, Falck B, Ingemansson S, Nobin A. The functional importance of sympathetic nerves to the liver and endocrine pancreas. Ann Surg 1979; 189: 96–100. 22. Yoshioka H, Wakabayashi T. Therapeutic neurotomy on head of pancreas for relief of pain due to chronic pancreatitis; a new technical procedure and its results. AMA Arch Surg 1958; 76: 546–54. 23. Levy MJ, Topazian MD, Wiersema MJ, et al. Initial evaluation of the efficacy and safety of endoscopic ultrasound-guided direct Ganglia neurolysis and block. Am J Gastroenterol 2008; 103: 98–103. 24. Kirchgessner AL, Liu MT, Gershon MD. In situ identification and visualization of neurons that mediate enteric and enteropancreatic reflexes. J Comp Neurol 1996; 371: 270–86. 25. Holst JJ, Schwartz TW, Knuhtsen S, Jensen SL, Nielsen OV. Autonomic nervous control of the endocrine secretion from the isolated, perfused pig pancreas. J Auton Nerv Syst 1986; 17: 71–84. 26. Kirchgessner AL, Gershon MD. Innervation and regulation of the pancreas by neurons in the gut. Z Gastroenterol Verh 1991; 26: 230–33.

23

3

Hepatic resection Ajay V. Maker and Michael D’Angelica
healthy individuals, excision of tumor can prolong life and in some cases provide long-term disease-free survival. Patient Selection Proper patient selection is critical to both the safety and efficacy of hepatic resection. One should evaluate the patient’s general state of health, the condition of the liver, and the volume of the future liver remnant to properly assess the risk of general anesthesia, major abdominal surgery, and liver resection. Subcostal and upper abdominal incisions are painful and may result in respiratory splinting and increased pulmonary complications compared to other incisions (16). For this reason, assessment of the patient’s ability to mobilize early and ambulate postoperatively must not be underestimated. Though there are many algorithms to evaluate liver function in patients with chronic liver disease, the Child-Pugh classification is a useful preoperative indicator, and patients with a Pugh score of B or C should generally not undergo liver resection. Hepatic resection in cirrhotic patients is particularly difficult with operative mortality increasing with advanced Child classification. Hepatic resection in the setting of portal hypertension is generally not recommended (17), as this condition predisposes the liver to higher portal pressures and diminished ability to increase portal flow to the liver remnant postoperatively, thereby inhibiting normal liver regeneration and increasing the risk of life-threatening bleeding. The cirrhotic liver has decreased regenerative capacity and impairment in liver function is greater, lasts longer, and can result in permanent liver failure. A low platelet count, splenomegaly, ascites, or evidence of varices on preoperative radiography may be the only findings to alert the surgeon to hepatic dysfunction. Many noncirrhotic patients that present for hepatic resection have abnormal liver function due to chemotherapy, diabetes, or obesity. These diseased livers carry an increased risk of functional impairment with large resections and may also have impaired function despite retained volume (18,19). In these livers, careful preoperative planning must be done to achieve a parenchymal sparing resection. Biopsy, if performed, can give clues to the fat content of the liver, as can preoperative imaging (20). Early data suggest that MRI spectroscopy can also accurately quantify hepatic fat content, and this may prove to be a useful tool in preoperative liver assessment and operative planning (20,21). In cases where liver function may be impaired, or where extended resection is necessary to gain tumor-free margins, portal vein embolization is being employed to induce hypertrophy of the proposed liver remnant (22,23). No absolute guidelines for embolization can be made; however, preoperatively induced liver hypertrophy is a valuable tool in planning and executing major liver resections (24). Furthermore, chronic biliary obstruction inhibits liver function and, thus patients with

introduction
Though liver anatomy and physiology have been studied for centuries, liver surgery still is a relatively young field. Just 30 years ago, the mortality of major hepatic resection neared 25%. This high mortality limited its utility and deterred patients and referring physicians from considering surgery. The current generation of hepatobiliary surgeons has an increased understanding of the segmental anatomy of the organ and has seen a dramatic decrease in the mortality of liver surgery to nearly 1% largely due to a dramatic decrease in blood loss (1). This chapter will address the basic principles and techniques to safely approach liver resection.

basic principles
Surgical Indications: Benign vs. Malignant Disease Though this chapter focuses on the technical aspects of hepatic resection, an understanding of when liver resection is indicated is of paramount importance. Due to advances in modern imaging techniques and an increased knowledge of the natural history of liver lesions, tumors that may have been resected in the past for diagnostic uncertainty are now often observed. Similarly, malignant lesions that were not resected in the past but referred for nonsurgical therapy are now being treated with resection. Indications for specific benign and malignant processes are outlined in other chapters; however, the general principles are mentioned here. Benign Disease Partial hepatectomy for benign conditions should be parenchymal preserving and reserved for lesions that are symptomatic, have premalignant potential, or carry an unclear diagnosis. Wide margins are not necessary, therefore in some cases, for example, focal nodular hyperplasia (FNH) or hemangiomas, enucleation may be safely performed, although in some instances an anatomic segmental resection may be the safest approach (2–4). This is addressed at the end of the chapter and detailed in other chapters. Malignant Disease Partial hepatectomy for malignant conditions must obtain a clear surgical margin, and is suitable for well-selected patients with both primary and metastatic cancer. We have found increased patient survival with margins of at least 1 cm in patients undergoing resection for metastatic colorectal cancer (5–13), though other series suggest that a negative margin, regardless of the distance, is sufficient (14,15). The exception may be in slow-growing tumors with multiple liver metastases, such as neuroendocrine tumors, where tumor debulking may be of value. As long as the functional remnant liver is adequate, usually about 25% liver volume in otherwise

24

HEPATIC RESECTION
hilar cholangiocarcinomas are also at increased risk of liver failure postoperatively. The functional residual liver volume should be calculated to insure adequate liver function postresection. A healthy, noncirrhotic individual requires a functional hepatic reserve of at least 20% of the original nontumoral liver volume. The regenerative capacity of the liver should enable full functional compensation within weeks of resection; once greater than 70% of liver volume is resected, however, there is a risk of clinically significant liver insufficiency. This risk is minimal if specimen volume has been replaced with tumor, in which case compensatory hypertrophy will have already occurred. Preoperative Imaging (See Also Chapters 3, 4, and 11) Fine-cut triphasic helical computed tomography (CT) with CT angiogram is the single most useful study in preoperative evaluation of liver tumors. When the study includes the chest, abdomen, and pelvis, preoperative staging is reliable and can identify areas outside of the liver that may need further evaluation or confirm nonoperative candidates. CT can define the vascular anatomy, identify anatomical variants, determine resectability, estimate the functional liver residual volume, and identify preoperative biliary drainage strategies, thereby obviating the need for further radiographic studies. CT angiography in particular has almost prevented the need for traditional angiography. 3-D reconstruction of the vasculature is particularly helpful in identifying vascular anomalies quickly and temporally. Furthermore, 3-D reconstruction of the vascular anatomy may lead to more accurate visualization of tumor– vessel relationships and may be a more accurate study to predetermine the operative line of transection (25). Magnetic resonance imaging can also provide high-quality vascular and volumetric assessments of the liver but its principal role is in characterizing liver tumors of unclear etiology. In experienced hands, ultrasound is a fast, inexpensive, and noninvasive modality that can quickly obtain information regarding tumor size and the amount of liver involvement, particularly in gallbladder and biliary tumors. It is especially helpful in distinguishing cysts from solid tumors and should be used in addition to CT to evaluate cysts for the presence of septations or mural thickening, which would suggest cystadenoma or a cystadenocarcinoma. Duplex ultrasound is also particularly helpful as a dynamic study to identify vasculature in relation to tumor masses. Anesthetic Techniques Operative and perioperative morbidity and mortality have been decreased in part due to changes in anesthetic practices over the evolution of hepatic resection. A focus on maintaining low central venous pressure (CVP) can greatly reduce blood loss and keep the operative field clean for proper visualization of the biliary and vascular anatomy during parenchymal transaction. This is accomplished by positioning the patient in mild Trendelenberg and minimizing intravenous fluid to maintain systolic blood pressures above 90 mmHg and urine output to about 25 mL/h. If the IVC is still distended after mobilization of the liver, parenchymal transection can wait until central venous pressure is decreased through use of narcotics, vasodilatory inhalation agents, or direct vasodilators. A central venous pressure of less than 5 mmHg can be maintained during the periods of liver mobilization and parenchymal transection. Though a cental venous catheter is a useful tool to follow the CVP, the surgeon can also look for a nondistended IVC and for blood coursing through flat intrahepatic veins. If transection is performed under Pringle control, bleeding is generally from hepatic veins, therefore, with a low hepatic venous pressure, even large tears in hepatic veins can be visualized to allow ligation or repair without massive hemorrhage. By Poiseuille’s law, blood flow is exponentially proportional to the radius of the vessel; therefore, even minor decreases in venous distention can decrease blood loss exponentially. With these techniques, the risk of postoperative renal failure has not been shown to be significant, nor has the risk of air embolism, which can be minimized, regardless, by keeping the patient in about 15° of Trendelenberg (26,27). Normal resuscitation is performed after the resection is completed and hemostasis has been achieved.

basic techniques
Positioning, Skin Incision, and Exposure The patient should be positioned supine with the arms extended at right angles to the body. Any self-retaining retractor can be utilized, however, we prefer the Goligher retractor to elevate the costal margin, and this crossbar can be fitted to the table to form a 45° angle from top of the crossbar to the xyphoid. The patient should be prepped from the mid-chest to below the umbilicus, and draped to expose the right chest in the event a right thoracotomy is necessary to gain additional exposure. Though some groups routinely make a J-shaped thoracoabdominal incision, in our experience a thoracoabdominal incision was rarely necessary in over 1800 cases (2). We employ selective use of diagnostic laparoscopy based on the risk of unresectable disease (28), and conform the type of incision to the expected resection. For access to both lobes of the liver, a bilateral subcostal incision can be used with or without vertical midline extension. For the great majority of liver resections, we employ a “hockey stick” incision, which includes a right subcostal incision with vertical midline extension to the xyphoid. These incisions, when combined with the Goligher retractor, provide good exposure of the suprahepatic IVC, even with large right-sided tumors. We have found a higher rate of incisional hernia with a “Mercedes” incision compared to a “hockey stick” incision (29). For left-sided resections, a midline incision may suffice. Occasionally, when there is severe right-sided hepatic atrophy or exposure to the suprahepatic IVC is necessary for safety, extension into the right chest can be helpful (Fig. 3.1). Mobilization The ligamentum teres is divided between clamps and ligated, leaving a long secure ligature that is used as a handle to further expose the porta hepatis. The thin veil of the falciform ligament is incised along its length to free it from the anterior abdominal wall and expose the ligamentum teres. In obese individuals, the area where the falciform is fused to the anterior abdominal wall may be invested within a large fat pad. This fat pad can be removed with diathermy from beneath

25

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
both sides of the exposed fascia, improving exposure and aiding with fascial reapproximation at the end of the case. The falciform ligament is divided up to the suprahepatic IVC (Fig. 3.2). Bimanual palpation of the liver should be performed to assess the extent of hepatic disease. Segment 4 should be carefully retracted cephalad to expose the clear veil of lesser omentum anterior to the caudate lobe and attaching to the ligamentum venosum. This is incised, allowing palpation of the caudate and celiac axis, and providing access through the foramen of Winslow to the porta hepatis. Intraoperative ultrasound is used at this point to define the extent of disease, vascular relationships, and to confirm resectability. To mobilize the right liver, the leaf of the right coronary ligament is dissected from the falciform ligament and carefully incised over the IVC and territory of the right hepatic vein. This should be done sharply with downward traction on the liver and superior traction on the diaphragm. Once the right hepatic vein is identified, the right coronary ligament is taken close to the liver surface to its furthest extent laterally and the right triangular ligament is divided. To complete the mobilization, the right liver must be freed inferiorly. Omental and peritoneal attachments to the liver and gallbladder are divided to expose the inferior extent of the right triangular ligament. The retroperitoneal attachments are incised off the right adrenal gland and the liver can then be rotated medially to expose the retrohepatic IVC. If the right liver is to be resected or control of the right hepatic vein is needed, the multiple small venous branches from the IVC to the posterior liver must be individually dissected, controlled, and divided. Large accessory inferior right hepatic veins are common and may require division with a vascular stapler or control with vascular clamps and ligatures. It is critical for the surgeon on the left side of the table to retract the right liver medially to expose these branches and prevent injury to the cava. When all of these branches are ligated and divided, all that is left to expose the right hepatic vein will be a fibrous band of tissue that runs lateral to the vein, encircles the IVC, and courses posteriorly to the left and posterior border of the caudate, known as the caval ligament (Fig. 3.3). A tunnel can be safely created medial to this ligament and lateral to the right hepatic vein with a Kelly clamp or renal pedicle clamp in order to allow either a ligature or a

F

A D

C

B

E

Figure 3.1 Incisions for liver resection. B-D, initial upper midline exploration. A-B-C, ideal for exposure of the whole liver (hockey stick). C-D-E, the classic chevron incision with A-D (Mercedes) extension. C-D, right subcostal incision. F, thoracoabdominal extension. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

Figure 3.2 Mobilization of the liver begins with downward traction on the liver and division of the falciform ligament to the inferior vena cava. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

26

HEPATIC RESECTION
vascular load endo-GIA staple fire. Once this is divided, the right liver is mobile and the lateral aspect of the right hepatic vein is exposed. The left liver is mobilized similarly, however since it does not lie on the vena cava, an extensive caval dissection is not necessary. Sharp and blunt dissection over the suprahepatic IVC will expose the groove between the right vein and the common trunk of the middle and left and middle hepatic veins. Downward traction on the liver and cephalad traction on the diaphragm help expose the left coronary ligament. The groove between the left and middle hepatic veins can be exposed with sharp dissection if there is no long intrahepatic common channel (Fig. 3.4). Care must be taken here to identify the phrenic vein as it courses on the underside of the diaphragm to enter the IVC, as it can be inadvertently injured if the triangular ligament is not properly exposed or not divided close to the liver surface (Fig. 3.5). As the left lateral segment is released from its peritoneal attachments, it is also useful to place a hand or laparotomy pad under the left lateral segment and anterior to the caudate to provide traction and to protect the stomach, bowel, and spleen from diathermic injury. Further mobilization of the left liver can be accomplished by dividing the lesser omentum as well as the ligamentum venosum either at the left portal vein or left hepatic vein insertions to expose these vessels and the underlying caudate lobe.

vascular isolation
Once the liver is mobilized, there are essentially three steps to safely perform a hepatectomy. These involve vascular inflow control, vascular outflow control, and parenchymal transection. Inflow Control All major hepatic resections require control of the vascular inflow to be accomplished safely. Furthermore, adequate

Figure 3.3 Multiple small venous branches from the IVC to the posterior liver must be individually dissected and divided. When all of these branches are controlled, all that is left to expose the right hepatic vein will be a fibrous band of tissue, the caval ligament. A tunnel can be safely created behind this ligament and above the right hepatic vein with a Kelly clamp. Once this is divided, the entire right liver is mobile and the venous outflow can be encircled and controlled. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

27

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
hepatic arterial and portal venous inflow must be maintained to the remnant liver. Selective inflow control may be achieved extrahepatically (30), intrahepatically during parenchymal transection (1,31), or by intrahepatic pedicle control via hepatotomies (32,33). In the extrahepatic approach, the hepatic artery and portal vein branches are dissected at the porta hepatis and controlled outside of the liver. In this approach, the individual artery and portal vein have to be separately identified and ligated since they have not yet entered the liver as a portal pedicle. The advantages of this approach are early vascular control prior to transection and demarcation of the liver on its surface. The disadvantages are a somewhat tedious dissection and the potential for injury to contralateral structures. The presence of tumor abutting the hilum may mandate extrahepatic inflow control. The right hepatic artery usually courses posterior to the common hepatic bile duct and can be dissected from the right side of the porta hepatis and controlled. Once divided, the proximal artery stump can be retracted anteriorly exposing the underlying portal vein. All branches must be carefully dissected and identified prior to division to insure that there is no compromise of flow to the future liver remnant, a potentially fatal complication. There is typically a small branch to the caudate process coming off the right portal vein proximally that may have to be controlled. As opposed to the short extrahepatic course from the hilum to the right liver, the vascular inflow to the left liver can be controlled in the umbilical fissure (34). The left portal vein and duct are mobilized by lowering the hilar plate. Here the left hepatic artery is typically found running cephalad along the left side of the porta hepatis anteriorly. It is prudent to insure that one has not inadvertently ligated the artery proximal to the right hepatic artery takeoff by confirming a pulse to the right liver. Once the left hepatic artery is divided, the underlying left portal vein can be dissected behind it. A branch to caudate lobe is very constant and should be preserved if the caudate is not going to be resected. Proximal dissection and identification of the right portal vein from the left side is worthwhile to confirm anatomy. Unless mandated by tumor proximity, we prefer to transect the bile duct (left or right) intrahepatically during parenchymal transection to absolutely avoid contralateral injury. This is especially important on the left side where there is often variant drainage of the major right sectoral ducts to the left hepatic duct. An alternative to extrahepatic inflow control at the hilum is intrahepatic control using a pedicle ligation technique. This technique is most appropriate for right-sided tumors not encroaching on the hilus. The portal triads carry Glisson’s capsule with them into the liver substance forming a sturdy pedicular sheath that can be dissected and clamped. Exposure of the pedicles can be accomplished by parenchymal transection down to the sheaths or by hepatotomies in the liver substance above the pedicle. For exposure of the right-sided inflow pedicle(s), hepatotomies are typically made along the inferior part of the gallbladder fossa and the caudate process and a large clamp is used to encircle the inflow structures. The whole right pedicle can be controlled this way or the right anterior and posterior sectoral pedicles can be encircled separately. The approach is rapid and avoids dissection of the contralateral structures in the hilum, but risks injury to the pedicle before encircling the triad, or hemorrhage from coursing veins, which commonly run close to the pedicles. Though total vascular isolation has been employed by some groups (35–38), we have found that total vascular isolation techniques were not necessary in more than 1800 consecutive liver resections (2). Outflow Control Though there are multiple small veins that drain the right lobe and segment I directly into the retrohepatic vena cava, the majority of hepatic blood flow drains into the inferior vena cava (IVC) via the left, middle, and right hepatic veins. In

Figure 3.4 Sharp and blunt dissection over the suprahepatic IVC exposes the right, middle and left hepatic veins. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

Left phrenic vein

Diaphragm

Left triangular ligament

Diathermy

Figure 3.5 Downward traction on the liver and cephalad traction on the diaphragm help expose the left coronary ligament. Care must be taken here to identify the phrenic vein as it courses on the underside of the diaphragm to enter the IVC, as it can be inadvertently injured if the triangular ligament is not properly exposed or not divided close to the liver surface. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

28

HEPATIC RESECTION
major hepatectomy, extrahepatic control of these vessels is preferred. Standard anatomy consists of a single right hepatic vein entering the vena cava, and a left and middle hepatic vein that is joined and entering the cava as a single trunk. Autopsy studies of the left and middle hepatic venous trunk have elucidated at least five types of hepatic vein trunk variants (39). The right hepatic vein is typically encircled after the dissection of the vena cava and caval ligament has been carried out as described earlier. The base of the right hepatic vein should be dissected sharply and once exposed, a clamp can be passed between the right and middle hepatic veins. Exposure of the left and middle hepatic vein extraheaptically can be challenging. The groove between the right and middle hepatic veins is initially developed from above the liver. The left liver is mobilized and the ligamentum venosum is divided just before its insertion into the left hepatic vein. Here a tunnel is carefully developed underneath the middle and left hepatic vein and they are encircled (Fig. 3.6). It is often difficult to individually encircle the left or middle hepatic vein extrahepatically but this depends on the anatomy of the common trunk. It is important to identify the hepatic venous anatomy on preoperative imaging and recognize variations in the branching patterns, since bleeding in this area can be difficult to control. Ligation of the hepatic venous outflow of the liver can also be accomplished during parenchymal transection with careful exposure of the cava and the origin of these veins once the liver has been transected to expose them. The exposures for specific resections are discussed later in the chapter. Parenchymal Transection Once vascular inflow and outflow to the lobe or segment has been controlled, all that remains is division of the liver parenchyma. There are many techniques to accomplish this. The instruments used are left to the surgeon’s preference, but it is imperative that the vessels and ducts divided be identified and dissected before division. Transection of the liver should be a deliberate dissection of intrahepatic structures rather than simply coagulation of liver tissue. In addition to the ability to confidently ligate each branch on the transection line, it allows one to identify the venous drainage and pedicle inflow to the remnant liver. Moreover, in cases where the tumor margin is adjacent to major hepatic veins and portal pedicles, it allows precise extirpation of the tumor. For these reasons, we prefer a simple crushing technique. Glisson’s capsule is scored with diathermy along the transection line and a Kelly clamp is used to crush the liver tissue and expose the vessels and ducts for clipping, ligation, or bipolar energy sealing. Larger pedicles are suture ligated or stapled (40). The operative surgeon crushes the tissue in small linear planes, the assistant clips or seals the vessels, and the surgeon divides the structures. In this fashion, the transection line is quickly and efficiently completed. Though not always necessary, inflow occlusion with a Pringle maneuver may be used to decrease blood loss, and an entire lobe can often be transected with three to four sessions of 10–15 minutes on Pringle with 5 minutes off. After removal of the specimen, the raw surface is carefully inspected for bile leaks, which are suture ligated or clipped. Some groups advocate injection of dye or intralipid via the cystic duct to identify open biliary tributaries for ligation. Since drainage is associated with prolonged hospital stay, increased infection, and no change in a need for interventional radiology directed drainage, we do not routinely place drains after hepatic resection in the absence of biliary reconstruction (41).

major hepatic resection: definitions and specific considerations
Multiple descriptions of liver anatomy and resections by anatomists and surgeons have resulted in terminologies that can be confusing and imprecise. A recent consensus conference in Brisbane, Australia, with the American Hepato-PancreatoBiliary Association has published new guidelines to clarify this nomenclature. When unclear, or if there is confusion about the description of a resection, one should revert to naming the numerical segments involved. The right liver is comprised of segments V–VIII and the left liver is comprised of segments II–IV. Appropriate terms for resection of the right or left liver would be “hepatectomy” or “hemi-hepatectomy.”

(A)

(B)

Figure 3.6 (A) Medial retraction of the left lateral segment exposes the ligamentum venosum. (B) The ligamentum venosum is divided sharply where it is tethered to the left hepatic vein, releasing the vein and enabling a tunnel to be dissected under the middle and left hepatic veins and anterior to the IVC. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

29

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Extending a right hepatectomy to include segment IV or a left hepatectomy to include segments V and VIII would be described as a “right/left trisectionectomy” or “trisegmentectomy.” Resection of segments II and III is often referred to as a “left lateral segmentectomy” or “sectionectomy.” There are essentially five types of major resection. The nomenclature of these resections is based on the anatomical classification (Table 3.1) (Fig. 3.7) (42–45). Right Hemihepatectomy (Right Hepatectomy, Right Hepatic Lobectomy) A right hemihepatectomy involves excision of segments V–VIII. The right lobe is completely mobilized and the right hepatic vein is isolated. The peritoneum overlying the common bile duct and extending into Calot’s triangle is incised to expose the cystic artery and duct. These are ligated and divided. A long tie is left on the proximal cystic stump and used as a retractor to help expose the common bile duct and dissect the vasculature. The hilar plate is lowered to protect the left hepatic duct. We typically do not dissect the right hepatic duct extrahepatically, but address it during parenchymal transection to absolutely avoid any potential for injury to the left hepatic duct. The right hepatic artery usually passes posterior to the common bile duct (Fig. 3.8) and is sharply dissected, ligated, and divided to the right of the common duct. Superior traction on the right hepatic artery stump will help expose the portal vein. The portal bifurcation is approached laterally and posteriorly. When dissecting the right portal vein, care should be taken to identify the first posterior branch to the right side of the caudate. Circumferential control of the right portal vein should not be attempted until this branch is identified and dissected as bleeding from this vein can be troublesome. Once a few centimeters of right portal vein are fully exposed and the left portal vein has been visualized, it is encircled and divided. Clamping of the right portal vein at this point should confirm demarcation of the right liver. Occasionally, the right anterior and posterior sectoral portal vein branches arise independently from the portal vein. In this instance, they must be individually dissected and ligated after confirming flow to the left liver. The right hepatic vein is isolated and divided as described previously. It is important that all the retrohepatic veins are first controlled and divided, that the dissection extends to the left of the IVC, and that the right hepatic vein is skeletonized completely right at the liver surface. It is especially important to gain extrahepatic control of the vein with large tumors near the hepatic venous confluence or in the posterior sector near the vena cava, where it can be difficult to obtain tumor clearance without excessive traction on the vein. Alternatively, the right hepatic vein can also be controlled from within the liver during parenchymal transection, however, this usually forces the hepatic transection to the right of the true principal midliver plane. After inflow ligation, a line of demarcation becomes evident. Figure-of-eight stay sutures are placed to either side of this line and parenchymal transection can begin safely. The surgeon’s left hand lifts the left lobe from above the IVC carefully as the transection plane is deepened. This will expose the middle hepatic vein, and division of the specimen can proceed to the right or left of the vein depending on tumor clearance. As the dissection proceeds superiorly, the segment V and then VIII hepatic veins are divided along the middle hepatic vein. The main right portal pedicle is exposed and divided with the endo-GIA stapler. This will control the right hepatic duct if it was not controlled extrahepatically. Alternatively, an anterior approach can be used to resect the right lobe of the liver. This approach is advantageous when the right lobe cannot be mobilized due to a large right-sided tumor, or there is a large mass adherent to the diaphragm or IVC (46). In this approach, after extrahepatic inflow division, the liver is transected without mobilization. It is then freed from its venous and ligamentous attachments to the IVC and peritoneum. The parenchyma is transected from the anterior liver surface to the IVC along the line of demarcation, and venous tributaries are controlled from the front, including the right hepatic vein (47,48). To help control bleeding in the deeper parenchymal plane, the “hanging maneuver” may be employed (49). In this maneuver, the anterior plane of the IVC is dissected from the liver undersurface. The most inferior veins draining the caudate are ligated and divided, and a tunnel is carefully created anterior to the IVC to the space between the right and middle hepatic veins with a Kelly clamp. This is a blind tunnel of 4 to 6 cm. A tape is passed that can then be used to elevate the liver away from the anterior surface of the IVC, helping to define the plane of transection and facilitating exposure of the deeper tissues. In this technique, the right portal pedicle

Table 3.1 Anatomy and Classification of Major Hepatic Resections
Anatomic Classification Couinaud Right hepatectomy Right lobectomya Left hepatectomy Extended left hepatectomya Left lobectomy
a b

Goldsmith and Woodburne Right hepatic lobectomy Extended right hepatic lobectomy Left hepatic lobectomy Extended left lobectomy Left lateral segmentectomy

Brisbane Right hemihepatectomy Right trisectionectomy Left hemihepatectomy Left trisectionectomy Left lateral sectionectomy

Segments resected V, VI, VII, VIII IV,V,VI, VII, VIIIb II, III, IV II, III, IV, V, VIIIb II, III

Often referred to as trisegmentectomy. May also include segment I.

30

HEPATIC RESECTION
is divided, parenchymal transection is completed to the IVC, the lateral venous attachments to the IVC are ligated and divided, the right hepatic vein is stapled, the coronary and triangular ligaments are divided, and the specimen is removed. Right Trisectionectomy (Right Lobectomy, Extended Right Lobectomy, Right Trisegmentectomy) A right trisectionectomy is a right hemihepatectomy extended to include segment IV. The liver is mobilized as described for a right hepatectomy. To approach the inflow and outflow of segment IV, the ligamentum teres is elevated to expose the umbilical fissure. If a bridge of tissue between segments III and IV is present concealing the fissure, this should be divided with diathermy. Here, the ligamentum teres can be traced to its embryologic origin at the left portal vein. Incising the fibrous tissue that tethers the left main pedicle to the base of the umbilical fissure releases the left-sided structures from the undersurface of segment IV, and it opens up the fissure. To safely perform a right trisectionectomy, the left hepatic duct should be freed clear of the proposed plane of transection. This is accomplished by lowering the hilar plate as previously described. The inflow and outflow to the right liver are controlled and divided as previously described. Once the right hepatic vein has been divided, the middle vein can usually be encircled. The liver tissue is divided to the right of the falciform ligament and the pedicles feeding segments IVa and IVb are ligated and divided as they come off the main left pedicle (Fig. 3.9). Unless tumor mandates, a deliberate dissection within the umbilical fissure is usually not necessary. As the plane of transection is deepened toward the IVC superiorly, the middle hepatic vein is encountered, dissected, and divided with a stapler. It is absolutely critical to protect the left hepatic vein as narrowing or transection of this vein will likely result in liver failure or massive hemorrhage secondary to a lack of other venous return from the liver.

(A)

(B)

(C)

(D)

(E) Figure 3.7 The anatomy and classification of major hepatic resections. (A) right hepatectomy, (B) left hepatectomy, (C) left lobectomy, (D) extended left hepatectomy, (E) right lobectomy.

31

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 3.9 To expose and control the portal pedicles to segment IV, the liver tissue is divided to the right of the falciform ligament and the pedicles feeding segments IVA and IVB are ligated and divided as they come off the main left pedicle. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

Figure 3.8 During right hepatectomy, the right hepatic artery usually passes posterior to the common bile duct and is sharply dissected, ligated, and divided to the right of the duct. After cholecystectomy, retraction of the cystic duct will expose the underlying artery. Source: Blumgart; Surgery of the Liver, Biliary Tract and Pancreas, 4th Edition; Chapter 80; copyright Elsevier.

Left Hemihepatectomy (Left Hepatectomy, Left Hepatic Lobectomy) A left hemihepatectomy involves excision of segments II– IV. The left lobe of the liver is mobilized, the umbilical fissure is exposed, and the hilar plate is lowered as previously described. The gastrohepatic ligament is entirely divided, with care taken to identify any accessory or replaced left hepatic arteries not identified on preoperative imaging. The left hepatic artery is dissected at the base of the umbilical fissure and divided. The caudate branch of the portal vein is identified before the left main portal vein enters the umbilical fissure. If the caudate lobe is to be spared, the portal vein is ligated and divided distal to this vein. A line of demarcation marking the right-sided border of segment IV corresponds with a plane that usually extends from the IVC to the base of the gallbladder fossa (“Cantlie’s line”). This “principle plane” is the same as that seen in a right hemihepatectomy. Segments II and III are reflected medially and the middle and left hepatic veins are identified, encircled and divided extrahepatically as described earlier. The left hepatic vein is often not amenable to circumferential extrahepatic exposure initially but can be exposed after splitting the liver back to its origin. Parenchymal transection completes the excision. Left Trisectionectomy (Extended Left Hepatectomy, Extended Left Lobectomy, Left Trisegmentectomy) A left trisectionectomy involves removal of segments II, III, IV, V, and VIII. The entire liver is mobilized. The inflow and outflow to the left lobe are controlled as previously

described for a left hemihepatectomy. The inflow to segments V and VII can be addressed in a few ways. The anterior sectoral pedicle can be encircled intrahepatically either through hepatotomies or after transection in the right scissura to the left of the right hepatic vein. The pedicle can be encircled and clamped confirming flow the posterior sector. Alternatively, an extensive hilar dissection can be carried out to identify and divide the arterial and portal branches to the right anterior sector. It is critical that preoperative imaging is reviewed for anatomic variations in the inflow and outflow to the right liver. Once the anterior sectoral inflow is divided, a near horizontal line of demarcation becomes evident anterior to the right hepatic vein and dividing the right anterior and posterior sectors. Parenchymal transection continues anterior to the right hepatic vein and the specimen is removed. The middle hepatic vein is necessarily taken as part of this resection and is addressed as described earlier. A left trisectionectomy is a challenging operation that requires significant experience with major hepatic resections. Left Lateral Sectionectomy (Left Lobectomy, Left Lateral Segmentectomy) A left lateral sectionectomy involves removal of segments II and III. The left lobe of the liver is mobilized and the hilar plate is lowered as previously described. Just to the left of the umbilical fissure, the portal pedicles to segments II and III are identified and divided. These can be identified and controlled through multiple hepatotomies or during parenchymal transaction in a plane just to the left of the falciform ligament. A deliberate dissection in the umbilical fissure is usually not necessary. The left hepatic vein is usually divided after parenchymal resection back to its origin but can also be controlled extrahepatically as described in the “outflow control” section of the chapter.

32

HEPATIC RESECTION

wedge vs. segmental resection
The segmental anatomy of the liver, as defined by Couinaud, divides the liver into eight independent segments, each with its own inflow and biliary drainage (see chapter 1) (42,50). As a result, each segment can be individually resected without affecting the inflow or outflow to the rest of the liver. Segment-oriented hepatectomy spares normal parenchyma and is particularly useful when bilateral noncontiguous segments are involved or in patients with chronic liver disease. Nonanatomic wedge resections can be useful for small peripheral tumors that are not close to major inflow pedicles or venous branches for which adequate tumor margins can be obtained. Though some groups have shown that anatomical resection is not superior to wedge resection for tumor clearance, pattern of recurrence, or survival (51), in our experience anatomic segmental resection resulted in improved tumor clearance and patient survival compared to wedge resection (52). Wedge excision may risk fracturing the plane between the tumor and normal liver, margin positivity, and intraoperative hemorrhage (12,53). Anatomic resection may provide better visibility, decrease the risk of major hemorrhage, and in many cases provide a wider margin of resection. Segmentectomy I (Caudate Resection) The caudate lobe is often resected with a right or left hemihepatectomy, however, isolated caudate resection may be performed for solitary tumors in segment I. The anatomy of the caudate lobe between the IVC, portal triad, and hepatic veins can make resection tedious and challenging. The caudate lobe straddles both hepatic lobes and therefore receives vascular inflow from both the right and left portal pedicles (54). Venous drainage is directly into the IVC via one to nine short hepatic veins (55). The left edge of the caudate fuses with the IVC via a fibrous band of tissue that encircles the IVC and attaches to segment VII. In many patients, this caval ligament may be composed of liver parenchyma. Dissection at the base of the umbilical fissure exposes the caudate branches of the left portal vein and hepatic artery for ligation and division. Segments II and III of the liver are mobilized and reflected to the right, exposing the caudate where it lies on the IVC. The left lateral attachments of the caudate to the IVC are divided (56). Exposure and division of the left caval ligament can be challenging and care should be taken to avoid injury to the cava inferiomedially and the base of the left and middle hepatic veins superiorly. With anterior traction on the caudate, the short hepatic veins draining into the IVC on the posterior aspect of the caudate can be visualized and controlled. If there is a bulky tumor in the caudate or anterior traction of the lobe is difficult, the retrohepatic veins can be approached from the right side, by mobilizing the right lobe and turning it to the left, then dissecting and dividing all the veins starting below the caudate and continuing onto the anterior surface of the IVC (57). The caudate branch from the right portal vein should also be identified and ligated. To complete the resection, the tissue joining the caudate to segment VII must be transected. Anteriorly and superiorly,

care must be taken to avoid injury to the middle and left hepatic veins. Segments II or III The approach to excising either segment II or III is the same as that for a left lateral segmentectomy, except the plane between the segments needs to be defined. This plane is identified by the course of the left hepatic vein, segment 3 being anterior and segment 2 being posterior. Inflow control to either segment is achieved in the umbilical fissure. Ligation of the portal pedicle will guide resection along the plane of demarcation. Care must be taken to divide the branches of the left hepatic vein draining the excised segment, but to leave the main left hepatic vein intact to drain the remnant liver. Segment IV As described in a right trisectionectomy, the inflow to segment IV is found to the right of the umbilical fissure. The hilar plate is lowered to protect the left bile duct and to provide access to the multiple pedicles to segment IV. Ligation of these pedicles will provide a line of demarcation along Cantlie’s line. During parenchymal transection, the venous drainage of segments IVa and IVb are divided sequentially to the left of the middle vein on the lateral border of the segment, and along the umbilical fissure on the medial border of the segment, where the umbilical vein often courses. The middle hepatic vein can be sacrificed in this operation if necessary with adequate drainage of the right liver and segments 2 and 3. Segments V and VIII (Anterior Sector) The inflow to segments V and VIII are from the right anterior sectoral pedicle. This can be approached and controlled extrahepatically or intrahepatically as described for a left trisectionectomy. If the anterior and posterior sectoral pedicles branch within the liver parenchyma, a hepatotomy over the anterior pedicle is necessary. Alternatively, the liver can be transected in the principal plane down to the base of the anterior sector where its origin can be controlled, typically posterior to terminal middle hepatic vein branches. The anterior sector lies between the right and middle hepatic veins, i.e., between Cantlie’s line and a transverse plane anterior to segments VI and VII. This horizontal plane of transection can be better defined by clamping the anterior pedicle to demarcate the right, left, and posterior borders. The transection line between V and VIII is demarcated and defined intrahepatically when control of the isolated segmental inflow is obtained. The middle hepatic vein can usually be safely divided in this operation if necessary, but in the absence of a large accessory right hepatic vein, the right hepatic vein must be preserved for adequate drainage of the posterior sector. Segments VI and VII (Posterior Sector) The inflow to segments VI and VII are from the posterior sectoral pedicle. This can often be approached and controlled in the fissure of Ganz, though the anatomy of anterior and posterior pedicles can be highly variable. If the anterior and posterior sectoral pedicles branch within the liver parenchyma,

33

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
the portal pedicles must be approached during parenchymal transection. The medial plane of transection can be better defined by clamping the posterior pedicle to demarcate the border. The classic description of a single pedicle from the posterior sectoral pedicle feeding either segment VI or VII is the exception rather than the rule (58), therefore, careful parenchymal dissection, preoperative study of the CT, and intraoperative ultrasound are critical to these resections. If a posterior sectorectomy is to be performed, the right hepatic vein can be sacrificed since the anterior sector drains into the middle hepatic vein. Central Hepatectomy (Segments IV, V, and VIII) A central hepatectomy with various amount of extension into any of the three segments can be performed combining the techniques described above. Typically this requires dividing the middle hepatic vein intrahepatically near its origin. The techniques of a segment IV resection and anterior sectorectomy are essentially combined. This is a challenging operation that requires a substantial surface of liver to be transected but can be very useful to spare parenchyma while removing centrally placed tumors.
2. Baer HU, Dennison AR, Mouton W, et al. Enucleation of giant hemangiomas of the liver: Technical and pathologic aspects of a neglected procedure. Ann Surg 1992;216(6):673–6. 3. Gedaly R, Pomposelli JJ, Pomfret EA, Lewis WD, Jenkins RL. Cavernous hemangioma of the liver: anatomic resection vs. enucleation. Arch Surg 1999 Apr;134(4):407–11. 4. Yoon SS, Charny CK, Fong Y, et al. Diagnosis, management, and outcomes of 115 patients with hepatic hemangioma. J Am Coll Surg 2003;197(3): 392–402. 5. Are C, Gonen M, Zazzali K, et al. The impact of margins on outcome after hepatic resection for colorectal metastasis. Ann Surg 2007 Aug;246(2): 295–300. 6. Cady B, Jenkins RL, Steele Jr GD, et al. Surgical margin in hepatic resection for colorectal metastasis: A critical and improvable determinant of outcome. Ann Surg 1998;227(4):566–71. 7. Cady B, Stone MD, McDermott Jr WV, et al. Technical and biological factors in disease-free survival after hepatic resection for colorectal cancer metastases. Arch Surg 1992;127(5):561–9. 8. Ekberg H, Tranberg KG, Andersson R. Determinants of survival in liver resection for colorectal secondaries. Br J Surg 1986;73(9):727–31. 9. Elias D, Cavalcanti A, Sabourin JC, et al. Resection of liver metastases from colorectal cancer: The real impact of the surgical margin. Eur J Surg Oncol 1998;24(3):174–9. 10. Kato T, Yasui K, Hirai T, et al. Therapeutic results for hepatic metastasis of colorectal cancer with special reference to effectiveness of hepatectomy: Analysis of prognostic factors for 763 cases recorded at 18 institutions. Dis Colon Rectum 2003;46:522–31. 11. Ohlsson B, Stenram U, Tranberg KG. Resection of colorectal liver metastases: 25-year experience. World J Surg 1998;22(3):268–77. 12. Scheele J, Stang R, Altendorf-Hofmann A, Paul M. Resection of colorectal liver metastases. World J Surg 1995;19(1):59–71. 13. Shirabe K, Takenaka K, Gion T, et al. Analysis of prognostic risk factors in hepatic resection for metastatic colorectal carcinoma with special reference to the surgical margin. Br J Surg 1997;84(8):1077–80. 14. Hamady ZZR, Cameron IC, Wyatt J, et al. Resection margin in patients undergoing hepatectomy for colorectal liver metastasis: A critical appraisal of the 1 cm rule. Eur J Surg Oncol 2006;32(5):557–63. 15. Pawlik TM, Scoggins CR, Zorzi D, et al. Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 2005;241(5):715–24. 16. Mimica Z, Pogorelic Z, Perko Z, et al. Effect of surgical incision on pain and respiratory function after abdominal surgery: a randomized clinical trial. Hepatogastroenterology 2007;54(80):2216–20. 17. Llovet JM, Fuster J, Bruix J. The Barcelona approach: diagnosis, staging, and treatment of hepatocellular carcinoma. Liver Transpl 2004;10 (2 Suppl 1):S115–20. 18. Zorzi D, Laurent A, Pawlik TM, et al. Chemotherapy-associated hepatotoxicity and surgery for colorectal liver metastases. Br J Surg. 2007;94(3): 274–86. 19. Kooby DA, Fong Y, Suriawinata A, et al. Impact of steatosis on perioperative outcome following hepatic resection. J Gastrointestinal Surg 2003;7(8):1034–44. 20. Siegelman ES, Rosen MA. Imaging of hepatic steatosis. Semin Liver Dis 2001;21(1):71–80. 21. Orlacchio A, Bolacchi F, Cadioli M, et al. Evaluation of the severity of chronic hepatitis C with 3-T1H-MR spectroscopy. AJR Am J Roentgenol 2008;190(5):1331–9. 22. Abulkhir A, Limongelli P, Healey AJ, et al. Preoperative portal vein embolization for major liver resection: a meta-analysis. Ann Surg 2008;247(1):49–57. 23. Covey AM, Brown KT, Jarnagin WR, et al. Combined portal vein embolization and neoadjuvant chemotherapy as a treatment strategy for resectable hepatic colorectal metastases. Ann Surg 2008;247(3):451–5. 24. Belghiti J. Arguments for a selective approach of preoperative portal vein embolization before major hepatic resection. J Hepatobiliary Pancreat Surg 2004;11(1):21–4. 25. Lamade W, Glombitza G, Fischer L, et al. The impact of 3-dimensional reconstructions on operation planning in liver surgery. Arch Surg 2000;135(11):1256–61.

enucleation of benign tumors ₍see chapters 28, 32, and 33₎
When indicated, hepatectomy for benign conditions should be parenchymal preserving. Though anatomic resection along segmental planes is sometimes necessary, some benign tumors may be enucleated, for example, adenomas, fibronodular hyperplasia, metastatic neuroendocrine tumors, and hemangiomas (2,4). Hemangiomas in particular push liver tissue away as they grow, and create a fibrolamellar plane of tissue that defines the border between cavernous tissue and normal liver parenchyma (2). The arterial supply to the hemangioma can be determined from preoperative imaging and is clamped, allowing the tumor to decompress via the venous outflow. The hepatic tissue over the mass is then incised to enter an avascular plane surrounding the tumor. Small vessels that traverse this plane are ligated and divided. The majority of the dissection can be done with the surgeon’s finger, and the mass is shelled out. This approach preserves normal parenchyma, eliminates the need for hepatic venous outflow control, limits blood loss, and has fewer complications than lobectomy (3,4). Management of benign lesions is covered in further detail in other chapters.

conclusion
Major hepatic resections for benign and malignant tumors can be accomplished safely and efficaciously. Proper patient selection, precise preoperative imaging, specific anesthetic techniques, and knowledge of the principal complications are essential. Study of each patient’s segmental anatomy will allow inflow and outflow control and the ability to tailor the resection needed for each individual.

references
1. Jarnagin WR, Gonen M, Fong Y, et al. Improvement in perioperative outcome after hepatic resection: Analysis of 1,803 consecutive cases over the past decade. Ann Surg 2002;236(4):397–407.

34

HEPATIC RESECTION
26. Cunningham JD, Fong Y, Shriver C, et al. One hundred consecutive hepatic resections: Blood loss, transfusion, and operative technique. Arch Surg 1994;129(10):1050–6. 27. Melendez JA, Arslan V, Fischer ME, et al. Perioperative outcomes of major hepatic resections under low central venous pressure anesthesia: Blood loss, blood transfusion, and the risk of postoperative renal dysfunction. J Am Coll Surg 1998;187(6):620–5. 28. D’Angelica M, Fong Y, Weber S, et al. The role of staging laparoscopy in hepatobiliary malignancy: prospective analysis of 401 cases. Ann Surg Oncol 2003 Mar;10(2):183–9. 29. D’Angelica M, Maddineni S, Fong Y, et al. Optimal abdominal incision for partial hepatectomy: increased late complications with Mercedes-type incisions compared to extended right subcostal incisions. World J Surg 2006;30(3):410–8. 30. Blumgart L. Hepatic resection. In: Dudley HAF, Rob C, Smith of Marlow RS, Pories WJ, eds. Rob & Smith’s operative surgery, 4th edn. London, Boston: Butterworth Scientific, 1982:v. 31. Tung TT. Les Re?sections Majeures et Mineures du Foie. 1979. 32. Launois B, Jamieson GG. The posterior intrahepatic approach for hepatectomy or removal of segments of the liver. Surg Gynecol Obstetrics 1992;174(2):155–8. 33. Launois B, Jamieson GG. The importance of Glisson’s capsule and its sheaths in the intrahepatic approach to resection of the liver. Surg Gynecol Obstetrics 1992;174(1):7–10. 34. Hepp J, Couinaud C. [Approach to and use of the left hepatic duct in reparation of the common bile duct.]. Presse Med. 1956;64(41):947–8. 35. Delva E, Camus Y, Nordlinger B, et al. Vascular occlusions for liver resections. Operative management and tolerance to hepatic ischemia: 142 cases. Ann Surg 1989;209(2):211–8. 36. Emond J, Wachs ME, Renz JF, et al. Total vascular exclusion for major hepatectomy in patients with abnormal liver parenchyma. Arch Surg 1995;130(8):824–30; discussion 30–1. 37. Emre S, Schwartz ME, Katz E, Miller CM. Liver resection under total vascular isolation. Variations on a theme. Ann Surg 1993;217(1):15–19. 38. Hannoun L, Borie D, Delva E, et al. Liver resection with normothermic ischaemia exceeding 1 h. Br J Surg. 1993;80(9):1161–5. 39. Nakamura S, Tsuzuki T. Surgical anatomy of the hepatic veins and the inferior vena cava. Surgery Gynecol Obstetrics 1981;152(1):43–50. 40. Fong Y, Blumgart LH. Useful stapling techniques in liver surgery. J Am Coll Surgeons 1997;185(1):93–100. 41. Fong Y, Brennan MF, Brown K, Heffernan N, Blumgart LH. Drainage is unnecessary after elective liver resection. Am J Surg 1996;171(1):158–62. 42. Couinaud C. Le foie. Etudes anatomiques et chirurgicales. Le Foie: Etudes Anatomiques et Chirurgicales. 1957. 43. Goldsmith NA, Woodburne RT. The surgical anatomy pertaining to liver resection. Surg Gynecol Obstet 1957;105:310–8. 44. Starzl TE, Iwatsuki S, Shaw BW, Jr, et al. Left hepatic trisegmentectomy. Surg Gynecol Obstet 1982;155(1):21–7. 45. Starzl TE, Koep LJ, Weil R, 3rd, et al. Right trisegmentectomy for hepatic neoplasms. Surg Gynecol Obstet 1980;150(2):208–14. 46. Chik BH, Liu CL, Fan ST, et al. Tumor size and operative risks of extended right-sided hepatic resection for hepatocellular carcinoma: implication for preoperative portal vein embolization. Arch Surg 2007;142(1):63–9; discussion 9. 47. Lai EC, Fan ST, Lo CM, Chu KM, Liu CL. Anterior approach for difficult major right hepatectomy. World J Surg 1996;20(3):314–7; discussion 8. 48. Lai ECS, Fan ST, Lo CM, et al. Hepatic resection for hepatocellular carcinoma: An audit of 343 patients. Ann Surg 1995;221(3):291–8. 49. Belghiti J, Guevara OA, Noun R, Saldinger PF, Kianmanesh R. Liver hanging maneuver: a safe approach to right hepatectomy without liver mobilization. J Am Coll Surg 2001;193(1):109–11. 50. Couinaud C. Bases anatomiques des hepatectomies gauche et droite reglees. J Chir 1954;70:933–66. 51. Zorzi D, Mullen JT, Abdalla EK, et al. Comparison between hepatic wedge resection and anatomic resection for colorectal liver metastases. J Gastrointest Surg 2006;10(1):86–94. 52. DeMatteo RP, Palese C, Jarnagin WR, et al. Anatomic segmental hepatic resection is superior to wedge resection as an oncologic operation for colorectal liver metastases. J Gastrointest Surg 2000;4(2):178–84. 53. Polk W, Fong Y, Karpeh M, Blumgart LH. A technique for the use of cryosurgery to assist hepatic resection. J Am Coll Surg 1995;180(2):171–6. 54. Mizumoto R, Suzuki H. Surgical anatomy of the hepatic hilum with special reference to the caudate lobe. World J Surg 1988;12(1):2–10. 55. Heloury Y, Leborgne J, Rogez JM, et al. The caudate lobe of the liver. Surg Radiol Anat 1988;10(1):83–91. 56. Takayama T, Makuuchi M. Intraoperative ultrasonography and other techniques for segmental resections. Surg Oncol Clin N Am 1996;5(2):261–9. 57. Lerut J, Gruwez JA, Blumgart LH. Resection of the caudate lobe of the liver. Surgery Gynecol Obstetrics 1990;171(2):160–2. 58. Hata F, Hirata K, Murakami G, Mukaiya M. Identification of segments VI and VII of the liver based on the ramification patterns of the intrahepatic portal and hepatic veins. Clin Anat 1999;12(4):229–44.

35

4

Ultrasound for HPB disorders Duan Li and Lucy Hann
is that ultrasound is operator-dependent; skilled technologists and radiologists are essential since diagnosis is made at image acquisition. For best results, the surgeon should communicate to the radiologist the specific clinical questions so that appropriate targeted images can be obtained at the time of the examination. This chapter will discuss ultrasound applications for diagnosis of hepatic, gallbladder, biliary, and pancreatic abnormalities. The role of specialized ultrasound techniques such as endoscopic ultrasound and intraoperative ultrasound will also be addressed.

introduction
Ultrasound is the initial study of choice in most clinical situations due to the lack of ionizing radiation, relatively low cost, and accessibility in varied settings such as at the bedside or in the operating suite. Ultrasound differs from other crosssectional imaging techniques in that it uses sound propagation and reflection from interfaces within tissue for imaging. Images are generated by piezoelectric material within the transducer that transmits and receives the sound signal. Higher frequency transducers provide the best resolution, but high frequencies are attenuated more rapidly in tissue. For that reason, transducer frequency is selected for the application. Superficial structures are evaluated at frequencies in the range of 6 to 18 MHz and transabdominal ultrasound, which requires better penetration, typically uses frequencies ranging from 3 to 6 MHz. Doppler is a unique feature of ultrasound for imaging vessels and blood flow. When moving blood is insonated, the frequency of the returning signal is proportional to blood velocity. A cursor is placed over a specific blood vessel and images are obtained in both gray scale and Doppler (termed “Duplex scanning”). The Doppler information can then be displayed in three different formats: (1) spectral Doppler, (2) color Doppler, and (3) power Doppler. Spectral Doppler shows a waveform with velocity changes and flow direction over time. Color Doppler displays mean velocities and direction of flow within vessels. The color codes assigned for velocities are usually displayed in the upper left aspect of the image. Power Doppler gives the amplitude of the Doppler signal without direction or frequency information; since it is not angledependent, it is very useful for imaging low flow and tortuous vessels. Ultrasound contrast agents further improve applications for vascular imaging. Current contrast agents use microbubbles encapsulated within thin lipid spheres. After intravenous injection, the microbubbles remain intravascular and do not diffuse into the interstitium as do MRI and CT contrast agents. After a low-power ultrasound signal is applied, the microbubbles oscillate (expand and contract) at harmonic frequencies that are detected by the transducer (1,2). With these ultrasound contrast agents, it is now possible to image tumor vasculature in exquisite detail (3–7) (Fig. 4.1). Despite the versatility of ultrasound, there are limitations. Sound is reflected at bone and air interfaces so scans are obtained from different positions to avoid intestinal air or rib artifact. This lack of standardized perspective compared to axial imaging format of CT and MRI may present difficulty for referring clinicians who are unfamiliar with the technique. To lessen bowel gas interference, 6-hour fast is recommended to improve visualization of the pancreas and liver and to provide sufficient gallbladder distension. Another significant limitation

liver
Anatomically the liver is divided into sectors that are defined by the scissurae that contain the hepatic veins; these sectors are then subdivided into individual hepatic segments that each contain intact portal and arterial inflow and hepatic venous outflow and draining bile ducts (8,9). Ultrasound hepatic anatomy is shown in Fig. 4.2. Diffuse Liver Disease Diffuse liver abnormalities include fatty infiltration, hepatitis, and cirrhosis. Hepatic steatosis is present in 17% to 33% of the general population and in 70% of overweight individuals (10). On ultrasound, the liver has diffusely increased echogenicity and in advanced cases significant sound beam attenuation obscures the deep liver. Areas of focal sparring may be seen anterior to the portal confluence and adjacent to the gallbladder. Hepatic steatosis impacts perioperative outcome and accurate preoperative diagnosis would be useful (11). Fatty infiltration increases liver stiffness, which can be measured by tissue displacement in response to the transmitted ultrasound wave. These elastography techniques hold promise for diagnosis of diffuse infiltrative liver diseases such as hepatic steatosis and early-stage hepatic fibrosis (12–14). Focal Hepatic Lesions Cystic lesions Ultrasound is the best modality to differentiate cystic from solid liver masses and to determine the internal architecture of cystic lesions. Simple cysts, found in 2% to 3% of patients (15), have thin wall, no internal echoes, and bright posterior enhancement. Even if the cyst is lobulated or has thin septation, benign diagnosis can be made (16). Symptomatic large simple cysts may be treated with ultrasound-guided aspiration and sclerosis (15,17), but it is extremely important to assess the cyst wall. Mural nodularity, thick tumor rim, and internal vascularity may indicate neoplasm such as biliary cystadenoma and these lesions should not be unroofed or aspirated since complete surgical resection is required. Cystic liver metastases present as complex cysts often with solid or

36

ULTRASOUND FOR HPB DISORDERS

(A)

(B)

Figure 4.1 Microbubble contrast enhanced ultrasound image of a hypervascular liver mass. (A) Contrast enhanced image shows the intense hypervascularity of this liver lesion (arrow) that proved to be focal nodular hyperplasia. (B) The lesion (arrows) is subtle on the corresponding grayscale image. (Complements of Siemens Medical Solutions, Ultrasound Division. Malvern, PA.)

irregular rim. These are typically from sarcoma, cystadenocarcinomas of the ovary and pancreas, and mucinous colon carcinoma primaries (16,18). Ovarian metastases are characteristically peripheral implants. Squamous cell tumors with necrosis appear as cystic masses and other metastases may cavitate in response to chemotherapy. The appearance of cyst contents on ultrasound can be used for differential diagnosis. Pyogenic abscess initially may be echogenic and later liquified with debris, fluid-fluid levels, and irregular wall (Fig. 4.3). Echogenic reflections with reverberations, seen in 20% to 30% of cases, suggest air within the abscess (18). The classic echinococcal cyst is a complex cyst with well-defined wall, containing double echogenic lines. Multiple, internal echogenic foci, “snowstorm signs” settle in the dependent portions of the cyst. Localized splits in the cyst wall, with floating, undulating membranes, are also characteristic and the cyst wall may calcify (19,20). Hematomas in the acute stage may be echogenic and then they have layering lowlevel echoes from blood, and later become honeycombed with septation. When a preexisting cyst becomes hemorrhagic, internal septation may be thick and irregular, but they float freely in real-time and are not rigid. Solid Liver Lesions Solid liver lesions are further characterized by lesion echogenicity, vascularity, and peripheral halo. Definitive diagnosis of benignity can be made for hemangiomas, focal fatty infiltration, and focal fatty sparing because of their classic ultrasound features. Benign focal nodular hyperplasia can also be identified when the characteristic “spokewheel” vascular pattern, tortuous feeding artery, and marked hypervascularity are seen on contrast-enhanced ultrasound or Doppler images (Fig. 4.1). Hypoechoic liver masses and lesions with a peripheral halo are suspicious for malignancy. Although CT and MRI are used for tumor staging, there can be added benefit from ultrasound to

(1) assess lesions that are “too small to characterize” by CT, (2) define the relationship of tumor to bile ducts, and (3) evaluate vascular encasement and tumor margin (Fig. 4.4). Typical hemangioma, seen in 70% to 80% of cases, is a uniformly echogenic mass with sharp margin (21) (Fig. 4.5A). The multiple vascular interfaces within the hemangioma cause the increased echogenicity and margin is well-demarcated since histopathologically hemangiomas lack a capsule. Hemangiomas have absent or minimal flow on Doppler imaging; they are never hypervascular. Another common appearance of hemangioma is a mass with thin peripheral echogenic rim with mixed central echogenicity (Fig. 4.5B). Giant hemangiomas > 5 cm often lack these characteristic ultrasound features because of central fibrosis, necrosis, and myxomatous degeneration. A study of 213 patients with typical hemangioma appearance and without risk for hepatic malignancy found only one patient with malignancy on long term follow-up and concluded that typical hemangiomas in low-risk patients do not require follow-up (22). This rule does not apply to patients with cirrhosis, hepatitis, or chronic liver disease that places them at increased risk for hepatocellular carcinoma, nor does it apply to patients who already have malignancies, and particularly not to those with primary tumors that exhibit echogenic metastases. Caturelli et al. (23) studied 2,000 patients with cirrhosis. Of these, 44 had hemangioma-like lesions. On follow-up, half proved to be hepatocellular carcinomas and half hemangiomas. Thus, in patients at risk for hepatocellular carcinoma, any echogenic lesion merits further evaluation or follow-up. Other benign conditions such as focal fatty infiltration and focal sparing are diagnosed by geographic margins and typical location in segment 4 anterior to the portal vein bifurcation or less commonly, adjacent to the gallbladder. Focal fat appears echogenic relative to normal liver and areas of focal sparing are less echogenic than fatty infiltrated liver. A useful finding on Doppler evaluation is that vessels cross undisturbed

37

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
without displacement through areas of focal fat or focal sparring (24,25). A hypoechoic halo around a liver lesion indicates a clinically significant mass, suspicious for malignancy, including hepatocellular carcinoma and hepatic adenoma and metastases from colorectal, gastrointestinal, neuroendocrine, renal cell, choriocarcinoma, and vascular primaries such as Kaposi sarcoma (Fig. 4.4). Pathologically, the halo is caused by proliferating malignant cells, compression of the liver parenchyma, and dilated sinusoids. The hypoechoic halo sign has a 95% positive predictive value and an 87% negative value for differentiating metastases from hemangioma (26,27). A hypoechoic halo may be seen even in small lesions <1.5 cm. The halo is detected when the tumor is hyperechoic relative to the surrounding liver. In hypoechoic tumors, the lesion and the halo have the similar echogenicity and therefore the halo sign is not evident. Hepatocellular carcinoma (HCC) has a variable sonographic appearance ranging from hypoechoic to echogenic, but a hyperechoic lesion with hypoechoic halo is the common presentation. Small satellite tumors are typically hypoechoic. Doppler evaluation is the key in diagnosing HCC since the tumor is hypervascular and invasion of the portal or hepatic

(A)

(B)

(C)

(D)

(E) Figure 4.2 Normal liver anatomy. (A) Transverse view of the right lobe. The middle hepatic vein separates the right from left hepatic lobes. The right hepatic vein divides the right anterior sector (segments 8 and 5) and the right posterior sector (segments 7 and 6). R = right hepatic vein, M = middle hepatic vein, IVC = inferior vena cava. (B) Longitudinal view of the right lobe reveals the right hepatic vein RHV and the hepatic segments. RK = right kidney. (C) Transverse view of the portal vein bifurcation. Segments are numbered. R = right portal vein, L = left portal vein, RK = right kidney, IVC = inferior vena cava, A = aorta. (D) Longitudinal view of the left lobe. The left hepatic vein separates the posterior left sector segment 2 from the anterior sector (segments 3 and 4). The caudate, segment 1, is demarcated anteriorly by the fissure for the ligamentum venosum (arrowhead) and the inferior vena cava posteriorly. IVC = inferior vena cava. RPV = right portal vein. (E) Color Doppler sagittal image of the portal vein reveals hepatopetal flow.

38

ULTRASOUND FOR HPB DISORDERS
veins is very common; about 40% of patients have portal venous involvement and 25% show hepatic venous involvement (28). Flow within the tumor is usually of high velocity and low resistance due to arterial-venous shunting within the tumor. Tumor thrombus from HCC can be distinguished from bland thrombus when arterial flow is detected within the thrombus (29). With ultrasound contrast agents, the hypervascularity and dysmorphic vessels in HCC are more apparent and there is washout in the portal venous phase. Hypoechoic liver masses are suspicious for malignancy. Liver metastases that are hypoechoic are most commonly from breast, lung, esophagus, stomach, pancreas, and non-Hodgkin lymphoma.

gallbladder and bile ducts
Ultrasound is the procedure of choice for evaluation of the gallbladder and bile ducts. Gallstones are mobile, echogenic, and have posterior acoustic shadows (Fig. 4.5). The shadowing is present with or without gallstone calcification; it is due to the acoustic mismatch between stone and surrounding bile. Stones <3 mm may not generate a shadow because of small size. Gallstones move quickly with positional variation in contrast to sludge, which moves slowly and lacks an acoustic shadow. Gallbladder polyps do not shadow and they are fixed in position. Management of gallbladder polyps depends on size and sonographic appearance. Cholesterol polyps, usually <5 mm and multiple, do not progress as shown in long-term studies, but larger polyps 1 cm or greater or those with irregular margins are at risk for malignancy (30,31).

Figure 4.3 A 50-year-old woman several years postpancreaticoduodenectomy for duodenal carcinoid developed fever after hepatic artery embolization for control of hepatic metastases. Right hepatic liquified abscess (asterisk) with posterior acoustic enhancement (arrowhead). Small posterior solid metastases (small arrows) are also seen.

(A)

(B)

Figure 4.4 Colorectal metastasis to left hepatic lobe was evident on ultrasound but not by CT done the same day. (A) The lateral left lobe was considered negative on CT. (B) Longitudinal ultrasound revealed a segment II metastases with peripheral halo (calipers) consistent with malignant lesion.

(A)

(B)

Figure 4.5 Hemangioma. (A) Typical hemangioma (arrow) is uniformly echogenic with no surrounding halo. (B) An atypical hemangioma with a thin bright rim (arrow) is shown in this longitudinal view of the right hepatic lobe. Another hemangioma (arrowhead) is noted peripherally.

39

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Normal gallbladder wall thickness is <3 mm. Mural thickening may occur with adenomyomatosis, inflammation, or neoplasm. Adenomyomatosis may be focal mass (adenomyoma) or diffuse thickening and gallbladder deformity with hourglass configuration. Ring down artifact from cholesterol crystals in Aschoff–Rokitansky sinuses is a diagnostic ultrasound feature of adenomyomatosis. Acute cholecystitis causes diffuse or focal gallbladder wall thickening with a layered appearance and in severe cases, the sloughed mucosa may be seen (Fig. 4.6). Marked edema in the pericholecystic space may mimic acute cholecystitis in patients with pancreatitis or hepatic inflammation. It is essential to carefully evaluate the gallbladder wall to exclude gallbladder carcinoma that may coexist with stones (Fig. 4.7). This is particularly important in patients being considered for laparoscopic cholecystectomy since surgical management of gallbladder carcinoma usually requires hepatic resection and recurrences in laparoscopic port sites are frequent (32). Gallbladder carcinoma may cause focal thickening or may obliterate the gallbladder lumen. Associated tumor extension into hepatic segments 4 and 5 and biliary obstruction at the hilus are common. Ultrasound is sensitive for detection of biliary dilation and to determine level of obstruction. Dilated intrahepatic bile

(A)

(B)

(C) Figure 4.6 Acute cholecystitis in a 56-year-old woman with abdominal pain. (A) Longitudinal view reveals a laminated appearance to the anterior gallbladder wall (arrowheads) and gallstone (arrow) with posterior acoustic shadow. (B) Transverse view shows thickened wall at 6 mm (calipers). (C) Longitudinal color Doppler image reveals vascular flow within the gallbladder wall.

(A)

(B)

Figure 4.7 Gallbladder carcinoma. (A) Longitudinal and (B) transverse sonogram of the gallbladder reveals a stone (arrowhead) with acoustic shadowing. The anterior fundus is narrowed and surrounded by hypoechoic soft tissue that infiltrates the adjacent liver (arrows).

40

ULTRASOUND FOR HPB DISORDERS
ducts produce the double duct sign and dilation of the common bile duct >6 mm is considered abnormal. There has been controversy regarding the size of the common bile duct with increasing age and postcholecystectomy, but recent studies have shown that even in the elderly, 98% of ducts are <6 mm and there is no compensatory dilation of the common duct after cholecystectomy (33–36). Cholangiocarcinomas cause biliary obstruction in characteristic patterns. Intrahepatic cholangiocarcinoma arises from the peripheral bile ducts and bile duct obstruction peripheral to the tumor is seen in almost one third of the cases. These tumors are also typically hypovascular, in contrast to HCC. Hilar cholangiocarcinoma are typically smaller since their critical location produces early jaundice. Associated vascular encasement is evident in nearly 50% of the cases (37). Ultrasound is useful for tumor staging that is determined by location of tumor along the ducts and the extent of vascular involvement (38). the onset of pancreatitis (46). Identification of infected pseudocyst is limited, but the presence of echogenic foci corresponding to gas bubbles is suggestive of infection. If there is clinical suspicion, ultrasound can provide image guidance for fluid aspiration or drainage. Venous thrombosis and pseudoaneurysms that occur secondary to pancreatitis can also be evaluated sonographically. Ultrasound findings in chronic pancreatitis include alteration of texture, calcification, pancreatic duct, and/or bile duct dilation, and chronic pseudocyst. The gland is usually atrophic and heterogeneous. Calcification, either focal or diffuse, and pancreatic duct dilation are the most classic sonographic features (47). When findings of chronic pancreatitis mimic neoplasm with ductal dilation, CT or MRI is needed to make the distinction. Pancreatic Neoplasms Characterization of pancreatic masses, aspiration and biopsy are increasingly being done with endoscopic ultrasound (EUS). Miniature ultrasound transducers mounted on endoscopes display radial or linear images of the pancreas. EUS is more sensitive for detection of small masses and biopsy can be performed through the posterior gastric wall (48). Adenocarcinoma appears on ultrasound as a focal mass with atrophy and pancreatic duct dilation distal to the mass. Vascular invasion is frequent and bile ducts are dilated commonly for masses in the pancreatic head. Ultrasound is considered reliable for diagnosis of nonresectable tumors and in such cases further imaging is not required; evaluation can proceed directly to biopsy for tissue diagnosis (40,49) (Fig. 4.9). Staging of pancreatic adenocarcinoma by EUS and CT were compared in a prospective study by DeWitt (50). EUS had higher sensitivity than CT for tumor detection (98% vs. 86%), better staging accuracy (67% vs. 41%), and both techniques were equivalent for nodal status. EUS is also useful for biopsy especially when CT-guided biopsy is negative. In a prospective study of patients with negative CT-guided biopsy of pancreatic masses, EUS biopsy had 95% sensitivity and 100% specificity for diagnosis (51). Neuroendocrine tumors such as insulinomas and gastrinomas usually have classical symptoms. When tumors are small, abdominal ultrasound is limited, but laparoscopic ultrasound and intraoperative ultrasound are extremely useful for tumor detection (52,53). Approximately one-third of endocrine tumors are nonfunctioning and these tumors are more likely malignant. Cystic pancreatic neoplasms (serous microcystic adenomas, mucinous adenomas, and solid and cystic pseudopapillary tumors) are best evaluated with EUS. Serous microcystic adenomas are benign tumors with multiple cysts ranging in size from 1 mm to 2 cm. These tumors may appear solid on ultrasound because of numerous interfaces produced by the microscopic cyst walls (54–57). Macrocystic mucinous tumors of the pancreas are malignant or potentially malignant and have cysts >2 cm. The cysts may have thick septation, mural nodules, and calcification may be present (47,56). It is not possible to distinguish between benign and malignant mucinous tumors, but in general, larger cysts and cystic masses with

pancreas
The echotexture of the normal pancreas is uniform and slightly higher echogenicity than liver. With aging and obesity, fatty infiltration of the pancreas may further increase echogenicity. The pancreatic duct is best seen transversely and is normally less than 2 mm in the body and 3 mm in the head (39). Diffuse Pancreatic Diseases In acute pancreatitis, the pancreas may become enlarged and hypoechoic with indistinct margins from edema. The edema may involve the entire gland or only a portion, usually the head. Peripancreatic fluid is a useful diagnostic feature; fluid and vascular mural thickening may also be observed (40) (Fig. 4.8). Pancreatic duct may be dilated. In the most acute stage, ileus limits ultrasound visualization and CT is more useful, but ultrasound has a role to exclude biliary calculi as an etiology for the pancreatitis (41–44). Severe inflammation progresses to inflammatory pancreatic mass or phlegmon with fluid collection, hemorrhage, and necrosis. Fluid is commonly seen within the lesser sac, anterior pararenal spaces, transverse mesocolon, small bowel mesentery, and parapancreatic spaces (45). Pseudocysts may persist for a minimum of 4 weeks after

Figure 4.8 Acute pancreatitis in a patient with AIDS. Transverse sonogram of the pancreas reveals heterogeneous pancreatic parenchyma and edema of the splenic vein (arrow). The vein has a layered appearance with a ring of hypoechoic fluid within the wall of the vessel. Fluid also is seen in the peripancreatic space (arrowheads).

41

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C) Figure 4.9 Unresectable pancreatic adenocarcinoma. (A) Longitudinal ultrasound image of the pancreas shows an enlarged pancreatic head (m) and dilated common bile duct (arrow), PV = portal vein, IVC = inferior vena cava. (B) Transverse sonogram reveals the pancreatic head mass (m) and dilated pancreatic duct (arrows) anterior to the splenic vein (SV). IVC = inferior vena cava, a = aorta. (C) Transverse sonogram reveals a left hepatic metastasis (arrows). R = right hepatic vein, M = middle hepatic vein, IVC = inferior vena cava.

significant solid component are more likely to be malignant (47,55,58). Aspirates of cystic lesions are relatively acellular but fluid analysis for tumor markers is useful. Brugge et al. (59) reported that elevated cyst fluid CEA level had 79% accuracy for diagnosis of mucinous tumors.

intraoperative ultrasound
Intraoperative ultrasound (IOUS) is an important tool for (1) assessment of tumors at the time of resection, (2) vascular mapping during hepatic resection or live split liver donor transplantation, and (3) guidance during intraoperative tumor ablation or biopsy (60–64) (Fig. 4.10). During hepatic resection, IOUS is used to characterize liver lesions that are indeterminate or occult on preoperative imaging. IOUS can accurately assess tumor extent relative to vascular structures and bile ducts (65–68); this is important since approximately 1 to 2 cm margin should be available between the tumor and vessels for optimal surgical outcome and vessel encasement or thrombosis may alter surgical approach (69–72). A prospective study by Cerwenka et al. (73) evaluated the role of IOUS in patients who had partial hepatectomy after standardized hepatic protocol preoperative MRI. Small additional lesions with mean size of 1.5 cm were found by IOUS in 7% of patients and in 5% of patients IOUS findings altered surgical strategy (73,74). IOUS altered management in 20% of patients who had resection for primary or secondary hepatic malignancies. Even with recent improvements in crosssectional imaging, there was no significant difference in resection

rate (72%), detection of unrecognized additional tumors (20% vs. 14% p=0.70), or detection of vascular involvement when groups in the years1999 to 2003 and 2003 to 2005 were compared (68,74–76). Special dedicated high-frequency (5–10 MHz) transducers are required for IOUS; these transducers can be applied directly on the area of interest to improve resolution compared to transabdominal ultrasound, which is limited by distance and abdominal wall artifact. Typically IOUS probes are small T-shaped linear and hockey-stick-shaped probes, which are easy to manipulate within restricted operative fields. Transducer specifications should include good near field resolution and Doppler capabilities. It is preferable to have the scanner connected to the hospital network to provide (1) access for consultation at remote sites during real-time scanning and (2) permanent image archiving in the electronic record. Review of preoperative imaging is essential before IOUS since preoperative planning increases the efficiency of the procedure. While performing IOUS, the surgeon should avoid applying excessive pressure. If vessels become compressed, it is difficult to assess patency or encasement. Light touch with the transducer can be used as a palpation method to differentiate between soft benign lesions such as hemangiomas, focal fat, and fat sparing versus malignant lesions, which are usually firm (77,78). IOUS may be limited for lesions in the high right lobe or in the posterior subdiaphragmatic location where access is

42

ULTRASOUND FOR HPB DISORDERS

(A)

(B)

(C) Figure 4.10 Intraoperative ultrasound reveals additional hepatic lesions. (A) A 2 cm segment 7 liver lesion (arrows) with peripheral halo and (B) an 8 mm segment 6 lesion (arrow) were seen on preoperative imaging. (C) A nonpalpable 6 mm lesion (arrow) in segment 4A was not evident on preoperative imaging. Lesions were resected with diagnosis of metastatic neuroendocrine carcinoma; primary site later identified in the pancreas. (Complements of Robert A. Kane, M.D., Professor of Radiology, Harvard Medical School, Chief, Body and Abdominal Ultrasound Imaging, Beth Israel Deaconess Medical Center, Boston, MA.)

difficult. In that situation, scanning from the opposite surface of the liver may improve visualization. Artifacts in the near field of the image may also obscure lesions near the hepatic surface. If this occurs, the surgeon can immerse the liver in a sterile saline bath, thereby changing the focus zone to better visualize the superficial anatomy. Another difficulty may be encountered when attempting to visualize lesions near a surgical margin. Echogenic foci from air bubbles in the parenchyma after cauterization or radiofrequency ablation may mimic echogenic mucin-containing colorectal metastases. This pitfall can be mitigated by imaging before intervention to accurately determine number, size, and location of lesions (79–81). New advances in intraoperative and interventional ultrasound techniques now allow fusion of ultrasound, CT, and MRI images and electromagnetic tracking to more precisely localize lesions for biopsy and thermal ablation procedures. For example, after initial CT data is entered, information from electromagnetic sensors is applied onto the needle device and the patient can guide the needle track in real time even when the needle is out of the ultrasound imaging plane. This process brings two data sets into spatial alignment. Such techniques have shown improved needle tracking for

interventional procedures and better three-dimensional visualization of tumor and treatment zone during radiofrequency ablation (63,82–84).

references
1. Cosgrove D. Ultrasound contrast agents: an overview. Eur J Radiol 2006; 60(): 324–30. 2. Wilson SR, Burns PN, Muradali D,et al. Harmonic hepatic US with microbubble contrast agent: initial experience showing improved characterization of hemangioma, hepatocellular carcinoma, and metastasis. Radiology 2000;215(1):153–61. 3. Albrecht T, Blomley MJ, Burns PN, et al. Improved detection of hepatic metastases with pulse-inversion US during the liver-specific phase of SHU 508A: multicenter study. Radiology. 2003; 227(2): 361–70. 4. Ascenti G, Mazziotti S, Zimbaro G, et al. Complex cystic renal masses: characterization with contrast-enhanced US. Radiology 2007; 243(1): 158–65. 5. Burns PN, Wilson SR. Focal liver masses: enhancement patterns on contrast-enhanced images--concordance of US scans with CT scans and MR images. Radiology 2007; 242(1): 162–74. 6. Correas JM, Claudon M, Tranquart, et al. The kidney: imaging with microbubble contrast agents. Ultrasound Q 2006; 22(1): 53–66. 7. Kabakci N, Igci E, Secil M, et al. Echo contrast-enhanced power Doppler ultrasonography for assessment of angiogenesis in renal cell carcinoma. J Ultrasound Med 2005; 24(6): 747–53.

43

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
8. Couinaud C. Le Foi: Etudes Anatomogiques et Chirurgicales. Paris: Masson, 1957. 9. Lafortune M, Madore F, Patriquin H, et al. G. Segmental anatomy of the liver: a sonographic approach to the Couinaud nomenclature. Radiology 1991; 181(2): 443–8. 10. Caballeria L, Auladell MA, Toran P, et al. Risk factors associated with nonalcoholic fatty liver disease in subjects from primary care units. A casecontrol study. BMC Gastroenterol 2008; 8: 44. 11. Kooby DA, Fong Y, Suriawinata A, et al. Impact of steatosis on perioperative outcome following hepatic resection. J Gastrointest Surg 2003; 7(8): 1034–44. 12. Fahey BJ, Nightingale KR, Nelson RC, et al. Acoustic radiation force impulse imaging of the abdomen: demonstration of feasibility and utility. Ultrasound Med Biol 2005; 31(9): 1185–98. 13. Kim DY, Kim SU, Ahn SH, et al. Usefulness of FibroScan for detection of early compensated liver cirrhosis in chronic hepatitis B. Dig Dis Sci 2008; 54(8): 1758–63. 14. Yoneda M, Mawatari H, Fujita K, et al. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with nonalcoholic fatty liver disease (NAFLD). Dig Liver Dis 2008; 40(5): 371–8. 15. Marn CS, Bree RL, Silver TM. Ultrasonography of liver. Technique and focal and diffuse disease. Radiol Clin North Am 1991; 29(6): 1151–70. 16. Middleton W. There are nonhyperechoic lesions in the liver: What does that mean? In: RSNA, editor. Categorical Course in Diagnostic Radiology: Findings at US—What do they mean? Oak Brook, IL: Radiological Society of North America, 2002. pp. 79–90. 17. Saini S, Mueller PR, Ferrucci JT, Jr.,et al. RJ. Percutaneous aspiration of hepatic cysts does not provide definitive therapy. AJR Am J Roentgenol 1983; 141(3): 559–60. 18. Mortele KJ, Ros PR. Cystic focal liver lesions in the adult: differential CT and MR imaging features. Radiographics 2001; 21: 895–910. 19. Czermak BV, Unsinn KM, Gotwald T, et al. Echinococcus multilocularis revisited. AJR Am J Roentgenol 2001; 176(5): 1207–12. 20. Pedrosa I, Saiz A, Arrazola J, et al. . Hydatid disease: radiologic and pathologic features and complications. Radiographics 2000; 20(3): 795–817. 21. Charboneau J. There is a hyperechoic mass in the liver. What does that mean? Categorical Course in Diagnostic Radiology: Findings at US— What Do They mean? RSNA; 2002: 73–8. 22. Liefer KM, Koster MI, Wang XJ, et al. Down-regulation of p63 is required for epidermal UV-B-induced apoptosis. Cancer Res 2000; 60(15): 4016–20. 23. Caturelli E, Pompili M, Bartolucci F, et al. Hemangioma-like lesions in chronic liver disease: diagnostic evaluation in patients. Radiology 2001; 220(2): 337–42. 24. Quinn SF, Gosink BB. Characteristic sonographic signs of hepatic fatty infiltration. AJR Am J Roentgenol 1985; 145(4): 753–5. 25. Ralls PW, Mack LA. Spectral and color Doppler sonography. Semin Ultrasound CT MR 1992; 13(5): 355–66. 26. Kruskal JB, Thomas P, Nasser I, et al. Hepatic colon cancer metastases in mice: dynamic in vivo correlation with hypoechoic rims visible at US. Radiology 2000; 215(3): 852–7. 27. Wernecke K, Vassallo P, Bick U, et al. The distinction between benign and malignant liver tumors on sonography: value of a hypoechoic halo. AJR Am J Roentgenol 1992; 159(5): 1005–9. 28. Reinhold C, Hammers L, Taylor CR, et al. Characterization of focal hepatic lesions with duplex sonography: findings in 198 patients. AJR Am J Roentgenol 1995; 164(5): 1131–5. 29. Wilson SR, Withers CE. The liver. In: Rumack CM, Wilson SR, Charboneau JW, Johnson JM, eds. Diagnostic Ultrasound. St. Louis: Elsevier Mosby, 2005: 77. 30. Levy AD, Murakata LA, Abbott RM, et al. From the archives of the AFIP. Benign tumors and tumorlike lesions of the gallbladder and extrahepatic bile ducts: radiologic-pathologic correlation. Armed Forces Institute of Pathology. Radiographics 2002; 22(2): 387–413. 31. Terzi C, Sokmen S, Seckin S, et al. Polypoid lesions of the gallbladder: report of 100 cases with special reference to operative indications. Surgery 2000; 127(6): 622–7. 32. Winston CB, Chen JW, Fong Y, et al. Recurrent gallbladder carcinoma along laparoscopic cholecystectomy port tracks: CT demonstration. Radiology 1999; 212(2): 439–44. 33. Perret RS, Sloop GD, Borne JA. Common bile duct measurements in an elderly population. J Ultrasound Med 2000; 19(11): 727–30; quiz 31. 34. Wilkinson ML. Are dilating bile ducts a cause for concern? Gut 1999; 45(5): 637–8. 35. Skalicky M, Dajcman D, Hojs R. Effect of cholecystectomy for gallstones on the surface of the papilla of Vater and the diamter of the common bile duct. Eur J Gastroenterol Hepatol 2002; 14(4): 399–404. 36. Majeed AW, Ross B, Johnson AG. The preoperatively normal bile duct does not dilate after cholecystectomy: results of a five year study. Gut 1999; 45(5): 741–3. 37. Are C, Gonen M, D’Angelica M, et al. Differential diagnosis of proximal biliary obstruction. Surgery 2006; 140(5): 756–63. 38. Jarnagin WR, Fong Y, DeMatteo RP, et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma. Ann Surg. 2001; 234(4): 507–17; discussion 17–19. 39. Hadidi A. Pancreatic duct diameter: sonographic measurement in normal subjects. J Clin Ultrasound 1983; 11: 17–22. 40. Ralls PW, Wren SM, Radin R, et al. Color flow sonography in evaluating the resectability of periampullary and pancreatic tumors. J Ultrasound Med 1997; 16(2): 131–40. 41. Balthazar EJ, Freeny PC, vanSonnenberg E. Imaging and intervention in acute pancreatitis. Radiology 1994; 193(2): 297–306. 42. Balthazar EJ, Ranson JH, Naidich DP, et al. Acute pancreatitis: prognostic value of CT. Radiology 1985; 156(3): 767–72. 43. Balthazar EJ, Robinson DL, Megibow AJ, et al. Acute pancreatitis: value of CT in establishing prognosis. Radiology 1990; 174(2): 331–6. 44. Clavien P, Hauser H, Meyer P, et al. Value of contrast-enhanced computerized tomography in the early diagnosis and prognosis of acute pancreatitis. Am J Surg 1988; 155: 457–66. 45. Jeffrey RB, Jr, Laing FC, Wing VW. Extrapancreatic spread of acute pancreatitis: new observations with real-time US. Radiology 1986; 159(3): 707–11. 46. Donovan PJ, Sanders RC, Siegelman SS. Collections of fluid after pancreatitis: evaluation by computed tomography and ultrasonography. Radiol Clin North Am 1982; 20(4): 653–65. 47. Atri M, Finnegan PW. The pancreas. Ultrasound. St. Louis, MO: Mosby, 1998: 241–56. 48. Catalano MF, Lahoti S, Geenen JE, et al. Prospective evaluation of endoscopic ultrasonography, endoscopic retrograde pancreatography, and secretin test in the diagnosis of chronic pancreatitis. Gastrointest Endosc 1998; 48(1): 11–17. 49. Yassa NA, Yang J, Stein S, et al. Gray-scale and color flow sonography of pancreatic ductal adenocarcinoma. J Clin Ultrasound 1997; 25(9): 473–80. 50. DeWitt J, Devereaux B, Chriswell M, et al. Comparison of endoscopic ultrasonography and multidetector computed tomography for detecting and staging pancreatic cancer. Ann Intern Med 2004; 141(10): 753–63. 51. Gress F, Gottlieb K, Sherman S, et al. Endoscopic ultrasonography— Guided fine-needle aspiration biopsy of suspected pancreatic cancer. Ann Int Med 2001; 134(6): 459–64. 52. Fendrich V, Bartsch DK, Langer P, et al. Diagnosis and surgical treatment of insulinoma—experiences in 40 cases. Dtsch Med Wochenschr 2004; 129(17): 941–6. 53. Gorman B, Charboneau JW. Benign pancreatic insulinoma: preoperative and intraoperative sonographics localization. AJR Am J Roentgenol 2003; 181: 787–92. 54. Buck JL, Hayes WS. From the archives of the AFIP. Microcystic adenoma of the pancreas. Radiographics 1990; 10(2): 313–22. 55. Fugazzola C, Procacci C, Bergamo Andreis IA, et al. Cystic tumors of the pancreas: evaluation by ultrasonography and computed tomography. Gastrointest Radiol 1991; 16(1): 53–61. 56. Johnson CD, Stephens DH, Charboneau JW, et al. . Cystic pancreatic tumors: CT and sonographic assessment. AJR Am J Roentgenol 1988; 151: 1133–8. 57. Friedman AC, Lichenstein JE, Dachman AH Cystic neoplasms of the pancreas. Radiology 1983; 149: 45–50. 58. Allen PJ, D’Angelica M, Gonen M, et al. A selective approach to the resection of cystic lesions of the pancreas: results from 539 consecutive patients. Ann Surg 2006; 244(4): 572–82.

44

ULTRASOUND FOR HPB DISORDERS
59. Brugge WR, Lewandrowski K, Lee-Lewandrowski E, et al. Diagnosis of pancreatic cystic neoplasms: a report of the cooperative pancreatic cyst study. Gastroenterology 2004; 126(5): 1330–6. 60. Torzilli G, Makuuchi M. Intraoperative ultrasonography in liver cancer. Surg Oncol Clin N Am 2003; 12(1): 91–103. 61. Torzilli G, Montorsi M, Donadon M, et al. “Radical but conservative” is the main goal for ultrasonography-guided liver resection: prospective validation of this approach. J Am Coll Surg 2005; 201(4): 517–28. 62. Berber E, Garland AM, Engle KL, et al. Laparoscopic ultrasonography and biopsy of hepatic tumors in 310 patients. Am J Surg 2004; 187(2): 213–8. 63. Kleemann M, Hildebrand P, Birth M, et al. Laparoscopic ultrasound navigation in liver surgery: technical aspects and accuracy. Surg Endosc 2006; 20(5): 726–9. 64. Zacherl J, Scheuba C, Imhof M, et al. Current value of intraoperative sonography during surgery for hepatic neoplasms. World J Surg 2002; 26(5): 550–4. 65. Jakimowicz JJ. Intraoperative ultrasonography in open and laparoscopic abdominal surgery: an overview. Surg Endosc 2006; 20 Suppl 2: S425–35. 66. Kane RA. Intraoperative ultrasonography: history, current state of the art, and future directions. J Ultrasound Med 2004; 23(11): 1407–20. 67. Minagawa M, Makuuchi M, Kubota K, et al. Intraoperative three-dimensional visualization of liver vasculature by ultrasonography. Hepatogastroenterology 2004; 51(59): 1448–50. 68. Sahani DV, Kalva SP, Tanabe KK, et al. Intraoperative US in patients undergoing surgery for liver neoplasms: comparison with MR imaging. Radiology 2004; 232(3): 810–4. 69. Santambrogio R, Opocher E, Ceretti AP, et al. Impact of intraoperative ultrasonography in laparoscopic liver surgery. Surg Endosc 2007; 21(2): 181–8. 70. Shukla PJ, Pandey D, Rao PP, et al. Impact of intra-operative ultrasonography in liver surgery. Indian J Gastroenterol 2005; 24(2): 62–5. 71. Silberhumer GR, Steininger R, Laengle F, et al. Intraoperative ultrasonography in patients who undergo liver resection or transplantation for hepatocellular carcinoma. Surg Technol Int 2004; 12: 145–51. 72. Thaler K, Kanneganti S, Khajanchee Y, et al. The evolving role of staging laparoscopy in the treatment of colorectal hepatic metastasis. Arch Surg 2005; 140(8): 727–34. 73. Cerwenka H. Intraoperative ultrasonography during planned liver resections remains an important surgical tool. Surg Endosc 2008; 22(4): 1137–8. 74. Ellsmere J, Kane R, Grinbaum R, et al.. Intraoperative ultrasonography during planned liver resections: why are we still performing it? Surg Endosc 2007; 21(8): 1280–3. 75. Long EE, Van Dam J, Weinstein S, et al. Computed tomography, endoscopic, laparoscopic, and intra-operative sonography for assessing resectability of pancreatic cancer. Surg Oncol 2005; 14(2): 105–13. 76. Ravi K, Britton BJ. Surgical approach to insulinomas: are pre-operative localization tests necessary? Ann R Coll Surg Engl 2007; 89(3): 212–7. 77. Kruskal JB, Kane RA. Intraoperative US of the liver: techniques and clinical applications. Radiographics 2006; 26(4): 1067–84. 78. Machi J, Oishi AJ, Furumoto NL, et al.. Intraoperative ultrasound. Surg Clin North Am 2004; 84(4): 1085–111. 79. Solbiati L, Ierace T, Tonolini M, et al. Guidance and monitoring of radiofrequency liver tumor ablation with contrast-enhanced ultrasound. Eur J Radiol 2004; 51 Suppl: S19–23. 80. Guimaraes CM, Correia MM, Baldisserotto M, et al. Intraoperative ultrasonography of the liver in patients with abdominal tumors: a new approach. J Ultrasound Med 2004; 23(12): 1549–55. 81. Fan RF, Chai FL, He GX, et al. Laparoscopic radiofrequency ablation of hepatic cavernous hemangioma. A preliminary experience with 27 patients. Surg Endosc 2006; 20(2): 281–5. 82. Wood BJ, Zhang H, Durrani A, et al. Navigation with electromagnetic tracking for interventional radiology procedures: a feasibility study. J Vasc Interv Radiol 2005; 16(4): 493–505. 83. Wood BJ, Locklin JK, Viswanathan A, et al. Technologies for guidance of radiofrequency ablation in the multimodality interventional suite of the future. J Vasc Interv Radiol 2007; 18(1 Pt 1): 9–24. 84. Krucker J, Xu S, Glossop N, et al. Electromagnetic tracking for thermal ablation and biopsy guidance: clinical evaluation of spatial accuracy. J Vasc Interv Radiol 2007; 18(9): 1141–50.

45

5

Liver surgery in elderly patients Gerardo Sarno and Graeme J. Poston
of arbitrary definition of “elderly,” in liver surgery the common practice is to identify as elderly a patient older than 70 years (6,13–18). This is due to the evidence of a rapid decrease of liver mass and portal blood flow from 70 years onward (17), which may affect liver function. A limited life expectancy in the elderly might argue against extending the indications for hepatectomy in these patients. However, life expectancy for people aged between 80 and 85 years is still 8 years, and 6 years for those over 85 years old. Moreover, the risk of cancer-related death diminishes with increasing age; it is estimated to be 40% for those aged between 50 and 70 years, falling to 10% for those over 90 years old (19). Recently, several studies reported comparable early and long-term results between young and aged patients undergoing liver resection. These studies highlighted that an age limit does not exist to contraindicate liver resection. After a careful evaluation of the operative risk, a similar chance of long-term survival can be offered also to well-selected elderly patients.

introduction
The recent increase in the geriatric population in society and increased life span have raised the expectation from surgeons to expand their operative indications to include geriatric patients. In liver surgery, the indications for hepatectomy have been expanded to include patients aged 70 and older, and several studies have demonstrated acceptable long-term survival of elderly patients after such surgery (1,2). In 1937, Brooks reported the results of surgery for 287 patients aged 70 and older. The operative mortality rate was high (19%), and onethird of patients who had abdominal operations died in hospital. Nevertheless, the author emphasized that with the rapid growth of the elderly population, and prolonged life expectancy, surgeons will increasingly be confronted with surgical problems among the elderly and must therefore strive to improve their results by studying physiologic processes in the aged (2,3). In the seven decades since Brooks’ paper, advancements in anesthesiology and intensive care, an increased knowledge of liver physiology, surgical hepatic anatomy, and resection techniques have encouraged hepatic resection in elderly patients, achieving improved surgical outcomes. Because of the high prevalence of liver cancers and aging of the world population, the elderly population considered for liver resection has increased (4,5). Also an effective multidisciplinary approach and better selection of elderly patients leads to reduced age-related perioperative morbidity and mortality (6). Moreover, the definition of “elderly patients” has been better defined so avoiding unnecessary confusion that has generated over the past years. Although advances in minimally invasive ablative techniques have increased the treatment options for patients with malignant hepatobiliary disease, liver resection remains the only treatment demonstrated to offer long-term survival (7–9). Also the past three decades have seen a dramatic decline in the mortality rate after liver resection in selected elderly patients, which is less than 5% in tertiary cancer care referral centers (2,6,10). Colorectal cancer has become a major public health problem that increasingly affects older people (11), and because the liver is the most common site of metastases, the number of elderly so affected is increasing (12). Liver resection is also successfully performed in aged patient suffering primary malignancies such as hepatocellular carcinoma (HCC) with results comparable to those seen in younger people. In this chapter, we highlight the main advances performed in liver surgery, taking into account all the issues that are still a matter of debate for elderly patients with primary or metastatic liver disease.

age-related liver changes
The effects of aging on the human liver have not been clearly determined (20). In general, aging is characterized by a progressive decline of cellular functions and also the liver undergoes physiological changes. Although some recent studies have shown that aging itself does not affect liver function, the amount of hepatic tissue that can be safely removed, and the consequent capacity of liver regeneration are often difficult to be precisely assessed (14). Aging has been shown to be associated with multiple changes in hepatic function, however the clinically relevant biochemical parameters of liver function remain generally normal in the elderly. Thus abnormalities of these parameters should be evaluated for the presence of liver disease (21). As a matter of fact, the liver function seems to be quite well maintained in old age, but numerous age-related changes in hepatic structure have been described (22). However, there have been few comprehensive studies of liver morphology during aging, and most of these have been performed in rodents (20). The most frequently cited morphological change in the human liver is a decrease in size. In elderly men, liver weight declines by about 6.5% and in women it decreases by 14.3% (23), which may be attributable to decreased hepatic blood flow (24–26). The decrease in blood flow is about 45% in subjects over 75 years when compared to those under 40 years (26,27). The classic gross appearance of the liver in older persons is known as “brown atrophy.” The brown color is due to accumulation of lipofuscin (ceroid) within hepatocyte and also associated with major degree of steatosis (21). Alteration in the hepatocyte morphology has been also described (28). It has been reported that the liver in elderly humans has histologically fewer, but larger hepatocytes (29).

definition of “elderly patients” The age at which persons become “elderly” depends on social, environmental, and individual factors. Nowadays, after years 46

LIVER SURGERY IN ELDERLY PATIENTS
In addition, hepatic clearance of many drugs is reduced in elderly persons (20,22). Traditional theories have attempted to attribute this observation to age-related reduction in liver mass and blood flow (24–26). More recently, it has been considered attributable to age-related changes in the sinusoidal endothelium and space of Disse, which may restrict the availability of oxygen and other substrates (30). Several other mechanisms have been described, among them the impaired enzymatic activities (31,32) due to oxidative protein damage sustained by free radicals (21). The rate of hepatic steatosis allowing safe liver surgery is not yet clearly defined, although a moderate to severe steatosis (involving more than 30% of the hepatocytes) seems to affect both postoperative morbidity and mortality (33,34). However, although it is impossible to exactly predict this feature before surgery without a liver biopsy specimen, this diagnostic tool should be considered when the presence of steatosis is suggested by imaging and a major resection is planned (14). All of these factors may reduce the functional reserve of the organ and therefore predisposing to postoperative liver failure (35). Thus in preoperative risk estimation prior to hepatic resection, it may be important to take into account the effect of aging upon liver function and structure, in addition to carrying out a qualitative and quantitative evaluation of liver parenchyma. The most frequently reported causes of death in elderly patients with no underlying liver disease undergoing liver resection are hepatic insufficiency, myocardial infarction, pneumonia, and gastrointestinal bleeding (4,42,43). The evaluation of associated medical disease has been widely investigated focusing in particular on American Society of Anesthesiology (ASA) scores. Advanced ASA grading is known as one of the most reliable predictors of postoperative complications and mortality (4). ASA scores measure major comorbid diseases easily and with minimal expense and are able to predict outcomes after major surgical procedures (4,44). Some authors have considered an ASA score higher than II (i.e., a patient with mild to moderate systemic disease) as a contraindication for surgery for HCC or for major hepatectomies (42,45). In such patients procedures other than surgery (radiofrequency ablation or transarterial chemoembolization) could be considered (16). The exact determination of the ASA score is highly operatordependent and the reported experiences of postoperative deaths for causes unrelated to surgery (i.e., myocardial infarction) (42,46) in subjects with an unremarkable history of cardiac or pulmonary disease suggest that this score should be applied more selectively during the evaluation of elderly patients with underlying liver disease (14). Also in the elderly, the performance status, especially if they were physically active before surgery, has to be taken into account since a significant lower risk of postoperative complications have been recorded (47). Finally, the morphologic characteristics of the underlying liver pathology and number and size of malignancies have to be carefully evaluated prior to performing hepatectomy, aiming to avoid overextensive resections and to minimize intraoperative haemorrhage (6). One other factor having deleterious effects on early and late outcome is intraoperative blood loss. It is well known that hemorrhage and the need for transfusion are closely associated with worse prognoses (48), and this may be even worse in an aged liver.

evaluation of the surgical risk
The stress of liver resection may not be well-tolerated in the elderly (4). Liver surgery is not without complication and, before considering liver resection in elderly patients, the increased risks and costs of such surgery must be balanced against the potential improvement of life expectancy. Elderly patients are more likely to have decreased life expectancy with comorbidity, so the decision to perform major hepatectomy has to be carefully balanced against the likelihood of benefit before undertaking such resections. However, most studies record small numbers of cases or have not distinguished between major and minor resections, making interpretation of results difficult (32). Factors other than age should be considered in evaluating surgical risk in the elderly. It is well known that in elderly patients, a preoperative decline in cardiac and pulmonary functions, also combined with cerebrovascular disease can be frequently seen (2,36). To achieve better results in the elderly population, proper patient selection in terms of liver functional reserve and comorbidities conditions is mandatory. This necessitates a close collaboration between surgeons, anesthesiologists, cardiologists, pulmonary physicians, and geriatric physicians (6). A clear preoperative selection process should be undertaken to minimize perioperative risks (37). The majority of elderly may suffer from more than one comorbid disease or for many reasons do not have a good performance status. Cardiovascular and pulmonary disease have a prevalence among the elderly of 20% to 27% and 14%, respectively (38). Moreover, cardiovascular disease and diabetes mellitus were reported to be significant risk factors especially when associated with cirrhosis (39–41).

colorectal liver metastases
Colorectal cancer is a major public health problem. In western society, it is expected to increase in incidence by over 30% over the next 20 years because of ever-growing elderly (>70 years of age) population (49,50). The liver is the most common site of metastases and is involved approximately in half of patients (12). By the time of initial diagnosis of colorectal cancer, nearly a quarter of patients will have clinically detectable liver metastases (CRLM), despite increasing patient and clinician awareness of the disease. Of those who undergo apparently successful resection of the primary tumor, nearly half will develop liver metastases, usually within the first three years after colectomy (49,51,52). Currently, over half of all cancers are diagnosed in elderly patients, and 76% of all colorectal cancer patients are diagnosed between 65 and 85 years old (53,54). Encouraging results of surgery for CRLM in the elderly have been reported with 5-year survival rates between 21% and 44% (2,4,55–58). In elderly patients, liver resection for CRLM provides, as with younger patients, the only chance of cure, compared with untreated patients who have a median survival of 4.5 to

47

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
6.5 months (59,60), or patients treated by chemotherapy alone who have a median survival of 9.2 to 16.5 months (60,61). The only true contraindication for liver resection is the technical nonfeasibility of hepatectomy, independent of the presence of other poor prognostic factors (8,62). In 2005, the reported percentage of patients over 70 years of age undergoing liver resection for CRLM was 26.5%, which was dramatically higher when compared to 6% in the early 1990s (6). This improvement is mostly related to developments in liver surgery since resection for CRLM can be performed with a mortality rate below 5% (6,13), with 5-year survival rates ranging from 28% to 39% (4,7,63,64). Better results can now also be achieved because of the extensive use of chemotherapy in the elderly. Elderly patients can receive protocols similar to younger ones (65). In general, since the introduction of oxaliplatin into chemotherapy regimens, a prolonged survival and a delay of progression of disease has been reported (13,66,67). The main issue of the use of oxaliplatin is hepatotoxicity (sinusoidal congestion and thrombosis), which could also prove to be a problem, especially in case of impaired liver function (13,68). However, no significant postoperative complications have been reported in elderly patients who did or did not receive chemotherapeutic treatment (13). Further evidence for offering hepatic resection to well-selected older patients is the evidence of similar benefit provided by repeat hepatectomy to elderly and younger patients (6). Liver failure is a worrying but thankfully rare complication after liver resection. Some authors have found elderly patients to be more at risk of developing this complication than younger ones, resulting in a more conservative surgical selection policy (69,70). Severe postoperative liver dysfunction may be present in fewer than 10% of elderly patients who have undergone major liver resection for malignancy (71). Postoperative liver failure due to large resections or sepsis is the most frequent cause of death (71). In general, liver resection should be avoided in the presence of bilobar or need for extended resections, especially when associated to concomitant extrahepatic disease and in medically compromised patients. In those cases, the indication for surgery should therefore be very carefully considered only in selected cases (6,72). In view of these findings, it is advisable to consider limited resection whenever possible from the oncologic perspective rather than extended surgery. The existing surgical literature on surgery for CRLM in elderly patients (2,4,55–58) should be interpreted with caution because of the small patient numbers treated at single centers. Often these series describe less than 50 patients. Only recently a large cohort study, collecting data from more than 100 centers, has been published (6). This study highlighted the evidence that hepatic resection for CRLM can be performed safely in elderly patients provided they are fit for such a procedure. The difference in survival between elderly and younger patients could in part be explained by the more limited survival expectancy of the elderly population, also reflecting the higher prevalence of comorbidity. The incidence of HCC is the fourth highest among all tumors (18), the number of patients affected has been increasing (73), and the age for detection of HCC is increasing in both men and women (17). Clarification of the optimal treatment strategy for extremely elderly patients with HCC has thus become an urgent necessity. Management of HCC with other modalities, such as percutaneous ethanol injection therapy (74), microwave therapy (75), and percutaneous radiofrequency ablation (RFA), may be an acceptable alternative to hepatic resection in the elderly, but the best treatment for patients in this age group remains controversial. Liver transplantation is theoretically the optimal treatment for HCC because it is the only method of treating both the tumor and the underlying liver cirrhosis. Replacement of the diseased liver is not only the best oncological treatment, but also the best method for preventing the development of new tumors and avoiding the life-threatening complications of cirrhosis. In patients with HCC and cirrhosis, transplantation based on the Milan criteria achieves a better outcome than hepatic resection with respect to both survival and disease recurrence (76–79). However, the limited availability of donor organs makes liver transplantation problematic (80,81) and as a consequence patients older than 70 years are excluded from transplantation programs (82). With advances in surgical treatment for HCC, hepatectomy for elderly HCC patients has become safer. There have been many reports of hepatectomies for elderly HCC patients (83–85). But because of the unclear data on long-term survival after local ablation of HCC, especially for large tumors, liver resection remains the preferred treatment, with 5-year survival rates ranging from 40% to 50% (86,87). Resection is considered to be a reasonable first-line treatment for patients with small tumors and underlying chronic liver disease, which may offer potential cure (80). Recent studies have shown the safety and feasibility of hepatectomy for HCC patients older than 70 years of age (88,89). It has been demonstrated that long-term outcome after resection of HCC is similar in older and younger patients (1,10,83,85). No operative mortality has been reported in a series of carefully selected octogenarians who underwent liver resection for HCC (45). Recent studies have identified some differences in the clinical pathological features of HCC between elderly and younger patients. Risk factors for HCC seem to be different in elderly people. A significant lower positive rate for HBsAg has been described among the elderly (88,90). Most HBV-related HCCs develop in patients in their early fifties. This may be the reason why there are few elderly HCC patients with HBV infection. On the other hand, HCV infection constitutes a major part of the etiology in elderly patients with HCC (17,91). Factors other than viral hepatitis infection, such as alcohol or genetic mutations, may contribute to the development of HCC in some elderly patients (88). Several studies have shown that elderly patients with HCC had good liver function and that only a small percentage of elderly patients with HCC had liver cirrhosis (90,92). It is possible that a large proportion of patients with cirrhosis and HCC die before reaching the age of 70 years, and those who survive have well-preserved hepatic function (93).

hepatocellular carcinoma
Primary tumors of the liver are among the most common solid tumours worldwide (4).

48

LIVER SURGERY IN ELDERLY PATIENTS
Some studies have demonstrated a close relation between HCC and alcohol abuse, that is, individuals who abuse alcohol have a significantly higher relative risk of developing HCC than those who do not. Although data about the role of alcohol in the development of HCC are inconsistent, the mechanisms that have been proposed include the induction of tumorigenesis secondary to alcoholic cirrhosis, a direct tumorinitiating and promoting effect of ethanol through induction of various enzymes, alterations of DNA repair, dietary deficiencies, immune suppression, and depletion of hepatic antitumor factors (17). There is no general agreement about the relation between alcohol abuse and postoperative recurrence of HCC or survival, but there have been a few reports of an interaction between alcohol abuse and postoperative recurrence (94). The mechanism by which alcohol abuse is related to HCC recurrence and a lower survival rate remains to be elucidated. However, at least two possible reasons can be suggested for the higher postoperative recurrence rate in patients with alcohol abuse. First, these patients may be more susceptible to developing new primary tumors after hepatectomy because chronic alcohol abuse enhances hepatocarcinogenesis. Second, they might have a higher incidence of unrecognized intrahepatic metastases at the time of initial hepatectomy because chronic alcohol abuse seems to be related to the aggressiveness of HCC, including the rate of metastasis (17). Heavy alcohol abuse and HCV infection are two leading causes of cirrhosis (91). Preoperative severe liver dysfunction carried a high risk for postoperative hepatic failure, and cirrhosis is associated with increased postoperative mortality in general (71). Advanced age is still related to poor early outcome (42), with operative mortality rates of up to 42%, attributable to liver failure in patients with cirrhosis (95). The significance of AFP has still not been well-defined. Some authors found a lower frequency of raised AFP level, compared with younger patients (17). Various other predictors have been reported to be risk factors for poor prognosis of postoperative HCC patients, such as liver cirrhosis, Child–Pugh grading, tumor size, satellite nodules, and vascular invasion (18). Some authors found a significantly higher frequency of tumor encapsulation in elderly HCC patients when comparing the histological characteristics of the resected tumors. Tumor encapsulation has been reported as a favorable prognostic factor for HCC (96). Also a higher frequency of tumor encapsulation might be an indicator for less malignant degree of the elderly patients with HCC (18). Tumor diameter should not be considered a prognostic factor. Patients over 70 years of age with large tumors should be scheduled for surgery with expected favorable results (85). For the elderly patients with HCC, predictors of postoperative survival are not well known. So far, only a few papers revealed differing findings by multivariate analysis. Hanazaki et al. (83) reported that liver cirrhosis and vascular invasion were independent prognostic factors for the survival of postresectional elderly HCC patients. Zhou et al. (97) found that Child–Pugh grading, portal vein tumor thrombus, and Edmondson–Steiner grading were prognostic factors. However, other authors failed to yield similar results. Postoperative recurrence of HCC is the most important factor affecting the survival of patients who underwent radical resection. Poor results in some series can be explained by a high proportion of patients with cirrhosis. If the amount of resected nontumors liver parenchyma is reduced, resection of the primary liver tumor is justified despite narrow surgical resection margins. A significant reduction in postoperative mortality, as well as morbidity can be achieved by this approach. When postoperative complications occur, they do not correlate with the amount of liver resected but with preoperative liver function and intraoperative haemorrhage (71). However, the prognosis following resection for HCC remains unsatisfactory because of the high incidence of recurrence in the liver remnant; the cumulative 5-year recurrence rates after curative hepatectomy are <70% (98). Therefore appropriate management of recurrent HCC is important to improve longterm outcomes after hepatectomy. Many studies have supported favorable results after repeat hepatectomy for recurrent HCC (89). Repeat hepatectomy is the first choice for patients with preserved liver function (18). Even for the elderly patients with recurrent HCC, repeated hepatectomy has been recommended to achieve better survival if these tumors were resectable (83). Patients with recurrent HCC are older than those with primary HCC. Repeat hepatectomy for recurrent HCC is safe even for patients aged more than 75 years, especially when they underwent limited hepatectomies (89). Recently, an alternative strategy of primary hepatectomy followed by liver transplantation for recurrent HCC (salvage liver transplantation) has been proposed (29). Percutaneous ablation therapy may also be a preferable therapeutic modality for small-sized or small-volume HCC; however, there have been only a few studies on ablation therapies for recurrent HCC, and the overall 3-year survival rate after ablation therapy was 43% to 48%, which is less than the survival rates obtained after repeat hepatectomies (>70%) in certain centers (89). Not only younger patients but also elderly patients with early-stage HCC might benefit from this modality, which is less invasive than hepatectomy. Selection criteria for elderly patients with recurrent HCC who are good candidates for repeat hepatectomy remain to be determined, and the age limitations for such an aggressive operative approach are not clear at present. Nevertheless, advanced age by itself does not have an adverse effect on operative outcomes, including postoperative complications and long-term prognosis, after repeat hepatectomies on patients with recurrent HCC. Repeat hepatectomy may therefore be justified for treating recurrent HCC in selected elderly patients. In conclusion, both the short-term and long-term outcome of resection of HCC seems similar to the younger in carefully selected elderly patients, even though elderly have a higher incidence of associated diseases. HCC in the elderly is less HBVassociated, less advanced, and less aggressive. Elderly patients with preoperative alcohol abuse should be followed up very closely, even after R0 surgery, since alcohol abuse is strongly correlated with postoperative recurrence and poor survival.

49

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Hepatectomy is safe for the elderly HCC patients without preoperative comorbidities or with well-controlled preoperative comorbidities.
15. Nagasue N, Chang YC, Takemoto Y, et al. Liver resection in the aged (seventy years or older) with hepatocellular carcinoma. Surgery 1993; 113(2): 148–54. 16. Mazzoni G, Tocchi A, Miccini M, et al. Surgical treatment of liver metastases from colorectal cancer in elderly patients. Int J Colorectal Dis 2007; 22(1): 77–83. 17. Kaibori M, Matsui K, Ishizaki M, et al. Hepatic resection for hepatocellular carcinoma in the elderly. J Surg Oncol 2009; 99(3): 154–60. 18. Huang J, Li BK, Chen GH, et al. Long-term outcomes and prognostic factors of elderly patients with hepatocellular carcinoma undergoing hepatectomy. J Gastrointest Surg 2009; 13(9): 1627–35. 19. Ganz PA. Does (or should) chronologic age influence the choice of cancer treatment? Oncology (Williston Park). 1992; 6(2 Suppl): 45–9. 20. Schmucker DL. Aging and the liver: an update. J Gerontol A Biol Sci Med Sci 1998; 53(5): B315–20. 21. Anantharaju A, Feller A, Chedid A. Aging liver. A review. Gerontology 2002 ; 48(6): 343–53. 22. Zeeh J, Platt D. The aging liver: structural and functional changes and their consequences for drug treatment in old age. Gerontology 2002; 48(3): 121–7. 23. Popper H. Aging and the liver. Prog Liver Dis 1986; 8: 659–83. 24. Woodhouse KW, Wynne HA. Age-related changes in liver size and hepatic blood flow. The influence on drug metabolism in the elderly. Clin Pharmacokinet 1988; 15(5): 287–94. 25. Marchesini G, Bua V, Brunori A, et al. Galactose elimination capacity and liver volume in aging man. Hepatology 1988; 8(5): 1079–83. 26. Wynne HA, Cope LH, Mutch E, et al. The effect of age upon liver volume and apparent liver blood flow in healthy man. Hepatology 1989; 9(2): 297–301. 27. Zoli M, Magalotti D, Bianchi G, et al. Total and functional hepatic blood flow decrease in parallel with ageing. Age Ageing 1999; 28(1): 29–33. 28. James OF. Gastrointestinal and liver function of old age. Clin Gastroenterol 1983; 12(3): 671–91. 29. Watanabe T, Tanaka Y. Age-related alterations in the size of human hepatocytes. A study of mononuclear and binucleate cells. Virchows Arch B Cell Pathol Incl Mol Pathol 1982; 39(1): 9–20. 30. Le Couteur DG, Cogger VC, Markus AM, et al. Pseudocapillarization and associated energy limitation in the aged rat liver. Hepatology 2001; 33(3): 537–43. 31. Kitani K. Aging of the liver: facts and theories. Arch Gerontol Geriatr 1991; 12(2–3): 133–54. 32. Menon KV, Al-Mukhtar A, Aldouri A, et al. Outcomes after major hepatectomy in elderly patients. J Am Coll Surg 2006; 203(5): 677–83. 33. Behrns KE, Tsiotos GG, DeSouza NF, et al. Hepatic steatosis as a potential risk factor for major hepatic resection. J Gastrointest Surg 1998; 2(3): 292–8. 34. Vauthey JN, Pawlik TM, Ribero D, et al. Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases. J Clin Oncol 2006; 24(13): 2065–72. 35. Mentha G, Huber O, Robert J, et al. Elective hepatic resection in the elderly. Br J Surg 1992; 79(6): 557–9. 36. Kahng KU, Roslyn JJ. Surgical issues for the elderly patient with hepatobiliary disease. Surg Clin North Am 1994; 74(2): 345–73. 37. Ashkanani F, Heys SD, Eremin O. The management of cancer in the elderly. J R Coll Surg Edinb 1999; 44(1): 2–10. 38. Repetto L, Granetto C, Venturino A. Comorbidity and cancer in the aged: the oncologists point of view. Rays 1997; 22(1 Suppl): 17–19. 39. Miyagawa S, Makuuchi M, Kawasaki S, Kakazu T. Criteria for safe hepatic resection. Am J Surg 1995; 169(6): 589–94. 40. Shimada M, Takenaka K, Fujiwara Y, et al. Risk factors linked to postoperative morbidity in patients with hepatocellular carcinoma. Br J Surg 1998 ; 85(2): 195–8. 41. Little SA, Jarnagin WR, DeMatteo RP, Blumgart LH, Fong Y. Diabetes is associated with increased perioperative mortality but equivalent longterm outcome after hepatic resection for colorectal cancer. J Gastrointest Surg 2002; 6(1): 88–94. 42. Ettorre GM, Sommacale D, Farges O, et al. Postoperative liver function after elective right hepatectomy in elderly patients. Br J Surg 2001; 88(1): 73–6. 43. Fortner JG, Lincer RM. Hepatic resection in the elderly. Ann Surg 1990; 211(2): 141–5.

financial cost
In the current climate of scarce health care resources, treatment for elderly patients has been under close scrutiny. Several studies have shown that elderly patients have benefited from liver resection for malignancy with results comparable to those younger than 70 years of age. The use of health care resources in terms of intensive care unit and in-hospital stays is no different than in the younger population, and some of this costsaving can be attributed to better support in terms of anesthesia and community nursing (32). Therefore careful selection of patients using the ASA grade and meticulous surgical technique are essential to achieve better outcomes after hepatic resection in patients over the age of 70 years.

conclusions
Age alone should not be considered a contraindication for liver resection: hepatectomy is safe, effective, and a curative therapy in the elderly. Major hepatectomies are the feasible procedure in patients older than 70 years who have preserved liver function and controllable medical conditions, yielding close to 0% operative mortality and low morbidity rates in specialized tertiary centers.

references
1. Takenaka K, Shimada M, Higashi H, et al. Liver resection for hepatocellular carcinoma in the elderly. Arch Surg. 1994; 129(8): 846–50. 2. Fong Y, Blumgart LH, Fortner JG, Brennan MF. Pancreatic or liver resection for malignancy is safe and effective for the elderly. Ann Surg 1995; 222(4): 426–34; discussion 34–7. 3. Brooks B. Surgery in patients of advanced age. Ann Surg 1937; 105(4): 481–95. 4. Fong Y, Brennan MF, Cohen AM, et al. Liver resection in the elderly. Br J Surg 1997; 84(10): 1386–90. 5. WHO. Ageing—Exploding the Myths. Ageing and Health Program (AHE). WHO. 1999. 6. Adam R, Frilling A, Elias D, et al. Liver resection of colorectal metastases in elderly patients. Br J Surg 2010; 97(3): 366–76. 7. Scheele J, Stang R, Altendorf-Hofmann A, Paul M. Resection of colorectal liver metastases. World J Surg 1995; 19(1): 59–71. 8. Jaeck D, Bachellier P, Guiguet M, et al. Long-term survival following resection of colorectal hepatic metastases. Association Francaise de Chirurgie. Br J Surg 1997; 84(7): 977–80. 9. Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 230(3): 309–18; discussion 18–21. 10. Poon RT, Fan ST, Lo CM, et al. Hepatocellular carcinoma in the elderly: results of surgical and nonsurgical management. Am J Gastroenterol 1999; 94(9): 2460–6. 11. Cooper GS, Yuan Z, Landefeld CS, Johanson JF, Rimm AA. A national population-based study of incidence of colorectal cancer and age. Implications for screening in older Americans. Cancer 1995; 75(3): 775–81. 12. Steele G, Jr., Ravikumar TS. Resection of hepatic metastases from colorectal cancer. Biologic perspective. Ann Surg 1989; 210(2): 127–38. 13. de Liguori Carino N, van Leeuwen BL, Ghaneh P, et al. Liver resection for colorectal liver metastases in older patients. Crit Rev Oncol Hematol 2008; 67(3): 273–8. 14. Cescon M, Grazi GL, Del Gaudio M, et al. Outcome of right hepatectomies in patients older than 70 years. Arch Surg 2003; 138(5): 547–52.

50

LIVER SURGERY IN ELDERLY PATIENTS
44. Cullen DJ, Apolone G, Greenfield S, Guadagnoli E, Cleary P. ASA Physical Status and age predict morbidity after three surgical procedures. Ann Surg 1994; 220(1): 3–9. 45. Wu CC, Chen JT, Ho WL, et al. Liver resection for hepatocellular carcinoma in octogenarians. Surgery 1999; 125(3): 332–8. 46. Yanaga K, Kanematsu T, Takenaka K, et al. Hepatic resection for hepatocellular carcinoma in elderly patients. Am J Surg 1988; 155(2): 238–41. 47. Seymour DG, Pringle R. Post-operative complications in the elderly surgical patient. Gerontology 1983; 29(4): 262–70. 48. Poon RT, Fan ST, Lo CM, et al. Improving perioperative outcome expands the role of hepatectomy in management of benign and malignant hepatobiliary diseases: analysis of 1222 consecutive patients from a prospective database. Ann Surg 2004; 240(4): 698–708; discussion 709–10. 49. Poston GJ. Surgical strategies for colorectal liver metastases. Surg Oncol 2004; 13(2–3): 125–36. 50. Primrose JN. Treatment of colorectal metastases: surgery, cryotherapy, or radiofrequency ablation. Gut 2002; 50(1): 1–5. 51. Stangl R, Altendorf-Hofmann A, Charnley RM, Scheele J. Factors influencing the natural history of colorectal liver metastases. Lancet 1994; 343(8910): 1405–10. 52. Sugarbaker PH. Surgical decision making for large bowel cancer metastatic to the liver. Radiology 1990; 174(3 Pt 1): 621–6. 53. Quaglia A, Capocaccia R, Micheli A, Carrani E, Vercelli M. A wide difference in cancer survival between middle aged and elderly patients in Europe. Int J Cancer 2007; 120(10): 2196–201. 54. Petrowsky H, Clavien PA. Should we deny surgery for malignant hepatopancreatico-biliary tumors to elderly patients? World J Surg 2005; 29(9): 1093–100. 55. Zacharias T, Jaeck D, Oussoultzoglou E, Bachellier P, Weber JC. First and repeat resection of colorectal liver metastases in elderly patients. Ann Surg 2004; 240(5): 85 8–65. 56. Brunken C, Rogiers X, Malago M, et al. [Is resection of colorectal liver metastases still justified in very elderly patients?]. Chirurg. 1998; 69(12): 1334–9. 57. Zieren HU, Muller JM, Zieren J. Resection of colorectal liver metastases in old patients. Hepatogastroenterology 1994; 41(1): 34–7. 58. Figueras J, Ramos E, Lopez-Ben S, et al. Surgical treatment of liver metastases from colorectal carcinoma in elderly patients. When is it worthwhile? Clin Transl Oncol 2007; 9(6): 392–400. 59. Moreaux J. [Hepatic metastases of colorectal cancer. Natural history and surgical treatment by excision]. Chirurgie 1985; 111(7): 528–37. 60. Cunningham D, Pyrhonen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998; 352(9138): 1413–8. 61. Scheithauer W, Kornek GV, Raderer M, et al. Randomized multicenter phase II trial of oxaliplatin plus irinotecan versus raltitrexed as firstline treatment in advanced colorectal cancer. J Clin Oncol 2002; 20(1): 165–72. 62. Minagawa M, Makuuchi M, Torzilli G, et al. Extension of the frontiers of surgical indications in the treatment of liver metastases from colorectal cancer: long-term results. Ann Surg 2000; 231(4): 487–99. 63. Nordlinger B, Guiguet M, Vaillant JC, et al. Surgical resection of colorectal carcinoma metastases to the liver. A prognostic scoring system to improve case selection, based on 1568 patients. Association Francaise de Chirurgie. Cancer 1996; 77(7): 1254–62. 64. Iwatsuki S, Dvorchik I, Madariaga JR, et al. Hepatic resection for metastatic colorectal adenocarcinoma: a proposal of a prognostic scoring system. J Am Coll Surg 1999; 189(3): 291–9. 65. Goldberg RM, Tabah-Fisch I, Bleiberg H, et al. Pooled analysis of safety and efficacy of oxaliplatin plus fluorouracil/leucovorin administered bimonthly in elderly patients with colorectal cancer. J Clin Oncol 2006; 24(25): 4085–91. 66. Goldberg RM, Sargent DJ, Morton RF, et al. A randomized controlled trial of fluorouracil plus leucovorin, irinotecan, and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004; 22(1): 23–30. 67. de Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18(16): 2938–47. 68. Pawlik TM, Choti MA. Surgical therapy for colorectal metastases to the liver. J Gastrointest Surg 2007; 11(8): 1057–77. 69. Kimura F, Miyazaki M, Suwa T, Kakizaki S. Reduction of hepatic acute phase response after partial hepatectomy in elderly patients. Res Exp Med (Berl) 1996; 196(5): 281–90. 70. Aalami OO, Fang TD, Song HM, Nacamuli RP. Physiological features of aging persons. Arch Surg 2003; 138(10): 1068–76. 71. Koperna T, Kisser M, Schulz F. Hepatic resection in the elderly. World J Surg 1998; 22(4): 406–12. 72. Wanebo HJ, Chu QD, Vezeridis MP, Soderberg C. Patient selection for hepatic resection of colorectal metastases. Arch Surg 1996; 131(3): 322–9. 73. El-Serag HB. Hepatocellular carcinoma: recent trends in the United States. Gastroenterology 2004; 127(5 Suppl 1): S27–34. 74. Kotoh K, Sakai H, Sakamoto S, et al. The effect of percutaneous ethanol injection therapy on small solitary hepatocellular carcinoma is comparable to that of hepatectomy. Am J Gastroenterol. 1994; 89(2): 194–8. 75. Seki T, Wakabayashi M, Nakagawa T, et al. Ultrasonically guided percutaneous microwave coagulation therapy for small hepatocellular carcinoma. Cancer 1994; 74(3): 817–25. 76. Figueras J, Jaurrieta E, Valls C, et al. Resection or transplantation for hepatocellular carcinoma in cirrhotic patients: outcomes based on indicated treatment strategy. J Am Coll Surg 2000; 190(5): 580–7. 77. Michel J, Suc B, Montpeyroux F, et al. Liver resection or transplantation for hepatocellular carcinoma? Retrospective analysis of 215 patients with cirrhosis. J Hepatol 1997; 26(6): 1274–80. 78. Sarasin FP, Giostra E, Mentha G, Hadengue A. Partial hepatectomy or orthotopic liver transplantation for the treatment of resectable hepatocellular carcinoma? A cost-effectiveness perspective. Hepatology 1998; 28(2): 436–42. 79. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334(11): 693–9. 80. Befeler AS, Di Bisceglie AM. Hepatocellular carcinoma: diagnosis and treatment. Gastroenterology 2002; 122(6): 1609–19. 81. Bismuth H, Majno PE, Adam R. Liver transplantation for hepatocellular carcinoma. Semin Liver Dis 1999; 19(3): 311–22. 82. Figueras J, Ibanez L, Ramos E, et al. Selection criteria for liver transplantation in early-stage hepatocellular carcinoma with cirrhosis: results of a multicenter study. Liver Transpl 2001; 7(10): 877–83. 83. Hanazaki K, Kajikawa S, Shimozawa N, et al. Hepatic resection for hepatocellular carcinoma in the elderly. J Am Coll Surg 2001; 192(1): 38–46. 84. Yeh CN, Lee WC, Jeng LB, Chen MF. Hepatic resection for hepatocellular carcinoma in elderly patients. Hepatogastroenterology 2004; 51(55): 219–23. 85. Ferrero A, Vigano L, Polastri R, et al. Hepatectomy as treatment of choice for hepatocellular carcinoma in elderly cirrhotic patients. World J Surg 2005; 29(9): 1101–5. 86. Arii S, Yamaoka Y, Futagawa S, et al. Results of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas: a retrospective and nationwide survey in Japan. The Liver Cancer Study Group of Japan. Hepatology 2000; 32(6): 1224–9. 87. Vivarelli M, Guglielmi A, Ruzzenente A, et al. Surgical resection versus percutaneous radiofrequency ablation in the treatment of hepatocellular carcinoma on cirrhotic liver. Ann Surg 2004; 240(1): 102–7. 88. Oishi K, Itamoto T, Kobayashi T, et al. Hepatectomy for hepatocellular carcinoma in elderly patients aged 75 years or more. J Gastrointest Surg 2009; 13(4): 695–701. 89. Tsujita E, Utsunomiya T, Ohta M, et al. Outcome of repeat hepatectomy in patients with hepatocellular carcinoma aged 75 years and older. Surgery 2009; 147: 696–703. 90. Tsukioka G, Kakizaki S, Sohara N, et al. Hepatocellular carcinoma in extremely elderly patients: an analysis of clinical characteristics, prognosis and patient survival. World J Gastroenterol 2006; 12(1): 48–53. 91. Dohmen K, Shirahama M, Shigematsu H, Irie K, Ishibashi H. Optimal treatment strategy for elderly patients with hepatocellular carcinoma. J Gastroenterol Hepatol 2004; 19(8): 859–65. 92. Nakamuta M, Morizono S, Kohjima M, Kotoh K, Enjoji M. Baseline characterization of patients aged 70 years and above with hepatocellular carcinoma. World J Gastroenterol 2005; 11(47): 7512–4.

51

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
93. Namieno T, Kawata A, Sato N, Kondo Y, Uchino J. Age-related, different clinicopathologic features of hepatocellular carcinoma patients. Ann Surg 1995; 221(3): 308–14. 94. Kubo S, Tamori A, Nishiguchi S, et al. Effect of alcohol abuse on polyamine metabolism in hepatocellular carcinoma and noncancerous hepatic tissue. Surgery 1998; 123(2): 205–11. 95. Yamamoto K, Takenaka K, Matsumata T, et al. Right hepatic lobectomy in elderly patients with hepatocellular carcinoma. Hepatogastroenterology. 1997; 44(14): 514–8. 96. Nagao T, Inoue S, Goto S, et al. Hepatic resection for hepatocellular carcinoma. Clinical features and long-term prognosis. Ann Surg 1987; 205(1): 33–40. 97. Zhou L, Rui JA, Wang SB, et al. Clinicopathological features, post-surgical survival and prognostic indicators of elderly patients with hepatocellular carcinoma. Eur J Surg Oncol 2006; 32(7): 767–72. 98. Sugimachi K, Maehara S, Tanaka S, Shimada M. Repeat hepatectomy is the most useful treatment for recurrent hepatocellular carcinoma. J Hepatobiliary Pancreat Surg 2001; 8(5): 410–6.

52

6

Small solitary hepatic metastases: when and how? David L. Bartlett and Yuman Fong
survival results for hepatic metastasectomy
While the purpose of this chapter is not to provide an in-depth review of the results of hepatic metastasectomy, a general sense of expected cure rate and prolongation of survival after hepatic metastasectomy for various histologies is required in order to make an informed decision regarding resection of small hepatic metastases. Colorectal Metastases Colorectal cancer, compared to other histologies, is more likely to present as disease isolated to the liver. The natural history of unresected solitary hepatic metastases from colorectal cancer was described by Wagner et al. where 39 patients with solitary metastases did not undergo therapy and the median survival was 24 months (3). Wood et al. described 15 patients with solitary hepatic metastases left untreated with a mean survival of 17 months (4). There is a considerable body of literature on the results of hepatic metastasectomy for colorectal cancer. The overall 5-year survival ranges from 22% to 39% (5). In many studies, low number and small size are associated with improved prognosis such that a small solitary metastasis from colorectal cancer has a greater than 50% of 5-year survival. Nuzzo et al. report 56% actuarial 5-year survival in patients with solitary metachronous hepatic metastases from colorectal cancer less than 4 cm in size (1). Table 6.1 reviews the results of the largest series for solitary metastasectomy. After resection of solitary metastases from colorectal cancer, 5-year survival ranges from 30% to 47% (6–10). These reports do not consider the small solitary metastases separately from the entire group of solitary metastases. The size of the lesion is expected to affect prognosis and, therefore, the actual results for small solitary hepatic metastases may be even better than the numbers reported in Table 6.1. Liver resection for hepatic colorectal metastases is, therefore, safe and effective, and may be curative. Neuroendocrine Metastases For cancers of other than colorectal origin, patients with hepatic metastases from neuroendocrine tumors have been thought to be the most likely to benefit from surgical resection. Certainly, if the tumor were symptomatic for either hormonal or physical reasons, resection should be considered even though cure is unlikely. Because of the indolent nature of these tumors, durable palliation can be achieved with cytoreduction. Fiveyear survival rates for untreated hepatic metastases from neuroendocrine tumors have ranged from 13% to 54% (11–15). In patients with no symptoms, the case for surgical resection, or any treatment for that matter, is less clear. We and others (16) have adopted a very aggressive approach even for asymptomatic tumors based only on retrospective data.

introduction
The management of patients with small hepatic metastases from colorectal cancer and other histologies requires the consideration of many diverse patient- and tumor-related factors. These factors include the natural history of the tumor type, the expected cure rate after surgical treatment, effectiveness of alternative treatments, and the morbidity of surgical resection. In general, the indications for any major surgical procedure include the potential for cure, prolongation of survival, and palliation of symptoms. For metastatic tumors to the liver in selected cases, the cure rate may be over 50% for colorectal cancer (1), but will be exceedingly rare for other histologies such as gastric cases, and melanoma and sarcoma. Small metastases to the liver generally do not cause symptoms (except for hormone secreting neuroendocrine tumors) and, therefore, palliation of symptoms is not a common indication for management of these lesions. Nevertheless, many issues remain unresolved. Does resection of a small solitary hepatic metastasis prolong survival in cases where the patient is likely to develop widespread metastases in the future? Is there any harm in allowing a tumor to go untreated for a period of time, knowing that with close follow-up the resection option may still be possible in the future? Do metastases metastasize such that a delay in management may obviate the curative option? Unfortunately, all of these difficult issues are only addressed by sparse data in the literature. The risk and extent of the surgical procedure plays a significant role in the decision making for management of small hepatic metastases. It is more reasonable to excise an enlarged subcutaneous lymph node for metastatic cancer than it is to perform a hepatic lobectomy when the chance of benefit is low in both cases. As other less invasive ablative options become routine therapy, it may be reasonable to consider these options in cases where surgical resection is unreasonable. These alternative options include percutaneous approaches at ablation such as radiofrequency ablation and percutaneous alcohol injection (2). Laparoscopic procedures may also be an alternative for the management of small hepatic metastases, including laparoscopic resection of tumors and laparoscopically directed ablation such as cryotherapy. If the risks, discomfort, and hospital stay are truly minimal, then it becomes reasonable to consider local treatment of these lesions, even with a small chance of overall benefit to the patient. This chapter will provide an overview of the data on survival benefit after resection of hepatic metastases and the techniques of surgical management. A brief discussion of minimally invasive and percutaneous procedures for management of small solitary hepatic metastases will follow. In addition, a discussion of the role for adjuvant therapy after resection or ablation of the hepatic metastases will be included.

53

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Chen et al. compared liver resection for neuroendocrine tumors with a retrospectively matched cohort who did not undergo resection, demonstrating improved survival after resection (17). The general recommendation is for aggressive surgical management of neuroendocrine metastasis (18). We acknowledge that the variable growth rate and sometimes indolent nature of these tumors make firm conclusions based on retrospective data without a nontreated control group suspect. The rarity of these tumors, however, does not allow for random assignment trials. Certainly for small hepatic metastases, aggressive surgical resection is indicated, while it is acknowledged that definitive proof of its benefit may never be achieved. Noncolorectal, Nonneuroendocrine Metastases For histologies other than colorectal or neuroendocrine cancer, the utility of hepatic metastasectomy is not as obvious. For these tumors, the liver is rarely the sole site of disease; liver metastases are rarely the ultimate cause of death, nor does it contribute significantly to symptoms prior to death. Nevertheless, selected cases of disease isolated to the liver after a long disease-free interval raise the possibility of a single site of metastatic disease that could be cured with surgical therapy. Table 6.2 reviews the largest series for hepatic metastasectomy with a variety of histologies. Breast cancer Many reviews have been published on hepatic metastasectomy for breast cancer. Due to the high incidence of breast cancer and the frequency of liver metastases for this histology, the first site of metastases is frequently observed to be hepatic. In highly selected patients, favorable results of section of such liver metastases have been reported. Raab et al. reported a 5-year survival of 18.4% in 34 patients after hepatic metastasectomy for breast cancer (19). Elias et al. reported 9% 5-year survival after resection in 21 patients (20). The relatively few patients in these reports compared to the total number of breast cancer patients in each institution during the study period reflect the degree of patient selection for surgery. The survival rates reported are actuarial survival rates and the actual cure rate is much lower. At most, hepatic metastasectomy for breast cancer should be considered cytoreductive. It may delay the development of symptoms and prolong survival, but it has very little chance of curing the disease. Sarcoma Similarly, hepatic resection for sarcoma metastases may be associated with long-term survival in highly selected patients, but it is unlikely to result in cure. In a series of 14 hepatic resections for metastatic sarcoma, recurrence was found in all patients during follow-up, and 11 of 14 failed in the liver (21). The median survival in that series was 30 months. Melanoma Metastatic cutaneous melanoma to the liver has been resected with long-term survival, but these tumors also ultimately recur (22). The erratic behavior of melanoma makes conclusions regarding the benefit of hepatic metastasectomy difficult.

Table 6.1 Survival After Hepatic Resection for a Solitary Colorectal Metastasis
Actuarial 5-year survival (%) 37 30 36a 47 47 Median survival (months) – – 45 54 –

Author Hughes et al. (6) Rosen et al. (7) Scheele et al. (8) Taylor et al. (9) Fong et al. (10)
a

Date 1988 1992 1995 1997 1997

N 509 185 180 A077 240

Actual 5-year survival.

Table 6.2 Survival Following Hepatic Metastasectomy for Noncolorectal Histologies
Author Chen et al. (17) Que et al. (57) Harrison et al. (26) Jaques et al. (21) Harrison et al.26 Elias et al. (20) Raab et al.19 Ochiai et al. (24) Bines et al. (25) Harrison et al. (26)
a b c

Histology Neuroendocrine Neuroendocrine Genitourinaryb Sarcoma Breast/melanoma/sarcoma Breast Breast Gastric Gastric Gastrointestinale

N 15 74 34 14 41 21 34 21 A07 A07

Actuarial 5-year survival (%) 73 73a 60 A00 26 A09 18 19c 14d A00

Median survival (months) NR NR NR 30 32 26 27 18 15 25

4-year survival. Includes renal (5), testicular (9), adrenal (7), ovary (7), uterine (4), cervix (2). 4 of 21 actual 5-year survivors. d 1 of 7 actual 5-year survivors. e Includes gastric (5), pancreatic (2). NR: not reached.

54

SMALL SOLITARY HEPATIC METASTASES: WHEN AND HOW?
Only in highly selected cases is it appropriate to consider resection of cutaneous melanoma. Ocular melanoma, on the other hand, has a unique natural history. Ocular melanoma preferentially metastasizes to the liver and the majority of patients die of liver failure as a direct result of tumor progression. Anecdotal reports exist of long-term survival after metastasectomy for ocular melanoma (23), although these tumors are also almost always multifocal and resection of what appears to be a solitary metastasis is most often associated with liver recurrence. These hepatic metastases may show up many years after the treatment of the primary tumor. A long disease-free interval reflects a slow tumor doubling time, and suggests resection may achieve durable palliation. Usually in this disease, however, the appearance of a solitary liver metastasis is merely a precursor of the later appearance of multiple metastases. Other gastrointestinal cancers In general, hepatic metastasectomy for gastrointestinal primaries other than colorectal is not associated with prolonged survival. For tumors such as esophageal, gastric, small bowel, and pancreatic cancer, the pattern of spread includes regional lymph nodes, the peritoneal cavity, and lung metastases in addition to liver metastases. It is unlikely that these patients will die of liver failure as a result of progression of hepatic metastases, but instead, suffer other gastrointestinal sequelae from extrahepatic tumor progression. A major operative procedure can be of significant detriment to these patients with aggressive cancers where survival is expected to be of the order of weeks to months. Nevertheless, even for these tumors, selected cases exist where one might consider resection, and the literature contains anecdotal reports of long-term survivors after liver resection (24,25). Genitourinary tumors For noncolorectal, nonneuroendocrine tumors, metastases from genitourinary primaries seem to have the best prognosis following hepatic metastasectomy. In a recent review by Harrison et al., 34 patients underwent hepatic resections for genitourinary primaries (including testicular, adrenal, ovary, renal, uterine, and cervix) with a 5-year actuarial survival of 60% (26). Other investigators have reported prolonged survival after resection for renal cell cancer (27) and adrenal cancer (28). While the natural history of genitourinary tumors contributes to these remarkable results, it does suggest a survival benefit to resection in selective cases. definitive treatment. Unfortunately, it is clear that metastatic tumors do have the potential to metastasize themselves, and this must be considered when recommending observation alone. Experimental evidence suggests that cells from spontaneous metastases are more likely to metastasize than cells populating the parent neoplasm (29). Clinically, the most obvious examples of metastases from metastatic colorectal cancer deposits are in the cases of perihepatic lymph node metastases (30) and satellite-tumor formation (31). Published data would indicate that metastases to periportal lymph nodes occur in 10% to 20% of cases of hepatic colorectal metastases (30). The presence of lymph node metastases portends a poor prognosis. Therefore, excision of liver tumors before they spread to regional lymph nodes would be advantageous. A recent paper examined the incidence of satellite micrometastasis in colorectal liver metastases by careful histologic examination of resection specimens and found that 56% of specimens had micrometastases as far as 3.8 cm away from the tumor being resected (31). In some cases, these satellites could be traced to the original metastasis by a trail of cells, suggesting spread from the original metastasis. As discussed previously, the presence of satellitosis is an important independent poor prognostic factor. It may be that a delay in resection allows for the development of satellitosis, which negatively impacts on prognosis. On the other hand, the presence of satellitosis may be an indicator of biologic aggressiveness, which portends a poor prognosis regardless of when the tumor is resected.

patient selection
Colorectal Metastases In order to decide when surgical resection is reasonable for small solitary hepatic metastases, it is important to review prognostic factors that are independent of size and number, which may influence the decision regarding management of these tumors. Many studies have examined data on prognostic factors for outcome after hepatic resection for colorectal metastases. The time to development of liver tumor after resection of the primary, pathologic margin, stage of the primary tumor, tumor number, carcinoembryonic antigen levels, satellitosis, extrahepatic disease, and positive surgical margin have all been shown to predict survival after hepatic resection for colorectal metastases independent of size (7,8,10,32). Extrahepatic disease is considered a contraindication to hepatic resection. Even the presence of perihepatic lymph nodes portends a poor prognosis and generally is felt to be a contraindication to resection. Particularly in the cases of small solitary hepatic metastases with extrahepatic disease, there would be no advantage to resection or ablation of the liver tumor because systemic disease will likely be the ultimate cause of death regardless of what is done with the liver metastases. Of the other various factors that are prognostic for outcome, surgical margin, and satellitosis are the least useful in patient selection. No one would subject a patient to surgical resection expecting a positive margin. Satellitosis cannot be easily assessed preoperatively and therefore is a poor selection criterion for surgery.

do metastases metastasize?
For small solitary hepatic metastases, where many months of growth would still not preclude resection, the question is whether a waiting period would allow for further spread of the tumor from the metastatic deposit itself. If metastatic tumors were unable to further metastasize, waiting for the first sign of progression prior to initiating treatment and allowing other metastatic disease to declare itself would seem a reasonable approach. If, however, metastases are able to spread during that waiting period, then the chance of potential cure may be adversely affected by the delay in

55

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
We analyzed our recent data on factors prognostic for outcome after resection of hepatic metastases from colorectal cancer (33). In data derived from our last 1001 liver resections for this disease, the seven factors found to be independent predictors of poor long-term outcome were 1. node positive outcome, 2. presentation of liver disease within 12 months of the primary cancer, 3. CEA > 200 ng/dl, 4. number of liver tumors > 1, 5. size > 5 cm, 6. positive margin, and 7. extrahepatic disease. From this, we formulated a clinical risk score (CRS) based on the first five of these factors for use in patient selection for surgery and for stratification of patients for clinical studies. Using one point for each criterion, a summed score of 0–2 puts patients in a low-risk group and is a strong indication for hepatectomy. In the patients with small tumors, a maximum score of 4 is possible. The 5-year survival of patients with small tumors and 0–2 points on the CRS is 47% and the median survival is 56 months (33). Patients with a score of 3–4 are in a high-risk group, with a median survival of 32 months and 5-year survival of 24% (Fig. 6.1). In these high-risk patients, a period of observation with no therapy or systemic chemotherapy allowing for the extent of metastases to declare themselves is reasonable. Improved imaging techniques such as fluorodeoxyglucose positron emission tomography (FDG PET) scanning should be considered and may help discover extrahepatic disease noninvasively in these patients at high risk for additional cancer (34). Finally, these patients should be considered for clinical studies of aggressive adjuvant chemotherapy after liver resection.
1.0

Neuroendocrine Tumors Patients with symptomatic neuroendocrine tumors should be considered for resection or ablation. For the small tumor, symptoms are most likely derived from hormonal secretion by the tumors, and such hormone levels will also provide a marker for effectiveness of the ablation or resection. For asymptomatic tumors, a period of observation to allow assessment of the pace and aggressiveness of the tumors is reasonable when the tumors are small. At the first signs of progression, resection or ablation should be considered. Noncolorectal, Nonneuroendocrine Tumors Harrison et al. defined prognostic factors involved in the resection of noncolorectal, nonneuroendocrine hepatic metastases (26). In this study, 96 patients underwent liver resection. The prognostic factors of significance on multivariate analysis included the disease-free interval (>36 months), curative resection (versus palliative incomplete resection), and primary tumor type. Their conclusions would suggest that regardless of histology, with a long disease-free interval patients may benefit from surgical resection.

resection techniques
For small solitary metastases to the liver, the goal of resection is to completely excise the tumor while preserving the maximum normal hepatic parenchyma. Preserving parenchyma facilitates postoperative recovery and also provides flexibility for further resections should intrahepatic recurrences occur (35). Small surface-oriented metastases can be excised using a nonanatomic wedge resection, whereas deeper lesions require formal segmentectomies or sectorectomies. A goal of at least a 1 cm margin is reasonable (36). The use of intraoperative ultrasound is important to rule out other small hepatic metastases, which may not be evident on preoperative scans and in defining the intersegmental planes for designing the approach to segmentectomy. Even for wedge resections, ultrasound is beneficial in defining the vascular anatomy around the lesion, which may help minimize blood loss. Wedge Resections Wedge resections must be performed meticulously to avoid inadvertently leaving a positive margin. Large chromic liver sutures can be placed and used for retraction during dissection. The parenchymal dissection should be performed along the lines used for other forms of liver resection. We prefer the Kelly clamp technique where the clamp is used to crush the normal parenchyma, exposing vessels that are then clipped, tied, suture ligated, or stapled using a vascular stapling device (37). The Pringle maneuver is used intermittently for 5 minutes at a time followed by reperfusion of the parenchyma, during which time the argon beam coagulator is used to coagulate small bleeding vessels on the surface. This technique is superior to the simple use of electrocautery for the dissection, which is often attempted for what seems to be routine wedge resections. The char effect of the electrocautery prevents adequate visualization of the anatomy, making it quite easy to stray into large vessels or into the tumor.

0.8

Survival

0.6

0.4

0.2

0.0 0 12 24 Months Figure 6.1 Prediction of long-term outcome for small (<3 cm) (N = 293) metastatic deposits based on clinical risk score (CRS). CRS is based on the following five criteria: (1) node positive primary cancer, (2) disease-free interval <12 months, (3) number of liver tumors >1, (4) size of liver tumor >5 cm and (5) CEA > 200 ng/dl. For score = 0–2 (N = 236) (open box), the median survival was 56 months and the 5-year survival 47%. For score = 3–4 (N = 57) (filled triangles), the median survival was 32 months and the 5-year survival 24%. 36 48 60

56

SMALL SOLITARY HEPATIC METASTASES: WHEN AND HOW?
The most difficult margin in performing a wedge resection is the deep margin of dissection. Using intraoperative ultrasound, the depth of dissection should be measured prior to the initiation of parenchymal dissection, including at least a 1-cm margin deep to the tumor. The dissection should be carried down perpendicular to the liver surface to the predetermined depth. At this point, the tumor can be lifted up and dissection can proceed horizontally across the base of the wedge. The tendency to resect with a “V-shaped approach” is more likely to be complicated by a positive deep margin. At the end of the dissection, the Pringle maneuver is removed and the argon beam coagulator is used to control bleeding vessels. Careful examination is made for any evidence of a bile leak, which is controlled with suture ligature. For larger lesions where it is especially difficult to achieve the deep margin safely, a cryoassisted wedge resection can be performed (38). The cryotherapy probe is inserted into the tumor and freezing is begun with real time ultrasound imaging. When the zone of freezing is confirmed by ultrasound to be at least 1 cm beyond the tumor, wedge resection is performed using the freeze margin as the margin of resection. The cryotherapy probe makes a ready retracting device and the parenchyma is usually easy to dissect at the margin of the ice-ball. Freezing must continue intermittently during dissection to ensure that the ice-ball does not retract and expose the tumor. Segmental Resections For all but the most superficial lesions, we prefer a segmental approach for the resection of tumor (39). Segmental resections have a significantly lower rate of pathologic positive margins, and this translates into improved long-term survival (40). Small, deep solitary metastases and surface lesions adjacent to major vascular structures lend themselves particularly well to segmentectomies or sectorectomies. The intersegmental planes can be identified intraoperatively using vascular landmarks with the aid of intraoperative ultrasound. Using these planes for parenchymal dissection will minimize blood loss and help ensure a safe margin. Inflow occlusion for the segment can almost always be performed first, thereby producing demarcation of the segmental planes to further enhance the dissection. The portal triad to segments II, III, and IV can be identified and controlled within the umbilical fissure with little parenchymal dissection (37). The right posterior sectoral pedicle can be found by dividing the parenchyma along a horizontal cleft (fissure of Gans) present on the inferior surface of the right lobe of the liver. The pedicle can be traced to its bifurcation to segments VI and VII for control of the individual segmental portal triads. The anterior sectoral pedicle can be dissected from an inferior or anterior approach. The major hepatic veins lie within the intersegmental planes and can be a source of significant blood loss during the parenchymal transection phase of a segmentectomy. The use of low central venous pressure (0–5 mmHg) during parenchymal dissection can decrease back bleeding in these veins (41). Extrahepatic control of the left, middle, and right hepatic veins can also be achieved and the vein of concern temporarily clamped at its junction with the vena cava during parenchymal transection to further minimize blood loss. When the solitary metastases lie near an intersegmental plane, two segments can be removed. This is most easily done as a formal sector such as the left lateral sectorectomy (segments II and III) and right posterior sectorectomy (segments VI and VII). The caudate lobe (segment I) can be resected as an isolated segmentectomy when the tumor is confined to this lobe (42). This requires a more extensive dissection, including complete division of all the perforating caudate veins draining directly into the vena cava as well as the numerous small portal triads extending off the main left pedicle at the base of the umbilical fissure. Figure 6.2 demonstrates a case of a small, solitary segment of hepatic metastasis for colorectal cancer, which was detected on an MRI scan used for screening because of a rising CEA. Although this was a surface lesion, intraoperative ultrasound revealed the segment VI triad immediately adjacent to the tumor. The segment VI triad was located by ultrasound and ligated at its origin with minimal parenchymal dissection. The intersegmental planes were then marked by electrocautery and a formal segmentectomy was performed with negative margins. While an aggressive resection was indicated and performed, the patient can still undergo a formal left or right hepatic lobectomy in the future if indicated. No dissection of the vena cava or porta hepatis was required. Morbidity and Mortality The mortality rates for major hepatic resection have decreased significantly over time to a common reporting of mortality in the 1% to 4% range (43). These values are even lower for wedge resections and segmentectomies. In a recent report of 270 wedge or segmental resections, the operative mortality was 0.5% (40). This low mortality is not surprising considering that the main cause of death in studies of liver resection is liver failure secondary to inadequate residual normal parenchyma, an unlikely event for resection of small solitary hepatic metastases where minimal normal parenchyma is sacrificed. While mortality rates are low, the complication rate for major hepatic resection is still relatively high, ranging from 20% to 50% (5). Bile leaks, perihepatic abscess, hemorrhage, cardiopulmonary complications, pleural effusions, pneumonia, and pulmonary embolism are among the most common complications (43). Many of these could be expected after segmentectomy and wedge resections as well as major hepatic resections. Even though these complications do not translate into a high mortality rate, they may affect recovery time and quality of life. While this is not a significant issue for patients expected to undergo a long-term disease-free interval or cure, it may be significant for patients whose survival is expected to be of the order of months. For those patients with aggressive tumors who are likely to fail outside the liver in the near future, less invasive techniques which are associated with a lower complication rate and quicker recovery time are more appealing. Ablative Techniques Other minimally invasive techniques include local ablative therapies such as laparoscopically directed cryotherapy (44) or

57

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C)

(D)

Figure 6.2 An example of a small, solitary colorectal metastasis to segment VI. (A) MRI reveals subtle abnormality not seen on CT scan. (B) Intraoperative ultrasound reveals the tumor and adjacent segment VI portal vein. (C) Intersegmental planes have been marked on the liver capsule with electrocautery and parenchymal dissection begun. (D) Resected segment with tumor (microscopic negative margins). (Special thanks to Dr Peter Choyke for MRI scan.)

radiofrequency ablation (45). These techniques will be discussed further in chapter 8. They provide ideal alternatives to laparotomy and major liver resection for the treatment of small solitary hepatic metastases, since the small tumor is the most likely to be completely treated by ablation techniques. Furthermore, treatment by ablative techniques does not preclude future resection. Percutaneous approaches to tumor ablation are even more attractive than laparoscopic procedures. Local injection of toxic agents such as ethanol has been shown to be effective for hepatocellular cancers, however these agents have not been proven for other histologies and are known to be poorly effective for colorectal cancer (2). Radiofrequency ablation can be performed percutaneously under ultrasound guidance with local anesthesia. Figure 6.3 demonstrates a case of a metastatic pancreatic cancer 2 years after a dramatic primary response to gemcitabine and radiation therapy. Because the patient will likely begin to fail in multiple sites in the near future with limited survival potential, a laparotomy and hepatic resection was not considered reasonable. She was treated with percutaneous radiofrequency ablation, achieving a good zone of necrosis encompassing the mass, and she spent only one day in the hospital with very minimal discomfort. How such procedures, which have low morbidity and which maintain quality of life, will factor

in the treatment of patients with small hepatic metastases must be addressed by studies with sufficient follow-up to define the local recurrence rate.

adjuvant chemotherapy
The role for adjuvant systemic chemotherapy after the removal of small solitary hepatic metastases is not well defined. Even for hepatic colorectal metastases, which are commonly treated with surgery, data on adjuvant chemotherapy after liver resection is sparse. Two retrospective studies have suggested a benefit of adjuvant systemic chemotherapy after metastasectomy, but others have not supported this (6,46–48). Use of systemic chemotherapy after resection of hepatic colorectal metastases is based mainly on data demonstrating adjuvant 5-fluorouracil (5-FU) and levamisol or 5-FU and leucovorin to decrease recurrence rate and improve survival when used after resection of the primary tumor (49). It is hoped that a similar benefit will be seen when 5-FUbased chemotherapy is used after metastasectomy. Current practice is to offer adjuvant 5-FU-based chemotherapy after hepatic resection to patients who have had no previous chemotherapy. There are currently no data to support the use of irinotecan and oxaliplatin in an adjuvant setting, although studies are in progress.

58

SMALL SOLITARY HEPATIC METASTASES: WHEN AND HOW?

(A)

(B)

(C) Figure 6.3 An example of a small, solitary pancreatic cancer metastasis treated with percutaneous radiofrequency ablation. (A) Pretreatment CT scan reveals hypodense 3 cm right lobe liver metastasis. (B) Ultrasound hoto with radiofrequency probe inserted into tumor. (C) Post-treatment scan (3 weeks) reveals large zone of necrosis replacing prior tumor. (Special thanks to Dr Thomas Shawker for ultrasound photo.)

For patients with hepatic colorectal metastases, the most common site of tumor recurrence after liver resection is the remnant liver (50). In the treatment of patients with small hepatic metastases, there is particular concern that even smaller undetected metastases may subsequently present as a liver tumor recurrence. Regional chemotherapy to treat the liver site is therefore a theoretically attractive option for adjuvant care. Data addressing the utility for such hepatic arterial infusional (HAI) chemotherapy had been sparse, consisting only of four small single-arm studies (51–53) and a single, small, randomized trial consisting of 36 patients (54). These preliminary studies demonstrated safety of such an approach, but efficacy data were insufficient to support the routine use of adjuvant intraarterial chemotherapy. Two large randomized trials examining adjuvant HAI have been completed. In the first trial (55), 224 patients from 25 centers were randomized to either no adjuvant therapy or adjuvant HAI 5-FU + systemic folinic acid. Although no difference was found between the groups, technical factors compromised this study such that only 34 of the 114 patients randomized to chemotherapy completed the adjuvant treatments. In another study, Kemeny et al. randomized 156 patients to either systemic 5-FU + leucovorin or HAI floxuridine (FUDR) + systemic 5-FU after complete resection of tumor (56). There was a significant survival advantage to HAI that is most likely related to local liver tumor control. We believe HAI chemotherapy is effective and

should be considered as an adjuvant to resection of hepatic colorectal metastases. For noncolorectal, nonneuroendocrine histologies metastatic to the liver, the most likely cause of death will be related to the disease outside the liver, regardless of how the liver is managed. For patients who are likely to develop systemic metastases in the near future, it may be reasonable to offer chemotherapy prior to resection. If the tumor responds, then a resection will be performed with confidence that other micrometastatic disease may be effectively treated with chemotherapy. If the tumor does not respond and the liver remains the only site of metastatic disease, resection is performed with increased confidence conferred by the longer period of observation. If the patient advances systemically during chemotherapy, then it is very unlikely that a resection would have been of benefit and the patient will have avoided the potential morbidity, pain, discomfort, and recovery time of an hepatic resection. That patient can go on to obtain second-line chemotherapy, investigational chemotherapy, or have no additional treatment.

conclusions
Algorithms for the management of small solitary hepatic metastases are shown in Figure 6.4. Both patient and tumor characteristics must be considered in making management decisions. The most important tumor-related characteristic is

59

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Colorectal metastases

Low CRS (0–2)

High CRS (3–4)

Resection

Observation or chemotherapy

Ablation

Resection

No extrahepatic progression

Extrahepatic progression

Adjuvant therapy protocol

Adjuvant therapy protocol

Resection or ablation (A) Neuroendocrine metastases

Chemotherapy

Non-colorectal non-neuroendocrine Asymptomatic Long disease-free interval Short disease-free interval

Symptomatic

Resection

Ablation

Observation Resection Progression No progression Effective chemotherapy (>20% response) No effective chemotherapy

Resection or ablation (B)

Observation (C)

Trial of chemotherapy

Ablation vs observation

Figure 6.4 Algorithms for the management of small hepatic metastases. (A) Algorithm for colorectal metastases (CRS, clinical risk score). (B) Algorithm for neuroendocrine metastases. (C) Algorithm for non-colorectal, non-neuroendocrine metastases.

histology. For patients with colorectal cancer (Fig. 6.4A), the prognostic factors for tumor recurrence after resection are well defined. Using the clinical risk score (CRS) as selection criterion, patients with CRS = 0–2 are ideal candidates for resection. Those with CRS = 3–4 should consider observation or chemotherapy prior to a definitive hepatic procedure. Immediate ablation or resection should be performed in the setting of a clinical trial, and most appropriately a trial examining adjuvant therapy. For neuroendocrine cancers (Fig. 6.4B), symptomatic tumors should be treated with resection and/or ablation when possible. When the cancer is found in an asymptomatic patient, a period of observation is not unreasonable because of the often indolent nature of these tumors. At resection, the principle should be to leave as much normal liver behind in order to minimize the risk of liver failure and in order to allow for

repeat anatomic liver resections in the future for recurrent disease. Enucleation with positive margins is acceptable for treatment of this histology because resection is almost never curative, and such cytoreduction can provide significant and durable palliation with minimum risk. For patients with small, solitary, noncolorectal nonneuroendocrine tumors, the most significant factor in terms of prognosis seems to be the disease-free interval (Fig. 6.4C). For patients with a long disease-free interval from primary resection a curative surgical resection is indicated as the most effective means of therapy. While it may be still unlikely that these patients can be cured, they must be given the benefit of the doubt and the most optimal procedure performed. The definition of “long” has been arbitrarily set at 36 months by Harrison et al. (26), but in reality it must vary according to histology. For gastric cancer, 12–24 months would be

60

SMALL SOLITARY HEPATIC METASTASES: WHEN AND HOW?
considered long, whereas for ocular melanoma, 3–5 years would be more reasonable. Patients with a short disease-free interval from a tumor with a poor prognosis should undergo a trial of chemotherapy if there is a known effective agent. If no effective agent exists (as is the case for most solid malignancies), then these patients are ideal for an experimental, minimally invasive, local ablative therapy. This provides an advantage to observation alone, given the low but definite risk of the metastases spreading during the observation period. It will be psychologically more comforting to the patient to know that the lesion has been ablated, and risk, pain, and recovery duration are minimal. Observation alone is also quite reasonable, but it is often not accepted by patients. Patient-related factors must also be taken into consideration. Patients who have concomitant illnesses that make them poor operative candidates may be better served with a minimally invasive or percutaneous technique, even in the case of potentially curable metastases from colorectal cancer. Because of improvements in diagnostic techniques and the routine use of serum tumor markers, the detection of small solitary hepatic metastases from various tumors will likely increase in the future. A uniform approach to these patients such as that which is outlined in the treatment algorithm should be considered.

references
1. Nuzzo G, Giuliante F, Giovannini I, Tebala GD, Clemente G, Vellone M. Resection of hepatic metastases from colorectal cancer. Hepato-gastroenterology 1997; 44: 751–9. 2. Alexander HR, Bartlett DL, Fraker DL, Libutti SK. Regional treatment strategies for unresectable primary or metastatic cancer confined to the liver. In: DeVita VT Jr, Hellman S, Rosenberg SA, eds. Cancer: Principles and Practice of Oncology. Philadelphia, PA: JB Lippincott, 1996: 1–19. 3. Wagner JS, Adson MA, van Heerden JA, Adson MD, Ilstrup DM. The natural history of hepatic metastases from colorectal cancer. A comparison with resective treatment. Ann Surg 1984; 199: 502–8. 4. Wood CB, Gillis CR, Blumgart LH. A retrospective study of the natural history of patients with liver metastases from colorectal cancer. Clin Oncol 1976; 2: 285–8. 5. Fong Y, Blumgart LH, Cohen AM. Surgical treatment of colorectal metastases to the liver. CA Cancer J Clin 1995; 45: 50–62. 6. Hughes KS, Simon R, Songhorabodi S. Resection of the liver for colorectal carcinoma metastases: a multi-institutional study of indications for resection. Surgery 1988; 103: 278–88. 7. Rosen CB, Nagorney DM, Taswell HF et al. Perioperative blood transfusion and determinants of survival after liver resection for metastatic colorectal carcinoma. Ann Surg 1992; 216: 493–505. 8. Scheele J, Stang R, Altendort-Hofmann A, Paul M. Resection of colorectal liver metastases. World J Surg 1995; 19: 59–71. 9. Taylor M, Forster J, Langer B, Taylor BR, Greig PD, Mahut C. A study of prognostic factors for hepatic resection for colorectal metastases. Am J Surg 1997; 173: 467–71. 10. Fong Y, Cohen AM, Fortner JG et al. Liver resection for colorectal metastases. J Clin Oncol 1997; 15: 938–46. 11. Moertel CG. An odyssey in the land of small tumors. J Clin Oncol 1987; 5: 1503–22. 12. Thompson GB, van Heerden JA, Grant CS, Carney JA, Ilstrup DM. Islet cell carcinomas of the pancreas: a twenty-year experience. Surgery 1988; 104: 1011–7. 13. Declore R, Friesen SR. Gastrointestinal neuroendocrine tumors. J Am Coll Surg 1994; 178: 188–211. 14. Godwin JD. Carcinoid tumors: an analysis of 2837 cases. Cancer 1975; 36: 560–9. 15. Sjoblom SM. Clinical presentation and prognosis of gastrointestinal carcinoid tumours. Scand J Gastroenterol 1988; 23: 779–87. 16. McEntee GP, Nagourney DMN, Kvols LK, Moertel CG, Grant CS. Cytoreductive hepatic surgery for neuroendocrine tumors. Surgery 1990; 108: 1091. 17. Chen H, Hardacre JM, Uzra A, Cameron JL, Choti MA. Isolated liver metastases from neuroendocrine tumors: does resection prolong survival? J Am Coll Surg 1998; 187: 88–92. 18. Ihse I, Persson B, Tibblin S. Neuroendocrine metastases of the liver. World J Surg 1995; 19: 76–82. 19. Raab R, Nussbaum KT, Behrend M, Weimann A. Liver metastases of breast cancer: results of liver resection. Anticancer Res 1998; 18: 2231–3. 20. Elias D, Lasser PH, Montrucolli D, Bonvallot S, Spielmann M. Hepatectomy for liver metastases from breast cancer. Eur J Surg Oncol 1995; 21: 510–3. 21. Jaques DP, Coit DG, Casper ES, Brennan MF. Hepatic metastases from soft-tissue sarcoma. Ann Surg 1995; 221: 392–7. 22. Schwartz SI. Hepatic resection for noncolorectal nonneuroendocrine metastases. World J Surg 1995; 19: 72–5. 23. Salmon RJ, Levy C, Plancer C, et al. Treatment of liver metastases from uveal melanoma by combined surgery-chemotherapy. Eur J Surg Oncol 1998; 24: 127–30. 24. Ochiai T, Sasako M, Mizuno S. Hepatic resection for metastatic tumours from gastric cancer: analysis of prognostic factors. Br J Surg 1994; 81: 1175–8. 25. Bines SD, England G, Deziel DJ, Witt TR, Doolas A, Roseman DL. Synchronous, metachronous and multiple hepatic resections of liver tumors originating from primary gastric tumors. Surgery 1993; 114: 799–805. 26. Harrison LE, Brennan MF, Newman E et al. Hepatic resection for noncolorectal, nonneuroendocrine metastases: a fifteen-year experience with ninety-six patients. Surgery 1997; 121: 625–32.

key points
Factors that determine management
● ● ● ●

Natural history of tumor type Expected cure rate after surgical treatment Effectiveness of alternative treatment strategies Morbidity of surgical resection

Survival rates following hepatic resection




Good evidence for long-term survival Colorectal metastases Neuroendocrine metastases Survival possible in highly selected cases Breast cancer Sarcoma (especially gastrointestinal stromal tumors) Melanoma

Patient selection factors in colorectal metastases




Contraindications Extrahepatic disease (except solitary pulmonary metastases) Positive hilar lymph nodes Relative contraindications Presentation within 12 months of resection of primary tumor CEA >200 ng/dl >1 liver tumor Tumor >5 cm in size Positive resection margin

61

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
27. Fujisaki S, Takayama T, Shimada K, et al. Hepatectomy for metastatic renal cell carcinoma. Hepato-gastroenterology 1997; 44: 817–9. 28. Iwatsuki S, Shaw BW, Starzl TE. Experience with 150 liver resections. Ann Surg 1983; 197: 247. 29. Talmadge JE, Fidler IJ. Enhanced metastatic potential of tumor cells harvested from spontaneous metastases of heterogeneous murine tumors. J Natl Cancer Inst 1982; 69: 975–80. 30. Elias D, Saric J, Jaeck D et al. Prospective study of microscopic lymph node involvement of the hepatic pedicle during curative hepatectomy for colorectal metastases. Br J Surg 1996; 83: 942–5. 31. Nanko M, Shimada H, Yamaoka H et al. Micrometastatic colorectal cancer lesions in the liver. Jpn J Surg 1998; 28: 707–13. 32. Hughes KS, Simon R, Songhorabodi S, Adson MA. Resection of the liver for colorectal carcinoma metastases: a multi-institutional study of patterns of recurrence. Surgery 1986; 100: 278–84. 33. Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 230(3): 309–18. 34. Delbeke D, Vitola JV, Sandler MP et al. Staging recurrent metastatic colorectal carcinoma with PET. J Nucl Med 1997; 38: 1196–201. 35. FernE1ndez-Trigo V, Shamsa F, Sugarbaker PH. Repeat liver resections from colorectal metastasis. Surgery 1995; 117: 296–304. 36. Shirabe K, Takenaka K, Gion T et al. Analysis of prognostic risk factors in hepatic resection for metastatic colorectal carcinoma with special reference to the surgical margin. Br J Surg 1997; 84: 1077–80. 37. Blumgart LH. Liver resection—liver and biliary tumours. In: Blumgart, LH ed. Surgery of the Liver and Biliary Tract. New York: Churchill Livingstone, 1994: 1495–538. 38. Polk W, Fong Y, Karpeh M, Blumgart LH. A technique for the use of cryosurgery to assist hepatic resection. J Am Coll Surg 1995; 180: 171–6. 39. Billingsley KG, Jarnagin WR, Fong Y, Blumgart LH. Segment-oriented hepatic resection in the management of malignant neoplasms of the liver. J Am Coll Surg 1999; 187: 471–81. 40. DeMatteo RP, Palese C, Jarnagin WJ, Sun RL, Blumgart LH, Fong Y. Anatomic segmental hepatic resection is superior to wedge resection as an oncologic operation for colorectal liver metastases. J Gastrointest Surg 2000; 4(2): 178–84. 41. Cunningham JD, Fong Y, Shriver C. One hundred consecutive hepatic resections: blood loss, transfusion and operative technique. Arch Surg 1994; 129: 1050–6. 42. Bartlett D, Fong Y, Blumgart LH. Complete resection of the caudate lobe of the liver: technique and results. Br J Surg 1996; 83: 1076–81. 43. Fong Y, Blumgart LH. Hepatic colorectal metastasis: current status of surgical therapy. Oncology 1998; 12: 1489–94. 44. Lezoche E, Paganini AM, Feliciotti F, et al. Ultrasound-guided laparoscopic cryoablation of hepatic tumors: preliminary report. World J Surg 1998; 22: 829–36. 45. Siperstein AE, Rogers SJ, Hansen PD, Gitomersky A. Laparoscopic thermal ablation of hepatic neuroendocrine tumor metastases. Surgery 1997; 122: 1147–55. 46. Fortner JG, Silva JS, Golbey RB. Multivariate analysis of a personal series of 247 consecutive patients with liver metastases from colorectal cancer: I. Treatment by hepatic resection. Ann Surg 1984; 196: 306–16. 47. Butler J, Attiyeh FF, Daly JM. Hepatic resection for metastases of the colon and rectum. Surg Gynecol Obstet 1986; 162: 109–13. 48. Pagana TJ. A new technique for hepatic infusional chemotherapy. Semin Surg Oncol 1986; 2: 99–102. 49. Moertel CG, Fleming TR, Macdonald JS et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 1990; 322: 352–8. 50. Blumbart LH, Fong Y. Surgical management of colorectal metastases to the liver. Curr Prob Surg 1995; 5: 333–428. 51. Goodie DB, Horton MD, Morris RW, Nagy LS, Morris DL. Anaesthetic experience with cryotherapy for treatment of hepatic malignancy. Anaes Int Care 1992; 20: 491–6. 52. Moriya Y, Sugihara K, Hojo K, Makuuchi M. Adjuvant hepatic intra-arterial chemotherapy after potentially curative hepatectomy for liver metastases from colorectal cancer: a pilot study. Eur J Surg Oncol 1991; 17: 519–25. 53. Curley SA, Roh MS, Chase JL, Hohn DC. Adjuvant hepatic artery infusion chemotherapy after curative resection of colorectal liver metastases. Am J Surg 1993; 166: 743–8. 54. Kemeny MM, Goldberg D, Beatty D et al. Results of a prospective randomized trial of continuous regional chemotherapy and hepatic resection as treatment of hepatic metastases from colorectal primaries. Cancer 1986; 57: 492–8. 55. Lorenz M, Muller HH, Schramm H et al. Randomized trial of surgery versus surgery followed by adjuvant hepatic arterial infusion with 5-fluorouracil and folinic acid for liver metastases of colorectal cancer. Ann Surg 1998; 228: 756–62. 56. Kemeny N, Huang Y, Cohen AM et al. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl Med 1999; 341(27): 2039–48. 57. Que FG, Nagorney DM, Batts KP, Linz LJ, Kvols LK. Hepatic resection for metastatic neuroendocrine carcinomas. Am J Surg 1995; 169: 36–43.

62

7

Managing complications of hepatectomy Fenella K. S.Welsh, Timothy G. John, and Myrddin Rees
postoperative morbidity and mortality on an initial univariate analysis, it failed to independently predict outcome on subsequent multivariate analysis. Similarly, Belghiti’s group found that the in-hospital mortality rate was significantly higher in those patients with cirrhosis (8.7%) compared to those without underlying liver disease (1%, p < 0.001), but this was not subjected to multivariate analysis (7). Thus while liver resection in cirrhotic patients is technically more challenging than resecting normal liver, with a higher incidence of bleeding, septic complications, and postoperative liver failure (14), these two studies would suggest that in experienced high-volume centers, liver resection can be safely performed in patients with early cirrhosis. The common complications of hepatectomy may be classified as specific to the procedure or of a more general nature (Table 7.3). This chapter will deal with these complications in turn, focusing on their definition, incidence, predisposing factors, prevention, presentation, investigation, and treatment.

introduction
The safety of elective liver surgery has improved dramatically in the past 30 years. A multicenter American series comprising 621 liver resections published in the late 1970s reported a 13% mortality (1). By contrast, recent large published series describe posthepatectomy mortality rates of 0% to 4.4%, with 19.6% to 45% morbidity (2–8) (Table 7.1). Furthermore, individual units have demonstrated a significant reduction in morbidity and mortality over time, despite ever-widening the indications for hepatectomy (6,8). This dramatic improvement in immediate postoperative outcome can be explained by increased specialization of liver surgery in high-volume centers (9), better selection of patients in terms of hepatic functional reserve and comorbid conditions, advances in surgical technique, including greater understanding of hepatic segmental anatomy and improved instrumentation for the parenchymal transection. Furthermore, anesthesia and critical care has improved enormously, the routine use of low central venous pressure (CVP) anesthesia being a particular advance. However, even a 20% complication rate remains significant, particularly if the indication for hepatectomy is for livingdonor transplantation. Furthermore, postoperative morbidity can also adversely affect disease-specific and disease-free survival (10–12). Thus the short- and long-term consequences of postoperative morbidity, coupled with increasing litigation, and limited health care resources, has renewed the drive to further improve the immediate outcome from liver resection, with emphasis on prevention of and improved management of complications, when they occur. The precise definitions of the specific complications such as bleeding, bile leak, and hepatic insufficiency are still without consensus. Moreover, the stratification of the severity of each complication is still unclear. Standardized definitions, grading, and reporting of the complications of hepatectomy are needed to allow an objective, quality assessment of outcome data from different units and further improve results. The system proposed and validated by Clavien, focusing on the therapeutic consequences of complications in order to rank their severity, is currently the best available (13). However, it is still not universally adopted within the surgical community. A number of studies have attempted to identify the risk factors associated with complications and death from hepatectomy, three of which are detailed in Table 7.2. From these studies, there is consensus that the estimated blood loss or blood transfusion rate, the extent of hepatic resection, and an additional extrahepatic procedure are all independent predictors of morbidity and mortality. In addition, medical comorbidity, an elevated preoperative creatinine, preoperative thrombocytopenia, or hypoalbuminemia also appear to increase the operative risk. However, in the Hong Kong study (8), while cirrhosis per se was associated with increased

bleeding
Incidence Bleeding is the most feared complication of hepatectomy, both on the operating table and in the immediate aftermath of surgery. In the 1960s and 1970s, it was the cause of major morbidity and mortality. The 1974 Liver Tumor Survey was a multicenter series of 621 hepatic resections performed in 98 U.S. centers, published in 1977. It reported a 13% mortality, with 15 of the 82 deaths (18%) due to exsanguinating hemorrhage in the operating room and bleeding being the documented primary cause of death in 26 of the 76 patients (34%), where the cause of death could be determined (1). However, bleeding is now relatively rare, with the median estimated blood loss for an elective hepatectomy being 345 to 600 ml (3,6) and the need for perioperative blood transfusion now being the exception rather than the rule. Indeed, the incidence of major hemorrhagic complications is rare, 0.7% (7/1005) in our own series (3). Of these seven cases, there were no on-table deaths, five patients were treated nonoperatively and two underwent reexploration for bleeding from a hepaticojejunal anastomosis and a left caudate branch of the portal vein respectively. In the Sloan-Kettering series of 1803 patients, the incidence is similar (1%) (6). Prevention Prevention remains the key to the management of bleeding. In the preoperative assessment, a careful drug history should be taken. If the patient is on drugs such as aspirin, clopidogrel, or warfarin, the indication for the treatment should be reviewed, and the drugs stopped where possible. Patients on warfarin as prophylaxis for thromboembolic events can be managed with an inferior vena cava (IVC) filter, placed preoperatively. It is

63

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 7.1 Morbidity and Mortality from Hepatic Resection in Recent Large Case-Series
Reference Imamura et al. Rees et al. Wei et al. Malik et al. Jarnagin et al. Belghiti et al. Years of study 1994–2002 1987–2005 1992–2002 1993–2006 1991–2001 1990–1997 No of centers 1 1 2 1 1 1 No of resections 1056 1005 423 687 1803 747 Case-mix 50% HCC 29% cirrhotic 100% CRLM 100% CRLM 100% CRLM 62% CRLM 10% HCC Elective & emergency. 35% benign 28% HCC 17% CRLM 32% cirrhotic 60% HCC 33% cirrhotic 0% 1.5% 1.7% 3.0% 3.1% 4.4% all 3.9% elective 8.7% cirrhotic 25.0% emergency 4.9% Mortality Morbidity 39% 25.9% 19.6% 29.5% 45.0% 22.0%

Poon et al.

1989–2003

1

1222

32.4%

Abbreviations: CRLM, colorectal liver metastases; HCC, hepatocellular carcinoma.

Table 7.2 Three Studies Reporting the Independent Predictors of Morbidity and Mortality after Hepatic Resection
Reference Jarnagin et al. Years of study 1991–2001 No of resections 1803 Predictors of morbidity Estimated blood loss Extent of resection + EH procedure ↑ preoperative creatinine Hypoalbuminemia Medical comorbidity Male gender ASA score Extent of resection Steatosis Blood transfusion + EH procedure Thrombocytopenia Blood transfusion + EH procedure Predictors of mortality Estimated blood loss Extent of resection + EH procedure ↑ preoperative bilirubin Thrombocyt openia Age + EH procedure (in patients with malignancy)

Belghiti et al.

1990–1997

478 elective resections, no cirrhotics

Poon et al.

1989–2003

1222

Hypoalbuminemia Thrombocytopenia ↑ preoperative creatinine Major resection Blood transfusion

Abbreviation:+EH Procedure, additional extra-hepatic procedure.

important to identify patients with tricuspid regurgitation or right heart disease, where the anesthetist may encounter difficulties in lowering the CVP, as this may influence the extent of resection. Careful evaluation and correction of coagulation abnormalities should be performed pre- and perioperatively, particularly in the cirrhotic patient. Two key maneuvers are used to prevent bleeding during hepatic transection: portal triad clamping and low CVP anesthesia. Portal triad clamping, first described by Pringle in 1908, reduces hepatic arterial and portal venous bleeding (15,16). Although a European survey demonstrated that the use of inflow occlusion is not universal, it did confirm that most hepatic surgeons resort to it in difficult cases and that experienced surgeons are more likely to use it routinely (17). However, a recent systematic review and meta-analysis of the effect of inflow occlusion on postoperative morbidity and mortality failed to demonstrate any significant outcome

benefit (18). This is confirmed by another systematic review published in 2009, which compared 166 patients with vascular occlusion to 165 patients with no vascular occlusion (19). However, despite the small numbers involved, this later study showed that blood loss was significantly lower in those patients who had vascular occlusion. A low CVP reduces back bleeding from hepatic veins during the transection (20–22) and is now accepted practice during liver resection worldwide. Indeed, following the introduction of low CVP anesthesia in our own unit, the mean blood loss was significantly reduced from 2116 to 426 ml (3). However, these techniques can test the patients’ cardiovascular reserve. Obstructing the portal blood flow causes venous congestion of bowel and in combination with warm ischemic liver injury, releasing a flush of anerobic metabolites and cytokines back into the circulation on release of the clamp (23). Low CVP anesthesia relies on patients being maintained in a

64

MANAGING COMPLICATIONS OF HEPATECTOMY
Table 7.3 Complications of Hepatectomy
General complications Immediate (on table) Early (days) • Hypothermia • Respiratory— atelectasis, pleural effusion, pneumonia • Cardiovascular— DVT, PE, MI, arrhythmias, CVA • Renal failure • Wound infection • Pain • Incisional hernia Specific complications • Bleeding • • • • Bleeding Bile leak Hepatic insufficiency Intra-abdominal abscess

Investigation and Treatment The hemoglobin concentration and clotting screen should be performed urgently, ensuring that the patient has an up-todate cross match. Any coagulopathy should be corrected. If the patient remains shocked, appropriate investigations may include endoscopy and mesenteric arteriography. Ultimately, as in our own series, small number of patients may need to return to the operating theatre for surgical control of hemorrhage.

biliary complications
• Biliary stricture

Late (weeks/ months)

Abbreviations: DVT, deep vein thrombosis; PE, pulmonary embolus; MI, myocardial infarction; CVA, cerebrovascular accident.

hypovolemic state until liver resection has been completed (20,21). This is in contrast to most other major surgical procedures, where patients have large volumes of crystalloid and colloid perioperatively. While a recent meta-analysis has confirmed that the use of the antifibrinolytic agent aprotinin can significantly reduce transfusion requirements during liver transplantation (24), there is no evidence for its routine use during liver resection (25). In contrast, a prospective double-blind randomized trial of tranexamic acid, another antifibrinolytic agent, has shown that its use perioperatively significantly reduced the blood loss and transfusion requirements in elective liver resection (26). Two prospective randomized controlled trials have failed to show any benefit of using recombinant factor VIIa in either noncirrhotic (27) or cirrhotic (28) patients undergoing hepatectomy. Since the early 1990s, the use of fibrin sealants has become a popular aid hemostasis at the hepatic parenchymal transection site. Two early randomized trials suggested some benefit in achieving hemostasis (29) and reducing postoperative blood loss (30), although the numbers involved were small. A more recent trial of a carrier-bound fibrin sealant (TachoSil®) suggested it was quicker and more effective hemostasis compared to argon beam coagulation (31). However, the numbers involved were again small (<65 patients in each group). A larger prospective randomized trial of fibrin glue versus control involving 300 patients undergoing hepatic resection was published in 2007 (32). This showed no difference in blood loss, blood transfusion, or overall morbidity between those who received the fibrin glue and those who did not. This study provides the best evidence to date that the routine use of such topical hemostats is not justified, although it is our own personal bias that fibrin glue needs to be combined with a collagen matrix to be effective. Presentation Postoperatively, patients who are bleeding may present with classical signs of shock, persistent blood loss in an abdominal drain, a drop in hemoglobin, or gastrointestinal bleeding.

Incidence and Definition of a Bile Leak Bile leak remains a persistent problem after hepatectomy, with a reported incidence of 1% to 12% (3,33). In addition, it appears to be the most common complication after living donor hepatectomy, with an incidence of 7.5% in the 731 donors in one Japanese series (34) and 9% in 381 donors in an American series (35). The variable incidence may be explained by the different patient populations analyzed and the lack of consensus regarding the definition. The Amsterdam group has defined bile leakage as one or more of the following criteria: the presence of persisting bile-stained effluent from an abdominal drain, leakage detected on radiological imaging, and occurrence of a bile collection drained percutaneously or found during relaparotomy (36). This definition of a bile leak, used in conjunction with Clavien’s severity grading (13), could be widely applied in clinical practice. Prevention Meticulous technique during the parenchymal transection, ensuring that both small and large bile ducts are adequately secured with clips, ties, or sutures, is vital for the prevention of bile leakage. Inspection of the cut surface and application of a clean white-gauze swab is usually enough to reveal a bile leak, which must then be sutured. Methods for testing for bile leakage have previously been advocated and include injection of the biliary system (usually via the cystic duct stump) with saline solution or methylene blue, or formal direct cholangiography after the transection has been completed. However, the only prospective randomized study has shown no evidence that such maneuvers reduce the bile-leak rate (37) and thus this technique cannot be recommended as a routine. While one study has show that topical fibrin glue significantly reduces the bile-leak rate following hepatectomy (38), other studies have failed to show any benefit (29,39). Thus there is no clear evidence that topical hemostatic agents used after the liver resection on the parenchymal surface reduce the bile-leak rate (33). Risk Factors for a Bile Leak There is consensus in the literature that extended hepatic resections are associated with an increased risk of bile leak (36). An Italian series of 610 liver resections without a concurrent hepaticojejunostomy reported a 3.6% incidence of postoperative bile leakage (38). On multivariate analysis, they found that resection of a peripheral cholangiocarcinoma (relative risk 5.5) and hepatectomies including segment 4 (relative risk 3.1) were the only independent risk factors for a bile leak.

65

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Other risk factors reported include a large transection area and operations, which expose the major Glissonian sheath around the hepatic hilum (major central resections including segments 4b, 5, or the caudate), with subsequent unrecognized injury to the bile duct (40). Gertsch and coworkers showed that patients who had postoperative ischemia of part of the remnant liver had a higher incidence of bile leakage (18.4%) compared to those with no ischemia (2.7%) (41). In addition, any hepatic resection, which includes resection of the extrahepatic biliary tree with concomitant hepaticojejunostomy has a significantly higher bile leak rate (36,40,42). The presence of cirrhosis appears to be associated with a lower risk of a bile leak, possibly because of a less aggressive surgical approach in these patients (38). Presentation of a Postoperative Bile Leak A bile leak can present as bile-stained effluent from an abdominal drain. Other patients will show signs of intra-abdominal sepsis, with a fever, abdominal pain, or right-sided chest signs, and leak bile into a secondarily placed drain. Management of a Postoperative Bile Leak Minor bile leaks may often resolve with no requirement for further intervention. In their case series, Vigano and coworkers found that 77% of bile leaks settled spontaneously. However, a drainage output greater than 100 ml on postoperative day 10 was the only independent risk factor for failure of conservative management (43). Percutaneous tube drainage should then be the intervention of choice. If percutaneous drainage fails because of persistent or recurrent bile leakage, endobiliary stenting should be undertaken, to reduce the intrabiliary pressure and promote rapid resolution of the bile leak (44,45). Clearly this can only be successful if there is communication between the leaking bile duct and the main biliary tree. In the face of failure of percutaneous and endoscopic approaches, relaparotomy should be undertaken, with a view to optimizing drainage, a further hepatic resection, or formation of a biliary enteric anastomosis. The precise intraoperative decision will depend on the volume of the liver remnant and functional liver reserve as well as the extent of local sepsis. We have had to perform a biliary enteric anastomosis for a persistent bile leak in one patient out of our entire cohort of liver resections (1/1600). The patient had undergone an extended left hepatectomy. Following an initially uncomplicated postoperative course with discharge home on day 4, she was readmitted three weeks later with intra-abdominal sepsis. After initial percutaneous drainage of a large bile collection, the leak failed to resolve with endoscopic biliary drainage. A laparotomy was therefore performed and the area of bile leakage identified at the resection edge, but no actual bile duct seen. A Roux-en-Y jejunal loop was therefore anastomosed to the resection edge with a successful outcome. Consequences of a Bile Leak The direct consequences of a postoperative bile leak include prolonged hospital stay and increased morbidity and mortality (39). Patients with persistent bile leakage are at risk of developing intra-abdominal sepsis, with the attendant risk of liver failure and death (46). Incidence of Biliary Stricture after Hepatectomy A biliary stricture is an uncommon, late complication of hepatectomy, with an incidence of 0.2% (4/1803) in the SloanKettering series (6). It is caused by unrecognized intrahepatic injury to the bile ducts, either directly or due to isolated devascularization of the biliary tree. A distal biliary stricture may be responsible for a persistent proximal bile leak. Management of Biliary Stricture After Hepatectomy A biliary stricture after hepatectomy should be managed in the same way as any iatrogenic biliary stricture. This will include radiological and/or endoscopic assessment of the level of the stricture and its relationship to the remaining biliary tree, together with an estimate of the volume and function of the hepatic remnant. Avoidance of sepsis or cholangitis is paramount, as is attention to the nutritional status of the patient. Potential treatments include endobiliary stenting, biliary reconstruction, or a further hepatic resection tailored to the individual circumstances.

hepatic insufficiency
Definition and Incidence There is currently no internationally accepted definition of postoperative liver failure or hepatic insufficiency. Belghiti’s group have proposed the “50-50 criteria,” which are a prothrombin index <50% of normal (corresponding to an International Normalized Ratio (INR) of 1.7 or more) and a serum bilirubin > 50 µmol/L on postoperative day 5, as a simple, accurate predictor of liver failure and death (47). On the fifth postoperative day, both prothrombin time and bilirubin should have returned to normal values. They found that the persistence of the “50-50 criteria” at this time indicated a significant impairment of liver function and was associated with a 59% risk of early postoperative mortality, compared with a 1.2% risk if the criteria were not met. They recently prospectively evaluated these criteria in a cohort of 436 elective hepatectomies and found that the “50-50 criteria” on postoperative days 3 and 5 were accurate predictors of death on multivariate analysis (48). The MD Anderson group reviewed data from 1059 noncirrhotic patients who underwent a major hepatectomy and found that a peak bilirubin of more than 120 µmol/L (7.0 mg/d/L), accurately predicted liver-related death and suggested that this be used as a definition (49). By this definition, the incidence of postoperative liver failure with or without multiorgan failure resulting in death in their series was 2.8%. In the French multicenter series of 1568 hepatectomies, the incidence of liver failure was 43/1568 (2.7%), however this was responsible for death in 7/43 (16%) of those patients (50). In the Hong Kong series (8), postoperative liver failure occurred in 47 out of the 1222 hepatic resections (3.8%), but again, it is not defined. This series had a higher incidence of patients (59.2%) with cirrhosis or chronic hepatitis compared to most Western case series. Overall, the reported incidence in the literature ranges from 0.7% to 9.1% (51).

66

MANAGING COMPLICATIONS OF HEPATECTOMY
Prevention The incidence of postoperative liver failure is related to the volume and quality of the remnant liver, the amount of blood lost during surgery (52,53), and the presence of comorbid conditions such as diabetes mellitus (51,52). As treatments are limited and the consequences life-threatening, preoperative assessment of the individual patient’s risk is vital and will affect the operative approach (54). Assessment of liver function can be achieved using liver biochemistry, coagulation studies, and the Child–Pugh classification (55). However, because of the limits of the Child–Pugh scoring system, surgeons have looked for other tests of hepatic function to help identify patients at risk of postoperative liver failure. The indocyanine green (ICG) retention test is the most widely used in clinical practice. ICG is a dye that is removed from plasma by the liver and rapidly excreted unchanged into bile. The ICG retention rate at 15 minutes (ICGR-15) provides a measure of hepatic function, with clearance said to be impaired when 15% or more of the ICG remains within the plasma at 15 minutes (56). Makuuchi’s group has successfully incorporated the ICGR-15 in their preoperative work-up of Child–Pugh class A patients to guide the extent of resection (54). The Hong Kong group also uses it in their preoperative assessment of patients with HCC, because of their high incidence of chronic liver disease. They use an ICGR of <20% as a cut-off for a major resection in cirrhotic patients (57). There is a close relationship between liver function and volume. The volume of the future liver remnant (FLR) may be assessed by computed tomography (CT) volumetry. Vauthey and colleagues reported the value of residual liver volume using CT as a predictor of hepatic dysfunction, noting a critical value of 25% below which they found a significant risk of hepatic dysfunction in patients with no underlying liver disease (58). These data have been confirmed by two other groups, who defined a critical FLR of 26.6% (53) and 26.5% (59), respectively. In patients with underlying liver disease, such as chemotherapy-associated steatohepatitis or cirrhosis, the FLR should be greater (59), with Vauthey’s group suggesting it should be at least 30% if the patient has received extensive preoperative chemotherapy and 40% if they have cirrhosis (60). A recent systematic review (61) looked at the association between chemotherapy type, liver injury, and the impact of liver injury on outcome following liver resection. This study found a significant association of irinotecan with steatohepatitis, especially in obese patients. These patients had a higher 90-day mortality rate compared to patients who did not have steatohepatitis (15% vs. 2%, p = 0.001) and a significantly higher risk of death from postoperative liver failure (6% vs. 1%, p = 0.01), highlighting that irinotecan appears to impair the functional reserve and regenerative capacity of the liver (60). Chemotherapy-associated hepatotoxicity and its impact on outcome after hepatectomy are covered in more detail in chapter 17. The use of portal vein embolization (PVE) to improve volume of the FLR and as a functional test of hepatic reserve PVE can be used to induce hypertrophy of the FLR and reduce the incidence of postoperative complications, including liver failure, in patients with a marginal FLR (62–65). PVE may also provide an important functional test of hepatic reserve in patients with a borderline FLR, with the degree of hypertrophy predicting outcome from hepatectomy (63). Certainly, if patients with borderline FLR remnants do not exhibit hypertrophy following PVE, they should not undergo hepatic resection because of the risk of postoperative liver failure—the so-called “trial of PVE.” A recent consensus statement suggests that PVE is indicated when the FLR is <20% in patients with normal liver, <30% in patients who have had chemotherapy, and <40% in patients with well-compensated cirrhosis (66). Optimization of Venous Drainage A key aspect of maximizing the function of the remnant liver and prevention of hepatic insufficiency is to preserve and optimize its venous drainage. Belghiti showed that following a right hepatectomy, left hepatic venous outflow was impaired if the left liver was not fixed in the anatomical position (defined as the position where the falciform ligament was in its strict medial position) (67). The consequent venous congestion could result in bleeding from the resection surface in the short-term and impaired function and regeneration of the liver remnant in the ensuing few days or weeks. The venous drainage on preoperative imaging should be carefully evaluated when planning any resection, therefore allowing optimization of the venous drainage of the future liver remnant. For example, preserving the umbilical vein when performing a right hepatectomy extended to segment 4a will allow adequate drainage of segment 4b. Belghiti’s group has demonstrated the importance of this, using the living donor hepatectomy as a model (68). They showed that 84% of donors who underwent right liver harvesting to include the middle hepatic vein, developed venous congestion of segment 4 postoperatively, compared to none of the donors who had right liver harvesting without including the middle hepatic vein. Furthermore, this was associated with impaired postoperative liver function and regeneration. Treatment of Postoperative Hepatic Insufficiency As emphasized above, the mainstay of management of operative hepatic insufficiency is prevention. However, should it occur, there are a number of important strategies to employ. Patients should be receiving best supportive care, to optimize other organ functions, in a minimum level 1 environment, with intensive escalation of care as required. It is important to avoid secondary septic insults such as pneumonia or intraabdominal sepsis, as any second “hit” will increase the risk of death. Liver failure and sepsis appear to be closely linked. Schindl and coworkers have reported a direct correlation between the extent of liver resection and the incidence of infective complications (53). This risk is further increased in the presence of cirrhosis or liver failure (53). Thus in a patient developing liver failure, infectious complications should be actively excluded by clinical assessment and radiological and bacteriological investigations. Infectious complications should be aggressively treated with appropriate antibiotics and drainage or reoperation as required. Other management strategies to combat posthepatectomy liver failure include dietary sodium restriction (<90 mmol

67

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
sodium per day), to reduce the sodium retention that can occur as a result of decreased renal excretion and enhanced sodium resorption (69). Some patients will develop hyponatremia and worsening ascites due to water retention. It is our practice to moderately restrict fluid for such patients to 1.5 to 2 L per day. If the serum sodium is > 126 mmol/L, patients should be commenced on spironolactone, an aldosterone antagonist, which acts on the distal tubules to increase natriuesis and conserve potassium. The initial dose should be 50 to 100 mg per day, increased up to 400 mg per day and limited by the development of hyperkalemia. The additional use of frusemide, a loop diuretic, at a dose of 40 mg per day, can enhance its natriuretic effect. In patients with a serum sodium 121 to 125 mmol/L, clinicians should consider stopping diuretics, particularly if there is evidence of renal impairment. In this scenario, patients should be given volume expansion, ideally with 20% salt-poor albumin. Other volume expanders such as Gelofusine® and 4.5% albumin solutions contain high concentrations of sodium (154 mmol/L) and their use will potentially worsen patients’ sodium retention. The management of patients with a serum sodium < 120 mmol/L is difficult and controversial. In this scenario, all diuretics should be stopped and patients should undergo volume expansion with colloid or saline. It is important that these patients are not taking nonsteroidal antiinflammatory drugs (NSAIDs), as these can also inhibit salt and water excretion and compound the problem (70,71). Hypoglycemia and hypophosphatemia should be aggressively corrected. These recommendations from the evolution of our own practice are reinforced by the current U.K. guidelines on the management of ascites in cirrhosis (69). A few small case series suggest that artificial liver support systems such as the molecular-adsorbent recirculating system (MARS) may be of value in treating posthepatectomy liver failure (72). However, a recent systematic review showed that there is currently insufficient evidence to support their use in these patients (73). leakage were all associated with postoperative infective complications (75). They show a reduction in their postoperative infection rate from 44.7% at the start of the study to 9.2% by the end, with improvements in clinical practice such as early enteral nutrition and aggressive management of bile leaks. There is no evidence that the use of postoperative systemic antibiotics reduces postoperative infective complications. In a prospective randomized trial, Wu and coworkers showed that postoperative systemic antibiotics after liver resection did not influence the incidence of infective complications, which was 23% in each group (76). Another prospective randomized trial investigated whether omentoplasty to the hepatic parenchymal transection surface reduced the incidence of deep abdominal complications (bleeding, hematoma, infection with or without purulent discharge through drains, or bile leakage). The authors found that while deep abdominal complications were significantly associated with major hepatic resections, omentoplasty did not reduce their incidence (77). Abdominal Drainage The use of routine drainage after liver resection and its role in preventing complications remains controversial. A prospective randomized trial involving 186 patients compared closed suction drainage with open drainage after elective hepatectomy. The trial showed that the incidence of infected subphrenic collections, postoperative ascites, and pleural effusion was significantly lower in the closed suction drainage group. However, both groups showed similar rates of subphrenic hematoma and biloma formation (78). In contrast, another trial prospectively randomized 120 patients undergoing elective hepatectomy to closed suction drainage or no drainage (79). This showed no difference in overall complication rate between the two groups. However, 18% of patients in the no drainage group subsequently required a percutaneous drain, compared to 8% in the drained group, but this was not statistically significant. The authors concluded that routine drainage was unnecessary after elective hepatectomy and adopted a selective drainage policy. A trial from Hong Kong, which randomized 104 patients with chronic liver disease to closed suction drainage or no drainage, showed that there was significantly higher morbidity in the drainage group (73%) compared to the no drainage group (38%) (80). Further, specifically there was a higher incidence of wound complications in the drainage group and a trend towards more septic complications. In conclusion, elective closed suction drainage in patients with chronic liver disease is not recommended. For all other patients, there is no evidence that routine abdominal drainage prevents postoperative abdominal septic complications. However, for patients at high risk of bile leakage (as outlined earlier in this chapter), routine drainage is recommended.

intra-abdominal infection
Importance and Incidence Posthepatectomy infections are important as they can precipitate liver failure and death, as discussed earlier. The incidence of infected perihepatic collections ranges from 2.7% to 6.1% in modern case series (6,8), but is higher (12.8%) in older series (74). The incidence of infected ascites is less than 1% (8). Factors Affecting the Incidence of Intra-abdominal Infection The decreasing incidence of intra-abdominal infections over time is a reflection of the evolution of liver surgery in the past 30 years. In Yanaga’s series of 149 liver resections performed between 1973 and 1984, 19 patients (12.8%) developed intraperitoneal septic complications, of whom 13 patients died of liver failure (74). They identified five risk factors for this, which were: (1) right or extended right hepatectomy, (2) age > 65 years, (3) operation time > 5 hours, (4) blood loss > 3L, and (5) postoperative bleeding, which required a laparotomy to achieve hemostasis. A further Japanese case series of 535 hepatectomies performed between 1992 and 2005 reported that advanced age, diabetes mellitus, the use of silk sutures, and bile

respiratory complications and pain relief
Incidence Respiratory complications such as pleural effusion and bronchopneumonia are common after hepatectomy. In the Hong Kong series, 7% of patients developed a postoperative pneumonia and 5% of patients had a pleural effusion requiring

68

MANAGING COMPLICATIONS OF HEPATECTOMY
aspiration (8). In the Sloan-Kettering series of 1803 resections, the corresponding incidence was 3% pneumonia and 8.5% symptomatic pleural effusion; 2.5% of patients had basal atelectasis, with a further 2.5% developing respiratory failure requiring support. In addition, 1% of patients suffered a pulmonary embolus postoperatively (6). Prevention and Management As with any abdominal operation, a patient’s risk for respiratory complications should be assessed preoperatively. Smokers should be encouraged to stop. Patients with chronic lung disease should have aggressive preoperative physiotherapy. Good postoperative pain relief to facilitate early mobilization, deep breathing, and coughing is paramount. Epidural analgesia is one of the best methods for provision of postoperative pain relief in patients recovering from major upper abdominal operations (81,82). However, the procedure itself is associated with complications such as hypotension, bradycardia, immediate or delayed respiratory depression, urinary retention, dural puncture and hematoma, and/or infection within the spinal cord. Furthermore, patients undergoing hepatectomy are at risk of a prolonged prothrombin time postoperatively and this may affect the timing of removal of the epidural catheter (83). A retrospective review of 367 patients who underwent elective hepatectomy showed that patients who had epidural analgesia had a significantly lower mean arterial blood pressure in the theater recovery area and were more likely to have a blood transfusion during their hospital course (84). Thus in our unit, for the past five years, we have moved away from epidural anesthesia to using a continuous intermuscular bupivacaine infusion combined with patientcontrolled analgesia (85). This is a safe, simple, and efficacious method of providing postoperative pain relief in patients after liver resection and is associated with a low incidence of pulmonary complications (85). To prevent the small but potentially fatal risk of thromboembolic complications, all our patients wear graduated compression stockings. Pneumatic foot pumps are worn in the operating theatre and continued until the patient is fully mobile. Lowdose subcutaneous low-molecular-weight heparin is given daily postoperatively, once the prothrombin time has returned to within three seconds of normal. When respiratory complications do occur, they should be managed aggressively and proactively to minimize the risk of sepsis precipitating hepatic insufficiency. Patients who develop cardiac complications following hepatectomy should be managed in conjunction with the local cardiologists. If required, aspirin, clopidogrel, and formal anticoagulation with heparin can be given within days of a hepatectomy, although it is advisable to avoid the administration of a large loading dose of warfarin, to minimize the risk of early secondary hemorrhage.

renal failure
Definition and Incidence Renal failure is defined as the need for renal replacement therapy. Studies have shown that 3% to 7% of patients require renal replacement therapy after liver resection (21,87). In our own case series, the incidence is 0.9% (unpublished data). Etiology of Renal Failure After Hepatic Resection There are three main factors which may contribute to the development of renal failure following liver resection. Elderly patients and those with conditions such as hypertension, atherosclerosis, or chronic kidney disease are at risk (88). These patients have a reduced capacity for neurohumoral autoregulation of glomerular blood flow during surgery and thus an increased risk of acute tubular necrosis (ATN) (88). Perioperative use of NSAIDs may also impair normal autoregulation of glomerular perfusion through inhibition of arteriolar dilatory prostaglandins (88) and should be avoided in patients with preoperative renal impairment. The second factor relates to the “hit” of surgery. Two key factors in the pathogenesis of ATN are hypovolemia and renal damage by inflammatory mediators (87). Both these events are predictable in every hepatic resection that employs low CVP anesthesia and portal inflow occlusion. Obstruction of the portal blood flow with the Pringle maneuver causes splanchnic venous congestion and, in combination with warm ischemic liver injury, results in a flush of anerobic metabolites and cytokines into the systemic circulation on release of the hepatic inflow clamp (23). Low CVP anesthesia relies on patients being maintained in a hypovolemic state until liver resection has been completed (20,21). This is in contrast to most other major surgical procedures, where patients are given significant volumes of crystalloid and colloid in the perioperative period. Moreover, vasodilators are often used to further reduce the CVP, leading to distributive changes in blood flow (20). Certainly, low CVP anesthesia with or without hepatic inflow occlusion can produce major circulatory changes, potentially resulting in ATN and subsequent renal impairment or failure (87). Another factor, which contributes to the etiology of renal failure following liver resection is a low perfusion state either secondary to cardiac dysfunction or distributive circulatory changes, such as sepsis or hepatorenal failure (87,88). Postoperative renal dysfunction is often multifactorial. Consequences of Postoperative Renal Failure The potential consequences of acute kidney injury include increased risk of mortality and may contribute to the development of chronic kidney disease (89).

cardiac complications
In our own series, the Sloan-Kettering and the Hong-Kong series, the most common cardiac complication of hepatectomy is arrhythmia, with an incidence of 2% to 5% (6,8). Myocardial infarction and heart failure will also occur in about 1% of patients. At-risk patients should be identified preoperatively and undergo a cardiac assessment with exercise or pharmacological stress echocardiography and coronary angiography. Cardiac function should be optimized preoperatively with medical therapy, coronary stenting, and coronary artery bypass grafting as required. We have also used a perioperative intra-aortic balloon pump (86).

69

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Hepatectomy in Patients with Preoperative Renal Impairment Patients with preoperative renal impairment, as defined by a raised preoperative serum creatinine, are at increased risk of both renal and non-renal complications (6). These patients require careful monitoring in the early postoperative period in order to optimize fluid balance and cardiac output and in some instances may require hemofiltration.
6. Jarnagin WR, Gonen M, Fong Y, et al. Improvement in perioperative outcome after hepatic resection. Analysis of 1803 consecutive cases over the past decade. Ann Surg 2002; 4: 397–407. 7. Belghiti J, Hiramatsu K, Benoist S, et al. Seven hundred and forty-seven hepatectomies in the 1990s: an update to evaluate the actual risk of liver resection. J Am Coll Surg 2000; 191: 38–46. 8. Poon RT, Fan ST, Lo CM, et al. Improving perioperative outcome expands the role of hepatectomy in management of benign and malignant hepatobiliary disease: analysis of 1222 consecutive patients from a prospective database. Ann Surg 2004; 240: 698–708. 9. Dimick JB, Cowan JA Jr, Knol JA, et al. Hepatic resection in the United States: indications, outcomes and hospital procedural volumes from a nationally representative database. Arch Surg 2003; 138: 185–91. 10. Laurent C, Sa Cunha A, Couderc P, et al. Influence of postoperative morbidity on long-term survival following liver resection for colorectal metastases. Br J Surg 2003; 90: 1131–6. 11. Ito H, Are C, Gonen M, et al. Effect of postoperative morbidity on longterm survival after hepatic resection for metastatic colorectal cancer. Ann Surg 2008; 247: 994–1002. 12. Chok KS, Ng KK, Poon RT, et al. Impact of postoperative complications on long-term outcome of curative resection for hepatocellular carcinoma. Br J Surg 2009; 96: 81–87. 13. Dindo D, Demartines N, Clavien PA. Classification of surgical complications. A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: 205–13. 14. Fan ST. Problems of hepatectomy in cirrhosis. Hepatogastroenterology 1998; 45: 1288–90. 15. Pringle JH. Notes on the arrest of hepatic haemorrhage due to trauma. Ann Surg 1908; 48: 541. 16. Man K, Fan ST, Ng IO, et al. Prospective evaluation of Pringle maneuver in hepatectomy for liver tumors by a randomised study. Ann Surg 1997; 226: 704–11. 17. van der Bilt JD, Livestro DP, Borren A, et al. European survey on the application of vascular clamping in liver surgery. Dig Surg 2007; 24: 423–35. 18. Rahbari NN, Wente MN, Schemmer P, et al. Systematic review and metaanalysis of the effect of portal triad clamping on outcome after hepatic resection. Br J Surg 2008; 95: 424–32. 19. Gurusamy KS, Kumar Y, Ramamoorthy R, et al. Vascular occlusion for elective liver resections. Cochrane Database of Systematic Reviews 2009; Issue 1. Art No: CD007530. DOI: 10.1002/14651858.CD007530. 20. Rees M, Plant G, Wells J, et al. One hundred and fifty hepatic resections: evolution of technique towards bloodless surgery. Br J Surg 1996; 83: 1526–9. 21. Melendez JA, Arslan V, Fischer ME, et al. Perioperative outcomes of major hepatic resections under low central venous pressure anesthesia: blood loss, blood transfusion, and the risk of postoperative renal dysfunction. J Am Coll Surg 1998; 187: 620–5. 22. Smyrniotis V, Kostopanagiotou G, Theodoraki K, et al. The role of central venous pressure and type of vascular control in blood loss during major liver resections. Am J Surg 2004; 187: 398–402. 23. Choukèr A, Schachtner T, Schauer R, et al. Effects of Pringle manoeuvre and ischaemic preconditioning on haemodynamic stability in patients undergoing elective hepatectomy: a randomized trial. Br J Anaesth 2004; 93:204–11. 24. Liu CM, Chen J, Wang XH. Requirements for transfusion and postoperative outcomes in orthotopic liver transplantation: a meta-analysis on aprotinin. World J Gastro 2008; 14: 1425–9. 25. Pereboom IT, de Boer MT, Porte RJ, et al. Aprotinin and nafamostat mesilate in liver surgery: effects on blood loss. Dig Surg 2007; 24: 282–7. 26. Wu CC, Ho WM, Cheng SB, et al. Perioperative parenteral tranexamic acid in liver tumour transection: a prospective randomised trial toward a “blood-transfusion”-free hepatectomy. Ann Surg 2006; 243: 173–80. 27. Lodge JP, Jonas S, Oussoultzoglou E, et al. Recombinant coagulation factor VIIa in major liver resection: a randomised placebo-controlled double-blind clinical trial. Anesthesiology 2005; 102:269–75. 28. Shao YF, Yang JM, Chau GY, et al. Safety and hemostatic effect of recombinant activated factor VII in cirrhotic patients undergoing partial hepatectomy: a multicentre randomised double-blind placebo-controlled trial. Am J Surg 2006; 191: 245–9.

wound complications
The incidence of wound infection was 5.2% in the SloanKettering series, with a further 10 patients (0.5%) having a wound dehiscence (6). The Hong Kong series of 1222 liver resections reports double these complication rates—with 115 patients (9.4%) developing a wound infection and 16 patients (1.3%) suffering wound dehiscence (8). An explanation of the higher incidence of wound complications in the Hong Kong series may be their higher percentage of cirrhotic patients (33% vs. 9%). A study from Japan of 626 liver resections, with a 7.7% incidence of incisional hernias, examined the risk factors for this (90). Risk factors included the type of incision, with a reversed T incision having a significantly higher incidence of an incisional hernia (21.7%) compared to midline (6.3%), J-shaped (4.7%), or a right transverse incision with long midline extension (5.4%). Furthermore, postoperative ascites, body mass index, repeat hepatectomy, and steroid use were also significant risk factors. The incidence of reported incisional hernia was 0.2% in our own series, with the two known patients who developed incisional hernias undergoing repair of these at the time of repeat liver resection. We believe this low incidence is related to the method of closure of the J-shaped wound, with a tension-free, 2-layer closure, using a 6:1 suture (looped 0-nylon) to wound–length ratio, as opposed to the traditional 4:1 ratio (85).

conclusions
The safety of elective liver surgery has improved dramatically in the past 30 years, despite ever-widening indications for hepatectomy. However, complications still happen and prevention is the key to minimizing their incidence. When complications do occur, they should be aggressively managed, in a high-dependency environment, by a multidisciplinary team. International consensus regarding definitions of complications and a severity classification is still required.

references
1. Foster JH and Berman MM. Solid liver tumours. Major Problems in Clinical Surgery, Vol. 22. Philadelphia, PA: WB Saunders, 1977: 1–342. 2. Imamura H, Seyama Y, Kokudo N, et al. One thousand fifty-six hepatectomies without mortality in 8 years. Arch Surg 2003; 138:1198–206. 3. Rees M, Tekkis PP, Welsh FK, et al. Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer: a Multifactorial model of 929 patients. Ann Surg 2008; 247: 125–35. 4. Wei AC, Greig PD, Grant D, et al. Survival after hepatic resection for colorectal metastases: a 10-year experience. Ann Surg Oncol 2006; 13: 668–76. 5. Malik HZ, Prasad KR, Halazun KJ, et al. Preoperative prognostic scoring for predicting survival after hepatic resection for colorectal liver metastases. Ann Surg 2007; 246: 806–14.

70

MANAGING COMPLICATIONS OF HEPATECTOMY
29. Noun R, Elias D, Balladur P, et al. Fibrin glue effectiveness and tolerance after elective liver resection: a randomised trial. Hepatogastroenterology 1996; 43: 221–4. 30. Liu M, Lui WY. The use of fibrin adhesive for haemostasis after liver resection. Chinese Medical Journal 1993; 51:19–22. 31. Frilling A, Stavrou GA, Mischinger HJ, et al. Effectiveness of a new carrierbound sealant versus argon beamer as haemostatic agent during liver resection: a randomised prospective trial. Langenbecks Arch Surg 2005; 390: 114–20. 32. Figueras J, Llado L, Miro M, et al. Application of fibrin glue sealant after hepatectomy does not seem justified: results of a randomised study in 300 patients. Ann Surg 2007; 245: 536–42. 33. Erdogan D, Busch ORC, Gouma DJ, et al. Prevention of biliary leakage after partial liver resection using topical haemostatic agents. Dig Surg 2007; 24: 294–9. 34. Shio S, Yazumi S, Ogawa K, et al. Biliary complications in donors for living donor liver transplantation. Am J Gastro 2008; 103:1393–8. 35. Ghobrial RM, Freise CE, Trotter JF, et al. Donor morbidity after living donation for liver transplantation. Gastroenterology 2008; 135: 468–76. 36. Erdogan D, Busch ORC, van Delden OM, et al. Incidence and management of bile leakage after partial liver resection. Dig Surg 2008; 25: 60–6. 37. Ijichi M, Takayama T, Toyoda H, et al. Randomised trial of the usefulness of a bile leakage test during hepatic resection. Arch Surg 2000; 135: 1395–1400. 38. Capussotti L, Ferrero A, Vigano L, et al. Bile leakage and liver resection: where is the risk? Arch Surg 2006; 141:690–4. 39. Lam CM, Lo CM, Liu CL, et al. Biliary complications during liver resection. World J Surg 2001; 25: 1273–6. 40. Yamashita Y, Hamatsu T, Rikimaru T, et al. Bile leakage after hepatic resection. Ann Surg 2001; 233: 45–50. 41. Gertsch P, Vandoni RE, Pelloni A, et al. Localised hepatic ischaemia after liver resection: a prospective evaluation. Ann Surg 2007; 246:958–65. 42. Vauthey JN, Baer HU, Guastella T, et al. Comparison of outcome between extended and nonextended liver resections for neoplasms. Surgery 1993; 114:968–75. 43. Vigano L, Ferrero A, Sgotto E, et al. Bile leak after hepatectomy: predictive factors of spontaneous healing. Am J Surg 2008; 196: 195–200. 44. Bhattacharjya S, Puleston J, Davidson B et al. Outcome of early endoscopic biliary drainage in the management of bile leaks after hepatic resection. Gastrointest Endosc 2003; 57:526–30. 45. Binmoeller KF, Katon RM, Shneidman R. Endoscopic management of postoperative biliary leaks: review of 77 cases and report of two cases with biloma formation. Am J Gastroenterol 1991; 86:227–31. 46. Yanaga K, Kanematsu T, Takenaka K, et al. Intraperitoneal septic complications after hepatectomy. Ann Surg 1986; 203: 148–52. 47. Balzan S, Belghiti J, Farges O, et al. The “50-50 criteria” on postoperative day 5: an accurate predictor of liver failure and death after hepatectomy. Ann Surg 2005; 242: 824–8. 48. Paugam-Burtz C, Janny S, Delefosse D, et al. Prospective validation of the “fifty-fifty” criteria as an early and accurate predictor of death after liver resection in intensive care unit patients. Ann Surg 2009; 249: 124–8. 49. Mullen JT, Ribero D, Reddy SK, et al. Hepatic insufficiency and mortality in 1059 noncirrhotic patients undergoing major hepatectomy. J Am Coll Surg 2007; 204: 854–62. 50. Nordlinger B, Guiguet M, Vailiant JC, et al. Surgical resection of colorectal carcinoma metastases to the liver: a prognostic scoring system to improve case selection based on 1568 patients. Cancer 1996; 77: 1254–62. 51. Van den Broek MA, Damink SW, Dejong CH, et al. Liver failure after partial hepatectomy: definition, pathophysiology, risk factors and treatment. Liver Int 2008; 28: 767–80. 52. Shirabe K, Shimada M, Gion T, et al. Postoperative liver failure after major hepatic resection for hepatocellular carcinoma in the modern era with special reference to remnant liver volume. J Am Coll Surg 1999; 188: 304–9. 53. Schindl MJ, Redhead DN, Fearon KC, et al. The value of residual liver volume as a predictor of hepatic dysfunction and infection after major liver resection. Gut 2005; 54: 289–96. 54. Imamura H, Sano K, Sugawara Y, et al. Assessment of hepatic reserve for indication of hepatic resection: decision tree incorporating indocyanine green test. J Hepatobiliary Pancreat Surg 2005; 12:16–22. 55. Pugh RNH, Murray-Lyon IM, Dawson JL, et al. Transection of the esophagus for bleeding oesophageal varices. Br J Surg 1973; 60: 646–9. 56. Schneider PD. Preoperative assessment of liver function. Surg Clin N Am 2004; 84: 355–73. 57. Poon RT, Fan ST, Lo CM, et al. Extended hepatic resection for hepatocellular carcinoma in patients with cirrhosis: is it justified? Ann Surg 2002; 236: 602–11. 58. Vauthey JN, Chaoui A, Do KA, et al. Standardised measurement of the future liver remnant prior to extended resection: methodology and clinical associations. Surgery 2000; 127: 512–9. 59. Ferrero A, Vigano L, Polastri R, et al. Postoperative liver dysfunction and future remnant liver: where is the limit? Results of a prospective study. World J Surg 2007; 31:1643–51. 60. Zorzi D, Laurent A, Pawlik TM, et al. Chemotherapy-associated hepatotoxicity and surgery for colorectal liver metastases. Br J Surg 2007; 94: 274–86. 61. Vauthey JN, Pawlik TM, Ribero D, et al. Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases. J Clin Oncol 2006; 24: 2065–72. 62. Abdalla EK, Hicks ME, Vauthey JN. Portal vein embolization: rationale, technique and future prospects. Br J Surg 2001; 88: 165–75. 63. Farges O, Belghiti J, Kianmanesh R, et al. Portal vein embolization before right hepatectomy: prospective clinical trial. Ann Surg 2003; 237: 208–17. 64. Kokudo N, Makuuchi M. Current role of portal vein embolization/ hepatic artery chemoembolization. Surg Clin N Am 2004; 84: 643–57. 65. Abulkhir A, Limongelli P, Healey AJ, et al. Preoperative portal vein embolization for major liver resection: a meta-analysis. Ann Surg 2008; 247: 49–57. 66. Abdalla EK, Adam R, Bilchik AJ, et al. Improving resectability of hepatic colorectal metastases; expert consensus statement. Ann Surg Oncol 2006; 13: 1271–80. 67. Ogata S, Kianmanesh R, Belghiti J. Doppler assessment after right hepatectomy confirms the need to fix the remnant left liver in the anatomical position. Br J Surg 2005; 92: 592–5. 68. Scatton O, Plasse M, Dondero F, et al. Impact of localised congestion related to venous deprivation after hepatectomy. Surgery 2008; 143:483–9. 69. Moore KP, Aithal GP. Guidelines on the management of ascites in cirrhosis. Gut 2006; 55 Suppl 6: 1–12. 70. Mirouze D, Zipser RD, Reynolds TB. Effect of inhibitors of prostaglandin synthesis on induced diuresis in cirrhosis. Hepatology 1983; 3:50–5. 71. Planas R, Arroyo V, Rimola A, et al. Acetylsalicylic acid suppresses the renal haemodynamic effect and reduces the diuretic action of furosemide in cirrhosis with ascites. Gastroenterology 1983; 84: 247–52. 72. Van de Kerkhove MP, de Jong KP, Rijken AM, et al. MARS treatment in post hepatectomy liver failure. Liver Int 2003; 23 Suppl 3: 44–51. 73. Liu JP, Gluud LL, Als-Nielsen D, et al. Artificial and bioartificial support systems for liver failure. Cochrane Database of Systematic Reviews 2004; Issue 1. Art No.: CD003628. DOI: 10.1002/14651858.CD003628. 74. Yanaga K, Kanematsu T, Takenaka K, et al. Intraperitoneal septic complications after hepatectomy. Ann Surg 1986; 203: 148–52. 75. Togo S, Matsuo K, Tanaka K, et al. Perioperative infection control and its effectiveness in hepatectomy patients. J Gastroent Hep 2007; 22: 1942–8. 76. Wu CC, Yeh DC, Lin MC, et al. Prospective randomised trial of systemic antibiotics in patients undergoing liver resection. Br J Surg 1998; 85: 489–93. 77. Paquet JC, Dziri C, Hay JM, et al. Prevention of deep abdominal complications with omentoplasty on the raw surface after hepatic resection. The French Association for Surgical Research. Am J Surg 2000; 179: 103–9. 78. Uetsuji S, Kwon AH, Komada H, et al. Clinical evaluation of closed suction drainage following hepatectomy. Surgery Today 1997; 27: 298–301. 79. Fong Y, Brennan MF, Brown K, et al. Drainage is unnecessary after elective hepatic resection. Am J Surg 1996; 171: 158–62. 80. Liu CL, Fan ST, Lo CM, et al. Abdominal drainage after hepatic resection is contraindicated in patients with chronic liver diseases. Ann Surg 2004; 239: 194–201. 81. Rodgers A, Walker N, Schug S, et al. Reduction of postoperative mortality and morbidity with epidural or spinal anaesthesia: results from overview of randomised trials. Br Med J 2000; 321: 1493. 82. Lui S, Carpenter RL, Neal JM. Epidural anaesthesia and analgesia. Their role in postoperative outcome. Anaesthesiology 1995; 82: 1474–506.

71

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
83. Matot I, Scheinin O, Eid A, et al. Epidural anesthesia and analgesia in liver resection. Anesth Analg 2002; 95:1179–81. 84. Page A, Rostad B, Staley CA, et al. Epidural analgesia in hepatic resection. J Am Coll Surg 2008; 206: 1184–92. 85. Basu S, Taamijmarane A, Bulters D, et al. An alternative method of wound pain control following hepatic resection: a preliminary study. HPB 2004; 6: 186–9. 86. Oliver JC, Welsh FKS, Bell J, et al. Elective intra-aortic balloon counterpulsation during a high-risk liver resection. Anaesth 2008; 63: 1365–8. 87. Saner F. Kidney failure following liver resection. Transplant Proc 2008; 40: 1221–4. 88. Abuelo JG. Normotensive ischemic acute renal failure. N Engl J Med 2007; 357: 797–805. 89. Mehta RL, Kellum JA, Shah SV, et al. Acute kidney injury network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007; 11: R31. 90. Togo S, Nagano Y, Masumoto C, et al. Outcome of and risk factors for incisional hernia after partial hepatectomy. J Gastrointest Surg 2008; 12: 1115–20.

72

8

Pancreatic resection Thilo Hackert, Moritz Wente, and Markus W. Büchler
dissection of the hepatoduodenal ligament, the lymph nodes along the common hepatic artery, portal vein, and the cranial portion of superior mesenteric vein as well as dissection of the right-sided lymph nodes of the celiac trunk and along the right side of the superior mesenteric artery (Fig. 8.2). Today, we have good evidence that there is no benefit for a more extended approach of lymphadenectomy. Meta-analysis from four randomized controlled clinical trials has shown no survival benefit after extended lymph node dissection but has demonstrated a significant increase in surgical morbidity (8). Completion of the dissection can then be done from the infracolic aspect by removing the first jejunal loop (20–25 cm) to ensure tensionless mobility of the next loop that is used for the following reconstruction and is transposed into the right upper quadrant transmesocolically. One of the most important operative steps to prevent severe postoperative complications is the pancreaticojejunostomy. We prefer to perform this anastomosis end-to-side in a twolayer fashion stitching the pancreatic duct separately (Fig. 8.3). Using this technique, insufficiency rates of less than 3.5% can be achieved (9,10). Bile duct reconstruction should be standardized as well to avoid leakage or postoperative bile collections. Although this complication is less frequent than pancreatic fistula, it may cause severe and long-lasting complications. An approach that can be performed even in technically challenging situations with small and deep ducts is the single-stitch distant suture of the posterior wall by a one-layer technique completed by single stitches of the anterior wall. Finally, an end-to-side duodenojejunostomy completes the reconstruction. Recent studies have shown that an antecolic reconstruction is much more favorable in terms of delayed gastric emptying (1,12). Drain placement seems to be another essential step at the end of the operation as there has been growing evidence that pancreatic leakage can be recognized and severe complications caused by intra-abdominal pancreatic fluid collections can be prevented by adequate drain positions. As there may be need for a long-lasting maintenance of intra-abdominal drains in case of fistulas, soft silicon drains should favorably be used. Drain removal—which can usually be done 48 hours postoperatively—should be preceded by analysis of pancreatic enzyme levels in the drain fluid. Amylase levels of more than 5000 iU/ml seem to represent a cutoff value for the recognition of pancreatic fistulas and should therefore be respected carefully (13–17). Distal Pancreatectomy Distal pancreatectomy is performed for tumors in the body or tail of the pancreas and includes—depending on the dignity of the underlying tumor—total splenectomy. From the surgical point of view, tumors above or on the left side of the superior

background
Pancreatic cancer remains—with an overall long-term survival rate of less than 1%—one of the most difficult cancers to treat. It is the fourth leading cause of cancer-related mortality in the Western world and is responsible for around 30,000 deaths per year in the United States and 65,000 per year in Europe (1,2). In only 10% to 20% of pancreatic cancer patients potentially curative surgery is possible, and even in these patients, the median survival is only 10 to 18 months with 5-year survival rates of approximately 20% to 25% (3,4). Nonetheless, surgery remains the only treatment option with the chance of cure. Pancreatic surgery has significantly changed during the past few years. Irrespectively, pancreas resections remain an intervention of particular significance, often technically challenging and with high logistic demands for preoperative diagnostics and perioperative management. Recently, the value of centralization of pancreatic surgery in “high volume institutions” has been demonstrated. The current mortality rates following pancreatic resections are well below 5% in specialized surgical centers (5,6).

standard resections
Whipple Resection Partial Pancreaticoduodenectomy (Whipple resection) with or without distal stomach resection is the surgical option for tumors of the pancreatic head, which account for the majority of pancreatic cancers (Fig. 8.1). Pylorus-preserving pancreaticoduodenectomy has been proven to be equal to the classical pancreaticoduodenectomy in terms of tumor recurrence or long-term survival, and should therefore be considered the standard procedure for tumors of the pancreatic head (7). Key steps of the surgical procedure are the postpyloric division of the duodenum, which is usually carried out by use of a stapling device and—meanwhile common in many centers—the supracolic division of the ascending duodenum as soon as this portion is reached during resection. This modification facilitates the resection procedure and allows manual control of the pancreatic head without switching positions between the supra- and infracolic department. Division of the pancreas is done sharply above the superior mesenteric vein after this has been tunneled to make sure that the vein is not injured during resection and that the dissection can be done without vein replacement (see below). After removing the specimen, tumor-free resection margins should be confirmed intraoperatively by frozen sections of the cut end of the bile duct and the cut end of the pancreatic remnant. Bleeding control along the pancreatic dissection margin is achieves by carefully stitching single bleeding sites with monofilament and nonabsorbable sutures. The pancreatic duct must be seen and protected during this procedure. During pancreaticoduodenectomy, a standardized lymphadenectomy needs to be carried out. This includes the complete

73

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 8.1 Partial pancreaticoduodenectomy. Classical Whipple resection (left) and pylorus-preserving modification (right).

Figure 8.2 Intraoperative situs after partial pancreaticoduodenectomy. Pancreatic remnant with probe introduced, dissected portal vein and hepatic artery. A jejunal loop is prepared for the pancreaticojejunostomy.

mesenteric vein are suitably located for this procedure. Dissection of the pancreas is performed above the vein after tunneling and lifting up the body of the gland. The dissection itself can be done sharply or by using a stapling device, preferably with a thickness-adopted adjustment of the stapler. To date, there are no high-power studies to support either procedure. In case of sharp dissection, we prefer a V-shaped transection line. As in other resections, tumor-free resection margins should be examined by intraoperative frozen section. The pancreatic duct is separately closed by a monofilament Z-shaped nonabsorbable suture and the transection line can afterward be closed by single stitches covering the complete margin by pancreatic capsular tissue. There is no need or evidence for any further covering of the resection margin by sealants or patches (18). This procedure implies a certain limitation concerning the extent of distal resection toward the head of the

pancreas. The larger the tissue area of the transected parenchyma gets, the more difficult it gets to close the parenchyma, which is associated with increased fistula rates. Therefore, the right margin of the superior mesenteric vein represents the limit to which a safe surgical closure of the pancreatic remnant can be performed. In case of transection by a stapling device, this limitation is technically implied by the length of the stapler line. No additional sutures are necessary after stapler dissection. Despite all approaches, fistula development after distal pancreatectomy remains an unsolved problem. Fistula rates range from 12% to 40% (8,19). To address this clinical problem, remnant closure by sutures after sharp dissection is currently compared to stapler dissection in a randomized controlled study (DISPACT trial) in a multicenter approach including 21 European centers and 360 patients by February 2009 (20). A spleen-preserving distal pancreatectomy can be performed in benign lesions or intraductal papillary mucinous neoplasias (IPMNs), if the splenic vessels are not involved in the tumor or cystic process. However, there are no clear advantages in preserving the spleen in adult patients (21). Possible advantages could be infection prophylaxis, less operative blood loss, fistula rates as well as fewer thromboembolic complications (22,23). By contrast, the risk of splenic infarction and portal hypertension has to be regarded whenever the spleen is preserved. From the currently available literature— mainly retrospective studies—none of these parameters is clearly proven, further studies have to address this topic in the future. By contrast, there is growing evidence that distal pancreatectomy can be performed with good results laparoscopically. This approach is usually performed using 5 trocars and stapler dissection of the pancreas. It is routinely performed in several centers with results comparable to the open approach in terms of operative morbidity and outcome (23). The possible advantages of laparoscopic operations, with faster patient recovery, less pain medication, and better cosmetic results are currently evaluated in larger series. Total Pancreatectomy The concept of total pancreatectomy has to be divided into the rescue procedure in not conservatively managed postoperative

74

PANCREATIC RESECTION

Figure 8.3 Pancreatico-jejunostomy. Preparation of duct sutures (upper left), position of the jejunal loop (upper right), anterior wall sutures (lower left), and completed anastomosis (lower right).

complications caused by the pancreatic remnant after head resections and the primarily performed total removal of the gland with or without the spleen (24). Completion pancreatectomy may be necessary in case of severe complication like insufficiency of the pancreaticojejunostomy with septic or bleeding complications. In this situation, an early completion operation can be life-saving for the patient and is technically similar to a distal pancreatectomy after disconnection of the pancreas anastomosis (25,26). Primary total pancreatectomy can be required in patients with a nonaltered pancreatic remnant due to the soft tissue texture, e.g., in distal bile dust cancer or duodenal tumors without congestion of the pancreatic duct, which can make the pancreatic anastomosis a dangerous reconstruction. The surgeon has to evaluate the cost–benefit relation carefully; in doubtful situations a risky anastomosis should rather be avoided. From the oncological point of view, extensive main-duct IPMNs, IPMNs with progression to carcinoma, familial or multifocal pancreatic cancer are indications for a primary total pancreatectomy. Furthermore, this procedure may be necessary if a tumor-free resection margin and R0 situation cannot be achieved otherwise (24–29). The resection can be performed as a two-part procedure with an

initial head resection similar to a Whipple procedure followed by the distal resection, which facilitates the surgical preparation, or with a removal of the gland as a complete specimen, if a pancreatic transection implies the risk of tumor cell spilling. Whenever possible, a pylorus-preserving reconstruction should be preferred. Duodenum-Preserving Pancreatic Head Resection The best technique for the surgical treatment of pancreatic head lesions in chronic pancreatitis is still under debate. Partial pancreatoduodenectomy with or without preservation of the pylorus have served for many years as the primary surgical procedure. However, these resections are unsatisfactory in terms of late morbidity with an incidence of up to 48% of postoperative diabetes mellitus (30). Today, duodenumpreserving pancreatic head resection duodenum-preserving pancreatic head resection (DPPHR), which was introduced by Beger in 1972 (31), has undergone several modifications and is considered the standard procedure for nonmalignant head lesions in chronic calcified pancreatitis (32). Whenever possible, depending on the extent of the calcified and fibrotic lesions, the Berne modification as the most tissue-sparing

75

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
approach should be performed (Fig. 8.4). The surgical procedure starts with an extensive Kocher maneuver of the pancreatic head to palpate the head of the pancreas and achieve bleeding control by compression during the resection phase. The anterior aspect of the head should be prepared under dissection of the right gastroepiploic vessels and ligation of the gastroduodenal artery to minimize blood loss during excision of the head. It is not necessary to tunnel the pancreas above the mesenteric vein, especially as this is often difficult due to the chronic inflammatory adherence of the parenchyma. The resection margin should be defined by circular sutures around the altered tissue area. Afterward, the head is sharply excised manually to control bleeding and perforation of the posterior parenchyma layer. All fibrotic and calcified tissue should be removed and the pancreatic duct has to be opened and inspected to extract stones and ensure free drainage into the resection cavity. Special attention has to be paid to the bile duct. In case of preoperative cholestasis and/or preceding stents, the bile duct needs to be opened by a T-shaped incision and the orifice should be fixed in the resection cavity to avoid postoperative recurrence of bile duct stenosis (33). Hemostasis in the resection cavity is achieved by selective single stitches with nonabsorbable sutures. The operation is completed by an anastomosis with a Roux-Y-transected jejunal loop in a side-to-side fashion by a two-layer running suture (Fig. 8.5). As in all other resections, drainage placement is important to monitor postoperative secretion and recognize possible fistula development soon. the DPPHR procedure is widely accepted nowadays and has proven to be equally efficient as the Whipple procedure in terms of long-term pain relief, overall morbidity and mortality combined with significantly less intraoperative blood replacement, shorter hospital stay, more postoperative weight gain, less exocrine insufficiency, better occupational rehabilitation, and quality of life in randomized controlled trials and a recent meta-analysis (32–37). Segmental Resection Segmental resections of the pancreas can be performed in benign lesions located in the body of the gland (38). Surgical technique includes a careful mobilization of the pancreatic segment under clipping of vessels followed by sharp dissection of the defined segment. Afterward, reconstruction was done by two-layer sutured anastomosis toward the tail of the pancreas similar to the Whipple anastomosis and V-shaped closure of the dissected margin toward the pancreatic head comparable to the left resection technique. In case of extended resections toward the head leading to a large resection margin, this can additionally be sealed with a seromuscular patch using the jejunal loop that has been anastomosed onto the pancreatic tail before. No fibrin glue or other sealants are required. At present, fistula rates between 8% and 63% are reported, which shows the heterogeneity of the present studies (38–41). However, a surgical mortality of 2% shows that segmental resections can be performed safely and offers a useful tissue-sparing tool in selected patients. Enucleation Especially benign tumors, cystic lesions or IPMNs do not necessarily require extensive pancreatic resections to

Figure 8.4 Pylorus-preserving pancreatic head resection (Berne modification). Note the incision and fixation of the bile duct in the resection cavity.

Figure 8.5 Pylorus-preserving pancreatic head resection (Berne modification). Resection cavity with first layer of the posterior wall of the pancreaticojejunostomy (left), completed anastomosis (right).

76

PANCREATIC RESECTION
achieve surgical cure. Limited resections represent a tissuesparing treatment option to minimize the risk of exocrine or endocrine pancreatic insufficiency postoperatively (42) and to reduce surgical morbidity and mortality by reduced operative trauma. One of the most important aspects to perform an enucleation successfully is the accurate localization of the tumor or cystic lesion. Besides preoperative localization by CT or MRI scan, the most important tool for tumor location is the experience of the surgeon performing the exploration (43–46). Mobilization of the pancreas is essential when tumors or cystic lesions have to be located to enable a careful digital examination of the suspected lesion. This should be supplemented by intraoperative ultrasound to exclude multifocal tumorous lesions especially in endocrine tumors or IPMNs. In addition, a possible relation to the pancreatic duct can only be clarified by ultrasound examination, if there is any doubt about it intraoperatively (47). A tumor size of 2.5 cm in diameter can be regarded as the limit for a safely performed enucleation. Tumors measuring more than 2.5 cm in size show malignant histological changes significantly more frequently, making a local surgical approach impossible. Besides, tissue trauma and wound surface following an enucleation reach a critical size for development of fistulas or other complications including bleeding or postoperative pancreatitis (47). Enucleation itself is performed by careful dissection along the tumor under clip ligation or stitching of vessels supplying the lesion (Fig. 8.6). There is no evidence for any sealant or glue application after completing of the enucleation. Drain placement is essential as currently fistula rates of approximately 20% are reported (48), most of them, however, clinically uncomplicated.

exceptional indications
Vessel Resections A common problem in pancreatic head resections is tumor adherence to the superior mesenteric or portal vein. Today, portal vein resection has become an established procedure and can be carried out with morbidity rates of that are comparable to standard Whipple procedures (49–56). Portal vein resection can be performed as a tangential resection with a direct suture or a patch reconstruction. In cases where a segmental resection is required due to a more extensive tumor adherence, either a direct anastomosis or the interposition of an autologous venous graft such as the saphenous vein or an allograft, e.g., a gore-tex tube. In case of a primary anastomosis, it is essential to mobilize the mesenteric root completely, which implies the complete mobilization of the right hemicolon. After this preparation, a tension-free reconstruction of defects up to 3 cm length is usually possible. The anastomosis is performed as a running suture of the posterior and anterior vessel wall with two 5-0 or 6-0 nonabsorbable sutures. When defects cannot be reconstructed by the patient’s vein alone, a size-adopted graft should be inserted in a similar end-to-end manner (52). Kinking of any venous anastomosis must be avoided to prevent intra- and postoperative vein or graft thrombosis with consecutive failure of the bowel circulation. In certain situations, it may be helpful not only to minimize the time of intraoperative occlusion of the mesenteric/portal vein but also to clamp the superior mesenteric artery for this period to avoid venous congestion and swelling of the small bowel and the right hemicolon (Fig. 8.7). Arterial resection is a rather uncommon surgical procedure during pancreatic cancer resection. If the superior mesenteric artery is involved in the tumor process, this is a general exclusion criterion for resection and has only been reported in few patients (56). By contrast, tumor adherence or infiltration along the celiac axis must not be considered as generally irresectable (43,53). In selected patients, the celiac trunk might be resected down to its aortic orifice in Whipple as well as in left resection or total pancreatectomies (54–56). As long as the proper hepatic artery can be preserved, a reconstruction is possible. The left gastric and splenic artery can usually be cut without reconstruction, a consecutive

Figure 8.6 Tumor enucleation in the body of the pancreas.

Figure 8.7 Examples of portal vein resections. Direct end-to-end anastomosis (left) and graft implantation (right).

77

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
splenectomy may be necessary in some patients. Restoration of the hepatic perfusion must be ensured by re-anastomosing the proper or common hepatic artery. This reconstruction can be done with an interposition of any arterial vessel of the celiac axis or a venous interposition graft. However, the arterial perfusion of the liver should be controlled by regular duplex examinations and restored aggressively in case of a vessel occlusion. Arterial hepatic perfusion failure may otherwise cause acute problems postoperatively in terms of liver ischemia, necrosis, and infection and is a risk factor for bile duct-associated complications in the long-term follow-up (54,55). Yet, it needs to be mentioned that there are no larger patient series on arterial resections in pancreatic surgery. Therefore, this procedure can be carried out safely in experienced hands but is not based on high-quality scientific data and outcome studies so far. Multivisceral Resections There are several studies (57–60) on the outcome after multivisceral resection for pancreatic cancer. In general, resection of adjacent organs, most commonly the stomach or left hemicolon in left resections and the right hemicolon in Whipple procedures as well as either adrenal gland or kidney in both types of resection can be performed safely to achieve a R0 situation. Technically, an en bloc resection should be performed without preparation along or injuring the tumor surface. This may result in “typical” resections such as right or left colectomies as well as individual segmental- or wedge-type resections. Multivisceral approaches can also be combined with vessel resections of the portal vein or the celiac axis. From the limited number of available studies, this approach is associated with an increased intraoperative blood loss and overall surgical morbidity as well as ICU and hospital stay (58,60). However, there seems to be a survival advantage in these patients and overall mortality is not increased compared to standard resections (59,60). Due to the limited number of patients reported so far, it is not possible to give valid data on long-term oncological outcome, making multivisceral resections an individually tailored approach that requires careful patient selection and surgical experience. Recurrence Resections Localized recurrence in pancreatic cancer may be an indication for relaparotomy and resection in selected patients. Although a large number of recurrences are located close to the arterial vessels, and therefore not resectable, recent studies support the concept of surgical exploration and resection whenever possible (61–63). This approach can be combined with intraoperative radiotherapy and radiation of the tumor bed to reduce the risk of another recurrence at the site of resection (Fig. 8.8). In case of local irresectability, intraoperative radiation can be performed with a palliative intention in terms of tumor reduction and pain control. An extended resection of the recurrent tumor with arterial vessels does not seem to be justified as the chance for a radical tumor removal is poor and patients do not seem to benefit from R1 or R2 resections. The available studies report successful resection rates of approximately 50% with acceptable surgical morbidity and suggest a survival benefit for those patients, especially in situations with a long time interval (>9–12 months) between the initial tumor diagnosis and the recurrence manifestation (63). As these are observational studies, there is no proven evidence for this approach today and larger controlled trials are required to evaluate long-term oncological value. Metastasis Resections Resection for metastatic pancreatic cancer is clearly restricted to exceptional indication and has only been reported anecdotally so far (64,65). Most commonly, the indication for metastasis resection arises in young patients with the accidental finding of a synchronous single liver lesion intraoperatively, which can be removed without increasing operative morbidity (64). Apart from this individual indication, metastasis resection can be performed in long-term survivors with

Figure 8.8 Pancreatic cancer recurrence resection. Intraoperative finding of the recurrence located in the interaortocaval space (left), situs after resection (right) prior to intraoperative radiation.

78

PANCREATIC RESECTION
localized metastastic disease, indicating favorable tumor biology and justifying the aggressive operative approach. This has to be embedded in a global oncological concept, must be decided highly individually and cannot be regarded as a standard procedure (65).
13. Pedrazzoli S, Liessi G, Pasquali C, et al. Postoperative pancreatic fistulas: Preventing severe complications and reducing reoperation and mortality rate. Ann Surg 2009; 249(1): 97–104. 14. Veillette G, Dominguez I, Ferrone C, et al. Implications and management of pancreatic fistulas following pancreaticoduodenectomy: The Massachusetts General Hospital experience. Arch Surg 2008; 143(5): 476–81. 15. Reid-Lombardo KM, Farnell MB, Crippa S, et al. Pancreatic anastomotic leakage after pancreaticoduodenectomy in 1,507 patients: A report from the Pancreatic Anastomotic Leak Study Group. J Gastrointest Surg 2007; 11(11): 1451–8; discussion 1459. Epub 2007 Aug 21. 16. Molinari E, Bassi C, Salvia R, et al. Amylase value in drains after pancreatic resection as predictive factor of postoperative pancreatic fistula: Results of a prospective study in 137 patients. Ann Surg 2007; 246(2): 281–7. 17. Bassi C, Dervenis C, Butturini G, et al. Postoperative pancreatic fistula: An international study group (ISGPF) definition. Surgery 2005; 138(1): 8–13. 18. Knaebel HP, Diener MK, Wente MN, et al. Systematic review and metaanalysis of technique for closure of the pancreatic remnant after distal pancreatectomy. Br J Surg 2005; 92(5): 539–46. 19. Andrén-Sandberg A, Wagner M, Tihanyi T, et al. Technical aspects of leftsided pancreatic resection for cancer. Dig Surg 1999; 16(4): 305–12. 20. Diener MK, Knaebel HP, Witte ST, et al. DISPACT trial: A randomized controlled trial to compare two different surgical techniques of DIStal PAnCreaTectomy—study rationale and design. Clin Trials 2008; 5(5): 534–45. 21. Fernández-Cruz L, Orduña D, Cesar-Borges G, Angel López-Boado M. Distal pancreatectomy: En-bloc splenectomy vs spleen-preserving pancreatectomy. HPB (Oxford) 2005; 7(2): 93–8. 22. Kimura W, Moriya T, Ma J, et al. Spleen-preserving distal pancreatectomy with conservation of the splenic artery and vein. World J Gastroenterol 2007; 13(10): 1493–9. 23. Distal pancreatectomy: radical or spleen-preserving? Chromik AM, Janot M, Sülberg D, Seelig MH, Uhl W. Chirurg 2008; 79(12): 1123–33. 24. Keck T, Hopt UT. Total pancreatectomy: Renaissance of a surgical procedure. Chirurg 2008; 79(12): 1134–40. 25. Wente MN, Shrikhande SV, Kleeff J, et al. Management of early hemorrhage from pancreatic anastomoses after pancreaticoduodenectomy. Dig Surg 2006; 23(4): 203–8. Epub 2006 Jul 26. 26. Büchler MW, Wagner M, Schmied BM, et al. Changes in morbidity after pancreatic resection: toward the end of completion pancreatectomy. Arch Surg 2003; 138(12): 1310–4; discussion 1315. 27. Sohn TA, Yeo CJ, Cameron JL, et al. Intraductal papillary mucinous neoplasms of the pancreas: an updated experience. Ann Surg 2004; 239(6): 788–97; discussion 797–9. 28. Inagaki M, Obara M, Kino S, et al. Pylorus-preserving total pancreatectomy for an intraductal papillary-mucinous neoplasm of the pancreas. J Hepatobiliary Pancreat Surg 2007; 14(3): 264–9. Epub 2007 May 29. 29. Müller MW, Friess H, Kleeff J, Dahmen R, Wagner M, Hinz U, BreischGirbig D, Ceyhan GO, Büchler MW. Is there still a role for total pancreatectomy? Ann Surg 2007; 246(6): 966–74; discussion 974–5. 30. Sakorafas GH, Farnell MB, Nagorney DM, et al. Pancreatoduodenectomy for chronic pancreatitis: long-term results in 105 patients. Arch Surg 2000; 135: 517–23. 31. Beger HG, Buchler M, Bittner RR, et al. Duodenum-preserving resection of the head of the pancreas in severe chronic pancreatitis. Early and late results. Ann Surg 1989; 209: 273–78. 32. Diener MK, Rahbari NN, Fischer L, et al. Duodenum-preserving pancreatic head resection versus pancreatoduodenectomy for surgical treatment of chronic pancreatitis: A systematic review and meta-analysis. Ann Surg 2008; 247(6): 950–61. 33. Cataldegirmen G, Bogoevski D, Mann O, et al. Late morbidity after duodenum-preserving pancreatic head resection with bile duct reinsertion into the resection cavity. Br J Surg 2008; 95(4): 447–52. 34. Büchler MW, Friess H, Muller MW, et al. Randomized trial of duodenumpreserving pancreatic head resection versus pylorus-preserving Whipple in chronic pancreatitis. Am J Surg 1995; 169: 65–9. 35. Farkas G, Leindler L, Daroczi M, et al. Prospective randomised comparison of organ-preserving pancreatic head resection with pylorus-preserving pancreaticoduodenectomy. Langenbecks Arch Surg 2006; 391: 338–42.

conclusion
Pancreatic surgery has undergone a remarkable development during the last decades. Appropriate surgical approaches have been established and can be used in differential indications today. In pancreatic cancer, standard resections include the classical Whipple operation and the pylorus-preserving modification, which should be preferred whenever possible as well as a distal or total pancreatectomy in extended tumors of the gland. All of these procedures can be carried out safely with surgical mortality rates well below 5% in specialized centers due to a high grade of standardization and experience. Modern tissue-sparing procedures such as the duodenum-preserving pancreatic head resection in chronic pancreatitis or tumor enucleations offer limited approaches for circumscribed nonmalignant pancreatic pathologies. Furthermore, extended resections for the treatment of pancreatic malignancies— including multivisceral and recurrence resections—are technically feasible although the oncological outcome of these procedures has to be further evaluated and pancreatic cancer treatment must always be embedded in an interdisciplinary concept of surgery and adjuvant therapy to ensure best possible outcome.

references
1. Jemal A, Siegel R, Ward E, et al. Cancer statistics. CA Cancer J Clin 2007; 57(1): 43–66. 2. Hariharan D, Saied A, Kocher HM. Analysis of mortality rates for pancreatic cancer across the world. HPB (Oxford) 2008; 10(1): 58–62. 3. Wagner M, Redaelli C, Lietz M, et al. Curative resection is the single most important factor determining outcome in patients with pancreatic adenocarcinoma. Br J Surg 2005; 91: 586–94. 4. Schnelldorfer T, Ware AL, Sarr MG, et al. Long-term survival after pancreatoduodenectomy for pancreatic adenocarcinoma: Is cure possible? Ann Surg 2008; 247(3): 456–62. 5. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002; 346(15): 1128–37. 6. McPhee JT, Hill JS, Whalen GF, et al. Perioperative mortality for pancreatectomy: A national perspective. Ann Surg 2007; 246(2): 246–53. 7. Diener MK, Knaebel HP, Heukaufer C, et al. A systematic review and meta-analysis of pylorus-preserving versus classical pancreaticoduodenectomy for surgical treatment of periampullary and pancreatic carcinoma. Ann Surg 2007; 245(2): 187–200. 8. Michalski CW, Kleeff J, Wente MN, et al. Systematic review and metaanalysis of standard and extended lymphadenectomy in pancreaticoduodenectomy for pancreatic cancer. Br J Surg 2007; 94(3): 265–73. 9. Büchler MW, Wagner M, Schmied BM, et al. Changes in morbidity after pancreatic resection: Toward the end of completion pancreatectomy. Arch Surg 2003; 138(12): 1310–4; discussion 1315. 10. Wente MN, Shrikhande SV, Kleeff J, et al. Management of early hemorrhage from pancreatic anastomoses after pancreaticoduodenectomy. Dig Surg 2006; 23(4): 203–8. Epub 2006 Jul 26. 11. Hartel M, Wente MN, Hinz U, et al. Effect of antecolic reconstruction on delayed gastric emptying after the pylorus-preserving Whipple procedure. Arch Surg 2005; 140(11): 1094–9. 12. Tani M, Terasawa H, Kawai M, et al. Improvement of delayed gastric emptying in pylorus-preserving pancreaticoduodenectomy: Results of a prospective, randomized, controlled trial. Ann Surg 2006; 243(3): 316–20.

79

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
36. Klempa I, Spatny M, Menzel J, et al. Pancreatic function and quality of life after resection of the head of the pancreas in chronic pancreatitis. A prospective, randomized comparative study after duodenum preserving resection of the head of the pancreas versus Whipple’s operation. Chirurg 1995; 66: 350–9. 37. Izbicki JR, Bloechle C, Knoefel WT, et al. Duodenum-preserving resection of the head of the pancreas in chronic pancreatitis. A prospective, randomized trial. Ann Surg 1995; 221: 350–8. 38. Müller MW, Friess H, Kleeff J, et al. Middle segmental pancreatic resection: An option to treat benign pancreatic body lesions. Ann Surg 2006; 244: 909–18; discussion 918–20. 39. Bassi C. Middle segment pancreatectomy: A useful tool in the management of pancreatic neoplasms. J Gastrointest Surg 2007; 11: 726–9. 40. Adham M, Giunippero A, Hervieu V, Courbière M, Partensky C. Central pancreatectomy: Single-center experience of 50 cases. Arch Surg 2008; 143: 175–180; discussion 180–1. 41. Christein JD, Smoot RL, Farnell MB. Central pancreatectomy: A technique for the resection of pancreatic neck lesions. Arch Surg 2006; 141: 293–9. 42. Fang WL, Su CH, Shyr YM, et al. Functional and morphological changes in pancreatic remnant after pancreaticoduodenectomy. Pancreas 2007; 35: 361–5. 43. Chung JC, Choi SH, Jo SH, et al. Localization and surgical treatment of the pancreatic insulinomas. ANZ J Surg 2006; 76: 1051–5. 44. Ritzel RA, Isermann B, Schilling T, et al. Diagnosis and localization of insulinoma after negative laparotomy by hyperinsulinemic, hypoglycaemic clamp and intra-arterial calcium stimulation. Rev Diabet Stud 2004; 1: 42–6. 45. Kisker O, Bastian D, Bartsch D, Nies C, Rothmund M. Localization, malignant potential, and surgical management of gastrinomas. World J Surg 1998; 22: 651–7; discussion 657–8. 46. Schmitz-Winnenthal FH, Z’graggen K, Volk C, Schmied BM, Büchler MW. Intraductal papillary mucinous tumors of the pancreas. Curr Gastroenterol Rep 2003; 5: 133–40. 47. Lee CJ, Scheiman J, Anderson MA, et al. Risk of malignancy in resected cystic tumors of the pancreas < or =3 cm in size: Is it safe to observe asymptomatic patients? A multi-institutional report. J Gastrointest Surg 2008; 12: 234–42. 48. Crippa S, Bassi C, Salvia R, et al. Enucleation of pancreatic neoplasms. Br J Surg 2007; 94: 1254–9. 49. Fuhrman GM, Leach SD, Staley CA, et al. Rationale for en bloc vein resection in the treatment of pancreatic adenocarcinoma adherent to the superior mesenteric-portal vein confluence. Pancreatic Tumor Study Group. Ann Surg 1996; 223(2): 154–62. 50. Hartel M, Niedergethmann M, Farag-Soliman M, et al. Benefit of venous resection for ductal adenocarcinoma of the pancreatic head. Eur J Surg 2002; 168(12): 707–12. 51. Harrison LE, Klimstra DS, Brennan MF. Isolated portal vein involvement in pancreatic adenocarcinoma. A contraindication for resection? Ann Surg 1996; 224(3): 342–7; discussion 347–9. 52. Weitz J, Kienle P, Schmidt J, Friess H, Büchler MW. Portal vein resection for advanced pancreatic head cancer. J Am Coll Surg 2007; 204(4): 712–6. Epub 2007 Feb 26. 53. Hartel M, Wente MN, Di Sebastiano P, Friess H, Büchler MW. The role of extended resection in pancreatic adenocarcinoma: is there good evidence-based justification? Pancreatology 2004; 4(6): 561–6. Epub 2004 Nov 15. 54. Martin RC 2nd, Scoggins CR, Egnatashvili V, et al. Arterial and venous resection for pancreatic adenocarcinoma: operative and long-term outcomes. Arch Surg 2009; 144(2): 154–9. 55. Nakao A, Takeda S, Inoue S, et al. Indications and techniques of extended resection for pancreatic cancer.World J Surg 2006; 30(6): 976–82; discussion 983–4. 56. Yekebas EF, Bogoevski D, Cataldegirmen G, et al. En bloc vascular resection for locally advanced pancreatic malignancies infiltrating major blood vessels: Perioperative outcome and long-term survival in 136 patients. Ann Surg 2008; 247(2): 300–9. 57. Sasson AR, Hoffman JP, Ross EA, et al. En bloc resection for locally advanced cancer of the pancreas: is it worthwhile? J Gastrointest Surg 2002; 6(2): 147–57; discussion 157–8. 58. Shoup M, Conlon KC, Klimstra D, Brennan MF. Is extended resection for adenocarcinoma of the body or tail of the pancreas justified? J Gastrointest Surg 2003; 7(8): 946–52; discussion 952. 59. Imamura M, Doi R. Treatment of locally advanced pancreatic cancer: Should we resect when resectable? Pancreas 2004; 28(3): 293–5. 60. Kleeff J, Diener MK, Z’graggen K, et al. Distal pancreatectomy: risk factors for surgical failure in 302 consecutive cases. Ann Surg 2007; 245(4): 573–82. 61. Meyers MO, Meszoely IM, Hoffman JP, et al. Is reporting of recurrence data important in pancreatic cancer? Ann Surg Oncol 2004; 11(3): 304–9. 62. Shibata K, Matsumoto T, Yada K, et al. Factors predicting recurrence after resection of pancreatic ductal carcinoma. Pancreas 2005; 31(1): 69–73. 63. Kleeff J, Reiser C, Hinz U, et al. Surgery for recurrent pancreatic ductal adenocarcinoma. Ann Surg 2007; 245(4): 566–72. 64. Shrikhande SV, Kleeff J, Reiser C, et al. Pancreatic resection for M1 pancreatic ductal adenocarcinoma. Ann Surg Oncol 2007; 14(1): 118–27. Epub 2006 Oct 25. 65. Gleisner AL, Assumpcao L, Cameron JL, et al. Is resection of periampullary or pancreatic adenocarcinoma with synchronous hepatic metastasis justified? Cancer 2007; 110(11): 2484–92.

80

9

Surgical complications of pancreatectomy Steven C. Katz and Murray F. Brennan
Delayed Gastric Emptying The incidence of delayed gastric emptying (DGE) following PD ranges from 4% to 29% (5,11,13) and is associated with other intraabdominal complications (Table 9.1). While DGE is not associated with an increased risk of death, it does prolong hospitalization time (5,20). Parameters used to define DGE include the volume of nasogastric tube output, the length of time before tolerance of oral feeding, and results of scintigraphic studies. At our institution, DGE is defined as failure to achieve oral intake sufficient to maintain adequate hydration by postoperative day 10 (9). DGE is thought to be due to numerous factors, including management of the pylorus, extent of retroperitoneal dissection, intraabdominal fluid collections, and decreased motilin activity (21). Early reports indicated that pylorus-preserving pancreaticoduodenectomy (PPPD) increased the risk of DGE (22,23) but subsequent studies have failed to confirm this (Table 9.4) (24,25). Radical resection or extended retroperitoneal dissection may also be associated with DGE (26). It is unclear if more extensive dissection has a direct effect or if higher rates of pancreatic leak, sepsis, or hemorrhage predispose to DGE (25). Expeditious management of fluid collections, infection, or bleeding may limit gastric dysmotility. An additional contributing factor to DGE may be reduced levels of circulating motilin following PD (27). In a randomized control trial (RCT) including 118 patients undergoing PD, erythromycin, the motilin analogue, reduced the DGE rate from 30% to 19% compared to placebo (21). By contrast, routine nasogastric decompression or withholding of oral feeding has not been shown to affect the rate of DGE. Based upon data from RCTs involving patients subjected to gastrectomy, routine nasogastric tube placement following pancreatic surgery is unnecessary (28,29). Furthermore, early oral feeding should be considered following major abdominal procedures (30,31). Postpancreatectomy Hemorrhage Postpancreatectomy hemorrhage (PPH) occurs in 2% to 9% of cases and the consequences may be severe (8,32–37). The initial evidence of hemorrhage may be the “sentinel bleed,” which is present in 30% to 100% of patients prior to massive PPH (38–41). Risk of PPH is related to inadequate intraoperative hemostasis, bile leak, pancreatic leak, intraabdominal infection, and sepsis (39,40,42–44). The presence of jaundice at the time of pancreatic resection may increase the risk of PPH, but this is not lessened by preoperative biliary drainage (37). The implications and management of PPH vary depending on the time of onset and source (4). Postoperative bleeding within the first 24 hours is most often due to a technical failure and requires reoperation if severe (35). The most appropriate course of action for PPH occurring beyond the immediate postoperative period will

Pancreatic resection and the associated complications remain challenging problems for patients and surgeons. Since the earliest reports describing the technique of pancreaticoduodenectomy (PD) by Kausch and Whipple, significant reductions in operative mortality and morbidity have been achieved (1,2). Postoperative mortality rates have been reduced from greater than 25% in the 1960s to less than 5% in specialized centers (3). The lower risk of death following pancreatic resection is due to advances in operative technique, improvements in perioperative care, percutaneous and endoscopic management of complications, and refinements in patient selection (4). Unfortunately, morbidity rates for PD continue to exceed 30% to 40% in large series (5–9). We discuss the prevalence, nature, predisposing factors, and management for major surgical complications that occur following pancreatic resection. While there are many nonsurgical complications that occur following pancreatic resection, these are not addressed. Right, left, central, and total pancreatectomies are discussed separately where appropriate. The most common individual complications are considered, followed by factors affecting morbidity rates. Throughout, we outline operative strategies and postoperative interventions that impact the risk and severity of surgical complications following pancreatectomy.

specific complications
Pancreatic Anastomotic Leak and Pancreatic Fistula Pancreatic leak occurs in 7% to 29% of patients following pancreatic resection (Tables 9.1 and 9.2) (5,7,10–14). The wide range in incidence is due in part to variability in defining the manifestations of pancreatic leaks and several classification systems have been proposed (9,15,16). Given similarity in management and clinical manifestations, pancreatic leak, fistula, fluid collection, and abscess will be considered together (12). Parenchymal consistency and the extent of operation are associated with pancreatic leak following right or left pancreatectomy (Table 9.3) (17). Small pancreatic duct diameter is a predictor of leak following PD (7,18). Management of fluid collections resulting from a pancreatic leak may involve operative drains, placement of postoperative drains, or reoperation (Fig. 9.1). Vin et al. reported that prolonged drainage was predicted by volume collected during the first 48 hours, fluid amylase >1000, or distal pancreatectomy (12). The magnitude of the pancreatic leak may also depend on whether the source is the main duct or parenchyma (19). Those patients who do develop pancreatic leaks are more likely to suffer from other complications or death, and this risk is exacerbated by superimposed infection (12). Numerous strategies have been attempted to minimize the chances of pancreatic leakage and these are discussed below.

81

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 9.1 Complications Following Pancreaticoduodenectomy
Author Balladur (42) Bottger (10) Gouma (11) Balcom (5) Muscari (7) Winter (13) Vin (12) House (49) Baker (47) N 223 228 300 489 300 1423 680 356 440 Pancreatic Fistula or Leak 13% 8% 7% 13% 17% 9% 18% 15% 16% Delayed Gastric Emptying NR NR 29% 12% NR 15% NR 4% 7% Hemorrhage 9% NR 5% NR 6% NR NR NR 2% Bile Leak NR <1% 2% 2% <1% 2% NR NR 2% Overall Complications 41% 26% 48% 39% 39% 38% NR 38% 36% Death 9% 6.1% 10.1% 11% 2% 2.0%* 1.7% 1.6%

*Includes pancreaticoduodenectomy, central pancreatectomy, and distal pancreatectomy.

Table 9.2 Complications Following Distal Pancreatectomy
Author Bottger (10) Balcom (5) Pannegeon (103) Siergaza (104) Ridolfini (105) Kleeff (50) Ferrone (14) Vin (12) N 72 190 175 132 64 302 462 220 Fistula or Leak 13% 13% 23% 14% 22% 12% 29% 13% Hemorrhage NR NR 2% 4% 3% 3% NR NR Overall Complications 27% 26% 42% 57% 37% 35% NR NR Death 1.7% 1.5% 0 5.0% 1.5% 2.0% 0.8% 2.0%*

*Includes pancreaticoduodenectomy, central pancreatectomy, and distal pancreatectomy.

Table 9.3 Predictors of Pancreatic Leak or Fistula
Following Pancreaticoduodenectomy Small duct diameter (7,10) Friable parenchyma (7,10) Extended resection (7) Placement of intraoperative drains (59) Blood loss (10) Obesity (49) Following Distal Pancreatectomy Multivisceral resection (14,104,105) Proximal (body) transaction (103) Friable parenchyma (105) Malnutrition (104) Obesity (14)

depend on its location. Extraluminal PPH may arise from the gastroduodenal artery (GDA), splenic artery, or tributaries of the superior mesenteric vessels. Intraperitoneal hemorrhage is often associated with a pancreatic leak and options include reoperation or angioembolization. When reoperation is selected, completion pancreatectomy and suture ligation of the bleeding vessel have been advocated (45). Operative intervention more than 1 week following pancreatic resection may be particularly challenging due to adhesions and tissue friability (46). Arterial embolization is valuable under these circumstances, with a success rate of approximately 80% (Fig. 9.2) (41). We advocate distal ligation of the GDA to

ensure the technical feasibility of angioembolization as it may not be possible when the bleeding point is in close proximity to the common hepatic or superior mesenteric artery. Intraluminal PPH should be initially addressed endoscopically and may originate from anastomoses, mucosal ulceration, or the cut pancreatic surface. When bleeding is found to originate from the cut pancreatic surface, hemostasis may be achieved during reoperation through a jejunotomy or gastrotomy (32). In summary, PPH may occur in up to 9% of patients. The timing and location of the bleeding in patients suffering from PPH are important factors in predicting outcome and determining appropriate management. The overall mortality of PPH is as high as 16% and delayed PPH is associated with a 47% chance of death (41,43). Appreciation of the clinical factors associated with PPH, including sentinel bleeding, sepsis, and pancreatic leak, facilitates prompt recognition. Bile Leak The incidence of choledochoenteric leak following PD is notably lower than pancreatic leak or fistula (Table 9.1). The larger size of the bile duct and more reliable tissue integrity may account for the relative infrequency of biliary leak when compared to pancreatic leak. Similar to pancreatic leak, bile leak is associated with both sterile and infected intraabdominal fluid collections (47). The vast majority of biliary leaks or fistulae can be managed by percutaneous, transhepatic, or transabdominal drainage (48).

82

SURGICAL COMPLICATIONS OF PANCREATECTOMY
Table 9.4 Pylorus Preservation
DGE % Author Van Berge Henegouwen (25) Lin (23) Jimenez (106) Seiler (24) N 200 31 62 114 PD 34 6 12 45 PPPD 37 38 33 32 Morbidity % PD 48 50 45 72 PPPD 44 56 44 57* Mortality % PD 6 0 0 5 PPPD 1 6 3 3 LOS (days) PD 20 – 12 24 PPPD 18* – 15* 25 PD 1580 687 723 2096 EBL PPPD 1247* 451 707 1453* OR TIME (minutes) PD 360 237 – 476 PPPD 288* 215 – 404*

*p < 0.05, DGE = delayed gastric emptying, PD = standard pancreaticoduodenectomy, PPPD = pylorus-preserving pancreaticoduodenectomy, LOS = length of stay, EBL = estimated blood loss.

Figure 9.1 The patient presented with fever and abdominal pain 2 weeks after a pancreaticoduodenectomy. A CT scan revealed a fluid collection in the RUQ (long arrow), which was managed with CT-guided percutaneous drainage (catheter indicated by short arrow). The amylase level in the aspirated fluid was consistent with a pancreatic leak (11,320 U/L).

Death Long-term survival following pancreatic resection is a function of the underlying disease, while perioperative mortality is related to the occurrence of complications, in addition to patient, institutional, and technical factors. Fortunately, the perioperative mortality rate following pancreatic resection has been reported to be less than 2% in the most recent large series (12–14,47,49,50). Pancreatic leak (11) and PPH (40,51,52) are the complications most frequently associated with perioperative mortality. As noted above, the improved mortality rates following pancreatic resection are due in large part to better management of complications. The vast majority of complications can be managed percutaneously, thereby reducing their severity and the risk of death (47).

factors affecting complication rates following pancreatectomy
Pancreatic Duct Management Following Resection Pancreatic leaks prolong hospitalization, and are associated with other complications including DGE, intraabdominal abscess, and cholangitis (3,53). Numerous strategies for management of the pancreatic duct following PD have been advocated, including pancreaticojejunostomy (PJ), pancreaticogastrostomy (PG), and duct ligation (DL). In addition, several technical modifications to distal pancreatectomy (DP) have been tested to reduce the leak rate.
Figure 9.2 Following a pancreaticoduodenectomy and hemodynamic instability, metallic coils were placed in the gastroduodenal artery stump (long arrow) to treat a suspected pseudoaneurysm. The catheter is positioned within the common hepatic artery (short arrow).

Investigators at Johns Hopkins compared PG and PJ in 145 patients who underwent PD and found that the two methods were associated with similar leak rates (53). Several

83

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
variations of PJ have been reported including invagination, end-to-side anastomosis, and side-to-side anastomosis. When compared to end-to-side (duct to mucosa) PJ, end-to end (invaginating) PJ was associated with a trend toward a higher pancreatic fistula rate (15% vs. 4%, p > 0.05). (54) Pancreatic duct ligation following PD as opposed to PG or PJ posed a greater risk of adverse outcomes (55). Marcus et al. also reported that duct ligation following PD was an independent risk factor for pancreatic leakage (18). Whether PG or PJ is employed, ensuring robust perfusion to the cut pancreatic surface prior to anastomosis is essential (56). Various techniques have been applied to both right and left pancreatic resections. Suc et al. (57) conducted an RCT with 182 patients undergoing DP or PD and determined that the use of fibrin glue did not affect the overall complication rate or incidence of pancreatic fistula. Thaker et al. reported that the use of absorbable mesh with a stapler reduced the leak rate significantly among 40 patients undergoing DP compared to the 40 control cases (58). Ferrone et al. did not confirm the efficacy of reinforcing pancreatic transection margins (14). Peritoneal Drainage The only randomized trial addressing the value of routine intraperitoneal drainage following pancreatic resection did not show a benefit (59). Patients who underwent pancreatic resection at the Memorial Sloan-Kettering Cancer Center were randomized to placement of closed suction drains (n = 88) or to no drain placement (n = 91). Those patients who had drains placed were significantly more likely to develop intraperitoneal sepsis, fluid collections, or fistulae (22% vs. 9%, p < 0.02). Thus, placement of drains following pancreatic resection should be considered on a selective basis. Octreotide The pathogenesis of pancreatic leaks has been thought to involve the enzymatic activity of the exocrine secretions. Thus, investigators have tested the ability of octreotide, a synthetic somatostatin analogue, to reduce the risk of postpancreatectomy complications (Table 9.5) (60). The majority of trials demonstrated that octreotide was associated with a significant reduction in perioperative morbidity (61–64). Two trials showed a significant reduction in the incidence of pancreatic fistula in patients receiving octreotide (62,63). The overall frequency of pancreatic fistula was particularly low in two of the trials in which octreotide and placebo were similar (65,66). The trials differ with respect to the proportion of patients undergoing right or left pancreatic resection, frequencies of various diagnoses, the dose of octreotide, and the definitions of pancreatic leak. Given the discrepant results among available trials, the routine use of octreotide for the prevention of pancreatic fistulas cannot be recommended. Individuals at high risk for pancreatic leak (61), such as those with ampullary cancer or soft, friable glands may benefit from exocrine inhibition. The cost of the drug must be balanced against its impact on length of stay and potential avoidance of additional procedures. Pylorus-Preserving Pancreaticoduodenectomy In a RCT comparing classic PD and pylorus-preserving PD (PPPD), the incidence of pancreatic fistula was not significantly different but PPPD was associated with more instances of DGE (23). This study was limited by small sample size and the difference in incidence of DGE between the two groups was not statistically significant. A subsequent RCT demonstrated that cumulative morbidity was significantly more frequent following classic PD when compared to PPPD (72% vs. 57%, p = 0.05) (24). Other trials failed to show significant differences in the rates of DGE or overall surgical complications when comparing classic PD to PPPD (Table 9.4). The decision to perform a PPPD or classic PD is a matter of surgeon preference as the two procedures do not result in markedly different perioperative outcomes. Extended Lymphadenectomy and Resection of Contiguous Structures Several investigators have studied the impact of extended retroperitoneal lymphadenectomy in patients with adenocarcinoma of the pancreas. In a multicenter prospective randomized trial involving 81 patients, extended lymphadenectomy did not significantly affect operative time, blood loss, morbidity, or mortality when compared to the standard dissection (67). The number of lymph nodes removed was similar among the two groups and the extent of resection did not correlate with locoregional control. Yeo et al. reported that radical PD increased operative times (68), as well as the rates of pancreatic fistula, delayed gastric emptying, and overall morbidity (26). Radical or extended PD does not appear to confer an oncologic benefit in patients with adenocarcinoma and may be associated with higher morbidity rates.

Table 9.5 Prophylactic Octreotide
Pancreatic Fistula % Author Buchler (61) Pederzoli (64) Montorsi (63) Friess (62) Lowy (65) Yeo (66) N 246 252 218 247 110 211 Placebo 38 19 20 22 6 11 Octreotide 18^ 9 9* 10* 12 9 55 29 36 30 25 34 Morbidity % Placebo Octreotide 32* 16* 22* 16* 30 40 Mortality % Placebo 5.8 3.8 5.6 0.8 0 0 Octreotide 3.2 1.6 8.1 1.6 2 1

*p < 0.05 versus the control group, ^statistical significance not indicated.

84

SURGICAL COMPLICATIONS OF PANCREATECTOMY
Among the 10% to 20% of patients with adenocarcinoma of the pancreas who are potentially curable, resection of contiguous structures, including the portal vein or spleen, may be necessary in up to 39% (69). While those undergoing resection of contiguous structures may experience higher degrees of intraoperative blood loss and longer hospital stays, perioperative and long-term outcomes are not significantly different (70). When portal vein involvement is the only factor precluding a potentially curative pancreatectomy, resection of the vessel with appropriate reconstruction may be performed without a significant change in operative mortality (71). In a separate study, splenectomy did not lead to increased perioperative morbidity but was associated with decreased survival in patients with pancreatic adenocarcinoma (72). Whether these findings are the result of direct immunologic effects of splenectomy or reflections of more aggressive tumor biology remains uncertain. Total and Central Pancreatectomy The incidence of multifocal pancreas adenocarcinoma is sufficiently low to render total pancreatectomy (TP) unnecessary in the vast majority of cases (73,74). The perioperative (75) and long-term outcomes (76) following TP for adenocarcinoma are even less favorable than those obtained following partial pancreatectomy. The lack of an incremental benefit of TP, along with the endocrine and exocrine sequelae, has limited the use of TP (77). However, increased recognition of intraductal papillary mucinous neoplasms (IPMNs) has led to increased interest in TP (78). Quality of life following TP may not be significantly different from patients with diabetes mellitus not undergoing pancreatic resection (79). Intermittent hypoglycemia is the most common endocrine complication, but fewer than 3% die following TP due to metabolic derangements (80,81). Although islet cell transplantation may delay or prevent the diabetic complications of total pancreatectomy, the role of the procedure is not fully defined (82). As the use of cross-sectional imaging has increased, the frequency of cystic and neuroendocrine lesions of the pancreas is growing. Given that cystic and neuroendocrine pancreatic tumors are often noninvasive, parenchyma-sparing pancreatic resections, such as central pancreatectomy (CP), may be appropriate (83). CP may pose a lower risk of diabetes mellitus than extended DP (84,85) and the rate of exocrine insufficiency is reported to be between 0% and 20% (84–87). The range of pancreatic fistula formation following CP is 14–62% (83–88), which is somewhat higher than what has been associated with right or left pancreatic resection. However, in a recent series, the overall rate of major complications was similar following CP when compared to extended DP (84). Laparoscopic Pancreatectomy Since initially reported (89), laparoscopic distal pancreatectomy is being performed with increasing frequency due to growing interest among patients and physicians. There are no RCTs from which to draw definitive conclusions about complications. Two series including a total of 286 laparoscopic left pancreatectomies indicate a pancreatic fistula rate of 16% to 17% (90,91). The oncologic equivalence of laparoscopic pancreatic resection to conventional approaches remains to be proven. One advantage of laparoscopic pancreatectomy appears to be a decreased length of stay (90). Laparoscopic right and central pancreatectomies are not widely performed and the literature is limited to case reports and small series. In properly selected patients, laparoscopic pancreatectomy may offer shortterm benefits when performed by experienced surgeons. Institutional Factors Hospital or surgeon volume and practice paradigms influence outcome and cost following pancreatic resection. Short-term mortality rates following PD are lower in high-volume compared to low-volume centers (92,93). Improved outcomes in high-volume centers are more likely a reflection of systematic factors rather than an independent effect of more experienced surgeons (93). Utilization of clinical pathways following pancreatic resection has been demonstrated to lower overall cost and decrease the average length of stay by 3 to 6 days (94,95). Clinical pathways have not been associated with significant reductions in morbidity or mortality in patients undergoing pancreatic resection (96). Patient Factors Numerous patient-related factors have been purported to increase the risks of complications and death following pancreatic resection. As noted elsewhere in this chapter, large duct diameter and firm pancreatic parenchymal texture may be associated with a lower risk of pancreatic leak (Table 9.3). Other patient variables that have been reported to increase morbidity rates include coagulopathy, severe jaundice, acute renal failure, obesity, and protein-calorie malnutrition (97–99). Age has not been shown to be an independent risk factor for morbidity and mortality following pancreatectomy (10). The impact of morbid obesity on the risk for postpancreatectomy complications deserves special attention given the scope of this problem in the U.S. population. House et al. determined that retrorenal visceral fat thickness was an independent predictor of overall morbidity, wound infection, and pancreatic fistula (49). Whether jaundice increases the risks of pancreatic resection and the impact of preoperative biliary drainage on perioperative outcomes remain an area of considerable controversy. Povoski et al. (100) demonstrated that in patients undergoing PD, preoperative biliary drainage was an independent predictor of postoperative infection, overall complications, and death. In contrast, Pisters et al. (101) reviewed their experience with preoperative biliary decompression in patients subjected to PD and determined that drainage did not increase the overall morbidity or mortality rates, but did increase the rate of wound infections. A recent meta-analysis of RCTs and comparative cohort studies concluded that there is no benefit to routine preoperative biliary drainage (102). Routine preoperative biliary drainage in jaundiced patients with pancreatic head tumors does not appear to be warranted, but may be appropriate in properly selected patients. Biliary decompression should be considered to address acute cholangitis, intractable pruritis, or to facilitate participation in studies investigating neoadjuvant therapy.

85

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 9.6 Recommendations for Prevention or Management of Pancreatectomy Complications
Evidence Category Radical pancreatectomy or extended retroperitoneal lymphadenectomy may be associated with a higher rate of certain complications and does not confer a significant benefit in oncologic outcome. Restoration of pancreaticoenteric continuity as opposed to ductal ligation is associated with significantly lower rates of pancreatic fistula and endocrine insufficiency. Pancreaticojejunostomy and pancreatogastrostomy following PD have similar complication rates. Pylorus preservation and PD with distal gastrectomy lead to similar perioperative outcomes. Utilization of absorbable mesh when transecting the distal pancreas with a stapling device has not been definitively shown to decrease the risk of pancreatic leak. Routine use of octreotide following pancreatic resection is not indicated but may be useful in selected, high-risk patients. Routine preoperative biliary drainage prior to pancreaticoduodenectomy is not indicated and should be performed in selected patients based upon the presence of symptoms, infection, or severe hyperbilirubinemia. Ib Recommendation Strength Category A

Ib Ib Ib III Ib Ia

A A A C A A

Recommended grading of categories of evidence: Ia, evidence from meta-analysis of randomised controlled trials; Ib, evidence from at least one randomised controlled trial; IIa, evidence from at least one controlled study without randomisation; IIb, evidence from at least one other type of quasi-experimental study; III, evidence from nonexperimental descriptive studies, such as comparative studies, correlation studies and case-control studies; IV, evidence from expert committee reports or opinions and/or clinical experience of respected authorities. Recommended strengths of management recommendation: A, directly based on category I evidence; B, directly based on category II evidence or extrapolated recommendation from category I evidence; C, directly based on category III evidence or extrapolated recommendation from category I or II evidence; D, directly based on category IV evidence or extrapolated recommendation from category I, II, or III evidence.

summary
While the mortality rates following pancreatic resection have improved dramatically, the incidence of complications remains high. Based upon the available literature, several recommendations have been proposed ( Table 9.6). Refinements in our abilities to detect and manage complications following pancreatectomy account, in large part, for improved perioperative mortality statistics. Further progress in enhancing the safety of pancreatic resection will depend upon the development of more effective measures to prevent and treat postpancreatectomy complications. Better understanding of the biology of the diseases we subject to pancreatic resection will allow for more precise patient selection and improve both perioperative and long-term outcomes.

references
1. Kausch W. Das carcinoma der papilla duodeni und seine radikale entfernung. Beitr Klin Chir 1912; 78: 439–51. 2. Whipple AO, Parsons WB, Mullins CR. Treatment of carcinoma of the ampulla of vater. Ann Surg 1935; 102(4): 763–79. 3. Stojadinovic A, Brooks A, Hoos A, et al. An evidence-based approach to the surgical management of resectable pancreatic adenocarcinoma. J Am Coll Surg 2003; 196(6): 954–64. 4. Wente MN, Veit JA, Bassi C, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery 2007;142(1): 20–5. 5. Balcom JH, Rattner DW, Warshaw AL, et al. Ten-year experience with 733 pancreatic resections: changing indications, older patients, and decreasing length of hospitalization. Arch Surg 2001; 136(4): 391–8. 6. Buchler MW, Wagner M, Schmied BM, et al. Changes in morbidity after pancreatic resection: toward the end of completion pancreatectomy. Arch Surg 2003; 138(12): 1310–4; discussion 1315.

7. Muscari F, Suc B, Kirzin S, et al. Risk factors for mortality and intraabdominal complications after pancreatoduodenectomy: multivariate analysis in 300 patients. Surgery 2006; 139(5): 591–8. 8. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997; 226(3): 248–57; discussion 257. 9. Grobmyer SR, Pieracci FM, Allen PJ, et al. Defining morbidity after pancreaticoduodenectomy: use of a prospective complication grading system. J Am Coll Surg 2007; 204(3): 356–64. 10. Bottger TC, Engelmann R, Junginger T. Is age a risk factor for major pancreatic surgery? An analysis of 300 resections. Hepatogastroenterology 1999; 46(28): 2589–98. 11. Gouma DJ, van Geenen RC, van Gulik TM, et al. Rates of complications and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume. Ann Surg 2000; 232(6): 786–95. 12. Vin Y, Sima CS, Gertrajdmen GI, et al. Management and outcomes of postpancreatectomy fistula, leak, and abscess: results of 908 patients resected at a single institution between 2000 and 2005. J Am Coll Surg 2008; 207(4): 490. 13. Winter JM, Cameron JL, Campbell KA, et al. 1423 pancreaticoduodenectomies for pancreatic cancer: A single-institution experience. J Gastrointest Surg 2006; 10(9): 1199–210; discussion 1210–1. 14. Ferrone CR, Warshaw AL, Rattner DW, et al. Pancreatic fistula rates after 462 distal pancreatectomies: staplers do not decrease fistula rates. J Gastrointest Surg 2008; 12(10): 1691–7; discussion 1697–8. 15. Clavien PA, Sanabria JR, Strasberg SM. Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery 1992; 111(5): 518–26. 16. DeOliveira ML, Winter JM, Schafer M, et al. Assessment of complications after pancreatic surgery: A novel grading system applied to 633 patients undergoing pancreaticoduodenectomy. Ann Surg 2006; 244(6): 931–7; discussion 937–9. 17. Bartoli FG, Arnone GB, Ravera G, Bachi V. Pancreatic fistula and relative mortality in malignant disease after pancreaticoduodenectomy. Review and statistical meta-analysis regarding 15 years of literature. Anticancer Res 1991; 11(5): 1831–48.

86

SURGICAL COMPLICATIONS OF PANCREATECTOMY
18. Marcus SG, Cohen H, Ranson JH. Optimal management of the pancreatic remnant after pancreaticoduodenectomy. Ann Surg 1995; 221(6): 635–45; discussion 645–8. 19. Nguyen JH. Distinguishing between parenchymal and anastomotic leakage at duct-to-mucosa pancreatic reconstruction in pancreaticoduodenectomy. World J Gastroenterol 2008; 14(43): 6648–54. 20. Berberat PO, Friess H, Kleeff J, et al. Prevention and treatment of complications in pancreatic cancer surgery. Dig Surg 1999; 16(4): 327–36. 21. Yeo CJ, Barry MK, Sauter PK, et al. Erythromycin accelerates gastric emptying after pancreaticoduodenectomy. A prospective, randomized, placebo-controlled trial. Ann Surg 1993; 218(3): 229–37; discussion 237–8. 22. Warshaw AL, Torchiana DL. Delayed gastric emptying after pyloruspreserving pancreaticoduodenectomy. Surg Gynecol Obstet 1985; 160(1): 1–4. 23. Lin PW, Lin YJ. Prospective randomized comparison between pyloruspreserving and standard pancreaticoduodenectomy. Br J Surg 1999; 86(5): 603–7. 24. Seiler CA, Wagner M, Sadowski C, et al. Randomized prospective trial of pylorus-preserving vs. classic duodenopancreatectomy (Whipple procedure): initial clinical results. J Gastrointest Surg 2000; 4(5): 443–52. 25. van Berge Henegouwen MI, van Gulik TM, DeWit LT, et al. Delayed gastric emptying after standard pancreaticoduodenectomy versus pylorus-preserving pancreaticoduodenectomy: an analysis of 200 consecutive patients. J Am Coll Surg 1997; 185(4): 373–9. 26. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma, part 2: randomized controlled trial evaluating survival, morbidity, and mortality. Ann Surg 2002; 236(3): 355–66; discussion 366–8. 27. Tanaka M, Sarr MG. Role of the duodenum in the control of canine gastrointestinal motility. Gastroenterology 1988; 94(3): 622–9. 28. Carrère N, Seulin P, Julio CH, et al. Is nasogastric or nasojejunal decompression necessary after gastrectomy? A prospective randomized trial. World J Surg 2007; 31(1): 122–7. 29. Doglietto GB, Papa V, Tortorelli AP, et al. Nasojejunal tube placement after total gastrectomy: a multicenter prospective randomized trial. Arch Surg 2004; 139(12): 1309–13; discussion 1313. 30. Brennan MF. Is nil per os still appropriate for patients undergoing upper gastrointestinal surgery? Nat Clin Pract Gastroenterol Hepatol 2008; 5(12): 660–1. 31. Han-Geurts IJ, Hop WC, Kok NF, et al. Randomized clinical trial of the impact of early enteral feeding on postoperative ileus and recovery. Br J Surg 2007; 94(5): 555–61. 32. Balladur P, Christophe M, Tiret E, Parc R. Bleeding of the pancreatic stump following pancreatoduodenectomy for cancer. Hepatogastroenterology 1996; 43(7): 268–70. 33. Bassi C, Dervenis C, Butturini G, et al. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005; 138(1): 8–13. 34. Cunningham JD, Weyant MT, Levitt M, et al. Complications requiring reoperation following pancreatectomy. Int J Pancreatol 1998; 24(1): 23–9. 35. de Castro SMM, Kuhlmann KFD, Busch ORC, et al. Delayed massive hemorrhage after pancreatic and biliary surgery: embolization or surgery? Ann Surg 2005; 241(1): 85–91. 36. Miedema BW, Sarr MG, van Heerden JA, et al. Complications following pancreaticoduodenectomy. Current management. Arch Surg 1992; 127(8): 945–9; discussion 949. 37. Rumstadt B, Schwab M, Korth P, et al. Hemorrhage after pancreatoduodenectomy. Ann Surg 1998; 227(2): 236–41. 38. Brodsky JT, Turnbull AD. Arterial hemorrhage after pancreatoduodenectomy. The ‘sentinel bleed’. Arch Surg 1991; 126(8): 1037–40. 39. Shankar S, Russell RC. Haemorrhage in pancreatic disease. Br J Surg 1989; 76(8): 863–6. 40. Tien YW, Lee PH, Yang CY, et al. Risk factors of massive bleeding related to pancreatic leak after pancreaticoduodenectomy. J Am Coll Surg 2005; 201(4): 554–9. 41. Yekebas EF, Wolfram L, Cataldegirmen G, et al. Postpancreatectomy hemorrhage: diagnosis and treatment: an analysis in 1669 consecutive pancreatic resections. Ann Surg 2007; 246(2): 269–80. 42. Balladur P, Christophe M, Tiret E, Parc R. Bleeding of the pancreatic stump following pancreatoduodenectomy for cancer. Hepatogastroenterology 1996; 43(7): 268–70. 43. Choi SH, Moon HJ, Heo JS, et al. Delayed hemorrhage after pancreaticoduodenectomy. J Am Coll Surg 2004; 199(2): 186–91. 44. Koukoutsis I, Bellagamba R, Morris-Stiff G, et al. Haemorrhage following pancreaticoduodenectomy: risk factors and the importance of sentinel bleed. Dig Surg 2006; 23(4): 224–8. 45. Farley DR, Schwall G, Trede M. Completion pancreatectomy for surgical complications after pancreaticoduodenectomy. Br J Surg 1996; 83(2): 176–9. 46. Otah E, Cushin BJ, Rozenblit GN, et al. Visceral artery pseudoaneurysms following pancreatoduodenectomy. Arch Surg 2002; 137(1): 55–9. 47. Baker TA, Aaron JM, Borge M, et al. Role of interventional radiology in the management of complications after pancreaticoduodenectomy. Am J Surg 2008; 195(3): 386–90; discussion 390. 48. Sohn TA, Yeo CJ, Cameron JL, et al. Pancreaticoduodenectomy: role of interventional radiologists in managing patients and complications. J Gastrointest Surg 2003; 7(2): 209–19. 49. House MG, Fong Y, Arnaoutakis DJ, et al. Preoperative predictors for complications after pancreaticoduodenectomy: impact of BMI and body fat distribution. J Gastrointest Surg 2008; 12(2): 270–8. 50. Kleeff J, Diener MK, Z’Graggen K, et al. Distal pancreatectomy: risk factors for surgical failure in 302 consecutive cases. Ann Surg 2007; 245(4): 573–82. 51. Trede M. The surgical treatment of pancreatic carcinoma. Surgery 1985; 97(1): 28–35. 52. van Berge Henegouwen MI, Allema JH, van Gulik TM, et al. Delayed massive haemorrhage after pancreatic and biliary surgery. Br J Surg 1995; 82(11): 1527–31. 53. Yeo CJ, Cameron JL, Maher MM, et al. A prospective randomized trial of pancreaticogastrostomy versus pancreaticojejunostomy after pancreaticoduodenectomy. Ann Surg 1995; 222(4): 580–8; discussion 588–92. 54. Chou FF, Sheen-Chen SM, Chen YS, et al. Postoperative morbidity and mortality of pancreaticoduodenectomy for periampullary cancer. European J Surg 1996; 162(6): 477–81. 55. Tran K, Van Eijck C, Di Carlo V, et al. Occlusion of the pancreatic duct versus pancreaticojejunostomy: a prospective randomized trial. Ann Surg 2002; 236(4): 422–8; discussion 428. 56. Strasberg SM, Drebin JA, Mokadam NA, et al. Prospective trial of a blood supply-based technique of pancreaticojejunostomy: effect on anastomotic failure in the Whipple procedure. J Am Coll Surg 2002; 194(6): 746–58; discussion 759–60. 57. Suc B, Msika S, Fingerhut A, et al. Temporary fibrin glue occlusion of the main pancreatic duct in the prevention of intra-abdominal complications after pancreatic resection: prospective randomized trial. Ann Surg 2003; 237(1): 57–65. 58. Thaker RI, Matthews BD, Linehan DC, et al. Absorbable mesh reinforcement of a stapled pancreatic transection line reduces the leak rate with distal pancreatectomy. J Gastrointest Surg 2007; 11(1): 59–65. 59. Conlon KC, Labow D, Leung D, et al. Prospective randomized clinical trial of the value of intraperitoneal drainage after pancreatic resection. Ann Surg 2001; 234(4): 487–93; discussion 493. 60. Klempa I, Schwedes U, Usadel KH. [Prevention of postoperative pancreatic complications following duodenopancreatectomy using somatostatin]. Chirurg 1979; 50(7): 427–31. 61. Büchler M, Friess H, Klempa I, et al. Role of octreotide in the prevention of postoperative complications following pancreatic resection. Amer J Surg 1992; 163(1): 125–30; discussion 130. 62. Friess H, Beger HG, Sulkowski U, et al. Randomized controlled multicentre study of the prevention of complications by octreotide in patients undergoing surgery for chronic pancreatitis. Br J Surg 1995; 82(9): 1270–3. 63. Montorsi M, Zago M, Mosca F, et al. Efficacy of octreotide in the prevention of pancreatic fistula after elective pancreatic resections: a prospective, controlled, randomized clinical trial. Surgery 1995; 117(1): 26–31. 64. Pederzoli P, Bassi C, Falconi M, Camboni MG. Efficacy of octreotide in the prevention of complications of elective pancreatic surgery. Italian Study Group. Br J Surg 1994; 81(2): 265–9.

87

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
65. Lowy AM, Lee JE, Pisters PW, et al. Prospective, randomized trial of octreotide to prevent pancreatic fistula after pancreaticoduodenectomy for malignant disease. Ann Surg 1997; 226(5): 632–41. 66. Yeo CJ, Cameron JL, Lillemoe KD, et al. Does prophylactic octreotide decrease the rates of pancreatic fistula and other complications after pancreaticoduodenectomy? Results of a prospective randomized placebo-controlled trial. Ann Surg 2000; 232(3): 419–29. 67. Pedrazzoli S, DiCarlo V, Dionigi R, et al. Standard versus extended lymphadenectomy associated with pancreatoduodenectomy in the surgical treatment of adenocarcinoma of the head of the pancreas: a multicenter, prospective, randomized study. Lymphadenectomy Study Group. Ann Surg 1998; 228(4): 508–17. 68. Yeo CJ, Cameron JL, Sohn TA, et al. Pancreaticoduodenectomy with or without extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma: comparison of morbidity and mortality and short-term outcome. Ann Surg 1999; 229(5): 613–22; discussion 622–4. 69. Brennan MF, Moccia RD, Klimstra D. Management of adenocarcinoma of the body and tail of the pancreas. Ann Surg 1996; 223(5): 506–11; discussion 511–2. 70. Shoup M, Conlon KC, Klimstra D, Brennan MF. Is extended resection for adenocarcinoma of the body or tail of the pancreas justified? J Gastrointest Surg 2003; 7(8): 946–52; discussion 952. 71. Harrison LE, Klimstra DS, Brennan MF. Isolated portal vein involvement in pancreatic adenocarcinoma. A contraindication for resection? Ann Surg 1996; 224(3): 342–7; discussion 347–9. 72. Schwarz RE, Harrison LE, Conlon KC, et al. The impact of splenectomy on outcomes after resection of pancreatic adenocarcinoma. J Am Coll Surg 1999; 188(5): 516–21. 73. Cooperman AM, Herter FP, Marboe CA, et al. Pancreatoduodenal resection and total pnacreatectomy—an institutional review. Surgery 1981; 90(4): 707–12. 74. Edis AJ, Kiernan PD, Taylor WF. Attempted curative resection of ductal carcinoma of the pancreas: review of Mayo Clinic experience, 1951– 1975. Mayo Clin Proc 1980; 55(9): 531–6. 75. Ihse I, Anderson H, Andren S. Total pancreatectomy for cancer of the pancreas: is it appropriate? World J Surg 1996; 20(3): 288–93; discussion 294. 76. Karpoff HM, Klimstra DS, Brennan MF, Conlon KC. Results of total pancreatectomy for adenocarcinoma of the pancreas. Arch Surg 2001; 136(1): 44–7; discussion 48. 77. Grace PA, Pitt HA, Tompkins RK, et al. Decreased morbidity and mortality after pancreatoduodenectomy. Am J Surg 1986; 151(1): 141–9. 78. Cuillerier E, Cellier C, Palazzo L, et al. Outcome after surgical resection of intraductal papillary and mucinous tumors of the pancreas. Am J Gastroenterol 2000; 95(2): 441–5. 79. Billings BJ, Christein JD, Harmsen WS, et al. Quality-of-life after total pancreatectomy: is it really that bad on long-term follow-up? J Gastrointest Surg 2005; 9(8): 1059–66; discussion 1066–7. 80. Assan R, Alexandre JH, Tiengo A, et al. Survival and rehabilitation after total pancreatectomy. A follow-up of 36 patients. Diabete Metab 1985; 11(5): 303–9. 81. Dresler CM, Fortner JG, McDermott K, Bajorunas DR. Metabolic consequences of (regional) total pancreatectomy. Ann Surg 1991; 214(2): 131–40. 82. Webb MA, Illouz SC, Pollard CA, et al. Islet auto transplantation following total pancreatectomy: a long-term assessment of graft function. Pancreas 2008; 37(3): 282–7. 83. Warshaw AL, Rattner DW, Fernandez-del Castillo C, Z’Graggen K. Middle segment pancreatectomy: a novel technique for conserving pancreatic tissue. Arch Surg 1998; 133(3): 327–31. 84. Ocuin LM, Sarmiento JM, Staley CA, et al. Comparison of central and extended left pancreatectomy for lesions of the pancreatic neck. Ann Surg Oncol 2008; 15(8): 2096–103. 85. Adham M, Giunippero A, Hervieu V, et al. Central pancreatectomy: single-center experience of 50 cases. Arch Surg 2008; 143(2): 175–80; discussion 180–1. 86. Iacono C, Bortolasi L, Serio G. Is there a place for central pancreatectomy in pancreatic surgery? J Gastrointest Surg 1998; 2(6): 509–16; discussion 516–7. 87. Rotman N, Sastre B, Fagniez PL. Medial pancreatectomy for tumors of the neck of the pancreas. Surgery 1993; 113(5): 532–5. 88. Celis J, Berrospi F, Ruiz E, et al. Central pancreatectomy for tumors of the neck and body of the pancreas. J Surg Oncol 2001; 77(2): 132–5. 89. Gagner M, Pomp A, Herrera MF. Early experience with laparoscopic resections of islet cell tumors. Surgery 1996; 120(6): 1051–4. 90. Kooby DA, Gillespie T, Bentrem D, et al. Left-sided pancreatectomy: a multicenter comparison of laparoscopic and open approaches. Ann Surg 2008; 248(3): 438–46. 91. Mabrut JY, Fernandez-Cruz L, Azagra JS, et al. Laparoscopic pancreatic resection: results of a multicenter European study of 127 patients. Surgery 2005; 137(6): 597–605. 92. Birkmeyer JD, Finlayson SR, Tosteson AN, et al. Effect of hospital volume on in-hospital mortality with pancreaticoduodenectomy. Surgery 1999; 125(3): 250–6. 93. Lieberman MD, Kilburn H, Lindsey M, Brennan MF. Relation of perioperative deaths to hospital volume among patients undergoing pancreatic resection for malignancy. Ann Surg 1995; 222(5): 638–45. 94. Kennedy EP, Rosato EL, Sauter PK, et al. Initiation of a critical pathway for pancreaticoduodenectomy at an academic institution—the first step in multidisciplinary team building. J Am Coll Surg 2007; 204(5): 917– 23; discussion 923–4. 95. Porter GA, Pisters PW, Mansyur C, et al. Cost and utilization impact of a clinical pathway for patients undergoing pancreaticoduodenectomy. Ann Surg Oncol 2000; 7(7): 484–9. 96. Vanounou T, Pratt W, Fischer JE, et al. Deviation-based cost modeling: a novel model to evaluate the clinical and economic impact of clinical pathways. J Am Coll Surg 2007; 204(4): 570–9. 97. Gilsdorf RB, Spanos P. Factors influencing morbidity and mortality in pancreaticoduodenectomy. Ann Surg 1973; 177(3): 332–7. 98. Warren KW, Cattell RB, Blackburn JP, Nora PF. A long-term appraisal of pancreaticoduodenal resection for peri-ampullary carcinoma. Ann Surg 1962; 155: 653–62. 99. Winter JM, Cameron JL, Yeo CJ, et al. Biochemical markers predict morbidity and mortality after pancreaticoduodenectomy. J Am Coll Surg 2007; 204(5): 1029–36; discussion 1037–8. 100. Povoski SP, Karpeh MS, Jr., Conlon KC, et al. Association of preoperative biliary drainage with postoperative outcome following pancreaticoduodenectomy. Ann Surg 1999; 230(2): 131–42. 101. Pisters PW, Hudec WA, Hess KR, et al. Effect of preoperative biliary decompression on pancreaticoduodenectomy-associated morbidity in 300 consecutive patients. Ann Surg 2001; 234(1): 47–55. 102. Sewnath ME, Karsten TM, Prins MH, et al. A meta-analysis on the efficacy of preoperative biliary drainage for tumors causing obstructive jaundice. Ann Surg 2002; 236(1): 17–27. 103. Pannegeon V, Pessaux P, Sauvanet A, et al. Pancreatic fistula after distal pancreatectomy: predictive risk factors and value of conservative treatment. Arch Surg 2006; 141(11): 1071–6; discussion 1076. 104. Sierzega M, Niekowal B, Kulig J, Popiela T. Nutritional status affects the rate of pancreatic fistula after distal pancreatectomy: a multivariate analysis of 132 patients. J Am Coll Surg 2007; 205(1): 52–9. 105. Ridolfini MP, Alfieri S, Gourgiotis S, et al. Risk factors associated with pancreatic fistula after distal pancreatectomy, which technique of pancreatic stump closure is more beneficial? World J Gastroenterol 2007; 13(38): 5096–100. 106. Jimenez RE, Fernandez-del Castillo C, Rattner DW, et al. Outcome of pancreaticoduodenectomy with pylorus preservation or with antrectomy in the treatment of chronic pancreatitis. Ann Surg 2000; 231(3): 293–300.

88

10

Laparoscopy in HPB surgery Nicholas O’Rourke and Richard Bryant
also enables the identification of choledocholithiasis, which in most cases can then be successfully managed during the same laparoscopic procedure (11). In the setting of acute cholecystitis, early laparoscopic cholecystectomy is preferred (12,13). The problem with a policy of delayed laparoscopic cholecystectomy is that a significant proportion of patients require an emergency cholecystectomy for recurrent or nonresolving acute cholecystitis in the difficult intermediate period with a higher rate of conversion. However, if symptoms have been present for more than a week or there is a mass present without generalized peritonism then it may be more prudent to manage the patient conservatively with a view to a delayed cholecystectomy. For mild gall stone pancreatitis, laparoscopic cholecystectomy with intraoperative cholangiogram should be performed during the same admission (14). A policy of interval cholecystectomy incurs a real risk of recurrent pancreatitis (15–21). Laparoscopic cholecystectomy has traditionally been performed with an overnight stay, but appropriately selected patients can be safely managed as a day case (22).

introduction
Laparoscopy offers great advantages to the patient with HPB disease. Although described in the early part of the 20th century, crude instrumentation limited its use. Progress seemed slow until the 1960s saw widespread uptake in the gynecologic community, with the Hopkins rod lens system greatly improving the optics. Sporadic reports of laparoscopic staging for HPB cancer soon followed, but it was not until the handheld camera development in the 1980s that the minimal access explosion began. Now surgeons could view the image on a monitor, and use two hands to operate instruments, while an assistant held the camera. Even the gall bladder could be removed using tiny incisions. The next 10 years saw almost every abdominal operation attempted, such that the interest now is not in what can be done, but in what should be done, and how best to do it.

laparoscopic cholecystectomy
Cholecystectomy was the first general surgical procedure to be widely performed laparoscopically. Following the first reports in 1985 and 1988 (1), the technique was rapidly popularized (2–5). Despite a possible increase in the incidence of severe bile duct injuries, the benefits of the laparoscopic approach have subsequently been confirmed by meta-analysis (6), demonstrating shorter hospital stay and faster convalescence with no difference in operating time or complications. There are various techniques in common usage, with the surgeon standing either on the patient’s right or left or between the legs, with the choice of technique depending on local teaching and personal preference. There are, however, fundamental principles to safely performing a laparoscopic cholecystectomy. Correct identification of the anatomy is fundamental. Most bile duct injuries are due to misperception rather than technical errors (7). It is important to understand the normal variations in biliary anatomy and how pathological changes may alter the relationships between the structures. The 30º telescope permits better visualization of Calot’s triangle. Hartmann’s pouch is retracted laterally and inferiorly so that the angle between the cystic and common hepatic ducts is increased rather than closed. Calot’s triangle is dissected high, just beneath the edge of the gall bladder, on both its anterior and posterior surfaces, to clearly identify the cystic duct and cystic artery as the only structures passing to the gall bladder (the “critical view” (8)). Dissection is never carried below the plane of Rouvière’s sulcus (9). Routine intraoperative cholangiography is recommended. This has been shown to decrease the risk and severity of biliary injury (10). It is essential that the full complement of upper duct anatomy is visualized to be certain that the common bile duct or an aberrant right hepatic duct is not being excised. It

laparoscopic common bile duct exploration
Most common duct stones can be managed laparoscopically (11,23–36). This allows treatment in one operation, rather than endoscopic retrograde cholangiopancreatography (ERCP) and sphincterotomy done as a separate procedure, either before or after laparoscopic cholecystectomy. Obviously, operative cholangiography, with fluoroscopy, and accurate interpretation is mandatory. Techniques used, in order of increasing complexity, are as follows:
● ● ● ● ●

Transcystic flushing Transcystic stone extraction Choledochotomy Transampullary stenting Choledochoduodenostomy

Flushing Small filling defects in the bile duct may be air bubbles, and only the dynamic image of fluoroscopy may allow visible distortion or coalescence of these bubbles. Small stones low in the bile duct may be flushed or pushed into the duodenum using the cholangiogram catheter, with intravenous glucagon occasionally helping by relaxing the sphincter. Transcystic Stone Extraction Formal duct exploration is performed, where possible (in about two-thirds of cases), via a transcystic approach. We prefer to use a purpose-built Nathanson CBD exploration catheter (Cook). This allows manipulation of a basket under contrast-assisted fluoroscopy. The cystic duct is dissected lower, close to the common bile duct. Balloon dilatation of the

89

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
cystic duct may occasionally be required. The stone to be extracted must not be larger than this diameter. If so, the stone may become stuck in the junction and require fragmentation or, worse, incision to remove. Choledochotomy as a primary procedure may be preferred for large or numerous stones. When inserting the Nathanson transcystic catheter, one must be careful that the basket is 1 to 2 cm inside the flexible tip of the catheter, to avoid turning the device into a spear, which can perforate the posterior aspect of the common duct. Under fluoroscopic guidance, the wire basket is deployed in the distal common bile duct, without traversing the ampulla. Gentle “jiggling” of the basket entraps the stone, which can then be retrieved by withdrawing the open basket. The stone can “flip” out of the duct and land anywhere in the right abdomen, often too quickly to be seen (Figs. 10.1 and 10.2). The characteristics of the basket employed are important. A fourwire steel basket will spring open in the bile duct such that with “jiggling” the stone is able to enter between the wires to then be trapped in the apex as the open basket is withdrawn. A softer nitonol basket will not tend to spring open in the same fashion and it is therefore often difficult to ensnare the stone under fluoroscopic guidance. An alternative transcystic approach is with a flexible choledochoscope. A 3-mm scope is normally required, as a 5-mm scope will only rarely pass trans-cystically. A grasper in the epigastric port provides traction to the right. A long 5-mm trocar in the right subcostal position is positioned against the cystic duct incision to prevent bowing of the choledochoscope within the abdomen. The choledochoscope is advanced into the common bile duct and the stone retrieved under direct vision. In these circumstances, a nitonol basket with a parachute arrangement at the apex is usually more effective as the stone entrapment is performed under direct vision. Clearance of the common bile duct can be confirmed by transcystic flexible choledochoscopy; however it is often difficult transcystically to introduce the choledochoscope to the common hepatic duct to confirm that there are no calculi above the cystic duct junction. If the transcystic approach fails, then a decision must be made between postoperative ERCP versus laparoscopic choledochotomy. A randomized controlled trial between these two options, after failure of transycstic CBDE, did not demonstrate any differences (27), and therefore the choice depends on individual patient factors and local expertise. If the common bile duct is narrow (<7 mm) then a choledochotomy should be avoided due to the risk of stricturing. Postoperative ERCP can be facilitated by the passage of an antegrade biliary stent (37). Choledochotomy In certain circumstances, a transcystic approach is unlikely to be successful, and if the CBD is of sufficient diameter, then it is reasonable to proceed straight to a laparoscopic choledochotomy. These circumstances include large stones (>10mm), multiple stones (>3) or stones above the cystic duct confluence. To perform laparoscopic choledochotomy, the anterior surface of the common bile duct is dissected just sufficiently to confidently identify the anatomy. A 1.5-cm vertical incision is made in the common bile duct below the cystic duct confluence. Filling of the duct system with saline via the transcystic catheter distends the collapsed duct and helps prevent injury to the posterior duct wall when the anterior wall is incised. Another method is to gently lift up the anterior wall with a suitable small atraumatic grasper (such as a “dolphin-nose”) introduced via the right subcostal port. This will create a small transverse ridge of the anterior duct wall, which can then be cut using scissors introduced via the epigastric port, thus creating a vertical incision that can then be extended with the scissors. A similar effect can also be created using stay sutures. Initial flushing via the choledochotomy with the suckeraspirator and massaging of the duct may remove the stones. A choledochoscope can be introduced, this time from the epigastric port. A 3 or 5 mm flexible scope may be employed, or even a rigid ureteroscope if the orientation is suitable. (On the rare occasions that a rigid ureteroscope is required, it can

Figure 10.1 Laparoscopic transcystic cholangiography demonstrating calculus in the distal common bile duct.

Figure 10.2 The calculus from Figure 10.1 after transcystic extraction utilizing the Nathanson basket. Inset: completion cholangiography.

90

LAPAROSCOPY IN HPB SURGERY
sometimes be introduced transcystically via the epigastric port if the orientation is suitable.) With the choledochoscope, the stones can be removed under direct vision, and the flexible choledochoscope can be maneuvered into both the upper ducts and lower CBD to confirm clearance of all calculi. On rare occasions, hydraulic lithotripsy may be required to break up impacted stones (27). Where there is confidence about stone clearance and biliary drainage, choledochotomy can be simply closed by suturing (25). If there is any doubt about biliary drainage or duct clearance, then choledochotomy should be closed after passage of an antegrade biliary stent, or a T-tube inserted. Choledochoduodenostomy For the elderly patient with a suspected benign stricture, and a reasonable stone load, laparoscopic choledochoduodenostomy is a good option (23). As in open surgery, a common duct diameter of greater than 10 mm is preferable. A continuous absorbable suture is used, and the operation mimics the open procedure with anastamosis of the choledochotomy to a longitudinal opening in the duodenum. imaging studies, but be found to have locally advanced disease or small liver or peritoneal metastases (imaging-occult metastases) that render the disease inoperable (Fig. 10.3). Staging laparoscopy can identify these patients and therefore spare the patient a laparotomy. Staging laparoscopy in its simplest form involves visual inspection of the peritoneal and liver surfaces, but may also include laparoscopic ultrasound, trial dissection, or peritoneal washing for cytology. Staging laparoscopy is preferable to a nontherapeutic laparotomy to identify unresectability. The hospital stay is shorter (51,52), and the patient is able to start chemotherapy sooner (53). The risks of a staging laparoscopy are low, with morbidity reported at 0% to 4% and mortality 0% to 0.15% (54). Port-site recurrences are uncommon, between 0% and 2% (54), and usually occur in patients with extensive peritoneal carcinomatosis. Staging laparoscopy may be performed as a prelude to resection in the same procedure or as a separate procedure prior to planned resection—there can be significant scheduling issues depending on the institution if an aborted procedure means allocated theater time is unable to be utilized. The yield of staging laparoscopy depends on many factors. The type and stage of the malignancy affects the likely presence of imaging-occult metastases, as does the quality and type of the imaging performed. The extent of the staging procedure is also important—whether laparoscopic ultrasound, peritoneal washings or trial dissection is included. It is also obviously influenced by what findings are considered to contraindicate resection; for example, localized peritoneal disease or porta hepatis nodes for colorectal liver metastases, or involvement of the portal vein requiring vein resection and grafting in pancreas cancer may not be considered contraindications to resection. The value of a positive staging laparoscopy also depends on whether any required palliative procedures, such as biliary or gastric bypass in carcinoma of the head of the pancreas, can be performed laparoscopically. In adenocarcinoma of the pancreas, after high-quality CT scanning, staging laparoscopy has been shown to identify

pancreatic pseudocyst
Pancreatic pseudocysts can be managed endoscopically with gastrotomy and stenting (perhaps the only current valid indication for NOTES [Natural Orifice Transabdominal Endoscopic Surgery]). Pancreatic pseudocysts can also be drained internally via a laparoscopic approach (38–46). Most commonly the pseudocyst is located in the lesser sac and the appropriate procedure is a cyst-gastrostomy. An anterior gastrotomy is made. The cyst can be seen bulging forward, adherent to the posterior stomach wall, which is incised with diathermy or the harmonic scalpel to enter the cyst. Cyst fluid will come flowing out under pressure at this point, and is important to have an instrument ready to pass into the cyst so that the point of communication is not lost. The cyst fluid is aspirated with the sucker and the cyst emptied. A linear stapler is then introduced into the small cyst-gastrotomy to create a wide cystgastrostomy, and the residual unstapled edges are sutured together. The pseudocyst can be entered with the laparoscope and inspected, and any debris removed. The anterior gastrotomy is then closed with sutures or a stapling device. In some cases, the position of the pseudocyst will require a side-to-side cyst-gastrostomy or Roux-en-Y cyst-enterostomy. Published reports suggest that laparoscopic cyst-gastrostomy has a higher initial success rate and lower recurrence rate than endoscopic cyst-gastrostomy (42,47). As the cyst-gastrostomy created via the endoscopic approach is only small, any large debris is unable to exit the cyst. However, endoscopic approaches can be improved with using balloon dilatation and multiple stents to maintain better drainage, endoscopic ultrasound to guide the procedure and avoid vessels (48,49), and with the development of stapling instrumentation for natural orifice surgery (50).

laparoscopic staging
A proportion of patients with hepatobiliary and pancreatic malignancies will appear to be resectable on noninvasive
Figure 10.3 Peritoneal metastases at staging laparoscopy and carcinoma of the head of the pancreas.

91

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
unresectability in 15% to 51% of patients, and spare 10% to 31% of patients an unnecessary laparotomy (51,55–61). Laparoscopic ultrasound has been shown to add information in 12% to 14% of patients (62–64). Patients with tumors larger than 3 cm are more likely to have unsuspected metastases at exploration (65), as are patients with a Ca 19.9 level greater than 150 kU/L (66,67). Positive peritoneal lavage has been found in 3% to 51% of patients (57,68–76), and is more likely in locally advanced or metastatic tumors (77), larger tumors, and tumors of the body or tail (70,78). Positive peritoneal cytology, which has the same prognosis as metastatic disease (79), is the only marker of unresectability in 1% to 14% of patients (57,69,70,76). Tumors of the body and tail of the pancreas are twice as likely as pancreatic head lesions to have imaging-occult metastases (57,69). Imaging-occult metastases are uncommon in nonpancreatic periampullary tumors (60,80,81) and routine laparoscopy in these patients is probably not indicated. Patients who on imaging have locally advanced, unresectable pancreatic cancer should also be considered for staging laparoscopy, as those without metastatic disease can be considered for chemoradiotherapy regimens aimed at local control or even downstaging followed by resection, regimens which would incur unnecessary treatmentrelated morbidity for those with metastatic disease (69,77,82). In colorectal liver metastases, laparoscopy will identify unresectable disease in 10% to 38% of patients, with a sensitivity of 39% to 75% (83–90). Laparoscopy is more likely to be positive in patients with a higher clinical risk score (83,86,91). In noncolorectal, nonneuroendocrine liver metastases, laparoscopy has been reported to identify unresectable disease in 25% of patients, with a sensitivity of 66% (92). Staging laparoscopy is useful for patients with primary biliary malignancies. For patients with suspected resectable gall bladder carcinoma on imaging, the yield for detecting unresectable disease is 56% to 62% (93,94), though the yield is less for intrahepatic cholangiocarcinoma (93) at 36% and hilar cholangiocarcinoma (93–95) at 25%. The yield for hilar cholangiocarcinomas is higher for T2 or T3 lesions than for T1 lesions (94) (36% vs. 9%). In hepatocellular carcinoma that is considered suitable for curative resection, peritoneal dissemination is uncommon, and standard laparoscopy is unlikely to add much information. Laparoscopy with laparoscopic ultrasound, however, can identify the extent of the primary tumor, additional imagingoccult tumors, portal or hepatic venous tumor thrombus or an inadequate hepatic remnant, with a yield for unresectability of 10% to 36% and a sensitivity of 63% to 96% (96–100). The results obtained will depend on the type and quality of preoperative imaging and the level of experience with laparoscopic ultrasound. gastric outlet obstruction, it is not necessary to perform a palliative biliary or gastric bypass (101). In many instances, the endoscopic approach is effective to relieve obstruction. Duodenal stenting is safer and provides a better quality of life than laparoscopic gastrojejunostomy in the short term (102), although laparoscopic gastrojejunostomy may provide a more durable result for patients with a longer life expectancy (103). ERCP with placement of a plastic biliary stent has a lower morbidity than traditional open surgical bypass, although plastic biliary stents have a tendency to occlude, resulting in recurrent biliary obstruction requiring a repeat procedure (104). Metallic stents, however, have a much higher patency rate in the longer term, and can serve many patients for the remainder of their survival (104–106). In some cases, however, stenting fails for technical reasons or due to inability to access the ampulla. In these cases, a laparoscopic bypass is a useful option (107–116), with the potential for lower morbidity and shorter hospital stay than an open surgical procedure (113,115). A laparoscopic biliary bypass is most easily performed as a stapled or sutured side-to-side cholecystojejunostomy. The main limitation of this approach is that the confluence of the cystic and common hepatic ducts must be well above the tumor to prevent recurrent biliary obstruction (117). This can be confirmed at the procedure by cholangiography via the fundus of the Gall bladder—a Verres needle with large syringe attached is used to empty the gall bladder of bile, which is then filled with contrast to confirm that the cystic duct confluence is more than 1 cm above the level of the tumor. If this is not the case, an hepaticojejunostomy is constructed. A gastrojejunostomy is typically fashioned in an antecolic, isoperistaltic stapled side-to-side manner.

laparoscopic pancreatectomy
Distal pancreatectomy is well suited to a laparoscopic approach. The usual indication is a solid or cystic tumor of the tail of the pancreas that is not clearly benign on preoperative imaging. The procedure may involve en-bloc resection of the spleen and splenic vessels; preservation of the spleen with preservation of the splenic vessels; or preservation of the spleen without preservation of the splenic vessels with the spleen supplied from the short gastric and gastroepiploic vessels (the Warshaw technique (118)). For lesions close to the spleen, when splenectomy is necessary, the approach can be similar to laparoscopic splenectomy, with the patient left side up, and the spleen and distal pancreas mobilized from lateral to medial. After division of the short gastric vessels and the gastrocolic omentum, the pancreas can be divided en bloc with the splenic vessels using a linear stapler. For a medial to lateral approach, the pancreatic neck is divided, either with a stapling device or with the harmonic scalpel with subsequent suture closure of the pancreatic stump. Where the splenic vessels are being resected, the splenic vein is divided with a stapling device and the splenic artery divided with a stapling device or locking clips. If the splenic vessels are to be preserved, then the tail of the pancreas is dissected carefully from them with control of the small vessels with clips and/or the harmonic scalpel or

laparoscopic palliative bypass
In patients with inoperable periampullary tumors, there is often biliary and/or gastric obstruction that requires relief. The traditional teaching in open surgery was to perform both a biliary and gastric bypass whether or not the patient was symptomatic. If at laparoscopy the tumor is found to be unresectable, in the absence of actual or impending biliary or

92

LAPAROSCOPY IN HPB SURGERY
electrosurgical sealing device. Otherwise the dissection continues in the relatively avascular plane behind the splenic vein. At this point, if the spleen is to be preserved with the Warshaw technique, then the splenic hilum is divided with a stapling device taking care to preserve the short gastric vessels, and the gastroepiploic arcade. Otherwise if the spleen is to be resected, the dissection continues in this plane behind the splenic vein to complete the mobilization of the spleen and complete the resection. The specimen is retrieved in a bag and a closed suction drain is placed. Laparoscopic distal pancreatectomy has been shown to be a safe procedure, with a shorter hospital stay and overall morbidity that is less than the open procedure (119–124). The main complication is a pancreatic fistula occurring in about 15% of patients, though this occurs at no greater rate than with an open resection (120,124). The application of fibrin glue to the stump (125) and the use of staple line mesh reinforcement (126) have both shown some benefit in small studies in reducing this rate, and in open surgery the placement of a transampullary stent (127) has shown some benefit, as has identification and direct suture of the main pancreatic duct (128), although the optimal management of the pancreatic stump is still to be determined. Preservation of the spleen by the Warshaw technique can be complicated by infarction of the lower pole of the spleen (129,130). Laparoscopic central pancreatectomy has been reported in the literature (131) and successfully been performed twice by one of the authors. The indication of a central tumor where diabetes is a risk postoperatively is not common. Laparoscopic enucleation of insulinomas has been reported in small series but is associated with a significant rate of pancreatic fistula (129,132,133). Intraoperative ultrasound is essential to ensure that the main pancreatic duct is not close to the resection line. Laparoscopic pancreaticoduodenectomy has been reported in small numbers (134–137). The procedure is feasible but prolonged and difficult, and the potential role for this procedure remains to be determined. Kentucky in November 2008. Agreed definitions of laparoscopic liver surgery include the following:








Pure laparoscopic: where the liver resection is completed laparoscopically and the specimen removed via a remote incision; Hand assisted: where the surgeon operates with his nondominant hand inside the abdomen, placed via an airtight device, through which the specimen is removed; Hybrid liver resection (145): where the liver is mobilized laparoscopically and most of the resection is done through a smaller than usual right upper quadrant incision; Conversion: where the surgeon changes to an open operation from one of the above. One can also convert from pure laparoscopic to hand assist or hybrid.

laparoscopic liver resection
The laparoscopic approach to liver resections presents certain technical challenges. It is a heavy solid organ that can be cumbersome to mobilize and manipulate, parenchymal transection requires the identification and control of large vessels with the potential for significant bleeding, and the paucity of external anatomical markers can make the maintenance of surgical orientation to ensure a satisfactory oncologic clearance difficult. Laparoscopic liver resection was initially reported in 1995 by Rau (138), Cuesta (139), and Hashizume (140). Anatomic resections in the form of left lateral sectionectomy were reported in 1996 by Azagra (141) and Kaneko (142), formal hemihepatectomies were reported in 1998 by Huscher (143), and Cherqui (144) reported the first significant series of 30 patients in 2000. The largest series were recently reported by Koffron (145) and Buell (146). Dr. Joe Buell organized the first international consensus meeting on laparoscopic liver resection, held in Louisville,

The most suitable cases for a laparoscopic approach are solitary small (<5 cm) lesions located in the peripheral segments (2–6) of the liver. Larger lesions are acceptable if they are pedunculated or located in the left lateral section. Multiple lesions may be suitable if they can be resected with a single anatomic hepatectomy with a clear margin, but not where multiple complicated or bilobar procedures are required. Hemihepatectomies can be considered for a laparoscopic approach where the plane of transection and major structures (pedicles, hepatic veins and inferior vena cava) are well clear of any lesions. Lesions located in segments 7 and 8 are difficult to approach laparoscopically for a tumorectomy as the costal margin limits the approach angles of the instruments, and there is a real risk of compromising the deep margin for fear of causing difficultto-control bleeding—they should only be considered for a laparoscopic tumorectomy if they are particularly small and superficial, otherwise they need to be considered for an open procedure or a laparoscopic right hemihepatectomy. The procedures are usually performed in the supine position, often with the surgeon standing between the legs. The left lateral decubitus position is useful for lesions in segments 6 and 7, which enables better exposure of the right posterior section of the liver. Hand ports may be used. These are most useful in right sided resections where mobilization is difficult, either in nonanatomical resections with a posterior tumor in the right lobe, or right hemihepatectomies with a bulky right lobe. Good quality laparoscopic equipment is vital. A good 10-mm laparoscopic right angle is also a very important tool. An initial laparoscopy is performed, and laparoscopic ultrasound is used to identify the lesions and their relationships to the appropriate anatomy (Fig. 10.4). A tape can be placed around the hepatic pedicle in readiness for a Pringle maneuver if required; this is usually reserved for situations where bleeding is encountered rather than used routinely, and uses an intermittent protocol (as the time of transection tends to be longer than in open surgery). The gall bladder is resected where indicated, but after division of the cystic duct and artery, it may be left attached to the liver until later in the procedure to help maneuver the liver, such that the gall bladder and round ligament become the two “handles of the liver.” It is a

93

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 10.4 Laparoscopic ultrasound used to mark out resection line for a tumor in segment 6.

useful point to divide the round ligament flush with the anterior abdominal wall such that there is not dangling tissue irritatingly obstructing the view and dirtying the camera through the whole procedure. There are many methods of parenchymal transection: harmonic scalpel (Ethicon), Ligasure device (Covidien), Gyrus (Gyrus ACMI), CUSA (Integra), TissueLink (Salient Surgical Technologies), stapling devices, water jet, and metal clips. Each have their advantages and disadvantages, and used appropriately each can have their place. Personal preference and experience as well as local teaching and availability determine the choice. The various energy-delivery devices will not control the large venous structures; these must be identified intraparenchymally and controlled with clips or stapling devices. It is also prudent to individually control the large pedicular branches. Stapling devices can be used en-masse across portions of the parenchyma to control the larger structures within, but a degree of finesse is lost and unexpected bleeding can be encountered. The combination of the harmonic scalpel for the superficial 2 cm of dissection with the CUSA for the deeper dissection is a good technique (147). A good alternative is the Ligasure device, which when used with a modified technique (closing while activating, using the cutting blade sparingly, with gentle saline irrigation to prevent charring) can be used to dissect out the larger intraparenchymal structures (148). Left lateral sectionectomy begins with mobilization of the falciform ligament, left coronary ligament, and lesser omentum to mobilize the left lobe. The parenchyma is divided so as to expose the upper surface of the segments 2 and 3 pedicles intrahepatically. The pedicles are then divided with a stapler. The parenchymal transection is then completed to expose the left hepatic vein intrahepatically, which is divided with a stapler. An alternative to this technique is mass stapling of the left lateral section (149,150). Left lateral sectionectomy is particularly suitable to a laparoscopic approach and arguments have been made that the laparoscopic approach should be used routinely for this resection (149,151).

Figure 10.5 Laparoscopic dissection of the right portal vein. The right hepatic artery has been divided. The cystic duct has been divided and is used to retract the bile duct. The right portal vein has been looped. The left portal vein is clearly demonstrated and the right portal vein can be seen dividing into its anterior and posterior branches.

For a left hepatectomy, the left liver is mobilized as above. The left hepatic artery and left portal vein are dissected extrahepatically, demonstrating the line of demarcation. The parenchymal transection is then begun, opening the liver to allow a good exposure of the left pedicle and sufficient space to introduce a stapling device to divide the left bile duct. The parenchymal transection is then continued, exposing the left hepatic vein intrahepatically, which is divided with a stapling device to complete the transection. A right hepatectomy can be performed either with an anterior or a traditional approach. An anterior approach begins with an extrahepatic dissection and division of the right hepatic artery and right portal vein (Fig. 10.5). The parenchymal transection is then begun, opening the liver to allow an intrahepatic division of the right bile duct. The parenchymal transection is then completed down to the anterior surface of the inferior vena cava. The minor hepatic veins are then divided between clips, followed by the right hepatic vein and hepatocaval ligament with stapling devices. The final step is mobilization of the liver and division of the right coronary ligament. In the traditional approach, there is the same extrahepatic division of the right hepatic artery and right portal vein, with full mobilization of the liver and division of the hepatocaval ligament and right hepatic vein before transection of the parenchyma (Fig. 10.6). Laparoscopic right hepatectomy is a difficult procedure that requires expertise in both laparoscopic and hepatic surgery. The specimen is removed intact in a bag, either through the hand port incision, a previous appendicectomy scar, or a Pfannenstiel incision. After this period of desufflation, the extraction incision is closed to allow re-establishment of the pneumoperitoneum to confirm hemostasis, as bleeding may have been tamponaded by the pressure of the pneumoperitoneum. In any type of laparoscopic liver resection, significant

94

LAPAROSCOPY IN HPB SURGERY

references
1. Reynolds W. The first laparoscopic cholecystectomy. J Soc Laparoendosc Surgeons 2001; 5(1): 89–94. 2. Dubois F, Berthelot G, Levard H. Laparoscopic cholecystectomy: historic perspective and personal experience. Surgical Laparosc Endosc 1991; 1(1): 52–7. 3. Schirmer BD, Edge SB, Dix J, et al. Laparoscopic cholecystectomy. Treatment of choice for symptomatic cholelithiasis. Ann Surg 1991; 213(6): 665–76; discussion 677. 4. Bailey RW, Zucker KA, Flowers JL, et al. Laparoscopic cholecystectomy. Experience with 375 consecutive patients. Ann Surg 1991; 214(4): 531– 40; discussion 540–1. 5. Fielding GA. Laparoscopic cholecystectomy. Aust NZ J Surg 1992; 62(3): 181–7. 6. Keus F, de Jong JAF, Gooszen HG, van Laarhoven CJHM. Laparoscopic versus open cholecystectomy for patients with symptomatic cholecystolithiasis. Cochrane Database of Systematic Reviews (Online). 2006; (4): CD006231. 7. Way LW, Stewart L, Gantert W, et al. Causes and prevention of laparoscopic bile duct injuries: analysis of 252 cases from a human factors and cognitive psychology perspective. Ann Surg 2003; 237(4): 460–9. 8. Strasberg SM, Hertl M, Soper NJ. An analysis of the problem of biliary injury during laparoscopic cholecystectomy. J Am Coll Surg 1995; 180(1): 101–25. 9. Hugh TB, Kelly MD, Mekisic A. Rouvière’s sulcus: a useful landmark in laparoscopic cholecystectomy. Br J Surg 1997; 84(9): 1253–4. 10. Fletcher DR, Hobbs MS, Tan P, et al. Complications of cholecystectomy: risks of the laparoscopic approach and protective effects of operative cholangiography: a population-based study. Ann Surg 1999; 229(4): 449–57. 11. Rhodes M, Nathanson L, O’Rourke N, Fielding G. Laparoscopic exploration of the common bile duct: lessons learned from 129 consecutive cases. Br J Surg 1995; 82(5): 666–8. 12. O’Rourke NA, Fielding GA. Laparoscopic cholecystectomy for acute cholecystitis. Aust NZ J Surg 1992; 62(12): 944–6. 13. Gurusamy KS, Samraj K. Early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Cochrane Database of Systematic Reviews (Online). 2006; (4): CD005440. 14. UK guidelines for the management of acute pancreatitis. Gut 2005; 54 Suppl 3: iii1–9. 15. Cameron DR, Goodman AJ. Delayed cholecystectomy for gallstone pancreatitis: re-admissions and outcomes. Ann Royal Coll Surgeons Engl 2004; 86(5): 358–62. 16. Cheruvu CVN, Eyre-Brook IA. Consequences of prolonged wait before gallbladder surgery. Ann Royal Coll Surgeons Engl 2002; 84(1): 20–2. 17. DeIorio AV, Vitale GC, Reynolds M, Larson GM. Acute biliary pancreatitis. The roles of laparoscopic cholecystectomy and endoscopic retrograde cholangiopancreatography. Surg Endosc 1995; 9(4): 392–6. 18. Frei GJ, Frei VT, Thirlby RC, McClelland RN. Biliary pancreatitis: clinical presentation and surgical management. Am J Surg 1986; 151(1): 170–5. 19. Ong S, Christie PM, Windsor JA. Management of gallstone pancreatitis in Auckland: progress and compliance. ANZ J Surg 2003; 73(4): 194–9. 20. Ranson JH. The timing of biliary surgery in acute pancreatitis. Ann Surg 1979; 189(5): 654–63. 21. Sargen K, Kingsnorth AN. Management of gallstone pancreatitis: effects of deviation from clinical guidelines. J Pancreas 2001; 2(5): 317–22. 22. Gurusamy K, Junnarkar S, Farouk M, Davidson BR. Meta-analysis of randomized controlled trials on the safety and effectiveness of day-case laparoscopic cholecystectomy. Br J Surg 2008; 95(2): 161–8. 23. Campbell-Lloyd AJM, Martin DJ, Martin IJ. Long-term outcomes after laparoscopic bile duct exploration: a 5-year follow up of 150 consecutive patients. ANZ J Surg 2008; 78(6): 492–4. 24. Fielding GA, O’Rourke NA. Laparoscopic common bile duct exploration. Aust NZ J Surg 1993; 63(2): 113–5. 25. Martin IJ, Bailey IS, Rhodes M, et al. Towards T-tube free laparoscopic bile duct exploration: a methodologic evolution during 300 consecutive procedures. Ann Surg 1998; 228(1): 29–34.

Figure 10.6 Mobilization of the right liver from the inferior vena cava.

bleeding can be encountered, and the surgeon must have the requisite laparoscopic skills to be able to control this situation. Good skills in laparoscopic suturing are essential. Conversion to laparotomy may certainly be required, but immediate conversion is not always the best response. Venous bleeding is often partly tamponaded by the pneumoperitoneum. The bleeding source is identified and controlled with a grasper, and hemostasis is achieved with suturing or a clip as appropriate. If initial maneuvers are not successful then conversion is required without persisting for too long or worsening the bleeding. These are potentially dangerous situations that require both skill and judgment. Assessment of laparoscopic liver resection has been based on series (144,147–149,151–173) and retrospective comparative studies (145,174–185). These reports from expert centers demonstrate that laparoscopic liver resection can be performed safely, and that despite a longer operating time there is the potential for reduced hospital stay and reduced bleeding. Despite initial concerns, CO2 embolus occurs uncommonly. The oncologic results in nonrandomized studies have been good (145,158,174–177,180), but care must be taken in interpreting these series, as those patients undergoing laparoscopic resection have been selected with smaller and fewer tumors that would normally also infer a better prognosis. There is a potential benefit in cirrhotic patients, with a lower incidence of ascites postoperatively (147,176), as well as fewer adhesions that facilitate subsequent transplantation (186).

conclusion
Laparoscopic approaches for the simplest of HPB procedures, cholecystectomy, have literally exploded around the world. More complex operations have been reported in small series, but have not been taken up with the same enthusiasm. As technology improves, and the skill set of surgeons increases, it seems inevitable to us that more and more will be done. As long as basic oncologic principles are adhered to, and the surgical maxim of conversion if concerned is followed, patients will continue to benefit from this exciting surgery.

95

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
26. Millat B, Atger J, Deleuze A, et al. Laparoscopic treatment for choledocholithiasis: a prospective evaluation in 247 consecutive unselected patients. Hepato-Gastroenterology 1997; 44(13): 28–34. 27. Nathanson LK, O’Rourke NA, Martin IJ, et al. Postoperative ERCP versus laparoscopic choledochotomy for clearance of selected bile duct calculi: a randomized trial. Ann Surg 2005; 242(2): 188–92. 28. Paganini AM, Lezoche E. Follow-up of 161 unselected consecutive patients treated laparoscopically for common bile duct stones. Surg Endosc 1998; 12(1): 23–9. 29. Paganini AM, Guerrieri M, Sarnari J, et al. Thirteen years’ experience with laparoscopic transcystic common bile duct exploration for stones. Effectiveness and long-term results. Surg Endosc 2007; 21(1): 34–40. 30. Petelin JB. Laparoscopic common bile duct exploration. Surg Endosc 2003; 17(11): 1705–15. 31. Rhodes M, Sussman L, Cohen L, Lewis MP. Randomised trial of laparoscopic exploration of common bile duct versus postoperative endoscopic retrograde cholangiography for common bile duct stones. Lancet 1998; 351(9097): 159–61. 32. Rojas-Ortega S, Arizpe-Bravo D, Marín López ER, et al. Transcystic common bile duct exploration in the management of patients with choledocholithiasis. J Gastrointest Surg 2003; 7(4): 492–6. 33. Taylor CJ, Kong J, Ghusn M, et al. Laparoscopic bile duct exploration: results of 160 consecutive cases with 2-year follow up. ANZ J Surg 2007; 77(6): 440–5. 34. Tokumura H, Umezawa A, Cao H, et al. Laparoscopic management of common bile duct stones: transcystic approach and choledochotomy. J Hepat Biliary Pancreatic Surg 2002; 9(2): 206–12. 35. Topal B, Aerts R, Penninckx F. Laparoscopic common bile duct stone clearance with flexible choledochoscopy. Surg Endosc 2007; 21(12): 2317–21. 36. Tranter SE, Thompson MH. Comparison of endoscopic sphincterotomy and laparoscopic exploration of the common bile duct. Br J Surg 2002; 89(12): 1495–504. 37. Martin CJ, Cox MR, Vaccaro L. Laparoscopic transcystic bile duct stenting in the management of common bile duct stones. ANZ J Surg 2002; 72(4): 258–64. 38. Barragan B, Love L, Wachtel M, Griswold JA, Frezza EE. A comparison of anterior and posterior approaches for the surgical treatment of pancreatic pseudocyst using laparoscopic cystogastrostomy. J Laparoendosc Adv Surg Tech Part A 2005; 15(6): 596–600. 39. Dávila-Cervantes A, Gómez F, Chan C, et al. Laparoscopic drainage of pancreatic pseudocysts. Surg Endosc 2004; 18(10): 1420–6. 40. Hindmarsh A, Lewis MPN, Rhodes M. Stapled laparoscopic cystgastrostomy: a series with 15 cases. Surg Endosc 2005; 19(1): 143–7. 41. Makris KI, St Peter SD, Tsao K, Ostlie DJ. Laparoscopic intragastric stapled cystgastrostomy of pancreatic pseudocyst in a child. J Laparoendosc Adv Surg Tech Part A 2008; 18(5): 771–3. 42. Melman L, Azar R, Beddow K, et al. Primary and overall success rates for clinical outcomes after laparoscopic, endoscopic, and open pancreatic cystgastrostomy for pancreatic pseudocysts. Surg Endosc [Internet]. 2008 November 27 [cited 2009 January 11]. Available from: http://www. ncbi.nlm.nih.gov/pubmed/19037696 43. Mori T, Abe N, Sugiyama M, Atomi Y. Laparoscopic pancreatic cystgastrostomy. J Hepat Biliary Pancreatic Surg 2002; 9(5): 548–54. 44. Obermeyer RJ, Fisher WE, Salameh JR, Jeyapalan M, Sweeney JF, Brunicardi FC. Laparoscopic pancreatic cystogastrostomy. Surg Laparosc Endosc Percutaneous Tech 2003; 13(4): 250–3. 45. Palanivelu C, Senthilkumar K, Madhankumar MV, et al. Management of pancreatic pseudocyst in the era of laparoscopic surgery—experience from a tertiary centre. Surg Endosc 2007; 21(12): 2262–7. 46. Teixeira J, Gibbs KE, Vaimakis S, Rezayat C. Laparoscopic Roux-en-Y pancreatic cyst-jejunostomy. Surg Endosc 2003; 17(12): 1910–3. 47. Aljarabah M, Ammori BJ. Laparoscopic and endoscopic approaches for drainage of pancreatic pseudocysts: a systematic review of published series. Surg Endosc 2007; 21(11): 1936–44. 48. Fockens P, Johnson TG, van Dullemen HM, Huibregtse K, Tytgat GN. Endosonographic imaging of pancreatic pseudocysts before endoscopic transmural drainage. Gastrointest Endosc 1997; 46(5): 412–6. 49. Krüger M, Schneider AS, Manns MP, Meier PN. Endoscopic management of pancreatic pseudocysts or abscesses after an EUS-guided 1-step procedure for initial access. Gastrointest Endosc 2006; 63(3): 409–16. 50. Romanelli JR, Desilets DJ, Earle DB. Pancreatic pseudocystgastrostomy with a peroral, flexible stapler: human natural orifice transluminal endoscopic surgery anastomoses in 2 patients (with videos). Gastrointest Endosc 2008; 68(5): 981–7. 51. Conlon KC, Dougherty E, Klimstra DS, et al. The value of minimal access surgery in the staging of patients with potentially resectable peripancreatic malignancy. Ann Surg 1996; 223(2): 134–40. 52. Holzman MD, Reintgen KL, Tyler DS, Pappas TN. The role of laparoscopy in the management of suspected pancreatic and periampullary malignancies. J Gastrointest Surg 1997; 1(3): 236–43; discussion 243–4. 53. Velanovich V, Wollner I, Ajlouni M. Staging laparoscopy promotes increased utilization of postoperative therapy for unresectable intraabdominal malignancies. J Gastrointest Surg 2000; 4(5): 542–6. 54. Stefanidis D, Grove KD, Schwesinger WH, Thomas CR. The current role of staging laparoscopy for adenocarcinoma of the pancreas: a review. Ann Oncol 2006; 17(2): 189–99. 55. Pietrabissa A, Caramella D, Di Candio G, et al. Laparoscopy and laparoscopic ultrasonography for staging pancreatic cancer: critical appraisal. World J Surg 1999; 23(10): 998–1002; discussion 1003. 56. Velasco JM, Rossi H, Hieken TJ, Fernandez M. Laparoscopic ultrasound enhances diagnostic laparoscopy in the staging of intra-abdominal neoplasms. Am Surg 2000; 66(4): 407–11. 57. Jimenez RE, Warshaw AL, Rattner DW, et al. Impact of laparoscopic staging in the treatment of pancreatic cancer. Arch Surg 2000; 135(4): 409–14; discussion 414–5. 58. Schachter PP, Avni Y, Shimonov M, et al. The impact of laparoscopy and laparoscopic ultrasonography on the management of pancreatic cancer. Arch Surg 2000; 135(11): 1303–7. 59. Menack MJ, Spitz JD, Arregui ME. Staging of pancreatic and ampullary cancers for resectability using laparoscopy with laparoscopic ultrasound. Surg Endosc 2001; 15(10): 1129–34. 60. Vollmer CM, Drebin JA, Middleton WD, et al. Utility of staging laparoscopy in subsets of peripancreatic and biliary malignancies. Ann Surg 2002; 235(1): 1–7. 61. Doran HE, Bosonnet L, Connor S, et al. Laparoscopy and laparoscopic ultrasound in the evaluation of pancreatic and periampullary tumours. Dig Surg 2004; 21(4): 305–13. 62. Minnard EA, Conlon KC, Hoos A, et al. Laparoscopic ultrasound enhances standard laparoscopy in the staging of pancreatic cancer. Ann Surg 1998; 228(2): 182–7. 63. Hünerbein M, Rau B, Hohenberger P, Schlag PM. [Value of laparoscopic ultrasound for staging of gastrointestinal tumors]. Der Chirurg; Zeitschrift Für Alle Gebiete Der Operativen Medizen. 2001; 72(8): 914–9. 64. Durup Scheel-Hincke J, Mortensen MB, Qvist N, Hovendal CP. TNM staging and assessment of resectability of pancreatic cancer by laparoscopic ultrasonography. Surg Endosc 1999; 13(10): 967–71. 65. Morganti AG, Brizi MG, Macchia G, et al. The prognostic effect of clinical staging in pancreatic adenocarcinoma. Ann Surg Oncol 2005; 12(2): 145–51. 66. Schlieman MG, Ho HS, Bold RJ. Utility of tumor markers in determining resectability of pancreatic cancer. Arch Surg 2003; 138(9): 951–5; discussion 955–6. 67. Connor S, Bosonnet L, Alexakis N, et al. Serum CA19-9 measurement increases the effectiveness of staging laparoscopy in patients with suspected pancreatic malignancy. Dig Surg 2005; 22(1–2): 80–5. 68. Warshaw AL. Implications of peritoneal cytology for staging of early pancreatic cancer. Am J Surg 1991; 161(1): 26–9; discussion 29–30. 69. Liu RC, Traverso LW. Diagnostic laparoscopy improves staging of pancreatic cancer deemed locally unresectable by computed tomography. Surg Endosc 2005; 19(5): 638–42. 70. Jimenez RE, Warshaw AL, Fernandez-Del Castillo C. Laparoscopy and peritoneal cytology in the staging of pancreatic cancer. J Hepat Biliary Pancreatic Surg 2000; 7(1): 15–20. 71. Fernández-del Castillo CL, Warshaw AL. Pancreatic cancer. Laparoscopic staging and peritoneal cytology. Surg Oncol Clin North Am. 1998; 7(1): 135–42.

96

LAPAROSCOPY IN HPB SURGERY
72. Merchant NB, Conlon KC. Laparoscopic evaluation in pancreatic cancer. Semin Surg Oncol 1998; 15(3): 155–65. 73. Leach SD, Rose JA, Lowy AM, et al. Significance of peritoneal cytology in patients with potentially resectable adenocarcinoma of the pancreatic head. Surgery 1995; 118(3): 472–8. 74. Makary MA, Warshaw AL, Centeno BA, et al. Implications of peritoneal cytology for pancreatic cancer management. Arch Surg 1998; 133(4): 361–5. 75. Schmidt J, Fraunhofer S, Fleisch M, Zirngibl H. Is peritoneal cytology a predictor of unresectability in pancreatic carcinoma? Hepat Gastroenterol 2004; 51(60): 1827–31. 76. Nieveen van Dijkum EJ, Sturm PD, de Wit LT, et al. Cytology of peritoneal lavage performed during staging laparoscopy for gastrointestinal malignancies: is it useful? Ann Surg 1998; 228(6): 728–33. 77. Liu RC, Traverso LW. Laparoscopic staging should be used routinely for locally extensive cancer of the pancreatic head. J Gastrointest Surg 2004; 8(8): 923–4. 78. Burdiles P, Rossi RL. Laparoscopy in pancreatic and hepatobiliary cancer. Surg Onco Clin North Am 2001; 10(3): 531–55, viii. 79. American Joint Committee on Cancer, American Cancer Society, American College of Surgeons. AJCC Cancer Staging Manual, 5th ed. Philadelphia, PA: Lippincott-Raven, 1997: 294. 80. Barreiro CJ, Lillemoe KD, Koniaris LG, et al. Diagnostic laparoscopy for periampullary and pancreatic cancer: what is the true benefit? J Gastrointest Surg 2002; 6(1): 75–81. 81. Brooks AD, Mallis MJ, Brennan MF, Conlon KCP. The value of laparoscopy in the management of ampullary, duodenal, and distal bile duct tumors. J Gastrointest Surg 2002; 6(2): 139–45; discussion 145–6. 82. Shoup M, Winston C, Brennan MF, Bassman D, Conlon KC. Is there a role for staging laparoscopy in patients with locally advanced, unresectable pancreatic adenocarcinoma? J Gastrointest Surg 2004; 8(8): 1068–71. 83. Grobmyer SR, Fong Y, D’Angelica M, et al. Diagnostic laparoscopy prior to planned hepatic resection for colorectal metastases. Arch Surg 2004; 139(12): 1326–30. 84. Jarnagin WR, Conlon K, Bodniewicz J, et al. A clinical scoring system predicts the yield of diagnostic laparoscopy in patients with potentially resectable hepatic colorectal metastases. Cancer 2001; 91(6): 1121–8. 85. Khan AZ, Karanjia ND. The impact of staging laparoscopy prior to hepatic resection for colorectal metastases. Eur J Surg Oncol 2007; 33(8): 1010–3. 86. Mann CD, Neal CP, Metcalfe MS, et al. Clinical Risk Score predicts yield of staging laparoscopy in patients with colorectal liver metastases. Br J Surg 2007; 94(7): 855–9. 87. Metcalfe MS, Close JS, Iswariah H, et al. The value of laparoscopic staging for patients with colorectal metastases. Arch Surg 2003; 138(7): 770–2. 88. Pilkington SA, Rees M, Peppercorn D, John TG. Laparoscopic staging in selected patients with colorectal liver metastases as a prelude to liver resection. HPB 2007; 9(1): 58–63. 89. Rahusen FD, Cuesta MA, Borgstein PJ, et al. Selection of patients for resection of colorectal metastases to the liver using diagnostic laparoscopy and laparoscopic ultrasonography. Ann Surg 1999; 230(1): 31–7. 90. Thaler K, Kanneganti S, Khajanchee Y, et al. The evolving role of staging laparoscopy in the treatment of colorectal hepatic metastasis. Arch Surg 2005; 140(8): 727–34. 91. Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 230(3): 309– 18; discussion 318–21. 92. D’Angelica M, Jarnagin W, Dematteo R, et al. Staging laparoscopy for potentially resectable noncolorectal, nonneuroendocrine liver metastases. Ann Surg Oncol 2002; 9(2): 204–9. 93. Goere D, Wagholikar GD, Pessaux P, et al. Utility of staging laparoscopy in subsets of biliary cancers : laparoscopy is a powerful diagnostic tool in patients with intrahepatic and gallbladder carcinoma. Surg Endosc 2006; 20(5): 721–5. 94. Weber SM, DeMatteo RP, Fong Y, Blumgart LH, Jarnagin WR. Staging laparoscopy in patients with extrahepatic biliary carcinoma. Analysis of 100 patients. Ann Surg 2002; 235(3): 392–9. 95. Connor S, Barron E, Wigmore SJ, et al. The utility of laparoscopic assessment in the preoperative staging of suspected hilar cholangiocarcinoma. J Gastrointest Surg 2005; 9(4): 476–80. 96. Lai ECH, Tang CN, Ha JPY, Tsui DKK, Li MKW. The evolving influence of laparoscopy and laparoscopic ultrasonography on patients with hepatocellular carcinoma. Am J Surg 2008; 196(5): 736–40. 97. Lang BHH, Poon RTP, Fan ST, Wong J. Influence of laparoscopy on postoperative recurrence and survival in patients with ruptured hepatocellular carcinoma undergoing hepatic resection. Br J Surg 2004; 91(4): 444–9. 98. Lo CM, Lai EC, Liu CL, Fan ST, Wong J. Laparoscopy and laparoscopic ultrasonography avoid exploratory laparotomy in patients with hepatocellular carcinoma. Ann Surg 1998; 227(4): 527–32. 99. Lo CM, Fan ST, Liu CL, et al. Determining resectability for hepatocellular carcinoma: the role of laparoscopy and laparoscopic ultrasonography. J Hepat Biliary Pancreatic Surg 2000; 7(3): 260–4. 100. Weitz J, D’Angelica M, Jarnagin W, et al. Selective use of diagnostic laparoscopy prior to planned hepatectomy for patients with hepatocellular carcinoma. Surgery 2004; 135(3): 273–81. 101. Espat NJ, Brennan MF, Conlon KC. Patients with laparoscopically staged unresectable pancreatic adenocarcinoma do not require subsequent surgical biliary or gastric bypass. J Am Coll Surg 1999; 188(6): 649–55; discussion 655–7. 102. Mehta S, Hindmarsh A, Cheong E, et al. Prospective randomized trial of laparoscopic gastrojejunostomy versus duodenal stenting for malignant gastric outflow obstruction. Surg Endosc 2006; 20(2): 239–42. 103. Jeurnink SM, Steyerberg EW, Hof GV, et al. Gastrojejunostomy versus stent placement in patients with malignant gastric outlet obstruction: a comparison in 95 patients. J Surg Oncol 2007; 96(5): 389–96. 104. Moss AC, Morris E, Mac Mathuna P. Palliative biliary stents for obstructing pancreatic carcinoma. Cochrane Database of Systematic Reviews (Online) 2006; (2): CD004200. 105. Davids PH, Groen AK, Rauws EA, Tytgat GN, Huibregtse K. Randomised trial of self-expanding metal stents versus polyethylene stents for distal malignant biliary obstruction. Lancet 1993; 340(8834–8835): 1488–92. 106. Weber A, Mittermeyer T, Wagenpfeil S, Schmid RM, Prinz C. Selfexpanding metal stents versus polyethylene stents for palliative treatment in patients with advanced pancreatic cancer. Pancreas 2009; 38(1): e7–e12. 107. Casaccia M, Diviacco P, Molinello P, Danovaro L, Casaccia M. Laparoscopic palliation of unresectable pancreatic cancers: preliminary results. Eur J Surg Acta Chirurg 1999; 165(6): 556–9. 108. Fletcher DR, Jones RM. Laparoscopic cholecystjejunostomy as palliation for obstructive jaundice in inoperable carcinoma of pancreas. Surg Endosc 1992; 6(3): 147–9. 109. Kazanjian KK, Reber HA, Hines OJ. Laparoscopic gastrojejunostomy for gastric outlet obstruction in pancreatic cancer. Am Surg 2004; 70(10): 910–3. 110. Khan AZ, Miles WFA, Singh KK. Initial experience with laparoscopic bypass for upper gastrointestinal malignancy: a new option for palliation of patients with advanced upper gastrointestinal tumors. J Laparoendosc Adv Surg Techn Part A 2005; 15(4): 374–8. 111. Kuriansky J, Sáenz A, Astudillo E, Cardona V, Fernández-Cruz L. Simultaneous laparoscopic biliary and retrocolic gastric bypass in patients with unresectable carcinoma of the pancreas. Surg Endosc 2000; 14(2): 179–81. 112. Nagy A, Brosseuk D, Hemming A, Scudamore C, Mamazza J. Laparoscopic gastroenterostomy for duodenal obstruction. Am J Surg 1995; 169(5): 539–42. 113. Navarra G, Musolino C, Venneri A, De Marco ML, Bartolotta M. Palliative antecolic isoperistaltic gastrojejunostomy: a randomized controlled trial comparing open and laparoscopic approaches. Surg Endosc 2006; 20(12): 1831–4. 114. Rhodes M, Nathanson L, Fielding G. Laparoscopic biliary and gastric bypass: a useful adjunct in the treatment of carcinoma of the pancreas. Gut 1995; 36(5): 778–80. 115. Röthlin MA, Schöb O, Weber M. Laparoscopic gastro- and hepaticojejunostomy for palliation of pancreatic cancer: a case controlled study. Surg Endosc 1999; 13(11): 1065–9.

97

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
116. Shimi S, Banting S, Cuschieri A. Laparoscopy in the management of pancreatic cancer: endoscopic cholecystojejunostomy for advanced disease. Br J Surg 1992; 79(4): 317–9. 117. Tarnasky PR, England RE, Lail LM, Pappas TN, Cotton PB. Cystic duct patency in malignant obstructive jaundice. An ERCP-based study relevant to the role of laparoscopic cholecystojejunostomy. Ann Surg 1995; 221(3): 265–71. 118. Warshaw AL. Conservation of the spleen with distal pancreatectomy. Arch Surg 1988; 123(5): 550–3. 119. Eom B, Jang J, Lee S, et al. Clinical outcomes compared between laparoscopic and open distal pancreatectomy. Surg Endosc [Internet] 2007 November 20 [cited 2009 January 3]. Available from: http://www.ncbi. nlm.nih.gov/pubmed/18027035 120. Kooby DA, Gillespie T, Bentrem D, et al. Left-sided pancreatectomy: a multicenter comparison of laparoscopic and open approaches. Ann Surg 2008; 248(3): 438–46. 121. Matsumoto T, Shibata K, Ohta M, et al. Laparoscopic distal pancreatectomy and open distal pancreatectomy: a nonrandomized comparative study. Surg Laparosc Endosc Percutaneous Techn 2008; 18(4): 340–3. 122. Nakamura Y, Uchida E, Aimoto T, et al. Clinical outcome of laparoscopic distal pancreatectomy. J Hepat Biliary Pancreatic Surg [Internet]. 2008 December 16 [cited 2009 January 3]. Available from: http://www. ncbi.nlm.nih.gov/pubmed/19083146 123. Tang CN, Tsui KK, Ha JPY, Wong DCT, Li MKW. Laparoscopic distal pancreatectomy: a comparative study. Hepat Gastroenterol 2007; 54(73): 265–71. 124. Velanovich V. Case-control comparison of laparoscopic versus open distal pancreatectomy. J Gastrointest Surg 2006; 10(1): 95–8. 125. Velanovich V. The use of tissue sealant to prevent fistula formation after laparoscopic distal pancreatectomy. Surg Endosc 2007; 21(7): 1222. 126. Thaker RI, Matthews BD, Linehan DC, et al. Absorbable mesh reinforcement of a stapled pancreatic transection line reduces the leak rate with distal pancreatectomy. J Gastrointest Surg 2007; 11(1): 59–65. 127. Fischer CP, Bass B, Fahy B, Aloia T. Transampullary pancreatic duct stenting decreases pancreatic fistula rate following left pancreatectomy. Hepat Gastroenterol 2008; 55(81): 244–8. 128. Bilimoria MM, Cormier JN, Mun Y, et al. Pancreatic leak after left pancreatectomy is reduced following main pancreatic duct ligation. Br J Surg 2003; 90(2): 190–6. 129. Fernández-Cruz L, Martínez I, Cesar-Borges G, et al. Laparoscopic surgery in patients with sporadic and multiple insulinomas associated with multiple endocrine neoplasia type 1. J Gastrointest Surg 2005; 9(3): 381–8. 130. Taylor C, O’Rourke N, Nathanson L, et al. Laparoscopic distal pancreatectomy: the Brisbane experience of forty-six cases. HPB 2008; 10(1): 38–42. 131. Sa Cunha A, Rault A, Beau C, Collet D, Masson B. Laparoscopic central pancreatectomy: single institution experience of 6 patients. Surgery 2007; 142(3): 405–9. 132. Sweet MP, Izumisato Y, Way LW, et al. Laparoscopic enucleation of insulinomas. Arch Surg 2007; 142(12): 1202–4; discussion 1205. 133. Toniato A, Meduri F, Foletto M, Avogaro A, Pelizzo M. Laparoscopic treatment of benign insulinomas localized in the body and tail of the pancreas: a single-center experience. World J Surg 2006; 30(10): 1916–9; discussion 1920–1. 134. Cuschieri SA, Jakimowicz JJ. Laparoscopic pancreatic resections. Semin Laparosc Surg 1998; 5(3): 168–79. 135. Dulucq JL, Wintringer P, Stabilini C, et al. Are major laparoscopic pancreatic resections worthwhile? A prospective study of 32 patients in a single institution. Surg Endosc 2005; 19(8): 1028–34. 136. Mabrut J, Fernandez-Cruz L, Azagra JS, et al. Laparoscopic pancreatic resection: results of a multicenter European study of 127 patients. Surgery 2005; 137(6): 597–605. 137. Sa Cunha A, Rault A, Beau C, et al. A single-institution prospective study of laparoscopic pancreatic resection. Arch Surg 2008; 143(3): 289–95; discussion 295. 138. Rau HG, Meyer G, Cohnert TU, et al. Laparoscopic liver resection with the water-jet dissector. Surg Endosc 1995; 9(9): 1009–12. 139. Cuesta MA, Meijer S, Paul MA, de Brauw LM. Limited laparoscopic liver resection of benign tumors guided by laparoscopic ultrasonography: report of two cases. Surg Laparosc Endosc 1995; 5(5): 396–401. 140. Hashizume M, Takenaka K, Yanaga K, et al. Laparoscopic hepatic resection for hepatocellular carcinoma. Surg Endosc 1995; 9(12): 1289–91. 141. Azagra JS, Goergen M, Gilbart E, Jacobs D. Laparoscopic anatomical (hepatic) left lateral segmentectomy-technical aspects. Surg Endosc 1996; 10(7): 758–61. 142. Kaneko H, Takagi S, Shiba T. Laparoscopic partial hepatectomy and left lateral segmentectomy: technique and results of a clinical series. Surgery 1996; 120(3): 468–75. 143. Hüscher CG, Lirici MM, Chiodini S. Laparoscopic liver resections. Semin Laparosc Surg 1998; 5(3): 204–10. 144. Cherqui D, Husson E, Hammoud R, et al. Laparoscopic liver resections: a feasibility study in 30 patients. Ann Surg 2000; 232(6): 753–62. 145. Koffron AJ, Auffenberg G, Kung R, Abecassis M. Evaluation of 300 minimally invasive liver resections at a single institution: less is more. Ann Sur 2007; 246(3): 385–92; discussion 392–4. 146. Buell JF, Thomas MT, Rudich S, et al. Experience with more than 500 minimally invasive hepatic procedures. Ann Surg 2008; 248(3): 475–86. 147. Cherqui D, Laurent A, Tayar C, et al. Laparoscopic liver resection for peripheral hepatocellular carcinoma in patients with chronic liver disease: midterm results and perspectives. Ann Surg 2006; 243(4): 499–506. 148. Dagher I, Proske JM, Carloni A, et al. Laparoscopic liver resection: results for 70 patients. Surg Endosc 2007; 21(4): 619–24. 149. Belli G, Fantini C, D’Agostino A, et al. Laparoscopic left lateral hepatic lobectomy: a safer and faster technique. J Hepat Biliary Pancreatic Surg 2006; 13(2): 149–54. 150. O’Rourke N, Shaw I, Nathanson L, Martin I, Fielding G. Laparoscopic resection of hepatic colorectal metastases. HPB 2004; 6(4): 230–5. 151. Chang S, Laurent A, Tayar C, Karoui M, Cherqui D. Laparoscopy as a routine approach for left lateral sectionectomy. Br J Surg 2007; 94(1): 58–63. 152. Ardito F, Tayar C, Laurent A, et al. Laparoscopic liver resection for benign disease. Arch Surg 2007; 142(12): 1188–93; discussion 1193. 153. Bachellier P, Ayav A, Pai M, et al. Laparoscopic liver resection assisted with radiofrequency. Am J Surg 2007; 193(4): 427–30. 154. Berends FJ, Meijer S, Prevoo W, Bonjer HJ, Cuesta MA. Technical considerations in laparoscopic liver surgery. Surg Endosc 2001; 15(8): 794–8. 155. Champault A, Dagher I, Vons C, Franco D. Laparoscopic hepatic resection for hepatocellular carcinoma. Retrospective study of 12 patients. Gastroentérol Clin Biol 2005; 29(10): 969–73. 156. Chen H, Juan C, Ker C. Laparoscopic liver surgery for patients with hepatocellular carcinoma. Ann Surg Oncol 2008; 15(3): 800–6. 157. Cherqui D, Soubrane O, Husson E, et al. Laparoscopic living donor hepatectomy for liver transplantation in children. Lancet 2002; 359(9304): 392–6. 158. Dagher I, Lainas P, Carloni A, et al. Laparoscopic liver resection for hepatocellular carcinoma. Surg Endosc 2008; 22(2): 372–8. 159. Descottes B, Lachachi F, Sodji M, et al. Early experience with laparoscopic approach for solid liver tumors: initial 16 cases. Ann Surg 2000; 232(5): 641–5. 160. Descottes B, Glineur D, Lachachi F, et al. Laparoscopic liver resection of benign liver tumors. Surg Endosc 2003; 17(1): 23–30. 161. Dulucq JL, Wintringer P, Stabilini C, Berticelli J, Mahajna A. Laparoscopic liver resections: a single center experience. Surg Endosc 2005; 19(7): 886–91. 162. Gayet B, Cavaliere D, Vibert E, et al. Totally laparoscopic right hepatectomy. Am J Surg 2007; 194(5): 685–9. 163. Gigot J, Glineur D, Santiago Azagra J, et al. Laparoscopic liver resection for malignant liver tumors: preliminary results of a multicenter European study. Ann Surg 2002; 236(1): 90–7. 164. Inagaki H, Kurokawa T, Nonami T, Sakamoto J. Hand-assisted laparoscopic left lateral segmentectomy of the liver for hepatocellular carcinoma with cirrhosis. J Hepato Biliary Pancreatic Surg 2003; 10(4): 295–8. 165. Kaneko H. Laparoscopic hepatectomy: indications and outcomes. J Hepato Biliary Pancreatic Surg 2005; 12(6): 438–43. 166. Koffron AJ, Kung R, Baker T, et al. Laparoscopic-assisted right lobe donor hepatectomy. Am J Transplant 2006; 6(10): 2522–5. 167. Mala T, Edwin B, Rosseland AR, et al. Laparoscopic liver resection: experience of 53 procedures at a single center. Journal of Hepat Biliary Pancreatic Surg 2005; 12(4): 298–303.

98

LAPAROSCOPY IN HPB SURGERY
168. O’Rourke N, Fielding G. Laparoscopic right hepatectomy: surgical technique. J Gastrointest Surg 2004; 8(2): 213–6. 169. Poultsides G, Brown M, Orlando R. Hand-assisted laparoscopic management of liver tumors. Surg Endosc 2007; 21(8): 1275–9. 170. Shimada M, Hashizume M, Maehara S, et al. Laparoscopic hepatectomy for hepatocellular carcinoma. Surg Endosc 2001; 15(6): 541–4. 171. Soubrane O, Cherqui D, Scatton O, et al. Laparoscopic left lateral sectionectomy in living donors: safety and reproducibility of the technique in a single center. Ann Surg 2006; 244(5): 815–20. 172. Tang CN, Tsui KK, Ha JPY, Yang GPY, Li MKW. A single-centre experience of 40 laparoscopic liver resections. Hong Kong Med J 2006; 12(6): 419–25. 173. Vibert E, Perniceni T, Levard H, et al. Laparoscopic liver resection. Br J Surg 2006; 93(1): 67–72. 174. Belli G, Fantini C, D’Agostino A, et al. Laparoscopic versus open liver resection for hepatocellular carcinoma in patients with histologically proven cirrhosis: short- and middle-term results. Surg Endosc 2007; 21(11): 2004–11. 175. Kaneko H, Takagi S, Otsuka Y, et al. Laparoscopic liver resection of hepatocellular carcinoma. Am J Surg 2005; 189(2): 190–4. 176. Laurent A, Cherqui D, Lesurtel M, et al. Laparoscopic liver resection for subcapsular hepatocellular carcinoma complicating chronic liver disease. Arch Surg 2003; 138(7): 763–9; discussion 769. 177. Lee KF, Cheung YS, Chong CN, et al. Laparoscopic versus open hepatectomy for liver tumours: a case control study. Hong Kong Med J 2007; 13(6): 442–8. 178. Abu Hilal M, McPhail MJW, Zeidan B, et al. Laparoscopic versus open left lateral hepatic sectionectomy: A comparative study. Eur J Surg Oncol 2008; 34(12): 1285–8. 179. Cai X, Wang Y, Yu H, Liang X, Peng S. Laparoscopic hepatectomy for hepatolithiasis: a feasibility and safety study in 29 patients. Surg Endosc 2007; 21(7): 1074–8. 180. Cai XJ, Yang J, Yu H, et al. Clinical study of laparoscopic versus open hepatectomy for malignant liver tumors. Surg Endosc 2008; 22(11): 2350–6. 181. Farges O, Jagot P, Kirstetter P, Marty J, Belghiti J. Prospective assessment of the safety and benefit of laparoscopic liver resections. J Hepat Biliary Pancreatic Surg 2002; 9(2): 242–8. 182. Lesurtel M, Cherqui D, Laurent A, Tayar C, Fagniez PL. Laparoscopic versus open left lateral hepatic lobectomy: a case-control study. J Am Coll Surg 2003; 196(2): 236–42. 183. Mala T, Edwin B, Gladhaug I, et al. A comparative study of the shortterm outcome following open and laparoscopic liver resection of colorectal metastases. Surg Endosc 2002; 16(7): 1059–63. 184. Morino M, Morra I, Rosso E, Miglietta C, Garrone C. Laparoscopic vs open hepatic resection: a comparative study. Surg Endosc 2003; 17(12): 1914–8. 185. Troisi R, Montalti R, Smeets P, et al. The value of laparoscopic liver surgery for solid benign hepatic tumors. Surg Endosc 2008; 22(1): 38–44. 186. Laurent A, Tayar C, Andréoletti M, et al. Laparoscopic liver resection facilitates salvage liver transplantation for hepatocellular carcinoma. J Hepatobiliary Pancreat Surg 2009; 16(3): 310–14.

99

11

Cross-sectional imaging for HPB disorders (MRI and CT) Lawrence H. Schwartz
In contrast to hepatic veins, which course between liver segments, portal veins and their accompanying artery and bileduct branches define the center of each segment. The main portal vein (PV) typically branches extrahepatically into the left and right portal veins (LPV and RPV, respectively); the RPV divides into the anterior and posterior sectoral branches, while the LPV enters the umbilical fissure and provides branches to the left lobe. While this describes standard anatomy, approximately one-third of patients have variant portal venous anatomy (Table 11.1) (4). In standard arterial anatomy the common hepatic artery arises from the celiac axis and divides into the gastroduodenal artery and the proper hepatic artery. The proper hepatic artery then gives rise to the right and left hepatic arteries. However, nearly one-half of patients have variations in their hepatic arterial system. A list of arterial variants demonstrated in a recent study is shown in Table 11.2 (1). Pancreas The pancreas is divided into the head, uncinate process, neck, body, and tail. These divisions are based upon external landmarks such as the superior mesenteric vein (SMV), which lies under the neck of the gland and defines the separation between the head and the body. The pancreas has a long course through the retroperitoneum and is intimately associated with multiple organs including the transverse colon, stomach, duodenum, and spleen. Additionally, it has a close relationship with the portal venous system that, in part, defines the resectability of pancreatic masses. The pancreas has a rich arterial supply that comes from multiple branches of the celiac and superior mesenteric arteries. As in hepatic surgery, preoperative determination of variations in peripancreatic vascular anatomy can greatly aid in operative planning.

overview
The maturation of surgery of the liver, biliary tract, and pancreas as field unto itself has happened concomitantly with, and partly as a result of, advances in cross-sectional imaging. Rather than relying on expected anatomy based on textbooks, surgeons can now plan operations based on the precise anatomical details of each individual patient. The ability to predict anatomical variations, which are present in nearly one-half of the patients, has taken the element of surprise out of the operating room and can help reduce operative morbidity (1). In this chapter, we briefly discuss the application of computed tomography (CT) and magnetic resonance imaging (MRI) to the surgical management of disease of the liver, biliary tract, and pancreas. We begin by reviewing the relevant cross-sectional anatomy of the organs being studied. Next, we discuss the various techniques used to obtain high-quality CT and MRI images of the liver, biliary tract, and pancreas. Finally, we review the cross-sectional imaging characteristics of important pathological entities commonly encountered by surgeons caring for patients with diseases of the liver, biliary tract, and pancreas.

cross-sectional anatomy
The effective use of cross-sectional imaging in the diagnosis and treatment of disorders of the liver, biliary tract, and pancreas mandates a strong understanding of anatomy. Developing this understanding of the complex three-dimensional structures and being able to extrapolate from two-dimensional representations requires a structured approach. A more detailed description of relevant anatomy can be reviewed in the previous chapters and elsewhere, therefore we will here focus on the interpretation of cross-sectional anatomy and its relation to in situ anatomy (2). Liver and Biliary Tract Segmental liver anatomy is the basis for modern liver surgery; therefore, we provide a framework with which to define this anatomy for each patient based upon cross-sectional imaging. Cantlie’s plane, also known as the main portal fissure, is an imaginary plane drawn from the gallbladder fossa toward the inferior vena cava (IVC) that divides the anatomical right and left lobes of the liver. The course of the middle hepatic vein (MHV) is fairly constant and lies in this otherwise potentially avascular plane; therefore, it can be used to delineate the two lobes on cross-sectional imaging (3). The right hepatic vein (RHV) defines the plane separating the anterior (segments 5 and 8) and posterior sectors (segments 6 and 7) of the right lobe. While the plane of the left hepatic vein (LHV) separates the anterior and posterior sectors of the left lobe, its anatomy is highly variable, making it a less useful landmark.

technique
Liver and Biliary Tract CT is commonly used as the primary modality to detect, characterize, and follow hepatic or biliary pathology. Modern multislice helical CT scanners allow for the rapid acquisition of large volumes of data in a single patient breath-hold, thereby allowing for the construction of high-resolution axial, coronal, sagittal, and three-dimensional images. Noncontrast CT allows us to make determinations about the character of the liver parenchyma based on changes in density. This is useful for detecting global hepatic abnormalities; however, it does not allow for the precise delineation of hepatic vascular structures nor the detection and characterization of subtle hepatic masses. CT examination of the liver, therefore, relies on iodinated contrast enhancement. Accurate CT imaging requires achieving maximal differences in attenuation between tissues, therefore understanding the contrast enhancement

100

CROSS-SECTIONAL IMAGING FOR HPB DISORDERS (MRI AND CT)
characteristics of the liver and each type of liver tumor is essential. The liver receives approximately 20% of its blood from the hepatic artery and the remaining 80% from the portal vein. Since intravenously injected contrast reaches the liver via the hepatic artery before it does via the portal vein, and takes some time to reach a state of equilibrium, a triphasic CT scan based on hepatic arterial, portal venous, and equilibrium phases is favored for examination of the liver. Although CT remains the most commonly used modality for obtaining cross-sectional images of the liver because of its lower cost and its greater ease of interpretation by clinicians, the indications for liver MRI continue to grow. As compared with triphasic liver CT, liver MRI has the advantages of not exposing patients to ionizing radiation, a greater measure of safety in patients with renal insufficiency, and an improved ability to characterize certain types of lesions. However, MRI is costly, more time intensive for patients, and contraindicated in patients with certain metal implants. In many centers, magnetic resonance cholangiopancreatography (MRCP) has nearly replaced diagnostic percutaneous transhepatic cholangiography (PTC) and endoscopic retrograde cholangiopancreatography (ERCP), thus reserving the latter studies for situations in which there is therapeutic intent or in which there is a need for tissue diagnosis. Using heavily T2-weighted sequences, MRCP represents stationary water with high signal intensity (5). As MRCP does not require the administration or biliary excretion of contrast, it works well even in the setting of hepatic dysfunction or obstructive jaundice. Pancreas As for liver and biliary tract imaging, contrast-enhanced CT remains the primary modality used in the setting of pancreatic disease; however, MRI again has some advantages. While CT has higher spatial resolution, MRI may have a better ability to characterize lesions based on tissue composition. Optimal CT imaging of the pancreas relies on the ability of multidetector CT scanners to rapidly capture large volumes of information during specific time periods after IV contrast administration. Thin slices and the ability to reformat images in multiple axes are helpful in preoperative preparation. Furthermore, water is administered as an oral contrast agent to improve differentiation among bowel, pancreas parenchyma,

Table 11.1 Anatomic Variations in Portal Vein Anatomy in 200 Patients
Patients Type 1 2 3(Z) 4 5 Portal vein variant Standard anatomy Trifurcation Right posterior portal vein as first branch of main portal vein Segment VII branch as separate branch of right portal vein Segment VI branch as separate branch of right portal vein Other No. 130 18 26 2 12 12 % 65 9 13 1 6 6

Table 11.2 Frequency of Different Arterial Variants Seen at CT Angiography in 371 Patients
Type of finding Classic celiac arterial anatomy Replaced RHA off SMA Replaced LHA off LGA Artery to segments 2 and 3 off LGA Artery to segments 4A and 4B off RHA Trifurcation of CHA into GDA, RHA, and LHA RHA off celiac axis Accessory LHA off LGA LGA directly oft abdominal aorta CHA off SMA CHA directly off the aorta RHA off GDA Accessory RHA Common trunk of celiac axis and SMA Medial and lateral branches separate off CHA LHA off CHA GDA off RHA SMA gives rise to GDA LHA off celiac axis RHA off aorta Segment 4 branch off GDA Extrahepatic branching of RHA into anterior and posterior with artery to segment 4 off anterior division of RHA No. of findings (o = 394|) 188 54 30 19 17 15 13 13 11 6 6 5 3 2 2 2 2 2 1 1 1 1 % of patients (n = 371) 51 15 8 5 5 4 4 4 3 2 2 1 1 <1 <1 <1 <1 <1 <1 <1 <1 <1

Note: Twenty patients had two variants seen at CT and one patient had four variants. Abbreviations: LHA, left hepatic artery; RHA, right hepatic artery; LGA, left gastric artery; SMA, superior mesenteric artery; CHA, common hepatic artery; GDA, gastroduodenal artery. Source: Reprinted with permission from Ref. 4.

101

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
and blood vessels. Importantly, modern CT scans may have even greater ability than endoscopic ultrasound to determine the involvement of major vascular structures by periampullary pancreatic cancers (6). MRI is of particular use in patients with contrast allergies or renal insufficiency, although optimal imaging with MRI still requires the administration of gadolinium as a contrast agent. As for biliary pathology, MRCP is often useful in assessing biliary and pancreatic ductal obstruction due to pancreatic masses. In clinical practice, these techniques are often supplemented by ERCP and endoscopic ultrasound. Hepatic Hemangioma Hepatic hemangiomata are common vascular lesions of the liver that receive their blood supply from the hepatic artery. Hemangiomata rarely cause symptoms; however, giant ones can be associated with abdominal pain or other compressive symptoms (7). Hemangiomata are diagnosed based on their nodular, clump-like pattern of early arterial enhancement on CT (Fig. 11.2). Although small ones are fairly homogeneous in appearance, large hemangiomata may have a heterogeneous appearance due to areas of thrombosis. MRI is the most accurate imaging modality for diagnosing hepatic hemangiomata. On T2-weighted imaging, they are hyperintense and have a lobulated appearance. Administration of gadolinium again shows early peripheral nodular enhancement. Focal Nodular Hyperplasia Focal nodular hyperplasia (FNH) is a common benign liver tumor made up of all elements of the hepatic parenchyma. FNH are completely benign and rarely, if ever, lead to symptoms. However, the fibrolamellar variant of hepatocellular carcinoma may be mistaken for FNH based on similar imaging characteristics (8). Therefore, accurate identification of FNH is of paramount importance. On pathological examination, FNH typically have a central scar that may be demonstrated on cross-sectional imaging. Contrast-enhanced CT show rapid homogeneous enhancement during the arterial phase with reduced attenuation during the portal venous phase (Fig. 11.3). MRI imaging of FNH reveals isointensity or slight T1 hypointensity or T2 hyperintensity, with a central scar that has even less T1 intensity or more T2 intensity (9). Contrast administration shows early enhancement with delayed enhancement of the central scar. Hepatocellular Adenoma Hepatocellular adenomas are benign proliferations of hepatocytes with a dramatically increased prevalence in patients with a history of oral contraceptive use. Although they are benign lesions, resection of hepatocellular adenomas is recommended because of their propensity for hemorrhage and, albeit rare, risk of malignant transformation. Adenomas are recognized

cross-sectional imaging characteristics of liver and biliary tract lesions
Cysts Nonparasitic simple hepatic cysts are fluid-filled thin-walled benign lesions that have no malignant potential, and are found in 1% to 5% of the population. Although treatment of large hepatic cysts may be undertaken to relieve compressive symptoms, most cysts require no treatment at all. Hepatic cysts are recognized on CT imaging by their spherical or near-spherical shape, water-attenuation fluid contents, and barely visible wall that lacks contrast enhancement. By MRI, simple cysts are homogeneous and have low-T1 and high-T2 signal intensity. Multiple hepatic cysts may also be present in the setting of polycystic kidney disease. Distinguishing simple cysts from cystadenoma, which is a very rare tumor, is important in that the latter lesion has the potential to compress the bile ducts, bleed, or develop into a cystadenocarcinoma. On crosssectional imaging, cystadenomas may demonstrate internal septations and a thick wall that enhances with contrast administration (Fig. 11.1). Echinococcal or hydatid cysts are common in certain parts of the world that are endemic for this echinococcus granulosis, which is a parasitic disease transmitted from dogs. Hydatid cysts are recognized as being well-circumscribed cystic lesions that often contain multiple smaller cysts known as daughter cysts. As the primary treatment for hydatid disease consists of administering the anthelmintic agent albendazole, recognition of this entity based on imaging characteristics is essential.

(A)

(B)

Figure 11.1 Biliary cystadenoma. (A) T2- and (B) postcontrast T1-weighted images of a biliary cystadenoma hanging off the inferior portion of the right lobe of the liver. Arrows indicate solid enhancing component of mass distinguishing this from a simple cyst.

102

CROSS-SECTIONAL IMAGING FOR HPB DISORDERS (MRI AND CT)

(A)

(B)

Figure 11.2 Hemangioma. (A) T1-weighted postcontrast imaging reveals a nodular peripheral enhancement (black arrow) pattern in the early arterial phase that is characteristic of hemangiomas. (B) T2-weighted imaging reveals a hyperintense, lobulated lesion.

(A)

(B)

(C)

(D)

Figure 11.3 Focal Nodular Hyperplasia (A) T1-weighted precontrast image is isointense to hepatic parenchyma (B) T2-weighted image is also isointense to hepatic parenchyma except the central scar (arrow), which is bright (C) T1-weighted postcontrast image in the arterial phase demonstrates homogenous, hyperintense enhancement with the central scar enhancing on (D) delayed postcontrast images.

103

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
on CT as hypervascular, heterogeneous lesions during arterial phase that become isodense or hypodense during the portal venous phase. MRI also shows hepatocellular adenomas to be heterogeneous, with T1 hyperintensity due to the presence of fat or hemorrhage and typically show arterial phase enhancement as with CT. Hepatocellular Carcinoma Hepatocellular carcinoma (HCC) ranks among the most common causes of cancer-related mortality worldwide; however, its incidence is markedly variable based on geography and prevalence of hepatitis B and C virus infection (10). As it occurs most commonly in the setting of cirrhosis, its radiological diagnosis can be challenging. This is due to the presence of fibrosis and regenerative nodules that can be difficult to distinguish from dysplastic nodules or HCC. Contrastenhanced CT helps provide some distinction, as small dysplastic nodules or HCC that can be mistaken for regenerative nodules typically enhance during the arterial phase and have contrast washout in the delayed venous phase (11). Larger HCC are more heterogeneous in their appearance and may not demonstrate contrast enhancement. MRI adds sensitivity to the diagnosis by showing differences in signal intensity between areas of carcinoma and cirrhotic liver (12). Although HCC are typically hypointense on T1-weighted images, welldifferentiated HCC may be hyperintense. T2-weighted images typically show HCC as hypointense lesions, however, this is variable as well. HCC usually enhance with gadolinium administration (Fig. 11.4). Fibrolamellar Carcinoma Fibrolamellar carcinoma (FLC) is a rare malignant tumor of the liver that typically arises in the absence of cirrhosis in relatively young patients (13). FLC is thought to be a variant of HCC, and is therefore also referred to as fibrolamellar HCC. On CT, FLC are usually large, hypoattenuating tumors with heterogeneous contrast enhancement and a nonenhancing central scar. The central fibrous scar usually show low signal intensity on both T1-weighted and T2-weighted images. Accurate diagnosis is essential since FLC may mimic FNH on cross-sectional imaging due to the presence of a fibrous central scar on both. Cholangiocarcinoma Adenocarcinoma that develops from epithelial cells lining the intrahepatic and extrahepatic bile ducts is termed cholangiocarcinoma. Hilar cholangiocarcinomas (Klatskin tumors), which arise at the confluence of the right and left hepatic ducts, are the most common type and typically present with jaundice (14). CT or MRI of patients with hilar cholangiocarcinoma shows intrahepatic biliary ductal dilatation, often in association with unilobar parenchymal atrophy, bile duct crowding, and portal vein impingement. An associated mass lesion may or may not be present, while a more prognostically favorable papillary variant may show a nodular mass within the biliary system (15). Extrahepatic cholangiocarcinoma, which arises in the common hepatic or common bile ducts, also presents with jaundice. Cross-sectional imaging demonstrates both intrahepatic and extrahepatic biliary ductal dilatation. Again, a mass lesion is rarely present, although papillary lesions are possible in this location as well. More commonly, the distal bile duct will show an area of focal thickening that enhances with contrast administration. Peripheral (intrahepatic) cholangiocarcinoma has an appearance on cross-sectional imaging that is more similar to that of more commonly encountered liver tumors. Biliary ductal dilatation is only focal in association with a low attenuation mass that shows peripheral enhancement (Fig. 11.5). Gallbladder Carcinoma Gallbladder carcinoma is a highly aggressive tumor that is notable for its highly variable incidence. Early gallbladder carcinoma may have few findings on cross-sectional imaging. CT of early lesions may demonstrate focal gallbladder-wall thickening or a polypoid mass within the lumen of the gallbladder. More advanced tumors may show a hypoattenuating mass in or replacing the gallbladder, which may be associated with hepatic involvement or biliary ductal dilatation. T2-weighted MRI images show heterogeneous signal intensity with irregular contrast enhancement (16) (Fig. 11.6). Metastatic Cancer to the Liver The most common indication for liver resection in the western world is metastatic disease, especially from colorectal cancer. Liver metastases from colorectal cancer, as well as those from other GI malignancies, are typically characterized by low attenuation relative to normal liver parenchyma; however, there is high variability in their appearance. Colorectal liver metastases are most readily appreciated on portal venous phase CT (Fig. 11.6), and have variable levels of rim enhancement. By contrast, neuroendocrine (Fig. 11.7) and other hypervascular metastases tend to show early arterial enhancement (17). MRI may help to characterize liver metastases, which are typically of low signal intensity on T1-weighted images and high signal intensity on T2-weighted images.

cross-sectional imaging characteristics of pancreatic lesions
Cysts The widespread use of high-quality abdominal cross-sectional imaging for a variety of indications has lead to an increased recognition of cystic lesions of the pancreas. Cystic pancreatic lesions may be non-neoplastic, as in the case of pseudocysts, or can be neoplasms that are completely benign, premalignant, or frankly cancerous. Given the broad differential diagnosis of pancreatic cysts, determining their histological origin, while challenging, is of paramount importance in deciding on management, especially for tumors greater than 3 cm in diameter (18,19). Pseudocysts, which are common sequelae of acute pancreatitis, are the most common cystic lesions of the pancreas. As such, differentiating pseudocysts from cystic neoplasms of the pancreas prior to treatment is desirable. Since a clinical history of prior episodes of acute pancreatitis is not perfectly correlated with the diagnosis of pseudocyst, radiological

104

CROSS-SECTIONAL IMAGING FOR HPB DISORDERS (MRI AND CT)

(A)

(B)

Figure 11.4 Hepatocellular carcinoma. (A) Arterial phase CT image demonstrating a dominate right-lobe mass. (B) Note the change in enhancement on the portal venous phase of imaging.

(A)

(B)

Figure 11.5 Gallbladder carcinoma. T2-weighted MRI (A) axial and (B) coronal images demonstrate a solid mass in the gallbladder with hyperintense surrounding liver parenchyma consistent with local extension of the tumor into the liver parenchyma.

differentiation is necessary. The presence on MRI of dependent debris within a cystic pancreatic lesion has been found to be highly suggestive of the diagnosis of pseudocyst (20). Serous cystadenoma (SCA) are the most common benign neoplasm of the pancreas and are typically asymptomatic findings, however a proportion of patients do present with symptoms due to mass effect (21). SCA are comprised of multiple smaller cysts and may have a variable appearance based on the size of the cysts that comprise them. In fact, microcystic tumors may have an appearance on CT more consistent with that of a solid tumor. MRI and ultrasound may be helpful in defining the cystic nature of such tumors. Since asymptomatic SCA do not require treatment, differentiating them from other malignant or premalignant lesions is critical. In the event that imaging characteristics are nondiagnostic, biopsy or resection may be indicated. Intraductal papillary mucinous neoplasms (IPMN) are premalignant tumors arising from the main pancreatic duct or its

Figure 11.6 Colorectal cancer metastasis. Irregular hypodense central lesion on contrast-enhanced CT represents a colorectal liver metastasis.

105

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

Figure 11.7 Neuroendocrine metastases. (A) Innumerable hypodense neuroendocrine metastatic nodules with variable levels of rim-enhancement on portal venous phase contrast-enhanced CT. (B) Coronal slice demonstrates direct extension of tumor into the portal vein (arrow).

(A)

(B)

Figure 11.8 Intraductal papillary mucinous neoplasm (IPMN). (A) T1-weighted postcontrast imaging reveals a low-intensity multilocular lesion in the head and uncinate process of the pancreas suspicious for a side-branch IPMN. (B) T2-weighted images demonstrate high signal intensity.

branches. IPMN contain epithelium ranging from benign adenoma to invasive adenocarcinoma. IPMN are differentiated based on whether they arise from side-branches or from the main pancreatic duct, with the latter having a higher potential for progressing to invasive malignancy. Cross-sectional imaging reveals a cystic region within or adjacent to the pancreatic parenchyma that may demonstrate continuity with the pancreatic ductal system. Factors that influence the decision to perform pancreatectomy for IPMN include size, growth, and the presence of fibrous septations or solid components. Mucinous cystic neoplasms (MCN) are less common lesions, typically seen in women, which are characterized by ovarian-type stroma. MCN are also felt to be premalignant lesions, therefore their resection is recommended. MCN have imaging characteristics similar to those of IPMN, with the absence of a definable connection to the main pancreatic ductal system (Fig. 11.8). Pancreatic Neuroendocrine Tumors Pancreatic neuroendocrine tumors (PNET), also known as islet cell tumors, are rare malignant neoplasms that have a relatively slow growth rate. While most PNET are nonfunctional, they

may secrete hormones that lead to clinical symptoms, especially in the setting of metastatic disease to the pancreas. PNET are typically hypervascular lesions that show early arterial phase enhancement on CT, but may be isodense on portal venous phase. Small functional tumors may prove difficult to identify on cross-sectional imaging studies, therefore accurate timing of imaging to the arterial phase of enhancement is important. Similarly, MRI imaging of PNET usually demonstrate high signal intensity on T2-weighted images and low intensity on T1-weighted images with significant contrast enhancement (22) (Fig. 11.9). Solid Pseudopapillary Tumor Solid pseudopapillary tumor (SPT) of the pancreas is a rare, indolent tumor that most commonly affects women in the first three decades of life. Although it has metastatic potential, malignant behavior is uncommon, therefore resection is considered curative. CT imaging is varied and may demonstrate a large lesion with internal hemorrhage or cystic degeneration. While vascular encasement, pancreatic ductal dilatation, and hepatic metastases are seen only in the carcinomatous variant

106

CROSS-SECTIONAL IMAGING FOR HPB DISORDERS (MRI AND CT)

Figure 11.9 Neuroendocrine pancreatic tumor. A well-preserved fat plane is seen separating this large hypervascular neuroendocrine tumor in the head of the pancreas from the superior mesenteric/portal vein. Areas of hypoattenuation towards the middle of the tumor are suggestive of central ischemia.

of this tumor, more typical findings based on size, capsule thickness, internal composition, and calcification pattern do not help to differentiate benign and malignant lesions (23). Acinar Cell Carcinoma Although acinar cells comprise the bulk of pancreatic tissue, acinar cell carcinomas (ACC) are very uncommon. ACC are typically well-circumscribed tumors that may be lobulated and may be heterogeneous or homogeneous in appearance on cross-sectional imaging (24). ACC range from being completely solid to mostly cystic with at least some solid components and central necrotic areas may be seen. Biliary or pancreatic ductal dilatation is occasionally seen. Contrast-enhanced CT shows homogeneous enhancement of solid components, but less than that of the surrounding pancreas. T2-weighted MRI images may show hyperintense signal in relation to pancreatic parenchyma. Pancreatic Adenocarcinoma Pancreatic adenocarcinoma, commonly referred to as pancreatic cancer, is the most common malignant neoplasm of the pancreas. It is a highly aggressive malignancy that carries with it an extremely high mortality rate, having an incidence that nearly equals its mortality rate. Because of the biologically aggressive nature of pancreatic cancer, the majority of patients present with metastatic or unresectable disease. Contrast-enhanced CT usually shows a hypoattenuating poorly defined mass with dilatation of the pancreatic duct distally and, in the case of tumors of the pancreatic head, biliary ductal dilatation as well. Biliary and/or pancreatic ductal dilatation may also be seen in the absence of an identifiable mass, and dilatation of both of these ductal systems—the double duct sign—is a classic sign of adenocarcinoma of the head of the pancreas. The double duct sign is not, however, pathognomonic for pancreatic cancer and may be associated with benign processes (25). MRI has the advantage of being usable in patients with diminished renal function and can be combined with MRCP to provide detail about the biliary and pancreatic ductal systems in a noninvasive fashion (Fig. 11.10). The primary determinants of the resectability of pancreatic lesions inevitably relate to vascular involvement, since the

Figure 11.10 Pancreatic adenocarcinoma. Contrast-enhanced CT reveals a hypointense solid-appearing mass in the tail of the pancreas. This appearance on cross-sectional imaging is characteristic of pancreatic adenocarcinoma.

pancreas is intimately associated with the portal venous system and in close proximity to the celiac artery and superior mesenteric artery (SMA). High-quality cross-sectional imaging clearly defines these relationships and predicts the likelihood of successful resection. Metastatic Cancer to the Pancreas In contrast to the liver, the pancreas is a rare site of metastatic disease. Although multiple types of cancer have been reported to metastasize to the pancreas, renal cell carcinoma, nonsmall cell lung carcinoma, and lymphoma are the most common sources of isolated pancreatic metastases. Renal cell carcinoma metastases to the pancreas are typically well-circumscribed, arterial-enhancing lesions, while other histologies tend to be more diffuse and variable in enhancement patterns.

conclusions
CT and MRI have become essential components in the diagnosis, perioperative management, and follow-up of hepatic, biliary, and pancreatic pathology. Therefore, the ability to appropriately order and interpret these studies, in consultation with radiologists, is a prerequisite to the surgical management of patients with such diseases.

references
1. Winston CB, Lee NA, Jarnagin WR, et al. CT angiography for delineation of celiac and superior mesenteric artery variants in patients undergoing hepatobiliary and pancreatic surgery. AJR Am J Roentgenol 2007; 189(1): W13–19. 2. Blumgart LH. Surgery of the Liver, Biliary Tract, and Pancreas, 4th edn. Philadelphia, PA: Saunders Elsevier, 2007. 3. Kamel IR, Lawler LP, Fishman EK. Variations in anatomy of the middle hepatic vein and their impact on formal right hepatectomy. Abdom Imaging 2003; 28(5): 668–74. 4. Covey AM, Brody LA, Getrajdman GI, Sofocleous CT, Brown KT. Incidence, patterns, and clinical relevance of variant portal vein anatomy. AJR Am J Roentgenol Oct 2004; 183(4): 1055–64.

107

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
5. Adam A, Dixon AK, Allison DJ, Grainger RG. Grainger & Allison’s Diagnostic Radiology: a Textbook of Medical Imaging 5th edn. Edinburgh: Churchill Livingstone, 2008. 6. Mansfield SD, Scott J, Oppong K, et al. Comparison of multislice computed tomography and endoscopic ultrasonography with operative and histological findings in suspected pancreatic and periampullary malignancy. Br J Surg 2008; 95(12): 1512–20. 7. Yoon SS, Charny CK, Fong Y, et al. Diagnosis, management, and outcomes of 115 patients with hepatic hemangioma. J Am Coll Surg 2003; 197(3): 392–402. 8. Blachar A, Federle MP, Ferris JV, et al. Radiologists’ performance in the diagnosis of liver tumors with central scars by using specific CT criteria. Radiology 2002; 223(2): 532–9. 9. Hussain SM, Terkivatan T, Zondervan PE, et al. Focal nodular hyperplasia: findings at state-of-the-art MR imaging, US, CT, and pathologic analysis. Radiographics 2004; 24(1): 3–17, discussion 18–19. 10. Pang RW, Joh JW, Johnson PJ, Monden M, Pawlik TM, Poon RT. Biology of hepatocellular carcinoma. Ann Surg Oncol 2008; 15(4): 962–71. 11. Takayama T, Makuuchi M, Kojiro M, et al. Early hepatocellular carcinoma: pathology, imaging, and therapy. Ann Surg Oncol 2008; 15(4): 972–8. 12. Coakley FV, Schwartz LH. Imaging of hepatocellular carcinoma: a practical approach. Semin Oncol 2001; 28(5): 460–73. 13. Stipa F, Yoon SS, Liau KH, et al. Outcome of patients with fibrolamellar hepatocellular carcinoma. Cancer 15 2006; 106(6): 1331–8. 14. Akoad M, Jenkins R. Proximal biliary malignancy. Surg Clin North Am 2008; 88(6): 1409–28, x–xi. 15. Jarnagin WR, Bowne W, Klimstra DS, et al. Papillary phenotype confers improved survival after resection of hilar cholangiocarcinoma. Ann Surg 2005; 241(5): 703–12, discussion 712–14. 16. Schwartz LH, Black J, Fong Y, et al. Gallbladder carcinoma: findings at MR imaging with MR cholangiopancreatography. J Comput Assist Tomogr 2002; 26(3): 405–10. 17. Tamm EP, Kim EE, Ng CS. Imaging of neuroendocrine tumors. Hematol Oncol Clin North Am 2007; 21(3): 409–32, vii. 18. Allen PJ, Brennan MF. The management of cystic lesions of the pancreas. Adv Surg 2007; 41: 211–28. 19. Allen PJ, D’Angelica M, Gonen M, et al. A selective approach to the resection of cystic lesions of the pancreas: results from 539 consecutive patients. Ann Surg 2006; 244(4): 572–82. 20. Macari M, Finn ME, Bennett GL, et al. Differentiating pancreatic cystic neoplasms from pancreatic pseudocysts at MR imaging: value of perceived internal debris. Radiology 2009; 251(1): 77–84. 21. Tseng JF, Warshaw AL, Sahani DV, et al. Serous cystadenoma of the pancreas: tumor growth rates and recommendations for treatment. Ann Surg 2005; 242(3): 413–19, discussion 419–21. 22. Rha SE, Jung SE, Lee KH, et al. CT and MR imaging findings of endocrine tumor of the pancreas according to WHO classification. Eur J Radiol 2007; 62(3): 371–7. 23. Lee JH, Yu JS, Kim H, et al. Solid pseudopapillary carcinoma of the pancreas: differentiation from benign solid pseudopapillary tumour using CT and MRI. Clin Radiol 2008; 63(9): 1006–14. 24. Tatli S, Mortele KJ, Levy AD, et al. CT and MRI features of pure acinar cell carcinoma of the pancreas in adults. AJR Am J Roentgenol 2005; 184(2): 511–19. 25. Plumley TF, Rohrmann CA, Freeny PC, Silverstein FE, Ball TJ. Double duct sign: reassessed significance in ERCP. AJR Am J Roentgenol 1982; 138(1): 31–5.

108

12

Liver metastases: detection and imaging Valérie Vilgrain, Ludovic Trinquart, and Bernard Van Beers
The role of Doppler techniques is often limited because flow signals in liver metastases are usually too low to be detected except in markedly hypervascular liver metastases. Because there are no specific features of metastases at conventional US, the differentiation of a single metastasis from other lesions is usually not possible but on the other hand, US is helpful to characterize benign lesions such as hepatic cysts and hemangiomas in oncological patients. While in many European countries, US was the recommended imaging follow-up method, CT or MRI is nowadays preferred in oncological patients.

The liver is the second most frequent site of metastases, after lymph nodes, providing a very suitable environment for the growth of metastases because of its rich blood supply from the systemic and splanchnic system. The overall extent of liver involvement in cancer patients is unknown but liver metastases have been found in 30% to 70% of patients who die because of cancer, depending on their primary tumor (1). Liver metastases frequently arise from colorectal cancer (CRC), with 15% to 20% of patients presenting with synchronous liver metastases and another 15% developing metachronous metastases to the liver within five years (2,3). But unlike many other types of cancers, the presence of distant metastases from CRC does not necessarily preclude curative treatment. In fact, CRC metastases are confined to the liver in 25% of patients (4). This confinement of metastatic disease to the liver has allowed progress in the treatment of these patients—via hepatic resection, regional chemotherapy, and thermoablative treatments, and the benefits of such approaches are demonstrated by the fact that survival of up to 25% of patients 10 year after resection of these metastases is possible. Isolated liver metastases also often arise from gastric and pancreatic cancers—because of the portal venous drainage to the liver—and less frequently from breast or lung cancers. But most non-CRC liver metastases are associated with distant metastatic spread to other organs and so require a more systemic therapeutic approach. However, metastases confined to the liver may also be seen in ocular melanoma, breast cancer, neuroendocrine tumors, renal cell cancers, and some sarcomas (5–7). In this context, the goal of imaging for liver metastases is twofold: first to establish an early and accurate diagnosis of liver metastases, second to stage preoperatively those patients with liver metastases confined to the liver, especially when the primary tumor is CRC. The diagnostic value of ultrasound (US), contrast-enhanced US, multidetector computed tomography (CT), and magnetic resonance (MR) imaging with nonspecific gadolinium chelates and liver-specific contrast agent is discussed. Pitfalls and limitations of imaging are shown. Lastly, the role of imaging in assessing number, localization, and size of metastases to determine resectability is emphasized.

contrast-enhanced ultrasound
The principle of this technique is to increase the lesion-to-liver contrast, using intravascular microbubble contrast agents, which allows enhanced detection of smaller liver metastases not seen on conventional US. Most contrast agents used nowadays provide strong and persistent signal enhancement due to harmonic resonance at low mechanical index, where minimal or no bubble destruction occurs. Examination includes a continuous evaluation of the lesion enhancement during the arterial (15–30 seconds delay), portal venous (30–60 seconds delay), and delayed (2–3 minutes delay) phases. Most liver metastases are hypovascular and exhibit no or minimal enhancement on the arterial phase. Interestingly, whatever the lesion enhancement is on the arterial phase, metastases show nonenhancing defects on the delayed phase, which seem to be the most useful determinant for both lesion detection and characterization (Fig. 12.1). This strong washout sign is caused by the biokinetics of the US contrast agents that are purely vascular effects; conversely to nonspecific CT and MR contrast agents that spread into the interstitium. Indeed, the use of contrast agents improves the sensitivity of US in detecting individual lesions by about 20% in comparison to baseline, independent of the type of contrast agent used (9). Contrast-enhanced US imaging is technically successful in most patients except those with severe obesity and marked steatosis in whom penetration of contrast-specific imaging is limited.

imaging techniques
Ultrasound Liver metastases are generally multiple, spherical, and have well-defined margins. Most lesions are hypoechoic. The most common hyperechoic metastases are observed in patients with CRC or neuroendocrine tumors. Large lesions often have more central hypoechogenicity related to areas of necrosis. A hypoechoic halo is seen surrounding the lesions in 40% of cases (8). More rarely, liver metastases may appear as diffuse infiltration.

computed tomography
Computed tomography (CT) is the most commonly used imaging modality for both detection and characterization of liver metastases. Multidetector helical CT is now the standard technique. It reduces the scan time, with high coverage and high-quality 3D reconstructions. The examination comprises of an unenhanced scan and, after intravenous administration of nonionic iodine contrast medium, two acquisitions at the late arterial phase and the portal venous phase. During the latter, the liver parenchyma enhances

109

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C) Figure 12.1 (A–C) Portal-phase CT shows a small liver tumor that is not characteristic of liver metastasis. This lesion is homogeneous and hyperechoic on ultrasound (B). Portal-phase contrast-enhanced ultrasound demonstrates washout, which is highly suggestive of liver malignancy (C).

the arterial phase. A key finding is the presence of a halo (11), which has been shown as a quite sensitive and specific finding for liver metastases. However, unenhanced, arterial-phase and delayed-phase imaging (which is optional) are helpful for differentiating benign lesions such as cyst, hemangioma, or hepatocellular tumors from metastases.

magnetic resonance imaging
The routine magnetic resonance (MR) imaging protocol includes nonenhanced T1- and T2-weighted pulse sequences and postcontrast sequences. On T1-weighted images, most liver metastases are hypointense, but isointense lesions are seen in approximately 10% of patients (12). Hyperintense liver metastases are very uncommon but may be seen in melanoma. On T2-weighted images, liver metastases are most commonly moderately hyperintense, whereas the remnants are isointense or markedly hyperintense (12). The areas of marked hyperintensity correspond to cystic changes or necrosis (12). It seems that the signal intensity on T2-weighted images is not related to metastases from a specific primary neoplasm. Contrast agents can improve diagnostic accuracy. Two groups of MR contrast agents may be used: nonspecific gadolinium chelates whose biokinetics are similar to iodine contrast agents, and liver-specific contrast agents, either for

Figure 12.2 Portal-phase CT demonstrating multiple heterogeneous lesions suggestive of liver metastases.

and it increases the lesion conspicuity of hypovascular tumors (10) (Fig. 12.2). Most liver metastases are hypoattenuating and hypovascular on unenhanced scans, meaning that they are better seen on the portal venous phase than on

110

LIVER METASTASES: DETECTION AND IMAGING

(A)

(B)

(C)

(D)

Figure 12.3 (A–D) MR imaging of a colorectal metastasis. The tumor is hyperintense on T2- and hypointense on T1-weighted imaging (A and B). Note the peripheral halo on portal-phase T1-weighted imaging and the delayed enhancement due to fibrous stroma (C and D).

reticuloendothelial system (ferumoxides) or hepatobiliary captation (Mn-DPDP or specific hepatobiliary gadolinium chelates). Briefly, the nonspecific gadolinium chelates are used for lesion characterization, while the others have been proposed for preoperative staging. The principle of the latter is to increase the lesion-to-liver contrast by decreasing markedly the signal of the liver on T2 sequences (ferumoxides) or increasing it on T1-weighted sequences (13). Some authors have also proposed double-contrast MR combining specific and nonspecific contrast agents. Similarly to CT, nonspecific gadolinium MR imaging should include baseline precontrast images and sequential acquisitions at arterial, portal, and equilibrium phases. In a large series of 516 liver metastases from various tumors in 165 consecutive patients, most liver metastases were hypovascular (64% of all patients and 91% of patients with colon cancer) ( 12 ). A hypervascular pattern of enhancement was identified in 36% of patients. During the arterial phase, peripheral ring enhancement was seen in 72% of patients. On the portal venous and delayed phase, incomplete central progression of lesion enhancement was found in two-thirds of patients ( 12 ) ( Fig. 12.3 ). Peripheral washout in metastases on delayed-phase images was identified in one-third

of patients with hypervascular metastases and almost never in hypovascular metastases ( 12 ) ( Fig. 12.4 ). After administration of liver-specific contrast agents, liver metastases that lack functioning hepatocytes or Kupffer cells do not enhance postcontrast, resulting in improved lesion conspicuity ( 14 ). Diffusion-weighted MR imaging is quite interesting in liver metastases. Nasu et al. (15) have shown increased detection of metastatic lesions with a combination of diffusion-weighted imaging and precontrast T1- and T2-weighted imaging when compared with liver-specific contrast MR imaging. Parikh et al. have shown that diffusion-weighted sequences were as accurate as T2-sequences for characterization of focal liver lesions including metastases (16) (Fig. 12.5).

positron emission tomography
Conversely to the other imaging modalities, which give more morphologic than functional information, positron emission tomography (PET) imaging is essentially functional imaging and provides a physiological survey for hypermetabolic tumors. PET scanning after administration of [18F] 2-fluoro-2-deoxyglucose (FDG) is based upon higher glycolytic activity of many tumors compared to normal tissue. [18F]FDG is transported into cells and phosphorylated by

111

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
the enzyme hexokinase to [18F]FDG-6-phosphate, which cannot proceed down the glycolytic pathway and is therefore accumulated in malignant tissue. This technique has improved markedly over the past decade, and many centers routinely incorporate PET imaging results in the staging of patients with liver metastases, especially when consideration is being given for liver resection. Most studies have focused on the diagnostic yield of fluorodeoxyglucose (FDG)-PET in patients with liver metastases from colon and rectal cancer. The two main limitations of PET are the lack of anatomical landmarks and poor spatial resolution. Development of PET/CT has overcome these drawbacks; unfortunately, most PET/CT examinations are performed with unenhanced CT images. Recently, some authors have investigated the role of IV iodinated contrast material in the evaluation of liver metastases at [18F]FDG PET/CT (17). They have shown that more liver metastases were detected on PET/contrast-enhanced CT compared with PET/unenhanced CT (83% and 67%, respectively). Similarly, liver metastases were more accurately characterized at PET/ contrast-enhanced CT compared with PET/unenhanced CT (73% and 57%, respectively).

(A)

(B)

(C)

(D)

Figure 12.4 (A–D) MR imaging of endocrine metastases. The tumors are strongly hyperintense on T2- and hypointense on T1-weighted imaging (A and B). Note the strong hypervascularity on arterial-phase T1-weighted imaging and the washout on portal-phase imaging (C and D).

(A)

(B)

Figure 12.5 MR imaging of liver metastases. Multiple tumors are seen on T2-weighted imaging (A). Conspicuity of small tumors is more evident on diffusionweighted imaging (B).

112

LIVER METASTASES: DETECTION AND IMAGING

perfusion imaging
Liver metastases induce changes in liver perfusion and have been shown to increase arterial blood flow and the arterial– portal flow ratio (hepatic perfusion index) compared to normal liver. Interestingly, animal studies have demonstrated that those changes may be detected at an early stage when liver metastases are occult on other imaging modalities (18). Hepatic perfusion index can be obtained using various techniques: nuclear medicine, US, CT, or MR imaging. Early results in patients were promising but lack of standardization in utilization, lack of consensus regarding the imaging modality, and the presence of multiple mathematical models have meant that perfusion has not been adopted in routine practice.

calcification during response to chemotherapy (19) (Fig. 12.6). Whether liver metastatic calcification carries a prognostic significance in CRC is still questionable. Intrabiliary Metastases Intrinsic bile duct involvement by metastases may occur either by growing from within or invading the lumen of the bile ducts. The most common intrabiliary metastases arise from colorectal carcinoma. Most patients present with various degree of biliary obstruction including jaundice (20). The presence of macroscopic intrabiliary extension seems to be a good indicator in patients with CRC showing less aggressive features (21). Pitfalls and Limitations The two most difficult situations in oncological patients are the changes in the liver such as steatosis that can create pseudolesions or hide true metastases, and the characterization of small lesions. Steatosis is not always homogeneous and may have either focal fatty sparing, or more rarely focal fatty deposit. Furthermore, fatty livers are often observed in cancer patients who have received chemotherapy. Clearly, in this context, MR is superior to CT by combining fat-suppressed T1 sequences and in- and opposed-phase T1-weighted imaging that can diagnose both focal and diffuse changes, and help to differentiate focal fat from metastases (Fig. 12.7). Another issue is the characterization of small liver lesions in cancer patients. Those lesions (smaller than 1 or 2 cm) are often deemed “too small to characterize,” and due to the high prevalence of benign lesions in the liver, are more frequently benign than malignant. Schwartz et al. reported in a series of 2978 cancer patients that metastases represented only 11.6% of patients with small liver lesions (22). Other authors have shown that the positive predictive value for malignancy increased notably using a cut-off of 20 mm compared to 10 mm (23). Yet, for an individual patient, we cannot rely only on lesion size for characterization. While CT is an excellent imaging modality for detection, it is not as good as US, contrast-enhanced US, and MR imaging when characterizing small liver lesions (24,25).

other imaging findings
Hypervascular Metastases Liver metastases from colon carcinoma, other gastrointestinal carcinoma, and pulmonary carcinoma are considered hypovascular, and those from thyroid carcinoma, neuroendocrine tumor, and renal cell carcinoma are hypervascular (12) (Fig. 12.5). Breast carcinoma metastases can be either hypovascular or hypervascular. Furthermore, in Danet’s series, patients with metastases that exhibit the nonclassical type of enhancement according to the primary tumor were not uncommon. For example, 9% of patients with colon carcinoma had hypervascular metastases (12). This observation emphasizes the role of imaging in characterizing liver lesions in cancer patients, and the importance of multiple complementary examinations. Cystic Metastases Liver metastases may have a cystic appearance with strong hyperintensity on heavily T2-weighted MR images. However, these tumors usually have other findings that are not seen in typical hepatic cysts such as internal septations, thick walls, and wall enhancement. Most cystic liver metastases arise from cystadenocarcinoma, neuroendocrine tumor, and sarcoma. Calcified Metastases Up to 11% of patients with colon carcinoma have calcified liver metastases at presentation. Patients may also develop

(A)

(B)

Figure 12.6 (A and B) Liver metastases before and after chemotherapy. Note the significant decrease in size of the tumors. Furthermore, the tumors present diffuse calcifications after chemotherapy.

113

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C)

Figure 12.7 (A–C) Liver metastases developed on a fatty liver. The tumor located in the posterior part of the segment 4 is barely visible on portal-phase CT (A) and is better seen on pre- and postcontrast T1-weighted imaging (B and C).

detection of liver metastases: which imaging modality?
Many different noninvasive imaging modalities are available for the preoperative detection of liver metastases: US and contrast-enhanced US, multidetector CT, MR imaging, and PET using FDG. Comparison of these imaging techniques is challenging and results have evolved over time due to technological improvements. Multidetector CT has notably increased the performance of CT in decreasing slice thickness and optimizing lesion enhancement on multiphasic studies after intravenous contrast. The use of liver-specific contrast agents in MR imaging has given new horizons for this imaging modality. Diffusion-weighted MR has markedly improved the detection of liver metastases. The use of US contrast agents has completely changed the role of US in oncology. Integrated PET/CT scanners combining metabolic and anatomical information has also resulted in an increased interest in PET studies, and it is likely that the role of IV iodinated contrast material in PET/CT scanners will be quite significant. Consequently, many studies have assessed and compared the diagnostic value of these imaging techniques, resulting in an extensive body of literature and the absence of any consensus on the diagnostic algorithm (26). Two systematic reviews (27,28) and one narrative review (29) have analyzed the available evidence (Tables 12.1 and 12.2). The first systematic review was published by Kinkel et al. in 2002, and aimed at comparing current noninvasive imaging methods such as US, CT, MR imaging, and FDG PET for the detection of hepatic metastases from colorectal, gastric, and esophageal cancers (28). Papers published between December 1985 and December 2000 were studied. Among the 54 studies included, 9 assessed US (686 patients, 74% with CRC), 25 assessed CT (1747 patients, 78% with CRC), 11 addressed MRI (401 patients, 100% with CRC), and 9 reported on PET (423 patients, 100% with CRC). In a “per-patient” meta-analysis, the authors concluded that [18F]FDG PET was the most sensitive examination (Level of evidence II): the combined per-patient sensitivity of [18F]FDG PET (0.90, 95% CI 0.82– 0.96) being significantly superior to that of US (0.66, 95% CI 0.54–0.77), CT (0.70, 95% CI 0.63–0.77), and MRI (0.71, 95%

Table 12.1 Levels of Evidence for Studies of Diagnostic Test Accuracy
Ia: Systematic review (with homogeneity)* of level-1 studies Ib: Level 1 studies II: Level 2 studies or systematic reviews of level-2 studies III: Level 3 studies or systematic reviews of level-3 studies IV: Expert committee reports or opinions Level 1 studies are studies that use a blind comparison of the test withw a validated reference standard (gold standard) In a sample of patients that reflects the population to whom the test would apply Level 2 studies are studies that have only one of the following: Narrow population (the sample does not reflect the population to whom the test would apply) Use a poor reference standard (defined as that where the “test” is included in the “reference” or where the “testing” affects the “reference”) The comparison between the test and reference is not blind Case–control studies Level 3 studies are studies that have at least two or three of the features listed above
*Homogeneity means there are no or minor variations in the directions and degrees of results between individual studies that are included in the systematic review.

Table 12.2 Grading of Recommendations on Diagnostic Tests
Grade A: Studies with level of evidence Ia or Ib Grade B: Studies with level of evidence II Grade C: Studies with level of evidence III Grade D: Studies with level of evidence IV

CI 0.61–0.80). Moreover, in the 35 studies with specificity higher than 85%, [18F]FDG PET was still the most sensitive technique, the combined sensitivity being 55% for US, 72% for CT, 76% for MR imaging, and 90% for FDG PET. The second systematic review was published by Bipat et al. in 2005 and aimed at evaluating CT, MR imaging, and [18F]

114

LIVER METASTASES: DETECTION AND IMAGING
FDG PET for the detection of colorectal liver metastases on a per-patient and per-lesion bases, reviewing articles from 1990 to 2003 (27). Among the 61 selected studies, 28 assessed nonhelical CT, 15 assessed helical CT, 5 concerned 1.0T MRI, 12 concerned 1.5T MRI, and 21 addressed [18F]FDG PET. On a “per-patient” basis, the combined sensitivity of PET (0.95, 95% CI 0.93–0.96) was significantly superior to that of nonhelical CT (0.60, 95% CI 0.58–0.65), helical CT (0.65, 95% CI 0.30–0.89), and 1.5T MRI (0.76, 95% CI 0.56–0.89) (Level of evidence II). On a “per–lesion” basis, nonhelical CT sensitivity (0.52, 95% CI 0.52–0.53) was significantly lower than that of helical CT (0.64, 95% CI 0.54–0.72), 1.0T MRI (0.66, 95% CI 0.66–0.66), 1.5T MRI (0.64, 95% CI 0.58–0.71), and PET (0.76, 95% CI 0.61–0.86). In other words, there was no evidence that the “per-lesion” sensitivities of PET, helical CT, and MRI differed significantly. For lesions of 1 cm or larger, SPIO-enhanced MR imaging was the most accurate modality. Therefore, considerable debate continues about which imaging modality offers the best noninvasive examination of the liver, and so some comments concerning the existing evidence need to be addressed. First, the diagnostic value of imaging techniques can be computed per patient (detection of at least one lesion per patient) or per lesion (detection of all lesions per patient). But, in cancer patients, the per-patient analysis is not adequate because the main question is not: “Does the patient have liver metastases?” But rather, “How many metastases are in the liver, and where?” As previously seen in the results of the prior metaanalyses, the per-lesion comparison of imaging modalities is still open. The second point worth considering is the frequent use of a suboptimal diagnostic reference standard. The most reliable reference standard is the combination of direct visualization and bimanual palpation of the liver, intraoperative US, and histopathological examination of resected liver tissue of each lesion found in the liver, so allowing for a lesion-by-lesion analysis. But only a minority of studies used this reference standard, resulting in underdetection of lesions and overestimation of sensitivity. Studies that analyzed the detection of liver metastases without this surgical reference standard (that is using imaging follow-up or a combination of other imaging modalities) are of limited value because sensitivity of the methodology will appear higher than the true one. It should be noted that even with this extensive pre- and intraoperative workup, lesions may be missed and in two series approximately 15% of patients were found to have “new” tumors on follow-up CT scans performed four to six months after hepatic resection (30,31). Third, the imaging modalities analyzed in the two metaanalyses extended for a long period of time from 1985 or 1990 to 2000 and their results are difficult to compare with up-to-date imaging. Research on hepatic contrast agents has advanced in two directions: first the development of US contrast agent, especially the more stable second generation of contrast media, has prompted a revival of interest in contrast-enhanced US; second tissue-specific MR contrast agents have been developed to target the main cell populations of the liver. These advances, as well as those of multidetector-row CT (32 or 64 slice systems) or PET/CT, were not integrated in the two available systematic reviews but were discussed in the Rappeport and Loft narrative review (29). Considering the available comparisons of modern MR imaging, multidetector-row CT and PET in the same group of patients with surgical reference standards, Rappeport and Loft questioned the conclusions of the two prior systematic reviews. The authors concluded that “state-of the-art anatomical imaging, e.g., liver-specific MR imaging and multidetector CT, must be considered more sensitive than PET in the detection of individual liver metastases” (Level of evidence II) (29). They also stated that “a preoperative PET/CT-study for detection of possible extra-hepatic tumor contraindicating liver surgery is also recommended.” Moreover, recent articles have pointed out the limitation of FDG-PET in detecting small liver metastases, with a significant superiority of CT and MR imaging (32–34). This is a key result because most lesions larger than 1 cm are depicted on all imaging techniques, but the detection of subcentimeter metastases remains disappointing and therefore the comparison of imaging modalities should focus on these small lesions. Consequently, the question seems to be which is the better imaging modality between CT and MR? And should we use nonspecific MR contrast or liver-specific contrast agents? We have to take into consideration the following:




Multidetector-row CT scanning is often the first choice for a “screening” liver examination at many institutions. This technique also enables rapid scanning of the chest and abdomen and allows evaluation of extrahepatic disease. MRI enhanced with SPIO is probably a more sensitive method than multidetector-row CT for detecting liver metastases (35,36), but to our knowledge, no study has evaluated the added value of MR imaging after multidetector-row CT examination and the consequences in treatment planning.

Based on the existing evidence, it is difficult to provide highstrength management recommendations. Our policy at our institution is to perform systematically a multidetector CT in patients with liver metastases. PET/CT is indicated for the detection of extrahepatic tumor before liver surgery. MR imaging is not routinely performed and is reserved for characterization of small liver lesions, in fatty livers, and in difficult cases after multidetector CT. For lesion characterization, we use nonspecific contrast MR agents, while for tumor detection and preoperative staging, we use liver-specific contrast agents. In both cases, our protocol includes diffusion-weighted MR sequences. Intraoperative US is routinely performed in our institution and intraoperative contrast-enhanced US is under investigation.

preoperative staging
Due to their biological properties, most liver metastases that are resected are secondary to CRC. Selected isolated liver metastases from breast cancer, sarcoma, renal cell cancer

115

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
and Wilms’ tumors, melanoma, other GI cancers, and, most frequently, endocrine tumors, may benefit from liver resection (5). Assessment of the number and location of liver metastases is a major issue of preoperative staging and has been described previously. Yet, imaging has to answer other questions relevant to technical success and R0 resection:








Does the future liver remnant have sufficient volume for the perioperative and postoperative course? This question is easily addressed by CT or MR volumetric measurement. Does the future liver remnant have satisfactory vascular inflow, venous outflow, and biliary drainage? Again CT or MR imaging is adequate to answer this question. Is there any contraindication to performing an R0 resection? This point is more difficult as classical contraindications such as bilobar metastases are overcome in most specialized centers either by twostep resection with preoperative portal embolization to increase the remaining normal liver parenchyma or downsizing with chemotherapy, or combination of resection and thermal ablation (37). However, it is crucial to evaluate the relationship between the liver tumors and important anatomical landmarks such as the inferior vena cava (IVC), hepatic venous confluence, and the main portal pedicles. Did the liver metastases respond to preoperative chemotherapy? Preoperative chemotherapy is currently performed in resectable patients as it has been shown to reduce the risk of events of progression-free survival in eligible and resected patients (38). In these patients, the role of imaging is to evaluate the tumor response according to RECIST (Response Evaluation Criteria in Solid Tumors) criteria. Association with targeted therapy may render more complex this assessment.

references
1. Pickren J, Tsukada Y, Lane W. Analysis of autopsy data. In: Weiss L, Gilbert H, eds. Liver Metastasis. Boston, MA: Hall, 1982: 2–18. 2. Kune GA, Kune S, Field B, et al. Survival in patients with large-bowel cancer. A population-based investigation from the Melbourne Colorectal Cancer Study. Dis Colon Rectum 1990; 33: 938–46. 3. Manfredi S, Lepage C, Hatem C, et al. Epidemiology and management of liver metastases from colorectal cancer. Ann Surg 2006; 244: 254–9. 4. Ballantyne GH, Quin J. Surgical treatment of liver metastases in patients with colorectal cancer. Cancer 1993; 71: 4252–66. 5. Elias D, Lasser P, Ducreux M, et al. Liver resection (and associated extrahepatic resections) for metastatic well-differentiated endocrine tumors: a 15-year single center prospective study. Surgery 2003; 133: 375–82. 6. Hughes K, Sugarbaker P. Resection of the liver for metastatic solid tumors. In: SA R, ed. Surgical Treatment of Metastatic Cancer. Philadelphia, PA: Lippincott, 1987; 125–64. 7. Leyvraz S, Spataro V, Bauer J, et al. Treatment of ocular melanoma metastatic to the liver by hepatic arterial chemotherapy. J Clin Oncol 1997; 15: 2589–95. 8. Albrecht T. Detection and characterisation of liver metastases. In: Lencioni R, ed. Enhancing the Role of Ultrasound with Contrast Agents. Milan: Springer, 2006; 53–67.

9. Albrecht T, Hoffmann CW, Schmitz SA, et al. Phase-inversion sonography during the liver-specific late phase of contrast enhancement: improved detection of liver metastases. AJR Am J Roentgenol 2001; 176: 1191–8. 10. van Leeuwen MS, Noordzij J, Feldberg MA, Hennipman AH, Doornewaard H. Focal liver lesions: characterization with triphasic spiral CT. Radiology 1996; 201: 327–36. 11. Nino-Murcia M, Olcott EW, Jeffrey RB, Jr., et al. Focal liver lesions: pattern-based classification scheme for enhancement at arterial phase CT. Radiology 2000; 215: 746–51. 12. Danet IM, Semelka RC, Leonardou P, et al. Spectrum of MRI appearances of untreated metastases of the liver. AJR Am J Roentgenol 2003; 181: 809–17. 13. Kim KW, Kim AY, Kim TK, et al. Small (< or = 2 cm) hepatic lesions in colorectal cancer patients: detection and characterization on mangafodipir trisodium-enhanced MRI. AJR Am J Roentgenol 2004; 182: 1233–40. 14. Schima W, Kulinna C, Langenberger H, Ba-Ssalamah A. Liver metastases of colorectal cancer: US, CT or MR? Cancer Imaging 2005; 5 Spec No A: S149–56. 15. Nasu K, Kuroki Y, Nawano S, et al. Hepatic metastases: diffusion-weighted sensitivity-encoding versus SPIO-enhanced MR imaging. Radiology 2006; 239: 122–30. 16. Parikh T, Drew SJ, Lee VS, et al. Focal liver lesion detection and characterization with diffusion-weighted MR imaging: comparison with standard breath-hold T2-weighted imaging. Radiology 2008; 246: 812–22. 17. Badiee S, Franc BL, Webb EM, et al. Role of IV iodinated contrast material in 18F-FDG PET/CT of liver metastases. AJR Am J Roentgenol 2008; 191: 1436–9. 18. Cuenod C, Leconte I, Siauve N, et al. Early changes in liver perfusion caused by occult metastases in rats: detection with quantitative CT. Radiology 2001; 218: 556–61. 19. Hale HL, Husband JE, Gossios K, Norman AR, Cunningham D. CT of calcified liver metastases in colorectal carcinoma. Clin Radiol 1998; 53: 735–41. 20. Povoski SP, Klimstra DS, Brown KT, et al. Recognition of intrabiliary hepatic metastases from colorectal adenocarcinoma. HPB Surg 2000; 11: 383–90, discussion 390–1. 21. Kubo M, Sakamoto M, Fukushima N, et al. Less aggressive features of colorectal cancer with liver metastases showing macroscopic intrabiliary extension. Pathol Int 2002; 52: 514–18. 22. Schwartz LH, Gandras EJ, Colangelo SM, Ercolani MC, Panicek DM. Prevalence and importance of small hepatic lesions found at CT in patients with cancer. Radiology 1999; 210: 71–4. 23. van Erkel AR, Pijl ME, van den Berg-Huysmans AA, et al. Hepatic metastases in patients with colorectal cancer: relationship between size of metastases, standard of reference, and detection rates. Radiology 2002; 224: 404–9. 24. Eberhardt SC, Choi PH, Bach AM, et al. Utility of sonography for small hepatic lesions found on computed tomography in patients with cancer. J Ultrasound Med 2003; 22: 335–43; quiz 345–46. 25. Mueller GC, Hussain HK, Carlos RC, Nghiem HV, Francis IR. Effectiveness of MR imaging in characterizing small hepatic lesions: routine versus expert interpretation. AJR Am J Roentgenol 2003; 180: 673–80. 26. Bipat S, van Leeuwen MS, Ijzermans JN, et al. Imaging and treatment of patients with colorectal liver metastases in the Netherlands: a survey. Neth J Med 2006; 64: 147–51. 27. Bipat S, van Leeuwen MS, Comans EF, et al. Colorectal liver metastases: CT, MR imaging, and PET for diagnosis: meta-analysis. Radiology 2005; 237: 123–31. 28. Kinkel K, Lu Y, Both M, Warren RS, Thoeni RF. Detection of hepatic metastases from cancers of the gastrointestinal tract by using noninvasive imaging methods (US, CT, MR imaging, PET): a meta-analysis. Radiology 2002; 224: 748–56. 29. Rappeport ED, Loft A. Liver metastases from colorectal cancer: imaging with superparamagnetic iron oxide (SPIO)-enhanced MR imaging, computed tomography and positron emission tomography. Abdom Imaging 2007; 32: 624–34.

116

LIVER METASTASES: DETECTION AND IMAGING
30. Ward J, Guthrie JA, Wilson D, et al. Colorectal hepatic metastases: detection with SPIO-enhanced breath-hold MR imaging: comparison of optimized sequences. Radiology 2003; 228: 709–18. 31. Ward J, Robinson PJ, Guthrie JA, et al. Liver metastases in candidates for hepatic resection: comparison of helical CT and gadolinium- and SPIOenhanced MR imaging. Radiology 2005; 237: 170–80. 32. Coenegrachts K, De Geeter F, Ter Beek L, et al. Comparison of MRI (including SS SE-EPI and SPIO-enhanced MRI) and FDG-PET/CT for the detection of colorectal liver metastases. Eur Radiol 2009; 19: 370–9. 33. Kong G, Jackson C, Koh DM, et al. The use of 18F-FDG PET/CT in colorectal liver metastases: comparison with CT and liver MRI. Eur J Nucl Med Mol Imaging 2008; 35: 1323–9. 34. Rappeport ED, Loft A, Berthelsen AK, et al. Contrast-enhanced FDG-PET/CT vs. SPIO-enhanced MRI vs. FDG-PET vs. CT in patients with liver metastases from colorectal cancer: a prospective study with intraoperative confirmation. Acta Radiol 2007; 48: 369–78. Onishi H, Murakami T, Kim T, et al. Hepatic metastases: detection with multi-detector row CT, SPIO-enhanced MR imaging, and both techniques combined. Radiology 2006; 239: 131–8. Ward J. New MR techniques for the detection of liver metastases. Cancer Imaging 2006; 6: 33–42. Gonzalez HD, Figueras J. Practical questions in liver metastases of colorectal cancer: general principles of treatment. HPB 2007; 9: 251–8. Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 2008; 371: 1007–16.

35.

36. 37.

38.

117

13

Surgery for metastatic colorectal cancer René Adam and E. Hoti
Although long-term survival and potential cure after surgical resection has not been demonstrated by randomized controlled trials, the evidence (uniformly poor results observed in untreated patients in contrast to extensive data documenting long-term survival after hepatectomy) supports a significant survival benefit from resection and has provided the rationale for increasing indications for liver surgery as the most effective treatment of CRLM.

introduction
Colorectal cancer (CRC) is a common malignancy with a very high incidence in Western countries. Approximately 150,000 new cases of CRC occur each year in the United States, accounting for more than 55,000 cancer-related deaths (1). Over half the patients diagnosed with CRC will develop liver metastases (CRLM) during the course of their disease (2), of which 15% to 25% will have liver metastases at the time of the diagnosis (3,4). In the absence of surgical treatment, 5-year survival is exceptional (5) and even with the best chemo- and bio-therapies, to date, median survival of unresected disease does not exceed two years (6,7). On the other hand, long-term survival and potential cure after surgical resection for CRLM has been demonstrated by numerous studies. Surgery is therefore considered as the treatment of choice for patients with resectable CRLM, yielding a 5-year survival between 35% and 52% (8,9). As a result, hepatic resection has evolved from a rare procedure associated with considerable mortality to a routine surgery with an operative mortality risk of around 2% (10,11). At present, the low operative mortality along with survival improvement has led to an expansion of more extensive liver surgery and to a clear change in surgical indications to a point where virtually no tumor should be considered unresectable provided that resection can be complete. These advances combined with novel systemic and regional ablative therapies have modified the course of the disease, transforming it from a uniformly fatal to increasingly curable for a majority of patients.

patient evaluation
Patient Selection for Surgery In deciding which patient will tolerate liver resection, a number of factors will need to be considered, including patient comorbidities. Age per se is not an independent factor for increased operative risk (15). This is a very important fact, considering that an increasing proportion of patients being evaluated for surgery for malignant disease, are elderly. On the other hand, scores like the American Society of Anaesthesiology (ASA) (16) or the preoperative Acute Physiology and Chronic Health Evaluation score do significantly influence the incidence of postoperative complications. Patients with an ASA score > 1 have been shown to have more than three times the mortality and twice the morbidity compared to those patients with an ASA of 1 (16). Therefore, a major goal of the preoperative evaluation is to identify patients who are at high operative risk so those who have a prohibitive risk can be excluded from surgery whereas those with manageable comorbidities can have these conditions addressed preoperatively in an attempt to reduce their operative risk. Definition of Tumor Resectability The earlier definition of the resectability (based on factors such as number of lesions, size, distribution, etc.) has been progressively challenged resulting on a concept shift, which now focuses on whether a macroscopic and microscopic complete (R0) resection of the liver lesion as well as complete resection of any extrahepatic diseases can be performed. At present, CRLM are defined as resectable if two aspects are fulfilled: (1) oncological anticipation that the disease can be completely resected without any residual hepatic or extrahepatic disease; (2) maintenance of an adequate volume of the future remnant liver with preserved vascular inflow, outflow, and biliary drainage. In general, at least 25% of the total liver is the minimum safe volume that can be left after liver resection in patients with normal liver parenchyma (17). Preoperative Imaging The complex decision to determine resectability requires detailed anatomic imaging to determine tumor location, exclude unresectable extrahepatic metastases, and assess the adequacy of the liver parenchyma after surgery. There are a

natural history of colorectal liver metastases
The natural history of untreated CRLM has been well studied. The median survival untreated following diagnosis is 6 to 12 months and 5-year survival is extremely rare. Most studies indicate that the prognosis is most closely related to tumor burden. Wood et al. showed that while the 1-year survival was only 5.7% for patients with widespread disease, 60% of the patients with solitary liver metastases were alive at 1 year with a mean survival of 25 months (12). Wagner et al. (5) reported the 3- and 5-year survival for untreated resectable disease to be 14% and 2%, compared to 4% and 0% for unresectable disease. Wilson and Adson (13) in their case-controlled study (60 patients treated with hepatic resection versus 60 comparable patients not subjected to surgery) demonstrated that hepatic resection was associated with 5-, 10- year survival of 25% and 19% compared to no 5-year survival in the non resectable group. Two other case-control studies demonstrated almost identical results (5,14). Scheele et al. (14) reported a 5-year survival rate 40% in patients who underwent tumor resections compared to 0% in 62 patients who had potentially resectable tumors but did not undergo resection.

118

SURGERY FOR METASTATIC COLORECTAL CANCER
myriad of diagnostic capabilities available to date, including three-dimensional CT scanning, CT angiography, magnetic resonance angiography (MRI), and CT volumetry. Nevertheless, despite the evolution of imaging modalities, difficulties still exist, especially when trying to differentiate between metastases and benign liver lesions or to detect small metastatic lesions. The current approach to address these pitfalls is to use a multimodality strategy (18). For example, although helical CT scanning provides information for the entire chest and abdomen during a single breath hold, up to 25% of the lesion can still be missed (19). MRI, on the other hand, is currently the most effective imaging modality in detecting and characterizing liver lesions and is often ordered prior to liver resection to characterize indeterminate lesions seen on a CT scan as it has a higher sensitivity to detect and characterize small lesions (20). Using liver-specific contrast agents, MRI has equivalent sensitivity to CT angiography (21) (Level of evidence: 1). Positron emission tomography (PET) is another useful modality for detecting liver metastases, especially when combined with CT scann. However, it is no more sensitive than MRI in detection, and it lacks the special resolution and the ability to characterize lesions. Truant et al. (22) correlated PET and CT findings in 53 patients with final pathologic diagnoses. They found that PET detected significantly more extrahepatic, intraperitoneal metastases than CT, with a sensitivity of 63% versus 25%. Another meta-analysis study, comparing helical CT, MRI, and fluorodeoxy-glucose PET (FDG-PET) in the detection of colorectal liver metastases, showed that the sensitivities on a per-patient basis were 64.7%, 78.8%, and 94.6%, respectively (23). In contrast, there are other reports that have questioned the superiority of the FDG-PET and consider MRI and helical CT more sensitive in detecting small liver metastases (24,25). PET and FDG-PET are, however, more advantageous in identifying extrahepatic and possible unresectable metastases, which could be a contraindication to liver resection (26) (Level of evidence 1). In addition, the ability of the later investigation to detect occult disease prevents unnecessary surgery in 21.5% of patients and changes the overall management in 25% (26). Hence, despite their pitfalls, the use of image overlays, combining FDG-PET and helical CT or MRI, can increase the accuracy of preoperative staging before hepatic resection (27) (Level of evidence: 1). surgery). When resecting ≥4 liver metastases, the limiting factor is not the number of metastases but whether it is possible to remove all of them (29–32). Similarly, the distribution (bilobar metastatic disease) is no longer considered as a prohibitive prognostic factor. Ercolani et al. (33) reported that the total tumor volume of liver metastases had a stronger influence on survival than did number and location. Also, data from LiverMetSurvey concerning patients with bilobar metastatic disease have demonstrated a 1-, 5-, and 10-year survival of 90%, 44%, and 22%, respectively (34). In general, if a complete resection of the metastases can be achieved with safe margins (R0 resection) while maintaining a sufficient volume of the residual liver, the number and location should not be considered as a contraindication to resection. Whereas achieving a negative resection margin is well established, the extent of this margin clearance remains controversial. Increasingly, studies are demonstrating that there is no significant difference in survival or recurrence related to the width of margin achieved. Elias et al. (35) demonstrated that the overall survival of patients with resection margins less than 1 cm was 27.8%, comparable to those with resection margins of ≥1 cm. Fong et al. (36), in his series of 426 patients undergoing hepatectomy for CRLM, reported an identical 5-year survival in the group with a clear margin of <1 cm compared to the group with a margin of 1 cm or greater. Similarly, Figueras et al. (37) reported that subcentimeter nonpositive surgical margin did not influence hepatic recurrence rates after hepatectomy for CLM. Kokudo et al. (38), in his study, went further on by demonstrating that a margin of 2 mm is clinically the minimum acceptable requirement, which carries approximately a 6% risk of margin-related recurrence. A recently published study from our center showed that despite a higher recurrence rate in patients with R1 resection (complete macroscopic resection with 0 mm free margin) compared to patients with R0 resection, the two groups had a similar overall and disease-free survival (61% vs. 57% and 28% vs. 17%) and recurrences were intrahepatic rather than being localized at the surgical margin (39).



prognostic factors and clinical risk scores
Prognostic Factors The importance of prognostic factors lies in two aspects: evaluation of the prognosis and selection of candidates for surgery. With the evolution, some accepted negative prognostic factors are no more considered, while new ones appear as substitutes. Factors that have been consistently considered as absolute or relative contraindications to liver resection are number/location of liver metastases, resection margin, presence of extrahepatic disease, and tumor involvement of portal lymph nodes.


Therefore, the absence of “safe” margins of resection should not be considered as an absolute contraindication to surgery provided that all tumors can be macroscopically resected. However, at the present surgeons should continue to plan hepatic resection with a preserved “safety zone” and avoid routine use of “minimum margin” surgery.


The number of metastases (≥4) is no longer considered a contraindication (28) to liver resection (based on the fact that long-term survival can be obtained for patients with four or more metastases treated with

The presence of extrahepatic disease reduces the hope of long-term survival and it has been considered as a contraindication to liver resection. Lately, however, resection in patients with extrahepatic disease with curative intent has been advocated by some groups. In a French series (40) of 84 patients who underwent complete resection of extrahepatic disease concurrently with hepatic resection, the

119

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
overall 5-year survival was 28% compared to 34% in the 224 patients undergoing liver resection alone. In addition, the study demonstrated that the total number of metastases has a stronger negative prognostic value after complete resection than their location. Other reports have suggested that patients with extrahepatic metastases may survive more than 5 years after a successful liver resection (41) (Level of evidence: 3). Encouraging results have been reported even after combined resection of concomitant liver metastases and peritoneal carcinomatosis, which has been traditionally considered as an absolute contraindication to liver resection (42,43). However, these results are observed in patients with a limited number of liver metastases (≤3 lesions). Involvement of the portal lymph nodes may be present in as many as 14% of patients with CRLM (44). Some authors have suggested that radical excision of involved portal nodes can produce a survival benefit (45). In a prospective study conducted by Jaeck et al. (46), the survival rate in patients with involved portal lymph nodes was significantly lower than in the control group (3-year survival 19% vs. 62%). However, patients with involved lymph nodes limited to the hepatoduodenal ligament and retropancreatic portion demonstrated a much better prognosis than those with involved lymph nodes around the common hepatic artery and celiac axis (3-year survival 38% vs. 0%). In a more recent study (47) conducted in patients responding to preoperative chemotherapy, we reported that combining liver resection and pedicular lymphadenopathy was justified in patients with involved pedicular lymph nodes (3- and 5-year survival 38% and 18%, respectively). Conversely, in patients with involved celiac or retroperitoneal lymph nodes, this approach was not justified (5-year survival 0%). In this group of patients, even the response to chemotherapy did not seem to change their usual poor prognosis. is the impossibility to remove all metastatic disease, while leaving sufficient functional hepatic parenchyma, regardless of the location, distribution, number and size of the metastases. Clinical Risk Scores Over 10 years ago, Nordlinger (48) introduced the first scoring system for patients with CRLM, based on a multicenter data from 1568 patients who accepted potentially curative resections. In this large series, they identified three groups of patients with low, intermediate and high risk for poor prognosis based on seven high risk factors (see Table 13.1A). Since then, at least six more scoring systems have been developed among which the proposal from Fong et al. (49) based on a single institution series of 1001 patients attracted the most attention. Seven parameters were found to be independent predictors of prognosis. These include presence of extrahepatic disease; positive resection margin; nodal metastases from primary cancer; short disease free interval; largest tumor greater than 5 cm; more than 1 liver metastases; CEA greater than 200 ng/ml (see Table 13.1A). The data for the first two parameters are not available preoperatively. However, using the last five criteria, a preoperative clinical risk score system was created with each positive criterion counting as 1 point. The total score out of 5 is highly predictive of a poor outcome (5-year survival 14%). Patients with a score of 0, 1, and 2 have a highly favorable outcome (5-year survival 60%, 44%, and 40%, respectively). Table 13.1B demonstrates the survival rates for each score grade.



Table 13.1A Prognostic Scoring Systems
Fong’s score* Nordlinger’s score* Node-positive primary tumor Stage of the primary tumor Disease-free interval Disease free interval (≥2 years vs. (<12 months between colon <2 years) resection and appearance of metastases) Size of largest lesion >5 cm Size of the largest metastasis (<5 cm vs. ≥5 cm) More than 1 tumor Number of liver nodules (1–3 vs. 4 or more) CEA >200 ng/mL Resection margin (>1cm vs. <1 cm) Age (<60 years vs. ≥ 60 years
*One point is assigned for each risk factor

Whereas preoperative factors may be generally instructive, these should not be used to exclude patients from surgical consideration. Patients with one or multiple negative prognostic factors can still derive a significant survival advantage from hepatic resection of their CRLM. To conclude, it is important to mention that at the present time the only unchallenged contraindication to liver resection Table 13.1B Survival Rates for Each Score Grade

Survival (%) Fong’s score Score 0 1 2 3 4 5 1 year 93 91 89 86 70 71 3 years 72 66 60 42 38 27 5 years 60 44 40 20 25 14 Median (mo) 74 51 47 33 20 22 Risk groups Low risk Intermediate risk High risk Nordlinger’s score Risk factors 0–2 3–4 5–7 2 years 79 60 43

120

SURGERY FOR METASTATIC COLORECTAL CANCER
The applicability of this score has been evaluated by independent investigators from Norway (50), indicating that the score is applicable to other populations outside of a large tertiary U.S. center. Recent data from Asia has shown that CRS is useful for predicting outcome after ablative therapy of liver metastases. Also the CRS can help to select the extent and sophistication of preoperative assessment (51) acting as a risk stratification tool in identifying patients who are most and least likely to have their management altered by the results of the test. In practice, although the scores are simple, easy to use and highly predictive of long-term outcome their clinical relevance in terms of indications and contraindications to surgery is low since even with poor prognostic factors, hepatic resection can provide a chance of long-term survival. Patients with poor scores could, however, be selected more appropriately for neoadjuvant and adjuvant therapy or for refined preoperative imaging (routine PET CT) to exclude those with contraindications to surgery. In addition, scores have proved useful for comparing results from different centers for surgical and ablative therapies as well as stratification of patients for trials. after the surgery (58). The preoperative cycles induced a complete response in 3.8% of patients and a partial response in 40.1% with a decrease in the diameter of the nodules of 29.5%. At 3 years, the disease-free survival was 28.1% in the group treated with surgery alone and 35.4% in the group that received perioperative chemotherapy (p = 0.058). The reduction of the size of the nodules could modify and facilitate the liver resection with a minor hepatectomy instead of a major liver resection. In patients presenting with five or more bilobar metastases, Tanaka et al. (59) showed that the 5-year survival rate was 38.9% in the group receiving neoadjuvant chemotherapy compared to 20.7% of the group treated with hepatectomy alone. In addition, multivariate analysis revealed neoadjuvant chemotherapy to be an independent predictive factor for survival. These results suggest a survival benefit of neoadjuvant chemotherapy in patients with resectable metastases. Whether the use of adjuvant chemotherapy would translate as the “gold standard” practice is still a matter of debate. Obviously, multinodular metastases are very likely to benefit from neoadjuvant chemotherapy owing to the potential of missing small metastases. Approaches to Surgery Assessment of Functional Hepatic Reserve The functional hepatic reserve can be assessed by Child–Pugh score and hepatic biological blood tests, however, to date the only test which has proven to have a good predictive value is the indocyanine green (ICG) clearance test (60). In candidates for liver resection with retention of less than 20% of ICG at 15 minutes, up to 60% of the volume of the parenchyma can be resected. Although liver metastases rarely develop in cirrhotic liver, with the ever increasing use of more efficient chemotherapy regimens and targeted agents, a rising number of patients are expected to present with damaged livers as a result of chemotherapy given before resection. Specific pathologic changes of the liver parenchyma (vascular changes and/or chemotherapy associated steatohepatitis) influencing the liver regeneration and function as well as the ability of the patient to recover have been observed, following administration of preoperative chemotherapy. Hence in this new context, evaluation of the functional reserve of the liver is becoming critical. Preoperative Biopsies Currently routine biopsy of liver lesions as part of the diagnostic process for patients who are thought to have potentially resectable lesions is not recommended. Although the seeding along the needle track has been believed to be very rare (incidence 0.003–0.07%) (61,62), it appears that it has been greatly underestimated. An incidence of needle track metastases ranging from 10% to 19% has recently been reported (63,64). Therefore, the potential benefits of liver biopsy in suspected patients are outweighted by the risk of these serious complications as well as the risk of deriving false reassurance from a false-negative result. Role of Laparoscopy and Laparoscopic Ultrasound (LUS) Evaluation In the recent years, many surgeons have advocated the use of laparoscopy for evaluation of CRLM preoperatively in order to

management of patients with resectable colorectal liver metastases
Preoperative management Neoadjuvant Chemotherapy Conventional first-line chemotherapeutic regimens for resectable colorectal liver metastases (CRLM) contain fluorouracil (5-FU) in addition to leucovorin. Using a bolus administration regimen for patients treated with 5-FU and leucovorin response rates ranging from 20% to 30% and a median survival of 11.5 months has been reported (52,53). No significant difference in median survival has been observed when the 5-FU was delivered by continuous infusion, despite improvement of the response rate and reduction of the toxicity. Combination of 5-FU with newer drugs such as irinotecan (topoisomerase I inhibitor) resulted in a higher response rate (39%), longer progression free and overall survival time (14.8) compared to 5-FU and leucovorin alone (54). In addition, it has been shown that irinotecan in combination with continuous infusion of 5-FU/ leucovorin (FOLFIRI) produces better response rates and longer progression free and overall survival compared to 5-FU/leucovorin alone (55). More recently, the combination of infusional 5-FU/leucovorin with oxaliplatin (cisplatin derivative) has been found to be less toxic and more efficacious than the bolus irinotecan/5-FU/ leucovorin regimen (56,57). Whether the combination of infusional 5-FU/leucovorin with oxaliplatin (FOLFOX) or FOLFIRI is better as first-line chemotherapy remains controversial as they have comparable response rates. What may be more persuasive is that when these regimens are used sequentially when progression or toxicity occurs, regardless the order, survival is prolonged. For patients with up to four liver metastases, a prospective trial conducted by the European Organization for the Research and Treatment of Cancer compared surgery alone versus surgery with perioperative chemotherapy (FOLFOX 4 – oxaliplatin/5-FU/leucovorin), six cycles before and six cycles

121

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
reduce the number of unnecessary surgical explorations. This procedure has been reported to exclude 25% to 48% of patients from laparotomy with a false-negative rate of less than 15% (65,66). Grobmyer et al. (67) suggested that patients should be considered for laparoscopic evaluation if they have two of the following characteristics: lymph node positive primary tumor, CEA levels greater than 200 ng/mL, >1 hepatic lesion, diseasefree interval <12 months, and hepatic metastatic lesion >5 cm. Patients with two or more of these factors have a 30% chance of having occult extrahepatic disease. However, with increasingly sensitive preoperative imaging and the increasing use of ablation and resection of extrahepatic sites, fewer patients are subjected to nontherapeutic laparoscopy (Level of evidence: 3). Role of Intraoperative Ultrasound (IOUS) IOUS is an essential adjunct to conventional imaging and is widely used to guide surgery and ablative techniques. In experienced hands, IOUS has been shown to contribute to acquisition of precise details regarding tumor size, location, extent of local invasion, and may alter or guide the surgery in up to 67% of cases (68). Also, when compared with preoperative radiological findings, IOUS has been found to be able to identify at least one additional malignant lesion in 10% to 12% of cases (68–70). As such, the use of IOUS should be considered as mandatory not only for intraoperative diagnostics but also for determining the type of surgical procedure (resection). Types of Liver Resection Generally liver resection can be divided into two groups: anatomic (resection of one or several segments) and nonanatomic “wedge” resections (resection of a portion of parenchyma surrounding the metastatic lesion). If more than three segments are resected, the hepatectomy is defined as major. Different types of anatomic liver resection are performed: right hepatectomy (segments V–VIII), left lateral lobectomy (segments II, III), and left hepatectomy (segments II–IV). Other types include central resection (segments IV, V, VIII) and bisegmentectomies (segments V, VI or segments IV, V). Resections exceeding the boundaries of a normal right or left are defined as extended hepatic resections and are divided in six different types. Right hepatectomies extended to segment IV, segment I, or both. Similarly, extended left hepatectomies may include segment I, segments V and VIII, or segments I, V, and VIII. Selecting the Resection Type The principles of hepatic resection (including the oncological goal which is to remove all metastatic sites with tumor free margins) are no different for colorectal metastases than for any other hepatic surgery. Rather than dogmatically adhering to an anatomical versus nonanatomic approach, the hepatobiliary surgeons now guide their decisions aiming ultimately at resecting all metastases with negative histologic margins. Therefore, the type of resection chosen for a particular patient is and should be individualized based on the size, number, and location of the metastases, their relation to main vascular pedicles, and the volume of future liver parenchyma. Whereas a small, superficial metastasis can be best treated with a nonanatomic resection, a large metastasis deeply located within the

Figure 13.1 Picture showing multiple metastasectomies with maximal preservation of the liver parenchyma.

liver may be treated with an anatomical, segment-oriented resection. The extent of liver resection (major vs. minor or anatomic vs. nonanatomic resection) is not by itself a prognostic factor. Therefore, independently of being anatomic or nonanatomic, resection should spare as much as possible the nontumoral parenchyma, bearing in mind that new recurrences could eventually develop for which surgery could possibly be indicated again (see Fig. 13.1).

postoperative management
Adjuvant Systemic Chemotherapy At present, despite several chemotherapy regimens, data would support the use of 5-FU and leucovorin as adjuvant chemotherapy after liver resection if patients have not previously failed this regimen. Portier et al., in a multicenter trial that randomized 173 patients after hepatectomy for CRLM to surgery alone or to surgery followed by chemotherapy (5-FU/ Leucovorin), demonstrated that patients who received adjuvant chemotherapy had a significantly better disease-free survival compared to that of patients treated with surgery alone (34% vs. 27%; p = 0.03) (71). A year later, Park et al., in a large two-center study comparing 518 patients treated with no chemotherapy (379 American, 139 European) to 274 patients treated (240 American, 34 European) with 5-FU-based adjuvant chemotherapy, demonstrated that systemic adjuvant chemotherapy prolongs survival after hepatic resection for colorectal metastases (72). Patients subjected to adjuvant chemotherapy had improved survival (p = 0.007) even after stratification by clinical risk score (p = 0.001). In every clinical risk score category, patients subjected to adjuvant chemotherapy had a higher chance of survival (range 1.3–2.0 times). Meanwhile, for those who have previously failed this regimen, an oxaliplatin- or irinotecan-based regimen should be considered. Despite that there has not been a clear demonstration of efficacy of any regimen, the higher response rate observed in patients treated with FOLFOX or FOLFIRI over the 5-FU/leucovorin has resulted in many groups to preferentially use these regimens in adjuvant settings.

122

SURGERY FOR METASTATIC COLORECTAL CANCER
Table 13.2 Reported Survival Outcomes after Resection of Colorectal Liver Metastases with Curative Intent
No. of patients 1568 1001 235 26 226 615 585 190 193 100 557 423 Operative mortality (%) 2 2.8 0.8 – 1 1 0 – 1 – 1 2 Patient survival (%) Postoperative morbidity (%) – – – – 18.6 18 – – 26 1 – 19.6 1 year – 89 – – 93 91 – – 69 86 97 93 3 years – 57 51 62 57 61 – 73 46 66 74 – 5 years 28 36 38 32 40 41 33 58 43 58 58 47 10 years – 22 38 – 26 – – – – – – 28

Author Nordlinger et al. Fong et al. Minagawa et al. Suzuki et al. Choti et al. Adam et al. Kato et al. Abdalla et al. Tanaka et al. Fernandez et al. Pawlik et al. Wei et al.

Year 1995 1999 2000 2001 2002 2003 2003 2004 2004 2005 2005 2006

Adjuvant Intra-arterial Chemotherapy A number of studies have reported the safety, efficacy, and feasibility of adjuvant regional hepatic chemotherapy. Kusunoki et al. conducted a nonrandomized trial of HAI versus systemic chemotherapy after radical liver surgery. He showed that the 5-year survival was significantly better for the HAI group compared to the 5-year survival of the systemic group (59% vs. 27%, p < 0.001) (73). Kemeny et al., in an intergroup study of 109 patients randomized to surgery alone or surgery and HAI-FUDR, demonstrated that the 4-year disease-free survival was significantly better in the HAI group (67% vs. 43%) (74). In another larger study, 156 patients were randomized to resection and systemic 5-FU or resection and combined systemic 5-FU and HAIFUDR. The patients who were treated with regional therapy had a significantly better 2-year survival (86% vs. 72%) and markedly improved liver disease control (75). In conclusion, convincing evidence currently exist to support the use of adjuvant chemotherapy, either systemic or regional, to prevent to some extent the risk of recurrence following liver resection. Outcomes of Resection for Colorectal Liver Metastases Morbidity and Mortality Overall, the perioperative mortality of liver resection for CRLM does not exceed 2%, ranging between 0% and 5% in most published series (10,11,76) and is strongly influenced by perioperative blood loss, liver function, and extent of liver resection. In experienced units, even major hepatic resections, constituting around 50% of cases have perioperative mortality not exceeding 2% (76). The principal causes of death are liver failure and sepsis. It has been observed that the mortality has changed little over the last two decades, however, this does not mean that there has not been progress made. With improved safety, surgeons are increasingly performing more extensive resections,

which explain the fact why the operative mortality and longterm survival have plateaued. In contrast, the perioperative morbidity rate is reported to be greater than 20% (28,77). The major morbidity associated with liver resection includes hemorrhage (1–3%), bile leak and/or fistula (4%), pleural effusion/pneumonia (5–10%/5–20%), and hepatic failure (3–8%). Of the nonliver-related complications, intra-abdominal sepsis is found to be the most frequent major complication, and pulmonary infection is the most frequent minor complication. Among liver-related complications, liver failure is the most serious and occurs in 3% to 8% of all major liver resections often being lethal. Similarly, intraoperative hemorrhage, although rare, is another major complication with a mortality as high as 17% (78). Long-term Survival Results Large series have reported a 5-year survival after hepatectomy for CRLM of 35% to 52% with a 33- to 46-month median survival (8,9,46,79) (see Table 13.2). However, recent data have shown an improved 5-year survival rate of 58% after complete resection of CRLM (35). Also, a number of series with sufficiently long-term follow-up indicate that the 10-year survival after resection can be expected in 20% to 30% of patients (12,46,80) (see Table 13.2 and Fig. 13.1). Similarly the International Registry of Hepatic Metastases of Colorectal Cancer (LiverMetSurvey), which to date includes more than 8000 patients, has demonstrated a 5- and 10-year survival of 41% and 26%, respectively. An important oncologic question is whether the recently improved systemic therapies can achieve the same results as resection for CRLM? This seems unlikely considering that longterm survival beyond 5-years is rare without liver resection (5) (Fig. 13.2). Indeed, the survival results can be questioned if considering that the patients who undergo resection are selected and may have better outcomes due to less aggressive disease.

123

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
However, there has never been a controlled trial to compare resection versus nonresection or conservative treatment of potentially resectable CRLM and this is unlikely to happen in the near future unless more efficacious systemic therapy regimens are discovered. Management of Nonresectable Metastatic Disease Despite the advances made so far in liver surgery, approximately 80% to 90% of patients with CRC liver metastases are not candidates for liver resection at the time of diagnosis. Apart from the fitness of the patients, the unresectability of liver lesions is due to the following reasons: technically unable to completely remove the lesions due to the number, size, and their distribution; or due to ill location of the metastatic lesion (infiltration of IVC, confluence of hepatic veins). As the most frequent cause responsible for technical unresectability is multinodular bilobar metastatic disease, different approaches used alone or in tailored combinations have been developed to improve the resectability rate by either reducing the tumor burden (in turn the extent of the hepatic resection) or by increasing the volume of remnant liver parenchyma. Instead, ill-located metastases are being increasingly treated by radical surgery such as liver resection combined with total vascular exclusion (TVE). Chemotherapy to Downstage Nonresectable Metastatic Disease Systemic Chemotherapy The improved efficacy of chemotherapy agents has not only allowed increased patient survival in the noncurative setting, but has allowed a subset of previously unresectable patients to undergo liver surgery after “tumor downstaging,” a concept first introduced by our team (81). By reconsidering the initial unresectability of patients who strongly respond to chemotherapy, several investigators have shown that survival can be achieved by liver resection in a significant proportion of patients who otherwise would have had a poor outcome (Level of evidence: 3). A study conducted by our group (81) demonstrated that an additional 16% of patients who had initially unresectable liver metastases became candidates for hepatic surgery after receiving systemic chemotherapy. The 3and 5-year survival rates were 54% and 40%, respectively, close to those observed after resection of initially resectable nodules. These results were confirmed by other studies including ours (82,83). In 2004, we reported that subsequent rescue surgery for unresectable CRLM downsized by chemotherapy resulted in a 5- and 10-year survival rate of 33% and 23%, respectively, with a disease-free survival of 17% at 10 years (82) (Figs. 13.3 and 13.4). In contrast, patients with tumor progression during preoperative chemotherapy have a significantly worse outcome, with a 5-year survival of 8% versus 37% and 30% for patients with objective tumor response or tumor stabilization (84). Patients with tumor progression still had a poor prognosis even when a potentially curative hepatic resection was performed. Another aspect worth mentioning about is the combination of chemotherapy with new molecular-targeted drugs (bevacizumab, cetuximab). These agents have had a significant impact on the survival of patients with advanced CRC disease when integrated with chemotherapy in trials. Using them in combination with oxaliplatin- or irinotecan-based regimens, these agents have produced tumor response rates greater than 50% to 60% (85). Disease control rates (complete response; partial response or stabilization of disease) exceeded 90% in the report of a phase II study of FOLFOX combined with cetuximab in nonoperable patients with epidermal growth factor receptor-expressing metastatic CRC (86). The objective response rate was 79% according to independent expert review (87). Data from the Paul Brousse series showed that the use of targeted agents in second line therapy also increases the number of patients eligible for resection. A total of 131 patients with epidermal growth factor receptor-positive CRLM who had progressed following two or more lines of FOLFOX or

100 90 80 70 60 50 40 30 20 10 0 0

Patient survival after a 1st liver operation for colorectal metastases: 8179 patients Log rank p = <0.0001

41% 7737 resected patients 26%

7% 1 2 Resected 3 4 5 Resected 6

442 nonresected patients 7 8 9 10

Nonresected

Figure 13.2 Five- and ten-year survival following hepatectomy for colorectal liver metastases. Source: www.livermetsurvey.org.

124

SURGERY FOR METASTATIC COLORECTAL CANCER
FOLFIRI regimens were treated with cetuximab, resulting in conversion of 7% unresectable patients to resectable. Certainly, recent results form the randomized trials (FOLFIRI vs. FOLFIRI/Cetuximab – CRYSTAL trial; and FOLFOX vs. FOLFOX/Cetuximab – OPUS trial) (88,89) add further evidence to the benefit conferred by cetuximab on the response and resection rates in patients with advanced CRLM treated with standard first-line therapies. As a result of using combined chemotherapy regimens, the resection rates have significantly increased compared to regimens of FOLFOX or FOLFIRI alone. Furthermore, in two other studies, cetuximab conferred an increase in response rate and resection rate over standard chemotherapy alone, with the benefits being the greatest for patients with KRAS wild-type tumors; CRYSTAL 59% versus 43% and OPUS 61% versus 37% (90,91). Intra-arterial Chemotherapy The interest in using intra-arterial chemotherapy in neoadjuvant setting has also progressively increased as it has been demonstrated to have a high response rate in both the firstand second-line settings. Clavien et al., using HAI-FUDR with or without leucovorin, induced resectability in 6 (26%) of 23 previously treated patients. The actuarial survival rates at 3 years were 84% for responders to neoadjuvant therapy compared with 40% for nonresponders (92). In a Memorial Sloan-Kettering study (93), 44 patients with extensive liver metastases received HAI-FUDR and dexamethasone plus oxaliplatin-based systemic chemotherapy as part of two Phase I trials. The study population in this trial had a high number of patients with more than 4 metastases, metastases greater than 5 cm, more than 25% liver involvement with tumor, a CEA level greater than 10 ng/dl and previously chemotherapy exposure. Despite these negative parameters, the objective response rate was 82%, resulting in complete gross resection of tumor in 9 (20%) of the 44 patients and a median survival for all patients of 26 months. Recently, preliminary data from

Paul Brousse Hospital – 473 patients (Apr. 88–Jul. 99) 100 80 66% Survival (%) 60 40 20 0 0 1 2 3 4 5 6 Years 52% 33% 23% No surgery 7 8 9 10 p = 0.01 48% 30% 91% Resectable: 335 Initially non-resectable: 138 No surgery

Figure 13.3 Curves demonstrating 5- and 10-year survival for initially resectable patients and for patients who underwent rescue surgery. Source : From Ref. (82).

(A)

(B)

(C)

(D) Figure 13.4 Unresectable colorectal liver metastases downsized by chemotherapy.

125

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
several clinical trials using the oxaliplatin or irinotecan via HAI have been promising. In summary, regardless the type of chemotherapy used in unresectable patients, a significant proportion (15–30%) is switched to resectability. This proportion of patients will probably expand with the increasing efficacy of chemotherapy and biological agents, justifying a close collaboration of oncologists and surgeons in the multidisciplinary treatment of these patients. Techniques to Improve Resectability In addition to preoperative chemotherapy, a number of interventional/surgical techniques are available to achieve a situation of resectability include tumor ablation techniques, portal vein embolization, two-stage liver resection, and extended liver surgery (total vascular exclusion and cooling). Tumor Ablation Techniques Locally ablative modalities such as radiofrequency ablation (RFA) and cryotherapy are both techniques used either independently or as adjunctive to surgery. These techniques are regarded as complementary to hepatectomy when complete resection cannot be achieved. The strategy of using them can result in an increased number of initially unresectable patients in whom the curative treatment can be accomplished.


found that the rate increases when treating multiple lesions (>8), large lesions (>3 cm), or tumors located to major blood vessels (blood warmth may impair the freezing process).

RFA is currently the most commonly applied ablation method. RFA involves localized application of conductive thermal energy to destroy tumor cells. Specifically alternating electric current in the range of radiofrequency waves (460 kHz) is applied from a generator through a needle electrode placed directly into the tumor.

Limitations of RFA are related to the lesion size (suitable for lesions ≤3 cm) or when a maximum of three tumors are present as well as the anatomical location of the tumor. In the vicinity of large hepatic vessels, the heat sink effect significantly increases the risk of incomplete ablation. Also, the risk of thermal injury is increased when nodules are close to main biliary structures or to extrahepatic organs. RFA procedure, when performed in combination with surgery, increases the resectability and curability for patients in whom hepatic resection alone is not curative. Adding RFA to hepatic resection has been reported to be well tolerated with a perioperative morbidity and mortality comparable to those seen after resection alone (94). For metastases considered as unresectable, RFA combined with hepatic resection can achieve a median survival as high 37 months (95).


Cryotherapy involves freezing and thawing of liver tumors by means of a cryoprobe. Tumor necrosis occurs by direct cellular freezing and indirectly through vascular thrombosis and tissue anoxia. Results of such treatment combined with hepatic resection for patients not eligible for hepatic resection alone have shown a 5-year survival rate of 24%, better than those obtained by palliative chemotherapy (96,97). Local recurrence at the site of cryotherapy occurs in 5% to 44% of patients and it has been

Portal Vein Embolization Portal vein embolization (PVE), which was first described by Makuuchi (98), is used to trigger a compensatory hypertrophy of the future remnant liver. In patients with an otherwise normal liver, current guidelines recommend preoperative PVE when the ratio of the remnant liver volume is <30%. Patients submitted to prolonged chemotherapy with a high risk of induced hepatic lesions should benefit from this method when this ratio is less than 40%. PVE can be performed percutaneously or using the ilocolic vein approach via a limited laparotomy. After PVE, hepatic volume is routinely evaluated using CT scanner volumetry, which gives information about the degree of the compensatory hypertrophy as well as the status of the metastatic disease. The optimal time interval necessary to induce maximum hypertrophy after PVE has not been established yet, although some Japanese teams use to perform resection as early as 2 weeks after the PVE. The majority of groups, however, would usually use a 4 to 6 weeks of interval between PVE and surgery. PVE is safe and does not add significant morbidity. In our series, a significant increase of liver volume following preoperative PVE was observed in 43% of patients, allowing 63% of originally unresectable liver metastases to be subsequently operated (99). The feasibility and the influence on the outcome in patients requiring an extended hepatectomy has been reported by other investigators also. Farges et al. (100) published the results of a prospective study of PVE performed in patients undergoing right hepatectomy for either primary liver cancer or metastatic liver disease. They demonstrated significantly fewer postoperative complications when PVE was used to increase the FLR volume in patients with chronic liver disease whose anticipated FLR was <40%. In contrast, patients with normal liver function who underwent a right hepatectomy did not benefit from PVE, as it was expected, since the remaining liver usually represents more than 30% of the functional liver volume. In summary, the PVE needs to be performed only in patients who are being considered for an extended right hepatic resection. PVE is rarely necessary prior to extended left hepatectomy because the right posterior sector typically constitutes about 30% of the total liver volume (101,102). On the other hand, in patients who have been treated with heavy neoadjuvant chemotherapies with a high risk of induced parenchymal liver lesions the PVE should be performed when the ratio of the remnant liver volume to the total estimated liver volume is less than 40%. The selective use of PVE may enable safe and potentially curative hepatic resection in a subset of patients with advanced colorectal metastases who would otherwise have been marginal candidates for resection because of an inadequate FLR or significant underlying liver disease.

126

SURGERY FOR METASTATIC COLORECTAL CANCER
Two-stage Hepatectomy The concept of “two-stage liver resection” to deal with multinodular CRC metastatic disease that cannot be resected in a single procedure owing to a too small volume of the future remnant liver was first described by our group (103). During the first stage, the less invaded hemiliver (usually left) is completely cleared of metastases by resection, which could be associated with a simultaneous portal vein ligation/embolization of the most involved hemiliver (usually right) – or percutaneous portal vein embolization 1 week later. The aim of this step is to minimize the risk of liver failure by performing a second and complete resection once regeneration induced by the portal vein embolization has taken place. Finally a second stage hepatectomy will be carried out to completely remove the liver harboring the remaining metastases (Fig. 13.5). The success of this method relies on the liver regeneration between the two interventions, allowing the second hepatectomy to be performed with a lower risk of complications, including liver failure. Our experience, as well as that of others, has demonstrated that this strategy can be carried safely and effectively in selected patients with initially nonresectable multiple bilobar CRLM (104–107) (Table 13.3). In our latest study, the 3- and 5-year survival rates were 60% and 42%. It should be mentioned that during the first stage performing nonanatomic

(A)

(B)

(C)

(D)

Figure 13.5 Radiological follow-up of a patient treated with combination of neoadjuvant chemotherapy and two-stage hepatectomy. 1A, hepatic metastases before chemotherapy treatment. 1B, planning of surgery after tumor downstaging (before the first hepatectomy). 1C, first hepatectomy. 1D, liver remnant following the second hepatectomy (segments IV and I).

Table 13.3 Reported Survival Outcomes after Two-Stage Hepatectomy for Colorectal Liver Metastases
Patient survival (%) Author/Institution Adam et al. (2000) Jaeck et al. (2004) Shimada et al. (2004) Togo et al. (2005) Adam et al. (2007) No of patients 16 33 12 11 45 Success rate (%) 81 76 100 100 69 Mortality rate (%) 15 0 0 0 6.5 Morbidity rate (%) 45 56 NA 18 48 Median (months) 31 – – 18 35 3 years 35 54 – 45 47 5 years – – – – 28

NA, not available. The mortality rates concern the second operation.

127

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
wedge resection is advantageous as it preserves a maximal amount of liver parenchyma that will hypertrophy after PVE to become the functional liver remnant. Also, our policy is to perform portal ligation and embolization with absolute alcohol during the first liver resection to avoid a second procedure before the definitive hepatectomy. Currently, the guidelines for two-stage hepatectomies include the following:
● ●

Patients with bilobar multinodular metastases for which a planned resection would leave more than three nodules or any nodule larger than 3 cm in the remnant liver could be candidates for two-stage hepatectomy (Fig. 13.6C).





No residual tumor should be left in the future remnant liver at the first intervention. If the resection alone cannot remove all lesions, one of the ablative methods (RFA, cryotherapy) has to be used for local tumor destruction in order to prevent tumor progression during the regeneration of the remnant liver. At the first intervention, portal dissection and mobilization of the lobe that is to be resected during the second intervention should be avoided.

In summary, based on the nature of the metastatic disease (number, size, and distribution), the treatment strategies, which can be applied with the aim of achieving a complete treatment of CRC liver metastases include the following:




Patients with unilobar multinodular metastases requiring resection of more than 60% to 70% of the functional liver parenchyma should undergo preoperative portal vein embolization. Following PVE, the induced hypertrophy of the future remnant liver allows for a curative resection while minimizing the risk of postoperative hepatic insufficiency (Fig. 13.6A). Patients with bilobar multinodular metastases for which a planned resection would leave no more than three nodules and none larger than 3 cm in the remnant liver are preferentially treated with a multimodal approach consisting of hepatic resection combined with RFA or cryotherapy of the unresectable nodules (Fig. 13.6B).

Extended Liver Surgery (Total Vascular Exclusion and Cooling) Involvement of the IVC and/or the confluence of hepatic veins by liver metastases is another situation that can be considered as a contraindication to liver resection. Currently, employing total vascular exclusion (TVE) of the liver and vascular reconstruction techniques can make surgery possible without taking further risks for this specific group of patients. As the experience has grown with TVE, an increasing number of patients are being operated with acceptable morbidity and mortality. Conventional TVE consists of clamping of the liver inflow (Pringle maneuver) as well as clamping of the supra and infrahepatic vena cava. Alternatively, in cases with no caval involvement by the tumor, selective control of the hepatic veins can be achieved allowing preservation of the caval flow. In cases whereby the caval clamping is associated with hemodynamic disturbances (hypotension), a venovenous bypass is necessary through which venous blood from femoral and portal vein is diverted to axillary or internal jugular vein. A drawback of these techniques, however, is that almost inevitably would induce warm ischemia for which the maximal duration of tolerance is assumed to be around 60 to 90 minutes. For cases which require interruption of hepatic blood flow for more than 60 minutes, hypothermic perfusion of the liver should be instituted to prevent the consequences of a long warm ischemic time. Such combination was evaluated in a study conducted in our center, which demonstrated that TVE combined with hypothermic perfusion was associated with a better ischemic tolerance and liver function as well as significantly lower complication rates compared to TVE ≥ 60 min. Combined liver and vena cava resection is another procedure facilitated by combined TVE and hypothermic perfusion. In

Right lobectomy Remnant liver <30%

Right hepatectomy ≤3 Metastases ≤ 30 mm (remnant liver)

Right hepatectomy >3 Metastases > 3 mm (remnant liver)

Portal vein embolization (A)

Hepatectomy + RFA or cryo (B) (C)

Two – stage hepatectomy

Figure 13.6 Diagrammatic illustration of the surgical strategies used when treating patients with “nonresectable” multimodal metastatic disease. (A) Multifocal unilobar metastases. (B) Multifocal bilobar metastatases. (C) Multifocal bilobar metastases. RFA, radio frequency ablation; Cryo, cryotherapy.

128

SURGERY FOR METASTATIC COLORECTAL CANCER
a series from our center (108), out of 22 patients who underwent such a procedure, one patient died (4.5%) during the perioperative course, whereas 14 patients (64%) developed complications. Overall 5-year survival for the operated patients was 38.3%, comparing favorably with other reported results. Hence, combining TVE with vascular reconstruction techniques has resulted in an increased number of patients undergoing surgery for CRC liver metastases involving vena cava and/or the confluence of hepatic veins (which not so long ago were considered as a contraindication to surgery). This approach, however, seems justified only for surgical teams experienced in both hepatobiliary and vascular surgery. Repeat Liver Resection for Recurrent Metastatic Disease Despite hepatic resections with a curative intent in wellselected patients, up to 60% subsequently will develop recurrent liver metastases. Of these, approximately 20% to 30% present with isolated recurrent liver metastases, which are potentially amenable to further resection. Regardless of the technical challenges due to adhesions and altered anatomy of the liver, the repeat hepatectomy is safe with a postoperative mortality and morbidity not different from those reported after a first resection (median survival approaches 2 years) ( Table 13.4). Five-year survival rates ranging from 16% to as high as 41% have been reported (109–111). Not surprisingly, the same prognostic factors that predict favorable outcome after primary resection apply to the repeated liver resection, including complete removal of metastatic lesions with satisfactory margins and no extrahepatic disease. Furthermore, a study conducted by our team demonstrated that a third hepatectomy is safe, with complication rates and survival benefit similar to first and second hepatectomies (112) (Fig. 13.7). The overall 5-year survival following the third hepatectomy was 32% and disease-free survival was 17%. Similarly, Pessaux et al. showed overall 5-year survival rates of 33%, 21%, and 36%, respectively, after a first, second, and third hepatectomy (113). Also, in repeat resections, the general rule applies that it does not matter how many lesions the patient has, provided that an R0 resection

Table 13.4 Reported Survival Outcomes after Repeat Liver Resection for Recurrent Colorectal Metastases
Patient survival (%) Author/Institution Fernandez et al. Adam et al. Yamamoto Muratore et al. Suzuke et al. Petrowsky et al. Adam et al. Shaw et al. Year 1995 1997 1999 2001 2001 2002 2003 2006 No of patients 170 64 75 29 26 126 199 66 1 years – 87 48 – – 86 89 – 3 years 45 60 31 35 62 51 46 – 5 years 32 41 – – 32 34 32 44

(%) 100 89% 88% 82% 54% 60 46% 40 42% 36% 32% 28% First hepatectomy Second hepatectomy Third hepatectomy

80

20

0 0 1 No 416 139 6 2 1 yr 267 80 49 3 2 yrs 169 37 31 3 yrs 120 27 15 4 4 yrs 83 19 10 5 yrs 60 13 6 5 Year

Patients at risk First hepatectomy Second hepatectomy Third hepatectomy

Figure 13.7 Survival after 1st, 2nd, and 3rd hepatectomy from the time of the index operation. Source: From Ref. (112).

129

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
can be achieved within limits of safety in terms of liver volume and function. Therefore, hepatic recurrences should be regarded as oncologically similar to metastatic disease at initial presentation and repeat hepatectomies should therefore be offered to patients based on the same criteria as those used for initial hepatectomy. Special Considerations Despite the advances made in the management of CRLM, there are still some areas of uncertainty or debatable. For instance, it is not clear what type of treatment is needed after a complete clinical response (total disappearance of metastases while on chemotherapy). Similarly, the management of synchronous presentation of primary colorectal cancer and hepatic metastases is still disputed (chemotherapy or upfront surgery) and so is the issue of which site should be operated first—bowel or liver? Treatment of the Lesions That Have Disappeared After Neoadjuvant Treatment With the advances in chemotherapy efficacy, the frequency of “missing metastases” has increased. Nevertheless, the treatment strategies concerning such lesions are not well defined, particularly so when trying to decide about the necessity to resect, the time, and type of resection. In a study conducted by our team (82), we initially reported that up to 7.2% of the patients with unresectable CRLM treated with systemic chemotherapy developed complete metastatic necrosis. Hence, we recommended that preferably all tumor-bearing sites must be resected during the surgery for CRLM. And, while later on, it was suggested that “missing metastases” are cured in 70% of the cases (114), increasingly, the evidence indicates the contrary—a persistence of histologically active tumor in as many as 83% of the lesions, which have a complete radiological response on imaging (115). Furthermore, a subsequent report from our unit (116) demonstrated that the actual number of patients with no more residual tumor cells (complete pathological response, CPR) in CRLM after neoadjuvant chemotherapy was even smaller (4.5%) than the one previously reported rate, which is in keeping with reports from other centers (117). Considering these results, we can say that a complete radiological response does not mean complete histological response, and despite the favorable long-term results associated with the CPR (5-year survival of 76%) the utility of surgery remains unchallenged. This view is supported by several reasons: (1) Confirmation of CPR depends primarily on the accuracy of the pathologic examination and on the exhaustivity of histologic sampling as undetected malignant cells could still be present in the resected lesion/s. By resecting all metastases, the possibility of leaving residual tumor cells behind is greatly reduced; (2) During the laparotomy it is often possible to diagnose additional metastatic disease, which otherwise would have remained undetected by the standard investigative tools; (3) The diagnosis of the CPR is a retrospective one as there is no imaging technique, which can reliably diagnose CPR preoperatively, hence, only surgical resection with concomitant pathological examination is able to make a definite diagnosis. In practice, due to very low overall incidence of CPR (4%), patients who have a complete radiological response should be operated on. Patients treated with neoadjuvant chemotherapy should be referred to surgeons before the initiation of the treatment, so as to avoid eventual difficult decisional situations brought about by the inability to localize previously seen radiological lesions. Synchronous Metastases ● Neoadjuvant chemotherapy or upfront surgery? The interest in using preoperative chemotherapy for resectable patients has been increasing. The rationale for this policy has been supported by the better prognosis obtained with neoadjuvant chemotherapy and surgery, compared to upfront surgery in patients with synchronous CRLM. Administering neoadjuvant chemotherapy not only can be associated with a lower rate of positive surgical margins compared to the rates observed in patients treated with upfront surgery (118), but also such approach provides time to identify a subgroup of patients who will develop progressive disease while on chemotherapy (119). Concerning the later group (patients with progressive disease while receiving neoadjuvant chemotherapy), a study conducted by our team showed that such patients had a 5-year survival of 8% versus that of 37% observed on patients with an objective tumor response to neoadjuvant chemotherapy (119). In addition, a study conducted by Rubbia-Brandt et al. (117), using a tumor regression grade scoring system, identified that resected patients with a poor histological response to chemotherapy had a lower disease-free survival at 3 years and a lower overall survival at 5 years. Certainly taking into consideration these results, the utility (benefit) of surgical intervention in this subgroup of patients has to be questioned. Therefore, the decision to give neoadjuvant chemotherapy should be individualized and based on specific clinical situations.








In patients who are chemonaïve with four or more synchronous CRLM and a nonocclusive primary, neodjuvant chemotherapy can be appropriate followed by repeated MRI and PET CT. Neoadjuvant chemotherapy can also be administered in patients with two to three bilobar CRLM. By contrast, if a patient belonging to this group has comorbidities or there is a concern about chemotherapy-related hepatotoxicity at a time when an extended resection is required, initial surgery would be indicated. Instead for patients with one to two unilobar metastatic diseases, upfront surgery should be considered first. Single or staged intervention? The optimal timing for resection of synchronous CRLM and the primary tumor remains a matter of controversy. Most surgeons

130

SURGERY FOR METASTATIC COLORECTAL CANCER
prefer a staged approach with initial resection of the colorectal primary followed by hepatic resection 8 to 12 weeks after. Supporters of this strategy argue that the combined approach is associated with increased morbidity and mortality (Level of evidence: 3). Nordlinger et al. (120) reported an operative mortality of 7% for combined resection compared to 2% for staged resection. Bolton and Fuhram (121) in their series reported a mortality rate of 12% for combined resections, which increased to 24% for those who underwent major liver resection. Reddy et al. (122), in a multi-institutional retrospective study comparing postoperative outcomes after simultaneous and staged colorectal and hepatic resections, concluded that caution should be exercised before performing simultaneous colorectal and major hepatic resections. For major hepatectomy, simultaneous colorectal resection increased mortality (8.3% vs. 1.4%, p < 0.05) and severe morbidity (36.1% vs. 15.1%, p < 0.05) as compared to combined minor liver and colorectal resection. Similarly, a recent study demonstrated that patients who underwent a combined resection had a higher mortality rate (10%) compared with patients treated by staged resection (1.1%), concluding that combined interventions should be performed in wellselected patients, <70 years old and not with rectal surgery (123). On the other hand, several studies have also reported that simultaneous resection of the colon and liver tumors results in morbidity and mortality comparable to staged resection (124,125). However, in the majority of these studies, the patients submitted to simultaneous resection, underwent limited liver resection, and were much more selected compared to those who underwent staged surgery by the same teams. In practice, simultaneous resections should be decided on an individual basis. Combined resections may be more appropriate in patients who require a straightforward colon resection and a limited liver resection (≤2 segments). Patients who require major liver resections particularly the elderly should be dealt with by staged resection. Ultimately, the final decision should be made by the operating surgeon based on the experience and the risk evaluation. In summary, it is recommended that colorectal and major liver resections (>3 segments) should not be performed during the same time. One-stage procedure (combined liver and colorectal resection) should be reserved for experienced teams sharing both colorectal and liver surgery expertise. Surgery for Synchronous Liver Metastases: Liver or Colon Resection First? The standard approach for synchronous CRLM consists of resection of the primary tumor followed by chemotherapy for 3 to 6 months with the goal of resecting the liver metastases if they stabilize or respond. However, this strategy has pitfalls as many patients have progression of their metastatic disease while being treated for their primary, precluding eventual surgery with curative intent. Based on this observation, Mentha et al. (126) designed a management strategy that involves highinduction chemotherapy first, followed by liver surgery, and completed by removal of the primary colorectal tumor. Such strategy aims at controlling the CRLM at the same time as the colorectal primary, optimize the chances of curative liver resection, and allowing the administration of well-programmed chemoradiotherapy before rectal surgery (when indicated). The authors have shown that the new “reverse” approach produced resectability and survival rates better than those expected from the published data on patients treated “conventionally” for disease of similar severity (3-year survival of 86%). The obvious candidate for this treatment would be a patient with nonobstructive primary colonic tumor. The rationale of this approach is that in the majority of patients the most life-threatening site is represented by the liver. In summary, the first treatment should focus on the global metastatic disease rather than locally treat the primary tumor: primary chemotherapy seems to be better than primary resection. For unresectable liver metastases with a primary colorectal cancer in place, chemotherapy as the first treatment line does not alter the survival expectancy. The first surgical procedure should logically deal with the tumor site, which is more difficult to resect and more likely to be life threatening for the patient.

conclusions
The surgical treatment of colorectal hepatic metastases represents the only potentially curative therapeutic option able to achieve long-term survival and a hope for cure. Newer treatment strategies have shifted from the traditional concept of successive lines of medical therapy to that of a continuum of care in which medical and surgical treatment combinations are tailored to the clinical settings. To optimize the treatment of CRLM, management by a multidisciplinary team consisting of oncologists, surgeons, and radiologists is of the utmost importance. Advances in body and hepatic imaging has allowed for more accurate selection of patients with colorectal liver metastases. Imaging modalities are now able to detect minimal metastatic, which not very long ago would have been very difficult to do so. The significance of the prognostic factors has changed, although helpful in stratifying patients with regards to prognosis, should not be used to exclude otherwise resectable patients from surgery. Data on the use of neoadjuvant and adjuvant therapy to decrease recurrence risk and improve survival in patients with initially resectable metastases are encouraging, while further evidence and assessment is needed. With newer chemotherapy regimens, a significant proportion of unresectable patients are currently switched to resectable, opening the way to a survival benefit, which is not very different to that of initially resectable patients. The use of modern surgical techniques has resulted in a reduction of perioperative mortality and morbidity, whereas tumor ablation techniques, PVE, and radical liver

131

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
surgery in combination with neoadjuvant chemotherapy have positively influenced the expansion of candidates for surgical resection. In addition, with the use of more active systemic chemotherapy as adjuvant therapy, we hope that an improved survival rate in resected patients will be observed. In patients with tumor recurrence following hepatectomy for CRLM, repeat hepatectomies provide long-term survival benefit similar to that of first hepatectomy. In future, better patient selection through improved imaging techniques and identification of genomic markers as well as further advances in pharmacotherapy will likely further improve the outcome for patients with CRLM.
20. Kamel IR, Bluemke DA. MR imaging of liver tumors. Radiol Clin North Am 2003; 41: 51–65. 21. Bipat S, van Leeuwen MS, Comans EF, et al. Colorectal liver metastases: CT, MR and PET for diagnosis: meta-analysis. Radiology 2005; 237: 123–31. 22. Truant S, Huglo D, Hebbar M, et al. Prospective evaluation of the impact of [18F] fluoro-2-deoxy-d-glucose positron emission tomography of resectable colorectal liver metastases. Br J Surg 2005; 92: 362–9. 23. Lubezky N, Metser U, Geva R, et al. The role and limitations of 18F fluoro-2-deoxy-d-glucose positron emission tomography (FDGPET) scan and computerized tomography in restaging patients with hepatic colorectal metastases following adjuvant chemotherapy: comparison with operative and pathological findings. J Gastrointest Surg 2007; 11: 472–8. 24. Rappeport ED, Loft A. Liver metastases from colorectal cancer: imaging with superparamagnetic iron oxide (SPIO) enhanced MR imaging, computer tomography and positron emission tomography. Abdom Imaging 2007; 32: 624–34. 25. Wiering B, Crabbe PF, Jager GJ, et al. The impact of 18F fluoro-2-deoxyd-glucose positron emission tomography in the management of colorectal liver metastases. Cancer 2005; 104: 2658–70. 26. Joyce DL, Wahl RL, Patel PV, et al. Preoperative positron emission tomography to evaluate potentially resectable hepatic colorectal metastases. Arch Surg 2006; 141: 1220–6. 27. Bipat S, van Leeuwen MS, Ijzermans JN, et al. Imaging and treatment of patients with colorectal liver metastases in the Netherlands: a survey. Neth J Med 2006; 64: 147–51. 28. Ekberg H, Tranberg KG, Anderson R, et al. Determinants of survival in liver resection for colorectal secondaries. Br J Surg 1986; 73: 727–31. 29. Doci R, Gennari L, Bignami P, et al. Morbidity and mortality after hepatic resection of metastases from colorectal cancer. Br J Surg 1995; 82: 377–81. 30. Scheele J, Strang R, Altendorfhofmann A, et al. resection of colorectal liver metastases. World J Surg 1995; 19: 59–71. 31. Gayowski TJ, Iwatsuki S, Madariaga SR, et al. Experience in hepatic resection of colorectal cancer – analysis of clinical and pathological risk factors. Surgery 1994; 116: 703–71. 32. Fortner JG, Silva JS, Golbey RB, et al. Multivariate analysis of a personal series of 247 consecutive patients with liver metastases from colorectal cancer. Treatment by hepatic resection. Ann Surg 1984; 199: 306–16. 33. Ercolani G, Grazi GL, Ravaioli M, et al. Liver resection for multiple colorectal metastases : influence of parenchymal involvement and total tumor volume versus number or location on long term survival. Arch Surg 2002; 137: 1187–92. 34. www.LiverMetSurvey.org. Assessed: January 2009. 35. Elias D, Cavalcanti A, Sabourin JC, et al. Results of 136 curative hepatectomies with a safety margin of less than 10 mm for colorectal metastases. J Surg Oncol 1998; 69: 88–93. 36. Fong Y, Cohen AM, Fortner JG, et al. Liver resection for colorectal metastases. J Clin Oncol 1997; 15: 938–46. 37. Figueras J, Burdio F, Torras J, et al. Effect of subcentimeter nonpositive resection margin on hepatic recurrence in patients undergoing hepatectomy for colorectal liver metastases. Evidence from 663 liver resections. Ann Oncol 2007; 18: 1190–5. 38. Kokudo N, Miki Y, Sugai S, et al. Genetic and histological assessment of surgical margins in resected liver metastases from colorectal carcinoma: minimum surgical margins for successful resection. Arch Surg 2002; 137: 833–40. 39. de Hass RJ, Wicherts DA, Flores E, et al. R1 resection by necessity for colorectal liver metastases is it still a contraindication to surgery? Ann Surg 2008; 248: 626–36. 40. Elias D, Liberale G, Vernerey D, et al. Hepatic and extrahepatic colorectal metastases: when resectable their localization does not matter, but their total number has a prognostic effect. Ann Surg Oncol 2005; 12: 900–9. 41. Kangmatsu Y, Kato T, Hirai T, et al. Preoperative probability model for predicting survival after resection of pulmonary metastases from colorectal cancer. Brit J Surg 2004; 91: 112–20.

references
1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55: 10–30. 2. Steele G, Jr., Ravikumar TS. Resection of hepatic metastases from colorectal cancer. Biologic perspective. Ann Surg 1989; 210: 127–38. 3. Blumgart LH, Allison DJ. Resection and embolization in the management of secondary hepatic tumors. World J Surg 1982; 6: 32–45. 4. Jatzko G, Wette V, Muller M, et al. Simultaneous resection of colorectal carcinoma and synchronous liver metastases in a district hospital. Int J Colorct Dis 1991; 6: 111–14. 5. Wagner JS, Adson MA, Van Heerden JA, et al. The natural history of hepatic metastases from colorectal cancer a comparison with respective treatment. Ann Surg 1984; 199: 502–8. 6. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leukovorin for metastatic colorectal cancer. N Eng J Med 2004; 350: 233542. 7. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan refractory metastatic colorectal cancer. N Eng J Med 2004; 351: 33745. 8. Ambiru S, Miyazaki M, Isono T, et al. Hepatic resection for colorectal metastases—analysis of prognostic factors. Dis Colon Rectum 1999; 42: 632–9. 9. Yazaki M, Ito H, Nakagava K, et al. Aggressive surgical resection for hepatic metastases involving the inferior vena cava. Am J Surg 1999; 177: 294–8. 10. Kato K, Yasui K, Hirai T, et al. Therapeutic results for hepatic metastasis of colorectal cancer with special reference to effectiveness of hepatectomy: analysis of prognostic factors for 763 cases recorded in 18 institutions. Dis Colon Rectum 2003; 46: S22–31. 11. Mutsaerts EL, van RS, Zoetmulder FA, et al. Prognostic factors and evaluation of surgical management of hepatic metastases from colorectal origin: a 10 year single institute experience. J Gastrointest Surg 2005; 9: 178–86. 12. Wood CB, Gillis CR, Blumgart LH. A retrospective study of the natural history of patients with liver metastases from colorectal cancer. Clin Oncol 1976; 2: 285–8. 13. Wilson SM, Adson MA. Surgical treatment of hepatic metastases from colorectal cancer. Arch Surg 1976; 111: 330–4. 14. Scheele J, Strangl R, Altendorf-Hofmann A, Gall FP. Indicators of prognosis after hepatic resection for colorectal secondaries. Surgery 1991; 110: 13–29. 15. Mentha G, Huber O, Robert J, et al. Elective hepatic resection in the elderly. Br J Surg 1992; 79: 557–9. 16. Belghiti J, Hiramatsu K, Benoist S, et al. Seven hundred forty seven hepatectomies in the 1990s: an update to evaluate the actual risk of liver resection. J Am Col Surg 2000; 191: 38–46. 17. Vauthey JN, Pawlik TM, Abdalla EK, et al. Is extended hepatectomy for hepatobiliary malignancy justified? Ann Surg 2004; 239: 722–30. 18. Sica GT, Ji H, Ross PR. CT and MR imaging of hepatic metastases. AJR Am J Roentgenol 2000; 174: 691–8. 19. Scott DJ, Guthrie JA, Arnold P, et al. Dual phase helical CT versus portal venous phase CT for detection of colorectal liver metastases: correlation with intraoperative sonography, surgical and pathological findings. Clin Radiol 2001; 56: 235–42.

132

SURGERY FOR METASTATIC COLORECTAL CANCER
42. Elias D, Benizri E, Pocard M, et al. Treatment of synchronous peritoneal carcinomatosis and liver metastases from colorectal cancer. Eur J Surg Oncol 2006; 32: 632–6. 43. Carmignani CP, Ortega Perez G, Sugarbaker PH. The management of synchronous peritoneal carcinomatosis and hematogenous metastasis from colorectal cancer. Eur J Surg Oncol 2004; 30: 391–8. 44. Elias D, Saric J, Jaeck D, et al. Prospective study of microscopic lymphnode involvement of the hepatic pedicule during curative hepatectomy for colorectal liver metastases. Br J Surg 1996; 83: 942–5. 45. Jaeck D, Nakano H, Bachellier P, et al. Significance of hepatic pedicle lymph node involvement in patients with colorectal liver metastases: a prospective study. Ann Surg Oncol 2002; 9: 430–8. 46. Jaeck D. The significance of hepatic pedicle lymph node metastases in surgical management of colorectal liver metastases and of other liver malignancies. Ann Surg Oncol 2003; 10: 1007–11. 47. Adam R, de Haas RJ, Wicherts D, et al. Is hepatic resection justified after chemotherapy in patients with colorectal liver metastases and lymph node involvement? J Clin Oncol 2008; 26: 3672–80. 48. Nordlinguer B, Guiguet M, Vaillant JC, et al. Surgical resection of colorectal carcinoma metastases to liver. A prognostic scoring system to improve case selection, based on 1568 patients. Cancer 1996; 77: 1254–62. 49. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999; 230: 309–18. 50. Mala T, Bohler G, Mathisen O, et al. Hepatic resection for colorectal metastases: can preoperative scoring predict patient outcome. World J Surg 2002; 26: 1348–53. 51. Schusser Fiorenza CM, Mahvi DM, Niederhuber J, et al. Clinical risk score correlates with yield of PET scan in patients with colorectal hepatic liver metastases. J Gastrointest Surg 2004; 8: 150–7. 52. Colucci G, Maiello E, Leo S, et al. Biochemical modulation of fluorouracil with high dose metotrexate or folinic acid in advanced colorectal cancer patients. Anticancer Res 1994; 14: 2157–62. 53. Leichmann CG, Fleming TR, Muggia FM, et al. Phase II study of fluorouracil and its modulation in advanced colorectal cancer: a southewest oncology group study. J Clin Oncol 1995; 13: 1303–11. 54. Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Eng J Med 2000; 343: 905–14. 55. Douillard JY, Cunningham D, Roth AD, et al. Irinotecan combined with fluorouracil compared with fluorouracil alone as first line treatment for metastatic colorectal cancer: a multicentre randomised trial. Lancet 2000; 355: 1041–7. 56. Goldberg RM, Sargent DJ, Morton RF, et al. A randomised controlled trial of fluorouracil, plus leucovorin, irinotecan and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004; 22: 23–30. 57. De Gramon A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18: 2938–47. 58. Nordlinger B, Sorbye H, Collette L, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer: a randomosed controlled trial. Lancet 2008; 371: 1007–16. 59. Tanaka K, Adam R, Shimada H, et al. Role of neoadjuvant chemotherapy in the treatment of multiple colorectal metastases to the liver. Br J Surg 2003; 90: 963–9. 60. Ozawa K. Hepatic function and liver resection. J Gastroenterol Hepatol 1990; 5: 296–309. 61. Fornari F, Civardi G, Cavanna L, et al. Complications of ultrasonically guided fine-needle biopsy. Results of a multicenter Italian study and rewied of the literature. Scand J Gastroenterol 1989; 24: 949–55. 62. Smith EH. Complications of percutaneous abdominal fine needle biopsy. Radiology 1991; 178: 253–8. 63. Rogers MS, Collinson R, Desai S, et al. Risk of dissemination with biopsy of colorectal liver metastases. Dis Colon Rectum 2003; 46: 454–9. 64. Ohlsson B, Nilsson J, Stenram U, et al. Percutaneous fine needle aspiration cytology in the diagnosis and management of live tumors. Br J Surg 2002; 89: 757–62. 65. Thaler K, Kunneganti S, Khajanchee Y, et al. The evolving role of staging laparoscopi in the treatment of colorectal hepatic metastasis. Arch Surg 2005; 140: 727–34. 66. Rahusen FD, Cuesta MA, Borgstein PJ, et al. Selection of patients for resection of colorectal metastases to the liver using diagnostic laparoscopy and laparoscopic ultrasounography. Ann Surg 1999; 230: 31–7. 67. Grobmyer SR, Fong Y, D’Angelica M, et al. Diagnostic laparoscopy prior to planned hepatic resection for colorectal metastases. Arch Surg 2004; 139: 1326–30. 68. Zacherl J, Scheuba C, Imhof M, et al. Current value of intraoperative sonography during surgery for hepatic neoplasms. World J Surg 2002; 26: 550–4. 69. Staren ED, Gambla M, Deziel DJ, et al. Intraoperative ultrasdound in the management of liver neoplasms. Am Surg 1997; 63: 591–6. 70. Machi J, Sigel B, Zaren HA, et al. Operative ultrasound during hepatobiliary and pancreatic surgery. World J Surgery 1993; 17: 640–5. 71. Portier G, Elias D, Bouche O, et al. Multicenter randomised trial of adjuvant fluorouracil and folonic acid compared with surgery alone after resection of colorectalliver metastases. J Clin Oncol 2006; 24: 4976–82. 72. Parks R, Gonen M, Kemeny N, et al. Adjuvant chemotherapy improves survival after resection of hepatic colorectal metastases: analysis of data from two continenets. J Am Coll Surg 2007; 204: 753–61. 73. Kusunoki M, Yanagi H, Noda M, et al. Results of pharmacokinetic modulating chemotherapy in combination with hepatic arterial 5-FU infusion and oral UFT after resection of hepatic colorectal metastases. Cancer 2000; 89: 128–35. 74. Kemeny MM, Adak S, Gray B, et al. Combined modality treatment for resectable metastatic colorectal carcinoma to the liver: surgical resection of the hepatic metastases in combination with continuous infusion of chemotherapy—an intergroup study. J Clin Oncol 2002; 20: 1499–1505. 75. Kemeny NE, Gonen M. Jarnagin WR, et al. Hepatic arterial infusion after liver resection. N Eng J Med 2005; 352: 734–5. 76. Wei AC, Greig PD, Grant D, et al. Survival after hepatic resection for colorectal metastases: a ten year experience. Ann Surg Oncol 2006; 13: 668–76. 77. Hughes KS, Rosenstein RB, Songhoradobi S, et al. Resection of the liver for colorectal carcinoma metastases – a multi institutional study of long term survivors. Dis Colon Rectum 1988; 31: 1–4. 78. Nordlinger B, Jaeck D, Guiget M, et al. Surgical resection of hepatic metastases: multicentric retrospective study by the French Association of Surgery. In: Nordlinger B, Jaeck D, eds. Treatment of Hepatic Metastases of Colorectal Cancer. Paris: Springer Verlag, 1992; 129–61. 79. Choti MA, Sitzmann JV, Tiburi MF, et al. Trends in long term survival following liver resection for hepatic colorectal metastases. Ann Surg 2002; 235: 759–66. 80. Minagawa M, Makuuchi M, Torzilli G, et al. Extension of the frontiers of surgical indications in the treatment of liver metastases from colorectal cancer: long term results. Ann Surg 2000; 231: 487–99. 81. Bismuth H, Adam R, Levi F, et al. Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. Ann Surg 1996; 224: 509–22. 82. Adam R, Delvart V, Pascal G, et al. Rescue surgery for unresectable colorectal metastases downstaged by chemotherapy. Ann Surg 2004; 240: 644–58. 83. Miyanari N, Mori T, Takahashi K, et al. Evaluation of aggressively treated patients with unresectable multiple liver metastases from colorectal cancer. Dis Colon Rectum 2002; 45: 1503–9. 84. Tumor progression while on chemotherapy: a contraindication to liver resection for multiple colorectal metastases? Ann Surg 2004; 240: 1052–61. 85. Folprecht G, Grothey A, Alberts S, et al. Neoadjuvant treatment of unresectable colorectal liver metastases: Correlation between tumor response and resection rates. Ann Oncol 2005; 16: 1311–19. 86. Tabernero JM, Van Cutsem E, Sastre J, et al. An international phase II study of cetuximab in combination with oxaliplatin/5-fluorouracil/ folinic acid (Folfox-4) in the first line treatment of patients with metastatic cancer expressing epidermal growth factor receptor (EGFR). Proc Am Soc Clin Oncol 2004; 23: 248.

133

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
87. Cervantes A. Cetuximab plus oxaliplatin/5-fluorouracil/folinic acid for the epidermal growth factor receptor expressing metastatic colorectal cancer in the first line setting: A phase II study. Eur J Cancer 2005; 3: 181–2. 88. Bokemeyer C, Staroslawsca E, Makhson A, et al. Cetuximab plus 5-FU/ FA/Oxaliplatin (FOLFOX 4) in the first line treatment of colorectal metastatic cancer: a large scale phase II study. OPUS. Eur J Cancer 2007; 5: 4. 89. Van Custem E, Bodoky G, Kyung Roh J, et al. CRYSTAL, a randomised phase III trial of cetuximab plus Folfiri vs Folfiri in first line treatment of metastatic colorectal cancer. Eur J Cancer 2007; 5: 4. 90. Van Custem E, Lang I, D’haens G, et al. KRAS status and efficacy in the first line treatment of patients with metastatic colorectal cancer treated with Folfiri with and without Cetuximab: the CRYSTAL experience. J Clin Oncol 2008; 26: (Abstr 2). 91. Bokemeyer C, Bondarenko I, Hartmann JT, et al. KRAS status and efficacy in the first line treatment of patients with metastatic colorectal cancer treated with Folfox with and without Cetuximab: the OPUS experience. J Clin Oncol 2008; 26: (Abstr 4000). 92. Clavien PA, Selzner N, Morse M, et al. Dowstaging of hepatocellular carcinoma and liver metastases from colorectal cancer by selective intraarterial chemotherapy. Surgery 2002; 131: 433–42. 93. Leonard GD, Fong Y, Jarnagin W, et al. Liver resection after hepatic arterial infusion plus sytemic oxaliplatin combinatins in pre treated patients with extensive unresectable colorectal liver metastases. J Cin Oncol 2004; 22: 3542–45. 94. Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection ablation for colorectal liver metastases. Ann Surg 2004; 239: 818–25. 95. Curley SA, Izzo F, Delrio P, et al. Radiofrequency ablation of unresectable primary and metastatic hepatic malignancies: results of 123 patients. Ann Surg 1999; 230: 1–8. 96. Adam R, Akpinar E, Johhan M, et al. Place of cryosurgery in the treatment of malignant liver tumors. Ann Surg 1997; 225: 39–50. 97. Brooks AJ, Wang F, Alfredson M, et al. syncronous liver resection and cryotherapy for colorectal metastases: survival analysis. Surgeon 2005; 3: 265–8. 98. Kokudo N, Makuuchi M. Current role of portal vein embolization/ hepatic artery embolization. Surg Clin North Am 2004; 84: 643–57. 99. Azoulay D, Castaing D, Ismail A, et al. Resection of nonresectable liver metastases from colorectal cancer after percutaneous portal vein embolization. Ann Surg 2000; 231: 480–6. 100. Farges O, Belghiti J, Kianmanesh R, et al. Portal vein embolization before right hepatectomy: Prospective clinical trial. Ann Surg 2003; 237: 208–17. 101. Abdalla EK, Denys A, Chevalier P, et al. Total and segmental liver volume variations: Implications for liver surgery. Surgery 2004; 135: 404–10. 102. Nagino M, Nimura Y, Kamiya J, et al. Right or left trisegment portal vein embolization before hepatic trisegmentectomy for hilar bile duct carcinoma. Surgery 1995; 117: 677–81. 103. Adam R, Laurent A, Azoulay D, et al. Two stage hepatectomy: a planed strategy to treat irresectable liver tumors. Ann Surg 2000; 232: 777–85. 104. Jaeck D, Oussoultzouglou E, Rosso E, et al. A two stage hepatectomy procedure combined with portal vein embolization to achieve curative resection for initially unresectable multiple and bilobar colorectal liver metastases. Ann Surg 2004; 240: 1037–51. 105. Adam R, Vinet E. Regional treatment of metastasis: surgery of colorectal liver metastases. Ann Oncol 2004; 15: 103–6. 106. Togo S, Nagano Y, Masui H, et al. Two stage hepatectomy for multiple bilobar liver metastases from colorectal cancer. Hepatogastroenterology 2005; 52: 913–19. 107. Jaeck D, Bachellier P, Nakano H, et al. One or two-stage hepatectomy combined with portal vein embolization for initially nonresectable colorectal liver metastases. Am J Surg 2003; 185: 221–9. 108. Azoulay D, Eshkenazy R, Andreani P, et al. In situ hypothermic perfusion of the liver versus standard total vascular exclusion for complex liver resection. Ann Surg 2005; 241: 277–86. 109. Adam R, Bismuth H, Castaing D, et al. Repeat hepatectomy for colorectal liver metastases. Ann Surg 1997; 225: 51–60. 110. Stone MD, Cady B, Jenkins RL, et al. Surgical therapy for recurrent liver metastases from colorectal cancer. Arch Surg 1990; 125: 718–21. 111. Petrowsky H, Gonen M, Jarnagin W, et al. Second liver resections are safe and effective treatment for recurrent hepatic metastases from colorectal cancer: a bi-institutional analysis. Ann Surg 2002; 235: 863–71. 112. Adam R, Pascal G, Azoulay D, et al. Liver resection for colorectal metastases: the third hepatectomy. Ann Surg 2003; 235: 863–71. 113. Pessaux P, Lermite E, Brehant O, et al. Repeat hepatectomy for recurrent colorectal liver metastases. J Surg Oncol 2006; 93: 1–7. 114. Elias D, Youssef O, Sideris L, et al. evolution of missing colorectal liver metastases following inductive chemotherapy and hepatectomy. J Surg Oncol 2004; 86: 4–9. 115. Benoist S, Brouquet A, Penna C, et al. Complete response of colorectal liver metastases after chemotherapy: does it mean cure? J Clin Oncol 2006; 24: 3939–45. 116. Adam R, Wicherts DA, de Haas R, et al. Complete pathological response after preoperative chemotherapy for colorectal liver metastases: myth or reality? J Clin Oncol 2008; 26: 1635–41. 117. Rubbia Brandt L, Giostra E, Brezault C, et al. Importance of histological tumor response assessment in predicting the outcome in patients with colorectal liver metastases treated with neoadjuvant chemotherapy followed by liver surgery. Ann Oncol 2007; 18: 299–304. 118. Parikh AA, Gentner B, Wu TT, et al. Perioperative complications in patients undergoing major liver resection with and without neoadjuvant chemotherapy. J Gastrointest Surg 2003; 7: 1082–8. 119. Adam R, Pascal G, Castaign D, et al. Tumor progression while on chemotherapy: a contraindication to liver resection for multiple colorectal metastases? Ann Surg 2004; 240: 1052–61. 120. Nordlinger B, Guiet M, Vaillant JC, et al. Surgical resection of colorectal carcinoma metastases to the liver. A prognostic scoring system to improve case selection, based on 1568 patients. Cancer 1996; 77: 1254–62. 121. Bolton J, Fuhram GM. Survival after resection of multiple bilobar hepatic metastases from colorectal carcinoma. Ann Surg 2000; 231: 743–51. 122. Reddy SK, Pawlik TM, Zorzi D, et al. Simultaneous resection of colorectal cancer and synchronous liver metastases: a multi-institutional analysis. Ann Surg Oncol 2007; 14: 3295–6. 123. Thelen A, Jonas S, Benckert C, et al. Simultaneous versus staged liver resection of synchronous liver metastases from colorectal cancer. Int J Colorectal Dis 2007; 22: 1269–76. 124. Tanaka K, Shimada H, Matsuo K, et al. Outcome after simultaneous colorectal and hepatic resection for colorectal cancer with synchronous metastases. Surgery 2004; 136: 650–9. 125. Weber JC, Bachellier P, Oussoultzoglou E et al. Simultaneous resection of colorectal primary tumor and synchronous liver metastases. Br J Surg 2003; 90: 956–62. 126. Mentha G, Majno P, Roth A. Neoadjuvant chemotherapy and resection of advanced synchronous liver metastases before treatment of the colorectal primary. Br J Surg 2006; 93: 872–8.

134

14

Chemotherapy for metastatic colorectal cancer Derek G. Power and Nancy E. Kemeny
schedule as randomized data has shown the superiority of this regimen compared with other 5-FU/LV schedules (16). Over the last few years, three new chemotherapeutic agents—irinotecan, oxaliplatin, and capecitabine (an oral version of 5-FU)—have been approved for the treatment of metastatic CRC. Irinotecan is a topoisomerase inhibitor and activity in the metastatic setting was established in randomized studies comparing irinotecan with best supportive care. In patients who progressed on fluorouracil therapy, one-year survival rates were increased from 14% to 36% with the use of single agent irinotecan (17). Combinations of irinotecan and 5-FU/ LV were then studied in the first-line setting. A randomized trial of irinotecan added to infusional 5-FU/LV compared to 5-FU/LV alone demonstrated an increased response rate (35% vs. 22%, respectively, p = 0.005) and a survival benefit of 3 months (17 vs. 14 months, respectively, p = 0.031). Grades 3 to 4 toxicities were more common in the irinotecan group, e.g. diarrhea (44% vs. 26%) and neutropenia (29% vs. 2%) (18). A phase III study of 683 patients compared weekly irinotecan and bolus 5-FU/LV (IFL) to 5-FU/LV alone. The IFL regimen increased response rate (39% vs. 21%, p < 0.001) and survival (14.8 vs. 12.6 months, respectively, p = 0.04) (19). The FOLFIRI regimen, that is, irinotecan combined with the deGramont 5-FU/LV combination, has been shown to be safe and efficacious in the first-line setting and is now accepted as the optimal way to combine irinotecan and FU/ LV. Response rates approaching 40% and median overall survival of 17 to 23 months have been reported (20,21). The randomized BICC-C trial (before the addition of bevacizumab) reported a median OS of 23.1 months for FOLFIRI vs. 17.9 months for mIFL and 18.9 months for CapIRI (capecitabine and irinotecan) with response rates of 47%, 42%, and 39%, respectively (21). Oxaliplatin is a platinum derivative and works by alkylating DNA. As a single agent, oxaliplatin is not superior to LV5FU-2 and has limited activity in advanced CRC (22,23). In the firstline setting, the FOLFOX regimen, that is, combination oxaliplatin and LV5FU-2 given as the deGramont schedule, was shown to be safe, efficacious, and superior to LV5FU-2 with response rates of 50% and median overall survival of 16 months (24). FOLFIRI compared with FOLFOX showed response rates of 56% and 54%, respectively, and no difference in median overall survival 20.6 versus 21.5 months (p = NS) (25). A phase III trial by Colucci and colleagues comparing FOLFIRI and FOLFOX4 also showed essentially equal efficacy in terms of response rate, time to progression, and overall survival (26). The intergroup trial showed that patients receiving FOLFOX had a median survival of 19.5 months compared to 17.4 months for irinotecan plus oxaliplatin (IROX) or 15 months for IFL (p = 0.001) (27,28). Efficacy of the FOLFOX

introduction
Colorectal cancer (CRC) is a major cause of cancer-related mortality worldwide and in Western countries, and it is the second most frequent cause of cancer-related death (1). The United States has the highest annual incidence of invasive CRC, and in 2008 an estimated 148,810 cases of CRC were diagnosed and 49,960 people died from the disease (2). At diagnosis, 20% to 25% of all patients will have synchronous liver metastases and at least another 60% of patients who develop metastatic disease will have metachronous liver metastases (3,4). The liver is the only metastatic site in about onethird of patients and this can be explained by the portal venous drainage of the colon and rectum to liver (5). Overall liver metastases are seen in approximately 20% to 70% of patients with CRC and lung metastases are seen in 10% to 20%. (6) For many years, the approach to patients with metastatic CRC was minimalist with fluorouracil-based chemotherapy being the only palliative option and median survival rarely exceeding one year (7). Surgical developments over the last 10 years in resection of liver metastases have resulted in improved long-term survival. Several large surgical series have shown 5-year survival rates averaging 30% to 40%, and in some patients a chance of cure, with 20% survival at 10 years after hepatic resection (8–10). Developments in chemotherapy, both systemic and regional, have resulted in a radically changed landscape for patients with metastatic CRC. Those patients who present with initially unresectable liver metastases, 80% to 85% of cases, may now have the chance of hepatic resection after chemotherapy downstaging. Even if liver resection is not possible, median survival has increased with modern chemotherapies (3,11). This review will focus on developments in chemotherapy and biologic therapy for the treatment of metastatic CRC and highlight how a true multidisciplinary approach has resulted in improved survival for this common disease.

systemic chemotherapy for unresectable liver disease (first and second line)
The fluorinated pyrimidine antimetabolites have been the cornerstone of systemic treatment for CRC for over 50 years. Fluorouracil (5-FU) was the only chemotherapeutic agent available for nearly 35 years. Response rates with bolus 5-FU were 10% to 20% with median survival around 10 to 12 months (12). Modifications of the 5-FU dosing schedule were studied and it was found that a protracted infusion of the drug increased response rates to 20% to 30% and median survival to 12 to 14 months (13,14). The addition of the biomodulator folinic acid [leucovorin (LV)] to 5-FU similarly increased response rates and median overall survival (14,15), and it has now become standard to combine 5-FU bolus plus 48-hour infusion with bolus LV (deGramont – LV5FU-2) in a bimonthly

135

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 14.1 Metastatic CRC Second-Line Regimens
Study Rothenberg (22) Rothenberg (32) Tournigand (25) Souglakos (33) Park (137) First-line IFL FOLFIRI FOLFIRI Oxaliplatin Irinotecan Second-line FOLFOX4 FOLFOX4 XELOX FOLFOX6 FOLFOX6 FOLFIRI FOLFOX RR (%) 9.9 12.4 15.3 15 FOLFIRI 18 15 mTTP 4.6 4.8 4.7 4.2 4 7.5 2 mOS 12.6 11.9 2.5 14 5

RR, response rate; mTTP, median time to progression (in months). mOS, median overall survival (in months).

regimen was also shown in the randomized TREE1 study, which compared mFOLFOX6 to bolus FU/LV/oxaliplatin and to capecitabine/oxaliplatin. Response rates were 41%, 20%, and 27%, respectively, and median OS was 19.2, 17.9, and 17.2 months, respectively (29). Overall, the FOLFOX and FOLFIRI regimens have improved response rate, time to progression, and overall survival compared to 5-FU/LV (30). Replacing 5FU/LV with capecitabine and combining with oxaliplatin (XELOX) has been shown to be noninterior to FOLFOX and thus is a third alternative for first-line treatment (31). The combination of capecitabine and irinotecan, however, is not well tolerated and is associated with high rates of severe vomiting and diarrhea. Therefore the bolus/infusional schedule of FU, LV5FU-2 is the preferred mode of administration in combination with irinotecan (21). In the second-line setting, after treatment failure with oxaliplatin or irinotecan-based regimens, results are less impressive (Table 14.1). Response rates of up to 18% and median overall survival of 6 to 14 months have been reported (32,33). It is noteworthy that in those patients who are 5-FU refractory, there is no difference in outcome if second-line therapy begins with either FOLFOX or irinotecan. In a phase III study of 491 5-FU-resistant patients with mCRC, median overall survival with second line FOLFOX was 13.8 versus 14.3 months for irinotecan alone (p = 0.38) (34).

Catheter in artery

Codman 3000 Pump

regional chemotherapy in unresectable liver disease
The rationale for hepatic arterial infusion (HAI) of chemotherapy is that the hepatic metastases receive their blood supply from the hepatic artery and the normal liver parenchyma is fed by the portal vein (35). The development of an implantable pump allowing continuous infusion of chemotherapy and long-term patency of the catheter and hepatic artery made the development of HAI therapy possible (Fig. 14.1). Floxuridine (FUDR) is the ideal drug for use via HAI as it has a high hepatic extraction, a short half life, and a steep dose–response curve. These properties give FUDR a 400-fold advantage when given via HAI (36,37). Ten randomized phase III trials have compared HAI FUDR with systemic FUDR or FU/LV in patients with unresectable CRC liver metastases. All of these trials showed superior response rates with HAI administration (42–62%) compared to systemic (9–21%) (38). Overall survival has been difficult to prove in many trials due to crossover design of small numbers in

Figure 14.1 Implantable HAI pump.

the study, initial extrahepatic disease in some studies, and the fact that HAI was not used in all cases though the patients are included in the survival data. The CALGB 9481 study compared HAI FUDR + Dexamethasone (Dex) with systemic intravenous (IV) FU/LV and did not have a crossover (39). Dexamethasone was added to the FUDR in the pump as it had previously been shown to decrease FUDR toxicity and increase efficacy (40). There was a significant increase in overall survival in the HAI FUDR Dex arm versus the systemic FU/LV arm (24.4 vs. 20 months, respectively, p = 0.0034). Quality of life assessment showed that the HAI arm experienced significantly better physical functioning compared with the systemic arm. The 51% 2-year survival compared favorably with systemic combinations of Oxaliplatin/5FU/LV (25), or irinotecan/5FU/LV (41). The use of HAI alone, however, in a new meta-analysis using the old flawed studies did not show an increased survival (38). With new systemic therapies, it is more appropriate to think of a combination of these regimens with HAI FUDR/Dex.

136

CHEMOTHERAPY FOR METASTATIC COLORECTAL CANCER
As over one-third of all patients with metastatic CRC will have liver-only disease, the combination of HAI FUDR to treat the liver disease and modern systemic chemotherapy to control potential extrahepatic micrometastases may show superior results as has been seen already in small studies. In a phase I study of 46 patients previously treated with systemic chemotherapy, HAI FUDR/Dex in combination with systemic irinotecan produced response rates of 74% and a median overall survival of 20 months (42). In an updated series of 49 patients with unresectable liver metastases treated with HAI FUDR/ Dex plus systemic oxaliplatin/irinotecan, 53% of whom were previously treated with systemic chemotherapy, Kemeny and colleagues report a 92% response rate and a median overall survival of 50.8 months and 35 months for chemotherapy in naive and pretreated patients, respectively. The resection rate in this study was 47%, in a population that was definitely unresectable at baseline (43) [see section “Converting unresectable liver disease to resection…” for further discussion on resection of liver metastases and for comment on using HAI with other chemotherapy besides FUDR, e.g., oxaliplatin]. The epidermal growth factor receptor (ERBB, EGFR) family compromises four molecules: EGFR, HER2, HER3, and HER4. EGFR is overexpressed in up to 70% of human CRCs and has been associated with advanced stage disease (50). EGFR activation mediates multiple cell-signaling pathways including PI3K/ AKT/mTOR and Ras/Raf/MEK/ERK resulting in resistance to apoptosis, proliferation, angiogenesis, and metastases. Two monoclonal antibodies that target the EGFR have been approved for the treatment of metastatic CRC. Panitumumab (Pmab) is a fully humanized IgG2 molecule, while cetuximab (Cmab) is an IgG1 chimeric molecule. A significant increase in response rate (22.9% vs. 10.8%) and time to progression (4.1 vs. 1.5 months), but not overall survival (8.6 vs. 6.9 months; p = 0.48) has been reported with Cmab in combination with irinotecan versus Cmab alone in patients refractory to irinotecan or oxaliplatin-based chemotherapy (51,52). The MABEL study of 1147 patients confirmed the results of the earlier Cmab studies. In patients who had progressed on previous irinotecan-containing regimens, the progression-free survival rate at 12 weeks was 61% with the irinotecan and Cmab combination and median overall survival was 9.2 months (53). Another secondline study, EPIC (Erbitux Plus Irinotecan in Colorectal Cancer), compared Cmab plus irinotecan with irinotecan alone in patients who had progressed on previous oxaliplatin-containing regimens. A significant improvement in PFS and response rate was found with the combination (3.98 vs. 2.56 months, p < 0.001; 16% vs. 4%, p < 0.001, respectively) with no difference in overall survival (10.7 vs. 10 months, p = 0.812) (54). The National Cancer Institute of Canada (NCIC) evaluated the effect of third-line Cmab in metastatic CRC patients who had previously received FU, irinotecan, and/or oxaliplatin. Compared to best supportive care, Cmab improved median overall survival from 4.6 months to 6.1 months (p = 0.005). (55) Results from these studies demonstrate that Cmab has activity as monotherapy, but is more effective when combined with irinotecan. This is likely due to modulation of irinotecan resistance by Cmab, which has been shown in preclinical work (56). Pmab was approved on the basis of an open label randomized phase III trial comparing Pmab with best supportive care in patients who had progressed on previous chemotherapies. Pmab significantly increased PFS (13.8 vs. 8.5 weeks, p = 0.001), but not overall survival (57). In the first-line setting, both Cmab and Pmab have shown activity as well. The phase III CRYSTAL trial of 1,217 patients compared FOLFIRI plus Cmab to FOLFIRI alone. Progression-free survival, the primary end-point, was significantly greater with the combination (8.9 vs. 8 months, p = 0.0479, respectively). Also there was a difference in response rate (47% vs. 39%, p = 0.0038) and median overall survival (19.9 vs. 18.6 months, p = 0.30) [intention to treat data] (58). The OPUS study randomized first-line FOLFOX4 plus Cmab with FOLFOX4 alone and showed a 10% increased response rate with the combination (46% vs. 36%, p = 0.084) and no change in median PFS (7.2 vs. 7.2 months) [intention to treat data] (59). Preliminary data from the phase III CALGB 80203 trial showed a higher overall response rate for FOLFOX/FOLFIRI plus Cmab versus chemotherapy alone (49% vs. 33%, p = 0.014) (60). There is less experience with Pmab, likely due to

systemic chemobiologic therapy for unresectable liver disease
Our increasing understanding of molecular pathways in carcinogenesis has led to the development of novel targeted therapy. Vascular endothelial growth factor (VEGF) plays a crucial role in physiologic and pathologic angiogenesis. Preclinical data with bevacizumab, a humanized monoclonal antibody against VEGF, showed inhibition of growth of human tumor xenografts as a result of inhibition of tumor angiogenesis (44) and improved delivery of chemotherapy to the tumor by altering tumor vasculature and decreasing elevated interstitial pressure in tumors (45). In a randomized phase III trial, the addition of bevacizumab (bev) to IFL versus IFL alone resulted in increased response rates, progression-free and overall survival (20.3 vs. 15.3 months, respectively, p < 0.001) (46). The results were not as promising in a randomized phase III study of 1400 patients where the addition of bevacizumab to both FOLFOX4 and XELOX in the first-line setting improved progression-free survival (9.4 vs. 8.0 months for the bev and placebo groups, respectively, p = 0.0023), but not response rates (47% vs. 45%) or median overall survival (21.3 and 19.9 months) in the bevacizumab and placebo groups, respectively (p = 0.077). (47) The TREE and BICC-C studies showed an increased response rate and overall survival when bevacizumab was added to oxaliplatin- and irinotecan-based regimens. However, these studies were sequential and not randomized (29,48). There has been no trials comparing FOLFOX/bevacizumab with FOLFIRI/bevacizumab. However, based on the fact that FOLFOX and FOLFIRI have virtually identical activity in the first-line setting, the addition of bevacizumab to either regimen is reasonable. Bevacizumab has also shown activity in the second-line setting. The European Cooperative Oncology Group (ECOG) showed that patients treated with FOLFOX4 and bevacizumab after progression on irinotecan and fluoropyrimidines had improved survival compared with FOLFOX4 alone (12.9 vs. 10.8 months, p = 0.0011). (49)

137

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
its more recent approval. In a phase II study in the first-line setting, an objective response rate of 47% and a median survival of 16.8 months was reported when Pmab was combined with irinotecan-based regimens (61). Studies combing Pmab with oxaliplatin in the first- and second-line settings are ongoing (62). There are no data direct comparisons between Cmab and Pmab and the decision to use one over the other may well come down to physician preference or decreased rate of hypersensitivity reactions seen with Pmab (63,64). Many of the above trials with EGFR antibodies accrued patients who had EGFR-expressing disease. It has since emerged that there is a lack of correlation between EGFR expression based on immunohistochemistry, gene expression, or gene copy number and response to Cmab and Pmab (65–67). More recently, the predictive value of KRAS mutations downstream of the EGFR has helped to define a subset of patients more likely to respond to EGFR monoclonal antibodies and possibly explains why the overall efficacy to EGFR inhibition has been so poor. Retrospective analyses of many of the trials discussed above show that response rates to EGFR antibodies in patients who are KRAS mutant is very low (68–73). As a result the drug licensing body in Europe (EMEA) and the American Society of Clinical Oncology (ASCO) has restricted the use of EGFR antibodies to KRAS wild-type patients in the treatment of metastatic CRC. The CRYSTAL trial retrospectively performed KRAS analysis and reported an increased response rate (59% vs. 43%, p = 0.0025) and an overall survival benefit in KRAS wild-type patients (24.9 vs. 21.0 months, p = 0.22) for the Cmab and chemotherapy versus chemotherapy alone groups, respectively (58,74,75). The OPUS study also reported the effect of EGFR inhibition in the KRAS wildtype (wt) population. The addition of Cmab to FOLFOX in wild type patients increased response rate (61% vs. 37%, p = 0.11) and median PFS from 7.2 to 7.7 months (p = 0.02) compared to FOLFOX alone. In the mutant KRAS population median PFS decreased from 8.6 months to 5.5 months with the addition of Cmab to FOLFOX (p = 0.02), suggesting a detrimental effect with the addition of Cmab to FOLFOX in KRAS mutant patients (59). The benefit of Cmab even in the select KRASwt population, however, is modest with an overall survival benefit of 3.9 months (in KRAS wild type tumors, HR = 0.84 [95% CI: 0.64–1.11]), as reported in the CRYSTAL trial and an improvement in PFS of 1.2 months, or approximately 37 days. Genetic and biochemical evidence indicates that BRAF is the principal downstream effector of KRAS and recent data has shown that the BRAF mutation V600E (present in approximately 10% of CRCs, thus leaving at least 40% of nonresponsive patients with no mutations in EGFR or BRAF associated with resistance to EGFR antibody therapy) (76). Intact expression of PTEN and expression levels of EGFR ligands (amphiregulin, epiregulin) may also play a role in identifying those who will benefit from anti-EGFR therapies (71,77). As data emerged on activity with biologic agents, combinations of biologics with chemotherapy was investigated. The phase II BOND-2 study showed that adding bevacizumab and Cmab to irinotecan in patients who were irinotecan refractory suggested a benefit for the two antibodies versus one (overall survival 14.5 vs. 11.4 months, respectively) (78). The randomized phase III PACCE study investigated the addition of Pmab to a combination of bevacizumab and chemotherapy (either oxaliplatin or irinotecan-based regimens) in the first-line setting demonstrated a decreased PFS for the Pmab/bevacizumab plus oxaliplatin versus bevacizumab/chemotherapy combination (10 vs. 11.4 months, respectively, p = 0.044). In the irinotecan-based chemotherapy group, median PFS was 10.1 months for those the received two antibodies versus 11.7 months for those receiving one (79). A trend toward worse OS was observed with Pmab/Bev plus systemic chemotherapy vs Bev plus chemotherapy in the KRAS wt group, 20.7 versus 24.5 months, respectively. Additional toxicity was also seen in the Pmab/ bevacizumab/chemotherapy group. The CAIRO-2 study compared the combination of Cmab with bevacizumab and XELOX in the first-line treatment and showed that the Cmab/ bevacizumab/XELOX resulted in a decreased median PFS compared with XELOX/bevacizumab (9.4 vs. 10.7 months, p = 0.018) (80). KRAS analysis was performed retrospectively and when compared to patients with KRAS mutations in the chemotherapy/bevacizumab group, cetuximab-treated patients with KRAS mutated tumors had significantly shorter PFS (8.1 vs. 12.5 months, p = 0.003) and OS (17.2 vs. 24.9 months, p = 0.03). In those patients with KRAS wild-type tumors, there was no difference in either PFS or OS with the addition of cetuximab. Thus the addition of VEGF/EGFR antibody combinations to chemotherapy suggests a lack of benefit. The SWOG/ CALGB 80405 trial looking at FOLFIRI or FOLFOX in combination with Cmab or bevacizumab or both may help to answer this question (60). To date, therefore, dual biologic therapy with bevacizumab and an anti-EGFR antibody should not be used in combination with chemotherapy in the first-line treatment of metastatic CRC outside of a clinical trial.

converting unresectable liver disease to resection
There is increasing literature supporting the use of modern systemic chemotherapy, as described above, to decrease the size and extent of liver disease, thus rendering previously unresectable metastases resectable (81–83). While this is not neoadjuvant therapy in the strictest sense of the word, the end point for patients with initially unresectable liver metastases from CRC should hopefully be hepatic resection. It has been shown that in nonresectable liver metastases, resection rate correlates with response (84). In the largest study to date, Adam and colleagues report an experience over 11 years in 1439 patients, of which 1104 had unresectable liver disease at presentation. Chemotherapy consisted of FOLFOX (70%), FOLFIRI (7%), or both (4%) and treatment was for an average of 10 courses. Hepatic resection was possible in 138 patients (12.5%) and the 5- and 10-year survival rates were 33% and 23%, respectively, which compares favorably to 335 patients who were resectable from the start and had 5- and 10-year survival rates of 48% and 30%, respectively (p = 0.01) (85). This study highlights the fact that modern chemotherapy can convert unresectable liver metastases to resection with good 5-year survival rates. Several other studies have looked at combinations/comparisons of FOLFOX, FOLFIRI, and FOLFOXIRI

138

CHEMOTHERAPY FOR METASTATIC COLORECTAL CANCER
Table 14.2 Neoadjuvant Systemic Chemotherapy for Unresectable Liver Metastases
Study Giacchetti (83) Regimen 5FU/LV/ Oxaliplatin 5FU/LV/ Oxaliplatin FOLFIRI FOLFOX4 FOLFOXIRI Resectability rate (%) 38 OS 5 years, 50% Med, 48m 5years, 54% All alive 19m f/u Med, 26m 4 years, 37% Med, 37m

Bismuth (82) Pozzo (138) Alberts (139) Masai (140)

16 32.5 33.3 26

m, months; Med, median; OS, overall survival; f/u, follow-up.

(oxaliplatin, irinotecan and bolus/infusional 5FU/LV) in patients with initially unresectable (or not optimally resectable) disease and report increasing rates of R0 hepatic resections and overall survival (Table 14.2). In a phase III study of 244 patients, conducted by the Gruppo Onclogico Nord-Ovest (GONO) group, FOLFOXIRI was compared with FOLFIRI. The rate of R0 hepatic resections was 36% for the triplet compared with 12% for the doublet (p = 0.017) (86). This group recently updated the long-term outcome of 196 patients with initially unresectable mCRC treated with FOLFOXIRI in two phase II and one phase I trials. The overall R0 resection rate was 19% and at 5 years, 29% of patients are free of disease (87). Another phase III study of 283 patients showed that the addition of oxaliplatin to FOLFIRI increased the resection rate of lung and liver metastases from 4% to 10%. Of those who underwent surgery after FOLFOXIRI, 86% had an R0 resection (88). Subset analysis of the oxaliplatin stop-go OPTIMOX-1 study showed that FOLFOX4 was superior to FOLFOX7 in terms of overall survival after an R0/R1 resection (51 vs. 38 months, respectively) (89). HAI combined with systemic therapy in nonrandomized studies has demonstrated high response and resection rates. In a retrospective series examining HAI FUDR Dex in patients who had all received prior oxaliplatin/5FU/LV and some had prior irinotecan as well, the response rate for 39 patients was 44%, and median OS from the time of initiation of HAI was 20.1 months, while it was 32 months from the initiation of treatment of their metastatic disease. Eighteen percent of patients proceeded to surgical resection or ablation (90). HAI FUDR/Dex combined with oxaliplatin and irinotecan based regimens produced resectability rates of up to 47% in patients who were definitely unresectable at presentation and 53% had received prior systemic therapy. The median survival for all patients was 41 months. Survival for the chemotherapy naive group was 50 months while it was 38 months for those previously treated (43). The benefit of HAI therapy given in a “neoadjuvant” setting has also been highlighted by Auer and colleagues. Radiologic complete response of liver metastases in patients treated with HAI FUDR was more likely to represent a true CR when

compared to systemic neoadjuvant chemotherapy (68% vs. 29%), and the liver recurrence rate was 14% in the HAI group versus 42% in the preoperative systemic chemotherapy group (p < 0.001) (91,92). The benefit of HAI in the “neoadjuvant” setting has also been reported with the use of other drugs besides FUDR. In a phase II study, Ducreux and colleagues reported the efficacy and relative safety of HAI Oxaliplatin plus 5FU/LV in 26 patients with initially unresectable liver metastases. Median overall survival and median disease free survival was 27 months and 27 months, respectively (93). The intention to treat response rate was 64% and 5 patients proceeded to R0 liver resection. Recent work by Boige and colleagues used HAI Oxaliplatin combined with systemic FU/LV after systemic failure with either FOLFOX or FOLFIRI or both (94). Median PFS and overall survival were 7 months and 16 months, respectively, and 7 out of 39 patients previously deemed unresectable were able to undergo an R0 liver resection. The use of preoperative HAI Oxaliplatin has also been reported by Elias and colleagues to be significantly associated with a true pathologic complete response even when “missing” metastases are left in place at hepatectomy (95). These studies suggest that regional therapy may produce a higher rate of true cured lesions than systemic therapy as described in the Benoist study where persistent macroscopic or microscopic residual disease or early recurrence in situ was observed in 83% of liver metastases having a complete response on imaging (96). The toxicity profile of HAI Oxaliplatin is abdominal pain (grades 3–4, 14%) and neutropenia (grades 3–4, 43%) (97). Irinotecan is not more useful via HAI route as the systemic levels of the active metabolite SN-38 are similar to that seen when irinotecan is given systemically (97–99). Biologic agents are also being used in combination with systemic chemotherapy in patients with initially unresectable liver disease. Recent data has shown that bevacizumab in combination with chemotherapy may increase hepatic resection rates and does not appear to impact adversely on surgical outcome or liver regeneration. In a nonrandomized phase II trial by Gruenberger and colleagues, the addition of bevacizumab to XELOX in patients with potentially resectable liver metastases resulted in an objective response rate of 73%, a resection rate of 93%, and no intraoperative or wound healing complications (100,101). To evaluate whether preoperative bevacizumab affects patients going for liver resection, the Bevacizumab Expanded Access Trial (BEAT) was designed and has thus far concluded that metastatectomy is feasible after bevacizumab treatment (102). Combination of EGFR antibodies with systemic chemotherapy may also have the potential to increase resection rates of unresectable or possibly resectable liver metastases. Adam and colleagues reported that combining Cmab with oxaliplatin- or irinotecan-based chemotherapy in chemorefractory patients with unresectable liver metastases can result in salvage liver resection rates of 17%. With a median follow-up of 16 months, 92% of resected patients (23/25) were alive and 10 patients (40%) were disease-free. There was no significant increase in operative mortality or liver injury. Median overall (OS) and progression-free survival (PFS) from initiation of cetuximab therapy was 20 and 13 months, respectively

139

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
(for resected patients) (103). A phase II study of FOLFOX4 plus cetuximab in the treatment of patients with EGFRexpressing initially unresectable liver metastases resection rates were 23.8% (8/10 of whom had liver metastases only), PFS of 12.8 months and median OS of 30 months (104). The CRYSTAL trial compared FOLFIRI + Cmab versus FOLFIRI alone and reported liver a resection rate (R0) of 9.8% for the investigational arm versus 4.5% for the control arm (58). The CELIM study reported an R0 resection rate of 34% for initially unresectable liver metastases (n = 106) with “neoadjuvant” FOLFOX/Cmab (n = 20) or FOLFIRI/Cmab (n = 16) (105). The NSABP is currently planning a trial to study the rates of conversion from unresectable to resectable liver disease using an EGFR antibody. Ongoing phase II trials in our institution are investigating the rate of conversion to complete resection with initially unresectable liver metastases after treatment with HAI FUDR Dex in combination with best systemic chemotherapy plus bevacizumab. Such studies will help to further define the role of HAI in the neoadjuvant setting and should lead to adequately powered phase III trials comparing HAI plus systemic chemobiologic therapy with chemobiologic therapy alone in this setting. Presently initial systemic chemotherapy with or without biologic therapy is reasonable as first-line therapy. If the liver disease is not resectable at this stage, then consideration should be given to HAI in combination with further systemic chemotherapy. If HAI therapy is not available, then chemotherapy with EGFR inhibitors should be used. The CAIRO (Capecitabine, Irinotecan, and Oxaliplatin in Advanced Colorectal Cancer) and FOCUS (Fluorouracil, Oxaliplatin, and CPT11 [irinotecan]-Use and Sequencing) trials investigated the strategy of sequential use of single agents and combination chemotherapy (108,109). The CAIRO study showed no significant difference in overall survival between sequential capecitabine followed by irinotecan followed by XELOX compared to XELIRI followed by XELOX (17.4 vs. 16.3 months, respectively, p = 0.3281). The FOCUS trial was a three-arm study comparing 5-FU/LV followed by irinotecan (control group) with either 5FU followed by 5-FU in combination with either irinotecan or oxaliplatin (group 1), or 5FU in combination with either irinotecan or oxaliplatin from the outset (group 2). Groups 1 and 2 achieved a longer overall survival time than the control group (13.9 months), but only the FOLFIRI regimen in group 2 achieved significance (16.7 months, p = 0.01). Both the CAIRO and FOCUS trials challenge the thinking that upfront combination regimens should be preferentially used. The staged approach upgraded to combination regimens has a role to play and can be considered. The literature on bevacizumab in first-or second-line setting has raised the question of continuing use of bevacizumab beyond progression of disease. Grothey and colleagues reported results from the large prospective observational study of 1445 patients who were enrolled in the BRiTE registry (Bevacizumab Regimens: Investigation of Treatment Effects and Safety). In multivariate analysis, bevacizumab beyond first progression (BBP) was strongly and independently associated with improved survival compared with no-BBP (31.8 vs. 19.9 months, p < 0.001) (110). Due to the substantial potential for selection bias in the BRiTE analysis such as patients with better performance scores or less disease receiving more Bev after progression, a phase III SWOG 0600 study bevacizumab continuation trial is planned to further investigate bevacizumab continuation beyond progression (Irinotecan Bevacizumab Continuation Trial iBET).

scheduling strategies for treatment of metastatic disease
Various strategies have been used in an attempt to improve the inconvenience and toxicity of chemotherapy. An especially troublesome toxicity is oxaliplatin-associated neurotoxicity. The OPTIMOX1 study compared FOLFOX4 [ARM A] given until progression with a “stop and go” regimen of FOLFOX7 (high-dose oxaliplatin and no bolus dose 5FU) X 6 cycles followed by maintenance 5FU X 12 cycles and then reintroduction of FOLFOX7 [ARM B] (89). There was an insignificant difference in prevalence of sensory neuropathy between the two arms, and median PFS (9 and 8.7 months for arms A and B, respectively) and overall survival (19.3 vs. 21.3) were equivalent. Maintenance 5FU (without oxaliplatin) is therefore a valid treatment option after initial exposure to FOLFOX. The OPTIMOX2 study evaluated a chemotherapy-free window compared to the “stop and go” schedule of OPTIMOX1. The maintenance chemotherapy group had significantly superior PFS and overall survival, the difference being especially seen in those patients with a poor prognosis (overall survival in the “stop and go” group versus chemotherapy free groups was 28.7 vs. 14.5 months, respectively), suggesting stopping all therapy was not effective (106). The OPTIMOX3 study will evaluate the use of targeted therapy with bevacizumab and erlotinib during the maintenance phase. Stopping irinotecan has been studied by Labianca and colleagues comparing FOLFIRI for 6 months with FOLFIRI X 2 months followed by 2 months of no treatment and then FOLFIRI for another 2 months. There was no significant difference in terms of efficacy between the two groups (107).

systemic therapy for resectable and unresectable liver disease
Reasons for giving neoadjuvant chemotherapy to those patients with clearly resectable liver metastases at presentation are (1) decreasing tumor size may make the surgery easier and (2) control micrometastatic disease. If a patient progresses in an extrahepatic site while on chemotherapy before liver resection, one can eliminate these patients from the risks and morbidity associated with hepatic resection. The LiverMetSurvey group found that those patients with ≥5 liver metastases survived longer if they were given neoadjuvant chemotherapy, with 5-year survival rates of 22% and 12% (p = 0.07) for the preoperatively and nonpreoperatively treated groups, respectively (3,111). (3) Assessment of chemotherapy activity preoperatively may help design appropriate postoperative therapy. (4) Preliver resection patients may tolerate chemotherapy better and full-dose treatment may impact on the ability to treat microscopic disease. (5) Response to neoadjuvant chemotherapy may reflect prognosis after liver resection (112). Adam and colleagues studied 131 patients (74% with synchronous CRC and resectable liver metastases) who underwent liver resection

140

CHEMOTHERAPY FOR METASTATIC COLORECTAL CANCER
for multiple lesions (>4) after systemic chemotherapy. In a multivariate analysis, tumor progression on chemotherapy and the number of chemotherapy regimens were independently associated with shorter survival duration (113). In contrast to this, a retrospective series at MSKCC of 111 patients with synchronous CRC and resectable liver metastases who received neoadjuvant chemotherapy were identified and it was shown that response to chemotherapy was not related to overall survival after hepatic resection. The median overall survival after liver resection was 62 months with a median follow-up of 63 months. Comparing response in three categories, that is, complete or partial response, stable disease, or progression of disease, median overall survival was similar (58 months – 65 months – 61 months, respectively, p = 0.98). Thus if response to neoadjuvant chemotherapy is used as a criterion for proceeding to liver resection, some patients may be denied potentially curative liver resection and therefore long-term survival (114). When patients with CRC and synchronous resectable liver metastases are undergoing treatment with neoadjuvant chemotherapy, it is important to scan frequently and consider short duration of preoperative chemotherapy. Potential disadvantages of neoadjuvant chemotherapy include: (1) liver toxicity from systemic chemotherapy, which includes steatosis, portal fibrosis, sinusoidal alterations, peliosis, and hemorrhagic centrilobular necrosis (3,115). These toxicities may increase the risk of liver resection, prevent liver resection, and impair the functioning of the remaining hepatic tissue (11). Oxaliplatin-based regimens are associated with a higher risk for vascular lesions and sinusoidal dilation, and irinotecan-based regimens are associated with higher risks for steatosis and steatohepatitis (116,117). There is relatively little data on the frequency or gravity of liver toxicity with the use of biologic agents prior to liver resection. D’Angelica and colleagues report no statistically significant increase in perioperative complications between perioperative bevacizumab versus matched-control groups (118). Klinger reports that when bevacizumab is added to oxaliplatin-based chemotherapy there was no impact on chemotherapy-induced hepatic steatosis and fibrosis, and bevacizumab decreased the severity of sinusoidal obstruction syndrome (101). (2) Secondary splenomegaly and resulting portal hypertension (3). (3) A complete radiologic response that may make it difficult for surgeons to resect appropriate areas (3). Benoist and colleagues report that persistent residual disease or early recurrence in situ were observed in 55 of 66 (83%) liver metastases having a complete response on imaging (119). The use of perioperative chemotherapy (FOLFOX) in patients with initially resectable liver metastases (≤4 metastases) was studied by the European Organization for the Research and Treatment of Cancer (EORTC 40983 study). In a 364-patient population, this trial showed that perioperative FOLFOX was compatible with major liver surgery, however, there was increased toxicities in treated group. The absolute increase in PFS in patients who underwent liver resection and perioperative FOLFOX was 9.2% (42.4% vs. 33.2%, p = 0.025). Reversible postoperative complications occurred more often after chemotherapy than after surgery (40/159 [25%] vs. 27/170 [16%]; p = 0.04) and included, biliary fistulae [output >100 ml/day for >10 days] (8% vs. 4%), hepatic failure [bilirubin >100 mg/ day for >3 days] (6% vs. 3%), and wound infection (3% vs. 2%). The clinical impact of these complications was not significant (120). Survival data are not yet available and the longterm benefit of neoadjuvant chemotherapy in those patients with initially resectable liver metastases is still not clear, especially since both pre- and postchemotherapy was given. At present, the EORTC 40051 BOS (Biologics, Oxaliplatin and Surgery) trial is assessing perioperative chemotherapy with FOLFOX6 and cetuximab with or without bevacizumab in patients with resectable hepatic metastases form CRC. Our advice, in clearly resectable lesions, is resection should be preformed first or if systemic chemotherapy is given, it should be for a short a period as possible (no more than six treatments at two-weekly intervals) to avoid liver toxicity, and liver lesions should be resected as soon as possible if the patient is a suitable surgical candidate. Consideration should then be given to “adjuvant” systemic chemotherapy in combination with HAI (see below).

˝adjuvant˝ systemic therapy after liver resection
The role of “adjuvant” systemic chemotherapy after liver resection is even more uncertain as the majority of data is with 5-FU/LV and it has been difficult to show a significant difference in disease-free and overall survival (121). Mitry and colleagues recently reported the combined results of two phase III trials comparing adjuvant FU/LV after liver resection with surgery alone. Median progression-free survival was 27.9 months in the chemotherapy (CT) arm as compared with 18.8 months in the surgery (S) arm (hazard ratio = 1.32; 95% CI: 1.00–1.76; p = .058). Median overall survival was 62.2 months in the CT arm compared with 47.3 months in the S arm (hazard ratio = 1.32; 95% CI: 0.95–1.82; p = .095). Adjuvant chemotherapy was independently associated with both progression-free survival and overall survival in multivariable analysis. Ychou and colleagues report no benefit with the addition of irinotecan to FU/LV after liver resection compared to FU/LV alone. The overall HR for DFS adjusted for the stratification factors was 0.89 (95% CI: 0.66–1.19, log-rank p = 0.47). Median DFS was 21.6 for FU/LV vs 24.7 months for FOLFIRI (122). No randomized data to support the use of “adjuvant” biologic therapy after liver resection have been published so far.

adjuvant regional therapy after liver resection
Recurrence of liver disease after liver resection is a significant problem with nearly 70% of patients developing recurrence in either hepatic or extrahepatic sites. It is estimated that up to 60% of recurrences will be in the liver (123). As microscopic liver disease is the most likely cause of this recurrence, there has been much interest in “adjuvant” HAI. Level I evidence for the role of HAI FUDR Dex in this setting is provided by Kemeny and colleagues. In a population of 156 patients who underwent complete liver resection, a phase III randomized trial was performed that compared HAI FUDR Dex plus systemic FU/LV with systemic FU/LV. After a median follow-up time of 10 years, 41% of the HAI arm are alive at 10 years

141

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
compared to 27.2% of the FU/LV alone arm. The median hepatic PFS has not yet been reached in the HAI arm and is 32.5 months in the systemic-only group (124,125). Other groups have also shown a clear benefit to “adjuvant” HAI plus systemic chemotherapy after liver resection (Table 14.3) (126–130). HAI FUDR Dex has been combined with modern systemic chemotherapy after liver resection. Kemeny and colleagues reported a two-year survival of 89% with HAI FUDR Dex in combination with irinotecan at a median follow-up of 26 months (131). A phase I trial examining adjuvant HAI FUDR Dex combined with systemic FOLFOX has been reported. Disease-free survival at 2 and 5 years was 59% and 50%, respectively, and 2- and 5-year overall survival rates were 90% and 86%, respectively (132). Based on this evidence, we recommend placement of an HAI pump at the time of liver resection and a 4 to 6 months period of “adjuvant” HAI FUDR/ Dex in combination with best available chemotherapy. Biologic therapy may be added in as part of a clinical trial. The benefits of adjuvant HAI after liver resection were recently demonstrated in a 1000-patient retrospective review by Ito and colleagues (133). In a multivariate analysis, one of the significant factors associated with survival after liver resection was HAI therapy. The median overall survival was 68 months with HAI therapy and 50 months for those that did not receive HAI (p = 0.0001). Another retrospective study of 250 patients who underwent liver resection compared adjuvant HAI FUDR + best available systemic chemotherapy (n = 125) with adjuvant systemic FOLFOX or FOLFIRI (n=125). Adjuvant HAI-FUDR plus modern systemic chemotherapy was associated with an improved liver recurrence-free survival (liver RFS) and disease-specific survival (DSS). For the adjuvant HAI-FUDR plus modern systemic group, the 5-year liver RFS, overall RFS, and DSS were 75%, 46%, and 72%, respectively, compared to 52%, 26%, and 55% for the modern sys alone group (p < 0.01) (134).

Table 14.3 True Randomized Trials of Adjuvant HAI FUDR
Chemotherapy Study Tono (141) MSKCC (124) ECOG (126) Lorenz (128) Lygidakis (129) HAI FU FUDR+IV FU/LV FUDR+IVFU/LV FU/LV Multidrug** IV po FU FU/LV FU None None No. Patients HAI 9 74 45 113 20 IV 10 82 30 113 20 2-year HPFS (%) HAI 78 90 67 67 – IV 30 60 43 63 – 2-year OS(%) HAI 78^ 86 62 62 – IV 50^ 72 53 65 – MS (Mths) HAI 63 72 64 44.8 20! IV 40 59 50 39.7 11!

PFS, hepatic progression-free survival; OS, overall survival; MS, median survival. ** Interleukin, carboplatin, mitomycin, epirubicin, LV, urographin. ! Mean survival. ^ 3-year overall survival.

Table 14.4 Dose Reductions for HAI FUDR
AST Ref value (ref) Current value$
*

FUDR dose >50 U/L 0 to <2 X ref 2 to <3 X ref 3 to <4 X ref ≥4 X ref <3 X ref >90 U/L 0 to <1.2 X ref 1.2 to 1.5 X ref ≥1.5 X ref <1.2 X ref >1.2 md/dl 0 to <1.2 X ref 1.2 to <1.5 X ref ≥1.5 X ref <1.2 X ref 100% 80% 50% HOLD 50% off last dose FUDR 100% 80% HOLD 25% off last dose FUDR 100% 50% HOLD 25% off last dose

If held,restart when: Ref value (ref) * Current value$

If held,restart when: Ref value (ref)* Current value$

If held, restart when:
* $

≤50 U/L 0 to <3 X ref 3 to <4 X ref 4 to <5 X ref ≥5 X ref <4 X ref Alk Phos ≤90 U/L 0 to <1.5 X ref 1.5 to >2 X ref ≥2 X ref <1.5 X ref Total Bilirubin ≤1.2 mg/dl 0 to <1.5 X ref 1.5 to <2 X ref ≥2 X ref <1.5 X ref

Reference value is the value obtained on the day patient received the last FUDR dose. Current value is that obtained at pump emptying or on the day of planned treatment (whichever is higher).

142

CHEMOTHERAPY FOR METASTATIC COLORECTAL CANCER
As preoperative chemotherapy for resectable or unresectable liver metastases becomes more effective, surgeons may be faced with the problem of “missing metastases,” that is, radiologic complete response of liver lesions. As mentioned above, radiologic response is more likely to be a true pathologic response when preoperative HAI is used. In the adjuvant setting, treatment with HAI is also more likely to result in a true pathologic complete response when “missing” liver metastases are left in place at hepatectomy. Elias and colleagues reported, in a series of 228 patients, that adjuvant HAI Oxaliplatin is significantly correlated with definitive eradication of “missing” metastases (p < 0.01) (95). Insertion of the implantable pump can be performed at the time of liver resection. The pump is placed percutaneously in the left lower quadrant and is fixed in position. The pump chamber is filled by accessing a subcutaneous septum and injecting. An expanding and contracting propellant liquid pushes on a bellows and infuses the drug, for example, FUDR, from the pump chamber via the hepatic artery catheter to the liver. It takes 2 weeks for the chamber to empty and it is then filled with glycerol or saline to keep the catheter patent. The cycle is then repeated after another 2 weeks, that is, 2 weeks of drug and 2 weeks of glycerol. Before insertion of the pump, hepatic arterial anatomy is viewed with a CT angiogram to make sure no aberrant vessels are present. Complication rates are low. Allen and colleagues reported a series of 544 patients with HAI pumps and complications were divided into early, for example, misperfusions and late, for example, dislodgement (135). Rates were less than 7% in both cases. A macroaggregated albumin nuclear scan is performed after all pump insertions to make sure the perfusion of the liver is adequate. Hepatotoxicity from HAI therapy depends on the drug being used and the duration of therapy. Raised transaminase is not uncommon (up to 70% of cases) and can be an early sign of liver damage. Raised bilirubin or alkaline phosphatise are a more serious sign of liver damage and may indicate sclerosing cholangitis (136). The addition of dexamethasone to FUDR has decreased the incidence of this side effect. An algorithm for does reductions based on liver blood tests has been drawn up and FUDR doses can be adjusted accordingly (Table 14.4). Treatment paradigms for metastatic CRC have changed dramatically over the last decade and involvement of a multidisciplinary team of surgeons, oncologists, radiologists, and pathologists can result in long-term survival becoming a reality.

references
1. Ferlay J, Autier P, Boniol M, et al. Estimates of the cancer incidence and mortality in Europe in 2006. Ann Oncol 2007; 18: 581–92. 2. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008; 58: 71–96. 3. Kemeny N. Presurgical chemotherapy in patients being considered for liver resection. Oncologist 2007; 12: 825–39. 4. Rees M, Tekkis PP, Welsh FK, et al: Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer: a multifactorial model of 929 patients. Ann Surg 2008; 247: 125–35. 5. Kemeny N. Management of liver metastases from colorectal cancer. Oncology (Williston Park). Oncology (Williston Park) 2006; 20: 1161–76. 6. Penna C, Nordlinger B. Colorectal metastasis (liver and lung). Surg Clin North Am 2002; 82: 1075–90, x–xi. 7. Meyerhardt JA, Mayer RJ. Drug therapy: Systemic therapy for colorectal cancer. N Engl J Med 2005; 352: 476–87. 8. Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 2004; 239: 818–27. 9. Fernandez FG, Drebin JA, Linehan DC, et al. Five-year survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with F-18 fluorodeoxyglucose (FDG-PET). Ann Surg 2004; 240: 438–50. 10. Tomlinson JS, Jarnagin WR, DeMatteo RP, et al. Actual 10-year survival after resection of colorectal liver metastases defines cure. J Clin Oncol 2007; 25: 4575–80. 11. Alberts SR, Wagman LD: Chemotherapy for colorectal cancer liver metastases. Oncologist 2008; 13: 1063–73. 12. Hoff PM, Cassidy J, Schmoll HJ. The evolution of fluoropyrimidine therapy: from intravenous to oral. Oncologist 2001; 6 Suppl 4: 3–11. 13. Meta-Analysis Group in Cancer. Toxicity of fluorouracil in patients with advanced colorectal cancer: effect of administration schedule and prognostic factors. J Clin Oncol 1998; 16: 3537–41. 14. Meta-Analysis Group in Cancer. Efficacy of intravenous continuous infusion of fluorouracil compared with bolus administration in advanced colorectal cancer. J Clin Oncol 1998; 16: 301–8. 15. Thirion P, Michiels S, Pignon JP, et al. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: an updated meta-analysis. J Clin Oncol 2004; 22: 3766–75. 16. de Gramont A, Bosset JF, Milan C, et al. Randomized trial comparing monthly low-dose leucovorin and fluorouracil bolus with bimonthly high-dose leucovorin and fluorouracil bolus plus continuous infusion for advanced colorectal cancer: a French intergroup study. J Clin Oncol 1997; 15: 808–15. 17. Cunningham D, Pyrhonen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998; 352: 1413–8. 18. Douillard JY, Cunningham D, Roth AD, et al. Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: a multicentre randomised trial. Lancet 2000; 355: 1041–7. 19. Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 2000; 343: 905–14. 20. Maiello E, Gebbia V, Giuliani F, et al. FOLFIRI regimen in advanced colorectal cancer: the experience of the Gruppo Oncologico dell’Italia Meridionale (GOIM). Ann Oncol 2005; 16 Suppl 4: iv56–60. 21. Fuchs CS, Marshall J, Mitchell E, et al. Randomized, controlled trial of irinotecan plus infusional, bolus, or oral fluoropyrimidines in first-line treatment of metastatic colorectal cancer: results from the BICC-C Study. J Clin Oncol 2007; 25: 4779–86.

conclusions
Liver resection should now be considered in all patients with liverconfined metastatic disease from CRC. Modern systemic chemotherapy with irinotecan- and oxaliplatin-based regimens can increase resection rates is a significant number of patients. The addition of biologic agents in patients with the appropriate molecular signature may increase repose rates and resectability rates even further. Long-term cures are possible in patients who undergo liver resection and series with 10-year survivors have been reported (10). In those patients in whom resection is not possible, overall survival has increased from less than 1 year with fluorouracil regimens to over 2 years with chemobiologic regimens (74). Hepatic arterial infusion with FUDR has consistently demonstrated increased response rates and hepatic progression-free survival compared to systemic chemotherapy. The combination of HAI FUDR with modern chemotherapy and perhaps chemobiologic therapy can result in increased resection rates of initially unresectable liver disease and also has a role to play after liver resection.

143

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
22. Rothenberg ML, Oza AM, Bigelow RH, et al. Superiority of oxaliplatin and fluorouracil-leucovorin compared with either therapy alone in patients with progressive colorectal cancer after irinotecan and fluorouracil-leucovorin: interim results of a phase III trial. J Clin Oncol 2003; 21: 2059–69. 23. Zori Comba A, Blajman C, Richardet E, et al. A randomised phase II study of oxaliplatin alone versus oxaliplatin combined with 5-fluorouracil and folinic acid (Mayo Clinic regimen) in previously untreated metastatic colorectal cancer patients. Eur J Cancer 2001; 37: 1006–13. 24. De Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18: 2938–47. 25. Tournigand C, Andre T, Achille E, et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol 2004; 22: 229–37. 26. Colucci G, Gebbia V, Paoletti G, et al. Phase III randomized trial of FOLFIRI versus FOLFOX4 in the treatment of advanced colorectal cancer: a multicenter study of the Gruppo Oncologico Dell’Italia Meridionale. J Clin Oncol 2005; 23: 4866–75. 27. Goldberg RM, Sargent DJ, Morton RF, et al. A randomized controlled trial of fluorouracil plus leucovorin, irinotecan, and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004; 22: 23–30. 28. Sanoff HK, Sargent DJ, Campbell ME, et al. Five-year data and prognostic factor analysis of oxaliplatin and irinotecan combinations for advanced colorectal cancer: N9741. J Clin Oncol 2008; 26: 5721–7. 29. Hochster HS, Hart LL, Ramanathan RK, et al. Safety and efficacy of oxaliplatin and fluoropyrimidine regimens with or without bevacizumab as first-line treatment of metastatic colorectal cancer: results of the TREE Study. J Clin Oncol 2008; 26: 3523–9. 30. Goldberg RM, Köhne CH, Seymour MT, et al. A pooled safety and efficacy analysis examining the effect of performance status (PS) on outcomes in nine first-line treatment (rx) trials (cts) of 6,286 patients (pts) with metastatic colorectal cancer (MCRC). J Clin Oncol 2007; 25: Abs# 4011. 31. Cassidy J, Clarke S, Diaz-Rubio E, et al. Randomized phase III study of capecitabine plus oxaliplatin compared with fluorouracil/folinic acid plus oxaliplatin as first-line therapy for metastatic colorectal cancer. J Clin Oncol 2008; 26: 2006–12. 32. Rothenberg ML, Navarro M, Butts C, et al. Phase III trial of capecitabine + oxaliplatin (XELOX) vs. 5-fluorouracil (5-FU), leucovorin (LV), and oxaliplatin (FOLFOX4) as 2nd-line treatment for patients with metastatic colorectal cancer (MCRC). J Clin Oncol 2007; 25: Abs# 4031. 33. Souglakos J, Vardakis N, Androulakis N, et al: Irinotecan plus weekly 5-fluorouracil and leucovorin as salvage treatment for patients with metastatic colorectal cancer: a phase II trial. Dig Dis 2007; 25: 100–5. 34. Kim GP, Sargent DJ, Mahoney MR, et al. Phase III noninferiority trial comparing irinotecan with oxaliplatin, fluorouracil, and leucovorin in patients with advanced colorectal carcinoma previously treated with fluorouracil: N9841. J Clin Oncol 2009; 20: 20. 35. Sullivan RD, Norcross JW, Watkins E, Jr. Chemotherapy of metastatic liver cancer by prolonged hepatic-artery infusion. N Engl J Med 1964; 270: 321–7. 36. Ensminger WD. Intrahepatic arterial infusion of chemotherapy: pharmacologic principles. Semin Oncol 2002; 29: 119–25. 37. Power DG, Kemeny NE. The role of floxuridine in metastatic liver disease. Mol Cancer Ther 2009; 21: 21. 38. Mocellin S, Pilati P, Lise M, et al. Meta-analysis of hepatic arterial infusion for unresectable liver metastases from colorectal cancer: the end of an era? J Clin Oncol 2007; 25: 5649–54. 39. Kemeny NE, Niedzwiecki D, Hollis DR, et al. Hepatic arterial infusion versus systemic therapy for hepatic metastases from colorectal cancer: a randomized trial of efficacy, quality of life, and molecular markers (CALGB 9481). J Clin Oncol 2006; 24: 1395–403. 40. Kemeny N, Conti JA, Cohen A, et al. Phase II study of hepatic arterial floxuridine, leucovorin, and dexamethasone for unresectable liver metastases from colorectal carcinoma. J Clin Oncol 1994; 12: 2288–95. 41. Goldberg RM, Sargent DJ, Morton RF, et al. Randomized controlled trial of reduced-dose bolus fluorouracil plus leucovorin and irinotecan or infused fluorouracil plus leucovorin and oxaliplatin in patients with previously untreated metastatic colorectal cancer: a North American Intergroup Trial. J Clin Oncol 2006; 24: 3347–53. Kemeny N, Gonen M, Sullivan D, et al. Phase I study of hepatic arterial infusion of floxuridine and dexamethasone with systemic irinotecan for unresectable hepatic metastases from colorectal cancer. J Clin Oncol 2001; 19: 2687–95. Kemeny NE, Huitzil FD, Capanu M, et al. Conversion to resectability utilizing hepatic artery infusion plus systemic chemotherapy for the treatment of unresectable liver metastases from colorectal cancer. J Clin Oncol Epub ahead of print, 2008. Kim KJ, Li B, Winer J, et al. Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumour growth in vivo. Nature 1993; 362: 841–4. Jain RK. Normalizing tumor vasculature with anti-angiogenic therapy: a new paradigm for combination therapy. Nat Med 2001; 7: 987–9. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350: 2335–42. Saltz LB, Clarke S, Diaz-Rubio E, et al. Bevacizumab in combination with oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal cancer: a randomized phase III study. J Clin Oncol 2008; 26: 2013–9. Fuchs CS, Marshall J, Barrueco J. Randomized, controlled trial of irinotecan plus infusional, bolus, or oral fluoropyrimidines in first-line treatment of metastatic colorectal cancer: updated results from the BICC-C study. J Clin Oncol 2008; 26: 689–90. Giantonio BJ, Catalano PJ, Meropol NJ, et al. Bevacizumab in combination with oxaliplatin, fluorouracil, and leucovorin (FOLFOX4) for previously treated metastatic colorectal cancer: results from the Eastern Cooperative Oncology Group Study E3200. J Clin Oncol 2007; 25: 1539–44. Goldstein NS, Armin M: Epidermal growth factor receptor immunohistochemical reactivity in patients with American Joint Committee on Cancer Stage IV colon adenocarcinoma: implications for a standardized scoring system. Cancer 2001; 92: 1331–46. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 2004; 351: 337–45. Saltz LB, Meropol NJ, Loehrer PJ, Sr., et al. Phase II trial of cetuximab in patients with refractory colorectal cancer that expresses the epidermal growth factor receptor. J Clin Oncol 2004; 22: 1201–8. Wilke H, Glynne-Jones R, Thaler J, et al. Cetuximab plus irinotecan in heavily pretreated metastatic colorectal cancer progressing on irinotecan: MABEL study. J Clin Oncol 2008; 26: 5335–43. Sobrero AF, Maurel J, Fehrenbacher L, et al. EPIC: phase III trial of cetuximab plus irinotecan after fluoropyrimidine and oxaliplatin failure in patients with metastatic colorectal cancer. J Clin Oncol 2008; 26: 2311–9. Jonker DJ, O’Callaghan CJ, Karapetis CS, et al. Cetuximab for the treatment of colorectal cancer. N Engl J Med 2007; 357: 2040–8. Prewett MC, Hooper AT, Bassi R, et al. Enhanced antitumor activity of anti-epidermal growth factor receptor monoclonal antibody IMCC225 in combination with irinotecan (CPT-11) against human colorectal tumor xenografts. Clin Cancer Res 2002; 8: 994–1003. Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol 2007; 25: 1658–64. Van Cutsem E, Kohne CH, Hitre E, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med 2009; 360: 1408–17. Bokemeyer C, Bondarenko I, Makhson A, et al. Fluorouracil, leucovorin, and oxaliplatin with and without cetuximab in the first-line treatment of metastatic colorectal cancer. J Clin Oncol 2008; 29: 29. Venook AP, Niedzwiecki D, Hollis D, et al. Phase III study of irinotecan/5FU/LV (FOLFIRI) or oxaliplatin/5FU/LV (FOLFOX) ± cetuximab for patients (pts) with untreated metastatic adenocarcinoma of the colon or rectum (MCRC): CALGB 80203 preliminary results. J Clin Oncol 2006; 26: Abs: 3509.

42.

43.

44.

45. 46.

47.

48.

49.

50.

51.

52.

53.

54.

55. 56.

57.

58.

59.

60.

144

CHEMOTHERAPY FOR METASTATIC COLORECTAL CANCER
61. Hecht J, Posey J, Tchekmedyian S, et al. Panitumumab in combination with 5-fluorouracil, leucovorin, and irinotecan (IFL) or FOLFIRI for first-line treatment of metastatic colorectal cancer (mCRC). ASCO Gastrointestinal Cancers Symposium: Abs 2006: 237. 62. Mano M, Humblet Y. Drug insight: panitumumab, a human EGFRtargeted monoclonal antibody with promising clinical activity in colorectal cancer. Nat Clin Pract Oncol 2008; 5: 415–25. 63. Helbling D, Borner M. Successful challenge with the fully human EGFR antibody panitumumab following an infusion reaction with the chimeric EGFR antibody cetuximab. Ann Oncol 2007; 18: 963–4. 64. Saif MW, Peccerillo J, Potter V. Successful re-challenge with panitumumab in patients who developed hypersensitivity reactions to cetuximab: report of three cases and review of literature. Cancer Chemother Pharmacol 2008; 10: 10. 65. Cappuzzo F, Finocchiaro G, Rossi E, et al. EGFR FISH assay predicts for response to cetuximab in chemotherapy refractory colorectal cancer patients. Ann Oncol 2008; 19: 717–23. 66. Chung KY, Shia J, Kemeny NE, et al. Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol 2005; 23: 1803–10. 67. Personeni N, Fieuws S, Piessevaux H, et al. Clinical usefulness of EGFR gene copy number as a predictive marker in colorectal cancer patients treated with cetuximab: a fluorescent in situ hybridization study. Clin Cancer Res 2008; 14: 5869–76. 68. Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol 2008; 26: 1626–34. 69. De Roock W, Piessevaux H, De Schutter J, et al. KRAS wild-type state predicts survival and is associated to early radiological response in metastatic colorectal cancer treated with cetuximab. Ann Oncol 2008; 19: 508–15. 70. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008; 359: 1757–65. 71. Khambata-Ford S, Garrett CR, Meropol NJ, et al. Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab. J Clin Oncol 2007; 25: 3230–7. 72. Lievre A, Bachet JB, Boige V, et al. KRAS mutations as an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab. J Clin Oncol 2008; 26: 374–9. 73. Lievre A, Bachet JB, Le Corre D, et al. KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Cancer Res 2006; 66: 3992–5. 74. Van Cutsem E, Lang I, D’haens G, et al. KRAS status and efficacy in the first-line treatment of patients with metastatic colorectal cancer (mCRC) treated with FOLFIRI with or without cetuximab: The CRYSTAL experience. J Clin Oncol 2008; 26: Abs 2. 75. Rougier P, Stroiakovski D, Köhne C, et al. Addition of cetuximab to FOLFIRI in first-line metastatic colorectal cancer (mCRC): Updated survival data and influence of KRAS status on outcome in the CRYSTAL study. American Society of Clinical Oncology—Gastrointestinal Cancer Symposium 2009; Abstract # 443. 76. Di Nicolantonio F, Martini M, Molinari F, et al. Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer. J Clin Oncol 2008; 10: 10. 77. Razis E, Briasoulis E, Vrettou E, et al. Potential value of PTEN in predicting cetuximab response in colorectal cancer: an exploratory study. BMC Cancer 2008; 8: 234. 78. Saltz LB, Lenz HJ, Kindler HL, et al. Randomized phase II trial of cetuximab, bevacizumab, and irinotecan compared with cetuximab and bevacizumab alone in irinotecan-refractory colorectal cancer: the BOND-2 study. J Clin Oncol 2007; 25: 4557–61. 79. Hecht JR, Mitchell E, Chidiac T, et al. A randomized phase IIIB trial of chemotherapy, bevacizumab, and panitumumab compared with chemotherapy and bevacizumab alone for metastatic colorectal cancer. J Clin Oncol 2008; 29: 29. 80. Tol J, Koopman M, Cats A, et al. Chemotherapy, bevacizumab, and cetuximab in metastatic colorectal cancer. N Engl J Med 2009; 360: 563–72. 81. Baize N, Gerard B, Bleiberg H, et al. Long-term survival of patients downstaged by oxaliplatin and 5-fluorouracil combination followed by rescue surgery for unresectable colorectal liver metastases. Gastroenterol Clin Biol 2006; 30: 1349–53. 82. Bismuth H, Adam R, Levi F, et al. Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. Ann Surg 1996; 224: 509–22. 83. Giacchetti S, Itzhaki M, Gruia G, et al. Long-term survival of patients with unresectable colorectal cancer liver metastases following infusional chemotherapy with 5-fluorouracil, leucovorin, oxaliplatin and surgery. Ann Oncol 1999; 10: 663–9. 84. Folprecht G, Grothey A, Alberts S, et al. Neoadjuvant treatment of unresectable colorectal liver metastases: correlation between tumour response and resection rates. Ann Oncol 2005; 16: 1311–9. 85. Adam R, Delvart V, Pascal G, et al. Rescue surgery for unresectable colorectal liver metastases downstaged by chemotherapy: a model to predict long-term survival. Ann Surg 2004; 240: 644–57; discussion 657–8. 86. Falcone A, Ricci S, Brunetti I, et al. Phase III trial of infusional fluorouracil, leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) compared with infusional fluorouracil, leucovorin, and irinotecan (FOLFIRI) as first-line treatment for metastatic colorectal cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol 2007; 25: 1670–6. 87. Masi G, Loupakis F, Pollina L, et al. Long-term outcome of initially unresectable metastatic colorectal cancer patients treated with 5-fluorouracil/leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) followed by radical surgery of metastases. Ann Surg 2009; 249: 420–5. 88. Souglakos J, Androulakis N, Syrigos K, et al. FOLFOXIRI (folinic acid, 5-fluorouracil, oxaliplatin and irinotecan) vs FOLFIRI (folinic acid, 5-fluorouracil and irinotecan) as first-line treatment in metastatic colorectal cancer (MCC): A multicentre randomised phase III trial from the Hellenic Oncology Research Group (HORG). Br J Cancer 2006; 94: 798–805. 89. Tournigand C, Cervantes A, Figer A, et al. OPTIMOX1: a randomized study of FOLFOX4 or FOLFOX7 with oxaliplatin in a stop-and-Go fashion in advanced colorectal cancer--a GERCOR study. J Clin Oncol 2006; 24: 394–400. 90. Gallagher DJ, Capanu M, Raggio G, et al. Hepatic arterial infusion plus systemic irinotecan in patients with unresectable hepatic metastases from colorectal cancer previously treated with systemic oxaliplatin: a retrospective analysis. Ann Oncol 2007; 24: 24. 91. Taylor RA, White RR, Kemeny N, et al. Predictors of a true complete response in colorectal liver metastases that disappear radiographically following chemotherapy. J Clin Oncol 2007; 25: Abstract:4058. 92. Auer RC, White RR, Kemeny N, et al. Predictors of a true complete response among disappearing liver metastases from colorectal cancer following chemotherapy. Cancer 2010; 116(6): 1502–9. 93. Ducreux M, Ychou M, Laplanche A, et al. Hepatic arterial oxaliplatin infusion plus intravenous chemotherapy in colorectal cancer with inoperable hepatic metastases: a trial of the gastrointestinal group of the Federation Nationale des Centres de Lutte Contre le Cancer. J Clin Oncol 2005; 23: 4881–7. 94. Boige V, Malka D, Elias D, et al. Hepatic arterial infusion of oxaliplatin and intravenous LV5FU2 in unresectable liver metastases from colorectal cancer after systemic chemotherapy failure. Ann Surg Oncol 2008; 15: 219–26. 95. Elias D, Goere D, Boige V, et al. Outcome of posthepatectomy-missing colorectal liver metastases after complete response to chemotherapy: impact of adjuvant intra-arterial hepatic oxaliplatin. Ann Surg Oncol 2007; 14: 3188–94. 96. Benoist S, Brouquet A, Penna C, et al. Complete response of colorectal liver metastases after chemotherapy: Does it mean cure? J Clin Oncol 2006; 24: 3939–45. 97. Boige V, Malka D, Elias D, et al. Hepatic arterial infusion of oxaliplatin and intravenous lv5fu2 in unresectable liver metastases from colorectal cancer after systemic chemotherapy failure. Ann Surg Oncol 2007; 26: 26. 98. De Jong FA, Mathijssen RH, Verweij J. Limited potential of hepatic arterial infusion of irinotecan. J Chemother 2004; 16 Suppl 5: 48–50. 99. van Riel JM, van Groeningen CJ, de Greve J, et al. Continuous infusion of hepatic arterial irinotecan in pretreated patients with colorectal cancer metastatic to the liver. Ann Oncol 2004; 15: 59–63.

145

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
100. Gruenberger B, Tamandl D, Schueller J, et al. Bevacizumab, capecitabine, and oxaliplatin as neoadjuvant therapy for patients with potentially curable metastatic colorectal cancer. J Clin Oncol 2008; 26: 1830–5. 101. Klinger M, Eipeldauer S, Hacker S, et al. Bevacizumab protects against sinusoidal obstruction syndrome and does not increase response rate in neoadjuvant XELOX/FOLFOX therapy of colorectal cancer liver metastases. Eur J Surg Oncol 2009; 4: 4. 102. Kretzschmar A, Cunningham D, Berry S. Feasibility of metastasectomy in patients treated with first-line bevacizumab for metastatic colorectal cancer: Preliminary results from the first BEAT study. 2007 Gastrointestinal Cancers Symposium, 2007. 103. Adam R, Aloia T, Levi F, et al. Hepatic resection after rescue cetuximab treatment for colorectal liver metastases previously refractory to conventional systemic therapy. J Clin Oncol 2007; 25: 4593–602. 104. Tabernero J, Van Cutsem E, Diaz-Rubio E, et al. Phase II trial of cetuximab in combination with fluorouracil, leucovorin, and oxaliplatin in the first-line treatment of metastatic colorectal cancer. J Clin Oncol 2007; 25: 5225–32. 105. Folprecht G, Gruenberger T, Hartmann JT, et al. Cetuximab plus FOLFOX6 or cetuximab plus FOLFIRI as neoadjuvant treatment of nonresectable colorectal liver metastases: A randomized multicenter study (CELIM-study). American Society of Clinical Oncology— Gastrointestinal Cancer Symposium 2009; Abstract # 296. 106. Maindrault-Goebel F, Lledo G, Chibaudel B, et al. Final results of OPTIMOX2, a large randomized phase II study of maintenance therapy or chemotherapy-free intervals (CFI) after FOLFOX in patients with metastatic colorectal cancer (MRC): A GERCOR study. J Clin Oncol 2007: Abstract # 4013. 107. Labianca R, Floriani I, Cortesi E, et al. Alternating versus continuous “FOLFIRI” in advanced colorectal cancer (ACC): A randomized “GISCAD” trial. J Clin Oncol 2006; 24: Abstract #3505. 108. Koopman M, Antonini NF, Douma J, et al. Sequential versus combination chemotherapy with capecitabine, irinotecan, and oxaliplatin in advanced colorectal cancer (CAIRO): a phase III randomised controlled trial. Lancet 2007; 370: 135–42. 109. Seymour MT, Maughan TS, Ledermann JA, et al. Different strategies of sequential and combination chemotherapy for patients with poor prognosis advanced colorectal cancer (MRC FOCUS): a randomised controlled trial. Lancet 2007; 370: 143–52. 110. Grothey A, Sugrue MM, Purdie DM, et al. Bevacizumab beyond first progression is associated with prolonged overall survival in metastatic colorectal cancer: results from a large observational cohort study (BRiTE). J Clin Oncol 2008; 26: 5326–34. 111. Adam R, Aloia T, Figueras J, et al. LiverMetSurvey: Analysis of clinicopathologic factors associated with the efficacy of preoperative chemotherapy in 2,122 patients with colorectal liver metastases. J Clin Oncol 2006; 24: Abstract #3521. 112. Leonard GD, Brenner B, Kemeny NE. Neoadjuvant chemotherapy before liver resection for patients with unresectable liver metastases from colorectal carcinoma. J Clin Oncol 2005; 23: 2038–48. 113. Adam R, Pascal G, Castaing D, et al. Tumor progression while on chemotherapy: A contraindication to liver resection for multiple colorectal metastases? Ann Surg 2004; 240: 1052–64. 114. Gallagher DJ, Zheng J, Capanu M, et al. Response to neoadjuvant chemotherapy does not predict overall survival for patients with synchronous colorectal hepatic metastases. Ann Surg Oncol 2009; 18: 18. 115. Chun YS, Laurent A, Maru D, et al. Management of chemotherapyassociated hepatotoxicity in colorectal liver metastases. Lancet Oncol 2009; 10: 278–86. 116. Aloia T, Sebagh M, Plasse M, et al. Liver histology and surgical outcomes after preoperative chemotherapy with fluorouracil plus oxaliplatin in colorectal cancer liver metastases. J Clin Oncol 2006; 24: 4983–90. 117. Vauthey JN, Pawlik TM, Ribero D, et al. Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases. J Clin Oncol 2006; 24: 2065–72. 118. D’Angelica M, Kornprat P, Gonen M, et al. Lack of evidence for increased operative morbidity after hepatectomy with perioperative use of bevacizumab: A matched case-control study. Ann Surg Oncol 2007; 14: 759–65. 119. Benoist S, Brouquet A, Penna C, et al. Complete response of colorectal liver metastases after chemotherapy: does it mean cure? J Clin Oncol 2006; 24: 3939–45. 120. Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 2008; 371: 1007–16. 121. Mitry E, Fields AL, Bleiberg H, et al. Adjuvant chemotherapy after potentially curative resection of metastases from colorectal cancer: a pooled analysis of two randomized trials. J Clin Oncol 2008; 26: 4906–11. 122. Ychou M, Hohenberger W, Thezenas S, et al. Randomized phase III trial comparing infused 5-fluorouracil/folinic acid (LV5FU) versus LV5FU+ irinotecan (LV5FU+IRI) as adjuvant treatment after complete resection of liver metastases from colorectal cancer (LMCRC). (CPT-GMA-301). J Clin Oncol 2008; 26: Abs: LBA4013. 123. Nakajima Y, Nagao M, Ko S, et al. Clinical predictors of recurrence site after hepatectomy for metastatic colorectal cancer. Hepatogastroenterology 2001; 48: 1680–4. 124. Kemeny N, Huang Y, Cohen AM, et al. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl J Med 1999; 341: 2039–48. 125. Kemeny NE, Gonen M. Hepatic arterial infusion after liver resection [7]. N Engl J Med 2005; 352: 734–5. 126. Kemeny MM, Adak S, Gray B, et al. Combined-modality treatment for resectable metastatic colorectal carcinoma to the liver: surgical resection of hepatic metastases in combination with continuous infusion of chemotherapy—an intergroup study. J Clin Oncol 2002; 20: 1499–505. 127. Lygidakis NJ, Sgourakis G, Vlachos L, et al. Metastatic liver disease of colorectal origin: the value of locoregional immunochemotherapy combined with systemic chemotherapy following liver resection. Results of a prospective randomized study. Hepatogastroenterology 2001; 48: 1685–91. 128. Lorenz M, Muller HH, Schramm H, et al. Randomized trial of surgery versus surgery followed by adjuvant hepatic arterial infusion with 5-fluorouracil and folinic acid for liver metastases of colorectal cancer. German Cooperative on Liver Metastases (Arbeitsgruppe Lebermetastasen). Ann Surg 1998; 228: 756–62. 129. Lygidakis NJ, Ziras N, Parissis J. Resection versus resection combined with adjuvant pre- and post-operative chemotherapy—immunotherapy for metastatic colorectal liver cancer. A new look at an old problem. Hepatogastroenterology 1995; 42: 155–61. 130. O’Connell MJ, Bolton JS, Mahoney MR, et al. Final results of hepatic arterial infusion (HAI) plus systemic (SYS) chemotherapy after multiple metastasectomy in patients with colorectal carcinoma metastatic (M-CRC) to the liver: A North Central Cancer Treatment Group (NCCTG) phase II study. J Clin Oncol 2004; 22: Abstract No 3527. 131. Kemeny N, Jarnagin W, Gonen M, et al. Phase I/II study of hepatic arterial therapy with floxuridine and dexamethasone in combination with intravenous irinotecan as adjuvant treatment after resection of hepatic metastases from colorectal cancer. J Clin Oncol 2003; 21: 3303–9. 132. Kemeny N, Capanu M, D’Angelica M, et al. Phase I trial of adjuvant hepatic arterial infusion (HAI) with floxuridine (FUDR) and dexamethasone plus systemic oxaliplatin, 5-fluorouracil and leucovorin in patients with resected liver metastases from colorectal cancer. Ann Oncol 2009; 20: 20. 133. Ito H, Are C, Gonen M, et al. Effect of postoperative morbidity on longterm survival after hepatic resection for metastatic colorectal cancer. Ann Surg 2008; 247: 994–1002. 134. House MG, Kemeny N, Jarnagin WR, et al. Comparison of adjuvant systemic chemotherapy with or without hepatic arterial infusional chemotherapy after hepatic resection for metastatic colorectal cancer. American Society of Clinical Oncology—Gastrointestinal Cancer Symposium 2009; Abstract # 383. 135. Allen PJ, Nissan A, Picon AI, et al. Technical complications and durability of hepatic artery infusion pumps for unresectable colorectal liver metastases: an institutional experience of 544 consecutive cases. J Am Coll Surg 2005; 201: 57–65. 136. Power DG, Healey-Bird BR, Kemeny NE. Regional chemotherapy for liver-limited metastatic colorectal cancer. Clin Colorectal Cancer 2008; 7: 247–59.

146

CHEMOTHERAPY FOR METASTATIC COLORECTAL CANCER
137. Park SH, Sung JY, Han SH, et al: Oxaliplatin, folinic acid and 5-fluorouracil (FOLFOX-4) combination chemotherapy as secondline treatment in advanced colorectal cancer patients with irinotecan failure: a Korean single-center experience. Jpn J Clin Oncol 2005; 35: 531–5. 138. Pozzo C, Basso M, Cassano A, et al. Neoadjuvant treatment of unresectable liver disease with irinotecan and 5-fluorouracil plus folinic acid in colorectal cancer patients. Ann Oncol 2004; 15: 933–9. 139. Alberts SR, Horvath WL, Sternfeld WC, et al. Oxaliplatin, fluorouracil, and leucovorin for patients with unresectable liver-only metastases from colorectal cancer: A North Central Cancer Treatment Group phase II study. J Clin Oncol 2005; 23: 9243–9. 140. Masi G, Marcucci L, Loupakis F, et al. First-line 5-fluorouracil/folinic acid, oxaliplatin and irinotecan (FOLFOXIRI) does not impair the feasibility and the activity of second line treatments in metastatic colorectal cancer. Ann Oncol 2006; 17: 1249–54. 141. Tono T, Hasuike Y, Ohzato H, et al. Limited but definite efficacy of prophylactic hepatic arterial infusion chemotherapy after curative resection of colorectal liver metastases: A randomized study. Cancer 2000; 88: 1549–56.

147

15

Multimodal approaches to the management of colorectal liver metastases Gerardo Sarno and Graeme J. Poston
purpose, and then the role(s) of each of the possible treatment modalities (surgery, ablation, systemic chemotherapy, regional chemotherapy/radiotherapy, and biological therapies) as well as the strategy of which treatment to use in which sequence.

This chapter focuses on the recent development of multimodal strategies intended to increase the pool of patients with colorectal liver metastases (CRLMs) for whom curative treatment may be possible. These strategies include improved preoperative staging, new standards for surgical resection, novel surgical strategies, the application of modern systemic chemotherapy in the neoadjuvant setting, an emerging role for ablative therapies, greater emphasis on the collaborative, multidisciplinary management of this disease, and most recently, the question of whether to resect the liver disease before the primary bowel tumor. It is now clear that an aggressive multidisciplinary approach to the management of this problem can result in one-third of these patients now being considered for treatment that even if not achieving complete cure, offers significant long-term survival. Colorectal cancer is globally a growing cause of public health concern (1,2). The prevalence is increasing at 5% per year among the burgeoning middle classes in both China and India, and in western society is expected to increase in incidence by over 30% over the next 20 years because of evergrowing elderly (>70 years of age) population (1,2).The liver is frequently the only site in 30% to 40% of patients with advanced disease (3). By the time of initial diagnosis of colorectal cancer, nearly a quarter of patients will have clinically detectable CRLMs, despite increasing patient and clinician awareness of the disease (1,4,5). Historically, these patients have a poorer prognosis when compared to those who subsequently develop metachronous diseases (1,5). Of those who undergo apparently successful resection of the primary tumor, nearly half will develop liver metastases, usually within the first 3 years after colectomy (1,4,5). Until recently, surgery was the only treatment that offered the chance of cure for CRLM, and until recently, only far less than 20% of these patients were considered suitable for attempted curative resection; historically, the remaining patients being offered palliative and symptomatic treatment (6). Recent data suggest that ablation therapy (radiofrequency or RFA, microwave) might achieve long-term survival, but with poorer overall results compared to surgical resection (7). The other major advance in recent years has been the availability of medical oncology strategies using chemotherapeutic and biologic agents not only to significantly prolong survival in incurable disease, but also to bring initially inoperable patients to surgical resection with curative intent (8). Recently, it has become a legal requirement in a number of European countries (UK, France, Belgium, and Spain) for all cancer patients to be discussed within the setting of a multidisciplinary team (MDT) before any treatment intervention commences. In order for such an MDT to be effective in the management of colorectal cancer liver metastases, the team must undertake a number of specific steps in determining the extent of spread of the cancer, and the best modalities for this

building an effective multidisciplinary team
An effective MDT needs to be built around a designated core membership. Once this core group is established, then other professionals from many disciplines can attend and become involved. For the management of patients with CRLM, the key disciplines of the core group include hepatobiliary surgery, medical oncology, diagnostic radiology, interventional radiology, and palliative care. It is essential that there is a designated team member from each of these disciplines. Other disciplines that we have found helpful in support of our CRLM MDT include gastroenterology/hepatology, our specialist nursing colleagues, and histopathology. However, having a dedicated clerical coordinator who can pull together all the relevant correspondence, documentation and investigations for each and every patient to be discussed is absolutely essential for the successful operation of such an MDT. With regard to diagnostic radiology, it is our experience that the radiologist presenting the images to the MDT will require at least 3 hours preparation time for every 20 patients to be discussed. This requirement for radiology time has major cost implications for the running of a busy radiology department.

preoperative staging: the key to selection of candidates for curative treatment
The uses of individual imaging techniques for diagnosing and staging CRLM have differing strengths and weaknesses. However, with all modalities we are rapidly improving our ability to detect low-volume metastatic disease much earlier in the disease process. It must be remembered that all metastases (those found at the time of initial presentation, and those subsequently found metachronously after apparently “curative” resection of the primary tumor) are synchronous to the time of diagnosis of primary colorectal cancer. There is now emerging consensus on the optimal choice of technique, and the sequence with which they should be employed (9–12).

computeed tomography
Recent advances in computed tomography (CT) technology (helical CT and multidetector row helical CT) have improved performance in speed of acquisition, resolution, and ability to image the liver during various phases of contrast enhancement with greater precision (9,12). Using intravenous iodinated contrast media these techniques characterize liver lesions based on their enhancement patterns during the various phases of contrast circulation in the liver (12). CT has limitations, including

148

MULTIMODAL APPROACHES TO THE MANAGEMENT OF COLORECTAL LIVER METASTASES
the need for a high radiation dose and low sensitivity in detecting and characterizing lesions smaller than 1 cm. 3. The volume of the liver remaining after resection, i.e., the “future remnant liver” (FRL) will be adequate (13). Clearly, the FRL limit for safe resection varies from patient to patient, and from institution to institution but in those with an otherwise normal liver, the safe FRL volume is 20% to 30% (19).

magnetic resonance imaging
Magnetic resonance imaging (MRI) is highly effective in detecting and characterizing smaller (<1 cm) liver lesions because of the high lesion to liver contrast, most frequently using gadolinium (9,12). The use of liver-specific contrast media, such as super paramagnetic iron oxide (SPIO), further improves the contrast between normal liver tissue and metastases (12,13). However, MRI is limited by low sensitivity for detecting extrahepatic disease, especially in the peritoneum and chest.

resection margins
Historically, resection was only considered if the hepatobiliary surgeon believed that the metastasis could be resected with a margin of healthy surrounding liver that was >1 cm. The new standards challenge the “1 cm rule.” More recent studies show that size of the resection margin has no effect on survival, as long as the margin is microscopically clear of disease (20,21).

positron emission tomography
Positron emission tomography (PET-CT) has emerged as an important diagnostic tool in detecting and staging metastatic colorectal cancer. Although the modality appears to be highly sensitive, specificity is lower because any focal area of hypermetabolism (including inflammation and abscesses) can generate false-positive results. Other disadvantages include higher cost, poorer lesion localization and limited sensitivity for lesions smaller than 1 cm. (13,14).

new strategies to improve resectability
Other strategies are being increasingly employed in patients with unresectable CRLM to improve resectability. Portal vein embolization induces atrophy of the liver to be resected with hypertrophy of the liver that is to remain (i.e., increases the FRL). Similarly, two-stage hepatectomies, employing delayed rehepatectomy after hypertrophy of the residual liver, may be used for large bilateral lesions in which a single-stage resection of all involved segments would result in acute liver failure (22,23). Disease outside the liver that may be resected with curative intent includes direct diaphragmatic invasion, adrenal metastases and lung metastases when few in number and readily resectable (1). Recent reports demonstrate that up to 35% of patients are still alive 5 years after resection of pulmonary colorectal metastases (24).

surgery
There are many substantial prospective and retrospective series of surgical resection of CRLM consistently show 5-year survival rates following liver resection of 30% to 50%, depending on selection criteria (15). The problem encountered when attempting to interpret these reports is that although there are more than 600 in the literature, barely 30 series are prospective studies, reporting more than 100 patients from reliable high-volume centers, and with median follow-up of >24 months (15). However, from these reports nearly all patients who survive for more than 5 years can usually be considered cured of the disease.

combining chemotherapy and surgery
Modern chemotherapeutic regimens utilizing oxaliplatin with 5-FU and folinic acid (FOLFOX), also irinotecan (FOLFIRI) are associated with high response rates of up to 50% and median survival in incurable disease that exceeds 2 years (25,26). Most significantly, such high response rates can now bring 10% to 30% of patients with disease initially considered unresectable to subsequent secondary liver resection (22,25,26). Within a consecutive series of 1104 patients with CRLM initially considered unresectable and treated with chemotherapy, 138 (12.5%) had a sufficiently good response to chemotherapy to enable potentially curative liver surgery to be performed in 93% of these cases (22). Survival was 33% and 23% at 5 and 10 years, respectively, with a median survival of 39 months, although this was significantly lower than that for patients resected primarily within the same period at the same institution (48% and 30% at 5 and 10 years, respectively) (22). Evaluation of these and other data suggest that the ability to achieve secondary liver resection of initially inoperable CRLM is directly proportional to the degree of response to the chemotherapy regimen (26). Phase II and III studies evaluating novel biological agents, such as the monoclonal antibodies directed against vascular endothelial growth factor (VEGF) (bevacizumab) and the epidermal growth factor receptor (cetuximab and panetumumab), suggest even greater response rates (and possibly higher

defining resectability of liver only disease
Historically, resectability of CRLM was relatively straightforward. The definition of resectability was based of old studies that identified certain adverse clinicopathological factors, and so liver resection was only attempted in patients who had one to three unilobar metastases, preferably presenting at least 12 months after resection of the primary tumor, whose disease was resectable with at least a 1 cm margin of healthy liver tissue and who had no hilar lymphadenopathy or extrahepatic disease (16). Such patients accounted for <10% of the total population with liver only metastatic disease (16). We now know that patients outside these traditionally accepted criteria can benefit from long-term survival following hepatectomy (17,18). Resectability is now based on whether a macroscopically and microscopically complete (R0) resection of the liver can be achieved. Therefore resectability is now defined by what healthy liver volume will remain. Our definition of liver resectability is now (19): 1. Disease can be completely resected. 2. At least two adjacent liver segments can be spared with adequate vascular inflow and outflow and biliary drainage.

149

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
100 90 80 70 60 50 40 Surgery only 30 20 10 0 0 O N 104 152 93 151 1 2 3 4 24 23 5 10 6 6 Number of patients at risk: 85 59 39 118 76 45 (years) 33.2% 42.4% Periop CT +9.2% At 3 years HR = 0.73; CI:0.55–0.97, p = 0.025

Figure 15.1 Three-year progression-free survival comparing surgery alone to surgery with perioperative chemotherapy in the EORTC 40983 (EPOC) trial (30).

secondary CRLM resection rates) when compared to conventional chemotherapy alone. Therefore, even more patients with initially unresectable CRLM may respond to treatment with combinations of systemic treatments in the future (27–29). Recent data from the German Phase II CELIM study have suggested that as many as 40% of patients with unresectable kras wild type colorectal cancer metastases confined to the liver may now be brought to liver resection with curative intent using combinations of cetuximab with either oxaliplatin-based or irinotecan-based chemotherapy regimens (30). Recent data have suggested that the addition of perioperative (both neoadjuvant with adjuvant) chemotherapy using FOLFOX to surgical resection confers improved disease-free survival when compared to surgery alone (31). These data need to be interpreted with caution since the study did not demonstrate its primary endpoint (3-year disease-free survival on intention to treat at the point of initial randomization), and only achieved significance on the analysis of operated patients, when ineligible patients were excluded (Fig. 15.1).

the role of tumor ablation
Much interest in tumor ablation (mostly using RFA) derives from its low morbidity and mortality (32). A recent metaanalysis of 95 published series reported complication rates of <9% (33), the commonest complications being intra-abdominal bleeding, sepsis, and biliary injury. Mortality rates range from 0% to 0.5%. However, the most reported disadvantage of RFA are the higher rates of local recurrence, ranging from 1.8% to 12% using the surgical approach, to as high as 40% with radiologic guided percutaneous placement of the probe. Undoubtedly, some of this higher local failure rate relates to the type of lesions being treated by percutaneous RFA. Ablative therapies are often used for the treatment of metastases that

are often too close to major vascular structures to be considered resectable with a clear margin. Just as a surgical margin would be likely to be compromised, high blood flow immediately adjacent to the tumor will conduct away heat, leading to incomplete ablation and tumor recurrence (2). The efficacy of RFA in unresectable CRLM has been established by several large cohort studies with median survivals of 28.9 to 36 months being achieved (32,33). Presently the dearth of prospective randomized controlled trials comparing RFA with chemotherapy over chemotherapy alone in unresectable CRLM is being addressed by the EORTC CLOCC trial (EORTC 40004). Early progression-free survival data from this study have suggested that the addition of RFA to FOLFOX-based chemotherapy confers a statistically significant improved progression-free survival of 17 months compared to 10 months for FOLFOX alone (T Ruers, personal communication). Microwave ablation therapy is now becoming commercially available. The major advantage of microwave ablation over RFA is speed. Whereas it may take 20 to 30 minutes to achieve an adequate ablation of a 3-cm metastasis using RFA, microwave can achieve the same degree of tumor destruction in only 3 to 4 minutes.

management strategies for synchronously detectable crlm
Patients who present with technically “easily” resectable primary tumor (right, transverse, left, and sigmoid colon) and peripherally placed, low-volume liver disease (segments 2, 3, 4B, 5, 6, and subcapsular lesions in segments 4A, 7, and 8) are amenable to synchronous resection of both primary tumor and metastatic liver disease at the same procedure, without significantly increased morbidity or mortality (34–37). Those patients (a decreasing minority) who present with large bowel obstruction, perforation or life-threatening hemorrhage and

150

MULTIMODAL APPROACHES TO THE MANAGEMENT OF COLORECTAL LIVER METASTASES
CRC metastases

Up-to-date CT or MRI of chest, abdomen, and pelvis

Hepatic metastases only

Determine if patient is candidate for hepatic resection or HR and RFA or RFA alone

Extrahepatic disease

Consider chemotherapy

Not candidate for HR or RFA Response Consider chemotherapy Hepatic resection Response
All tumor(s) resectable

No response

HR and RFA
Some tumor(s) resectable and some ablatable
(Laparotomy – HR and RFA)

RFA alone
All tumor(s) ablatable but not resectable
Lapartomy, Laparoscopic or Percutaneous)

No response Consider HR and/or RFA

Consider HR and/or RFA

Follow-up CT or MRI of the chest, abdomen, and pelvisand CRC surveillance

Incomplete ablation or new metastases

Consider repeat or serial RFA and/or repeat HR

Figure 15.2 The possible treatment strategy algorithm for patients with colorectal liver metastases (40).

synchronous CRLM should have immediate definitive lifesaving treatment (endoscopic stenting, resection with either a stoma or immediate reconstruction). Most surgical oncologists would recommend that in situations where resection of the primary tumor may be more demanding (T2–T3 rectal carcinoma), or when the management strategy for the primary tumor requires neoadjuvant treatment (chemoradiotherapy for T3–T4 rectal carcinoma), or the liver disease (albeit technically resectable) is of such an extent that it requires at least a hemi-hepatectomy or more, then planned sequential staged procedures carry lower perioperative risk (36,37). However, when considering staged sequential treatment strategies, concerned must remain about the risk of tumor progression at both sites during treatment (31,38–40). For

patients presenting with asymptomatic primary tumors in the presence of unresectable liver metastases, it would be reasonable to propose a course of systemic chemotherapy and base subsequent treatment strategies on the degree of response (38). Those patients whose chemotherapy response is sufficient that their liver disease is now amenable to potential hepatectomy can now be considered for surgery with curative intent (31,38–40). For the 6% to 10% of patients with present with inoperable disease and continue to progress while on chemotherapy (23,31), consideration can be given to further lines of chemotherapy, but overall the outlook is poor and futile surgery can be avoided. For patients with primary colon cancer (as opposed to primary rectal tumors) with initially unresectable liver whose disease responds so well that an R0 resection of all tumor sites

151

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
can be achieved using a relatively minor liver resection, then synchronous liver–bowel surgery is feasible (38).
12. Martinez L, Puig I, Valls C. Colorectal liver metastases: Radiological diagnosis and staging. Eur J Surg Oncol 2007; 33(S2): S5–S16. 13. Charnsangavej C. Selection for resection: Preoperative imaging evaluation. Program of the AHPBA 2006 consensus conference; January 25, 2006; San Francisco, CA. 14. Israel O, Mor M, Gaitini D, et al. Combined structural and functional evaluation of cancer patients with a hybrid camera based PET/CT system using (18) F-FDG. J Nucl Med 2002; 43: 1129–36. 15. Simmonds PC, Primrose JN, Colquitt JL, et al. Surgical resection of hepatic metastases from colorectal cancer: A systematic review of published studies. Br J Cancer 2006; 94: 982–99. 16. Hughes KS, Simon R, Songhorabodi S, et al. Resection of the liver for colorectal carcinoma metastases: A multi-institutional study of patterns of recurrence. Surgery 1986; 100: 278–284. 17. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: Analysis of 1001 consecutive cases. Ann Surg 1999; 230: 309–18. 18. Minagawa M, Makuuchi M, Torzilli G, et al. Extension of the frontiers of surgical indications in the treatment of liver metastases from colorectal cancer: Long-term results. Ann Surg 2000; 231: 487–99. 19. Vauthey JN, Pawlik TM, Abdalla EK, et al. Is extended hepatectomy for hepatobiliary malignancy justified? Ann Surg 2004; 239: 722–32. 20. Scheele J, Stangl R, Altendorf-Hofmann A, Paul M. Resection of colorectal liver metastases. World J Surg 1995; 19: 59–71. 21. Pawlik TM, Scoggins CR, Zorzi D, et al. Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 2005; 241: 715–24. 22. Adam R, Delvart V, Pascal G, et al. Rescue surgery for unresectable colorectal liver metastases downstaged by chemotherapy: A model to predict long-term survival. Ann Surg 2004; 240: 644–57. 23. Petrowsky H, Gonen M, Jarnagin W, et al. Second liver resections are safe and effective treatment for recurrent hepatic metastases from colorectal cancer: A bi-institutional analysis. Ann Surg 2002; 235: 863–71. 24. Kanemitsu Y, Kato T, Hirai T, Yasui K. Preoperative probability model for predicting overall survival after resection of pulmonary metastases from colorectal cancer. Br J Surg 2004; 91: 112–20. 25. Pozzo C, Basso M, Cassano A, et al. Neoadjuvant treatment of unresectable liver disease with irinotecan and 5-fluorouracil plus folinic acid in colorectal cancer patients. Ann Oncol 2004; 15: 933–39. 26. Folprecht G, Grothey A, Alberts S, et al. Neoadjuvant chemotherapy of unresectable colorectal liver metastases: correlation between tumour response and resection rates. Ann Oncol 2005; 16: 1311–9. 27. Wicherts DA, de Haas RJ, Adam R. Bringing unresectable liver disease to resection with curative intent. Eur J Surg Oncol 2007; 33(S2): S42–S51. 28. Adam R, Aloia T, Levy F, et al. Hepatic resection after rescue cetuximab treatment for colorectal liver metastases previously refractory to conventional systemic chemotherapy. J Clin Oncol 2007; 25: 4593–602. 29. Gruenberger B, Tamandl D, Schueller J, et al. Bevacizumab, capecitabine and oxaliplatin as neoadjuvant treatment for patients with potentially curable metastatic colorectal cancer. J Clin Oncol 2008; 26: 1830–5. 30. Folprecht G, Gruenberger T, Hartmann JT, et al. Cetuximab plus FOLFOX6 or cetuximab plus FOLFIRI as neoadjuvant treatment of nonresectable colorectal liver metastases: a randomized multicenter study CELIM-study. ASCO GI, San Francisco 2009, abstract 296. 31. Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX-4 and surgery for respectable liver metastases from colorectal cancer. Lancet 2008; 371: 1007–16. 32. Feliberti EC, Wagman LD. Radiofrequency ablation of liver metastases from colorectal cancer. Cancer Control 2006; 13: 48–51. 33. Mulier S, Mulier P, Ni Y, et al. Complications of radiofrequency coagulation of liver tumors. Br J Surg 2002; 89: 1206–22. 34. Poston GJ, Byrne C. Decision making for patients with colorectal cancer liver metastases. Ann Surg Oncol 2006; 13: 10–1. 35. Verghese M, Pathak S, Poston GJ. Increasing long-term survival in advanced colorectal cancer. Eur J Surg Oncol 2007; 33(S2): S1–S4. 36. Nesbitt C, Glendinning RJ, Byrne C, Poston GJ. Factors influencing treatment strategies in advanced colorectal cancer. Eur J Surg Oncol 2007; 33(S2): S88–S94.

should the liver resection take precedence over the bowel surgery?
The fundamental question is now whether or not, having achieved a window of therapeutic opportunity to deal with the liver disease, does the liver disease takes precedence over the primary tumor (39,40)? It has been proposed that the liver disease should be resected first, and then following eradication of the liver disease, subsequently deal with the primary bowel tumor. Using this strategy, potentially curative surgery for both primary and secondary disease has been achieved in 16 of 20 (80%) such patients in small singlecenter series (38).

conclusions
If feasible, surgical resection remains the gold standard of treatment for CRLM. Unfortunately, patients still present with advanced colorectal cancer. Modern chemotherapy regimens offer increasing numbers of patients with initially unresectable CRLM the chance of being brought to potentially curative liver surgery (Fig. 15.2) (41). The remaining controversies in this field are the timing of such surgery and the strategic decisions of which operation (bowel first, liver first, or synchronous combined surgery) is now the first procedure of choice? The role of the MDT in the management of colorectal cancer liver metastases is to collate all the available data that can lead to an accurate assessment of disease spread and stage, then using these data, to plan an effective treatment strategy that ideally is focused on possible cure, but in any event is aimed at gaining maximal survival advantage for our patients.

references
1. Poston GJ. Surgical strategies for colorectal liver metastases. Surg Oncol 2004; 13: 125–36. 2. Primrose JN. Treatment of colorectal metastases: Surgery, cryotherapy or radiofrequency ablation. Gut 2002; 50: 1–5. 3. Weiss L, Grundmann E, Torhorst J, et al. Hematogenous metastatic patterns in colonic carcinoma: An analysis of 1541 necropsies. J Pathol 1986; 150: 195–203. 4. Sugarbaker PH. Surgical decision making for large bowel cancer metastatic to the liver. Radiol 1990; 174: 621–6. 5. Stangl R, Altendorf-Hofmann A, Charnley RM, Scheele J. Factors influencing the natural history of colorectal liver metastases. Lancet 1994; 343: 1405–10. 6. Geoghegan JG, Scheele J. Treatment of colorectal liver metastases. Br J Surg 1994; 86: 158–69. 7. Abdalla E, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation and combined resection/ablation for colorectal liver metastases. Ann Surg 2004; 239: 818–25. 8. Bismuth H, Adam R, Levy F, et al. Resection of nonresectable metastases from colorectal cancer after neoadjuvant chemotherapy. Ann Surg 1996; 224: 509–20. 9. Sahani DV, Kalva SP. Imaging the liver. Oncologist 2004; 9: 385–97. 10. McLoughlin JM, Jensen EH, Malafa M. Resection of colorectal liver metastases. Cancer Control 2006; 13: 32–41. 11. Vauthey JN. Patients with hepatic colorectal metastases. Program of the AHPBA 2006 consensus conference; January 25, 2006; San Francisco, CA.

152

MULTIMODAL APPROACHES TO THE MANAGEMENT OF COLORECTAL LIVER METASTASES
37. Weber JC, Bachellier P, Oussoultzoglou E, Jaeck D. Simultaneous resection of colorectal primary tumour and synchronous liver metastases. Br J Surg 2003; 90: 956–62. 38. Benoist S, Pautrat K, Mitry E, et al. Treatment strategy for patients with colorectal cancer and synchronous irresectable liver metastases. Br J Surg 2005; 92: 1155–60. 39. Mentha G, Majno PE, Andres A, et al. Neoadjuvant chemotherapy and resection of advanced synchronous liver metastases before treatment of the colorectal primary. Br J Surg 2006; 93: 872–8. 40. Mentha G, Majno P, Terraz S, et al. Treatment strategies for the management of advanced colorectal liver metastases detected synchronously with the primary tumour. Eur J Surg Oncol 2007; 33(S2): S76–S83. 41. Blokhius TJ, van der Schaaf MP, van den Tol MP, et al. Results of radiofrequency ablation of primary and secondary liver tumors: Longterm follow-up with computed tomography and positron-emission tomography-18F-deoxyglucose scanning. Scand J Gastroenterol 2004; 241: 93–7.

153

16

Management of neuroendocrine tumor hepatic metastasis Kaori Ito
Wessels et al. proposed the Carcinoid Symptom Severity Scale (Table 16.4) (16). This scaling system well illustrates the disease severity and is used to evaluate the symptomatic change before/after treatment (17). Carcinoid crisis is an acute life-threatening presentation, which is precipitated by anesthesia or interventional procedures (18,19). When large amounts of hormonal products are suddenly released into the systemic circulation, they trigger hypotension, tachyarrhythmias, bronchospasm, and neurological abnormalities. Carcinoid crisis is treated by the intravenous administration of somatostatin (SST) (50–100 μg). Premedication with SST analogs before interventional therapies can prevent crisis (19–23). Carcinoid-associated fibrosis should also be of concern. Carcinoid tumors cause fibrosis of the surrounding tissue. The pathogenesis of fibrosis is not well studied. Fibrosis in the peritoneum leads to bowel ischemia or mechanical bowel obstruction (24,25), in the retroperitoneum leads to hydronephrosis (24,26), in heart leads to tricuspid/pulmonary valve disease (carcinoid heart disease) (27), and in the thorax leads to thickening of pleura (28). These lesions can be the major cause of morbidity and mortality of the patients in the advanced stage. Laboratory Investigation Blood Patients with suggestive symptoms of NET should undergo laboratory tests to confirm the diagnosis of NET. Elevated plasma chromogranin A (CgA) level is the most sensitive marker of carcinoid tumors. CgA is a water-soluble acidic glycoprotein, which is stored in the secretory granules of NET cells. The sensitivity of plasma CgA level in NETs is reported up to 100%, however, it is not specific because the elevation of CgA is observed in prostate carcinoma too (1,29–31). Other biomarkers include bradykinin, serum substance P, neurotensin, human chorionic gonadotropin (hCG), neuropeptide K, and neuropeptide PP (1). Urine The measurement of 24-hour urinary 5-hydroxyindoleacetic acid (5-HIAA) can provide a summation of paroxysmal tumor secretion activity. 5-HIAA is a metabolite of serotonin, which is released by carcinoid tumors. The specificity of this test is around 90%. False-positive can will occur with consumption of serotonin-rich foods (bananas, avocados, plums, eggplant, tomatoes, plantain, pineapples, and walnuts) (32,33). If the laboratory findings were equivocal, a provocative test such as a pengastrin test (injection) or alcohol ingestion might be performed under the careful monitoring (31) (Grade III. Recommendation C).

introduction
Neuroendocrine tumors (NET), including both carcinoid tumors and islet tumors, are derived from primitive neuroectodermal cells that are distributed throughout the body during embryonic development (1–4). Therefore, NETs originate from various organs but most commonly involve the lungs, bronchi, and gastrointestinal tract (2,5). NETs were traditionally classified into foregut, midgut, and hindgut derivatives based on their presumed origin of gut. Currently, it is replaced by the WHO classification system of 2000 according to the histological differentiation (6). Clinical presentations widely differ depending on both their organ and excess hormone production (e.g., serotonin, histamine, tachykinins, and prostaglandins) (3,5). The overall incidence of NET has been reported to be 1 to 2 cases per 100,000 people (5) (Tables 16.1 and 16.2). Hepatic metastasis is the second common metastasis following lymph node metastasis in NETs (3) (Fig. 16.1). Up to 45% of patients with abdominal carcinoid will present with bowel obstruction and more than half of patients who were explored for bowel obstruction due to NETs are found to have hepatic metastases. Despite of the fact that liver is the common metastatic site of NETs, primary hepatic NET is extremely rare (0.6% of all NETs) (2). In this chapter, we describe specifically about the management of hepatic metastases of NETs.

diagnosis
Clinical Features Hepatic metastases of NET could be diagnosed preoperatively following investigation of a specific hormonal syndrome or following the incidental finding of hepatomegaly or an abdominal mass. Or it could be discovered at the abdominal exploration for primary gastrointestinal NETs. Besides hormonal symptoms, patients will complain local symptoms due to tumor bulk (pain, early satiety, or palpable mass). Subclinical hepatic metastasis does not require treatments, however, lifestyle-altering symptoms or biologically aggressive tumors require treatment (7). Demographics, presentation, symptoms, tumor histology, and primary tumor location of patients with NET-hepatic metastasis are summarized in Table 16.3 (8). The most representative symptom of patients with NETs is carcinoid syndrome. It is caused by systemic circulation of hormonal products from bulky metastatic NETs. This syndrome is a manifestation of late stage of NET and 5% to 10% of all NET patients present with this syndrome (1,5,9,10). In patients with NET-hepatic metastasis, carcinoid syndrome is frequently evident, at least biochemically (1). Common symptoms and signs include cutaneous flushing (71–80%), diarrhea (76–80%), hepatomegaly (71%), carcinoid heart disease (41–70%), asthma (9–25%), pellagra (2%) (3,11–15).

154

MANAGEMENT OF NEUROENDOCRINE TUMOR HEPATIC METASTASIS
Table 16.1 WHO Classification of Neuroendocrine Tumors (NET) of the Gastrointestinal Tract with Portal Venous Drainage
Stomach, ileum, colon Well-differentiated Endocrine Tumor (Carcinoid) (1) Benign behavior Non-functioning Confined to mucosa-submucosa, nonangioinvasive Size ≤1 cm (stomach or small intestine) or ≤2 cm (colon) (2) Uncertain behavior Nonfunctioning Confined to mucosa-submucosa, nonangioinvasive Size >1 cm (stomach or small intestine) or >2 cm (colon) Well-differentiated Endocrine Carcinoma (Malignant Carcinoid) Low-grade malignant tumor Deeply invasive (muscularis propria or beyond) or with metastases (liver) Poorly-differentiated endocrine carcinoma Small-cell carcinoma High-grade malignant tumor Mixed Endocrine/Exocrine Carcinoma Moderate to high-grade malignant tumor Pancreas Well-differentiated endocrine tumor (1) Benign behavior Confined to the pancreas, non-angioinvasive Size: <2 cm Mitosis: ≤2 Ki67 positive cells/10 HPF: ≤2% (2) Uncertain behavior Confined to the pancreas Size: ≥2 cm Mitosis: >2, or angioinvasive Ki67 positive cells/10 HPF: >2% Well differentiated Endocrine Carcinoma Functioning or non-functioning Low-grade malignant tumor with gross local invasion and/or metastases (liver) Ki67 positive cells/10 HPF: >5% Poorly-differentiated Endocrine Carcinoma Small cell carcinoma High-grade malignant tumor Ki67 positive cells/10 HPF: >15%

Source: Adapted from Ref. (6).

Table 16.2 Anatomical Location of NETs (Carcinoid Only, Except for Islet Cell Tumors) from SEER 1973–1999
% Lung, bronchi, and trachea Stomach Duodenum Jejunum Ileum Appendix Cecum Colon Rectum Other
Source: Adapted from Ref. (2).

28 5 3 2 15 5 4 5 14 6

the primary site. Sensitivity of CT/MRI is reported to be about 80%. The detection rates are 76% to 100% for CT alone and 67% to 81% for MRI alone (31). Triple phase spiral CT is the most informative to evaluate NET-hepatic metastasis. NEThepatic metastasis-associated findings are defined as mass lesions with calcification and radiating strands of fibrosis (36) (Fig. 16.3) (Grade III. Recommendation C). Positron Emission Tomography (PET) [18F]Fluoro-2-deoxy-d-glucose (FDG)-PET scan became an essential tool for many cancers to detect cancer cells, which were not seen in other imaging, or to quantify metastatic sites by the whole body image. The utility of 18F-PET scan for NETs is not well supported. Because typically NETs are slow growing with a low metabolic rate, the uptake of 18F by NETs cannot be visualized (37,38). The detection rates are reported as 25% to 73% (1,39). Instead, PET scan with the radioactive serotonin precursor 11C-5 HT, and 68Ga/64Cu coupled to octreotide revealed an excellent detection rate (40). Since there were no large studies to assess the efficacy of PET scan compared to other diagnostic imaging (41), the role of PET scan for NETs is still unclear (Grade III. Recommendation C).
Radiolabeled Metaiodobenzylguanidine (MIBG) Because NETs concentrate MIBG, the administration of 123 I-MIBG is another option to detect metastatic NETs. Sensitivity and specificity were reported as 55% to 70% and 95%, respectively (42,43). Although the accuracy of this test is inferior to that of SSTR scintigraphy, MIBG is a useful alternative of SSTR scintigraphy in patients on long-acting SST analog in whom SSTRs may have been occupied by SST analogue already (42,43) (Grade III. Recommendation C).

Diagnostic Imaging Somatostatin Receptor (SSTR) Scintigraphy Given the clinical presentation and biochemical confirmation of NETs, topographical localization of primary tumor and metastases should be pursued. SSTR scintigraphy utilizing 111 In-labelled SST analog (octreotide) can detect NETs that express SSTRs. Sensitivity of this test is reported as 84% (57– 93%) (1,34,35) (Fig. 16.2). The simultaneous use of single positron emission computed tomography (SPECT) can enhance the sensitivity. This is the first choice of diagnostic imaging test to find the primary site of carcinoid tumors (Grade III. Recommendation C). Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) CT and MRI are utilized to obtain more precise image of local extent of tumor if surgery is contemplated, but not to look for

155

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 16.1 Macroscopic (A) and microscopic (B) views of large resected hepatic neuroendocrine liver metastasis (stained for chromogranin-A). Note the hypervascularity of the metastasis compared to the background liver and comparable colorectal liver metastases.

Table 16.3 Clinical Features of NET-Hepatic Metastasis
Demographics Gender Men Women Age Metastatic presentation Synchronous Metachronous Symptoms Hormonal Pain Mechanical, progressive tumor growth Weight loss Jaundice Asymptomatic with progressive tumor growth Gastrointestinal bleeding Tumor histology Carcinoid Nonfunctional islet cell tumor Functional islet cell tumor Primary tumor location Pancreas Gastrointestinal bleeding Lung Unknown
Source: Adapted from Ref. (8).

% 40 60 52 years % 74 26 % 55 55 36 15 5 4 1 % 48 31 21 % 49 25 9 16

Hepatic Resection Surgical resection of NET-hepatic metastasis is categorized into curative intent resection or palliation intent resection (debulking/cytoreductive surgery). Both of approaches contribute to the improvement of symptoms and the prolonged survival. Curative intent resection is indicated for patients with solitary or localized hepatic metastasis. Only 10% to 25% of NET-hepatic metastasis is found in this category. Palliation intent resection is applied to patients with considerable symptom due to multiple, bulky extended tumor. The removal of more than 90% of the tumor bulk allows a significant palliation. Concurrent resection of extrahepatic tumors is often performed. Regional lymphadenectomy should be done because lymph node metastasis is common in NETs. Administration of SST analogs will be required for patients with residual tumors. Curative resection is associated with longer survival than noncurative resection (91% vs. 76% at 5 years; median 30–50 vs. 16–32 months, respectively). Summary of literature review regarding postoperative outcomes is shown in Table 16.5 (8,47–58). Postoperative morbidity and mortality rates are 22% (3-26%) and 3% (0–6%). Although perioperative carcinoid crisis is not very frequent (0–3%), precaution should be always taken. Relief of symptom was achieved in 92% (46–100) of patients. Disease-free survival is 17 to 60 months (median 41 months) and 36% (16–42%) at 5 years. Hepatic recurrence is reported in 82% of patients (52). Survivals have an estimated median of 67 (52–81) months. Five-year survival rate extends up to 73% (31–85%). Ten-year survival is reported as 35% (Grade III. Recommendation C).

treatment
Treatment of NET-hepatic metastasis is required for patients with lifestyle-altering symptoms, or biologically aggressive tumors. The principal requirements are to remove the primary and metastatic sites in order to reduce levels of bioactive agents (27,44–46). Therefore, surgical resection is the first choice as long as patients fit to surgery. Treatment options include hepatic resection, hepatic artery occlusion, radiofrequency ablation, cryoablation, liver transplantation, and medical therapy. Indications and timing of therapy are still controversial.

Hepatic Artery Embolization Hepatic artery embolization (HAE) is a rational approach against liver malignancies by using the discrepancy of blood supply between liver tumor and normal liver. The selective occlusion of hepatic artery causes hypoxic damage of tumor. Patients with NET-hepatic metastasis who don’t fit for surgery will benefit from this therapy. Indications include (i) rapid enlargement of tumor mass, (ii) increasing symptoms, and (iii) patient preference for the procedure in lieu of other treatment (46). Occlusive and/or chemotherapeutic agents are infused into the hepatic artery through an angiography catheter (5).

156

MANAGEMENT OF NEUROENDOCRINE TUMOR HEPATIC METASTASIS
Table 16.4 Carcinoid Symptom Severity Scale
Score 1 Description No symptoms Description Symptoms: None Frequency: None Lifestyle effects: None Symptoms: Diarrhea, flushing, or wheezing Frequency: Up to 4 times daily Lifestyle effects: None to minimal Symptoms: Diarrhea, flushing, or wheezing Frequency: 5–7 times weekly Lifestyle effects: Restricts patient from leaving home for prolonged periods of time. Symptoms: Diarrhea, flushing, or wheezing Frequency: Multiple daily episode Lifestyle effects: Symptoms require significant recognization of daily activities to accommodate for these symptoms; patients rarely leave home, must be close to bathroom facilities and medical supplies. Symptoms: Diarrhea, flushing, or wheezing (severe) or of sufficient severity to warrant hospitalization for treatment of dehydration, electrolyte imbalance, refractory hypertension, or asthma Frequency: Numerous (>4) daily Lifestyle effects: Symptoms are disabling; patients are unable to leave home or require hospitalization.

2

Mild symptoms

3

Symptoms impact daily living Sever symptoms

4

5

Disabling symptoms

Source: Adapted from Ref. (16).

Figure 16.3 Computed tomography scan of multiple neuroendocrine hepatic metastases from a primary small bowel carcinoid. Figure 16.2 Indium-111 Octreotide scan demonstrating octreotide avid liver metastases.

A diagnostic angiography should be obtained from a femoral approach to confirm the anatomy of artery and the patency of portal vein before the administration of the therapeutic agents (Fig. 16.4). Patients should receive SST analogs before the procedure to prevent hormonal adverse events (59). There are no definitive data to support the agents for embolization such as Gelfoam, Ivalon, starch particles, lipidol, or radio isotope-loaded spheres. Selection of chemotherapeutic agents is also still inconclusive. Cisplatin, doxorubicin, and mitomycin are most commonly utilized. Almost all of patients experience postprocedural abdominal pain, nausea, vomiting, and fever. Transaminase levels will shoot up dramatically, and then followed by the elevation of

alkaline phosphatase and or serum bilirubin. Tumor-related hormone level will increase temporarily, however, it can be obviated with SST analogs administration (59). Major complications include gastrointestinal bleeding, gastric and duodenal ulceration, hepatic abscess, ischemic necrosis of gallbladder or small intestine, pancreatitis, sepsis, renal failure, hepatorenal syndrome, portal vein thrombosis, sclerosing cholangitis, arterial thrombosis, and arrythmias (60). Since these complications are common and even can be fatal, patient selection should be strict and postprocedural hospital stay with careful monitoring is warranted. Overall, morbidity rate ranges from 3% to 20%, mortality rate ranges from 0% to 7% (Table 16.6). Successful symptomatic relief and the reduction of tumor size can be achieved; however, the duration of palliation may

157

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 16.5 Hepatic Resection (Literature review: 1996–2008)
Follow-up period Morbidity Mortality (months) (%) (%) 6–108 24 6 Relief of symptoms (%) 100 Disease free survival (%) 36% at 5 years Disease free survival (months) 17 Disease specific survival (%) 46% at 5 years Disease specific survival (months) NA

Year Dousset (Paris, France) (47) 1996

No. of patients 17

Chen (MD, USA) (48) Chamberlain (NY, USA) (8) Jaeck (Strasbourg, France) (49) Yao (IL, USA) (50) Chung (CA, USA) (51) Sarmiento (MN, USA) (52) Knox (TN, USA) (53) Mazzaferro (Milan, Italy) (54) Osborne (FL, USA) (55) Musunuru (WI, USA) (56) Landry (KT, USA) (112) Eriksson (Uppsala, Sweden) (58) Median

1998 2000 2001 2001 2001 2003 2004 2007 2006 2006 2008 2008

15 34 13 16 31 170 17 36 61 13 23 42

27 27 42 30 26 NA NA NA NA 20 NA 18 27

NA NA NA 12 26 14 24 NA 3 NA 26 20 22

0 6 0 0 3 1 0 0 2 NA 0 0 3

NA 86–100 46 100 90 96 82 NA 92 100 NA 71 92

NA NA 69% at 3 years 42% at 5 years NA 16% at 5 years NA 19% at 10 years NA NA NA NA 36% at 5 years

21 NA NA

60 46 9–104 NA 35 50 NA NA 41

73% at 5 years 76% at 5 years 68% at 6 years 73% at 5 years 31% at 5 years 61% at 5 years 85% at 5 years 59% at 10 years NA 83% at 3 years 75% at 5 years NA 73% at 5 years

not reached not reached NA not reached NA 81 135 NA 43 NA 52 NA 67

be limited due to recurrence or rearterization of tumors. The occlusive agent alone is associated with a relief of symptom rate 49% (33–100%), median disease-free survival at 15 (8–37) months, and median disease-specific survival at 24 (24–120) months (55,56,61–68). Whereas the embolization with chemotherapeutic agents results in a relief of symptom rate 75% (61–92%), median disease-free survival is 14 (10–19) months, and median disease-specific survival 33 (25–49) months. The summary of recent literature review is shown in Table 16.7 (Grade III. Recommendation C). Hepatic Radiofrequency Ablation Radiofrequency ablation (RFA) can be performed percutaneously or laparoscopically for patients who are unfit for hepatic resection or intraoperative RFA in addition to hepatic resection is also advantageous. High-frequency alternating current causes ionic agitation that is converted into heat, and leads coagulation necrosis of tumor. The probes can deploy 4 to 5 cm in a single session with up to 200 W of power (Fig. 16.5).

Figure 16.4 Typical angiogram of multiple neuroendocrine hepatic metastases prior to embolization.

158

Table 16.6 Hepatic Artery Occlusion (Literature Review: 1989–2008)

Year 1989 1998 1999 2002 2006 2007 3 0 49 38% at 1 year NA NA NA 15 55 35 24 59 15 Gelfoam Polyvinyl alcohol particles Lipiodol/Gelfoam Polyvinyl alcohol particles Trisacryl gelatin microsphere (embosphere) NA 17 0 5 0 0 6 0 0 0 38–52 89–100 64 59 33 8 15 NA 37 NA 8 Gelfoam NA 0 38 NA

No. of patients Agent(s) used

Morbidity (%)

Mortality (%)

Relief of symptoms (%) Disease free survival (%) Disease specific survival (%)

Disease free survival (months)

Disease specific survival (months) NA 80–120 24 NA 24 NA

Occlusive agent alone (HAE) Nobin (Lund, Sweden) (63) 38% at 1 year NA NA NA NA NA 72

Eriksson (Uppsala, Sweden) (64) Brown (NY, USA) (113) Schell (FL, USA) (17) Osborne (FL, USA) (55) Granberg (Uppsala, Sweden) (65)

13% died in 5 months 83% at 5 years 54% at 5 years 72% at 5 years NA NA

Median

24

HAE with systemic chemotherapy Moertel (MN, USA) (68) Loewe (Vienna, Austria) (67) 1994 2003 12 7 98 111 23 NA, including surgical ligation N-Butyl-2-cyanoacrylate, Lipiodol 12 NA 5 9 98.0 NA

NA NA NA

NA 65% at 5 years 65% at 5 years

49 39 44

Median Combination with chemotherapeutic agents (HACE) Ruszniewski (Paris, France) (114) 1993 Drougas (TN, USA) (60) 1998 24 15 14 46 122 23 14 20 0 4 5 8 60 0 0 NA 75 64 78 92 2003 2007 2007

NA NA NA NA 5% at 5 years

13 NA 17 14 10

NA NA 83% at 5 years 29% at 5 years 28% at 5 years

NA 25 48 33 33

Roche (Villejuif, France) (115) Ho (MO, USA) (116)

Bloomston (OH, USA) (117)

Christante (OR, USA) (118)

2008

59

Doxorubicin, Lipiodol Doxorubicin, cisplatin, mitomycin C +5FU Doxorubicin, Lipiodol Cisplatin, doxorubicin, mitomycin, Lipiodol Cisplatin, doxorubicin, mitomycin, ioxaglate sodium, Lipiodol Cisplatin, doxorubicin, mitomycin NA 20 7 5 9 0

61 75 NA

NA 5% at 5 years NA

19 14 23

27% at 5 years 29% at 5 years NA

39 33 34

MANAGEMENT OF NEUROENDOCRINE TUMOR HEPATIC METASTASIS

Median HAE and HACE Gupta (TX, USA) (66) 2005 123

Musunuru (WI, USA) (56)

2006

18

HAE, polyvinyl alcohol particle or gelfoam powder. HACE, for carcinoid tumor: cisplatin+doxorubicin. For islet cell tumor: 5FU+streptozocin NA

NA 9

NA 0

83 83

25 24

NA NA

NA 34

Median

31% at 3 years 31% at 3 years

159

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 16.7 Milan Criteria
Inclusion criteria 1. Confirmed histology of carcinoid tumor (low-grade neuroendocrine tumors) with or without syndrome 2. Primary tumor drained by the portal system (pancreas and intermediate gut: from distal stomach to sigmoid colon) removed with a curative resection (pre-transplant removal of all extra-hepatic tumor deposits) through surgical procedures different and separate from transplantation 3. Metastatic diffusion to liver parenchyma < or + 50% 4. Good response or stable disease for at least 6 months during the pre-transplant period 5. Age < or + 55 years Exclusion criteria 1. Small-cell carcinoma and high-grade neuroendocrine carcinomas (non-carcinoid tumors) 2. Other medical/surgical conditions contraindicationg liver transplantation, including previous tumors 3. Non-gastrointestinal carcinoids or tumors not-drained by the portal system
Source: Adapted from Ref. (54).

study by Mazzaglia et al. (69) detailed the outcomes of 80 laparoscopic RFA sessions in 63 patients who underwent RFA alone for NET-hepatic metastasis (54). Relief of symptoms was achieved more than 90% of the patients. Median diseasefree survival was 11 months. Median disease-specific survival was close to 4 years. Five-year survival rate was 48% (Grade III. Recommendation C). Hepatic Cryoablation Cryoablation can be applied for patients with unresectable refractory NETs. Intraoperative approach combined with hepatic resection is common rather than cryoablation alone. The cryoprobe is inserted into tumor under ultrasound guidance. The freezing temperature of cryoprobe is maintained liquid nitrogen perfusing in the uninsulated tip. Tumor is monitored until the “ice ball” enveloped the tumor with a 1-cm margin of normal tissue. Multiple freezing–thaw cycles lead to tumor destruction (73). Indications for this procedure are still unclear. Complications include coagulopathy, bleeding, acute renal failure, and pulmonary embolism. Morbidity rate is reported variously but at a minimum of 23%. Mortality rate is 0% to 2% (74–76). Of note, this procedure is usually combined with hepatic resection; therefore, the outcomes of reported studies are not specific for cryoablation. Almost all patients experienced the relief of symptoms and biochemical response. Local recurrence at the ablated site is reported as 17% in a study from Seifert et al. (75). Median recurrence-free survival is 10 months. Median disease-specific survival is 20 to 49 months. Three-year survival rate is up to 91% (74–76) (Grade III. Recommendation C). Liver Transplantation Liver transplantation is a therapeutic alternative of hepatic resection for unresectable NET-HM patients. Whereas the results of liver transplantation for other metastatic tumors are poor (77), patients with NET-hepatic metastasis have been more likely benefit from liver transplantation (57,78–85). Although this approach is still controversial, “Milan criteria” for indication to liver transplantation in patients with NEThepatic metastasis are widely referred (Table 16.7) (54). Patients will receive a full graft, a split graft or a domino graft from deceased or living donor (78). The general principle of complete resection of both primary and metastatic tumor has to be pursued in the setting of this treatment. Therefore, transplantation could be performed with concurrent resection of extrahepatic tumor including lymphadenectomy of hepatic pedicle and hepato-duodenal ligament (54,78,86). Standard immunosuppression should be administered postoperatively. Adjuvant chemotherapy or long-acting SST analogs will be applied as appropriately (54). Table 16.8 shows the summary of literature review. Postoperative morbidity includes acute rejection episodes, acute cholangitis, and bacteremia. Overall morbidity rate are reported as 56% (32–75%). Mortality rate is 10% (5–44%). Recent studies report that 5-year survival of 21% (36–90%), with symptomatic relief occurring in all of the patients.

Figure 16.5 CT guided radiofrequency ablation of single small liver metastasis 1 day after arterial embolization.

This treatment is suitable for relatively small tumors. Indications for RFA include (i) fewer than 4 in number, (ii) smaller than 5 cm, (iii) accessible location in liver, and (iv) not in contiguity to vascular structures, bowel, or the gall bladder (46). Complications include bleeding, sepsis, and intrahepatic biliary duct damage. Morbidity rate is around 5%. No RFArelated mortality has been reported (58,69). RFA is associated with the high incidence of the recurrence at previously ablated sites (58,70,71). Local recurrence rate is reported to be 5 to 6% (69,70,72). The assessment of outcome of this procedure is somewhat difficult to be specific because many of RFA are combined with surgical resection. A large

160

MANAGEMENT OF NEUROENDOCRINE TUMOR HEPATIC METASTASIS
Table 16.8 Liver Transplantation (Literature Review, 1996–2008)
FollowDiseaseup Relief of free period Morbidity Mortality symptoms survival (months) (%) (%) (%) (%) NA NA 44 NA Rec rate 11% NA NA 8 100 Rec rate 57.1% 60 NA 14 NA 24% at 5 years 60 NA 5 NA 21% at 5 years 34 NA 27 NA 9% at 5 years 22 32 5 NA 80% at 1 years 60 NA NA NA 77% at 5 years 54 NA 7 NA Rec rate 33.3% 36 75 10 NA 57% at 3 years 46 NA 14 NA 20% at 5 years 50 56 10 100 21% at 5 years Disease specific survival (%) NA NA 47% at 5 years 80% at 5 years 36% at 5 years 87% at 1 years 90% at 5 years NA 57% at 3 years 47% at 5 years 47% at 5 years Disease specific survival (months) 24 55 NA NA NA NA NA NA NA 56 55

Dousset (Paris, France) (47) Lang (Gottingen, Germany) (57) Lehnert(Heidelberg, Germany) (87) Rosenau (Hannover, Germany) (88) Florman (NY, USA) (89) V.Vilsteren (MN, USA) (86) Mazzaferro (Milan, Italy) (54) Olausson (Goteborg, Sweden) (90) Marin (Murcia, Spain) (91) Le Treut (Marseille, France) (78) Median

Year 1996 1997 1998 2002 2004 2006 2007 2007 2007 2008

No. of patients 9 12 103 19 11 19 24 15 10 85

43.795 pt 58 6 weeks to 48 months NA 11 NA NA NA 25 NA NA 25

Disease-free survival is 21% (9–77%) at 5 years. Median time to recurrence is 25 (1.5–58) months (47,54,57,78,86–91) (Grade III. Recommendation C). Medical Treatment SSTR-targeted Therapy SST analogs are effective in improving hormonal symptoms due to NETs. SST inhibits the release of serotonin and other hormones from NETs (92). Because SST has a short half-life (about 2 minutes), it is not suitable for clinical use (92). Longacting SST analogues (octreotide and lanreotide) are widely applied. The response rate ranges 70% to 80% when administered subcutaneously every 6 to 12 hours (93,94). Dosage should be adjusted with clinical use from 50 to 500 μg 3 times a day. Adverse effects include gallstones, steatorrhea, sinus bradycardia, cardiac conduction abnormalities, arrhythmias, hypothyroidism, hypoglycemia, and hyperglycemia (92,95). SSTR analogs are utilized for preoperative symptomatic control, preprocedural medication to prevent carcinoid crisis, and postoperative supportive therapy if residual tumors were evident (1) (Grade III. Recommendation C). Chemotherapy Chemotherapeutic agents for NETs include streptozotocin, 5-FU, doxorubicin, cyclophosphamide, etoposide, cisplatin, temozolomide, thalidomide, paclitaxel, and docetaxel (1,4). Overall response rate of chemotherapeutic alone is reported to be only 20% to 40%. At least there is one randomized trial comparing streptozocin +5FU and doxorubicin +5FU (96). The patients were enrolled this study were 249. Response rate

in two groups were similar (16% vs. 15.9%). Streptozocin + 5 FU was associated a subtle increase of survival (24.3 vs. 15.7 months), however, renal toxicity was significantly frequent in that group. Unfortunately, there are no data existing to reveal the benefit of each chemotherapeutic agent, or the combination of agents (92,97–99) (Grade Ib. Recommendation B). Interferon Interferon inhibits tumor growth by directly blocking the G0/G1 phase of cell cycle. Applications of interferon to NETs have been investigated since 1982 (100–103). Interferon alpha alone resulted in biochemical response rate of 7% to 66%, and tumor response rate 0% to 25%. The combination of interferon alpha and SST analogs failed to be effective (104) (Grade III. Recommendation C). Radionuclide Therapy Receptor targeted therapy with radionuclides is an emerging treatment for patients with disseminated NET metastases. 131 I-MIBG, [111In-DTPA-D-Phe] octreotide, 90yttrium, and 177 lutetium-labeled SST analogs are utilized (35,105–111). Agents will be selected by uptake at diagnostic imaging. This treatment is specific and tolerated. Fair levels of biochemical response and volume reduction are reported. Symptomatic relief can be achieved. Reported adverse effect is renal damage. Adequate renal protection should be added before treatment (Grade III. Recommendation C).

summary
Summary of treatment for NET-hepatic metastasis is shown in Table 16.9.

161

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 16.9 Summary of Treatment for NET-Hepatic Metastasis
Relief of symptoms (%) 92 Diseasefree survival (%) 36% at 5 years Diseasefree survival (months) 41 Diseasespecific survival (%) 73% at 5 years Diseasespecific survival (months) 67

Indication Hepatic resection (8,47–56,58,112) Hepatic artery occlusion HAE (17,55, 63–65,113) HACE (60,114–118) Tumor restricted to one lobe (1) Rapid enlargement of tumor mass, (2) Increasing symptoms, (3) Patient preference for the procedure in lieu of other treatment, (4) Patients with adequate liver function and patent portal vein (1) Fewer than 4 in number, (2) Smaller than 5cm, (3) Accessible location in liver, and (4) Not in contiguity to vascular structures, bowel, or the gall bladder Small lesions. Usually combined with hepatic resection Milan criteria. See Table 16.7.

Morbidity (%) 22

Mortality (%) 3

3 20

0 5

49 75

38% at 1 year 5% at 5 years

15 14

72% at 5 years 29% at 5 years

24 33

Hepatic radiofrequency ablation (58,69,119,120)

5

0

82

NA

11

NA

48

Hepatic cryotherapy (74–76) Liver transplantation (47,54,57,78,86–91)

23

0

95

32% at 3 years 21% at 5 years

10

91% at 3 years 47% at 5 years

45

56

10

100

25

55

key points
Diagnosis:








Hormonal symptoms (carcinoid syndrome) and/or symptoms due to hepatic mass Laboratory investigations: CgA (blood) and 5-HIAA (urine) Identify the primary and metastatic sites by SSTR scintigraphy Assess the resectability of hepatic metastasis by CT or MRI

Treatment:
● ●



● ●



Surgical resection (curative or palliative) Other liver targeted therapy (HAE/HACE, RFA, and cryotherapy) Liver transplantation for selected unresectable patients SST analogues for symptomatic control Radionuclide therapy is emerging for the disseminated disease No proven survival benefit in chemotherapy

references
1. Modlin IM, Kidd M, Latich I, et al. Current status of gastrointestinal carcinoids. Gastroenterology 2005; 128(6): 1717–51. 2. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer 2003; 97(4): 934–59. 3. Moertel CG. Karnofsky memorial lecture. An odyssey in the land of small tumors. J Clin Oncol 1987; 5(10): 1502–22. 4. Kvols LK. Revisiting C.G. Moertel’s land of small tumors. J Clin Oncol 2008; 26(31): 5005–7.

5. Kulke MH, Mayer RJ. Carcinoid tumors. N Engl J Med 1999; 340(11): 858–68. 6. Solcia E, Capella C, Kloppel G, Heitz PU, Sobin LH, Rosai J. Endocrine tumours of the gastrointestinal tract. In: Solcia E, Kloppel G, Sobin LH, eds. Histologic typing of endocrine tumours. WHO international histological typing of endocrine tumours. Heidelberg, New York: Springer Verlag, 2000: 57–67. 7. DeMatteo RP, Fong Y, Blumgart LH. Surgical treatment of malignant liver tumours. Baillieres Best Pract Res Clin Gastroenterol 1999; 13(4): 557–74. 8. Chamberlain RS, Canes D, Brown KT, et al. Hepatic neuroendocrine metastases: does intervention alter outcomes? J Am Coll Surg 2000; 190(4): 432–45. 9. Moertel CG, Sauer WG, Dockerty MB, et al. Life history of the carcinoid tumor of the small intestine. Cancer 1961; 14: 901–12. 10. Burke AP, Thomas RM, Elsayed AM, et al. Carcinoids of the jejunum and ileum: an immunohistochemical and clinicopathologic study of 167 cases. Cancer 1997; 79(6): 1086–93. 11. Grahame-Smith DG. The carcinoid syndrome. Am J Cardiol 1968; 21(3): 376–87. 12. Feldman JM. Carcinoid tumors and the carcinoid syndrome. Curr Probl Surg 1989; 26(12): 835–85. 13. Tilson MD. Carcinoid syndrome. Surg Clin North Am 1974; 54(2): 409–23. 14. Lundin L. Carcinoid heart disease. A cardiologist’s viewpoint. Acta Oncol 1991; 30(4): 499–502. 15. Berge T, Linell F. Carcinoid tumours. Frequency in a defined population during a 12-year period. Acta Pathol Microbiol Scand [A] 1976; 84(4): 322–30. 16. Wessels FJ, Schell SR. Radiofrequency ablation treatment of refractory carcinoid hepatic metastases. J Surg Res 2001; 95(1): 8–12. 17. Schell SR, Camp ER, Caridi JG, et al. Hepatic artery embolization for control of symptoms, octreotide requirements, and tumor progression in metastatic carcinoid tumors. J Gastrointest Surg 2002; 6(5): 664–70. 18. Debas HT, Mulvihill SJ. Neuroendocrine gut neoplasms. Important lessons from uncommon tumors. Arch Surg 1994; 129(9): 965–71; discussion 971–2. 19. Kinney MA, Warner ME, Nagorney DM, et al. Perianaesthetic risks and outcomes of abdominal surgery for metastatic carcinoid tumours. Br J Anaesth 2001; 87(3): 447–52.

162

MANAGEMENT OF NEUROENDOCRINE TUMOR HEPATIC METASTASIS
20. Kvols LK, Martin JK, Marsh HM, et al. Rapid reversal of carcinoid crisis with a somatostatin analogue. N Engl J Med 1985; 313(19): 1229–30. 21. Marsh HM, Martin JK, Jr., Kvols LK, et al. Carcinoid crisis during anesthesia: successful treatment with a somatostatin analogue. Anesthesiology 1987; 66(1): 89–91. 22. Muralidhar V, Sharma A. Carcinoid crisis during a partial hepatic resection; lack of essential drugs: a cause for concern in the tropics. Trop Gastroenterol 1996; 17(1): 26–9. 23. Dery R. Theoretical and clinical considerations in anaesthesia for secreting carcinoid tumors. Can Anaesth Soc J 1971; 18(3): 245–63. 24. Warner TF, O’Reilly G, Lee GA. Mesenteric occlusive lesion and ileal carcinoids. Cancer 1979; 44(2): 758–62. 25. Marshall JB, Bodnarchuk G. Carcinoid tumors of the gut. Our experience over three decades and review of the literature. J Clin Gastroenterol 1993; 16(2): 123–9. 26. Sakai D, Murakami M, Kawazoe K, et al. Ileal carcinoid tumor complicating carcinoid heart disease and secondary retroperitoneal fibrosis. Pathol Int 2000; 50(5): 404–11. 27. Norheim I, Oberg K, Theodorsson-Norheim E, et al. Malignant carcinoid tumors. An analysis of 103 patients with regard to tumor localization, hormone production, and survival. Ann Surg 1987; 206(2): 115–25. 28. Moss SF, Lehner PJ, Gilbey SG, et al. Pleural involvement in the carcinoid syndrome. Q J Med 1993; 86(1): 49–53. 29. O’Connor DT, Deftos LJ. Secretion of chromogranin A by peptideproducing endocrine neoplasms. N Engl J Med 1986; 314(18): 1145–51. 30. Nobels FR, Kwekkeboom DJ, Coopmans W, et al. Chromogranin A as serum marker for neuroendocrine neoplasia: comparison with neuronspecific enolase and the alpha-subunit of glycoprotein hormones. J Clin Endocrinol Metab 1997; 82(8): 2622–8. 31. Modlin IM, Tang LH. Approaches to the diagnosis of gut neuroendocrine tumors: the last word (today). Gastroenterology 1997; 112(2): 583–90. 32. Tormey WP, FitzGerald RJ. The clinical and laboratory correlates of an increased urinary 5-hydroxyindoleacetic acid. Postgrad Med J 1995; 71(839): 542–5. 33. Feldman JM, Lee EM. Serotonin content of foods: effect on urinary excretion of 5-hydroxyindoleacetic acid. Am J Clin Nutr 1985; 42(4): 639–43. 34. Raderer M, Kurtaran A, Leimer M, et al. Value of peptide receptor scintigraphy using (123)I-vasoactive intestinal peptide and (111)In-DTPA-DPhe1-octreotide in 194 carcinoid patients: Vienna University Experience, 1993 to 1998. J Clin Oncol 2000; 18(6): 1331–6. 35. Hoefnagel CA. Metaiodobenzylguanidine and somatostatin in oncology: role in the management of neural crest tumours. Eur J Nucl Med 1994;21(6):561–81. 36. Paulson EK, McDermott VG, Keogan MT, et al. Carcinoid metastases to the liver: role of triple-phase helical CT. Radiology 1998; 206(1): 143–50. 37. Eriksson B, Lilja A, Ahlstrom H, et al. Positron-emission tomography as a radiological technique in neuroendocrine gastrointestinal tumors. Ann N Y Acad Sci 1994; 733: 446–52. 38. Rege SD, Hoh CK, Glaspy JA, et al. Imaging of pulmonary mass lesions with whole-body positron emission tomography and fluorodeoxyglucose. Cancer 1993; 72(1): 82–90. 39. Jadvar H, Segall GM. False-negative fluorine-18-FDG PET in metastatic carcinoid. J Nucl Med 1997; 38(9): 1382–3. 40. Sundin A, Eriksson B, Bergstrom M, et al. PET in the diagnosis of neuroendocrine tumors. Ann N Y Acad Sci 2004; 1014: 246–57. 41. Adams S, Baum RP, Hertel A, et al. Metabolic (PET) and receptor (SPET) imaging of well- and less well-differentiated tumours: comparison with the expression of the Ki-67 antigen. Nucl Med Commun 1998; 19(7): 641–7. 42. Krenning EP, Kooij PP, Bakker WH, et al. Radiotherapy with a radiolabeled somatostatin analogue, [111In-DTPA-D-Phe1]-octreotide. A case history. Ann N Y Acad Sci 1994; 733: 496–506. 43. Hanson MW, Feldman JM, Blinder RA, et al. Carcinoid tumors: iodine-131 MIBG scintigraphy. Radiology 1989; 172(3): 699–703. 44. McEntee GP, Nagorney DM, Kvols LK, et al. Cytoreductive hepatic surgery for neuroendocrine tumors. Surgery 1990; 108(6): 1091–6. 45. Martin JK, Jr, Moertel CG, Adson MA, et al. Surgical treatment of functioning metastatic carcinoid tumors. Arch Surg 1983; 118(5): 537–42. 46. Garrot C, Stuart K. Liver-directed therapies for metastatic neuroendocrine tumors. Hematol Oncol Clin North Am 2007; 21(3): 545–60; ix–x. 47. Dousset B, Saint-Marc O, Pitre J, et al. Metastatic endocrine tumors: medical treatment, surgical resection, or liver transplantation. World J Surg 1996; 20(7): 908–14; discussion 914–5. 48. Chen H, Hardacre JM, Uzar A, et al. Isolated liver metastases from neuroendocrine tumors: does resection prolong survival? J Am Coll Surg 1998; 187(1): 88–92; discussion 92–3. 49. Jaeck D, Oussoultzoglou E, Bachellier P, et al. Hepatic metastases of gastroenteropancreatic neuroendocrine tumors: safe hepatic surgery. World J Surg 2001; 25(6): 689–92. 50. Yao KA, Talamonti MS, Nemcek A, et al. Indications and results of liver resection and hepatic chemoembolization for metastatic gastrointestinal neuroendocrine tumors. Surgery 2001; 130(4): 677–82; discussion 682–5. 51. Chung MH, Pisegna J, Spirt M, et al. Hepatic cytoreduction followed by a novel long-acting somatostatin analog: a paradigm for intractable neuroendocrine tumors metastatic to the liver. Surgery 2001; 130(6): 954–62. 52. Sarmiento JM, Heywood G, Rubin J, et al. Surgical treatment of neuroendocrine metastases to the liver: a plea for resection to increase survival. J Am Coll Surg 2003; 197(1): 29–37. 53. Knox CD, Feurer ID, Wise PE, et al. Survival and functional quality of life after resection for hepatic carcinoid metastasis. J Gastrointest Surg 2004; 8(6): 653–9. 54. Mazzaferro V, Pulvirenti A, Coppa J. Neuroendocrine tumors metastatic to the liver: how to select patients for liver transplantation? J Hepatol 2007; 47(4): 460–6. 55. Osborne DA, Zervos EE, Strosberg J, et al. Improved outcome with cytoreduction versus embolization for symptomatic hepatic metastases of carcinoid and neuroendocrine tumors. Ann Surg Oncol 2006; 13(4): 572–81. 56. Musunuru S, Chen H, Rajpal S, et al. Metastatic neuroendocrine hepatic tumors: resection improves survival. Arch Surg 2006; 141(10): 1000–4; discussion 1005. 57. Lang H, Oldhafer KJ, Weimann A, et al. Liver transplantation for metastatic neuroendocrine tumors. Ann Surg 1997; 225(4): 347–54. 58. Eriksson J, Stalberg P, Nilsson A, et al. Surgery and radiofrequency ablation for treatment of liver metastases from midgut and foregut carcinoids and endocrine pancreatic tumors. World J Surg 2008; 32(5): 930–8. 59. Caplin ME, Buscombe JR, Hilson AJ, et al. Carcinoid tumour. Lancet 1998; 352(9130): 799–805. 60. Drougas JG, Anthony LB, Blair TK, et al. Hepatic artery chemoembolization for management of patients with advanced metastatic carcinoid tumors. Am J Surg 1998; 175(5): 408–12. 61. Brown KT, Nevins AB, Getrajdman GI, et al. Particle embolization for hepatocellular carcinoma. J Vasc Interv Radiol 1998; 9(5): 822–8. 62. Lauffer JM, Zhang T, Modlin IM. Review article: current status of gastrointestinal carcinoids. Aliment Pharmacol Ther 1999; 13(3): 271–87. 63. Nobin A, Mansson B, Lunderquist A. Evaluation of temporary liver dearterialization and embolization in patients with metastatic carcinoid tumour. Acta Oncol 1989; 28(3): 419–24. 64. Eriksson BK, Larsson EG, Skogseid BM, et al. Liver embolizations of patients with malignant neuroendocrine gastrointestinal tumors. Cancer 1998; 83(11): 2293–301. 65. Granberg D, Eriksson LG, Welin S, et al. Liver embolization with trisacryl gelatin microspheres (embosphere) in patients with neuroendocrine tumors. Acta Radiol 2007; 48(2): 180–5. 66. Gupta S, Johnson MM, Murthy R, et al. Hepatic arterial embolization and chemoembolization for the treatment of patients with metastatic neuroendocrine tumors: variables affecting response rates and survival. Cancer 2005; 104(8): 1590–602. 67. Loewe C, Schindl M, Cejna M, et al. Permanent transarterial embolization of neuroendocrine metastases of the liver using cyanoacrylate and lipiodol: assessment of mid- and long-term results. AJR Am J Roentgenol 2003; 180(5): 1379–84. 68. Moertel CG, Johnson CM, McKusick MA, et al. The management of patients with advanced carcinoid tumors and islet cell carcinomas. Ann Intern Med 1994; 120(4): 302–9. 69. Mazzaglia PJ, Berber E, Milas M, et al. Laparoscopic radiofrequency ablation of neuroendocrine liver metastases: a 10-year experience evaluating predictors of survival. Surgery 2007; 142(1): 10–19. 70. Berber E, Siperstein A. Local recurrence after laparoscopic radiofrequency ablation of liver tumors: an analysis of 1032 tumors. Ann Surg Oncol 2008; 15(10): 2757–64.

163

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
71. Ahmad A, Chen SL, Kavanagh MA, et al. Radiofrequency ablation of hepatic metastases from colorectal cancer: are newer generation probes better? Am Surg 2006; 72(10): 875–9. 72. Hellman P, Lundstrom T, Ohrvall U, et al. Effect of surgery on the outcome of midgut carcinoid disease with lymph node and liver metastases. World J Surg 2002; 26(8): 991–7. 73. Siperstein AE, Berber E. Cryoablation, percutaneous alcohol injection, and radiofrequency ablation for treatment of neuroendocrine liver metastases. World J Surg 2001; 25(6): 693–6. 74. Bilchik AJ, Sarantou T, Foshag LJ, et al. Cryosurgical palliation of metastatic neuroendocrine tumors resistant to conventional therapy. Surgery 1997; 122(6): 1040–7; discussion 1047–8. 75. Seifert JK, Cozzi PJ, Morris DL. Cryotherapy for neuroendocrine liver metastases. Semin Surg Oncol 1998; 14(2): 175–83. 76. Goering JD, Mahvi DM, Niederhuber JE, et al. Cryoablation and liver resection for noncolorectal liver metastases. Am J Surg 2002; 183(4): 384–9. 77. Penn I. Hepatic transplantation for primary and metastatic cancers of the liver. Surgery 1991; 110(4): 726-34; discussion 734–5. 78. Le Treut YP, Gregoire E, Belghiti J, et al. Predictors of long-term survival after liver transplantation for metastatic endocrine tumors: an 85-case French multicentric report. Am J Transplant 2008; 8(6): 1205–13. 79. Makowka L, Tzakis AG, Mazzaferro V, et al. Transplantation of the liver for metastatic endocrine tumors of the intestine and pancreas. Surg Gynecol Obstet 1989; 168(2): 107–11. 80. Arnold JC, O’Grady JG, Bird GL, et al. Liver transplantation for primary and secondary hepatic apudomas. Br J Surg 1989; 76(3): 248–9. 81. Routley D, Ramage JK, McPeake J, et al. Orthotopic liver transplantation in the treatment of metastatic neuroendocrine tumors of the liver. Liver Transpl Surg 1995; 1(2): 118–21. 82. Le Treut YP, Delpero JR, Dousset B, et al. Results of liver transplantation in the treatment of metastatic neuroendocrine tumors. A 31-case French multicentric report. Ann Surg 1997; 225(4): 355–64. 83. Ihse I, Persson B, Tibblin S. Neuroendocrine metastases of the liver. World J Surg 1995; 19(1): 76–82. 84. McDermott EW, Guduric B, Brennan MF. Prognostic variables in patients with gastrointestinal carcinoid tumours. Br J Surg 1994; 81(7): 1007–9. 85. Frilling A, Rogiers X, Malago M, et al. Liver transplantation in patients with liver metastases of neuroendocrine tumors. Transplant Proc 1998; 30(7): 3298–300. 86. van Vilsteren FG, Baskin-Bey ES, Nagorney DM, et al. Liver transplantation for gastroenteropancreatic neuroendocrine cancers: Defining selection criteria to improve survival. Liver Transpl 2006; 12(3): 448–56. 87. Lehnert T. Liver transplantation for metastatic neuroendocrine carcinoma: an analysis of 103 patients. Transplantation 1998; 66(10): 1307–12. 88. Rosenau J, Bahr MJ, von Wasielewski R, et al. Ki67, E-cadherin, and p53 as prognostic indicators of long-term outcome after liver transplantation for metastatic neuroendocrine tumors. Transplantation 2002; 73(3): 386–94. 89. Florman S, Toure B, Kim L, et al. Liver transplantation for neuroendocrine tumors. J Gastrointest Surg 2004; 8(2): 208–12. 90. Olausson M, Friman S, Herlenius G, et al. Orthotopic liver or multivisceral transplantation as treatment of metastatic neuroendocrine tumors. Liver Transpl 2007; 13(3): 327–33. 91. Marin C, Robles R, Fernandez JA, et al. Role of liver transplantation in the management of unresectable neuroendocrine liver metastases. Transplant Proc 2007; 39(7): 2302–3. 92. Oberg K, Kvols L, Caplin M, et al. Consensus report on the use of somatostatin analogs for the management of neuroendocrine tumors of the gastroenteropancreatic system. Ann Oncol 2004; 15(6): 966–73. 93. Welin SV, Janson ET, Sundin A, et al. High-dose treatment with a longacting somatostatin analogue in patients with advanced midgut carcinoid tumours. Eur J Endocrinol 2004; 151(1): 107–12. 94. Garland J, Buscombe JR, Bouvier C, et al. Sandostatin LAR (long-acting octreotide acetate) for malignant carcinoid syndrome: a 3-year experience. Aliment Pharmacol Ther 2003; 17(3): 437–44. 95. Schnirer, II, Yao JC, Ajani JA. Carcinoid--a comprehensive review. Acta Oncol 2003; 42(7): 672–92. 96. Sun W, Lipsitz S, Catalano P, et al. Phase II/III study of doxorubicin with fluorouracil compared with streptozocin with fluorouracil or dacarbazine in the treatment of advanced carcinoid tumors: Eastern Cooperative Oncology Group Study E1281. J Clin Oncol 2005; 23(22): 4897–904. 97. Saltz L, Kemeny N, Schwartz G, et al. A phase II trial of alpha-interferon and 5-fluorouracil in patients with advanced carcinoid and islet cell tumors. Cancer 1994; 74(3): 958–61. 98. Moertel CG, Hanley JA. Combination chemotherapy trials in metastatic carcinoid tumor and the malignant carcinoid syndrome. Cancer Clin Trials 1979; 2(4): 327–34. 99. Hughes MJ, Kerr DJ, Cassidy J, et al. A pilot study of combination therapy with interferon-alpha-2a and 5-fluorouracil in metastatic carcinoid and malignant endocrine pancreatic tumours. Ann Oncol 1996; 7(2): 208–10. 100. Oberg K. Neuroendocrine gastrointestinal tumours. Ann Oncol 1996; 7(5): 453–63. 101. Dirix LY, Vermeulen PB, Fierens H, et al. Long-term results of continuous treatment with recombinant interferon-alpha in patients with metastatic carcinoid tumors—an antiangiogenic effect? Anticancer Drugs 1996; 7(2): 175–81. 102. Di Bartolomeo M, Bajetta E, Zilembo N, et al. Treatment of carcinoid syndrome with recombinant interferon alpha-2a. Acta Oncol 1993; 32(2): 235–8. 103. Jacobsen MB, Hanssen LE, Kolmannskog F, et al. Interferon-alpha 2b, with or without prior hepatic artery embolization: clinical response and survival in mid-gut carcinoid patients. The Norwegian carcinoid study. Scand J Gastroenterol 1995; 30(8): 789–96. 104. Janson ET, Oberg K. Long-term management of the carcinoid syndrome. Treatment with octreotide alone and in combination with alpha-interferon. Acta Oncol 1993; 32(2): 225–9. 105. Waldherr C, Pless M, Maecke HR, et al. The clinical value of [90Y-DOTA]-d-Phe1-Tyr3-octreotide (90Y-DOTATOC) in the treatment of neuroendocrine tumours: a clinical phase II study. Ann Oncol 2001; 12(7): 941–5. 106. McCarthy KE, Woltering EA, Espenan GD, et al. In situ radiotherapy with 111In-pentetreotide: initial observations and future directions. Cancer J Sci Am 1998; 4(2): 94–102. 107. De Jong M, Breeman WA, Bernard HF, et al. Therapy of neuroendocrine tumors with radiolabeled somatostatin-analogues. Q J Nucl Med 1999; 43(4): 356–66. 108. Anthony LB, Woltering EA, Espenan GD, et al. Indium-111-pentetreotide prolongs survival in gastroenteropancreatic malignancies. Semin Nucl Med 2002; 32(2): 123–32. 109. Buscombe JR, Caplin ME, Hilson AJ. Long-term efficacy of high-activity 111 In-pentetreotide therapy in patients with disseminated neuroendocrine tumors. J Nucl Med 2003; 44(1): 1–6. 110. Kwekkeboom DJ, Bakker WH, Kam BL, et al. Treatment of patients with gastro-entero-pancreatic (GEP) tumours with the novel radiolabelled somatostatin analogue [177Lu-DOTA(0), Tyr3]octreotate. Eur J Nucl Med Mol Imaging 2003; 30(3): 417–22. 111. Mukherjee JJ, Kaltsas GA, Islam N, et al. Treatment of metastatic carcinoid tumours, phaeochromocytoma, paraganglioma and medullary carcinoma of the thyroid with (131)I-meta-iodobenzylguanidine [(131) I-mIBG]. Clin Endocrinol (Oxf) 2001; 55(1): 47–60. 112. Landry CS, Scoggins CR, McMasters KM, et al. Management of hepatic metastasis of gastrointestinal carcinoid tumors. J Surg Oncol 2008; 97(3): 253–8. 113. Brown KT, Koh BY, Brody LA, et al. Particle embolization of hepatic neuroendocrine metastases for control of pain and hormonal symptoms. J Vasc Interv Radiol 1999; 10(4): 397–403. 114. Ruszniewski P, Rougier P, Roche A, et al. Hepatic arterial chemoembolization in patients with liver metastases of endocrine tumors. A prospective phase II study in 24 patients. Cancer 1993; 71(8): 2624–30. 115. Roche A, Girish BV, de Baere T, et al. Trans-catheter arterial chemoembolization as first-line treatment for hepatic metastases from endocrine tumors. Eur Radiol 2003; 13(1): 136–40. 116. Ho AS, Picus J, Darcy MD, et al. Long-term outcome after chemoembolization and embolization of hepatic metastatic lesions from neuroendocrine tumors. AJR Am J Roentgenol 2007; 188(5): 1201–7.

164

MANAGEMENT OF NEUROENDOCRINE TUMOR HEPATIC METASTASIS
117. Bloomston M, Al-Saif O, Klemanski D, et al. Hepatic artery chemoembolization in 122 patients with metastatic carcinoid tumor: lessons learned. J Gastrointest Surg 2007; 11(3): 264–71. 118. Christante D, Pommier S, Givi B, et al. Hepatic artery chemoinfusion with chemoembolization for neuroendocrine cancer with progressive hepatic metastases despite octreotide therapy. Surgery 2008; 144(6): 885–93; discussion 893–4. 119. Berber E, Tsinberg M, Tellioglu G, et al. Resection versus laparoscopic radiofrequency thermal ablation of solitary colorectal liver metastasis. J Gastrointest Surg 2008; 12(11): 1967–72. 120. Hellman P, Ladjevardi S, Skogseid B, et al. Radiofrequency tissue ablation using cooled tip for liver metastases of endocrine tumors. World J Surg 2002; 26(8): 1052–6.

165

17

Noncolorectal, nonneuroendocrine metastases C. Kahlert, R. DeMatteo, and J.Weitz
gynecological tumors
The most common gynecological types of cancer comprise of ovarian cancer, endometrial cancer, and cervical cancer. Ovarian cancer is the leading cause of tumor-related death among gynecological malignancies. Epithelial tumors are the most frequent, followed by sarcomas, germ cell and stromal tumors (10). These tumors commonly metastasize to the peritoneal cavity and lymph nodes, however, liver metastases are also a common site of systemic spread (10). Since there is an inverse correlation between the volume of the residual tumor and the overall patient survival, resection of liver metastases should be performed with optimal cytoreduction of extrahepatic lesions (11,12). By this, median overall survival can be prolonged significantly (11). Despite a significant decline of the incidence of uterine cancer for the last 70 years, it still remains the fourth most frequent tumor among women (1). Endometrial cancer usually metastasizes by lymphatic vessels. The most common sites for hematogenous dissemination are bone, lung, and liver (13). Recently, a multicenter study by Adam et al. reported that resection of uterine liver metastases resulted in a 5-year survival of 35% and a median overall survival of 32 months. Similar results were observed in smaller study by Kollmar et al. (14). Cervical cancer, similar to uterine cancer, spreads most frequently via the lymphatic system, whereas hematogenous dissemination is a rare event. This may explain why liver metastases occur only in 1.2% to 2.2% of the patients (15,16). Furthermore, in many cases, liver metastases are accompanied by uncontrolled locoregional disease or extrahepatic lesions (15). Only in 0.3% to 5% of cases, isolated liver metastases are found. Up to date, only a few case reports report data regarding liver resection in hepatic metastasized cervical cancer (17–19), proving the feasibility of hepatectomy in this tumor stage, but still lacking profound data regarding the medical benefit. In summary, as for other noncolorectal, nonneuroendocrine liver metastases, liver resection for hepatic lesions of gynecological tumors can be accomplished safely. For evaluating the benefit of a surgical intervention, more data should be acquired by further clinical studies. Despite the lack of more data, liver resection is offered to carefully selected patients with secondary liver disease of these types of cancer.

introduction
Approximately 90% of malignant hepatic lesions are metastases of extrahepatic primary tumors. Despite enormous progress of multimodal therapeutical options, surgical resection remains the only option for curative treatment in most of these cases. However, in contrast to colorectal or neuroendocrine hepatic metastases, the surgical approach for noncolorectal and nonneuroendocrine hepatic metastases is highly controversial. From a critical point of view, it is argued that noncolorectal, nonneuroendocrine liver metastases often belong to very aggressive types of cancer. In addition, they partly derive from extraabdominal primary tumors. In contrast to intraabdominal tumors metastasizing through the portal vein with the liver theoretically being the first filter organ, liver metastases of extraabdominal primary tumors imply a simultaneous systemic spread of tumor cells. However, resection of hepatic tumors can be accomplished safely with an appropriate risk of perioperative mortality and morbidity, and patients with favorable tumor biology might benefit from a surgical approach. Therefore, proper selected patients should be offered resection of noncolorectal, nonneuroendocrine liver metastases. Since noncolorectal, nonneuroendocrine hepatic metastases encompass a heterogeneous group of primary tumors, the management of these metastases needs to be discussed individually for each primary tumor type.

breast cancer
Breast cancer is the most frequent malignant tumor and the second most common cause of cancer death in women (1). Patients with breast cancer rarely present with isolated liver metastases, in only 10% to 20% of metastatic breast cancer, metastases are restricted solely to the liver (2,3) (Fig. 17.1). Therefore, the risk of systemic tumor relapse after removal of liver metastases is high and ought to be accounted before a surgical liver resection is contemplated. Retrospective studies report a median survival between 36 and 63 months (4–6) ( Table 17.1) when patients underwent surgical treatment in addition to systemic chemotherapy. In contrast, in patients treated with chemotherapy alone, the median overall survival will rarely exceed two years (7). Predictive risk factors, which should be considered for selecting appropriate patients, are lymph node status, extensive hepatic lesions requiring a major resection (8), recurrence of liver metastases within one year after resection of the primary tumor (9), R2 resection, and failure to respond to preoperative chemotherapy (4). Applying these criteria on patients with breast cancer and isolated hepatic metastases enables to select a subset of patients where liver resection may improve progression-free and overall survival compared to systemic treatment alone.

renal cell cancer
Since 1950, the incidence of renal cell cancer (RCC) has more than doubled. Based on advances in renal imaging, improved staging, and refinement in surgical technique the 5-year survival has notably improved (20). The clinical outcome in early-stage RCC is excellent with a high probability of complete remission. However, since renal cell carcinoma is characterized by high resistance to chemotherapy, the median

166

NONCOLORECTAL, NONNEUROENDOCRINE METASTASES
overall survival in metastatic disease remains still unsatisfactory. Even by introduction of immunotherapy and targeted therapy, median overall survival reaches approximately only two years (21), making systemic therapy a merely palliative approach. Resection of isolated liver metastases (4,22) (Fig.17.2) as well as other distant metastases of renal cell cancer (23,24) resulted in 5-year survival rates between 38% and 88%. Though the results of these retrospective studies are certainly influenced by a selection bias, they should be used to evaluate carefully whether in patients with metastatic renal cancer a complete surgical resection of the distant disease is possible. These patients will most likely benefit from a surgical approach if all disease can be resected. Whether the results of surgical resection can be further improved by a multimodal therapeutic regimen including immunotherapy and moleculary targeted therapy needs to be further evaluated.

Figure 17.1 Solitary liver metastases originating from breast cancer.

Table 17.1 Selected Retrospective Studies Reporting Clinical Outcome in Patients after Resection of Liver Metastases from Breast Cancer
Number of patients included 12 454 31 Median overall survival (months) 35.9 45 63

pancreatic cancer
Pancreatic cancer is one of the most aggressive tumors. It is the fourth leading cause of cancer death in females and the fifth leading cause in males worldwide (1). When pancreatic cancer is first diagnosed, the majority of patients are not amenable to surgical treatment according to the established standard criteria. Approximately only 15% to 25% of patients are eligible to curative operation procedures. One of the most frequent exclusion criteria for a curative surgical intervention is the presence of distant metastases, namely, liver metastases. In metastasized pancreatic cancer, palliative systemic chemotherapy is considered to provide the best therapeutical option. Yet, in many cases, R0 resection of liver metastases in addition to resection of the primary tumor would be technically feasible. By palliative systemic chemotherapy, median overall survival reaches approximately 6 months (25–27). The impact of a curative-intent surgical intervention is still unanswered. In several retrospective studies of resection of liver metastases of pancreatic cancer, the median overall survival ranges from 6 months to 20 months (4,19,28–37) (Table 17.2). In single cases, single patients have even survived longer than 5 years (30). Though, despite some encouraging results, the decision for the resection of pancreatic cancer liver metastases should be made on an individual basis where the patient is aware of a nonstandard treatment approach. Currently, resection of liver metastases is highly controversial and certainly far from being accepted by the medical community.

Author Caralt et al. (5) Adam et al. (4) Vlastos et al. (6)

Year 2008 2006 2004

5-year survival 33% 41% 61%

Figure 17.2 Solitary liver metastases originating from renal cancer.

gastric cancer
Table 17.2 Selected Retrospective Studies Reporting About Clinical Outcome in Patients after Resection of Liver Metastases from Pancreatic Cancer
Number of patients included 17 10 40 Median overall survival 5.9 11.4 20 5-year survival Not published Not published 25%

Author

Year

Gleisner et al. (29) 2007 Shrikande et al. (28) 2006 Adam et al. (4) 2006

The incidence of gastric cancer has steadily decreased in Europe and the United States, however, it still remains the second most common cancer worldwide. For locally advanced gastric cancer, the overall survival has been improved by treating patients with perioperative chemotherapy (38). However, for gastric adenocarcinoma with distant metastases, overall survival is still not favorable. The presence of liver metastases (Fig. 17.3) is generally considered to define a noncurative state of the disease. Patients treated by palliative chemotherapy survive approximately 9 to 10 months on average

167

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
in two retrospective multicenter studies with an appropriate number of patients (4,49). In general, the histological type of these tumors was squamous cell carcinoma. The clinical outcome after hepatic resection was unfavorable with median overall survival of 16 and 18 months, respectively, and a 5-year survival below 20%. This may lead to the conclusion that for hepatic metastasized tumors of the lungs and the head and neck region, nonsurgical interventions should be favored for management of these patients. However, bearing in mind that surgical resection is the only hope for cure, patients should not be categorically excluded, but carefully evaluated for a surgical approach as an individual curative attempt.

sarcoma
For surgical evaluation, liver metastases originating from sarcomas can be divided into two subtypes: gastrointestinal stromal tumors and non-GIST sarcomas. Gastrointestinal stromal tumors emerge most often in the stomach, second most in the small bowl and in the colon and most rarely in the duodenum, the esophagus or nonintestinal organs (50). Almost half of the patients suffer from distant metastases and in more than 50%, the metastatic disease is isolated to the liver (51). One decade ago, a large retrospective study regarding hepatic resection for GIST metastatic to the liver reported about a median overall survival of 39 months and 5-year survival of 30% (52). Since the introduction of imatinib mesylate in the therapy of GIST (53), the oncological management has changed in favor of a multimodal treatment, improving the patient outcomes significantly. In recent retrospective studies including patients treated with imatinib mesylate, the 5-year survival was 70% (4) and median overall survival was not reached despite long periods of followup (54) (Table 17.4). However, DeMatteo et al. and Gronchi et al. observed in two retrospective studies that mainly patients with metastatic GIST responding to a preoperative tyrosine kinase inhibitor therapy profit by a surgical approach whereas nonresponder do not seem to benefit by tumor resection (55,56). These data should be taken into account when selecting appropriate patients for surgery with liver metastases of gastrointestinal stromal tumors. In summary, gastrointestinal stromal tumors metastatic to the liver require interdisciplinary therapy regimens. Besides to surgical resection, this should involve application of new medical agents such as imatinib or radiofrequency ablation for small lesions not accessible to surgical resection (54,57). Non-GIST sarcomas have a worse prognosis than gastrointestinal stromal tumors. While extremity and trunk soft tissue sarcomas most frequently metastasize to the lung, primary visceral and retroperitoneal sarcomas often disseminate to the liver (51) (Fig. 17.4). After surgical resection of liver metastases, patients have a median overall survival of 32 to 37 months and 5-year survival probability of 27% to 32% (4,54) (Table 17.4). These data demonstrate that most patients with hepatic metastases of sarcomas will succumb to their disease. Patients with a disease-free interval exceeding two years, however, seem to have a better prognosis after hepatic resection (52). These data again point to the fact that selected patients should be offered

Figure 17.3 Solitary liver metastases originating from gastric cancer.

Table 17.3 Selected Retrospective Studies Reporting about Clinical Outcome in Patients after Resection of Liver Metastases from Gastric Cancer
Number Median of patients overall included survival 42 64 40 34 15 12 5-year survival Not published 27% 18%

Author Koga et al. (41) Adam et al. (4) Ambiru et al. (46)

Year 2007 2006 2001

(39,40). Retrospective studies, reporting outcome after hepatic resection as treatment for liver metastases of gastric adenocarcinoma, estimate the overall survival between 19 and 34 months (41–45) (Table 17.3). These numbers exceed the outcome of patients having merely received systemic chemotherapy. As each of these studies have included only a small number of patients probably affected by a selection bias, the data are still insufficient. To select patients with liver metastases of gastric cancer who might benefit from a surgical approach, different strategies have been implemented. Summarizing the available literature, resection of hepatic metastases seems to be associated with a better survival if solitary or metachronous lesions are being resected. As in many other tumor types, patient selection therefore seems to be the most important factor ensuring a benefit for the patients undergoing liver resection for liver metastases of gastric cancer.

respiratory tract⁄head and neck tumors
Both tumors deriving from the lungs and from the head and neck regions are usually associated with poor prognosis in the presence of distant metastases. Therefore, albeit tumors from these sites often form liver metastases (46–48), only recently the impact of surgical resection has been documented

168

NONCOLORECTAL, NONNEUROENDOCRINE METASTASES
Table 17.4 Selected Retrospective Studies Reporting about Clinical Outcome in Patients after Resection of Liver Metastases from GIST- And Non-GIST Sarcoma
Author Adam et al. (4) Pawlik et al. (54) DeMatteo et al. (52) Year 2006 2006 2001 Histological type GIST Non-GIST sarcoma GIST and leiomyosarcoma Non-GIST sarcoma GIST Non-GIST sarcoma Number of patients included 33 125 54 12 34 22 Median overall survival (months) Not reached 32 Not reached 37 39* 5-year survival 70% 31% Ca. 50% No survivor after 5 years 30%*

*Including patients with GIST- and non-GIST sarcoma.

Figure 17.4 Solitary liver metastases originating from a non-GIST-sarcoma.

resection of liver metastases, as they most likely will benefit from this treatment.

melanoma
Melanomas belong to the most frequent types of tumors with an increasing incidence over the last 30 years. Of the melanomas, 90% derive from the skin, 5% have an ocular origin, and 5% develop at other sites (58). Noteworthy, depending on the primary site, melanomas display a different metastasizing pattern. Cutaneous melanomas disseminate to the liver only in 15% to 20% of patients with metastatic disease (Fig. 17.5). This often happens with simultaneous metastatic decay of other organs (59). By contrast, in 40% of patients with liver metastases from uveal melanoma, the liver is the only site of the disease (60). Therefore, the number of liver metastases resected from uveal melanoma is nearly equivalent to that of cutaneous melanoma. While excision of early stage melanoma results in an excellent prognosis, chemotherapy achieves barely a median overall survival of 12 months in disseminated tumor stage (60,61). Hence, surgical resection of metastases offers the only chance for cure. However, patients amenable to a surgical intervention at the liver account for approximately only 2% to 3% of all patients representing with liver metastases of melanoma (62,63). In these cases, the median overall survival is estimated to be between 19 and 28 months and the median 5-year survival reaches 20% (4,49,62) (Table 17.5). This may justify the

Figure 17.5 Solitary liver metastases originating from cutaneous melanoma.

resection of liver metastases from cutaneous or uveal melanoma in individual patients.

predictive factors determining clinical outcome
For a successful preoperative assessment, several predictive factors can be taken into account to decide whether a patient with noncolorectal, nonneuroendocrine hepatic metastases might benefit from surgical resection. Several studies examined factors associated with improved survival after resection of noncolorectal, nonneuroendocrine liver metastases. In the study of Weitz et al. (19), 141 patients undergoing hepatic resection between April 1981 and April 2002 were analyzed, length of disease-free interval, primary tumor type, and completeness of resection were independent prognostic factors regarding cancer-specific survival (Table 17.6). Adam et al. analyzed the outcome of 1452 patients from 41 centers undergoing hepatectomy between 1983 and 2004. The majority of primary tumors were breast cancer (32%), gastrointestinal cancers (16%), and urologic cancers (14%). Five-year overall survival for the entire cohort of patients was

169

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 17.5 Selected Retrospective Studies Reporting About Clinical Outcome in Patients After Resection of Liver Metastases from Cutaneous and Ocular Melanoma
Author Pawlik et al. (49) Herman et al. (62) Year 2006 2001 Site of melanoma Cutaneous Ocular Cutaneous Number of patients included 24 16 5 Median overall survival (months) 24 29 22* 5-year survival No survivor 20% 70%†

*Including patients with both cutaneous and ocular melanoma. † Within a mean follow-up of 25.4 months.

Table 17.6 Analysis of Prognostic Factors for Cancer-Specific Survival
Univariate Median CSS No. Gender Age (years) Presentation Disease-free interval Primary minor Male Female ≤50 >50 Synchronous (1) Metachronous ≤24 mo >24 mo Adrenocortical Breast Gastrointestinal Reproductive tract Melanoma Renal Other Unknown 48 93 55 86 39 102 71 70 15 29 12 (mo) 52 40 42 41 37 48 34 52 40 48 21 p-Value NS NS NS 0.04 1.4(1.0–1.8) Reference 0.7(0.4–1.3) 1.0(0.6–1.7) 0.8 (0.3–1.5) 0.03 Multivariate Hazard Ratio (95% CI) p-Value

0.0!

0.02

Primary tumor

Prior metastases* Extrahepatic disease Size Number Distribution Margin status

Resection type Blood transfusion Postoperative complies (1) Lions

Reproductive tract Nonreproductive tract Yes No Yes No ≤5 cm >5 cm I >l Unilobar Bi lobar RO Rl R2 Minor major* Yes No Yes No

39 17 II 13 5 39 102 24 117 41 100 88 53 88 53 100 41 116 19 6 65 76 79 62 46 95

115 17 48 32 11 115 36

0.4 (0.2–0.7) 1.5(0.7–2.7) 0.7 (0.3–1.3) 1.7(0.3–1.3) Reference 0.02

48 40 42 46 42 37 49 34 46 40 49 17 10 40 52 52 37 40 49

NS NS NS NS NS <0.0I Reference 2.1 (1.1–4.1) 2.7 (0.8–7.9) <0.01

NS NS NS

Source: From Ref. (19). *Major liver resection: resection of 3 or more liver segments. Prior extrahepatic metastases. Presentation of the liver metastases: the same time as the primary tumor. CSS indicates cancer-specific survival; CI, confidence interval; NS, not significant.

170

NONCOLORECTAL, NONNEUROENDOCRINE METASTASES
36% and the 10-year overall survival of 23%. The following independent adverse prognostic factors could be identified: patient age over 60 years, nonbreast origin, melanoma or squamous histology, disease-free interval of less than 12 months, extrahepatic disease, incomplete macroscopic resection, and major hepatectomy. The authors went on to stratify patients according to the number of adverse prognostic factors present and stratified them into low-risk patients (0–3 points, 5-year survival: 46%), mid-risk patients (4–6 points, 5-year survival: 33%), and high-risk patients (>6 points, 5-year survival of less than 10%). This model may help to guide the decision regarding the best approach to patients with noncolorectal, nonneuroendocrine liver metastases.
11. Bristow RE, Montz FJ, Lagasse LD, Leuchter RS, Karlan BY. Survival impact of surgical cytoreduction in stage IV epithelial ovarian cancer. Gynecol Oncol 1999; 72: 278–87. 12. Lim MC, Kang S, Lee KS, et al. The clinical significance of hepatic parenchymal metastasis in patients with primary epithelial ovarian cancer. Gynecol Oncol 2009; 112: 28–34. 13. Aalders JG, Abeler V, Kolstad P. Recurrent adenocarcinoma of the endometrium: a clinical and histopathological study of 379 patients. Gynecol Oncol 1984; 17: 85–103. 14. Kollmar O, Moussavian MR, Richter S, et al. Surgery of liver metastasis in gynecological cancer - indication and results. Onkologie 2008;31: 375–9. 15. Kim GE, Lee SW, Suh CO, et al. Hepatic metastases from carcinoma of the uterine cervix. Gynecol Oncol 1998; 70: 56–60. 16. Carlson V, Delclos L, Fletcher GH. Distant metastases in squamous-cell carcinoma of the uterine cervix. Radiology 1967; 88: 961–6. 17. Kaseki H, Yasui K, Niwa K, et al. Hepatic resection for metastatic squamous cell carcinoma from the uterine cervix. Gynecol Oncol 1992; 44: 284–7. 18. Wolf RF, Goodnight JE, Krag DE, et al. Results of resection and proposed guidelines for patient selection in instances of non-colorectal hepatic metastases. Surg Gynecol Obstet 1991; 173: 454–60. 19. Weitz J, Blumgart LH, Fong Y, et al. Partial hepatectomy for metastases from noncolorectal, nonneuroendocrine carcinoma. Ann Surg 2005; 241: 269–76. 20. Pantuck AJ, Zisman A, Belldegrun AS. The changing natural history of renal cell carcinoma. J Urol 2001; 166: 1611–23. 21. Motzer RJ, Bukowski RM, Figlin RA, et al. Prognostic nomogram for sunitinib in patients with metastatic renal cell carcinoma. Cancer 2008; 113: 1552–8. 22. Thelen A, Jonas S, Benckert C, et al. Liver resection for metastases from renal cell carcinoma. World J Surg 2007; 31: 802–7. 23. Piltz S, Meimarakis G, Wichmann MW, et al. Long-term results after pulmonary resection of renal cell carcinoma metastases. Ann Thorac Surg 2002; 73: 1082–7. 24. Zerbi A, Ortolano E, Balzano G, et al. Pancreatic metastasis from renal cell carcinoma: which patients benefit from surgical resection? Ann Surg Oncol 2008; 15: 1161–8. 25. Moore MJ, Goldstein D, Hamm J, et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007; 25: 1960–6. 26. Heinemann V, Quietzsch D, Gieseler F, et al. Randomized phase III trial of gemcitabine plus cisplatin compared with gemcitabine alone in advanced pancreatic cancer. J Clin Oncol 2006; 24: 3946–52. 27. Louvet C, Labianca R, Hammel P, et al. Gemcitabine in combination with oxaliplatin compared with gemcitabine alone in locally advanced or metastatic pancreatic cancer: results of a GERCOR and GISCAD phase III trial. J Clin Oncol 2005; 23: 3509–16. 28. Shrikhande V, Kleeff J, Reiser C, et al. Pancreatic resection for M1 pancreatic ductal adenocarcinoma. Ann Surg Oncol 2007; 14: 118–27. 29. Gleisner AL, Assumpcao L, Cameron JL, et al. Is resection of periampullary or pancreatic adenocarcinoma with synchronous hepatic metastasis justified? Cancer 2007; 110: 2484–92. 30. Yamada H, Hirano S, Tanaka E, et al. Surgical treatment of liver metastases from pancreatic cancer. HPB (Oxford) 2006; 8: 85–8. 31. Karavias DD, Tepetes K, Karatzas T, et al. Liver resection for metastatic non-colorectal non-neuroendocrine hepatic neoplasms. Eur J Surg Oncol 2002; 28: 135–9. 32. Laurent C, Rullier E, Feyler A, et al. Resection of noncolorectal and nonneuroendocrine liver metastases: late metastases are the only chance of cure. World J Surg 2001; 25: 1532–6. 33. Benevento A, Boni L, Frediani L, et al. Result of liver resection as treatment for metastases from noncolorectal cancer. J Surg Oncol 2000; 74: 24–9. 34. Lang H, Nussbaum KT, Weimann A, et al. [Liver resection for non-colorectal, non-neuroendocrine hepatic metastases]. Chirurg 1999; 70: 439–46. 35. Takada T, Yasuda H, Amano H, et al. Simultaneous hepatic resection with pancreato-duodenectomy for metastatic pancreatic head carcinoma: does it improve survival? Hepatogastroenterology 1997; 44: 567–73.

summary
Resection of noncolorectal, nonneuroendocrine liver metastases is associated with an improved progression-free and overall survival in a selected subgroup of patients. However, until now, these data have been mainly obtained by retrospective studies and probably are affected by selection bias, as patients with lower performance status and poorer prognosis are less likely to have undergone surgery. Due to these limitations, a conclusion regarding a direct comparison to nonsurgical approaches cannot be drawn. Each individual case needs to be carefully assessed prior to a decision regarding a surgical approach. Furthermore, to reduce the risks of postoperative morbidity and mortality, it is recommendable to perform the surgical intervention on the liver at high-volume centers (64).

references
1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008; 58: 71–96. 2. Atalay G, Biganzoli L, Renard F, et al. Clinical outcome of breast cancer patients with liver metastases alone in the anthracycline-taxane era: a retrospective analysis of two prospective, randomised metastatic breast cancer trials. Eur J Cancer 2003; 39: 2439–49. 3. Er O, Frye K, Kau Sd CW, et al. Clinical course of breast cancer patients with metastases limited to the liver treated with chemotherapy. Cancer J 2008; 14: 62–68. 4. Adam R, Chiche L, Aloia T, et al. Hepatic resection for noncolorectal nonendocrine liver metastases: analysis of 1,452 patients and development of a prognostic model. Ann Surg 2006; 244: 524–35. 5. Caralt M, Bilbao I, Cortes J, et al. Hepatic resection for liver metastases as part of the “oncosurgical” treatment of metastatic breast cancer. Ann Surg Oncol 2008; 15: 2804–10. 6. Vlastos G, Smith D L, Singletary S E, et al. Long-term survival after an aggressive surgical approach in patients with breast cancer hepatic metastases. Ann Surg Oncol 2004; 11: 869–74. 7. Alba E, Martin M, Ramos M, et al. Multicenter randomized trial comparing sequential with concomitant administration of doxorubicin and docetaxel as first-line treatment of metastatic breast cancer: a Spanish Breast Cancer Research Group (GEICAM-9903) phase III study. J Clin Oncol 2004; 22: 2587–93. 8. Pocard M, Pouillart P, Asselain B, et al. [Hepatic resection for breast cancer metastases: results and prognosis (65 cases)]. Ann Chir 2001; 126: 413–20. 9. Selzner M, Morse M A, Vredenburgh J J, et al. Liver metastases from breast cancer: long-term survival after curative resection. Surgery 2000; 127: 383–9. 10. Rose PG, Piver MS, Tsukada Y. Metastatic patterns in histologic variants of ovarian cancer. An autopsy study. Cancer 1989; 64: 1508–13.

171

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
36. Howard JM. Pancreatoduodenectomy (Whipple resection) with resection of hepatic metastases for carcinoma of the exocrine pancreas. Arch Surg 1997; 132: 1044. 37. Klempnauer J, Ridder GJ, Piso P, et al. [Is liver resection in metastases of exocrine pancreatic carcinoma justified?]. Chirurg 1996; 67: 366–70. 38. Cunningham D, Allum WH, Stenning S P, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006; 355: 11–20. 39. Cunningham D, Starling N, Rao S, et al. Capecitabine and oxaliplatin for advanced esophagogastric cancer. N Engl J Med 2008; 358: 36–46. 40. Van Cutsem E, Moiseyenko VM, Tjulandin S, et al. Phase III study of docetaxel and cisplatin plus fluorouracil compared with cisplatin and fluorouracil as first-line therapy for advanced gastric cancer: a report of the V325 Study Group. J Clin Oncol 2006; 24: 4991–7. 41. Koga R, Yamamoto J, Ohyama S, et al. Liver resection for metastatic gastric cancer: experience with 42 patients including eight long-term survivors. Jpn J Clin Oncol 2007; 37: 836–42. 42. Morise Z, Sugioka A, Hoshimoto S, et al. The role of hepatectomy for patients with liver metastases of gastric cancer. Hepatogastroenterology 2008; 55: 1238–41. 43. Okano K, Maeba T, Ishimura K, et al. Hepatic resection for metastatic tumors from gastric cancer. Ann Surg 2002; 235: 86–91. 44. Roh HR, Suh KS, Lee HJ, et al. Outcome of hepatic resection for metastatic gastric cancer. Am Surg 2005; 71: 95–9. 45. Sakamoto Y, Ohyama S, Yamamoto J, et al. Surgical resection of liver metastases of gastric cancer: an analysis of a 17-year experience with 22 patients. Surgery 2003; 133: 507–11. 46. Kotwall C, Sako K, Razack MS, et al. Metastatic patterns in squamous cell cancer of the head and neck. Am J Surg 1987; 154: 439–42. 47. Pieterman RM, van Putten JW, Meuzelaar JJ, et al. Preoperative staging of non-small-cell lung cancer with positron-emission tomography. N Engl J Med 2000; 343: 254–61. 48. Spector G J. Distant metastases from laryngeal and hypopharyngeal cancer. ORL J Otorhinolaryngol Relat Spec 2001; 63: 224–8. 49. Pawlik TM, Zorzi D, Abdalla EK, et al. Hepatic resection for metastatic melanoma: distinct patterns of recurrence and prognosis for ocular versus cutaneous disease. Ann Surg Oncol 2006; 13: 712–20. 50. Wente MN, Buchler MW, Weitz J. [Gastrointestinal stromal tumors (GIST). Surgical therapy]. Chirurg 2008; 79: 638–43. 51. DeMatteo RP, Lewis JJ, Leung D, et al. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000; 231: 51–8. 52. DeMatteo RP, Shah A, Fong Y, et al. Results of hepatic resection for sarcoma metastatic to liver. Ann Surg 2001; 234: 540–7; discussion 547–8. 53. Demetri GD, von Mehren M, Blake CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002; 347: 472–80. 54. Pawlik TM, Vauthey JN, Abdalla EK, et al. Results of a single-center experience with resection and ablation for sarcoma metastatic to the liver. Arch Surg 2006; 141: 537–43; discussion 543–4. 55. DeMatteo RP, Maki RG, Singer S, et al. Results of tyrosine kinase inhibitor therapy followed by surgical resection for metastatic gastrointestinal stromal tumor. Ann Surg 2007; 245: 347–52. 56. Gronchi A, Fiore M, Miselli F, et al. Surgery of residual disease following molecular-targeted therapy with imatinib mesylate in advanced/metastatic GIST. Ann Surg 2007; 245: 341–6. 57. Gomez D, Al-Mukthar A, Menon KV, et al. Aggressive surgical resection for the management of hepatic metastases from gastrointestinal stromal tumours: a single centre experience. HPB (Oxford) 2007; 9: 64–70. 58. Chang AE, Karnell LH, Menck HR. The National Cancer Data Base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer 1998; 83: 1664–78. 59. Leiter U, Meier F, Schittek B, et al. The natural course of cutaneous melanoma. J Surg Oncol 2004; 86: 172–8. 60. Becker JC, Terheyden P, Kampgen E, et al. Treatment of disseminated ocular melanoma with sequential fotemustine, interferon alpha, and interleukin 2. Br J Cancer 2002; 87: 840–5. 61. Avril MF, Aamdal S, Grob JJ, et al. Fotemustine compared with dacarbazine in patients with disseminated malignant melanoma: a phase III study. J Clin Oncol 2004; 22: 1118–25. 62. Herman P, Machado M A, Montagnini A L, et al. Selected patients with metastatic melanoma may benefit from liver resection. World J Surg 2007; 31: 171–4. 63. Rose D M, Essner R, Hughes T M, et al. Surgical resection for metastatic melanoma to the liver: the John Wayne Cancer Institute and Sydney Melanoma Unit experience. Arch Surg 2001; 136: 950–5. 64. Weitz J, Koch M, Friess H, et al. Impact of volume and specialization for cancer surgery. Dig Surg 2004; 21: 253–61.

172

18

Chemotherapy-associated hepatotoxicity Martin Palavecino, Daria Zorzi, and Jean-Nicolas Vauthey
chemotherapy-associated nonalcoholic fatty liver disease
Nonalcoholic fatty liver disease (NAFLD) encompasses different types of pathological changes in the liver, ranging from steatosis to steatohepatitis. NAFLD affects up to 24% of the general population and increases to 75% in patients with a body mass index equal or greater to 30 kg/m2 (16). Usually, NAFLD is asymptomatic, but it may progress to cirrhosis and develop hepatocellular carcinoma in later stages of disease (17,18). The diagnosis of NAFLD can be suspected by laboratory routine tests and imaging findings. However, the gold standard diagnostic method is the histological assessment of the liver. Steatosis is defined as the fat accumulation in the hepatocytes. It can be graded according to the percentage of affected cells (mild when less than 30% of the hepatocytes are involved, moderate with involvement of 30% to 60% of the hepatocytes, and severe with >60% hepatocytes involved). Steatohepatitis is defined as steatosis associated with inflammation (neutrophilic portal and lobular infiltration, perisinusoidal fibrosis, hepatocellular ballooning, glycogenated nuclei) (8). Kleiner et al. proposed a score based on three features (steatosis, lobular inflammation, and ballooning) evaluated semiquantitatively. A Kleiner score of 5 or greater correlates with a diagnosis of steatohepatitis, while a score of 3 or 4 is considered borderline (19). Steatosis Different chemotherapeutic regimens, such as intraarterial floxuridine (20), 5-FU and folinic acid (21), interferon α and 5-FU (22), 5-FU, and levamisole (23) were reported to induce steatosis. However, none of these early studies reported the effects of steatosis on surgical outcomes (8). In 1998, Behrns et al. (24) evaluated outcomes after major hepatectomy in patients with steatosis. The authors found that patients with moderate to severe steatosis (>30%) had a higher BMI, longer operative times, and higher rates of postoperative morbidity, mortality, and intraoperative blood transfusion. Similarly, Kooby et al. (25) analyzed a cohort of patients with steatosis who underwent liver resection. Steatosis was associated with infectious complications but not with major complications or postoperative mortality. Two studies were carried out at The University of Texas M.D. Anderson Cancer Center. The first showed an increased rate of steatosis in patients treated with irinotecan, but Parikh et al. found no increased mortality in patient with steatosis, even when severe (26). In the second study, Vauthey et al. (9) studied 406 resected patients using the Kleiner’s score (19), and no agent was found to be associated with steatosis and there was no increased postoperative morbidity or mortality rate (Table 18.1). However, many patients with steatosis have other

introduction
Hepatic resection is the only therapy that offers a chance for cure in patients with colorectal liver metastases, but only 20% of the patients are candidates for resection at the time of diagnosis (1–3). For those patients who are resected, the 5-year survival has been reported up to 58% (4–8). After hepatic resection, the majority of patients will develop recurrence within the liver with or without extrahepatic metastases. Systemic chemotherapy has been used preoperatively (9) or as an adjuvant therapy after surgery to decrease the risk of disease recurrence (10). During the last decade, an increasing number of new therapeutic agents has been developed to improve the response rates of the existing drugs (Fig. 18.1). Initially, 5-fluorouracil (5-FU)-based chemotherapy had a 20% response rate with a modest improvement in survival (11). Capecitabine was introduced as an oral alternative to intravenous 5-FU. The addition of oxaliplatin and irinotecan in combination with 5-FU increased the response rate to 50% and the conversion of unresectable metastases to resectable was subsequently reported in up to 38% of patients (12,13). Most recently, bevacizumab and cetuximab, antibodies vascular endothelial growth factor and an antiepidermal growth factor receptor, respectively, have been associated with response rates of up to 70%, when combined with standard chemotherapy (14). The rationale for preoperative chemotherapy includes (i) the downsizing of metastases, thus decreasing the amount of resected parenchyma and increasing the rate of curative resection; (ii) the identification of patients who will not benefit from surgical resection due to disease progression during chemotherapy; (iii) the early treatment of micrometastases (8–10). Nordlinger et al. (15) reported the results of a multicentric randomized controlled trial (EORTC Intergroup trial 40983), evaluating the outcome of patients with resectable colorectal liver metastases (no more than four metastases, no extrahepatic disease) with two arms: surgery alone versus six cycles of FOLFOX4 before and after surgery. The trial showed an increased progression-free survival at 3 years of 8.1% (from 28.1% to 36.2%, p = 0.041) in all eligible patients; and 9.2% (from 33.2% to 42.4%, p = 0.025) in all patients undergoing resection (15). Clinical and pathological studies have established associations between specific chemotherapeutic agents and histologic changes in the liver. Current evidence suggests there are two broad categories of chemotherapy-induced liver injury: nonalcoholic fatty liver disease (NAFLD), including steatosis and steatohepatitis, and sinusoidal obstruction syndrome (9) (Figs. 18.2 and 18.3). The use of sequential or combined treatments may result in mixed patterns of injuries. The objective of the present chapter is to summarize the changes induced in the liver parenchyma by chemotherapy and its effects on surgical outcomes.

173

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
comorbid conditions, such as obesity and diabetes that can increase the risk of complications. In a recent study of patients who underwent major hepatectomy, patients with steatosis had increased blood loss, morbidity, and intensive care unit stays compared to matched control patients with normal livers (27). The prevalence of obesity (BMI ≥ 30 kg/m2) was 26% in the steatotic patients compared with 2% in controls, which may have contributed to the poorer outcome in steatotic patients. Steatohepatitis An increased rate of steatohepatitis in patients undergoing preoperative chemotherapy was first observed by Fernandez et al. (28). Multivariate analysis showed that treatment with irinotecan or oxaliplatin and high BMI were independent risk factors for steatohepatitis. In the previously mentioned study, Vauthey et al. (9) analyzed the relationship between preoperative chemotherapy and liver injury. Using the Kleiner’s score (19), 8% of the patients had steatohepatitis. Steatohepatitis rate was higher in those patients treated with irinotecan-based chemotherapy (20% vs. 4%, p < 0.001). The incidence of steatohepatitis was higher in patients with BMI higher than 25 kg/m2. The 90-day mortality rate for patients with steatohepatitis was 15%, compared to 2% for patients without steatohepatitis (p = 0.001). The main cause of death was liver failure. The conclusion of the study was to cautiously use irinotecan in patients with BMI higher than 25 kg/m2, especially in patients undergoing major hepatic resections (Table 18.1). Unlike simple steatosis, which does not significantly impact postoperative outcome, steatohepatitis is an ominous finding and a relative contraindication to major liver resection. Given the associations between irinotecan, steatohepatitis, and increased postoperative mortality, major hepatic resection should probably not be performed in patients with known steatohepatitis, and irinotecan should be avoided in patients with known steatosis or steatohepatitis or the features of metabolic syndrome if major hepatic resection is anticipated. Sinusoidal obstruction syndrome The association between sinusoidal obstruction syndrome and oxaliplatin-based chemotherapy was first described by Rubbia-Brandt et al. (29) in 2004. Changes associated with

1980

1985

1990

1995

2000

2005

RR% Median survival (months) 20–25 ~55 ~70 13 20–22 >24?

5-FU Capecitabine Irinotecan Oxaliplatin Cetuximab Bevacizumab

Figure 18.1 During the last 10 years, several new drugs were incorporated to the armamentarium for the treatment of colorectal liver metastases. 5-FU, 5-fluorouracil; RR, response rate. Source: Modified from Ref. (33).

(A)

(B)

(C)

(D)

Figure 18.2 Nonalcoholic fatty liver disease (NAFLD). (A) Macroscopic view of a fatty liver (yellow liver). (B) Pathology specimen showing the aspect of a fatty liver. (C) Microscopic view of a simple steatosis: accumulation of large globules of fat in the cells. (D) Microscopic view of steatohepatitis: different degrees of inflammation in the field (ballooned and apoptotic cells). Source: Modified from Ref. (8).

174

CHEMOTHERAPY-ASSOCIATED HEPATOTOXICITY

(A)

(B)

(C)

Figure 18.3 Sinusoidal obstruction syndrome. (A) Macroscopic view of a liver with oxaliplatin-related sinusoidal injury (blue liver). (B) Pathology specimen showing the aspect of a liver with sinusoidal injury. (C) Microscopic view of sinusoidal injury: centrilobular sinusoidal dilatation with scattered macrovesicular steatosis. Source: Modified from Ref. (8).

Table 18.1 Published Data on Chemotherapy-Associated Hepatotoxicity and Its Effect on Postoperative Outcomes
Author, year Behrns, 1998 (24) Kooby, 2003 (25) Parikh, 2003 (26) Fernandez, 2005(28) Karoui, 2006(39) Vauthey, 2006 (9) Nordlinger, 2008(15) Nakano, 2008(40) Reddy, 2008 (44) Number of patients 135 325 chemo, 160 controls 61 chemo, 47 controls 37 45 chemo, 22 controls 248 chemo, 158 controls 151 chemo, 152 controls 36 chemo 39 chemo, 57 controls Major hepatectomy 100% 69% chemo, 63% controls 100% 49% 100% 68% N/A 100% 69% Drugs 5-FU ± irinotecan 5-FU ± irinotecan 5-FU ± irinotecan/ oxaliplatin 5-FU ± irinotecan/ oxaliplatin 5-FU ± irinotecan 5-FU ± oxaliplatin 5-FU ± oxaliplatin 5-FU ± oxaliplatin Bevacizumab + oxaliplatin Type of liver injury Steatosis Steatosis Steatosis Steatohepatitis Sinusoidal injury Steatohepatitis Sinusoidal injury N/A Sinusoidal injury N/A Morbidity NS Higher NS N/A Higher NS NS Higher Higher NS Mortality NS NS NS N/A NS NS+ NS NS N/A NS

Source: Modified from Ref. (45). NS, not significant; chemo, chemotherapy; 5-FU, 5-fluorouracil; N/A, not available. + Subset of patients with steatohepatitis had increase 90-day mortality.

175

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
sinusoidal injury (dilation and congestion, venous occlusion, and fibrosis) were in found in 78% of patients treated with oxaliplatin. This study did not analyze the effects of the injury on outcomes after resection. The association between oxaliplatin and sinusoidal injury has been confirmed in other studies, in which the incidence of sinusoidal injury in patients treated with oxaliplatin ranges from 19% to 52%. In the study by Vauthey et al. (9), the incidence of sinusoidal dilatation was higher in those patients with oxaliplatin-based chemotherapy compared to patients with other chemotherapy regimen (19% vs. 2%, p < 0.001). The analysis also demonstrated that oxaliplatinbased chemotherapy for a median of 12 weeks preoperatively was not associated with increased morbidity or mortality after surgery. Likewise, the pathological findings of sinusoidal injury itself were not associated with an increased rate of perioperative complications. In patients receiving preoperative 5-FU ± oxaliplatin, Aloia et al. (30) found severe forms of vascular alterations, specifically hemorrhagic centrilobular necrosis and regenerative nodular hyperplasia in patients treated for greater than 6 months preoperatively (12 cycles). In these patients, a higher rate of preoperative blood transfusions was noted. The EORTC Intergroup trial 40983 showed an increase in postoperative complications with perioperative chemotherapy with oxaliplatin plus surgery compared to surgery alone (26% vs. 16%, respectively, p = .04). However, this complication rate of 26% for combined chemotherapy and surgery compares favorably with the 36% complication rate previously reported for resection without preoperative chemotherapy in single center studies (31). Of note, in the surgery only arm of the EORTC Intergroup Trial 40983, 18 patients (11%) underwent an unnecessary laparotomy (open and close) compared to only 8 patients (5%) in the perioperative chemotherapy arm. If the total open and close is added to the total complications in each arm of the study, the difference between the two arms becomes nonsignificant for all unfavorable events (surgical complications plus open and close) in the comparison between perioperative chemotherapy plus surgery versus surgery alone (30% vs. 26%, p = 0.5). Taken together, limited preoperative chemotherapy (four to six cycles preoperatively) remains a valid option and is used at major centers in patients with resectable and unresectable liver metastases. Short-course preoperative chemotherapy is currently used at our institution in most patients with colorectal liver metastases considered for resection (Fig. 18.4) (32).
Diagnosis of colorectal liver metastasis

Resectable Preoperative therapy 2–3 months Hepatectomy (one-stage or two-stage) ± PVE*

Unresectable

Resectable

First-line chemotherapy re-evaluate 2–3 months

Second-line chemotherapy

Postoperative therapy 3–4 months

Third-line chemotherapy

Figure 18.4 Treatment recommendation for liver metastases of colorectal cancer. Source: Adapted from Ref. (32).

the effects of monoclonal antibodies
Targeted biologic agents are increasingly being used for the systemic treatment of colorectal liver metastases. In the past 5 years, bevacizumab and cetuximab were approved by the Food and Drug Administration for the treatment of colorectal liver metastases (33). Bevacizumab is a monoclonal antibody against vascular endothelial growth factor (VEGF). D’Angelica et al. (34) studied the effects of bevacizumab on outcomes after liver surgery. The authors compared patients who underwent surgery with or without preoperative bevacizumab. They found no increase in morbidity and suggested a waiting time of 6 to 8 weeks between the last dose of bevacizumab and

surgery. More recently, a study by Gruenberger et al. provided evidence to suggest that this interval may be shortened to 5 weeks without increase in perioperative complications (35). Ribero et al. (36) analyzed the effect of bevacizumab in patients receiving oxaliplatin-based chemotherapy. The response to therapy was measured with the percentage of viable cells in the surgical specimen. Patients who received preoperative bevacizumab had a significant lower rate of viable cells compared to those patients who did not receive preoperative bevacizumab (33% vs. 45%, p = 0.02). The incidence and severity of sinusoidal injury were lower in patients receiving preoperative bevacizumab (27% vs. 54%, p = 0.006). The antiangiogenic effects of bevacizumab have raised concerns regarding potential effects on bleeding, wound healing, and liver regeneration. A study from the MDACC reported that the addition of bevacizumab to chemotherapy before portal vein embolization did not impair liver regeneration (37). Cetuximab is a monoclonal antibody against epidermal growth factor receptor (EGFR). No specific liver injury has been so far identified and related to the preoperative administration of cetuximab. Preclinical data in animal models investigated the effects of anti-EGFR antibodies after partial hepatectomy in mice and found that their blockade does not impair liver regeneration (38). Future investigations are needed to further study possible specific histologic changes in the liver in patients treated with biologic agents.

diagnosis
Liver function tests cannot be used to assess chemotherapyassociated liver injury, since many patients have normal laboratory values despite significant hepatic injury. A heightened index of suspicion for chemotherapy-associated hepatic injury is necessary in patients at risk for NAFLD due to obesity, diabetes, or hyperlipidemia, as well as patients who have received prolonged courses of chemotherapy. Computed tomography can identify patients with fatty infiltration by determining the density of the liver compared to the spleen (at least 10 Hounsfield units lower than the spleen). Magnetic resonance

176

CHEMOTHERAPY-ASSOCIATED HEPATOTOXICITY
imaging (MRI) accurately predicts steatosis on the basis of signal differences between fat and water. However, modern imaging methods cannot differentiate between steatosis and steatohepatitis. For these reasons, liver biopsy is the gold standard diagnostic procedure to confirm liver injury. Percutaneous liver biopsy may be associated with false-negative results, due to the patchy distribution of the injuries. To overcome this issue, laparoscopy with direct inspection and core biopsy may be an alternative to image-guided percutaneous biopsy in patients suspected of chemotherapy-associated liver injury, especially in those patients who are candidates for major hepatic resection. Grossly, sinusoidal injury results in the so-called blue liver syndrome, characterized by a bluish, edematous, spongiform appearance and consistency (Fig. 18.3), while steatosis results in a yellow liver (Fig. 18.2). tomography. Briefly, the contours of the FLR are delineated on the screen, and volume is calculated by adding each slice’s volume, determined by the surface area, slice thickness, and space between slices (42). To calculate the total liver volume, Vauthey et al. (42) determine a formula based on body surface area. The estimated liver volume is calculated using the following formula: total liver volume = –794.41 + 1267.28 × body surface area. The ratio of the FLR to total estimated liver volume is defined as the standardized FLR (sFLR), which has been shown to reflect the function of the remnant liver and correlate with surgical outcome. When the sFLR is predicted to be insufficient for safe hepatic resection, portal vein embolization (PVE) is a strategy to induce hypertrophy of the FLR (42). In normal livers, if the standardized future liver remnant is ≤20% of total liver volume, portal vein embolization (PVE) should be considered. In patients who received extensive chemotherapy, preoperative PVE should be considered when the standardized future liver remnant is ≤30% of total liver volume (13). In this context, PVE is used as a procedure to test the capacity of the injured liver to regenerate. In a study by Ribero et al. (36), a degree of hypertrophy (DH = sFLR post-PVE – sFLR pre-PVE) ≤ 5% predicted the occurrence of postoperative complications, either overall, liver-related complications, or liver dysfunction. Patients with significant chemotherapyassociated liver injury who have inadequate liver hypertrophy after a technically successful PVE are not candidates for a major liver resection. Another strategy for patients with chemotherapy-associated hepatotoxicity to undergo complete resection of metastases is two-stage liver resection. This approach allows resection in patients with extensive bilateral liver metastases that have responded or remain stable on chemotherapy. In the first stage, metastases in the FLR are removed with a minor resection. After the first surgery, the hepatic regenerative capacity is assessed and PVE should be performed, if the sFLR is insufficient. After adequate regeneration, a second-stage major resection is performed up to 8 weeks after PVE. In a study from MDACC, using this approach, in patients with a median of seven liver metastases, the 3-year overall and disease-free survival rates were 86% and 51%, respectively, after perioperative chemotherapy and two-stage hepatectomy (43).

prevention
Several issues should be taken into account to prevent chemotherapy-associated liver injuries. First of all, prolonged unnecessary courses of preoperative chemotherapy should be avoided. Different studies demonstrated that hepatotoxicity is strongly related to chemotherapy duration. Karoui et al. (39) analyzed two groups of patients who underwent liver resection with or without chemotherapy (5-FU ± irinotecan/oxaliplatin). In the chemotherapy group, five patients developed liver insufficiency versus none in the control group. Morbidity was higher in patients who received at least six cycles of chemotherapy compared to those who received five cycles or less (54% vs. 19%, p = 0.047). In another study, Aloia et al. (30) concluded that patients who received more than 12 cycles of oxaliplatin-based chemotherapy had a higher rate of reoperations and a longer length of stay compared to patients who received 12 or fewer cycles. The optimal duration of preoperative chemotherapy to maximize therapeutic benefit, while avoiding hepatotoxicity, is likely up to 4 months (i.e., 8 cycles). In the study from MDACC, patients received relatively short-course oxaliplatin for 3 to 4 months, which was not associated with increased morbidity or mortality after hepatic resection (9). Another issue to be considered is the duration of the interval between chemotherapy and liver resection. Several studies show that a longer interval between chemotherapy and hepatic resection for CLM reduces hepatotoxicity and surgical complications. However, this interval should be balanced with the risk of tumor progression during the treatment-free interval. In the European trial, Nordlinger et al. (15) reported an interval between the last dose of chemotherapy and liver resection (in the chemotherapy arm) of 2 to 5 weeks. Nakano et al. (40) observed a mean interval between the last chemotherapy and surgery of 6.5 months in patients without sinusoidal injury compared to 3.6 months in patients with sinusoidal injury. Welsh et al. (41) observed a morbidity rate of 2.6%, 5.5%, and 11% when the intervals between the last chemotherapy and surgery was 9 to 12 weeks, 5 to 8 weeks, and ≤5 weeks, respectively (p = 0.009). In patients with suspected chemotherapy-associated liver injury, the functional future liver remnant should be assessed prior to major liver resection to minimize postoperative complications. The future liver remnant (FLR) can be assessed using three-dimensional contrast-enhanced computed

summary
During the last decade, several new chemotherapeutics agents were introduced in the armamentarium for the treatment of colorectal liver metastases. These new drugs were used as adjuvant treatment as well as preoperative treatment before liver resection. The rationale for preoperative chemotherapy is as follows:






To increase resectability in patients initially deemed unresectable, by downsizing the metastases To improve progression-free survival in patients with resectable metastases, when compared to surgery alone To select patients who may not benefit from surgery due to tumor progression while on chemotherapy.

177

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
However, preoperative treatment may increase the incidence of chemotherapy-associated hepatotoxicity, including steatosis, steatohepatitis, and sinusoidal injury. Each injury has been associated with the use of specific agents, such as steatohepatitis in patients who received irinotecan-based chemotherapy and sinusoidal injury in patients who received oxaliplatin-based chemotherapy. Steatohepatitis is associated with increased risk of mortality due to liver failure and represents a relative contraindication to major hepatic resection. Significant sinusoidal injury with fibrosis and regenerative nodular hyperplasia may increase the risk of bleeding from liver resection, but no increased mortality was associated with sinusoidal injury and oxaliplatin. Preoperative chemotherapy should be administrated only in short courses. Several studies suggested limiting the use of chemotherapy to less than 4 months. In patients receiving bevacizumab, the period between the last dose of the drug and surgery should be at least 5 weeks. Before liver surgery, the future liver remnant should be assessed. In patients who have received prolonged chemotherapy, PVE is indicated if the standardized future liver remnant is ≤30%. Preoperative chemotherapy should be coordinated by a multidisciplinary team and should be adjusted according to patient, tumor, and liver characteristics.
14. Adam R, Aloia T, Levi F, et al. Hepatic resection after rescue cetuximab treatment for colorectal liver metastases previously refractory to conventional systemic therapy. J Clin Oncol 2007; 25: 4593–602. 15. Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 2008; 371: 1007–16. 16. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002;346: 1221–31. 17. Adams LA, Lymp JF, St Sauver J, et al. The natural history of nonalcoholic fatty liver disease: a population-based cohort study. Gastroenterology 2005; 129: 113–21. 18. Bugianesi E, Leone N, Vanni E, et al. Expanding the natural history of nonalcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology 2002; 123: 134–40. 19. Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005; 41: 1313–21. 20. Zeiss J, Merrick HW, Savolaine ER, et al. Fatty liver change as a result of hepatic artery infusion chemotherapy. Am J Clin Oncol 1990; 13: 156–60. 21. Peppercorn PD, Reznek RH, Wilson P, Slevin ML, Gupta RK. Demonstration of hepatic steatosis by computerized tomography in patients receiving 5-fluorouracil-based therapy for advanced colorectal cancer. Br J Cancer 1998; 77: 2008–11. 22. Sorensen P, Edal AL, Madsen EL, Fenger C, Poulsen MR, Petersen OF. Reversible hepatic steatosis in patients treated with interferon alfa-2a and 5-fluorouracil. Cancer 1995; 75: 2592–6. 23. Moertel CG, Fleming TR, Macdonald JS, Haller DG, Laurie JA. Hepatic toxicity associated with fluorouracil plus levamisole adjuvant therapy. J Clin Oncol 1993; 11: 2386–90. 24. Behrns KE, Tsiotos GG, DeSouza NF, et al. Hepatic steatosis as a potential risk factor for major hepatic resection. J Gastrointest Surg 1998; 2: 292–8. 25. Kooby DA, Fong Y, Suriawinata A, et al. Impact of steatosis on perioperative outcome following hepatic resection. J Gastrointest Surg 2003; 7: 1034–44. 26. Parikh AA, Gentner B, Wu TT, et al. Perioperative complications in patients undergoing major liver resection with or without neoadjuvant chemotherapy. J Gastrointest Surg 2003; 7: 1082–8. 27. McCormack L, Petrowsky H, Jochum W, Furrer K, Clavien PA. Hepatic steatosis is a risk factor for postoperative complications after major hepatectomy: a matched case-control study. Ann Surg 2007; 245: 923–30. 28. Fernandez FG, Ritter J, Goodwin JW, et al. Effect of steatohepatitis associated with irinotecan or oxaliplatin pretreatment on resectability of hepatic colorectal metastases. J Am Coll Surg 2005; 200: 845–53. 29. Rubbia-Brandt L, Audard V, Sartoretti P, et al. Severe hepatic sinusoidal obstruction associated with oxaliplatin-based chemotherapy in patients with metastatic colorectal cancer. Ann Oncol 2004; 15: 460–6. 30. Aloia T, Sebagh M, Plasse M, et al. Liver histology and surgical outcomes after preoperative chemotherapy with fluorouracil plus oxaliplatin in colorectal cancer liver metastases. J Clin Oncol 2006; 24: 4983–90. 31. Metreveli RE, Sahm K, Denstman F, Abdel-Misih R, Petrelli NJ. Hepatic resection at a major community-based teaching hospital can result in good outcome. Ann Surg Oncol 2005; 12: 133–7. 32. Kopetz S, Vauthey JN. Perioperative chemotherapy for resectable hepatic metastases. Lancet 2008; 371: 963–5. 33. Chun YS, Vauthey JN. Extending the frontiers of resectability in advanced colorectal cancer. Eur J Surg Oncol 2007; 33 Suppl 2: S52–8. 34. D’Angelica M, Kornprat P, Gonen M, et al. Lack of evidence for increased operative morbidity after hepatectomy with perioperative use of bevacizumab: a matched case-control study. Ann Surg Oncol 2007; 14: 759–65. 35. Gruenberger B, Tamandl D, Schueller J, et al. Bevacizumab, capecitabine, and oxaliplatin as neoadjuvant therapy for patients with potentially curable metastatic colorectal cancer. J Clin Oncol 2008; 26: 1830–5. 36. Ribero D, Wang H, Donadon M, et al. Bevacizumab improves pathologic response and protects against hepatic injury in patients treated with

references
1. Steele G, Jr., Ravikumar TS. Resection of hepatic metastases from colorectal cancer. Biologic perspective. Ann Surg 1989; 210: 127–38. 2. Scheele J. Hepatectomy for liver metastases. Br J Surg 1993; 80: 274–6. 3. Bismuth H, Adam R, Levi F, et al. Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. Ann Surg 1996; 224: 509–20. 4. Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 2004; 239: 818–25. 5. Choti MA, Sitzmann JV, Tiburi MF, et al. Trends in long-term survival following liver resection for hepatic colorectal metastases. Ann Surg 2002; 235: 759–66. 6. Fernandez FG, Drebin JA, Linehan DC, et al. Five-year survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with F-18 fluorodeoxyglucose (FDGPET). Ann Surg 2004; 240: 438–47; discussion 447–50. 7. Pawlik TM, Scoggins CR, Zorzi D, et al. Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 2005; 241: 715–22. 8. Zorzi D, Laurent A, Pawlik TM, et al. Chemotherapy-associated hepatotoxicity and surgery for colorectal liver metastases. Br J Surg 2007; 94: 274–86. 9. Vauthey JN, Pawlik TM, Ribero D, et al. Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases. J Clin Oncol 2006; 24: 2065–72. 10. Chun YS, Vauthey JN. Risks of neoadjuvant chemotherapy for resectable colorectal carcinoma hepatic metastases. Curr Colorectal Cancer Rep 2008; 4: 87–92. 11. Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000; 343: 905–14. 12. Adam R, Avisar E, Ariche A, et al. Five-year survival following hepatic resection after neoadjuvant therapy for nonresectable colorectal. Ann Surg Oncol 2001; 8: 347–53. 13. Adam R, Delvart V, Pascal G, et al. Rescue surgery for unresectable colorectal liver metastases downstaged by chemotherapy: a model to predict longterm survival. Ann Surg 2004; 240: 644–57.

178

CHEMOTHERAPY-ASSOCIATED HEPATOTOXICITY
oxaliplatin-based chemotherapy for colorectal liver metastases. Cancer 2007; 110: 2761–7. Zorzi D, Chun YS, Madoff DC, Abdalla EK, Vauthey JN. Chemotherapy with bevacizumab does not affect liver regeneration after portal vein embolization in the treatment of colorectal liver metastases. Ann Surg Oncol 2008; 15: 2765–72. Van Buren G, 2nd, Yang AD, Dallas NA, et al. Effect of molecular therapeutics on liver regeneration in a murine model. J Clin Oncol 2008; 26: 1836–42. Karoui M, Penna C, Amin-Hashem M, et al. Influence of preoperative chemotherapy on the risk of major hepatectomy for colorectal liver metastases. Ann Surg 2006; 243: 1–7. Nakano H, Oussoultzoglou E, Rosso E, et al. Sinusoidal injury increases morbidity after major hepatectomy in patients with colorectal liver metastases receiving preoperative chemotherapy. Ann Surg 2008; 247: 118–24. 41. Welsh FK, Tilney HS, Tekkis PP, John TG, Rees M. Safe liver resection following chemotherapy for colorectal metastases is a matter of timing. Br J Cancer 2007; 96: 1037–42. 42. Vauthey JN, Abdalla EK, Doherty DA, et al. Body surface area and body weight predict total liver volume in Western adults. Liver Transpl 2002; 8: 233–40. 43. Chun YS, Vauthey JN, Ribero D, et al. Systemic chemotherapy and twostage hepatectomy for extensive bilateral colorectal liver metastases: perioperative safety and survival. J Gastrointest Surg 2007; 11: 1498–1504. 44. Reddy SK, Morse MA, Hurwitz HI, et al. Addition of bevacizumab to irinotecan- and oxaliplatin-based preoperative chemotherapy regimens does not increase morbidity after resection of colorectal liver metastases. J Am Coll Surg 2008; 206: 96–106. 45. Chun YS, Laurent A, Maru D, Vauthey JN. Management of chemotherapyassociated hepatotoxicity in colorectal liver metastases. Lancet Oncol 2009; 278–86.

37.

38.

39.

40.

179

19

Thermal ablation of liver metastases Samir Pathak and Graeme J. Poston
comparing ablative techniques against each other, and only one trial has shown survival superiority of ablation combined with chemotherapy over chemotherapy alone in liver only disease (see later). This chapter aims to detail the mechanism of action of the various ablative therapies available, and also to review the available literature regarding the implementation of these techniques.

introduction
Colorectal cancer is a common malignancy and as many as 25% of patients will have liver metastasis (CRLM) at presentation and a further 20% to 30% will develop metachronous disease following colorectal surgery (1). The vast majority of disease-related deaths are due to metastatic disease. In metastatic disease the median length of survival without treatment is approximately between 5 and 12 months (2). Currently hepatic resection is established as the treatment modality of choice for colorectal liver metastases (CRLM) with 5-year survival rates of up to 60% being reported by some groups (3–6). Unfortunately at the time of presentation only 20% to 30% are deemed suitable to resection because of tumor location, number of metastases, other comorbidities, and lack of hepatic reserve (7). Consequently, in recent times there has been considerable interest in the use of oxaliplatin-based neoadjuvant therapy to reduce tumor burden, so increasing the probability of achieving a curative resection and hence improve overall disease-free survival. Even patients who initially had unresectable hepatic disease may respond to chemotherapy and become resectable (8). However, despite a more aggressive approach to surgical resection and the use of combination regimens of highly active chemotherapy drugs, a significant proportion of patients are still not eligible for resection. Additionally, the high rate of recurrence seen in the liver, affecting 53% to 68% of patients requiring repeat resections can only be tolerated by a subset of patients (9). Hence, recently there has been considerable interest in thermoablative techniques and their potential role in the management of CRLM. Ablative technologies involve the delivery of localized treatment via open, laparoscopic, or percutaneous route. Theoretical advantages include less physiological stress, making treatments suitable for patients who may not otherwise be appropriate for formal resection. The potential for either percutaneous or laparoscopic approach offers an alternative for patients unfit or unwilling to undergo major abdominal surgery and general anaesthesia (9–13). Formal resection is guided by vasculobiliary anatomy, with significant amounts of healthy parenchyma being removed along with disease. Targeted ablations minimize the removal of healthy parenchyma, making it useful in patients with borderline parenchymal volume and function. Anatomically difficult lesions may not be amenable to formal resection, but accessible by probe ablation. Ablation can also be performed as an adjunct to surgical resection in patients with bilobar disease, where patients have the majority of their tumor burden formally resected with remnant disease burden being ablated. However, there remains a need for more long-term survival data regarding ablative therapies. There have been no randomized control trials comparing any ablative therapy to resections in patients who would be candidates for either therapy or

limitations of the literature
Apart from the European CLOCC Study (EORTC 40004) (see later), there are no published randomized controlled trials (RCTs) comparing the use of radiofrequency ablation (RFA), ethanol ablation, or cryoablation with hepatectomy in the treatment of colorectal liver metastases. Furthermore, there is only one study comparing microwave ablation (MCT) with partial hepatectomy for the treatment of CRLM (14). Therefore the majority of the data reviewed come from single-arm, retrospective, and prospective single-center studies. The results derived from each study must be viewed with caution as the number and ethnicity of patients in each study varied greatly (e.g., studies from Japan vs. Western populations). Possible selection bias and varying end-points between studies also existed. Additionally, it must be borne in mind that the definition of resectability has evolved over the course of time and this resultantly would have led to differing cohorts of patients between studies. Clinical results for distinct patient populations (neuroendocrine metastases, noncolorectal GI metastases, hepatocellular carcinoma, etc.) were often combined. Some articles reported independent tumor outcomes whereby others gave a combined outcome. Furthermore, other baseline characteristics such as extrahepatic disease and the use of neoadjuvant or adjuvant chemotherapy varied greatly between studies. End points were not always reported uniformly. Some considered survival or recurrence from the time of diagnosis, whereas other studies looked at these end points from time of first treatment. We have therefore tried to assess comparative treatment efficacy using 1-, 2-, 3-, 4-, and 5-year survival rates, median survival rates, complication rates, local (hepatic) recurrence, site-specific (at ablation site) recurrence, and overall recurrence. No single study reported all of the above. The heterogeneity of the studies and the absence of long-term data for microwave ablation, in particular, make it difficult to offer an evidence based recommendation for the ablative management of unresectable colorectal hepatic metastases.

cryotherapy
Cryoablation of hepatic metastases using insulated probes containing liquid nitrogen/argon have been used for the destruction of CLRM (15). They are placed into each metastasis, whereupon

180

THERMAL ABLATION OF LIVER METASTASES
liquid nitrogen or argon is used to freeze the lesion, using temperatures as low as −30°C. The progress of the enlarging ice ball may be monitored radiologically using MRI/CT or ultrasound scanning (16). Initially this was seen to be advantageous, however subsequently it has been demonstrated that the peripheries of the ice ball may not have reached a sufficiently low temperature to cause cellular death (17). Furthermore, histopathological assessment of lesions produced by cryotherapy has also shown that there is significant amount of tissue adjacent to blood vessels that remains undamaged by the ablation (18). This “heat sink” effect may result in viable tumor remaining in seemingly “treated” lesions, explaining high local recurrence rates. The physiological basis of cryotherapy has been well investigated and is dependent upon the rapid formation of ice crystals during the freezing process. Additionally, cellular hypoxia due to disruption of the surrounding microvascular structures also induces cell destruction and enhances the direct damage resulting from ice ball formation (19). There are no clearly defined indications for the use of cryotherapy but patients with unresectable metastatic disease secondary to either extensive bilobar involvement or difficult anatomical location may benefit. Tumors, which lie in close proximity to major blood vessels such as the inferior vena cava, large portal branches or large hepatic veins, may make cryotherapy difficult due to the “freeze-thaw” effect previously described. Conventionally, a laparotomy was required for the direct application of probes; however probes have now been developed that are small enough to be placed percutaneously. In the studies reviewed, reports of 3- and 5-year survival were sparse, and varied between 30.9% to 44% and 13% to 26%, respectively. Median survival ranged between 22.9 and 94.2 months (Table 19.1). However, the marked heterogeneity of these studies makes direct comparison difficult. Major complication rates (defined as complications requiring the patient to remain in hospital) were high, ranging from 22% to 70%. The major concern with use of cryotherapy is the “cryoshock” phenomenon, where patients develop a systemic response to ablation, consisting of marked thrombocytopenia leading to coagulopathy, pleural effusion, acute respiratory distress syndrome, and myoglobinuria (30–33). The true incidence of cryoshock is difficult to establish, though reports from the literature suggest a mortality of 0% to 8%. A large multicentre survey estimated that it was responsible for 18% of perioperative deaths (31). The high rate of local complications (hemorrhage from a cracked liver, subphrenic abscess, bilomas, and biliary fistulae) and fear of cryoshock has led to this technique falling out of favor as other safer and equally efficient techniques have evolved. with a median survival of 29 to 39 months. Major complication rates were reported as 0.22% to 25%. Again, these outcomes reflect the heterogeneity of these studies. Major hepatic resections have an overall complication rate (major and minor complications) of around 20% (5). The fact that two studies report major complication rate of a similar magnitude (35,36), with survival data in keeping with most major series assessing resection alone, would suggest that edge cryotherapy presents a theoretical advantage for patients deemed “unresectable.”

radiofrequency ablation
Radiofrequency ablation (RFA) uses radiofrequency radiation to produce heat locally within the hepatic parenchyma. The radiofrequency current generates ionic agitation, which in turn is translated into heat, resulting in the subsequent breakdown of proteins and cell membranes (43). The main advantage when compared to cryotherapy is that the probes can be placed percutaneously. However, as with all locally ablative techniques, the efficacy of the treatment diminishes with increasing size of the lesion. Hence, manufacturers have designed a variety of electrodes that can be deployed in situ to produce a number of tips. RFA refers to coagulation from all electromagnetic resources with a frequency less than 900 kHz, with the majority functioning within the parameters of 300 to 500 kHz. Initially, problems existed with early radiofrequency designs due to the effects of high temperatures in the tissue surrounding the probe. This is due to tissue impedance secondary to tissue charring. Subsequently, this impedance results in reduced dissipation of current (44,45). This problem has been the major drawback of RFA, though it has been countered somewhat by the use of cooled electrode tips. However, the principle limiting factor of RF ablation remains the size of the achievable ablated tissue. This is because only the tissue immediately adjacent to the tip is heated by ionic agitation. The remainder of the tissue is ablated via heat produced via thermal conduction. This effect is magnified in the presence of large blood vessels, which further reduce heat via a phenomenon known as “the heat sink” effect (19). Both normal liver parenchyma and metastatic liver are water-rich and also have an extensive blood supply (via angiogenesis in the case of metastases). Hence, thermal conduction is facilitated, but as mentioned previously, this is a less efficient means of ablation than ionic agitation. Therefore, current opinion suggests that RFA is more susceptible to the heat sink effect than microwave ablation. Various measures have been used previously to reduce the heat sink effect, such as occlusion of the portal vein and hepatic artery at the time of ablation. Although the ablative area is increased, the risk of bile duct damage and portal vein thrombosis is increased. Because of the relative simplicity of the technique, the fact that it can be performed percutaneously and the comparatively cheap devices employed, RFA is a technique that remains widely practised (32).

edge cryotherapy
Edge cryotherapy employs the application of cryotherapy to the resection margins posthepatectomy in order to extend the margins of resectability. Several studies describe the use of cryotherapy when a histopathologically positive margin is expected. During this procedure, flat cryoprobes are placed against the resection edge of the remnant liver, whereupon remnant liver tissue is frozen to a depth of at least 1 cm (34). Reported 3- and 5-year survival for these patients was 43% to 60% and 26% to 44%, respectively,

review of the literature regarding rfa
Previous reviews have suggested that there are no compelling data supporting the use of RFA in patients with viable extrahepatic disease (EHD) (46). EHD is known to be a poor prognostic

181

182
N 44 116 55 172 56 86 30 61 20 49 15 1 3.9 — 4.2 — 4 3 3.38 1.7 5.1 1.4 5 4.4 — 3.6 — — 2 4 2 — — — 9.5 18 16 — — 0 16 — 0 0 — 82 — 89 — 85 76 87 — — — — 56 — 65 67 — 61 54 — — — — 32.3 44 41 43 43 — 36 — 30.9 — — — 24 — — — — — — — — 13 26 19 22 19 — — — — 33 26 29 28 30 33 — 26 — 22.9 94.2 Metastases (n) Size (cm) EHD 1 year 2 years 3 years 4 years 5 years 22% 31% — 28% — — 30% — 30% 26% 70% Median survival (months) Major Minor — — — — — — 20% 55% 53%

Table 19.1 Summary of Studies Looking at Cryotherapy (Survival and Complications)

Author

Year

Seifert (20) Seifert (21) Seifert (22) Yan (23) Kerkar (24) Brooks (25) Joosten (26) Chen (27) Kornprat (28) Paganini (open) (29) Paganini (29) (lap)

1998 1998 2003 2003 2004 2005 2005 2006 2007 2007 2007

Table 19.2 Summary of Studies Looking at Edge Cryotherapy (Survival and Complications)

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Author 63 26 44 75 86 24 55 124 — 1 1 2 2 2.9 2.4 4.1 3.1 5 5 — 5 4.5 3.5 5 0 — — 0 0 — 18 15

Year

N

Metastases (n) Size (cm) — — — — — 92 — 84

Extrahepatic disease %

1 year survival %

2 years survival % — — — 50.5 — — — 61

3 years survival % 60 — — — 54.7 58 — 43

4 years survival % — — — — — 37 — 28

5 years survival % 44 — — — — — 26 24

Median survival (months) — 29 33 33 — 39 29 29

Major 25 0.27 0.22 — 0.34 21 — —

Minor — — — — 34% — — —

Korpan (35) Dwerryhouse (37) Seifert (38) Finlay (39) Gruenberger (40) Rivoire (36) Seifert (41) Niu (42)

1997 1998 1998 2000 2001 2002 2005 2006

THERMAL ABLATION OF LIVER METASTASES
indicator, predicting decreased overall survival and disease-free survival compared to patients without EHD (3,47). The authors remain unconvinced of this conclusion, since several studies consisting of patients with EHD (range 8.7–30%) have demonstrated reasonable median survival (range 18–37 months). Berber et al. (48) evaluated their 10-year experience with RFA in 234 patients who had a variety of neoplastic process occurring within the hepatic parenchyma. Their results showed that 80% of patients had progressive disease, despite aggressive chemotherapy, with the oncological intervention failing approximately 8 months before RFA. The authors also found no significant difference in patients who had EHD at the time of treatment, against those who did not. This observation strengthens our opinion that EHD is not an accurate predictor of outcome in patients with unresectable colorectal liver metastases. Hence, patients should not be denied RFA on this basis alone. There does not appear to be a maximum number of metastases that may be treated via RFA. However, there is a perception that local recurrence and survival rates appear to be negatively correlated with number and size of treated metastases. This review did not identify evidence clearly supporting this hypothesis, although a trend toward it is evident. The reasons for this conclusion are not obvious but it may be that patient factors, such as age, comorbidity, and operator experience have a significant influence. Generally, the highest ablation success rates were achieved in patients with solitary colorectal liver metastases or patients with a few metastases smaller than 3 cm (49–53). As with formal resection, the aim of any tumor eradication therapy is to achieve a clear negative margin. It follows therefore that that the best results are obtainable when the tumor is smaller than the size of coagulative necrosis produced by a single ablation probe, and it is therefore the size of ablation zone that limits the use of RFA. Most ablation devices can produce single ablations of around 4 cm in diameter. As probe delivery is performed by hand, either “blindly” or using two-dimensional imaging techniques (USS, CT), probes may be inadvertently placed away from the geometric centre of a lesion, resulting in a rim of untreated tissue. This opinion may explain higher recurrence rates in lesions larger than 3 cm. Attempts have been made to increase the ablation size and overcome the inherent limit of RFA by developing probes that deploy multiple “tines” around a lesion, as well as adopting techniques that reduce blood flow through parenchyma, another method known to increase lesion size by increasing the area which reaches sufficient temperature by indirect heating (54). The location of metastases within the liver is an important factor in determining the success of RFA. Tumors adjacent to large hepatic vessels are problematic, as larger vessels act as a heat sink, making it more difficult to ablate the tumor. Several studies commented on the increased failure rates in tumors adjacent to major blood vessels (26,55). Ablation near portal vein pedicles is also associated with an increased risk of major bile duct injuries, possibly as a result of de-epithelialization injuries related to heat. coagulation of tissue surfaces was slower and produced deeper areas of tissue necrosis, compared to normal electrocautery units. This led to it being investigated as a technique for the treatment of unresectable hepatic malignancies (81). Microwave radiation lies between infrared radiation and radiofrequency, with frequencies from 900 to 2450 MHz. Tissue heating is based on the agitation of water molecules, which in turn cause cellular death via coagulation necrosis. Thus it is different from RFA as the frequency of the electromagnetic radiation used is considerably higher. This results in a greater ability to localize the dissipation of energy, though the tissue penetration is reduced (81–83). The microwave generators available for clinical use have an output of between 70 and 90 W. The microwave emitting needle is placed directly into the tumor, usually under radiological guidance. The emitting needle is attached to the microwave generator and when the generator is activated, each area of the tumor is treated for 30 to 60 seconds at 70 to 90 W. The rapid generation of heat using MCT produces 10 to 25 mm zones of coagulative necrosis after only 30 to 60 seconds. The rapid development of coagulative necrosis precludes the further dissipation of heat to surrounding tissues. Thus, microwave offers many of the benefits of RFA, with some substantial theoretical advantages. These benefits include higher intratumoral temperatures, faster ablation times, larger tumor ablation volumes, ability to use simultaneous multiple applicators and less procedural pain (30,32). With RFA, the zone of active tissue heating is limited to a zone of a few millimeters surrounding the active electrode, with the remainder of the ablation zone being created via thermal conduction. However, via a superior convection profile, microwave produces a larger zone of active heating, allowing a more uniform destruction of cells within the target area. RFA is also limited by the impedance with tissue boiling and charring, because water vapor and char act as electrical insulators. Due to the electromagnetic nature of microwaves, microwave ablations appear unaffected by the effect of water vapor and charring. MCT technology allows for open, laparoscopic, and percutaneous routes of delivery. Ablation is performed using a thin antenna that is attached to the microwave generator. In the literature, different protocols for time and power of ablation have been proposed, dependent on the tissue and antenna type (84). Seki et al. (85) treated 15 patients with solitary colorectal liver metastases who declined formal resection using percutaneous microwave ablation. Ten patients were alive at the end of the follow-up period (9–37 months), with a median survival of 24.2 months. This is broadly similar to best chemotherapy, but obviously direct comparisons are difficult between such homogeneous studies. No recurrence was detected in adequately treated lesions, although two patients experienced recurrence due to inadequate treatment at presentation as defined by incomplete destruction on posttreatment imaging. Another Japanese group (14) performed a small randomized control study on 30 patients comparing hepatic resection versus MCT. One, 2- and 3-year survival rates for the microwave group were 71%, 57%, and 14% compared to

microwave ablation
Microwave coagulation (MCT) was initially developed in the early 1980s by Tabuse et al. in order to optimize haemostasis along the plane of dissection during hepatic resection. The microwave

183

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
69%, 56%, and 23% for the resection group. The mean survival time was 25 months for the microwave group versus 23 months for the resection group. Statistically, there was no difference in survival between the two groups. A significant proportion of patients in both arms of this study developed hepatic failure without explanation as to why this occurred. Tanaka et al. (88) performed a retrospective analysis of 53 patients who underwent hepatectomy or hepatectomy plus microwave ablation. Their results suggested no difference between the two groups in terms of recurrence or survival. This suggests that ablative therapies may be used to extend the margins of resectability. In a review of 31 patients by Bhardwaj et al. (93), of whom the majority had unresectable colorectal metastases, median survival was 29 months, with a 3-year survival of 40%, and local recurrence was only 2%. Despite the variety of primary and secondary lesions, these figures illustrate the potential for MCT in the treatment of hepatic metastases.

Ultrasound probe

Ultrasound beam Iceball surrounding and encompassing tumor Cryoprobe Probe within liver
Ultrasound

(A)

(B)

Figure 19.1 Cryoablation (surgical view A) under intra-operative ultrasound control (B).

percutaneous ethanol injection
Percutaneous ethanol injection (PEI) involves passing single or multiple fine needles followed by intratumoral injection of pure ethanol, causing cytotoxic cell death mainly via dehydration. The main disadvantage with PEI is that treatment in patients with large metastases has been found to be inadequate due to incomplete alcohol diffusion within the tumor mass. Additionally, the results for PEI in the treatment of CLRM have not been as promising as for primary hepatocellular carcinoma, due mainly to the difference in tumor characteristics (94). HCC is usually hypervascular and may be encapsulated, qualities which will reduce leakage into the surrounding hepatic parenchyma, while ensuring diffusion through the lesion. However, CRLM tend to be dense and infiltrative, making the diffusion of ethanol more unpredictable, resulting in pockets of untreated tumor. For this reason, thermoablative techniques are preferred for CRLM (16). There was a paucity of data regarding use of PEI for the treatment of colorectal liver metastases. Only three series were identified, which looked at the feasibility of PEI as a treatment, as opposed to survival and complication rates. It is unlikely that future research will be channeled in this direction as the tumor characteristics of metastases make them unfavorable for PEI.

Figure 19.2 Radiofrequency ablation.

the clocc study
Finally, at the time of coming to press, Ruers and colleagues presented the final results of the EORTC CLOCC (EORTC 40004) Study at the 2010 ASCO meeting in Chicago (97). This study was conceived as a 400-patient Phase III randomization of patients with up to nine unresectable liver-only metastases to receive either oxaliplatin-based chemotherapy or chemotherapy plus RFA (open, laparoscopic, or percutaneous) with or without concomitant resection of easily resectable lesions. The primary end-point was powered to test for a 38% overall survival benefit in the RFA arm. This was an extremely ambitious project, and recruitment was understandably extremely difficult. After a period of extreme frustration, the trial objective was reduced to a 100-patient

randomized Phase II, with an actual accrual of 119 patients. However, it remains a unique landmark study, probably never to be repeated, and the only prospective study to address the question of the real survival benefit of thermal ablation therapy for metastatic liver disease. Although there was a significant improvement in 3-year progression free-survival (PFS) of 27.6% for RFA + chemotherapy compared to 10.7% for chemotherapy alone (p = 0.025) (Fig. 19.3), a secondary end point, overall survival (OS) at 30 months (the primary study end point) was no different for RFA + chemotherapy (63.8%) over chemotherapy alone (58.6%) (p = 0.218) (Fig 19.4). It must be remembered that when designed, the study was never powered to demonstrate a significant result for its primary end point with such low numbers, and it is extremely unlikely that any investigators will ever be bold enough to try to repeat such a study. Therefore in the pragmatic real world, we must accept the evidence that we have, which in our opinion suggests a survival benefit for thermal ablation therapies in the treatment of relatively low-volume unresectable liver metastases.

184

THERMAL ABLATION OF LIVER METASTASES
Progression free survival 100 90 80 Overall logrank test: p = 0.025 70 60 50 18.08% 40 30 20 10 0 0 O 53 44 N 59 60 1 2 3 4 5 6 7 Treatment CT RF+\-resection+CT RF + Chemo

Chemo (years)

Number of patients at risk: 24 12 5 34 20 13

4 8

2 3

1 0

Figure 19.3 Three year PFS in the CLOCC study comparing RFA + chemotherapy (27.6%) to chemotherapy alone (10.7%) (p = 0.025).

100 90 80 70 60 50 40 30 20 10 0 0 O 39 31 N 59 60 1 2 Overall logrank test: p = 0.218

Overall survival

RF + Chemo

Chemo (years) 3 4 5 6 7 Treatment CT RF+\-resection+CT

Number of patients at risk: 52 43 27 53 44 27

15 19

5 7

1 1

Figure 19.4 Thirty month OS comparing RFA+chemotherapy (63.8%) to chemotherapy alone (58.6%) (p = 0.218).

185

186
Number 54 14 30 36 102 37 25 14 122 30 25 30 66 56 68 235 73 135 52 109 16 47 23 1.8 6.2 1.2 — 3.25 — — 2 1.63 — — 1.9 2.8 3.5 1 2.8 4.1 3.2 2.7 1.6 1.75 3.12 5.57 1.3 1.4 2 2.1 2.2 2.25 2.5 2.7 2.9 3 3 3.1 3.2 3.5 3.7 3.9 3.9 4.1 5.2 — — — — — — No No No — No — 20.5 — 28 — 15 — 38 23 No 30 — — — — 8.7 81 — — — 87 — — 90 79 — — 75 — 92 — — 91 — 87 — 84 88 — — 95 — — 62 — — 54 — — — 45 — 67 — — — — 77 67 68 80 — — — — 26 46 — 60 54 38 57 52.6 7 — 42 20.6 20.2 28 — 50 33 — 57 — — — — — 26 — — — — — — 3 — — — — — — — — — — — — — — 34 — 46.5 25.5 — 22 27 — — 21 — 30 18.4 25 — — — — — — Metastases (n) Size (cm) EHD 1 year 2 years 3 years 4 years 5 years Median survival (months) — — 36 39 32 40 — — 31.5 — 37 23.2 27 28 20.5 24 overall 38 28.9 — 30 — 39 18 Major — 33.3 — 11 6.9 — 0 11.5 1.1 — 4 — 10 3.4 — — 4 — 7.1 0.9 2.9 13 7 Minor — — — — 4 — 0 — 7 — — 5 49.1 — 2.9 — 6 — — 6.4 2.9 —

Table 19.3 Summary of Studies Looking at RFA (Survival and Complications)

Author

Year

Ablative type

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

De Baere (56) Elias (57) Park (58) Knudsen (59) Sorensen (60) Lee (61) Hur (62) Lermite (51) Veltri (63) Aloia (64) Oshowo (65) Suppiah (66) Reuter (67) Hildebrand (68) Berber (69) Siperstein (70) Gilliams (53) Berber (71) Iannitti (72) Solbiati (73) Terraz (74) Abitabile (52) Stippel (75)

2000 2000 2008 2009 2007 2008 2009 2006 2008 2006 2003 2007 2009 2006 2008 2007 2005 2005 2002 2001 2007 2007 2002

RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA RFA

Table 19.4 Summary of Studies looking at RFA ± Resection (Survival and Complications)

Author

Year

Ablative type

Number

Metastases (n) size (cm) EHD 1 year 2 years 3 years 4 years 5 years

Median survival (months)

Major

Minor — — — — — — — 20 22 — 27 11 13 — — 11

Pawlik (76) Scaife (77) Abdalla (3) Elias (78) Joosten (26) Amersi (49) Kornprat (28) Gleisner (79) Nikfarjam (80)

2003 2003 2004 2005 2005 2006 2007 2008 2009

RFA ± resection RFA ± resection RFA ± resection RFA ± resection RFA ± resection RFA ± resection RFA ± resection RFA ± resection RFA ± resection

124 50 158 63 28 74 19 66 23

3 2 — 2.4 3 3.3 5 2 —

1.8 2 — 15 2 3.56 2 2.5 —

No No — No No No — — —

— — — 92 93 — — 92 —

— 66 — 67 75 — — — —

— — 43/37 47 — — — 51.2 —

— — 36/22 — — — — — —

— — — — — — — 28 68

37.3 — — 36 — 29.7 — 38.1 —

THERMAL ABLATION OF LIVER METASTASES

Table 19.5 Summary of Studies Looking at MCT (Survival and Complications)
Minor (%) 6.7 — 1 year — 71 91.4 — 80 — — — 82.1 40 — 2 years — — 59.5 — — — — — — — — 3 years — 57 46.4 — 51 — — — — — — 4 years — — 29 — — — — — — — — 5 years — 14 — — 17 — — 32 — — — — — — 7.8 — 76.3 0 80.1

Author Seki (85) Shibata (86) Liang (11) Yokoyama (87) Tanaka (88) Iannitti (89) Kuang (90) Ogata (91) Zhang (92) Bhardwaj (93)

Year 1999 2000 2003 2003 2005 2007 2007 2008 2008 2009

Ablative type MCT MCT MCT MCT/RFA MCT MCT MCT MCT/RFA MCT MCT N 15 14 28 12 16 33 11 32 34 24 Size (cm) 2.1 2.7 3.12 2.4 4.8 3.6 2.75 2.8 — 2 — EHD (%) 0 0 5 — 5 — 0 22 — 0 —

Metastases (n) 1 4.1 2 2.8 2.2 2.57 1.47 4 — 2.87

Median survival (months) 24.2 27 20.5 — 28 — — 43 — 29 —

Major (%) — 14.0 0.0 — 19.0 16.1 4.0 3.4 — 0.0 2.6

187

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 19.6 Summary of Studies Looking at PEI (Survival and Complications)
Author Kessler (95) Giorgio (96) Year 2002 2003 Number 13 (inc HCC) 47 Metastases (n) Unknown Unknown Size (cm) Unknown Unknown EHD Unknown Unknown Recurrence rates 50% (inc HCC) Unknown Median survival Unknown Unknown Major 8% 3% Minor 25% 17.7%

Table 19.7 Overall Comparison of Studies Reviewed (Cumulative)
Recurrence (Range %) Ablative technique Cryotherapy Ethanol MCT RFA Local 12–39 — 5–13 10–31 Overall 78–88 — 50–78 47–86 1 year 84 — 73 85 2 years 59 — 60 67 Survival rates (%) 3 years 37 — 30 36 4 years 21 — 29 30 5 years 17 — 16 24 Major complication rates (%) 29 5 7 6

Median

RFA + resection Edge cryotherapy

5 Year

4 Year

3 Year

2 Year

1 Year 0 10 20 30 40 50 60 70 80 90 100

Figure 19.5 Survival figures for studies reviewed (ablation as adjunct to surgery).

conclusions
The ideal ablative therapy should cause complete tumor ablation, yet be parenchyma sparing, reproducible, safe, and be minimally invasive. Current advancements, particularly in RFA and MCT, are promising but the perfect ablative model is still elusive. The literature cannot support the use of percutaneous ethanol injection for the treatment of colorectal metastases, though we accept that it has a role in the management of hepatocellular carcinoma. Similarly, the literature demonstrates that although cryoablation has acceptable survival figures, its ongoing use cannot be advocated given the high rate of local complications (Table 19.7). Ablative therapies offer great potential for lesions that cannot be formally resected. The increasing burden of metastatic colorectal disease means that a growing number of patients will have unresectable metastases and hence will be candidates for ablation. It is important that the ablation causes complete tumor destruction within the treatment zone. Heat sink effect may

result in tumor viability even within a seemingly completed ablation. The ability to accurately place the probe is also vital to ensure that the treatment zone encompasses the focus of disease. Currently, most centers use RFA or microwave ablation as treatment of choice. Microwave offers the theoretical advantage of larger ablation volumes, shorter ablation duration, and the ability to perform multiple simultaneous ablations to increase ablation volume as well as more predictable ablation zones around vessels. The lower local recurrence rate found in this review probably reflects the more predictable ablation characteristics on MCT. Conversely the larger body of evidence surrounding RFA is probably a manifestation of its maturity as a technology, rather than an implicit endorsement of its superiority over other technologies. The role of ablative technology in a palliative setting is unclear. However, 3-year survival of between 30% and 37% compares favorably with best supportive chemotherapy (Figs. 19.5 and 19.6). The underlying mechanism behind this remains unclear, though it may be related to decreasing the tumor burden.

188

THERMAL ABLATION OF LIVER METASTASES
Median Radiofrequency ablation Microwave ablation 5 Year Cryotherapy

4 Year

3 Year

2 Year

1 Year 0 10 20 30 40 50 60 70 80 90

Survival (%) Figure 19.6 Survival figures for ablative studies reviewed.

Ethanol ablation Radiofrequency ablation Microwave ablation Cryotherapy RFA+resection Edge cryotherapy

0

5

10

15 20 25 Complication rates(%)

30

35

40

Figure 19.7 Complication rates for all studies reviewed.

The safety profiles of RFA and MCT appear similar and in the current climate, both are safe and effective therapies, which should be deployed (Fig 19.7). Further studies are needed to demonstrate long-term outcomes and ongoing research will ensure that ablative technologies continue to evolve rapidly.

references
1. (2003)CRUIC. CancerStats. Cancer Research UK Information Centre (2003)]. Available from: http://www.info.cancerresearch.org/cancerstats. 2. Hugh TJ, Kinsella AR, Poston GJ. Management strategies for colorectal liver metastases—Part I. Surg Oncol 1997; 6(1): 19–30. 3. Abdalla EK, Vauthey JN, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg. 2004; 239(6): 818–25; discussion 25–7. 4. Fernandez FG, Drebin JA, Linehan DC, et al. Five-year survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with F-18 fluorodeoxyglucose (FDG-PET). Ann Surg 2004; 240(3): 438–47; discussion 47–50.

5. Choti MA, Sitzmann JV, Tiburi MF, et al. Trends in long-term survival following liver resection for hepatic colorectal metastases. Ann Surg 2002; 235(6): 759–66. 6. Pawlik TM, Scoggins CR, Zorzi D, et al. Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 2005; 241(5): 715–22, discussion 22–4. 7. Adam R, Huguet E, Azoulay D, et al. Hepatic resection after down-staging of unresectable hepatic colorectal metastases. Surg Oncol Clin N Am 2003; 12(1): 211–20, xii. 8. Nordlinger B, Van Cutsem E, Rougier P, et al. Does chemotherapy prior to liver resection increase the potential for cure in patients with metastatic colorectal cancer? A report from the European Colorectal Metastases Treatment Group. Eur J Cancer 2007; 43(14): 2037–45. 9. Goldberg SN, Solbiati L, Hahn PF, et al. Large-volume tissue ablation with radio frequency by using a clustered, internally cooled electrode technique: laboratory and clinical experience in liver metastases. Radiology 1998; 209(2): 371–9. 10. Jiao LR, Habib NA. Experimental study of large-volume microwave ablation in the liver (Br J Surg 2002; 89: 1003–7). Br J Surg 2003; 90(1): 122.

189

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
11. Liang P, Dong B, Yu X, Yang Y, Yu D, Su L, et al. Prognostic factors for percutaneous microwave coagulation therapy of hepatic metastases. AJR Am J Roentgenol 2003; 181(5): 1319–25. 12. Sato M, Watanabe Y, Kashu Y, et al. Sequential percutaneous microwave coagulation therapy for liver tumor. Am J Surg 1998; 175(4): 322–4. 13. Shen P, Geisinger KR, Zagoria R, Levine EA. Pathologic correlation study of microwave coagulation therapy for hepatic malignancies using a threering probe. J Gastrointest Surg 2007; 11(5): 603–11. 14. Shibata T, Niinobu T, Ogata N, Takami M. Microwave coagulation therapy for multiple hepatic metastases from colorectal carcinoma. Cancer. 2000; 15: 89(2): 276–84. 15. Hugh TJ, Kinsella AR, Poston GJ. Management strategies for colorectal liver metastases—Part II. Surg Oncol 1997; 6(1): 31–48. 16. Evans J. Ablative and catheter-delivered therapies for colorectal liver metastases (CRLM). Eur J Surg Oncol 2007; 33 Suppl 2: S64–75. 17. Mala T, Edwin B, Samset E, et al. Magnetic-resonance-guided percutaneous cryoablation of hepatic tumours. Eur J Surg 2001; 167(8): 610–7. 18. Bhardwaj N, Strickland AD, Ahmad F, Atanesyan L, West K, Lloyd DM. A comparative histological evaluation of the ablations produced by microwave, cryotherapy and radiofrequency in the liver. Pathology. 2009; 41(2): 168–72. 19. Bhardwaj N, Strickland AD, Ahmad F, Dennison AR, Lloyd DM. Liver ablation techniques: a review. Surg Endosc 2010; 24(2): 254–65. 20. Seifert JK, Morris DL. Indicators of recurrence following cryotherapy for hepatic metastases from colorectal cancer. Br J Surg 1999; 86(2): 234–40. 21. Seifert JK, Morris DL. Prognostic factors after cryotherapy for hepatic metastases from colorectal cancer. Ann Surg 1998; 228(2): 201–8. 22. Seifert JK, Junginger T. Cryotherapy for liver tumors: current status, perspectives, clinical results, and review of literature. Technol Cancer Res Treat 2004; 3(2): 151–63. 23. Yan DB, Clingan P, Morris DL. Hepatic cryotherapy and regional chemotherapy with or without resection for liver metastases from colorectal carcinoma: how many are too many? Cancer 2003; 15: 98(2): 320–30. 24. Kerkar S, Carlin AM, Sohn RL, et al. Long-term follow up and prognostic factors for cryotherapy of malignant liver tumors. Surgery. 2004; 136(4): 770–9. 25. Brooks AJ, Wang F, Alfredson M, Yan TD, Morris DL. Synchronous liver resection and cryotherapy for colorectal metastases: survival analysis. Surgeon 2005; 3(4): 265–8. 26. Joosten J, Jager G, Oyen W, Wobbes T, Ruers T. Cryosurgery and radiofrequency ablation for unresectable colorectal liver metastases. Eur J Surg Oncol 2005; 31(10): 1152–9. 27. Chen YY, Perera DS, Yan TD, Schmidt LM, Morris DL. Applying Fong’s CRS liver score in patients with colorectal liver metastases treated by cryotherapy. Asian J Surg 2006; 29(4): 238–41. 28. Kornprat P, Jarnagin WR, DeMatteo RP, et al. Role of intraoperative thermoablation combined with resection in the treatment of hepatic metastasis from colorectal cancer. Arch Surg 2007; 142(11): 1087–92. 29. Paganini AM, Rotundo A, Barchetti L, Lezoche E. Cryosurgical ablation of hepatic colorectal metastases. Surg Oncol 2007; 16 Suppl 1: S137–40. 30. Seifert JK, Achenbach T, Heintz A, Bottger TC, Junginger T. Cryotherapy for liver metastases. Int J Colorectal Dis 2000; 15(3): 161–6. 31. Seifert JK, Morris DL. World survey on the complications of hepatic and prostate cryotherapy. World J Surg. 1999; 23(2): 109–13; discussion 13–4. 32. Primrose JN. Treatment of colorectal metastases: surgery, cryotherapy, or radiofrequency ablation. Gut 2002; 50(1): 1–5. 33. Stubbs RS, Alwan MH, Booth MW. Hepatic cryotherapy and subsequent hepatic arterial chemotherapy for colorectal metastases to the liver. HPB Surg 1998; 11(2): 97–104. 34. Seifert JK, Junginger T, Morris DL. A collective review of the world literature on hepatic cryotherapy. J R Coll Surg Edinb 1998; 43(3): 141–54. 35. Korpan NN. Hepatic cryosurgery for liver metastases. Long-term followup. Ann Surg 1997; 225(2): 193–201. 36. Rivoire M, De Cian F, Meeus P, et al. Combination of neoadjuvant chemotherapy with cryotherapy and surgical resection for the treatment of unresectable liver metastases from colorectal carcinoma. Cancer 2002; 95(11): 2283–92. 37. Dwerryhouse SJ, Seifert JK, McCall JL, et al. Hepatic resection with cryotherapy to involved or inadequate resection margin (edge freeze) for metastases from colorectal cancer. Br J Surg 1998; 85(2): 185–7. 38. Seifert JK, Morris DL. Cryotherapy of the resection edge after liver resection for colorectal cancer metastases. Aust N Z J Surg 1998; 68(10): 725–8. 39. Finlay IG, Seifert JK, Stewart GJ, Morris DL. Resection with cryotherapy of colorectal hepatic metastases has the same survival as hepatic resection alone. Eur J Surg Oncol 2000; 26(3): 199–202. 40. Gruenberger T, Jourdan JL, Zhao J, King J, Morris DL. Reduction in recurrence risk for involved or inadequate margins with edge cryotherapy after liver resection for colorectal metastases. Arch Surg. 2001; 136(10): 1154–7. 41. Seifert JK, Junginger T. Prognostic factors for cryotherapy of colorectal liver metastases. Eur J Surg Oncol 2004; 30(1): 34–40. 42. Niu R, Yan TD, Zhu JC, Black D, Chu F, Morris DL. Recurrence and survival outcomes after hepatic resection with or without cryotherapy for liver metastases from colorectal carcinoma. Ann Surg Oncol 2007; 14(7): 2078–87. 43. Ruers T, Bleichrodt RP. Treatment of liver metastases, an update on the possibilities and results. Eur J Cancer 2002; 38(7): 1023–33. 44. Livraghi T, Goldberg SN, Monti F, et al. Saline-enhanced radio-frequency tissue ablation in the treatment of liver metastases. Radiology 1997; 202(1): 205–10. 45. Wright AS, Sampson LA, Warner TF, Mahvi DM, Lee FT, Jr. Radiofrequency versus microwave ablation in a hepatic porcine model. Radiology 2005; 236(1): 132–9. 46. Wong SL, Mangu PB, Choti MA, et al. American Society of Clinical Oncology 2009 clinical evidence review on radiofrequency ablation of hepatic metastases from colorectal cancer. J Clin Oncol 2010; 28(3): 493–508. 47. Pawlik TM, Choti MA. Surgical therapy for colorectal metastases to the liver. J Gastrointest Surg 2007; 11(8): 1057–77. 48. Berber E, Siperstein A. Local recurrence after laparoscopic radiofrequency ablation of liver tumors: an analysis of 1032 tumors. Ann Surg Oncol 2008; 15(10): 2757–64. 49. Amersi FF, McElrath-Garza A, Ahmad A, et al. Long-term survival after radiofrequency ablation of complex unresectable liver tumors. Arch Surg 2006; 141(6): 581–7; discussion 7–8. 50. Jiang HC, Liu LX, Piao DX, et al. Clinical short-term results of radiofrequency ablation in liver cancers. World J Gastroenterol 2002; 8(4): 624–30. 51. Lermite E, Lebigot J, Oberti F, et al. Radiofrequency thermal ablation of liver carcinoma. Prospective study of 82 lesions. Gastroenterol Clin Biol 2006; 30(1): 130–5. 52. Abitabile P, Hartl U, Lange J, Maurer CA. Radiofrequency ablation permits an effective treatment for colorectal liver metastasis. Eur J Surg Oncol 2007; 33(1): 67–71. 53. Gillams AR, Lees WR. Radio-frequency ablation of colorectal liver metastases in 167 patients. Eur Radiol 2004; 14(12): 2261–7. 54. Kim SK, Lim HK, Ryu JA, et al. Radiofrequency ablation of rabbit liver in vivo: effect of the pringle maneuver on pathologic changes in liver surrounding the ablation zone. Korean J Radiol 2004; 5(4): 240–9. 55. Lu DS, Raman SS, Limanond P, et al. Influence of large peritumoral vessels on outcome of radiofrequency ablation of liver tumors. J Vasc Interv Radiol 2003; 14(10): 1267–74. 56. de Baere T, Elias D, Dromain C, et al. Radiofrequency ablation of 100 hepatic metastases with a mean follow-up of more than 1 year. AJR Am J Roentgenol 2000; 175(6): 1619–25. 57. Elias D, Goharin A, El Otmany A, et al. Usefulness of intraoperative radiofrequency thermoablation of liver tumours associated or not with hepatectomy. Eur J Surg Oncol 2000 Dec; 26(8): 763–9. 58. Park IJ, Kim HC, Yu CS, et al. Radiofrequency ablation for metachronous liver metastasis from colorectal cancer after curative surgery. Ann Surg Oncol 2008; 15(1): 227–32. 59. Knudsen AR, Kannerup AS, Mortensen FV, Nielsen DT. Radiofrequency ablation of colorectal liver metastases downstaged by chemotherapy. Acta Radiol 2009; 50(7): 716–21. 60. Sorensen SM, Mortensen FV, Nielsen DT. Radiofrequency ablation of colorectal liver metastases: long-term survival. Acta Radiol 2007; 48(3): 253–8. 61. Lee WS, Yun SH, Chun HK, et al. Clinical outcomes of hepatic resection and radiofrequency ablation in patients with solitary colorectal liver metastasis. J Clin Gastroenterol 2008; 42(8): 945–9. 62. Hur H, Ko YT, Min BS, et al. Comparative study of resection and radiofrequency ablation in the treatment of solitary colorectal liver metastases. Am J Surg 2009; 197(6): 728–36.

190

THERMAL ABLATION OF LIVER METASTASES
63. Veltri A, Sacchetto P, Tosetti I, et al. Radiofrequency ablation of colorectal liver metastases: small size favorably predicts technique effectiveness and survival. Cardiovasc Intervent Radiol 2008; 31(5): 948–56. 64. Aloia TA, Vauthey JN, Loyer EM, et al. Solitary colorectal liver metastasis: resection determines outcome. Arch Surg 2006; 141(5): 460–6; discussion 6–7. 65. Oshowo A, Gillams A, Harrison E, Lees WR, Taylor I. Comparison of resection and radiofrequency ablation for treatment of solitary colorectal liver metastases. Br J Surg 2003; 90(10): 1240–3. 66. Suppiah A, White TJ, Roy-Choudhury SH, et al. Long-term results of percutaneous radiofrequency ablation of unresectable colorectal hepatic metastases: final outcomes. Dig Surg 2007; 24(5): 358–60. 67. Reuter NP, Woodall CE, Scoggins CR, McMasters KM, Martin RC. Radiofrequency ablation vs. resection for hepatic colorectal metastasis: therapeutically equivalent? J Gastrointest Surg 2009; 13(3): 486–91. 68. Hildebrand P, Leibecke T, Kleemann M, et al. Influence of operator experience in radiofrequency ablation of malignant liver tumours on treatment outcome. Eur J Surg Oncol 2006; 32(4): 430–4. 69. Berber E, Tsinberg M, Tellioglu G, Simpfendorfer CH, Siperstein AE. Resection versus laparoscopic radiofrequency thermal ablation of solitary colorectal liver metastasis. J Gastrointest Surg 2008; 12(11): 1967–72. 70. Siperstein AE, Berber E, Ballem N, Parikh RT. Survival after radiofrequency ablation of colorectal liver metastases: 10-year experience. Ann Surg 2007; 246(4): 559–65; discussion 65–7. 71. Berber E, Pelley R, Siperstein AE. Predictors of survival after radiofrequency thermal ablation of colorectal cancer metastases to the liver: a prospective study. J Clin Oncol 2005; 23(7): 1358–64. 72. Iannitti DA, Dupuy DE, Mayo-Smith WW, Murphy B. Hepatic radiofrequency ablation. Arch Surg 2002; 137(4): 422–6; discussion 7. 73. Solbiati L, Ierace T, Tonolini M, Osti V, Cova L. Radiofrequency thermal ablation of hepatic metastases. Eur J Ultrasound 2001; 13(2): 149–58. 74. Terraz S, Constantin C, Majno PE, et al. Image-guided multipolar radiofrequency ablation of liver tumours: initial clinical results. Eur Radiol 2007; 17(9): 2253–61. 75. Stippel DL, Bohm S, Beckurts KT, Brochhagen HG, Holscher AH. Intraoperative radiofrequency ablation using a 3D navigation tool for treatment of colorectal liver metastases. Onkologie 2002; 25(4): 346–50. 76. Pawlik TM, Izzo F, Cohen DS, Morris JS, Curley SA. Combined resection and radiofrequency ablation for advanced hepatic malignancies: results in 172 patients. Ann Surg Oncol 2003; 10(9): 1059–69. 77. Scaife CL, Curley SA, Izzo F, et al. Feasibility of adjuvant hepatic arterial infusion of chemotherapy after radiofrequency ablation with or without resection in patients with hepatic metastases from colorectal cancer. Ann Surg Oncol 2003; 10(4): 348–54. 78. Elias D, Sideris L, Pocard M, et al. Incidence of unsuspected and treatable metastatic disease associated with operable colorectal liver metastases discovered only at laparotomy (and not treated when performing percutaneous radiofrequency ablation). Ann Surg Oncol 2005; 12(4): 298–302. 79. Gleisner AL, Choti MA, Assumpcao L, et al. Colorectal liver metastases: recurrence and survival following hepatic resection, radiofrequency ablation, and combined resection-radiofrequency ablation. Arch Surg 2008; 143(12): 1204–12. 80. Nikfarjam M, Shereef S, Kimchi ET, et al. Survival outcomes of patients with colorectal liver metastases following hepatic resection or ablation in the era of effective chemotherapy. Ann Surg Oncol 2009; 16(7): 1860–7. 81. Izzo F. Other thermal ablation techniques: microwave and interstitial laser ablation of liver tumors. Ann Surg Oncol 2003; 10(5): 491–7. 82. Wemyss-Holden SA, Dennison AR, Berry DP, Maddern GJ. Local ablation for unresectable liver tumors: is thermal best? J Hepatobiliary Pancreat Surg 2004; 11(2): 97–106. 83. Martin LW, Warren RS. Current management of colorectal liver metastases. Surg Oncol Clin N Am 2000; 9(4): 853–76; discussion 77–8. 84. Carrafiello G, Lagana D, Mangini M, et al. Microwave tumors ablation: principles, clinical applications and review of preliminary experiences. Int J Surg 2008; 6 Suppl 1: S65–9. 85. Seki T, Wakabayashi M, Nakagawa T, et al. Percutaneous microwave coagulation therapy for solitary metastatic liver tumors from colorectal cancer: a pilot clinical study. Am J Gastroenterol 1999; 94(2): 322–7. 86. Shibata T, Murakami T, Ogata N. Percutaneous microwave coagulation therapy for patients with primary and metastatic hepatic tumors during interruption of hepatic blood flow. Cancer. 2000; 88(2): 302–11. 87. Yokoyama T, Egami K, Miyamoto M, et al. Percutaneous and laparoscopic approaches of radiofrequency ablation treatment for liver cancer. J Hepatobiliary Pancreat Surg 2003; 10(6): 425–7. 88. Tanaka K, Shimada H, Nagano Y, Endo I, Sekido H, Togo S. Outcome after hepatic resection versus combined resection and microwave ablation for multiple bilobar colorectal metastases to the liver. Surgery. 2006; 139(2): 263–73. 89. Iannitti DA, Martin RC, Simon CJ, et al. Hepatic tumor ablation with clustered microwave antennae: the US Phase II Trial. HPB (Oxford) 2007; 9(2): 120–4. 90. Kuang M, Lu MD, Xie XY, et al. Liver cancer: increased microwave delivery to ablation zone with cooled-shaft antenna—experimental and clinical studies. Radiology 2007; 242(3): 914–24. 91. Ogata Y, Uchida S, Hisaka T, et al. Intraoperative thermal ablation therapy for small colorectal metastases to the liver. Hepatogastroenterology 2008; 55(82–83): 550–6. 92. Zhang X, Zhou L, Chen B, et al. Microwave ablation with cooled-tip electrode for liver cancer: an analysis of 160 cases. Minim Invasive Ther Allied Technol 2008; 17(5): 303–7. 93. Bhardwaj N, Strickland AD, Ahmad F, et al. Microwave ablation for unresectable hepatic tumours: clinical results using a novel microwave probe and generator. Eur J Surg Oncol 2010; 36(3): 264–8. 94. Bartolozzi C, Lencioni R. Ethanol injection for the treatment of hepatic tumours. Eur Radiol 1996; 6(5): 682–96. 95. Kessler A, Blank A, Merhav H, Orron D, Konikoff F, Oren R, et al. Minimally invasive techniques in the treatment of liver tumors. Isr Med Assoc J 2002; 4(12): 1106–10. 96. Giorgio A, Tarantino L, de Stefano G, et al. Complications after interventional sonography of focal liver lesions: a 22-year single-center experience. J Ultrasound Med 2003; 22(2): 193–205. 97. Ruers T, Punt C, v Coevorden F, et al. Final results of the EORTC Intergroup randomized study 40004 (CLOCC) evaluating the benefit of radiofrequency ablation combined with chemotherapy for unresectable colorectal liver metastases. Proc ASCO 2010; J Clin Oncol 2010; 28 (15): A-3526.

191

20 Resection for hepatocellular carcinoma
introduction
Hepatocellular carcinoma (HCC) is one of the most common malignancies with more than a million cases reported every year worldwide. It is a common cause of death in the Far East because of high endemecity of hepatitis B virus (HBV) infection. In the West, the rising trend in the incidence of HCC is parallel to the epidemic of Hepatitis C (HCV) virus infection (1). Alcohol excess, genetic hemochromatosis, aflatoxin B1 (2), and primary biliary cirrhosis are other associated risk factors. In the West, the majority of HCC patients have associated cirrhosis of which a significant number is alcohol related. More than a third of these cases present with HCC as the initial presentation in contrast to the Far East where HCC is often diagnosed at an early stage by surveillance of the at-risk population (3). Symptomatic HCC has a poor prognosis with a median survival of 1 to 8 months (4). A multi-disciplinary approach involving the surgeons, hepatologist, clinical oncologist, and radiologist is needed to formulate the best treatment options. Surgical resection offers the best possible treatment outcome but a large proportion of patients are not suitable for such an approach.

Rajesh Satchidanand, Stephen W. Fenwick, and Hassan Z. Malik
staging systems
A staging system allows for separation of patients into groups and selection of appropriate treatment modality. A number of staging systems are used in HCC using tumor characteristics and underlying liver pathology. The most commonly used method of tumor, nodes, and metastases (TNM) in the American Joint Committee on Cancer–Tumor–Metastases (AJCC), TNM staging dependant on resection and postoperative histology (9). The Okuda Classification includes variables related to the tumor and liver function (10). The Cancer of the Liver Italian Program Investigators staging systems includes ChildTurcotte-Pugh (Child A/B/C) score, tumor morphological characteristics, AFP, and vascular invasion/portal vein thrombosis. The Japan Integrated Staging score uses a combination of Child A/B/C score and TNM staging system. By far, The Barcelona Clinic Liver Cancer (BCLC) using tumor variables and the current available treatment options gives a better prognostic value in early cases (Fig. 20.2) (11).

treatment
A number of treatment options are available for patients with HCC. These include 1. Liver resection (LR) 2. Orthotopic liver transplant (OLT) including deceased/ cadaveric donor liver transplant (DDLT/CLT) and living donor liver transplant (LDLT) 3. Treatment prior to OLT: bridging the gap 4. Less invasive procedure involving chemical or thermal destruction of liver parenchyma 5. Regional or systemic chemotherapy 6. Radiotherapy including external beam irradiation or embolization with radioactive particles Before selecting a treatment option, careful consideration should be given to preoperative staging, underlying condition of the liver and the general fitness of the patient. Staging laparoscopy is a mandatory prior to LR or OLT to rule out extrahepatic disease. Assessing the residual liver function in chronic liver disease is very important before LR, as any major resection in a cirrhotic patient may result in fatal liver failure in the immediate post-operative period. Traditionally, the Child A/B/C scoring system has been used to assess the residual liver function. However, the Model for End-stage Liver Disease (MELD) scoring is used as an alternative in United States. Ascites on CECT, bilirubin of >2 mg/dL, and iodocyanine green retention test (used extensively in the East) (8) at 15 minutes of <30% bodes ill for residual liver function. Clinically relevant portal hypertension with hepatic vein gradient of >10 mm of mercury, esophageal varices, splenomegaly, and a platelet count of <100 × 109/L are accurate predictors of post-operative liver decompensation (12). Patients with Child

diagnosis
Asymptomatic HCC is diagnosed either as an incidental finding or on routine surveillance of at-risk population. Ultrasound scan (with or without contrast enhancement) with measurement of serum alfa-feto protein (AFP) is routinely used for screening (5,6). Once a suspicion of a focal lesion is raised, further assessment with contrast-enhanced computerized tomography (CECT) and/or magnetic resonance imaging (MRI) with contrast enhancement is needed to confirm the diagnosis of HCC. Tumor biopsy is rarely needed and in fact should be avoided in potentially resectable lesion due to the risk of tumor seeding along the needle track and intra-celomic spread. Furthermore, CECT is a good modality to look for the presence of cirrhosis, ascites, and metastases. A typical HCC shows hyper vascular enhancement with characteristic feature of early uptake of contrast and portal venous washout, an enhanced pseudocapsule, vascular invasion on CECT which gives more than 80% accuracy in diagnosing these lesions (7). MRI is more sensitive in detecting lesions 1 to 2 cm in size. A quarter of intra-hepatic lesions smaller than 10 mm is miss-diagnosed on pre-operative investigations. Diagnosing any lesion ≥2 cm with characteristics CECT/MRI findings is possible with a high degree of accuracy. In lesions 1 to 2 cm without concordance with two radiological investigations, a raised AFP of ≥400 µ/L and one radiological modality with positive features, diagnosis is possible (8). In lesions <10 mm, expectant follow-up with repeat imaging at 3 to 6 months is an appropriate management algorithm (Fig. 20.1).

192

RESECTION FOR HEPATOCELLULAR CARCINOMA
HCC within Milan criteria

Child-Pugh A

Child-Pugh B–C

Solitary < 3 cm

Solitary 3–5 cm

2–3 nodules < 3 cm

Solitary < 5 cm 2–3 nodules < 3 cm

Deep location

Peripheral location

Deep location Portal hypertension Varices, platelets < 100,000/mm3

No

Yes

Portal vein embolization If right hepatectomy required

RF ablation

Laparoscopic resection

Open resection

Transplantable Recurrence Not transplantable

Transplantation

RF, TACE, resection, New drugs, supportive care

Figure 20.1 Algorithm for management of transplantable hepatocellular carcinoma used at Henri-Mondor Hospital. Source: From Ref. (29).

A can withstand up to 50% liver resection, but in those with Child B, a future remnant liver value of less than 75% is associated with major complications. LR remains the treatment of choice in early cases of noncirrhotic HCC, in tumors of <5 cm size, or up to three tumor nodules each <3 cm in size. Even though early experience with OLT yielded good results, it was fraught with recurrence (13). Increasing incidence of HCC with scarcity of donor livers available for transplant meant stringent criteria for patient selection. Hence, with the adoption of the more restrictive Conventional Milan Criteria (CMC: 1 lesion <5 cm, 2–3 lesions <3 cm), OLT has resulted in better long-term results (14). With increasing experience, some groups have suggested expanding the boundaries of CMC. One such recommendation is University of California, San Francisco (UCSF), criteria for patients with one lesion ≤6.5 cm or two to three lesions ≤4.5 cm with a total tumor diameter of <8 cm (15).

early stage hcc
Patients with one lesion <5 cm in size or two to three lesions <3 cm in size with good residual liver functions are considered as having early stage disease. In these patients LR, OLT or percutaneous ablative therapy with a curative intent yielding high response rate are possible (16). Both LR and OLT have the best outcomes and treatment of choice is dependent on the availability of a donor liver. Tumor progression while waiting for a donor liver may decide the treatment option.

Liver Resection Liver resection in early HCC can be used in three different settings: (a) primary therapy, (b) to obtain material for morphological assessment of the tumor and to select patients who would benefit OLT, and (c) as a bridge therapy for those who are enlisted for OLT.

193

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

HCC
Stage 0
PST 0, child-pugh A, okuda 1

Stage A–C
Okuda 1–2, PST 0–2, Child-Pugh A–B

Stage D
Okuda 3, PST > 2, Child-Pugh C

Very early stage (0) single < 2 cm. Carcinoma in situ

Early stage (A) Single or 3 nodules < 3 cm, PS 0

Intermediate stage (B) Multinodular, PS 0

Advanced stage (C) Portal invasion, N1, M1, PS 1–2

Terminal stage (D)

Single Portal pressure/bilirubin Increased Normal

3 nodules ≤ 3 cm

Associated diseases

Portal invasion, N1, M1

No Liver transplantation (CLT/LDLT)

Yes

No

Yes New agents Symptomatic treatment

Resection

PEI/RF

Chemoembolization

Curative treatments

Randomized controlled trials

Figure 20.2 Barcelona Clinic Liver Cancer staging classification and treatment schedule. Stage 0: Patients who have very early HCC are optimal candidates for resection. Stage A: Patients who have early HCC are candidates for radical therapies (resection and ablation, liver transplantation, or percutaneous treatments). Stage B: Patients who have intermediate HCC may benefit from chemoembolization. Stage C: Patients who have advanced HCC may receive new agents in the setting of a randomized, controlled trial. Stage D: Patients who have end stage disease receive symptomatic treatment. Abbreviations: LDLT, living-related donor liver transplantation; PEI, percutaneous ethanol injection; RF, radiofrequency. Source: From Ref. (30).

Primary Resection Therapy LR for HCC has come a long way from early attempts with more than 50% mortality and no 5-year survival. With better understanding of tumor morphology, availability of advanced imaging modalities, patient selection, greater understanding of liver anatomy, improvement in surgical techniques, intraoperative ultrasound scan, and well-trained dedicated liver surgeons has resulted in up to 70% 5-year survival rates comparable to OLT, but a recurrence of 40% to 70% still represents a major cause of death (8,16). LR can be performed as a wedge resection, segmentectomy, or major resection. The extent of the liver resection is dependent on the size of the tumor, whether it is unifocal or multifocal and the presence or absence of cirrhosis in the residual liver. A peripherally located small lesion especially in segment 2 or 3 of liver can be safely resected either laparoscopically or by open resection. But in the presence of cirrhosis, wedge resection or non-anatomical resection in a rigid liver can be difficult and associated with significant blood loss. A hemi hepatectomy can achieve a good tumor clearance with least postoperative complications in lesions measuring 2 to 3 cm. This could be managed either laparoscopically or by handport assisted laparoscopic surgery (17). Deeply seated lesions or large single lesions measuring 3 to 5 cm with no evidence of portal hypertension need major

hepatectomy. Preoperative portal vein embolization (PVE) can be used to increase the functional residual volume of liver. Though theoretically this could overcome the problems of postoperative hepatic decompensation, barring few smaller studies, randomized controlled trials have not shown benefit from PVE (18). The ratio of functional residual volume to total liver volume should be more than 25% in non-cirrhotic livers and >40% in cirrhotic livers. In high-volume centers and in the Far East, major hepatectomies are undertaken with minimal postoperative complication rates. Tumor recurrence following primary resection has an incidence of 70%. Recurrence is more common after resection in the cirrhotic liver due to the ongoing process of carcinogenesis. It is more common in multifocal lesions, vascular invasion, and positive resection margin. Salvage OLT can be used for those with recurrence following resection, although patients will still have to fulfil the criteria for transplant. Resection Prior to OLT LR can be used to help select patients for OLT (19). This gives an opportunity to examine not only the surrounding liver but also the specimen for histo-pathological examination. A tumor with adverse morphological features (such as satellite nodules, vascular invasion) could preclude a patient for OLT even though it meets the CMC. Conversely, a large tumor which

194

RESECTION FOR HEPATOCELLULAR CARCINOMA
falls just outside the CMC, but with good prognostic features could be considered for transplant. With more experience in managing this condition, there is constant urge to push the boundaries of CMC. Resection to Bridge the Gap Prior to OLT Tumor progression in patients waiting for OLT is a common problem, especially in aggressive tumors. In centers with a long wait for OLT, traditionally transarterial chemo-embolization (TACE), percutaneous ablation with ethanol injection (PEI), or radiofrequency (RFA) is used. The amount of tumor necrosis cannot be accurately quantified. Moreover, inadequate tumor necrosis can lead to tumor recurrence following OLT. LR can be used instead to bridge the gap prior to OLT (20). LR is better at tumor control than either TACE or RFA while waiting for OLT. This strategy is restricted to Child A and to a lesser extent in Child B and subsequent OLT can be technically challenging. Liver Transplant In patients with early HCC which are unresectable due to underlying chronic liver disease, OLT offers the best possible outcome. This not only removes the tumor but also the underlying causative factor. In carefully selected patients who meet the CMC, 5-year survival rates in excess of 70% can be achieved (11,14). With a paucity of donor liver, especially in the Far East where the incidence of HCC is high, tumor progression leads to dropout from the waiting list. Though it is difficult to ascertain the exact rate of dropout, about 22% on the waiting list for OLT drop out in the first year due to tumor progression (5,21). Tumors with more than two nodules on presentation and those measuring >3 cm have a higher likelihood of drop out from the waiting list. OLT can be used as (a) primary therapy and (b) salvage OLT Primary Therapy Primary OLT without any pre-transplant treatment, in patients with early HCC who meet the CMC within first 6 months of diagnosis, is the ideal treatment. But due to scarcity of available donor liver, this is not always possible. Traditionally, in the West, DDLT is the method of choice. To make the status of patient amenable for OLT while on the waiting list and to prevent drop out various adjuvant therapies can used. Commonly used are TACE, RFA, and to a lesser extent LR. Furthermore, LDLT has been used, more so in the Far East to overcome the lack of donor organs. The survival of the graft in low-volume liver transplant in LDLT is dependent not only on the extent of ischemia–reperfusion injury but also on the presence or absence of portal hypertension in the recipient. Early interest in the West has not been sustained due to adverse publicity following donor mortality risk. Even in the Far East, LR seems to be initial choice of therapy with LDLT being used in case of tumor recurrence. Pushing the boundaries of CMC has resulted in more and more patients falling just outside the accepted norms being referred to the transplant units. The idea of expanded criteria is mooted on the basis of Metro Rail paradigm “the farther you travel, the higher the price” (22). Salvage OLT In patients who have had LR, PEI, or RFA as the primary treatment survival figures at the end of 5 years is 70% (11,14), 53% (23), and 60% (24), respectively. In those with recurrence, salvage surgery in the form of OLT can be offered. In the Far East, with perpetual shortage of donor liver, LDLT is being used more frequently for salvage OLT. The selection criteria for LDLT are far more liberal than the stringent CMC used for DDLT. Chemical or Thermal Ablation In patients with small tumor located deeply within the liver parenchyma, tumor ablation performed percutaneously or trans-arterially is possible. RFA (24,25) is used routinely not only as a primary therapy but also as a pre-transplant therapy to reduce the dropout rate. The limiting factor is the tumor size and presence of larger vessel close to the tumor with complications occurring in 8% to 23% including abscess formation, biliary injury, and a potential for tumor seeding along the track. PEI (26) is a cheaper alternative with fewer side effects, but the use is limited by tumor size given the fact that best results are seen for tumors <2 cm. Both provide a good cumulative survival benefits.

intermediate and advanced stage hcc
Those patients who have larger asymptomatic tumor which does not fall into CMC category, Child B, compensated chronic liver disease and the absence of extra hepatic spread is considered to have intermediate stage HCC. LR though controversial has been used as an initial therapy option in large HCC with comparable outcomes (27). TACE and RFA either exclusively or in combination have been used to downstage HCC with good results (28). This could be used not only as a prognostic indicator for post transplant outcome, but also for selection of patients for OLT. Application of expanded criteria such as UCSF still needs full validation and has been used in relatively few centers. However, some centers do routinely offer primary OLT with acceptable 5-year survival figures. In the Far East, LR is being offered as a first-line therapy with salvage OLT being used for recurrence. Patients with unresectable HCC with vascular invasion and/or extra hepatic spread are considered to have advanced HCC. Treatment for advanced HCC is restricted to TACE, RFA, or radio sphere embolization. TACE has shown significant benefit in unresectable HCC with good response rates. Patients have a transient post-embolization syndrome with pain, fever, and transient raise in liver enzymes. Major complications such as ischemic necrosis of gall bladder, liver abscess, and biliary stricture are rare. Systemic chemotherapy is rarely being used because of poor response rates. Furthermore, in patients with cirrhosis, hypersplenism, worsening of portal hypertension, major variceal bleeding, or bleeding from gastrointestinal tract and onset of encephalopathy are some of the serious complications

references
1. Okuda K, Fujimoto I, Hanai A, et al. Changing incidence of hepatocellular carcinoma in Japan. Cancer Res 1987; 47: 4967–72. 2. Ross, RK, Yuan JM, Yu MC, et al. Urinary aflatoxin biomarkers and risk of hepatocellular carcinoma. Lancet 1992; 339: 943–6

195

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
3. Okuda K. Clinical presentation and natural history of hepatocellular carcinoma and other liver cancers. In: Okuda K, Tabor E, eds. Liver Cancer. New York: Churchill Livingstone, 1997: 1–12. Forner A, Hessheimer AJ, Real MI, et al. Treatment of hepatocellular carcinoma. Crit Rev Oncol Hematol 2006; 60(2): 89–98. Makuuchi M, Sano K. The surgical approach to HCC: our progress and results in Japan. Liver Transpl 2004; 10(2 Suppl 1): S46–52. Spangenberg HC, Thimme R, Blum HE. Serum markers of hepatocellular carcinoma. Semin Liver Dis 2006; 26(4): 385–90. Jacobs JE, Birnbaum BA. Computed tomography imaging for focal hepatic lesions. Semin Roentgenol 1995; 30: 308–323. Cormier JN, Thomas KT, Chari RS, et al. Management of hepatocellular carcinoma. J Gastrointest Surg 2006; 10(5): 761–80. Vauthey J, Lauwers G, Esnaola N, et al. Simplified staging for hepatocellular carcinoma. J Clin Oncol 2002; 20: 1527–36. Okuda K, Ohtsuki T, Obata H, et al. Natural history of hepato cellular carcinoma and prognosis in relation to treatment: a study of 850 patients. Cancer 1985; 56: 918–28. Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999; 19: 329–38. Bruix J, Castells A, Bosch J, et al. Surgical resection of hepatocellular carcinoma in cirrhotic patients: prognostic value of preoperative portal pressure. Gastroenterology 1996; 111(4): 1018–22. Ringe B, Pichlmayr R, Wittekind C, Tusch G. Surgical treatment of hepatocellular carcinoma: experience with liver resection and transplantation in 198 patients. World J Surg 1991; 2: 270–85. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334: 693–699. Yao FY, Ferrell L, Bass NM, et al. Liver transplantation for hepatocellular carcinoma: Expansion of the tumor size limits does not adversely impact survival. Hepatology 2001; 33: 1394–1403. Lopez PM, Villanueva A, Llovet JM. Systematic review: evidence-based management of hepatocellular carcinoma–an updated analysis of randomized controlled trials [review]. Aliment Pharmacol Ther 2006; 23(11): 1535–47. Cherqui D, Laurent A, Tayar C, et al. Laparoscopic liver resection for peripheral hepatocellular carcinoma in patients with chronic liver disease: mid-term results and perspectives. Ann Surg 2006; 243(4): 499–506. 18. Kianmanesh R, Regimbeau JM, Belghiti J. Selective approach to major hepatic resection for hepatocellular carcinoma in chronic liver disease. Surg Oncol Clin N Am 2003; 12(1): 51–63. 19. Belghiti, J, Carr BI, Greig PD, Lencioni R, Poon RT. Treatment before liver transplantation for HCC. Ann Surg Oncol 15(4): 993–1000. 20. Sala M, Fuster J, Llovet JM, et al. High pathological risk of recurrence after surgical resection for hepatocellular carcinoma. An indication for salvage transplantation. Liver Transpl 2004; 10: 1294–300. 21. Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology 1999; 30: 1434–40. 22. Yao FY. Liver transplantation for hepatocellular carcinoma: beyond the Milan. Crit Am J Transplant 2008; 8: 1982–9. 23. Livraghi T, GiorgioA, Marin G, et al. Hepatocellular carcinoma and cirrhosis in 746 patients : long term results of percutaneous ethanol injection. Radiology 1995; 197: 101–8. 24. Choi D, Lim HK, Rhim H, et al. Percutaneous radiofrequency ablation for early-stage hepatocellular carcinoma as a first-line treatment: long-term results and prognostic factors in a large single-institution series. Eur Radiol 2007; 17(3): 684–92. 25. Lencioni R, Pina CD, Bartolozzi C. Percutaneous image-guided radiofrequency ablation in the therapeutic management of hepatocellular carcinoma. Abdom Imaging 2005; 30(4): 401–8. 26. Ebara M, Okabe S, Kita K, et al. Percutaneous ethanol injection for small hepatocellular carcinoma: therapeutic efficacy based on 20-year observation. J Hepatol 2005; 43(3): 458–64. 27. Pandey D, Lee K-H, Wai C-T, Wagholikar G, Tan K-C. Long term outcome and prognostic factors for large hepatocellular carcinoma (10 cm or more) after surgical resection. Ann Surg Oncol 14(10): 2817–23. 28. Cheng BQ, Jia CQ, Liu CT, et al. Chemoembolization combined with radiofrequency ablation for patients with hepatocellular carcinoma larger than 3 cm: a randomized controlled trial. J Am Med Assoc 2008; 299(14): 1669–77 29. Cherqui D, Laurent A, Mocellin N, et al. Liver resection for transplantable hepatocellular carcinoma: long term survival and role of secondary liver transplantation. Ann Surg 2009; 250(5): 738–46. 30. Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003; 362: 1914.

4. 5. 6. 7. 8. 9. 10.

11. 12.

13.

14.

15.

16.

17.

196

21

Treatment of laparoscopically discovered gallbladder cancer Jason K. Sicklick, David L. Bartlett, and Yuman Fong
diagnoses each year (13). It has an annual incidence of 1.3 per 100,000 in females and 0.8 per 100,000 in males, with an average incidence of 1.2 cases per 100,000 population per year (14). This cancer is responsible for approximately 2,800 deaths per year. The most obvious associated conditions for gallbladder cancer are gallstone disease and chronic cholecystitis. Between 75% and 98% of all patients with carcinoma of the gallbladder have cholelithiasis (15). Most importantly, gallbladder cancer will be found once in every 100 cases of presumed gallstone disease. The natural history of gallbladder cancer has been defined through many retrospective reviews and large surveillance programs. The overall 5-year survival is consistently less than 5%, with a median survival of 5 to 8 months. Piehler et al. (5) reviewed 5,836 cases in the world’s literature from 1960 to 1978. They reported an overall 5-year survival of 4.1% and a 1-year survival of 11.8%. Only about 25% were resectable for cure, and of those resected for cure, 16.5% survived 5 years. Perpetuo et al. (4) reviewed the M.D. Anderson Cancer Center experience with gallbladder cancer over 36 years and reported a 5-year survival rate of less than 5% and median survival of 5.2 months. Cubertafond et al. (16) reported the results of a French Surgical Association Survey of 724 carcinomas of the gallbladder. They reported a median survival of 3 months, a 5-year survival rate of 5%, and a 1-year survival rate of 14%. They observed no differences among the different surgical procedures adopted, and concluded that no progress had been made in the treatment of gallbladder cancer. A survey of gallbladder cancer in Wessex, United Kingdom, revealed only four patients out of 95 surviving from 8 to 72 months after the time of diagnosis (17). A review of gallbladder cancer from Australia revealed a 12% 5-year survival rate, with all survivors having stage I or II disease. The median survival for patients with stage III or IV disease was only 46 days (18). SEER data from the United States demonstrated similarly unsatisfying results, with only marginal improvement over earlier studies with median overall survival time being 10 months (95% CI 9 to 11 months), as well as 1-year, 2-year, 3-year, and 5-year overall survival rates of 46%, 30%, 23%, and 17%, respectively, in 4,180 patients (19). A multi-institutional review from Japan, on the other hand, reported a 50.7% 5-year survival for 984 patients undergoing radical resection versus 6.2% for 702 patients undergoing more conservative management (20). These results suggest that it may be possible for surgery to have a role in changing the natural history of this tumor. Therefore, it is clear that radical liver resection, or extended liver resection, in gallbladder cancer does have survival benefit in selected cases (7,21,22). Despite this data, it is important to emphasize that there has been only one small randomized, prospective trial on the treatment of gallbladder cancer. Moreover, there are no

introduction
Traditionally a great deal of pessimism has been associated with the treatment of gallbladder cancer (1). There are many reasons for the skepticism associated with this disease entity since its first description in 1778 (2). Foremost is the aggressive nature of this cancer for dissemination. Gallbladder cancer spreads early by direct invasion into the liver, as well as through lymphatics to regional nodes, by peritoneal dissemination to produce carcinomatosis, and by hematogenous means to produce synchronous liver and other distant metastases. As a result, gallbladder cancer often presents at a time when surgical excision is either no longer possible or is technically difficult while alternative therapies including chemotherapy and radiation are generally ineffective. Therefore, it is not surprising that in 1924 Blalock recommended that surgery be avoided for gallbladder cancer if the diagnosis could be made preoperatively (3). In fact, until recently, the 5-year survival in most large series was less than 5%, and the median survival was less than 6 months (4,5). In the modern era, liver resection has become increasingly safe. More recent experience has demonstrated that radical surgery may be a sensible and potentially curative option in the treatment of this disease (6,7). The data have demonstrated that surgical excision is the treatment option of choice for those patients whose gallbladder cancers are confined to the local region of the liver and porta hepatis (8–10). Beginning in late 1980s, when the techniques for laparoscopic cholecystectomy were introduced, a new presentation for gallbladder cancer was conceived. With the advent and popularization of laparoscopic cholecystectomy, increasing number of cases of gallbladder cancer were being discovered laparoscopically. Currently, approximately 750,000 cholecystectomies are performed in the United States annually for presumed calculous biliary disease (11). Since gallbladder cancer is encountered in 1% of cholecystectomies for cholelithiasis (7), a significant number of patients will present with this clinical scenario. Therefore, meticulous inspection of the gallbladder should be mandatory (12). The current chapter will review data addressing the utility of subsequent radical resection for laparoscopically discovered gallbladder cancer. We will begin with a brief general review of gallbladder cancer, which focuses on the natural history and results of surgical treatment. Summarized data on presentation and results of treatment for laparoscopically discovered disease will be discussed, including the differences of discovery by an open rather than laparoscopic operation.

epidemiology
Gallbladder cancer is the most common biliary tract malignancy and the fifth most common gastrointestinal malignancy in the United States. In fact, there are approximately 5,000 new

197

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
randomized trials comparing extended resection to conservative management. The routine use of more radical resections, including those of segments IV and V and the common bile duct, despite a negative cystic duct margin, has gained some popularity. There is no randomized data in the literature to show that this is mandatory in patients with Tis, T1, or T2 disease where a negative margin is obtained. Tumor may then pass to lymph nodes posterior to the pancreas, portal vein, and common hepatic artery. Advanced disease may ultimately reach the interaortocaval, celiac axis, and superior mesenteric artery lymph nodes. Gallbladder cancer also has a remarkable propensity to seed and grow within the peritoneal cavity, which may account for its ability to grow along the tracts of needle biopsy sites and laparoscopic port sites. Growth in those sites may be further exacerbated by bile spillage during laparoscopic cholecystectomy (31,32). In fact, another group demonstrated that the incidence of port site/ peritoneal recurrence was higher in patients with gallbladder perforation (3/7, 43%) than in those without (0/21, 0%; p = 0.011) (33). The long-term survival was worse in the seven patients with gallbladder perforation (cumulative 5-year survival of 43%) as opposed to those without perforation (cumulative 5-year survival of 100%; p < 0.001). Hematogenous spread is less common but will present most often as noncontiguous liver metastases, and more rarely as lung or brain metastases. At postmortem examination, Perpetuo et al. (4) reported that 91% of patients had liver metastasis and 82% had intra-abdominal lymph node involvement, 60% had peritoneal spread, 32% had lung metastases, and 5% had brain metastases.

pathology
At early stages, gallbladder carcinomas are difficult to grossly differentiate from chronic cholecystitis. As a result, they are often found incidentally upon pathologic section. Even at late stages, when the tumor can obstruct the common bile duct and produce jaundice, gallbladder cancer is often mistaken for benign disease since associated gallstones and Mirizzi’s syndrome are common (23). Therefore, a long-term obstruction of the mid-common bile duct should be considered a gallbladder cancer until proven otherwise. Tumors that arise in the neck and within Hartmann’s pouch may also infiltrate the common hepatic duct, making them clinically and radiographically indistinguishable from hilar cholangiocarcinomas. Approximately 60% of tumors originate in the fundus of the gallbladder, 30% in the body, and 10% in the neck (24). These tumors grow most commonly in a diffusely infiltrative form (25), with a tendency to involve the entire gallbladder, and spread in a subserosal plane, which is the same as the surgical plane used for routine cholecystectomy. If such a tumor is unrecognized at the time of surgery, a simple cholecystectomy will not completely excise the disease and may lead to dissemination of tumor. Although the nodular type of tumor may show early invasion through the gallbladder wall into the liver or adjacent structures, it may be easier to control surgically than the infiltrative type because the margins are better defined. The papillary growth pattern has the best prognosis because even large tumors have only minimal invasion of the gallbladder wall (14). The most common histologic cell type of gallbladder cancers is adenocarcinoma (26). Other rare subtypes of gallbladder cancer include papillary carcinoma, mucinous carcinoma, clear cell carcinoma, signet ring carcinoma, squamous cell carcinoma, small cell (oat cell) carcinomas (27), adenosquamous tumors (28), sarcomas, carcinosarcoma, carcinoid, lymphoma, melanoma, and gastrointestinal stroma tumors (GIST) (29,30).

staging
The multitude of staging systems (Table 21.1) used for this disease has made it difficult to compare treatment results. Nevin et al. (34) originally classified patients into five stages based primarily on the thickness of invasion, and combined patients with direct liver extension or distant metastases into stage V. Donahue et al. (35) modified the Nevin system to include tumors with contiguous liver invasion as stage III and noncontiguous liver involvement as stage V. Stage IV continued to include lymph node metastases. The Japanese Biliary Surgical Society staging system separated tumors into four stages according to the degree of lymph node metastasis, serosal invasion, peritoneal dissemination, hepatic invasion, and bile duct infiltration. The main weakness of this staging system is that lymph node metastases are considered in the same stage as microinvasion of the liver. Despite these various systems, the most common system for evaluating gallbladder cancer worldwide has been the American Joint Committee on Cancer (AJCC) TNM staging system for gallbladder cancer (26). Unfortunately, the 6th edition of the AJCC staging system underwent radical changes due to a desire to match the staging of other biliary cancers. The staging system was therefore not consistent with data. A recent paper documented the deficiencies of the 6th edition staging system using 10,705 cases of this disease from the National Cancer Database (36). Thus, the new 7th edition staging will revert to a system much more in line with past staging (Table 21.1). According to this system, tumors without perimuscular invasion are considered stage I. Tumors with invasion into the perimuscular connective tissue but without extension beyond the serosa or into the liver are considered stage II. Tumors that perforate the serosa and/or directly invade the liver and/or adjacent structures, such as the stomach, duodenum, colon, pancreas, omentum, or extrahepatic biliary tree are stage IIIA if

patterns of spread
Gallbladder carcinoma commonly disseminates by four modes: 1. Direct extension and invasion of the liver and adjacent organs 2. Lymphatic spread 3. Shedding and peritoneal dissemination, and 4. Hematogenous spread to distant sites. The gallbladder lies on segments IVb and V of the liver and these segments are involved early in tumors of the fundus and body. Direct extension into the portal structures (i.e., portal vein, hepatic artery, and bile duct) commonly occurs and is a major cause of symptoms. Lymphatic spread is also common and most often involves cystic and pericholedochal nodes.

198

TREATMENT OF LAPAROSCOPICALLY DISCOVERED GALLBLADDER CANCER
Table 21.1 Summary of Most Commonly Used Staging Systems
Stage I AJCC 5th edition (TNM) Mucosal (T1N0M0) AJCC 6th edition (TNM) IA: Mucosal or muscular invasion (T1N0M0) IB: Perimuscular invasion (T2N0M0) IIA: Perforate the serosa and/ or directly invade the liver and/or adjacent structures (T3N0M0 IIB: Tumors with regional nodal lymph node metastases but no invasion of the main portal vein, hepatic artery, or multiple extrahepatic organs/ structures (T1-3N1M0) Tumor invades the main portal vein, hepatic artery, or multiple extrahepatic organs/ structures. (T4NxM0) (Distant metastases (TxNxM1) Modified Nevin In situ carcinoma Japanese Confined to gallbladder capsule N1 lymph nodes; minimal liver orbile duct invasion Proposed 7th edition AJCC Mucosal (T1N0M0)

II

Muscular invasion (T2N0M0)

Mucosal or muscular invasion

Muscular invasion (T2N0M0)

III

Liver invasion <2 cm; lymph node mets (T3N1M0)

Transmural direct liver invasion

N2 lymph nodes; marked liver or bile duct invasion Distant metastases

IV

V

(1) Liver invasion >2 cm (T4N0M0, TxN1M0 (2) Distant metastases (TxN2M0, TxNxM1) N2 lymphadenopathy [peripancreatic (head only), periduodenal, perioportal, celiac, superior mesenteric, or paraaortic nodes] [—]

Lymph node metastasis

Transmural (T3N0M0), or T-3 with nodal involvement Metastatic disease, or vascular involvement with nodal metastases (T4N1M0)

[—]

Distant metastases

[—]

there is no regional lymph node metastasis. Stage IIIB tumors have nodal metastases but no vascular invasion. Stage IVA includes those patients with vascular invasion, and stage IVB includes patients with distant metastases or those with vascular invasion and nodal metastases.

clinical presentation
The clinical presentation of gallbladder cancer is often identical to biliary colic and/or chronic cholecystitis, making it difficult to diagnose preoperatively. It is also difficult to easily distinguish gallbladder cancer from benign gallstone disease from blood tests. Elevated alkaline phosphatase and/or bilirubin levels are found in cases of advanced tumors, but may also be found for patients with gallstones. A CEA greater than 4 ng/ml is 93% specific for the diagnosis of gallbladder cancer, but is only 50% sensitive (37). A serum Ca 19–9 level (38) greater than 20 units/ml has 79.4% sensitivity and 79.2% specificity, but neither test is routine in patients suspected of having benign disease. Vigilance for cancer in examination of preoperative sonograms or CT scans is essential. Any mass or polyp associated with the gallbladder (Fig. 21.1) or the presence of a porcelain gallbladder should raise concerns of a gallbladder cancer.Figure 21.1 It is often difficult to make the diagnosis of gallbladder cancer based upon clinical history as it often presents similarly to

Figure 21.1 CT scan demonstrating a papillary carcinoma of the gallbladder. This patient was subjected to a laparoscopic cholecystectomy in spite of this scan and required a subsequent reoperation for a potentially curative radical resection.

benign calculous disease. In a report of 42 laparoscopically discovered gallbladder cancers, in only two of the cases did the laparoscopic surgeon suspect a cancer prior to the surgical procedure (39). The laparoscopic procedure consisted of 19

199

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
cases of laparoscopic cholecystectomy, one laparoscopic cholecystectomy with intraoperative cholangiogram, and six laparoscopic biopsies only. There were 16 cases that were converted to open procedures, including 12 open cholecystectomies, three open cholecystectomies with common bile duct exploration, and one cholecystectomy with hepaticojejunostomy. Even at the conclusion of the laparoscopic procedure, in only 20 of the 42 cases (47.6%) was there any suspicion of cancer. This underscores the difficulties in diagnosing gallbladder cancer and the ease with which this aggressive malignancy is confused with benign stone disease. Recent studies (40–42) have evaluated the role of meticulous inspection of gallbladder specimens at the time of laparoscopic cholecystectomy. Akyurek et al. (40) inspected 548 laparoscopic cholecystectomy specimens by making an incision in the gallbladder wall and palpating the mucosa after removing the gallbladder from the abdominal cavity. If an abnormal mucosa was observed or palpated, it was marked and then histopathologic examination was performed. They identified 50 cases to be suspicious and histopathologic examination of frozen sections revealed incidental pathologies in 15 specimens. Moreover, five of these specimens had gallbladder cancers. The sensitivity and specificity of the procedure was 78.9% and 93%, respectively, suggesting that this is a simple method for identifying incidental gallbladder cancers and may allow for a definitive resection to be performed at the time of the initial operation. In another study of 983 cases, 11 cancers were identified. Based upon frozen sections, cancer was diagnosed in 40% of Tis lesions, whereas it was found in 83% of T2 or T3 lesions, which required conversion to a more radical operation (41). A larger cohort of 1452 patients identified four patients with gallbladder cancers and, in all cases, there was either preoperative or intraoperative suspicion (43). Together, these studies would suggest that careful inspection and selective evaluation of suspicious gallbladder lesions using frozen sections should be performed. traditional methods for such assessment. For small tumors, the pattern of obstruction seen on PTC or ERCP may assist in differentiating gallbladder cancer from other tumors or benign disease (46). In the last decade, MR cholangiopancreatography (MRCP) (Fig. 21.2) has improved to become a suitable, noninvasive substitute for direct cholangiography (47). Historically, clinical suspicion for main portal venous and/or hepatic arterial involvement by tumor usually prompted angiography to definitively demonstrate resectability. Improvements in Doppler ultrasound and in MR angiography provide noninvasive substitutes for such assessment. We will often assess a patient presenting with known gallbladder cancer with a single MR scan (48). Detailed information on liver involvement, biliary extension, vascular proximity and involvement, and nodal disease can all be gleaned from this single noninvasive test. With the quality of current cross-sectional imaging, it is rare that direct cholangiography or angiography is necessary. Recently, a role for fluorodeoxyglucose positron emission tomography in the management of patients with gallbladder cancer has been established. This test is useful in diagnosing nodal, peritoneal, and distant metastases (49). In a series of 31 patients with gallbladder cancer, 7 (23%) had therapy altered by staging with FDG-PET.

surgical management
A wide range of operations has been advocated for gallbladder cancer from simple cholecystectomy to combined extended hepatectomy, common bile duct resection, and pancreaticoduodenectomy (50). Debate still exists as to the extent of surgery (51). A survey of prominent gastrointestinal surgeons in the United States indicated that 49% recommended lymph node dissection and 64% recommended some form of liver resection for stage T2–4 disease. The cynical attitude toward this disease is reflected by the recommendation of 21% of surgeons to perform only a simple cholecystectomy for nodepositive disease (52). Studying the earliest stages of the disease, incidental Tis or T1A gallbladder cancer discovered in specimens following

radiologic workup
Most patients with laparoscopically discovered gallbladder cancer will have had an ultrasound performed for suspected cholelithiasis. Review of this ultrasound may provide information concerning liver involvement by tumor, biliary extension of tumor, and/or vascular involvement. However, it is most often that another cross-sectional imaging test is indicated for further assessment of these sites for disease, as well as to assess for presence of nodal disease. The combination of CT scanning and ultrasonography (44) is the most common combination for initial assessment, although MRI scanning can be substituted for CT (45). If the initial assessment suggests evidence of laboratory or radiologic signs of biliary obstruction, assessment of the extent of biliary involvement by another imaging technique may be necessary. Gallbladder cancer can cause obstructive jaundice by direct invasion of the common hepatic duct, or by compression and involvement of the common hepatic duct by pericholedochal lymph nodes. A high correlation between Mirizzi’s syndrome and gallbladder cancer exists (23). Endoscopic or percutaneous cholangiograms (PTC) are the

Figure 21.2 Magnetic resonance cholangiopancreatography demonstrating extent of gallbladder cancer. Extension of tumor within and obstructing the common bile duct is shown with isolation of the left and right hepatic duct. The portal vein (white arrow) is patient and not involved by tumor.

200

TREATMENT OF LAPAROSCOPICALLY DISCOVERED GALLBLADDER CANCER
laparoscopic cholecystectomy does not warrant further surgery if the cancer is limited to the lamina propria-muscularis layer and if a subsequent staging workup is negative. These patients have a 5-year survival rate ranging from 90% to 100% (53). Data would indicate that a potentially curative approach for gallbladder cancer, except for disease at the earliest stages, would require a liver resection and a lymphadenectomy. In the past, some argued that T2 cancers with negative margins may only require a simple cholecystectomy. More recent data suggests that this is not the case. A study from Memorial Sloan-Kettering Cancer Center (MSKCC) demonstrated that even in T2 gallbladder cancer, extended or radical resection affords improved survival over cholecystectomy alone (54). It is clear from pathologic data that T2, T3, or T4 tumors were all associated with greater than a 50% chance of metastases to the regional lymph nodes (Table 21.2). As liver resections have become increasingly safe, increasing numbers of surgical centers are performing radical resections for this disease and data are consequently accumulating that justifies such an aggressive approach. Unless a patient has clear contraindications to resection, including medical comorbidities or unresectable disease, surgical exploration should be attempted. We will review the data supporting radical resection for gallbladder cancer at various stages of disease. Then a discussion of the justification of such treatment in patients with laparoscopically discovered gallbladder cancer will be presented. The most practical way of thinking about laparoscopically discovered gallbladder cancer is to base therapy upon clinical T stage of disease. Not only is there a close correlation of T stage with prognosis, but patients presenting in this setting will usually have had the gallbladder excised and the extent of Table 21.2 Findings Related to T stage of Disease
Stage T2 T3 T4 Total metastases (%) 9 16 16 Peritoneal metastases (%) 12 43 68 Nodal metastases (%) 50 50 66

local disease defined pathologically. Knowing the likelihood of further local, nodal, peritoneal disease will allow for rational therapeutic choices. Tumors Confined to the Muscular Propria (T1 Tumors) There are abundant data to indicate that early gallbladder cancer, which has not penetrated through the muscular layer of the gallbladder, is adequately treated by simple cholecystectomy. Tsukada et al. (55) demonstrated that in 15 cases with T1 lesions, there were no cases with lymph node metastasis. Table 21.3 (6,20,21,28,35,56–61) summarizes results of resection for stage I disease. After simple cholecystectomy alone, the 5-year survival was 78% to 100% (59,62). In a report of 56 patients treated with simple cholecystectomy alone, only two patients recurred and subsequently died of their disease. Both had submucosal spread of the tumor to involve the cystic duct margin (21). When patients present after laparoscopic cholecystectomy with a pathologic diagnosis of T1 gallbladder cancer, a careful review of the pathology is imperative. Care must be taken to verify both negative margins including the cystic duct stump and that there are no areas of deeper invasion. If the gallbladder margin is involved by tumor, a liver resection is required. If the cystic duct stump is involved, an excision of the common bile duct, including the junction with the cystic duct, is indicated. No nodal dissection is necessary. Tumor Invading into the Subserosal Layer (Stage II) By definition, T2 tumors do not transgress the serosal plane. However, the recommended management for T2 disease is an extended or radical cholecystectomy to include a liver resection and regional lymph node dissection including periportal, peripancreatic, and celiac nodes. This recommendation is based on the pattern of spread of disease. In the most common infiltrative form of gallbladder cancer (25), the cancer often spreads in a subserosal plane, which is the same as the surgical plane used for routine cholecystectomy. This results in a higher likelihood of positive margins after simple cholecystectomy. In the review by Yamaguchi and Tsuneyoshi (59), patients had tumor extending into the subserosal layer and 11 of these had positive microscopic margins after simple cholecystectomy. Furthermore, the likelihood of metastatic disease to regional

T2, submucosal invasion; T3, full thickness invasion through gallbladder wall with <2 cm extension into liver; T4, > 2 cm extension into liver. Source: From Ref. (39).

Table 21.3 Actuarial Survival Results Reported In Retrospective Reviews after Resection of Stage I Gallbladder Cancers
Author Ouchi et al. (56) Yamaguchi and Enjoji (28) Donohue et al. (35) Gall et al. (57) Ogura et al. (20) Shirai et al. (58) Yamaguchi and Tsuneyashi (59) Shirai et al. (21) Matsumoto et al. (6) Oertli et al. (60) de Aretxabala et al. (61) Year 1987 1988 1990 1991 1991 1992 1992 1992 1992 1993 1992 N 14 11 6 7 366 39 6 56 38 4 6 32 Procedure Not specified Not specified Simple cholecystectomy: 83% Simple cholecystectomy Not specified Simple cholecystectomy Simple cholecystectomy Simple cholecystectomy Extended cholecystectomy Extended cholecystectomy Simple cholecystectomy Simple cholecystectomy: 69% 3-Year survival (%) 78 100 100 86 87 100 100 100 100 100 100 94 5-Year survival (%) 71.4 Not reported 100 86 78 100 100 100 100 100 100 94

201

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
lymph nodes exceeds 50% (7,39,52). Indeed, it is perhaps this group of T2 lesions which will have the best chance of benefiting from definitive extended re-resection (61). Five-year survival of patients subjected to simple cholecystectomy is 20% to 57% while the survival of patients subjected to radical resection is 70% to 100% (Table 21.4) (63–66). For patients presenting with T2 gallbladder cancer discovered at laparoscopic cholecystectomy, a re-exploration with the intent to perform a liver resection and regional nodal dissection is recommended. During this re-exploration, inflammation from the previous operation will make it difficult to determine the exact extent of disease. Furthermore, all of the laparoscopic port sites should be excised in a full thickness manner. The patients must be informed that final pathology may not demonstrate residual tumor. Advanced Tumors (stages III and IV) Patients with T3 or T4 gallbladder cancer will present after laparoscopic cholecystectomy not only with an obvious pathologically positive margin for tumor, but also with a hepatic mass on cross-sectional imaging. Debate raged in the past regarding justification of radical surgery for such advanced disease. As radical resections have become increasingly safe, reports of long-term survivors after aggressive surgical management are abundant in the literature; Table 21.5 reviews these studies. Onoyama et al. (67) reported a 63.6% 5-year survival for Japanese Biliary Surgical Society stage II and 44.4% 5-year survival for stage III disease after extended cholecystectomy (AJCC 5th edition stage III). They reported a 5-year survival rate of 8.3% for stage IV disease. In addition, they noted a 5-year survival rate of 60% for patients having metastatic disease to N1 nodes. Shirai et al. (21) reported a 45% 5-year survival for patients with node-positive tumors, documenting nine patients surviving over 5 years after radical resection. Gall et al. (57) reported that four of eight patients undergoing curative resection for AJCC stages III and IV gallbladder carcinoma at the initial operation were alive after 81, 50, 13, and 8 months. Our data from MSKCC revealed a median overall survival for the 435 patient cohort of 10.3 months. The median survival for those presenting with stages Ia–III disease was 12.9 months and 5.8 months for those presenting with stage IV disease (68). We previously reported a 67% actuarial 5-year survival for patients with completely resected stage III and 33% 5-year survival for patients with completely resected stage IV tumors (7, 66) These results represent marked alteration of the natural history of this tumor. These data would indicate that radical surgery for advanced gallbladder cancer may be potentially curative (Table 21.5) (64–66). Patients presenting with T3 and T4 disease after laparoscopic cholecystectomy should have imaging performed to rule out signs of unresectable disease, including noncontiguous liver metastases or signs of carcinomatosis. Barring any contraindications to surgery (i.e., medical contraindications to major abdominal surgery, cirrhosis, or insufficient remnant liver volume to maintain adequate hepatic function), patients should be re-explored for radical resection of tumor, which usually requires a major liver resection and regional lymphadenectomy. Re-resection after Laparoscopic Cholecystectomy Data available for re-resection for gallbladder cancer treated initially with open simple cholecystectomy suggest that, for tumors with a depth of penetration greater or equal to the perimuscular coat (i.e., T2), a radical re-resection is warranted (62). However, the prognosis for patients subjected to two operations for gallbladder cancer is thought to be less favorable than for patients treated with a single procedure. Gall et al. (57) reported a median survival of 42 months for patients

Table 21.4 Actuarial Survival Results Reported in Retrospective Reviews after Resection of Stage II Gallbladder Cancers
Author Yamaguchi and Enjoji (28) Donohue et al. (35) Ogura et al. (20) Gall et al. (57) Shirai et al. (58) Yamaguchi and Tsuneyashi (59) Matsumoto et al. (6) Oertli et al. (60) Cubertafond et al.a (16) Bartlett et al. (7) Paquet (107) Shih (63) Kai (64) Jensen (65) D’Angelica (66)
Multi-institutional survey. Chole, cholecystectomy.
a

Year 1988 1990 1991 1991 1992 1992 1992 1993 1994 1996 1998 2007 2007 2008 2008

N 73 12 499 7 35 10 25 9 17 52 8 5 34 9 25 769 196 41

Procedure Not specified 67% Extended chole Not specified 86% Simple chole Simple chole Extended chole Simple chole Extended chole Simple chole 88% Simple chole Extended chole Extended chole Extended chole Simple chole Extended chole Simple chole Extended chole Extended chole

3-year survival (%) 40.1 58 53 86 57 90 36 100 29 20 100 100 49 22 60 40 55 84

5-year survival (%) Not reported 22 37 86 40.5 90 36 100 24 Not reported 88 80 49 22 60 29 42 79

202

TREATMENT OF LAPAROSCOPICALLY DISCOVERED GALLBLADDER CANCER
undergoing a curative resection at the first operation versus 12.5 months for those undergoing a curative resection at a second operation. Our experience over a 10-year period demonstrated a median survival of 15.7 months for those discovered incidentally at laparoscopic cholecystectomy (68). More recent data would suggest that there is no difference in outcomes in patients who undergo laparoscopic cholecystectomy for unsuspected gallbladder cancer (69,70). Moreover, there appears to be no difference in survival or recurrence between patients that have undergone initial open or laparoscopic cholecystectomy (71). However, it is clear that obtaining an R0 resection significantly improves survival in patients undergoing re-resection (72,73). To that end, a study from Johns Hopkins Hospital (63) showed that there was no survival difference between patients who were immediately converted to an open resection when identified to have gallbladder cancer intraoperatively (N = 6) versus those patients who had a completed laparoscopic cholecystectomy and were re-explored at a later point after discovery of a gallbladder cancer at histopathological review (N = 33). This study would suggest that gallbladder carcinoma discovered during a laparoscopic cholecystectomy does not require immediate conversion to an open resection and should be referred to a tertiary care center for further exploration. In a recent series of 206 cases of laparosocpically discovered gallbladder cancer, 136 patients were re-explored (68). Thirtyfive of these patients were found to have no cancer on exploration or in the final re-excision specimen, while 101 had residual tumor re-excised. Of note, those without residual disease still had a 50% chance of eventually dying of cancer. The 5-year survival of the patients with no residual disease was 63% and median survival 72 months. Those completely resected of residual disease had a 5-year survival of 22% and a median survival of 19 months. Those with incompletely resected residual disease had a median survival of 12.7 months, and no patient survived 5 years. Liver Resection Except for the patient with T1 tumors who has a positive cystic duct margin, because of the possibility of residual disease remaining within the gallbladder bed, all other patients undergoing re-exploration for re-resection should have some form of liver resection (i.e., a radical or extended cholecystectomy). Even patients with T2 tumors have a likelihood of residual gallbladder bed disease because the most common plane for simple cholecystectomy is subserosal. Recommendations for liver resection for gallbladder cancer have ranged from a limited wedge excision of 2 cm of liver around the gallbladder bed to routine extended right hepatic lobectomy. We prefer an anatomic segment IVb and V resection when possible, because this anatomic operation allows the greatest chance of tumor clearance while minimizing the amount of functional liver removed. In cases of previous cholecystectomy, such a limited resection may not be possible. Scars from the previous surgery may be difficult to distinguish from tumor and a more radical resection may be necessary to ensure complete eradication of disease. It

Table 21.5 Actuarial Survival Results reported in Retrospective Reviews after Resection of Stage III and IV Gallbladder Cancers
Author Matsumoto et al. (6) Chijiiwa and Tanaka (102) Onoyama et al. (67) Bartlett et al. (7) Ouchi et al. (56) Nakamura et al. (103) Donohue et al. (35) Gall et al. (57) Shirai et al. (58) Ogura et al. (20) Todoroki et al. (92) Nimura et al. (50) Matsumoto et al. (6) Chijiiwa and Tanaka (102) Onoyama et al. (67) Bartlett et al. (7) Kai (64) D’Angelica (66) Jensen (65) Year 1992 1994 1995 1996 1987 1989 1990 1991 1992 1991 1991 1991 1992 1994 1995 1996 2007 2008 2008 N 8 12 12 8 12 13 17 8 20 453 27 14 27 11 14 7 16 63 119 Stage III III III III III/IV III/IV III/IV III/IV III/IV IV IV IV IV IV IV IV III/IV III/IV III 3-Year survival (%) 38 80 44 63 17 16 50 50 – 18 7 10 25 11 8 25 40 45 18 5-Year survival (%) – – 44 63 – 16 29 – 45 8 – – – – 8 25 36 28 9 Comments Majority with common bile duct resection Extended resections only Extended resections only Extended resections only Extended resections only Includes 5 HPD, 10 extended hepatectomy Extended resections only Includes only curative resection at initial surgery All patients have lymph node metastases Multi-institutional series with 25% simple cholecystectomy All patients had IORT All patients underwent HPD Includes 3 HPD, 6 extended hepatectomy, 11 CBD resection Extended resections only Japanese staging Long-term survivors with no lymph node metastases

CBD, common bile duct; HPD, hepatopancreatoduodenectomy; IORT, intraoperative radiation therapy.

203

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
must be emphasized that complete excision of any suspicious areas must be performed since any residual tumor will result in recurrence that is usually rapidly fatal. Often, therefore, a right lobectomy or trisegmentectomy will be necessary. Lymph Node Dissection Studies of lymphatic spread of gallbladder cancer have been published and reviewed (74). Recommendations for lymph node dissection for gallbladder cancer have ranged from excision of the cystic duct node alone to en bloc portal lymphadenectomy with pancreaticoduodenectomy (6). Justification for more radical procedures comes from the propensity for early spread to the superior, anterior, and posterior pancreaticoduodenal nodes. Combined liver and pancreatic resections have high operative mortalities of near 20%, and are not justified by long-term results. Portal lymphadenectomy for tumor penetrating the gallbladder beyond T2 is supported by our findings of positive nodes in over 50% of patients with T2, T3, or T4 gallbladder cancer (39). We also believe that an adequate portal lymphadenectomy requires resection of the common bile duct. Particularly in patients who have just had a cholecystectomy, the portal lymph nodes are often intimately associated with the bile duct. Resection of the common bile duct greatly facilitates nodal clearance. In general, a full Kocher maneuver should be performed. The lymphatic tissue should then be dissected behind the duodenum and pancreas and subsequently swept superiorly. Any interaortocaval nodes or superior mesenteric nodes should be included in the specimen if possible. The common bile duct should be transected as it courses posterior to the duodenum into the pancreas. The portal vein and hepatic artery should be skeletonized and all tissue swept superiorly along with the transected duct. At the confluence of the right and left hepatic ducts, the common bile duct should be divided again (assuming the cystic duct does not enter the right hepatic duct). A Roux-en-Y hepaticojejunostomy should be performed to re-establish biliary– enteric continuity. Laparoscopic Port Sites A number of studies have demonstrated the propensity of tumor to recur in the laparoscopic port sites because gallbladder cancer has a great potential for peritoneal seeding and dissemination (Table 21.6) (75–82). Indeed, our preliminary report of the first ten patients we encountered with laparoscopically discovered gallbladder cancer included two patients in whom the tumor recurrence was found in a port site (79). The incidence of peritoneal metastases is higher than reported in the pre-laparoscopic era. One report found a 32% recurrence rate appearing as a new or enlarging abdominal wall mass on physical examination and/or CT scanning for followup of disease (83). Another study by Paolucci (84) found 174 cases of port site metastasis after laparoscopic cholecystectomy and 12 recurrences in the surgical scar after converted or open cholecystectomy. This report found a 14% incidence of port site metastases at 7 months after laparoscopic cholecystectomy for cancer. Therefore, it has become our standard practice to excise laparoscopic port sites at the re-exploration. During Table 21.6 Recurrence of Tumor in Laparoscopic Port Sites
Author Drouard et al. (75) Clair et al. (76) Landen (77) Fligelstone et al. (78) Fong et al. (79) Nduka et al. (80) Nally and Preshaw (81) Kim and Roy (82) Antonakis et al. (42) Cucinotta et al. (108) Hamila et al. (109) Paolucci (84) N 1 1 1 1 2 1 1 1 3 4 174 Port Umbilical Umbilical Umbilical Umbilical Umbilical Epigastric Umbilical Umbilical 0 Not specified Not specified Various

that operation, care must be taken to perform a full abdominal inspection to rule out peritoneal disease (85). Whether such excision of port sites is useful requires further investigation, since port recurrence may be just a marker for diffuse peritoneal dissemination of disease. Complications The operations described above are extensive procedures with substantial risks. In particular, the majority of patients undergoing treatment for gallbladder cancer are in their seventh or eighth decade of life and may be at increased risk as a consequence of concomitant medical comorbidites. In a multi-institutional review of 1686 gallbladder cancer resections from Japan, a comparison of morbidity by procedure was made (20). A morbidity of 12.8% was reported for cholecystectomy, 21.9% for extended cholecystectomy, and 48.3% for hepatic lobectomy. The mortality rates were 2.9%, 2.3%, and 17.9%, respectively. There were 150 hepatopancreatoduodenectomies for gallbladder cancer, with a 54% morbidity rate and a 15.3% mortality rate. The morbidity and mortality rates of major liver resections have decreased in later reports, even in the aged population (86). In our report of re-resection for laparoscopically discovered gallbladder cancer, all resected patients were subjected to some form of liver resection and the operative mortality was 5% (39). The most common complications are bile collections, liver failure, intra-abdominal abscess, and respiratory failure. The risk of resection for each patient and for each type of resection needs to be weighed against the chance of benefiting from the procedure based on the stage of disease.

adjuvant therapy
Because of the rarity of gallbladder cancer in general, as well as the rarity of completely resected disease, there is only one prospective, randomized trial examining the utility of adjuvant therapy for gallbladder cancer. This trial assessed 5-year overall survival in patients following noncurative resection who received postoperative adjuvant chemotheraoy using mitomycin C and 5-FU. Survival was improved with adjuvant therapy (26% vs. 1%, P = 0.03). (87) However, most data available is derived from retrospective series. Conclusive data do not support the routine use of chemotherapy (88–90).

204

TREATMENT OF LAPAROSCOPICALLY DISCOVERED GALLBLADDER CANCER
Data regarding radiation therapy are more substantial, but still far from conclusive (91). Todoroki et al. (92) examined intraoperative radiation therapy after complete resection for stage IV gallbladder cancer. They reported a 10.1% 3-year survival for patients receiving intraoperative radiation therapy versus 0% for surgery alone. Bosset et al. (93) examined postoperative external beam irradiation after complete resection in seven patients. They concluded that it was a safe treatment, and five of the seven patients were still alive at a median follow-up of 11 months. Hanna and Rider (94) reported radiation therapy in 51 patients and reported survival to be significantly longer in patients receiving postoperative radiotherapy compared with those who had surgery alone. In a retrospective review from Finland, the median survival of patients receiving postoperative radiation was 63 months compared with 29 months for patients receiving surgery alone (95). Another small study from the Mayo clinic evaluated 21 patients following curative resection along with adjuvant combined modality therapy with external beam radiation and 5-FU (96). These 21 patients had a 5-year survival rate of 64% versus a historical surgical cohort with a 5-year survival rate of 33% after R0 resection alone. Currently, in patients with node-positive disease, we are recommending radiation therapy. Chemotherapy is only used as a potential radiation-sensitizing agent. upon evaluation of the patient’s general medical fitness as well as rigorous radiologic workup to rule out disseminated disease. Evidence of distant nodal (i.e., N2) disease on preoperative workup precludes a curative resection as no long-term survivors have been reported with gross N2 disease. These patients should be treated only as symptoms develop but should not be offered a reoperation for curative intent. Those re-explored for resection should undergo a standard extended cholecystectomy, including an extensive nodal dissection to include the superior pancreaticoduodenal nodes and a skeletonization of the vessels in the porta hepatis. If the nodal dissection is compromised by the presence of the common bile duct, then this should be resected. In addition, a segment IVb and V resection of the liver or extended resection of the liver should be included, as dictated by the location of the tumor as well as surrounding inflammation and scar tissue.

key points
Gallbladder cancer will be found in 1 per 100 cholecystectomy specimens (incidence 1.2 cases per 100,000 population per year).
● ●

palliative management
Palliative therapy should be considered in the context that the median survival for patients presenting with unresectable gallbladder cancer is 2 to 4 months (60,97). The goal of palliation should be relief of pain, jaundice, and bowel obstruction, as well as prolongation of life. These should be done as simply as possible given the aggressive nature of this disease. Biliary bypass for obstruction can be difficult because of advanced disease in the porta hepatis. A segment III bypass is usually necessary if surgical bypass is chosen to relieve jaundice (98,99). However, such bypasses have a 12% 30-day mortality rate (99) In the event of a preoperative diagnosis of advanced, unresectable gallbladder cancer in the jaundiced patient, therefore, a noninvasive radiologic approach to biliary drainage is justified. Systemic chemotherapy (100) and radiation therapy (101) have little effect on these tumors. Patients with unresectable disease and good functional status who desire therapy should be directed to investigational studies to determine whether any novel therapies may be of benefit.





75% to 98% association with cholelithiasis. A long obstruction of the mid-common bile duct is gallbladder cancer until proven otherwise. Radiologic investigation of gallbladder cancer: Ultrasound MRCP CT ERCP/PTC if jaundiced Surgical management: Stage I (T1N0M0): Simple cholecystectomy alone Stage II (T2N0M0): Radical cholecystectomy Stage III (T3N0M0) ± hepatic invasion <2 cm: Radical cholecystectomy Stage IV (T4N0M0) ± liver invasion >2 cm – No dissemination: extended hepatectomy – Widespread dissemination: no surgical option

references
1. Gourgiotis S, Kocher HM, Solaini L, et al. Gallbladder cancer. Am J Surg 2008;196(2):252–64. 2. Kato S, Nakagawa T, Kobayashi H, Arai E, Isetani K. Septum formation of the common hepatic duct associated with an anomalous junction of the pancreaticobiliary ductal system and gallbladder cancer: report of a case. Surgery Today 1994;24(6):534–7. 3. Blalock AA. A statistical study of 888 cases of biliary tract disease. Johns Hopkins Hosp Bull 1924;35:391–409. 4. Perpetuo MD, Valdivieso M, Heilbrun LK, et al. Natural history study of gallbladder cancer: a review of 36 years experience at M. D. Anderson Hospital and Tumor Institute. Cancer 1978;42(1):330–5. 5. Piehler JM, Crichlow RW. Primary carcinoma of the gallbladder. Surg Gynecol Obstetr 1978;147(6):929–42. 6. Matsumoto Y, Fujii H, Aoyama H, et al. Surgical treatment of primary carcinoma of the gallbladder based on the histologic analysis of 48 surgical specimens. Am J Surg 1992;163(2):239–45.

summary
Gallbladder cancer is an aggressive disease with a dismal prognosis. It should not, however, be approached with a fatalistic attitude. Appropriate workup and extended resection can result in a cure. Gallbladder cancer will be encountered approximately once every 100 times that a gallbladder is removed for presumed benign gallstone disease. For those patients discovered to have a T1 cancer during pathologic analysis, no further therapy is indicated as long as all the margins, including the cystic duct margin, are negative (56,58). However, T2, T3, or T4 tumors deserve consideration for reexploration (54,61,102–106). Selection for re-resection relies

205

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
7. Bartlett DL, Fong Y, Fortner JG, Brennan MF, Blumgart LH. Long-term results after resection for gallbladder cancer. Implications for staging and management. Ann Surg 1996;224(5):639–46. 8. Steinert R, Nestler G, Sagynaliev E, et al. Laparoscopic cholecystectomy and gallbladder cancer. J Surg Oncol 2006;93(8):682–9. 9. Misra MC, Guleria S. Management of cancer gallbladder found as a surprise on a resected gallbladder specimen. J Surg Oncol 2006;93(8):690–8. 10. Sun CD, Zhang BY, Wu LQ, Lee WJ. Laparoscopic cholecystectomy for treatment of unexpected early-stage gallbladder cancer. J Surg Oncol 2005;91(4):253–7. 11. Nakeeb A, Comuzzie AG, Martin L, et al. Gallstones: genetics versus environment. Ann Surg 2002;235(6):842–9. 12. Akatsu T, Ueda M, Shimazu M, et al. Long-term survival of patients with gallbladder cancer detected during or after laparoscopic cholecystectomy. World J Surg 2005;29(9):1106–9, discussion. 13. Pawlik TM, Scoggins CR, Thomas MB, Vauthey JN. Advances in the surgical management of liver malignancies. Cancer J 2004;10(2):74–87. 14. Carriaga MT, Henson DE. Liver, gallbladder, extrahepatic bile ducts, and pancreas. Cancer 1995;75(1 Suppl):171–90. 15. Wanebo HJ, Vezeridis MP. Treatment of gallbladder cancer. Cancer Treat Res 1994;69:97–109. 16. Cubertafond P, Gainant A, Cucchiaro G. Surgical treatment of 724 carcinomas of the gallbladder. Results of the French Surgical Association Survey. Ann Surg 1994;219(3):275–80. 17. Carty NJ, Johnson CD. Carcinoma of the gallbladder: a survey of cases in Wessex 1982–1989. J Royal Coll Surgeons Edinburgh 1991;36(4):238–41. 18. Wilkinson DS. Carcinoma of the gall-bladder: an experience and review of the literature. Austr NZ J Surg 1995;65(10):724–7. 19. Wang SJ, Fuller CD, Kim JS, et al. Prediction model for estimating the survival benefit of adjuvant radiotherapy for gallbladder cancer. J Clin Oncol 2008;26(13):2112–7. 20. Ogura Y, Mizumoto R, Isaji S, et al. Radical operations for carcinoma of the gallbladder: present status in Japan. World J Surg 1991;15(3):337–43. 21. Shirai Y, Yoshida K, Tsukada K, Muto T, Watanabe H. Radical surgery for gallbladder carcinoma. Long-term results. Ann Surg 1992;216(5):565–8. 22. Tsukada K, Hatakeyama K, Kurosaki I, et al. Outcome of radical surgery for carcinoma of the gallbladder according to the TNM stage. Surgery 1996;120(5):816–21. 23. Redaelli CA, Buchler MW, Schilling MK, et al. High coincidence of Mirizzi syndrome and gallbladder carcinoma. Surgery 1997;121(1):58–63. 24. Albores-Saavedra J, Nadji M, Henson DE. Intestinal-type adenocarcinoma of the gallbladder. A clinicopathologic study of seven cases. Am J Surg Pathol 1986;10(1):19–25. 25. Sumiyoshi K, Nagai E, Chijiiwa K, Nakayama F. Pathology of carcinoma of the gallbladder. World J Surg 1991;15(3):315–21. 26. American College of Surgeons. American Joint Committee on Cancer, Staging System of Bile Duct Cancers, 6th ed. 2003. 27. Henson DE, Albores-Saavedra J, Corle D. Carcinoma of the gallbladder. Histologic types, stage of disease, grade, and survival rates. Cancer 1992;70(6):1493–7. 28. Yamaguchi K, Enjoji M. Carcinoma of the gallbladder. A clinicopathology of 103 patients and a newly proposed staging. Cancer 1988;62(7):1425–32. 29. Peerlinck ID, Irvin TT, Sarsfield PT, Harington JM. GIST (gastro-intestinal stromal tumour) of the gallbladder: a case report. Acta Chir Belg 2004;104(1):107–9. 30. Park JK, Choi SH, Lee S, et al. Malignant gastrointestinal stromal tumor of the gallbladder. J Korean Med Sci 2004 Oct;19(5):763–7. 31. Chan CP, Chang HC, Chen YL, et al. A 10-year experience of unsuspected gallbladder cancer after laparoscopic cholecystectomy. Int Surg 2003;88(3):175–9. 32. Yamamoto H, Hayakawa N, Kitagawa Y, et al. Unsuspected gallbladder carcinoma after laparoscopic cholecystectomy. J Hepatobiliary Pancreat Surg 2005;12(5):391–8. 33. Wakai T, Shirai Y, Hatakeyama K. Radical second resection provides survival benefit for patients with T2 gallbladder carcinoma first discovered after laparoscopic cholecystectomy. World J Surg 2002 Jul;26(7):867–71. 34. Nevin JE, Moran TJ, Kay S, King R. Carcinoma of the gallbladder: staging, treatment, and prognosis. Cancer 1976;37(1):141–8. 35. Donohue JH, Nagorney DM, Grant CS, et al. Carcinoma of the gallbladder. Does radical resection improve outcome? Arch Surg 1990;125(2):237–41. 36. Fong Y, Wagman L, Gonen M, et al. Evidence-based gallbladder cancer staging: changing cancer staging by analysis of data from the National Cancer Database. Ann Surg 2006;243(6):767–71. 37. Strom BL, Maislin G, West SL, et al. Serum CEA and CA 19-9: potential future diagnostic or screening tests for gallbladder cancer? Int J Cancer 1990;45(5):821–4. 38. Ritts RE, Jr., Nagorney DM, Jacobsen DJ, Talbot RW, Zurawski VR, Jr. Comparison of preoperative serum CA19-9 levels with results of diagnostic imaging modalities in patients undergoing laparotomy for suspected pancreatic or gallbladder disease. Pancreas 1994;9(6):707–16. 39. Fong Y, Heffernan N, Blumgart LH. Gallbladder carcinoma discovered during laparoscopic cholecystectomy: aggressive reresection is beneficial. Cancer 1998;83(3):423–7. 40. Akyurek N, Irkorucu O, Salman B, et al. Unexpected gallbladder cancer during laparoscopic cholecystectomy. J Hepatobiliary Pancreat Surg 2004;11(5):357–61. 41. Aoki T, Tsuchida A, Kasuya K, et al. Is frozen section effective for diagnosis of unsuspected gallbladder cancer during laparoscopic cholecystectomy? Surg Endosc 2002;16(1):197–200. 42. Antonakis P, Alexakis N, Mylonaki D, et al. Incidental finding of gallbladder carcinoma detected during or after laparoscopic cholecystectomy. Eur J Surg Oncol 2003;29(4):358–60. 43. Darmas B, Mahmud S, Abbas A, Baker AL. Is there any justification for the routine histological examination of straightforward cholecystectomy specimens? Ann R Coll Surg Engl 2007;89(3):238–41. 44. Chijiiwa K, Sumiyoshi K, Nakayama F. Impact of recent advances in hepatobiliary imaging techniques on the preoperative diagnosis of carcinoma of the gallbladder. World J Surg 1991;15(3):322–7. 45. Wilbur AC, Gyi B, Renigers SA. High-field MRI of primary gallbladder carcinoma. Gastrointest Radiol 1988;13(2):142–4. 46. Collier NA, Carr D, Hemingway A, Blumgart LH. Preoperative diagnosis and its effect on the treatment of carcinoma of the gallbladder. Surg Gynecol Obstetr 1984;159(5):465ina70. 47. Schwartz LH, Coakley FV, Sun Y, Blumgart LH, Fong Y, Panicek DM. Neoplastic pancreaticobiliary duct obstruction: evaluation with breath-hold MR cholangiopancreatography. Am J Roentgenol 1998;170(6):1491–5. 48. Schwartz LH, Lefkowitz RA, Panicek DM, et al. Breath-hold magnetic resonance cholangiopancreatography in the evaluation of malignant pancreaticobiliary obstruction. J Comput Assist Tomogr 2003;27(3):307–14. 49. Corvera CU, Blumgart LH, Akhurst T, et al. 18F-fluorodeoxyglucose positron emission tomography influences management decisions in patients with biliary cancer. J Am Coll Surg 2008;206(1):57–65. 50. Nimura Y, Hayakawa N, Kamiya J, et al. Hepatopancreatoduodenectomy for advanced carcinoma of the biliary tract. Hepato-Gastroenterology 1991;38(2):170–5. 51. Sicklick JK, Choti MA. Controversies in the surgical management of cholangiocarcinoma and gallbladder cancer. Semin Oncol 2005;32(6 Suppl 9):S112–17. 52. Gagner M, Rossi RL. Radical operations for carcinoma of the gallbladder: present status in North America. World J Surg 1991;15(3):344–7. 53. Shoup M, Fong Y. Surgical indications and extent of resection in gallbladder cancer. Surgical Oncology Clinics of North America 2002;11(4):985–94. 54. Fong Y, Jarnagin W, Blumgart LH. Gallbladder cancer: comparison of patients presenting initially for definitive operation with those presenting after prior noncurative intervention. Ann Surg 2000;232(4):557–69. 55. Tsukada K, Kurosaki I, Uchida K, et al. Lymph node spread from carcinoma of the gallbladder. Cancer 1997;80(4):661–7. 56. Ouchi K, Owada Y, Matsuno S, Sato T. Prognostic factors in the surgical treatment of gallbladder carcinoma. Surgery 1987;101(6):731–7. 57. Gall FP, Kockerling F, Scheele J, Schneider C, Hohenberger W. Radical operations for carcinoma of the gallbladder: present status in Germany. World J Surg 1991;15(3):328–36. 58. Shirai Y, Yoshida K, Tsukada K, Muto T, Watanabe H. Early carcinoma of the gallbladder. Eur J Surg 1992;158(10):545–8. 59. Yamaguchi K, Tsuneyoshi M. Subclinical gallbladder carcinoma. Am J Surg 1992;163(4):382–6. 60. Oertli D, Herzog U, Tondelli P. Primary carcinoma of the gallbladder: operative experience during a 16 year period. Eur J Surg 1993;159(8):415–20. 61. de A, X, Roa I, Burgos L, et al. Gallbladder cancer in Chile. A report on 54 potentially resectable tumors. Cancer 1992 Jan 1;69(1):60–5.

206

TREATMENT OF LAPAROSCOPICALLY DISCOVERED GALLBLADDER CANCER
62. Shirai Y, Yoshida K, Tsukada K, Muto T. Inapparent carcinoma of the gallbladder. An appraisal of a radical second operation after simple cholecystectomy. Ann Surg 1992;215(4):326–31. 63. Shih SP, Schulick RD, Cameron JL, et al. Gallbladder cancer: the role of laparoscopy and radical resection. Ann Surg 2007;245(6):893–901. 64. Kai M, Chijiiwa K, Ohuchida J, et al. A curative resection improves the postoperative survival rate even in patients with advanced gallbladder carcinoma. J Gastrointest Surg 2007;11(8):1025–32. 65. Jensen EH, Abraham A, Habermann EB, et al. A critical analysis of the surgical management of early-stage gallbladder cancer in the United States. J Gastrointest Surg 2009;13(4):722–7. 66. D’Angelica M, Dalal KM, DeMatteo RP, et al. Analysis of the extent of resection for adenocarcinoma of the gallbladder. Ann Surg Oncol 2009;16(4):806–16. 67. Onoyama H, Yamamoto M, Tseng A, Ajiki T, Saitoh Y. Extended cholecystectomy for carcinoma of the gallbladder. World J Surg 1995;19(5):758–63. 68. Duffy A, Capanu M, bou-Alfa GK, et al. Gallbladder cancer (GBC): 10-year experience at Memorial Sloan-Kettering Cancer Centre (MSKCC). J Surg Oncol 2008;98(7):485–9. 69. Lam CM, Yuen AW, Wai AC, et al. Gallbladder cancer presenting with acute cholecystitis: a population-based study. Surg Endosc 2005;19(5):697–701. 70. Sarli L, Contini S, Sansebastiano G, et al. Does laparoscopic cholecystectomy worsen the prognosis of unsuspected gallbladder cancer? Arch Surg 2000;135(11):1340–4. 71. de A, X, Roa IS, Mora JP, et al. Laparoscopic cholecystectomy: its effect on the prognosis of patients with gallbladder cancer. World J Surg 2004;28(6):544–7. 72. Kang CM, Choi GH, Park SH, et al. Laparoscopic cholecystectomy only could be an appropriate treatment for selected clinical R0 gallbladder carcinoma. Surg Endosc 2007;21(9):1582–7. 73. Kang CM, Lee WJ, Choi GH, et al. Does “clinical” R0 have validity in the choice of simple cholecystectomy for gallbladder carcinoma? J Gastrointest Surg 2007;11(10):1309–16. 74. Boerma EJ. Towards an oncological resection of gall bladder cancer. Eur J Surg Oncol 1994;20(5):537–44. 75. Drouard F, Delamarre J, Capron JP. Cutaneous seeding of gallbladder cancer after laparoscopic cholecystectomy. N Engl J Med 1991;325(18):1316. 76. Clair DG, Lautz DB, Brooks DC. Rapid development of umbilical metastases after laparoscopic cholecystectomy for unsuspected gallbladder carcinoma. Surgery 1993;113(3):355–8. 77. Landen SM. Laparoscopic surgery and tumor seeding. Surgery 1993;114(1):131–2. 78. Fligelstone L, Rhodes M, Flook D, Puntis M, Crosby D. Tumour inoculation during laparoscopy. Lancet 1993;342(8867):368–9. 79. Fong Y, Brennan MF, Turnbull A, Coit DG, Blumgart LH. Gallbladder cancer discovered during laparoscopic surgery. Potential for iatrogenic tumor dissemination. Arch Surg 1993;128(9):1054–6. 80. Nduka CC, Monson JRT, Menzies-Gow N, Darzi A. Abdominal wall metastases following laparoscopy. Br J Surg 1994;81:648–52. 81. Nally C, Preshaw RM. Tumour implantation at umbilicus after laparoscopic cholecystectomy for unsuspected gallbladder carcinoma. Can J Surg 1994;37(3):243–4. 82. Kim HJ, Roy T. Unexpected gallbladder cancer with cutaneous seeding after laparoscopic cholecystectomy. Southern Med J 1994;87(8): 817–20. 83. Winston CB, Chen JW, Fong Y, Schwartz LH, Panicek DM. Recurrent gallbladder carcinoma along laparoscopic cholecystectomy port tracks: CT demonstration. Radiology 1999;212(2):439–44. 84. Paolucci V. Port site recurrences after laparoscopic cholecystectomy. J Hepatobiliary Pancreat Surg 2001;8(6):535–43. 85. Giuliante F, Ardito F, Vellone M, Clemente G, Nuzzo G. Port-sites excision for gallbladder cancer incidentally found after laparoscopic cholecystectomy. Am J Surg 2006;191(1):114–6. 86. Fong Y, Blumgart LH, Fortner JG, Brennan MF. Pancreatic or liver resection for malignancy is safe and effective for the elderly. Ann Surg 1995;222(4):426–34. 87. Takada T, Amano H, Yasuda H, et al. Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A phase III multicenter prospective randomized controlled trial in patients with resected pancreaticobiliary carcinoma. Cancer 2002;95(8):1685–95. 88. Chao TC, Jan YY, Chen MF. Primary carcinoma of the gallbladder associated with anomalous pancreaticobiliary ductal junction. J Clin Gastroenterol 1995;21(4):306–8. 89. Oswalt CE, Cruz AB, Jr. Effectiveness of chemotherapy in addition to surgery in treating carcinoma of the gallbladder. Rev Surg 1977;34(6):436–8. 90. Morrow CE, Sutherland DE, Florack G, Eisenberg MM, Grage TB. Primary gallbladder carcinoma: significance of subserosal lesions and results of aggressive surgical treatment and adjuvant chemotherapy. Surgery 1983;94(4):709–14. 91. Todoroki T. Radiation therapy for primary gallbladder cancer. HepatoGastroenterology 1997;44(17):1229–39. 92. Todoroki T, Iwasaki Y, Orii K, et al. Resection combined with intraoperative radiation therapy (IORT) for stage IV (TNM) gallbladder carcinoma. World J Surg 1991;15(3):357–66. 93. Bosset JF, Mantion G, Gillet M, et al. Primary carcinoma of the gallbladder. Adjuvant postoperative external irradiation. Cancer 1989;64(9):1843–7. 94. Hanna SS, Rider WD. Carcinoma of the gallbladder or extrahepatic bile ducts: the role of radiotherapy. Can Med Assoc J 1978;118(1):59–61. 95. Vaittinen E. Carcinoma of the gall-bladder. A study of 390 cases diagnosed in Finland 1953-1967. Annales Chirurgiae et Gynaecologiae Fenniae - Supplement 1970;168:1–81. 96. Kresl JJ, Schild SE, Henning GT, et al. Adjuvant external beam radiation therapy with concurrent chemotherapy in the management of gallbladder carcinoma. Int J Radiation Oncol Biol Phys 2002;52(1):167–75. 97. Wanebo HJ, Castle WN, Fechner RE. Is carcinoma of the gallbladder a curable lesion? Ann Surg 1982;195(5):624–31. 98. Bismuth H, Corlett MB. Intrahepatic cholangioenteric anastomosis in carcinoma of the hilus of the liver. Surg Gynecol Obstet 1975;140:170–6. 99. Kapoor VK, Pradeep R, Haribhakti SP, Sikora SS, Kaushik SP. Early carcinoma of the gallbladder: an elusive disease. J Surg Oncol 1996;62(4):284–7. 100. Taal BG, Audisio RA, Bleiberg H, et al. Phase II trial of mitomycin C (MMC) in advanced gallbladder and biliary tree carcinoma. An EORTC Gastrointestinal Tract Cancer Cooperative Group Study. Ann Oncol 1993;4(7):607–9. 101. Houry S, Schlienger M, Huguier M, et al. Gallbladder carcinoma: role of radiation therapy. Br J Surg 1989;76(5):448–50. 102. Chijiiwa K, Tanaka M. Carcinoma of the gallbladder: an appraisal of surgical resection. Surgery 1994;115(6):751–6. 103. Nakamura S, Sakaguchi S, Suzuki S, Muro H. Aggressive surgery for carcinoma of the gallbladder. Surgery 1989;106(3):467–73. 104. Fong Y, Malhotra S. Gallbladder cancer: recent advances and current guidelines for surgical therapy. Adv Surg 2001;35:1–20. 105. Whalen GF, Bird I, Tanski W, Russell JC, Clive J. Laparoscopic cholecystectomy does not demonstrably decrease survival of patients with serendipitously treated gallbladder cancer. J Am Coll Surgeons 2001;192(2):189–95. 106. Todoroki T, Takahashi H, Koike N, et al. Outcomes of aggressive treatment of stage IV gallbladder cancer and predictors of survival. HepatoGastroenterology 1999;46(28):2114–21. 107. Paquet KJ. Appraisal of surgical resection of gallbladder carcinoma with special reference to hepatic resection. J Hepatobiliary Pancreat Surg 1998;5(2):200–6. 108. Cucinotta E, Lorenzini C, Melita G, Iapichino G, Curro G. Incidental gall bladder carcinoma: does the surgical approach influence the outcome? Aust NZ J Surg 2005;75(9):795–8. 109. Hamila F, Letaief R, Khnissi M, et al. [Port site recurrence after laparoscopic cholecystectomy]. Tunis Med 2006;84(11):697–700.

207

22

Liver transplantation for HCC: Asian perspectives Shin Hwang, Sung-Gyu Lee, Vanessa de Villa, and Chung Mao Lo
zation, local ablation, and injection therapy. Surgical resection has been regarded as the treatment of choice for HCC, but it has definite limitations in both tumor resectability and patient safety when hepatic functional reserve is markedly reduced. Although it has been shown that liver resection can be achieved safely in selected patients with cirrhosis (11,12), it is widely accepted that it carries a significant risk of postoperative morbidity, even in patients with Child–Pugh grade A cirrhosis. With reasonable patient selection, liver resection is associated with relatively low operative mortality rates, ranging in recent studies from 0% to 4% (11–13). Long-term survival results after surgical resection, however, are affected by tumor recurrence and progressive deterioration of liver function (14,15). Non-surgical treatments often result in incomplete tumor control and also reveal high recurrence rates when used for advanced HCC lesions. The natural history of untreated and non-surgically treated HCC in Asia is worse than that of similar cases in Western countries. A study from Hong Kong (16) showed that the 1-year and 2-year survival rates of patients with untreated HCC were 7.8% and 0.9%, respectively, whereas a comparable study from Spain (17) showed 54% and 40%, respectively (3).

introduction
Hepatocellular carcinoma (HCC) is the fifth most common and the third most deadly cancer worldwide (1). HCC is closely associated with chronic liver disease and more than 80% of cases occur in cirrhotic livers (2). Since functional reserve of the liver is often significantly impaired, application of treatment modalities such as liver resection, transarterial chemoembolization, or local ablation therapy is limited. Not only HCC recurrence, but also progressive deterioration of liver function decrease patient survival. Liver transplantation (LT) is the only treatment that offers a chance of cure for the tumor and the underlying liver cirrhosis at the same time, although there is an additional risk of accelerated tumor recurrence from immunosuppression. The outcome of LT for HCC in Western countries is encouraging, but availability of liver grafts still remains the main limitation for LT. Since the incidence of HCC combined with chronic liver diseases is much higher, and organ donation from deceased donors is much fewer in Asian countries compared to Western countries, it is difficult to use Deceased Donor LT (DDLT) as one of the main treatment modalities for HCC in Asia. In fact, organ donation from deceased donors remains below five per million per year in most Asian countries because of various social and cultural reasons (Fig. 22.1) (3). Because of these constraints, Living Donor LT (LDLT) is now the main method of LT in a number of Asian countries, and is an alternative to DDLT in every indication for LT. Numerous technical innovations have been achieved to secure donor safety as well as ensure patient survival. The outcomes of LT for HCC in major Asian LT centers are acceptably favorable and also comparable to those seen in Western countries (3). Patient selection criteria for HCC have also been proposed by several major Asian LT centers based on their own data (4–7).

history of lt in asia
LT in Asia started early and yet progressed slowly when compared with Western countries (3). The first liver transplant in Asia was performed in 1964 by Nakayama with a graft from a non-heart beating donor in Chiba, Japan (18). This was only 1 year after Starzl’s first attempt in the world for human LT in Denver, USA (19). It was a highly controversial exploit because organ donation from a deceased person was not accepted in Japanese culture at that time. The second transplant in Asia was performed in 1978 in Shanghai, P.R. China (3,20). In 1984, Chen, in Taiwan, performed the first liver transplant with long-term survival in Asia at a time when there was still no brain death law in that country (21). Legislation regarding brain death was passed in Singapore (22) and Taiwan in 1987, and subsequently in Japan and Korea (23). In Asia, the serious scarcity of deceased donor organ donation and strong demand for LT provided a powerful driving force for the development of LDLT as a practical alternative in replacing DDLT. The first LDLT in Asia was performed by Nagasue of Shimane University, Japan, in 1989 (24), only 1 year after the first attempt of LDLT by Raia in Brazil in 1988 (25). Subsequently Hong Kong, Korea, and Taiwan rapidly initiated pediatric LDLT programs transplanting a left lateral section or a left lobe graft from a parent donor to a child. However, the ultimate need for LDLT was in adult patients. Transplant centers in Asia have repeatedly advanced the frontier of adult-to-adult LDLT. In 1993, Makuuchi in Japan performed the first successful adult LDLT using a left lobe

hepatocellular carcinoma in asia
Hepatitis B and C are the most common etiologic agents for HCC (8). Since Hepatitis B Virus (HBV) is more prevalent in most Asian countries, 80% of HCC occurs in areas where HBV infection is endemic, and the geographic distribution of HCC largely follows that of HBV infection (2). Chronic HBV infection is endemic in Asia, where an estimated 300 million individuals are infected by HBV, representing about 75% of the world’s chronic HBV carriers (9). China has the highest ageadjusted incidence of HCC (37.9 per 100,000 in males and 14.2 per 100,000 in females), and accounts for 50% of all cases in the world (1,2). Unlike other Asian countries, prevalence of HBV infection is exceptionally low in Japan, where hepatitis C is more prevalent and similar to that seen in Europe and the United States (8,10). Conventional standard treatments for HCC include surgical resection and non-surgical treatments such as chemoemboli-

208

LIVER TRANSPLANTATION FOR HCC: ASIAN PERSPECTIVES
Japan 0.07 1.9 Korea 4.2 Hong Kong 4.7 Singapore 6.6 Taiwan Australia United Kingdom Canada Germany Portugal Italy France Belgium Austria USA Spain 0 5

10 10.7 12.8 14.4 19 20.9 22.2 22.8 24.6 25.2 35.1 35 40

10 15 20 25 30 Number of deceased donors per million of population

Figure 22.1 Comparison of organ donation rates from deceased donors in different countries in the East and West in 2005 (3).

graft (26). The left lobe graft, which usually comprises less than 40% of the whole liver volume, is often too small for an adult recipient. For successful LDLT for adult patients, graft size is the most important factor in determining the outcome of LDLT and is still a controversial concern for selection of the ideal donor. The Kyoto group first reported the use of a right lobe graft for a pediatric recipient as a result of a variance in the donor’s arterial anatomy in 1994 (27). The first case report (28), and subsequently the first series (29) of successful adult LDLT, using a right lobe graft was reported from Hong Kong in 1997. Further technical advances in adult LDLT, including the addition of the caudate lobe to a left lobe graft (30) and the use of right lateral section grafts (31), were reported from Japan. In 1999, authors proposed conservation of the middle hepatic vein trunk to the donor, and the reconstruction of hepatic venous drainage of the right anterior section of the liver graft to avoid congestion injury of both the right lobe liver graft and the donor’s remnant left lobe (32). Middle hepatic vein reconstruction has apparently increased the success rate of the right lobe LDLT in adult patients and widened the safety margin and the pool of living donors. As an attempt to provide adequate graft volume for an adult recipient and minimize the risk of an individual donor, authors also performed implantation of dual grafts from two donors for one recipient in 2000 (33,34). In Asia, where the supply of deceased donor organs remains seriously limited and the demand for LT is persistently increasing, the applicability of LDLT will continue. For successful LDLT, the risk to the donor should be balanced by the greater benefit to the recipient. Every effort must be taken to minimize donor morbidities, making this procedure beneficial to the donor and the recipient (35–38).

selection criteria of hcc and outcome
LT offers the opportunity to eliminate both tumor and underlying liver cirrhosis at the same time. From the viewpoints of quality of life and tumor recurrence, any other treatment

modality cannot achieve such a favorable result comparable to that of LT. However, it should be emphasized that HCC recurrence is the most common cause of late patient death after LT (4,39–41). The clinical sequence of post-transplant HCC recurrence is usually much worse than in non-transplant patients since the response to treatment for HCC recurrence after LT is disappointing. To date, the most effective measure to prevent post-transplant HCC recurrence is strict selection of transplant candidates after exclusion of patients with known risk factors for HCC recurrence. Eligibility guidelines for transplantation, such as the Milan and University of California at San Francisco (UCSF) criteria, have been adopted to reduce the post-transplant HCC recurrence and the wasting of donor organs (42,43). The Milan criteria were originally established on the basis of pre-transplant imaging findings but were re-evaluated on the basis of explant liver pathology, whereas the UCSF criteria were based on explant pathology but validated by pre-transplant imaging findings. Each of these two sets of criteria is derived from the experience with DDLT (Table 22.1). Application of these criteria to LDLT has resulted in patient survival outcome, very similar to that following DDLT, as shown in high-volume multi-center cohorts from Japan and Korea (40,41). Moreover, the prognostic powers of the Milan and UCSF criteria were reported to be the same in both DDLT and LDLT (41). When selecting transplant candidates, there is a real risk of discrepancies between the pre-transplant radiological and explant pathologic staging (4,40,41). Candidates for DDLT should be selected after consideration of the extent of HCC at the time of listing, and any further progression of HCC during the waiting period. LDLT, because of its shorter waiting time, is less affected by tumor progression, permitting more flexible selection of transplant candidates than DDLT. A substantial proportion of adult LDLT patients not fulfilling the Milan or UCSF criteria has been found to survive longer than expected after LT (4,40,41,44,45). Therefore, it seems reasonable to attempt further reduction of unnecessary dropouts arising from the strict application of narrow selection criteria.

209

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 22.1 Criteria for Indication of LT in the Setting of Hepatocellular Carcinoma.
Type of donor LDLT: 100% Number of patients 221 Patient survival rate within criteria 1-year: 94.3%, 3-years: 87.5%, 5-years: 81.6% 5-years: 86.7% Patient survival rate exceeding criteria 1-year: 71.9%, 3-years: 37.2%, 5-years: 20.7% 5-years: 34.4%

Center Asan, Korea (4)

Tumor criteria Diameter ≤5 cm; Number of lesions ≤6; No gross vascular invasion Diameter ≤5 cm; Number of lesions ≤10; PIVKA-II ≤400 mAU/mL Diameter ≤5 cm; Number of lesion ≤5 Total diameter ≤8 cm; or total diameter >8 cm and histopathologic grades I and II, and AFP ≤400 ng/mL Diameter ≤5 cm if single lesion; or Diameter ≤3 cm if multiple lesions and number of lesions ≤3 Diameter ≤6.5 cm if single lesion; or Diameter ≤4.5 cm if 2 lesions with total diameter ≤8 cm

Viral hepatitis HBV: 93%, HCV: 7% HBV: 34%, HCV: 53% HCV: 62% HBV: 100%

Kyoto, Japan (5)

LDLT: 100%

125

Tokyo, Japan (6) Hangzhou, P.R. China (7)

LDLT: 100% Not defined

78 195

Recurrence-free 3-years: 94% 1-year: 92.8%, 3-years: 70.7%, 5-years: 70.7%

Recurrence-free 3-years: 50% 1-years: 49.9%, 3-years: 27.0%, 5-years: 18.9%

Milan, Italy (42)

DDLT: 100%

HBV: 23%, HCV: 67%

48

4-years: 85%

4-years: 50%

UCSF, USA (47)

DDLT: 93%, LDLT: 7%

HBV: 23%, HCV: 65%

168

Recurrence-free 1-year: 98.6%, Recurrence-free 5-years: 96.7%

Recurrence-free 1-year: 80.4%, Recurrence-free 5-years: 59.5%

Abbreviations: AFP, alpha-fetoprotein; HBV, hepatitis B virus; HCV, hepatitis C virus; LDLT, living donor liver transplantation; PIVKA-II, prothrombin induced by vitamin K absence II; UCSF, University of California at San Francisco.

The golden standard to date for selection of HCC patients for DDLT and LDLT are the Milan criteria, while the UCSF criteria are regarded as acceptable alternative guidelines. The UCSF criteria, which expand the maximal tumor size to 6.5 cm for a single HCC lesion, produce survival rates comparable to those of the Milan criteria. However, the study populations in the original studies evaluating these two criteria do not seem to be large enough, introducing the possibility of bias from small numbers of patients with recurrent tumors (42,46). The prognostic power of these criteria appears sufficient, but their discriminatory power for those HCC patients who do not meet these criteria is not sufficiently high (4). Many highvolume LT center series or multi-center studies revealed that long-term patient survival rates look uniformly favorable when extent of HCC met the indication criteria, but the recurrence rates showed great differences among indication criteria (Fig. 22.2). Currently proposed criteria for indication of LT are summarized in Table 22.1 (4–7,42,47). Considering that these proposed expanded criteria also showed favorable outcomes in recent studies (4–7), it is probable that the Milan criteria are becoming outdated. However, the question is which criteria should be the new golden standard instead of the Milan criteria (48). It is reported that hepatitis C is a significant predictor of tumor recurrence and impaired survival after LT in patients with HCC (49), implicating that there may be a benefit in the expansion of the Milan criteria for HCC in the non-hepatitis C population. The clinical sequence of HCC in patients with HBV infection has been reported to be favorable compared to

that in patients with hepatitis C virus infection because HBV infection is more effectively controllable than hepatitis C infection after LT (50,51). Uniquely, some major LT centers in Korea and Japan challenged the Milan criteria, accepting a much higher number of nodules (five or more) (4–6,52). On the other hand, a number of major LT centers in the United States and Europe have mainly focused on enhanced criteria regarding the tumor diameter (more than 5 cm) (47,53–55). LT is also indicated for small HCC in Child–Turcotte–Pugh (CTP) class B or C cirrhosis, or CTP class A with portal hypertension (42,56–59). However, the optimal treatment strategy for patients with small single HCC, cirrhosis with preserved liver function, and absence of portal hypertension is not yet established. No prospective randomized study has been reported which compares liver resection and LT for small HCC in cirrhotic patients who could be eligible for both treatment modalities. Authors have reported the results of liver resection in 100 cirrhotic patients with single, less than 3 cm-sized HCC who would have been eligible for primary LT, comparing with a group of 17 patients who underwent LT for HCC with similar tumor stage during diagnosis and preserved liver function (60). These results showed that 37 of 39 HCC recurrences initially occurred in the liver after resection, but only one recurrence occurred in lung after LT. Overall patient survival was not proven to be statistically different, but diseasefree survival was significantly different between the two groups (Fig. 22.3). The influence of these encouraging results of LT for HCC may be reflected on the proportion of liver recipients

210

LIVER TRANSPLANTATION FOR HCC: ASIAN PERSPECTIVES
1.0 1.0 Beyond Asan Proportion of recurrence Proportion of recurrence 0.8 0.8

0.6

Beyond UCSF

0.6 Beyond Milan, within Asan Within Milan

0.4 Beyond Milan, within UCSF 0.2 Within Milan 0.0 0 12 24 36 48 Posttransplant months 60

0.4

0.2

0.0

0

12

24 36 48 Posttransplant months

60

(A)

(B)

Figure 22.2 Comparison of the hepatocellular carcinoma recurrence curves between the Milan and UCSF criteria (A) and between the Milan and Asan criteria (B) (4). 1.0 0.9 0.8 Proportion of survival Proportion of survival 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 (A) 20 40 60 80 100 120 140 (B) Postoperative months P = 0.27 Hepatectomy LT 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 20 40 60 80 100 120 140 Postoperative months P = 0.047 Hepatectomy LT

Figure 22.3 Comparison of the overall patient survival (A) and recurrence-free survival (B) between the patients who undergone hepatic resection and liver transplantation for hepatocellular carcinoma (HCC) <3 cm-sized, single nodule with Child–Turcotte–Pugh class A cirrhosis (60).

undergoing LT per year in Asia (Fig. 22.4). Hepatic resection is still a good treatment modality for selected patients because of lower cost and no requirement for donor organs. However, if a donor is available, primary LT can be considered as a preferred treatment modality for single small HCC of CTP class A cirrhosis in the presence of portal hypertension, because LT may provide excellent disease-free survival and maintenance of normal quality of life (60).

pretransplant treatment for hcc
Pretransplant neoadjuvant therapy has been often attempted to the HCC patients waiting for LT by the way of percutaneous ablation or transarterial chemoembolization (61–63). In patients enrolled for DDLT, pretransplant neoadjuvant therapy is important since the waiting period is usually too long to leave the patient without any treatment, and downstaging of advanced tumors may allow expansion of the current criteria without adversely affecting survival. However, the effect of downstaging is not strongly supported from the data of a considerable number of clinical studies (64–67). In reality, the

response to neoadjuvant therapy may be potentially influenced by the selection effect for tumors with better biological behavior. Such a selection bias also influences post-transplant outcome (68–70). Salvage LT has been performed for recurrent HCC or deterioration of liver function after primary liver resection. From the viewpoint of pre-transplant treatment, prior liver resection has two roles: primary treatment and bridging to LT. Considering the incidence of deceased donors and high proportion of LDLT, many of prior liver resection may be intended for primary curative treatment rather than bridge treatment. There are two important points to be taken into account before performing salvage LT (71). The first is the technical feasibility of LT, especially for LDLT. The surgical condition of patients with recurrent HCC after prior liver resection is not very different to that of patients who underwent non-surgical treatment, except that adhesion and anatomical distortion exist within the abdomen, which could be overcome by technical skill (71). Some authors claim that every combination of prior hepatectomy and living donor graft is feasible for patients

211

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
1,200 1,000 Number of operations HCC 800 600 400 200 0 1994 1995 1996 1997 1998 1999 2000 Year 2001 2002 2003 2004 2005 Non-HCC

Figure 22.4 Annual proportions of liver transplants for hepatocellular carcinoma (HCC) in Asia (excluding mainland China) (3).

undergoing salvage LDLT. The second is the main indication for salvage LT, in which there is still no consensus on eligibility criteria. Authors also suggest that the survival outcome of salvage LDLT, based on explant tumor staging, was equivalent to that of primary LDLT, and it may be reasonable that primary and salvage LDLT share the same indication criteria for HCC. For salvage LDLT, determination of the timing of LT is of practical importance. There are proposals to perform LDLT as early as the situation permits in order to avoid tumor progression, but it is unclear whether patients who had early HCC recurrence after liver resection are adequate candidates for salvage LT. If early recurrence is associated with a well-advanced tumor or unfavorable pathology, the patient may not benefit from salvage LT because there is a very high probability of post-transplant HCC recurrence. In contrast, if early recurrence is related to incomplete local control of small HCC lesions, the patient may be a candidate for salvage LT. Patients with a longer interval between prior liver resection and salvage LDLT show more favorable survival than those with a shorter interval. A thorough pre-transplant HCC workup should be performed because extra-hepatic recurrence was not uncommon in patients with recurrent HCC after resection (37,72,73). For salvage DDLT, there are two important reports revealing contradictory results. Belghiti and colleagues (74) compared 18 patients who underwent secondary LT following liver resection with 70 undergoing primary LT and assessed postoperative outcomes and long-term survival. They concluded that in selected patients, liver resection prior to LT does not increase morbidity or impair long-term survival following primary LT, and, accordingly, liver resection prior to LT can be integrated into the treatment strategy for HCC. On the other hand, Adam and colleagues (75) compared the results of secondary LT for tumor recurrence following resection of initially transplantable HCC in 17 patients with cirrhosis with those of primary LT in 195 patients. They reported that LT after liver resection was associated with a higher operative mortality, an increased risk of recurrence, and a poorer outcome compared with primary LT.

treatment for recurrent hcc after lt
The clinical sequence of post-transplant HCC recurrence is usually much worse than in non-transplant patients, since the response to treatment for HCC recurrence after LT is disappointing (Fig. 22.5). Strict recurrence surveillance for early timely detection for recurrence is recommended. Thus, annual risk-based adjustment of follow-up intervals should be established. Early detection and aggressive treatment for posttransplant HCC recurrence can result in prolongation of patient survival in a considerable proportion of patients (76,77). Such a survival gain may rationalize vigorous post-transplant surveillance for HCC recurrence. It is reported that organ transplant recipients given mammalian target of rapamycin (mTOR) inhibitor have reduced incidence of recurrent or de novo post-transplant malignancies (78). Kneteman and colleagues (79) performed DDLT on 40 patients using sirolimus-based immunosuppression that minimized exposure to calcineurin inhibitors and steroid. Tumor-free 4-year survival was 81.1% and the tumor recurrence rate was less than 5.3% in 19 recipients with tumors that satisfied the Milan criteria and were 76.8% and 19% for patients with extended criteria tumors. From these results, Wall (80) noted that the results obtained with HCC satisfying the Milan criteria were good, although not essentially different from those obtained by others using standard immunosuppressive therapy (81–83), whereas the results with extended criteria tumors were excellent, strongly suggesting mTOR inhibitor therapy inhibited tumor growth. Elsharkawi and colleagues (84) reported a DDLT case in whom three HCC pulmonary metastases underwent complete regression after conversion from cyclosporine and azathioprine to sirolimus and mycophenolate mofetil and who remained tumor-free 18 months later. The recent demonstration of survival benefits for HCC patients treated with the oral agent sorafenib is encouraging progress in the development of molecularly targeted anticancer agents in HCC (85,86). Several new targeted agents have been developed and are under clinical trial in patients undergoing non-surgical treatment, resection, or LT at present. In

212

LIVER TRANSPLANTATION FOR HCC: ASIAN PERSPECTIVES
1.0 5. Ito T, Takada Y, Ueda M, et al. Expansion of selection criteria for patients with hepatocellular carcinoma in living donor liver transplantation. Liver Transpl 2007; 13: 1637–44. 6. Sugawara Y, Tamura S, Makuuchi M. Living donor liver transplantation for hepatocellular carcinoma: Tokyo University Series. Dig Dis 2007; 25: 310–12. 7. Zheng SS, Xu X, Wu J, et al. Liver transplantation for hepatocellular carcinoma: Hangzhou experiences. Transplantation 2008; 85: 1726–32. 8. Raza SA, Clifford GM, Franceschi S. Worldwide variation in the relative importance of hepatitis B and hepatitis C viruses in hepatocellular carcinoma: a systematic review. Br J Cancer 2007; 96: 1127–34. 9. Lesmana LA, Leung NW, Mahachai V, et al. Hepatitis B: Overview of the burden of disease in the Asia-Pacific region. Liver Int 2006; 26(Suppl 2): 3–10. 10. El-Serag HB. Hepatocellular carcinoma: an epidemiologic view. J Clin Gastroenterol 2002; 35(Suppl 2): S72–8. 11. Fan ST, Lai ECS, Lo CM, et al. Hospital mortality of major hepatectomy for hepatocellular carcinoma associated with cirrhosis. Arch Surg 1995; 130: 198–203. 12. Lee SG, Hwang S. How I do it: assessment of hepatic functional reserve for indication of hepatic resection. J Hepatobiliary Pancreat Surg 2005; 12: 38–43. 13. Poon RTP, Fan ST, Lo CM, et al: Long term survival and pattern of recurrence after resection of small hepatocellular carcinoma in patients with preserved liver function. Implications for a strategy of salvage transplantation. Ann Surg 2002; 235: 373–82. 14. Bigourdan JM, Jaeck D, Meyer N, et al: Small hepatocellular carcinoma in Child A cirrhotic patients: hepatic resection versus transplantation. Liver Transpl 2003; 9: 513–20. 15. Margarit C, Escartin A, Castells L, et al. Resection for hepatocellular carcinoma is a good option in Child-Turcotte-Pugh class A patients with cirrhosis who are eligible for liver transplantation. Liver Transpl 2005; 11: 1242–51. 16. Yeung YP, Lo CM, Liu CL, et al. Natural history of untreated nonsurgical hepatocellular carcinoma. Am J Gastroenterol 2005; 100: 1995–2004. 17. Llovet JM, Bustamante J, Castells A, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: rationale for the design and evaluation of therapeutic trials. Hepatology 1999; 29: 62–7. 18. Shimada M, Fujii M, Morine Y, et al. Living-donor liver transplantation: present status and future perspective. J Med Invest 2005; 52: 22–32. 19. Starzl TE, Marchioro TL, Vonkaulla KN, et al. Homotransplantation of the liver in humans. Surg Gynecol Obstet 1963; 117: 659–76. 20. Huang J. Ethical and legislative perspectives on liver transplantation in the People’s Republic of China. Liver Transpl 2007; 13: 193–6. 21. Chen C, Wang K, Lee M, et al. Liver transplantation for Wilson’s disease: report of the first successful liver transplant in Taiwan. Jap J Transplant 1987; 22: 178–84. 22. Shum E, Chern A. Amendment of the Human Organ Transplant Act. Ann Acad Med Singapore 2006; 35: 428–32. 23. de Villa VH, Lo CM, Chen CL. Ethics and rationale of living-donor liver transplantation in Asia. Transplantation 2003; 75(Suppl 3): S2–5. 24. Nagasue N, Kohno H, Matsuo S, et al. Segmental (partial) liver transplantation from a living donor. Transplant Proc 1992; 24: 1958–9. 25. Raia S, Nery JR, Mies S. Liver transplantation from live donors. Lancet 1989; 2: 497. 26. Hashikura Y, Makuuchi M, Kawasaki S, et al. Successful living-related partial liver transplantation to an adult patient. Lancet 1994; 343: 1233–4. 27. Yamaoka Y, Washida M, Honda K, et al. Liver transplantation using a right lobe graft from a living related donor. Transplantation 1994; 57: 1127–30. 28. Lo CM, Fan ST, Liu CL, et al. Extending the limit on the size of adult recipient in living donor liver transplantation using extended right lobe graft. Transplantation 1997; 63: 1524–8. 29. Lo CM, Fan ST, Liu CL, et al. Adult-to-adult living donor liver transplantation using extended right lobe grafts. Ann Surg 1997; 226: 261–9. 30. Miyagawa S, Hashikura Y, Miwa S, et al. Concomitant caudate lobe resection as an option for donor hepatectomy in adult living related liver transplantation. Transplantation 1998; 66: 661–3. 31. Sugawara Y, Makuuchi M, Takayama T, et al. Liver transplantation using a right lateral sector graft from a living donor to her granddaughter. Hepatogastroenterology 2001; 48: 261–3.

Proportion of survival

0.8

0.6 DDLT

0.4

0.2 LDLT 0.0 0 12 Months Figure 22.5 Cumulative survival curves after the first detection of hepatocellular carcinoma recurrence in DDLT and LDLT groups. There is no statistically significant difference between these two groups (41). 24 36

contrast to previous cytotoxic agents, most targeted agents do not induce regression of tumors but stabilize disease progression. Sorafenib is the first drug proven to prolong survival in advanced stage patients, but the survival benefit is still relatively short. Therefore, future trials should be conducted regarding the combination of sorafenib with other pre-existing treatment modalities or new targeted agents, especially in the potential role of assisting in a bridge to LT (35,87).

conclusion
The high prevalence of HCC and subsequent high incidence of HCC in the situation of very low organ donation rate in Asia lead to making a unique pattern of indication and strategy in the application of LT. HCC has begun to be a main indication of LT, especially in the form of LDLT. Effective promotion of deceased organ donation is essential in every Asian country. Although LDLT has become the main form of LT in Asia, donor safety should always be emphasized. The selection indication of LT for HCC is likely to be expanded further, but it should be prudently adopted after qualified risk–benefit analyses. Application for small HCC in patients with preserved liver function has also been an expanded indication after gradual improvements in peri-operative mortality. As before, new innovative surgical techniques will emerge to solve currently intractable technical problems. Emergence of new effective treatment modalities for HCC recurrence will expand the selection criteria further without compromising disease recurrence rate. The practice of LT in Asia will continue to evolve further, giving more benefits to patients with HCC.

references
1. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55: 74–108. 2. McGlynn KA, London WT. Epidemiology and natural history of hepatocellular carcinoma. Best Pract Res Clin Gastroenterol 2005; 19: 3–23. 3. de Villa V, Lo CM. Liver transplantation for hepatocellular carcinoma in Asia. Oncologist 2007; 12: 1321–31. 4. Lee SG, Hwang S, Moon DB, et al. Expanded indication criteria of living donor liver transplantation for hepatocellular carcinoma at one largevolume center. Liver Transpl 2008; 14: 935–45.

213

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
32. Lee SG, Lee YJ, Park KM, et al. Anterior segment congestion of a right liver lobe graft in living donor liver transplantation and its strategy to prevent congestion. J Korean Soc Transplant 1999; 13: 213–9. 33. Lee S, Hwang S, Park K, et al. An adult-to-adult living donor liver transplant using dual left lobe grafts. Surgery 2001; 129: 647–50. 34. Lee SG. Asian contribution to living donor liver transplantation. Gastroenterol Hepatol 2006; 21: 572–4. 35. Hwang S, Lee SG, Lee YJ, et al. Lessons learned from 1,000 living donor liver transplantations in a single center: how to make living donations safe. Liver Transpl 2006; 12: 920–7. 36. Lo CM. Complications and long-term outcome of living liver donors: a survey of 1,508 cases in five Asian centers. Transplantation 2003; 75: S12–5. 37. Akabayashi A, Slingsby BT, Fujita M. The first donor death after livingrelated liver transplantation in Japan. Transplantation 2004; 77: 634. 38. Hsu HT, Hwang SL, Lee PH, Chen SC. Impact of liver donation on quality of life and physical and psychological distress. Transplant Proc 2006; 38: 2102–5. 39. Kaihara S, Kiuchi T, Ueda M, et al. Living-donor liver transplantation for hepatocellular carcinoma. Transplantation 2003; 75: S37–40. 40. Todo S, Furukawa H, and the Japanese Study Group on Organ Transplantation. Living donor liver transplantation for adult patients with hepatocellular carcinoma: experience in Japan. Ann Surg 2004; 240: 451–9. 41. Hwang S, Lee SG, Joh JW, et al. Liver transplantation for adult patients with hepatocellular carcinoma in Korea: comparison between cadaveric donor and living donor liver transplantations. Liver Transpl 2005; 11: 1265–72. 42. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334: 693–9. 43. Marsh JW, Dvorchik I, Bonham CA, Iwatsuki S. Is the pathologic TNM staging system for patients with hepatoma predictive of outcome? Cancer 2000; 88: 538–43. 44. Gondolesi GE, Roayaie S, Munoz L, et al. Adult living donor liver transplantation for patients with hepatocellular carcinoma: extending UNOS priority criteria. Ann Surg 2004; 239: 142–9. 45. Takada Y, Ueda M, Ito T, et al. Living donor liver transplantation as a second-line therapeutic strategy for patients with hepatocellular carcinoma. Liver Transpl 2006; 12: 912–9. 46. Yao FY, Ferrell L, Bass NM, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. Hepatology 2001; 33: 1394–403. 47. Yao FY, Xiao L, Bass NM, et al. Liver transplantation for hepatocellular carcinoma: validation of the UCSF-expanded criteria based on preoperative imaging. Am J Transplant 2007; 7: 2587–96. 48. Weber M, Clavien PA. Hepatocellular carcinoma and liver transplantation: entering the area after the Milan and University of California at San Francisco criteria? Liver Transpl 2008; 14: 911–4. 49. Bozorgzadeh A, Orloff M, Abt P, et al. Survival outcomes in liver transplantation for hepatocellular carcinoma, comparing impact of hepatitis C versus other etiology of cirrhosis. Liver Transpl 2007; 13: 807–13. 50. Sasaki Y, Yamada T, Tanaka H, et al. Risk of recurrence in a long-term follow-up after surgery in 417 patients with hepatitis B- or hepatitis C-related hepatocellular carcinoma. Ann Surg 2006; 244: 771–80. 51. Hwang S, Lee SG, Ahn CS, et al. Prevention of hepatitis B recurrence after living donor liver transplantation: primary high-dose hepatitis B immunoglobulin monotherapy and rescue antiviral therapy. Liver Transpl 2008; 14:770–8. 52. Soejima Y, Taketomi A, Yoshizumi T, et al. Extended indication for living donor liver transplantation in patients with hepatocellular carcinoma. Transplantation 2007; 83: 893–9. 53. Duffy JP, Varadanian A, Benjamin E, et al. Liver transplantation criteria for hepatocellular carcinoma should be expanded: a 22-year experience with 467 patients at UCLA. Ann Surg 2007; 246: 502–11. 54. Onaca N, Davis GL, Goldstein RM, et al. Expanded criteria for liver transplantation in patients with hepatocellular carcinoma: a report from the International Registry of Hepatic Tumors in Liver Transplantation. Liver Transpl 2007; 13: 391–9. 55. Jonas S, Mittler J, Pascher A, et al. Living donor liver transplantation of the right lobe for hepatocellular carcinoma in cirrhosis in a European center. Liver Transpl 2007; 13: 896–903. 56. Llovet JM, Bruix J, Gores GJ. Surgical resection versus transplantation for early hepatocellular carcinoma: clues for the best strategy. Hepatology 2000; 31: 1019–21. 57. Bruix J, Llovet JM. Prognostic prediction and treatment strategy in hepatocellular carcinoma. Hepatology 2002; 35: 519–24. 58. Bigourdan JM, Jaeck D, Meyer N, et al. Small hepatocellular carcinoma in Child A cirrhotic patients: hepatic resection versus transplantation. Liver Transpl 2003; 9: 513–20. 59. Margarit C, Escartin A, Castells L, et al. Resection for hepatocellular carcinoma is a good option in Child-Turcotte-Pugh class A patients with cirrhosis who are eligible for liver transplantation. Liver Transpl 2005; 11: 1242–51. 60. Moon DB, Lee SG, Hwang S. Liver transplantation for hepatocellular carcinoma: single nodule with Child-Pugh class A sized less than 3 cm. Dig Dis 2007; 25: 320–8. 61. Soderdahl G, Backman L, Isoniemi H, et al. A prospective, randomized, multi-centre trial of systemic adjuvant chemotherapy versus no additional treatment in liver transplantation for hepatocellular carcinoma. Transpl Int 2006; 19: 288–94. 62. Porrett PM, Peterman H, Rosen M, et al. Lack of benefit of pre-transplant locoregional hepatic therapy for hepatocellular cancer in the current MELD era. Liver Transpl 2006; 12: 665–73. 63. Decaens T, Roudot-Thoraval F, Bresson-Hadni S, et al. Impact of pretransplantation transarterial chemoembolization on survival and recurrence after liver transplantation for hepatocellular carcinoma. Liver Transpl 2005; 11: 767–75. 64. Lubienski A. Hepatocellular carcinoma: interventional bridging to liver transplantation. Transplantation 2005; 80(Suppl 1): S113–S119. 65. Perez Saborido B, Meneu JC, Moreno E, et al. Is transarterial chemoembolization necessary before liver transplantation for hepatocellular carcinoma? Am J Surg 2005; 190: 383–7. 66. Decaens T, Roudot-Thoraval F, Bresson-Hadni S, et al. Impact of pretransplantation transarterial chemoembolization on survival and recurrence after liver transplantation for hepatocellular carcinoma. Liver Transpl 2005; 11: 767–75. 67. Obed A, Beham A, Pullmann K, et al. Patients without hepatocellular carcinoma progression after transarterial chemoembolization benefit from liver transplantation. World J Gastroenterol 2007; 13: 761–7. 68. Yao FY, Hirose R, LaBerge JM, et al. A prospective study on downstaging of hepatocellular carcinoma prior to liver transplantation. Liver Transpl 2005; 11: 1505–14. 69. Millonig G, Graziadei IW, Freund MC, et al. Response to preoperative chemoembolization correlates with outcome after liver transplantation in patients with hepatocellular carcinoma. Liver Transpl 2007; 13: 272–9. 70. Otto G, Herber S, Heise M, et al. Response to transarterial chemoembolization as a biological selection criterion for liver transplantation in hepatocellular carcinoma. Liver Transpl 2006; 12: 1260–7. 71. Hwang S, Lee SG, Moon DB, et al. Salvage living donor liver transplantation after prior liver resection for hepatocellular carcinoma. Liver Transpl 2007; 13: 741–6. 72. Tanaka H, Kubo S, Tsukamoto T, et al. Recurrence rate and transplantability after liver resection in patients with hepatocellular carcinoma who initially met transplantation criteria. Transplant Proc 2005; 37: 1254–6. 73. Poon RT, Fan ST, Lo CM, et al. Long-term survival and pattern of recurrence after resection of small hepatocellular carcinoma in patients with preserved liver function: implications for a strategy of salvage transplantation. Ann Surg 2002; 235: 373–82. 74. Belghiti J, Cortes A, Abdalla EK, et al. Resection prior to liver resection for hepatocellular carcinoma. Ann Surg 2003; 238: 885–93. 75. Adam R, Azoulay D, Castaing D, et al. Liver resection as a bridge to transplantation for hepatocellular carcinoma on cirrhosis. Ann Surg 2003; 238: 508–19. 76. Roayaie S, Schwartz JD, Sung MW, et al. Recurrence of hepatocellular carcinoma after liver transplant: patterns and prognosis. Liver Transpl 2004; 10: 534–40. 77. Schwartz M, Konstadoulakis M, Roayaie S. Recurrence of hepatocellular carcinoma after liver transplantation: is immunosuppression a factor? Liver Transpl 2005; 11:494–6.

214

LIVER TRANSPLANTATION FOR HCC: ASIAN PERSPECTIVES
78. Monaco AP. The role of mTOR inhibitors in the management of posttransplant malignancy. Transplantation 2009; 87: 157–63. 79. Kneteman NM, Oberholzer J, Al Saghier M, et al. Sirolimus based immunosuppression for liver transplantation in the presence of extended criteria for hepatocellular carcinoma. Liver Transpl 2004; 10: 1301–11. 80. Wall WJ. Hepatocellular cancer, transplantation, and sirolimus. Liver Transpl 2004; 10: 1312–4. 81. De Carlis L, Giacomoni A, Pirotta V, et al. Surgical treatment of hepatocellular cancer in the era of hepatic transplantation. J Am Coll Surg 2003; 196: 887–97. 82. Yao FY, Ferrell L, Bass NM, et al. Liver transplantation for hepatocellular carcinoma: comparison of the proposed UCSF criteria with the Milan criteria and the Pittsburgh modified TMN criteria. Liver Transpl 2002; 8: 765–74. 83. Herrero JI, Sangro B, Quiroga J, et al. Influence of tumor characteristics on the outcome of liver transplantation among patients with liver cirrhosis and hepatocellular carcinoma. Liver Transpl 2001; 7: 631–6. 84. Elsharkawi M, Staib L, Henne-Bruns D, et al. Complete remission of posttransplant lung metastases from hepatocellular carcinoma under therapy with sirolimus and mycophenolate mofetil. Transplantation 2005; 77: 855–7. 85. Thomas M. Molecular targeted therapy for hepatocellular carcinoma. J Gastroenterol 2009; 44(Suppl 19): 136–41. 86. Wang Z, Zhou J, Fan J, et al. Effect of rapamycin alone and in combination with sorafenib in an orthotopic model of human hepatocellular carcinoma. Clin Cancer Res 2008; 14: 5124–30. 87. Lee HC. Systemic chemotherapy of hepatocellular carcinoma–Korean experience. Oncology 2008; 75(Suppl 1): 114–8.

215

23

Non-surgical treatment of hepatocellular carcinoma Ghassan K. Abou-Alfa and Karen T. Brown
beyond the scope of this chapter. There are currently three main categories of percutaneously administered intra-arterial therapy: bland embolization, chemo-embolization, and radioembolization. Recently drug-eluting microspheres that can be loaded with chemotherapeutic agents have been used for treating HCC as well. Chemo-embolization and Bland Embolization In last 30 years many papers have been published describing myriad techniques for performing hepatic arterial embolotherapy. Chemotherapeutic agents mixed with lipiodol and injected into the artery supplying the tumor, referred to as Trans-catheter Arterial Chemo-Embolization, or TACE, has been extensively studied and used, despite the lack of convincing pharmacokinetic data favoring this method. Studies clearly demonstrating high and prolonged concentration of chemotherapeutic agents within tumor were performed using mitomycin C, doxorubicin, and aclarubicin dissolved in hydrocarbon solvents and then in lipiodol (6), or using a lipophilic agent (7), methods which are not clinically used. When the chemotherapeutic agent is dissolved in water and then mixed with lipiodol and administered as an emulsion, concentration of drug in the tumor is immediately high, but low at 6 hours, 1 day, and 7 days (3). In a study by Raoul et al. (8) doxorubicin was given to patients intra-arterially either alone as an infusion, or emulsified with lipiodol, or with lipiodol and gelatin sponge. There was no significant difference in total amount of doxorubicin released into the circulating blood, but patients in whom gelatin sponge was used had less release within the first hour of treatment. In a prospective randomized clinical study performed to evaluate the effect of lipiodol when added to intraarterial cis-platinum and doxorubicin, there was no difference in response to treatment between the groups given only intraarterial chemotherapy compared to those who received chemotherapy emulsified with lipiodol (9). Another study evaluated intra-arterial doxorubicin versus doxorubicin with lipiodol, and found no difference in the area under the concentration–time curve, or terminal half life, and no difference in pharmacokinetic profile or systemic toxicity using the same dose schedule but administering the doxorubicin intravenously (10). Pharmacokinetic data supporting the use of intra-arterial chemotherapy, or chemotherapy plus lipiodol administered as an emulsion, as commonly used today, remains debatable. What has been demonstrated, in both pharmacokinetic and clinical studies (7,11), is the benefit derived from the addition of an embolic agent, such as Gelfoam, for chemo-embolization. This has led some authors to postulate that the primary tumoricidal effect of embolotherapy might result from ischemia induced by the embolization rather than the chemotherapy or lipiodol. Two well-known randomized trials of

introduction
Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide (1). Added to its worldwide prevalence, there is a continued rising incidence of HCC in the western hemisphere (2). More than 80% of patients present with advanced or unresectable disease, and the recurrence rates after surgical resection is as high as 50% (3). Effective therapies for advanced HCC remain an immediate need. In the field of local therapies for locally advanced disease, extensive work has been done in regard to the feasibility and effectiveness of different ablative modalities, with several issues still in debate. In the more advanced and metastatic disease settings, recent data showed an improvement in survival in advanced HCC with sorafenib. This landmark study not only helped in defining a standard of care for advanced HCC, but also generated several clinical questions among whom to treat, and how to account for the cirrhosis and underlying liver dysfunction. Second line therapies and future clinical trials are prime time discussions.

local regional therapies
While surgical treatment remains the best hope for cure in patients with primary HCC cancer, many patients are not candidates for resection or transplantation at the time of presentation. This may be due to the extent or distribution of disease, underlying liver function, or the general medical condition of the patient. Although transplantation is essentially curative, patients spend time on a transplant list awaiting a donor organ. During this period they are at risk for progression. Patients who undergo surgical resection have a relatively high risk of developing recurrent disease in the remaining liver (4). Thus, local regional therapies, including trans-arterial embolization and percutaneous chemical or thermal ablation play a significant role in the care of these patients. Arterial Embolization The basic physiologic principle that makes hepatic intra-arterial therapies for HCC feasible is the dual blood supply to the liver. The portal vein provides over 75% of blood flow to the hepatic parenchyma, and provides the primary trophic blood supply, although in cirrhotic patients with portal hypertension there is a shift toward more dependence upon arterial blood flow. Conversely, studies performed in the early 1950s established that the primary blood supply to liver tumors was from the hepatic artery (5). Thus malignant tumors may be targeted by delivering treatments intra-arterially; theoretically treatments administered in this fashion would have little effect on the hepatic parenchyma and would result in fewer systemic side effects. Intra-arterial infusion therapies are usually carried out using permanently implanted ports or pumps and are

216

NON-SURGICAL TREATMENT OF HEPATOCELLULAR CARCINOMA
chemo-embolization which provide level 1b evidence have shown an improvement in survival compared to best supportive care (13,14). The study by Llovett et al. randomized patients into three arms: bland embolization, also referred to as hepatic arterial embolization (HAE), TACE, and best supportive care (13). Although demonstrating a convincing survival benefit from TACE, it is unfortunate that the study was stopped before enough patients had been accrued to the HAE arm to permit any statement to be made about bland embolization, in particular with respect to no treatment. When the trial was stopped, there was no significant difference in survival between the bland and chemo-embolization groups, despite the fact that Gelfoam pledgets were used as the embolic agent in the bland group. It is known that more proximal vessel occlusion within the liver leads to almost immediate development of flow distal to the occlusion via collateral vessels, as demonstrated by Michels in 1953 (15). Bland arterial embolization results in ischemic cell death, therefore the goal must be to cause terminal vessel blockade. The Gelfoam pledgets used in this trial result in more proximal vessel occlusion. The fact that there was no significant difference in survival between the chemo-embolization group and the group embolized with Gelfoam pledgets alone is thus noteworthy. Bland embolization using small particles known to cause terminal vessel blockade is the primary method of intra-arterial therapy at some institutions, and level llb evidence as supported by the data of Maluccio et al. (16). The results of bland embolization are essentially immediate, and significant tumor necrosis can be demonstrated by imaging within hours of the procedure (Fig. 23.1A–C). This is a particularly useful feature in patients who present with significant tumor burden where further progression may render them untreatable. This is also effective for treating tumor thrombus within the portal vein (Fig. 23.2A and B). Bland particle embolization can provide prompt control of disease. Little is written about the response time line in patients treated with TACE. TACE is not performed to “stasis” and the primary effect is considered to be chemotherapy related. Proponents of this method of treatment often point out that the intent is not to induce necrosis. These facts lead one to suspect that maximum response to treatment might not be immediately seen post-embolization. The bland embolization or chemo-embolization debate has been going on for quite some time now. Proponents of TACE continue to assert that this treatment results in deposition of

(A)

(B)

(C)

Figure 23.1 (A) Pre-treatment CT (arterial phase) in patient with multi-focal HCC. (B) Non-treatment CT performed 12 hours after left hepatic and phrenic artery particle (bland) embolization. Note retained contrast material and small “gas bubbles” in treated tumor. (C) Post-treatment CT (arterial phase) demonstrating imaging findings consistent with necrosis of treated tumor corresponding to treated tumor that had retained contrast and “gas bubbles” on immediate postembolization CT scan.

(A)

(B)

Figure 23.2 (A) Pre-treatment CT in patient with large HCC in right liver with tumor thrombus extending into portal vein (B) post-treatment CT 6 weeks after bland embolization demonstrating imaging findings of necrosis of tumor within liver and portal vein.

217

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
high concentrations of chemotherapy within the tumor that stays there for prolonged period of time and, rightly, point out that this method is the only one that has been proven to extend survival in the Llovett and Lo randomized trials mentioned previously (13,14). Excellent results (level IIa evidence) following chemo-embolization have been reported from Japan on 8510 patients treated between 1994 and 2001, with 1-, 3-, and 5-year survival of 82%, 47%, and 26% (17). About 74% of the patients had positive hepatitis C serology. Patients were excluded if they had evidence of nodal disease or distant metastases, and only 4% of patients had more than secondorder portal vein involvement; however, almost half of the patients were Child class B or C. Llovett et al. (13) reported the probability of survival following chemo-embolization in their group of 40 patients, the majority of whom had positive hepatitis C serology, to be 82%, 63%, and 29% at 1, 2, and 3 years, respectively, while Lo et al. reported survivals of 57%, 31%, and 26% at 1, 2, and 3 years, respectively, in their study of 40 patients; 90% of whom were hepatitis B serology positive. The cumulative survival rates at 1, 2, and 3 years in a recent report on triple drug chemo-embolization by Buijs et al. (18) who treated 190 patients were 58%, 39%, and 29%. In this level IIb evidence study 40% of patients were hepatitis C serology positive whereas 21% had positive hepatitis B serology. Finally, in a large series of patients treated with bland embolization, Maluccio et al. (16) reported 1-, 2-, and 3-year survival in 159 patients who were without extra-hepatic disease or portal vein involvement by tumor to be 84%, 66%, and 51%; all of these patients were treated without the addition of chemotherapy or lipiodol to the embolic agent. What seems clear from this data is that arterial embolotherapy is an effective method of treating HCC in an effort to prolong the patient’s survival. This has been proven in randomized studies comparing chemo-embolization to supportive care (13,14). Comparable, or better, survival results can be obtained with bland embolization, as demonstrated by Maluccio et al. (16). Drug-Eluting Microspheres Currently available drug-eluting beads, or DEBs, are preformed microspheres available in diameters ranging from 40 to 1200 µm. These spheres are deformable and made from a macromere derived from Polyvinyl Alcohol (PVA). They are typically loaded with doxorubicin when used to treat HCC, using 150 mg per treatment. The pharmacokinetic profile of the DEBs is significantly different than that which is seen with conventional TACE, with level IIb evidence that the peak drug concentration in the serum is an order of magnitude lower for DEBs and the area under the curve (AUC) is significantly lower as well (19). Objective response by EASL criteria has been reported in 70% to 80% of patients (19–21) and 1- and 2-year survival of 92.5% and 88.9% have been reported in a level IIa study of 27 patients with large or multi-focal tumors (19). Radio-embolization Yttrium 90 is a pure β emitter that can be loaded in glass or resin microspheres and delivered to the tumor intra-arterially. Only the glass microspheres are approved by the Food and Drug Administration (FDA) for use in treating HCC. Within the tumor vasculature these microspheres can deliver high dose radiation to the tumor. The microspheres can be delivered selectively to the anatomic segment or segments that contain the tumor or, in the case of multi-focal disease, can be administered in a lobar distribution. Unlike TACE or bland embolization, prior to treatment with 90Y patients must undergo planning angiography to evaluate vascular and tumor anatomy and blood flow dynamics. Non-hepatic branches in the territory to be treated that may result in gastrointestinal blood flow are occluded with coil embolization and finally a technetium 99 macroaggregated albumin scan is performed both to test for the presence of gastrointestinal blood flow and to estimate the percent of injected material that is shunted to the pulmonary vasculature. Radiation dose to the lungs of 30 Gy for a single administration or >50 Gy accumulated in multiple administrations, as well as any detectable GI blood flow not correctable by vessel occlusion precludes treatment with intra-arterial 90Y. Recent level IIb data for 90Y glass microspheres used to treat 140 patients with HCC by Atassi et al. reported a 68% response rate by EASL criteria. Median survival for Okuda stage I or II was 800 days versus 368 days and for Child A or B + C was 800 days versus 258 days (22). Level IIa evidence supports the use of radio-embolization for downstaging HCC to other methods of treatment. In a group of 35 patients who were not transplant candidates, Kulik et al. reported that 66% were downstaged to transplant, resection, or ablation and the median survival for the entire group was 800 days (23). Although 90Y has been shown to be an effective method for treating HCC, grade 3 or 4 liver toxicity is seen in up to 1/3 of low risk patients. Radio-embolization is expensive and more complicated to administer than either bland or chemo-embolization. The possibility that radiation-related hepatic fibrosis superimposed on the patients underlying liver disease may result in compromised hepatic function several years after treatment remains unknown. Percutaneous Chemical or Thermal Ablation Since the 1980s, Percutaneous Ethanol Injection Therapy (PEIT) has been widely used to treat patients with HCC (24–27). There is level IIa evidence that this method is effective, particularly when used to treat small tumors <3 to 5 cm (28). In the late 1990s there was a flurry of interest in using acetic acid injection to treat HCC. Despite encouraging results showing significant decrease in local recurrence and increase in survival after acetic acid injection when compared to ethanol, this agent was never in widespread use (29). Chemical ablation has been used primarily to treat patients with three or fewer tumors less than 3 to 5 cm. Even the smallest lesions injected with ethanol typically required several treatment sessions, although methods of treating even large tumors in one session have been described (30,31). In the late 1990s percutaneous methods of thermal ablation, particularly Radiofrequency Ablation (RFA), became available. In 1999, Livraghi et al. published a study of 86 patients who received either RFA or PEIT “randomized” by the distance the patient lived from the hospital (32). This study, which should be considered either Ib or IIa evidence, concluded that

218

NON-SURGICAL TREATMENT OF HEPATOCELLULAR CARCINOMA
“RF ablation results in a higher rate of complete necrosis and requires fewer treatment sessions than percutaneous ethanol injection. However the complication rate is higher with RF ablation than with PEIT. RF ablation is the treatment of choice for most patients with HCC.” Providing level lb evidence of the effectiveness of RFA, in 2003 Lencioni et al. reported a statistically significant difference in local recurrence-free survival in a randomized trial of 102 patients who underwent either RFA or PEIT, with no significant difference in complication rate (33). Thermal ablation has moved to the forefront of treatment for patients with HCC (34) although PEIT is still used for lesions that are unsuitable for thermal ablation, usually because of their location. The concept of combining radiofrequency ablation with embolization as a means of achieving a more complete ablation, and perhaps shortening treatment time has been explored. Better results might be expected when embolization is performed first, since the embolization procedure devascularizes the tumor, eliminating much of the “heat sink” and potentially improving the efficiency of the deposited energy. In 2000, Rossi et al. showed that HCC tumors 3.5 to 8.5 cm in diameter could be ablated in one or two sessions after occlusion of the tumor blood supply with a balloon or gelatin sponge (35). In 2002, Yamakado et al. performed chemoembolization followed by RFA in 64 patients with tumors as large as 12 cm and reported complete necrosis by imaging of all lesions regardless of size with only two instances of local recurrence (36). In a case–control retrospective study published in 2005 Maluccio et al. provided level III evidence that there was no statistically in survival in a group of 40 patients who underwent surgical resection, when compared to significant difference a group of 33 patients treated with bland embolization followed by either PEIT or RFA in tumors up to 7 cm in size (37). The latest level 1b evidence for the effectiveness of combined therapy emerged in 2008 when Cheng et al. a published a randomized-controlled trial of TACE–RFA, TACE alone or RFA alone in 291 patients with HCC larger than 3 cm (38). This paper reported statistically significant improved overall survival in patients treated with TACE–RFA compared to either method alone. The significance persisted when patients were stratified into those with one tumor or more than one tumor. Combining embolization with an ablative method seems to make sense when treating HCC. Performing the embolization prior to the ablation has the advantage of accurately defining number and size of tumors as well as making the tumors much more conspicuous for targeting with CT when the ablation is performed within 24 to 48 hours. treatment, and a control arm for several comparative trials testing either other single agents or combination regimens (41). PIAF (cisPlatin, Interferon, Adriamycin, and 5-Fluorouracil) Given the disappointing results of single agent doxorubicin and other single agent therapies, combination regimens have also been investigated. A combination of cisplatin, interferon, doxorubicin, and 5-fluorouracil (PIAF) has demonstrated promising activity in a phase II study. This regimen yielded a response rate of 26% and a median survival of 9 months (42). Of note, 13 patients (26%) who had a partial response, 9 underwent surgery, and 4 (9%) were found to have had a complete pathologic response to chemotherapy. These data were encouraging enough to consider evaluating PIAF versus doxorubicin as part of a large randomized phase III study that failed to show any survival advantage for PIAF (43). More importantly, the data of the phase II study of PIAF raised the possibility of using the combination in the neoadjuvant setting. This approach would be recommended in that specific setting of medically fit patients with good liver function in whom tumor cytoreduction is necessary to permit respectability (Level of evidence IIa, category C). Anti-angiogenic Therapies HCC is a highly vascular solid tumor, and a vascular endothelial growth factor (VEGF) has been shown to promote HCC development and metastasis in preclinical models (44). Sorafenib is a novel molecular targeted agents that inhibits both pro-angiogenic (VEGFR-1, -2, -3; PDGFR-β) and tumorigenic (RET, Flt-3, c-Kit) receptor tyrosine kinases (RTKs). Sorafenib also inhibits the serine or threonine kinase Raf-1 in vitro (45). A phase II trial of sorafenib evaluating response in patients with advanced HCC showed 33.6% of patients to have stable disease (≥16 weeks) commensurate with a median time-to-progression (TTP) of 4.2 months and the median overall survival of all patients was 9.2 months (46). The reported stable disease was commensurate with an interesting observation of central tumor necrosis was found in many patients in the study (Fig. 23.3). A sub-analysis evaluating the correlation between tumor necrosis and response was performed (47). The ratio of tumor necrosis and volume (N/T) was significantly associated with response, with responders having greater increase in the ratio between necrosis and tumor volume relative to baseline, as compared to nonresponders (p = 0.02), This data stresses the need for radiologic techniques other than RECIST to evaluate HCC response such as dynamic imaging. For practical reasons, it is recommended that patients on soarfenib be followed with triphasic CT scans or MRI and any decision to continue or stop therapy should be based on a multiple factors including patient clinical evaluation, imaging studies, and serum markers (where applicable). (Level of evidence III, category C). This phase II study led to the development of a large doubleblinded, randomized phase III trial evaluating single agent sorafenib versus placebo in patients with advanced HCC and no more than Child–Pugh A cirrhosis (48). The trial demonstrated an improvement in survival of 10.7 months in the

systemic therapies
Historical Background Chemotherapy has been studied extensively in HCC. Despite reported responses ranging between 10% and 20%, no study has shown an impact on survival. Doxorubicin has been studied extensively and there is still no agreement about its use in advanced HCC, with response rates ranging between 0% (39) and 79% (40). Despite the poor or debatable outcome of most of these trials, doxorubicin became the default standard for

219

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 23.3 Baseline and serial follow-up scans demonstrate tumor necrosis in a hepatocellular carcinoma patient.

sorafenib group versus 7.9 months in the placebo group (p < 0.001, HR = 0.69). The study led to the approval of sorafenib by the FDA for the treatment of unresectable HCC (49) and thus sorafenib became the standard of care in this setting (Level of evidence Ib, category A). The drug-related grade 3–4 toxicity profile included diarrhea (8%) and hand foot syndrome (8%). Despite the infrequency of bleeding events (<1%), one should still use caution in this regard considering the anti-angiogenic nature of sorafenib. Treating Patients with Advanced Cirrhosis The exciting results of the phase III study apply mainly to patients with Child–Pugh A score, similar to the population evaluated in the study. The safety and efficacy of sorafenib in patients with Child–Pugh B or C cirrhosis is still however being evaluated. In the phase II study evaluating sorafenib in HCC (50), 28% of patients had Child–Pugh B cirrhosis. In 28 patients from which pharmacokinetic samples were obtained, AUC (0–8) (mg h/L) was comparable between the Child–Pugh A (25.4) and Child–Pugh B (30.3) patients. Cmax (mg/L) were 4.9 and 6 for Child–Pugh A and B patients, respectively, with similar drug-related side effects profiles. However, Child–Pugh B patients had worsening of their liver function at a more frequent rate. A transient increase of serum bilirubin was reported in 40% of the patients with Child–Pugh B compared to 18% of Child–Pugh A patients. It is unclear and tough if this elevation in bilirubin is drug related or disease progression. Sorafenib acts as a substrate for UGT1A1, and the study did not collect direct bilirubin measurements, so it remains unclear if this total bilirubin may also be due to an inhibitory effect of UGT1A1 and decreased bilirubin glucuronidation. Another study evaluating sorafenib in patients with liver dysfunction may help in giving certain guidance on the use of sorafenib in patients with liver cirrhosis (51). The most commonly reported Drug-Limiting Toxicity (DLT) among patients with elevated bilirubin at baseline was further elevation of bilirubin. In the lack of any further data, one may consider the recommended doses of sorafenib reported in this study: 400 mg PO twice per day for bilirubin up to 1.5 x upper limit

of normal (ULN); 200 mg PO twice per day (or 400 mg PO daily) for bilirubin 1.5 to 3 × ULN; and to avoid sorafenib for bilirubin above 3 × ULN (Level of evidence IIa, category C). Nonetheless the safety of sorafenib in patients with HCC and advanced cirrhosis needs to be further studied. Future Developments In the advanced disease setting, the next step will be to evaluate other novel therapies or combinations of different agents looking for further improvement in survival. Bevacizumab, another potent anti-angiogenic agent, has been studied extensively in patients with advanced HCC (52,53). In one of the two studies of 28 patients, 4 had to discontinue therapy because of serious adverse events, including 1 transient ischemic attack and 3 serious esophageal bleeding events, which led to a modification of the study to identify and manage esophageal varices prior to enrollment (54). Bevacizumab has also been studied in combination with chemotherapy (55–57) and other biologic agents. The most promising bevacizumab doublet data are in combination with erlotinib (58). Patients with HCC and CLIP ≥ 3 were treated with bevacizumab and erlotinib. Based on the intent-to-treat analysis, 7 of 34 patients had radiographic responses and 27 (79.4%) had stability of disease for as long as 8 weeks. Median progression-free survival (PFS) was 9 months and median overall survival (OS) 19 months. Grade 3 and 4 fatigue and hypertension were each reported in 15% of the cases and similar grade gastrointestinal bleeds were reported in 9% of the cases. The positive outcome of this study supports the biologic relevance of this combination of anti-angiogenic therapy and tyrosine kinase inhibitor in HCC and should be explored further. Sunitinib, another multi-targeted tyrosine kinase inhibitor, has also been tested in HCC (59). Of 26 patients treated with sunitinib at 37.5 mg daily dose, 10 (38.5%) showed stability of disease, with a median PFS of 4.1 months. Another study showed similarly promising results at the dose of 50 mg, with median TTP of 21 weeks and median OS of 45 weeks (60). Sorafenib has been evaluated in combination with doxorubicin as part of a randomized double-blinded phase II

220

NON-SURGICAL TREATMENT OF HEPATOCELLULAR CARCINOMA
study of doxorubicin plus sorafenib compared to doxorubicin plus placebo in chemotherapy-naïve HCC patients (61). The primary endpoint, median time to progression, was 9 months for the doxorubicin plus sorafenib arm and 5 months for the doxorubicin plus placebo arm. An exploratory comparison of OS between the two arms showed a significant difference of 13.7 months in favor of doxorubicin plus sorafenib versus 6.5 months for doxorubicin plus placebo (p = 0.0049, HR = 0.45). The toxicity profile was similar between the two arms of the study and with toxicities commonly seen with single agent doxorubicin and sorafenib. Grade 3–4 toxicities included fatigue (15%) and neutropenia (50%) in both arms. Sorafenib-related toxicity included grade 3–4 diarrhea (11%) and grade 3–4 hand–foot syndrome (9%) in the combination arm. There was more left ventricular dysfunction in the doxorubicin plus sorafenib arm, having been reported in 19% of the cases (all grades) with 2% grade 3–4. The questionable synergy between sorafenib and doxorubicin will be best answered in the currently underway large randomized trial evaluating the combination versus sorafenib alone. Other clinical scenarios where sorafenib has the potential of being evaluated are surgical adjuvant setting, peri-trans-arterial chemo-embolization of embolization, and pre-transplant. In a study aimed at defining angiogenic activity in tumors subjected to Trans-Arterial Embolization (TAE) by evaluating the Tumor Microvessel Density (MVD) in an animal model (62), it was found that tumors treated with TAE showed varying degrees of central necrosis with residual viable tumor cells in the periphery. Tumor MVD in animals treated with TAE was significantly higher than that in the control group (23.6 vs. 17.5; p = 0.001). The animals treated with TAE showed a statistically significant increase in VEGF levels compared with the control group. The use of an anti-angiogenic therapy like sorafenib seems plausible to curb this angiogenic drive. A study evaluating sorafenib in the adjuvant setting is underway. Post-TACE is also underway. In an attempt to reduce on the dropout rate on transplant lists (63), a stabilizing agent like sorafenib seems plausible. However, the potential bleeding concern from an anti-angiogenic agent remains an issue that needs to be studied further.

references
1. Parkin DM, Bray F, Ferlay J, et al. Estimating the world cancer burden: Globocan 2000. Int J Cancer 2001; 94: 153–6. 2. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999 Mar 11; 340(10): 745–50. 3. Thomas MB, O’Beirne JP, Furuse J, Chan AT, Abou-Alfa G, Johnson P. Systemic therapy for hepatocellular carcinoma: cytotoxic chemotherapy, targeted therapy and immunotherapy. Ann Surg Oncol 2008 Apr; 15(4): 1008–14. 4. Poon R T-P, Fan S-T, lo C-M, et al. Intrahepatic recurrence after curative resection of Hepatocellular carcinoma: long-Term results of treatment & prognostic factors. Ann Surg 1999; 229(2): 216–22. 5. Breedis C, Young G. Blood supply of neoplasms of the liver. Am J Pathol 1954; 30: 969–85. 6. Konno T. Targeting cancer chemotherapeutic agents by use of lipiodol contrast medium. Cancer 1990; 66: 1897–903. 7. Egawa H, Maki A, Mori K, et al. Effects of intra-arterial chemotherapy with a new lipophilic anticancer agent, Estradiol-Chlorambucil (KM2210), dissolved in lipiodol on experimental liver tumors in rats. J Surg Oncol 1990; 44: 109–14. 8. Raoul JL, Heresbach D, Bretagne JF, et al. Chemoembolization of hepatocellular carcinomas. A study of biodistribution and pharmacokinetics of doxorubicin. Cancer 1992; 70(3): 585–90 9. Carr BI, Iwatsuki S, Baron R, et al. Intrahepatic arterial cisplatinum and doxorubicin with or without lipiodol for advanced hepatocellular carcinoma (HCC): a prospective randomized study. Proc Annu Meet Am Soc Clin Oncol 1993; 12: A668. 10. Johnson PJ, Kalayci C, Dobbs N, et al. Pharmacokinetics and toxicity of intraarterial Adriamycin for hepatocellular carcinoma: effect of coadministration of lipiodol. Hepatology 1991; 13: 120–7. 11. Nakao N, Kamino K, Miura K, et al. Recurrent hepatocellular carcinoma after partial hepatectomy: value of treatment with transcatheter chemoembolization. AJR 1991; 156: 1177–9. 12. Ngan H, Lai C, Fan S, et al. Transcatheter arterial chemoembolization in inoperable hepatocellular carcinoma: four year follow-up. J Vasc Interv Radiol 1996; 7: 419–5. 13. Llovet JM, Real MI, Montana X, et al. Arterial embolization or chemoembolization versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomized controlled trial. Lancet 2002; 359: 1734–9. 14. Lo C-M, Ngan H, Tso W-K, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 2002; 35(5): 1164–71. 15. Michels NA. Collateral arterial pathways to the liver after ligation of the hepatic artery and removal of the coeliac axis. Cancer 1953; 6: 708. 16. Maluccio MA, Covey AC, Ben Porat L, et al. Transcatheter arterial embolization with only particles for the treatment of unresectable hepatocellular carcinoma. J Vasc Interv Radiol 2008; 19: 862–9. 17. Takayasu K, Arii S, Ikai I, et al. Prospective cohort study of transarterial chemoembolization for unresectable hepatocellular carcinoma in 8510 patients. Gastroenterology 2006; 131: 461–9. 18. Buijs M, Vossen JA, Frangakis C, et al. Nonresectable hepatocellular carcinoma: Long-term toxicity in patients treated with transarterial chemoembolization – single-center experience. Radiology 2008; 249(1): 346–54. 19. Varela M, Real MI, Burrel M, et al. Chemoembolization of hepatocellular carcinoma with drug eluting beads: efficacy and doxorubicin pharmacokinetics. J Hepatol 2007; 46: 474–81. 20. Poon RTP, Tso WK, Pang RWC, et al. A phase I/II trial of chemoembolization for hepatocellular carcinoma using a novel intra-arterial drug-eluting bead. Clin Gastroenterol Hepatol 2007; 5: 1100–8. 21. Malagari K, Chatzimichael K, Alexopoulou E, et al. Transarterial chemoembolization of unresectable hepatocellular carcinoma with drug eluting beads: results of an open-label study of 62 patients. Cardiovasc Intervent Radiol 2008; 31: 269–80. 22. Atassi B, Lewandowski RJ, Kulik L, et al. Treatment of unresectable hepatocellular carcinoma using intra-arterial Y90 (TheraSphere®): long-term follow-up. Presented at the Society of Interventional Radiology Annual Meeting 2006, Toronto, ON, Canada.

conclusion
HCC remains prevalent worldwide and is showing a steady rise in incidence in the western hemisphere. A high number of patients are in need for either local or systemic therapies. When disease is limited to the liver, a local-regional method of treatment such as embolization or ablation should be the initial treatment. In some cases, combining these methods of treatment or using them sequentially is appropriate. Sorafenib is the first systemic treatment to demonstrate a survival advantage in a large, randomized, controlled phase III study and is now approved by the FDA for patients with unresectable HCC. Future studies are to look into improving the outcome with sorafenib further by combining with other systemic therapies or local therapies. Other uses of systemic therapies, that is, adjuvant therapy, are also under investigation.

221

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
23. Kulik L, Atassi B, van Holsbeeck L, et al. Yttrium-90 microspheres (TheraSphere®) treatment of unresectable hepatocellular carcinoma: downstaging to resection, RF ablation and bridge to transplantation. J Surg Oncol 2006; 94: 572–86. 24. Livraghi T, Salmi A, Bolondi L, et al. Small hepatocellular carcinoma: percutaneous alcohol injection: results in 23 patients. Radiology 1988; 168: 313–7. 25. Ebara M, Ohto M, Sugiura N, et al. Percutaneous ethanol injection for the treatment of small hepatocellular carcinoma: study of 95 patients. J Gastroenterol Hepatol 1990; 5: 616–26. 26. Giorgio A, Tarantino L, Francica G, et al. Percutaneous ethanol injection under sonographic guidance of hepatocellular carcinoma in compensated and decompensated cirrhotic patients. J Ultrasound Med 1992; 11: 587–95. 27. Shiina S, Imamura M, Omata M. Percutaneous ethanol injection therapy (PEIT) for malignant liver neoplasms. Sem Intervent Radiol 1997; 14(3): 295–302. 28. Livraghi T, Giorgio A, Marin G, et al. Hepatocellular carcinoma and cirrhosis in 746 patients: long-term results of percutaneous ethanol injection. Radiology 1995; 197: 101–8. 29. Ohnishi K, Yoshioka H, Ito S, Fujiwara K. Prospective randomized controlled trial comparing percutaneous acetic acid injection and ethanol injection for small hepatocellular carcinoma. Hepatology 1998; 27(1): 67–72. 30. Livraghi T, Lazzaroni S, Pellicano S, et al. Percutaneous ethanol injection of hepatic tumors: single-session therapy with general anesthesia. AJR 1993; 161: 1065–9. 31. Giorgio A, Tarantino L, Francica, et al. One-shot percutaneous ethanol injection of liver tumors under general anesthesia: preliminary data on efficacy and complications. Cardiovasc Intervent Radiol 1996; 19: 27–31. 32. Livraghi T, Goldberg SN, Lazzaroni S, et al. Small hepatocellular carcinoma: treatment with radiofrequency ablation versus ethanol injection. Radiology 1999; 210: 655–61. 33. Lencioni RA, Allgaier H-P, Cioni D, et al. Small Hepatocellular carcinoma in cirrhosis: randomized comparison of radio-frequency thermal ablation versus percutaneous ethanol injection. Radiology 2003; 228: 235–40. 34. Raut CP, Izzo F, Marra P, et al. Significant long-term survival after radiofrequency ablation of unresectable hepatocellular carcinoma in patients with cirrhosis. Ann Surg Oncol 22005; 128: 616–28. 35. Rossi S, Garbagnati F, Lencioni R, et al. Percutaneous radio-frequency thermal ablation of nonresectable hepatocellular carcinoma after occlusion of tumor blood supply. Radiology 2000; 217: 119–26. 36. Yamakada K, Nakatsuka A, Ohmori S, et al. Radiofrequency ablation combined with chemoembolization in hepatocellular carcinoma: treatment response based on tumor size and morphology. J Vasc Interv Radiol 2002; 13: 1225–32. 37. Maluccio M, Covey AM, Gandhi R, et al. Comparison of survival rates after bland arterial embolization & ablation versus surgical resection for treating solitary hepatocellular carcinoma up to 7cm. J Vasc Interv Radiol 2005; 16: 955–61. 38. Cheng B-Q, Jia C-Q, Liu C-T, et al. Chemoembolization combined with radiofrequency ablation for patients with hepatocellular carcinoma larger than 3 cm: a randomized controlled trial. JAMA 2008; 299(14): 1669–77. 39. Barbare JC, Ballet F, Petit J, et al. Hepatocellular carcinoma with cirrhosis: treatment with doxorubicin. Phase II evaluation. Bull Cancer 1984; 71(5): 442–5. 40. Olweny CL, Med M, Toya T, et al. Treatment of hepatocellular carcinoma with adriamycin. Preliminary communication. Cancer 1975; 36: 1250–7. 41. Simonetti RG, Liberati A, Angiolini C, et al. Treatment of hepatocellular carcinoma: a systematic review of randomized controlled trials. Ann Oncol 1997; 8: 117–36. 42. Leung TW, Patt YZ, Lau WY, et al. Complete pathological remission is possible with systemic combination chemotherapy for inoperable hepatocellular carcinoma. Clin Cancer Res 1999 Jul; 5(7):1676–81,. 43. Yeo W, Mok TS, Zee B, et al. A randomized phase III study of doxorubicin versus cisplatin/interferon α-2b/doxorubicin/fluorouracil (PIAF) combination chemotherapy for unresectable hepatocellular carcinoma. J Natl Cancer Inst 2005; 97: 1532–8. 44. Yoshiji H, Kuriyama S, Yoshii J, et al. Halting the interaction between vascular endothelial growth factor and its receptors attenuates liver carcinogenesis in mice. Hepatology 39:1517–24, 2004. 45. Wilhelm SM, Carter C, Tang L, et al. BAY 43-9006 exhibits broad spectrum oral anti-tumor activity and targets the Raf/MEK/ERK pathway and receptor tyrosine kinases involved in tumor progression and angiogenesis. Cancer Res 2004; 64: 7099–109. 46. Abou-Alfa GK, Schwartz L, Ricci S, et al. Phase II study of sorafenib in patients with advanced hepatocellular carcinoma. J Clin Oncol 2006; 24: 1–8. 47. Abou-Alfa GK, Zhao B, Capanu M, et al. Tumor Necrosis as a Correlate for Response in Subgroup of Patients with Advanced Hepatocellular Carcinoma (HCC) Treated with Sorafenib. Stockholm: ESMO, 2008. 48. Llovet JM, Ricci S, Mazzaferro V, et al.; SHARP Investigators Study Group. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008 Jul 24; 359(4): 378–90. 49. http://www.fda.gov/CDER/Offices/OODP/whatsnew/sorafenib.htm 50. Abou-Alfa GK, Amadori D, Santoro A, et al. Is sorafenib (S) safe and effective in patients (pts) with hepatocellular carcinoma (HCC) and Child-Pugh B (CPB) cirrhosis? J Clin Oncol 2008 (May 20 suppl; abstr 4518); 26. 51. Miller A, Murry DJ, Owzar K, et al. Pharmacokinetic (PK) and phase I study of sorafenib (S) for solid tumors and hematologic malignancies in patients with hepatic or renal dysfunction (HD or RD): CALGB 60301. J Clin Oncol 2007; ASCO Annual Meeting Proceedings Part I. Vol 25, No. 18S (June 20 Suppl): 3538. 52. Siegel AB, Cohen EI, Ocean A, et al. Phase II trial evaluating the clinical and biologic effects of bevacizumab in unresectable hepatocellular carcinoma. J Clin Oncol 2008; 26: 2992–8. 53. Malka D, Dromain C, Farace F, et al. Bevacizumab in patients (pts) with advanced hepatocellular carcinoma (HCC): preliminary results of a phase II study with circulating endothelial cell (CEC) monitoring. J Clin Oncol 2007; ASCO Annual Meeting Proceedings Part I. Vol 25, No. 18S (June 20 Suppl): 4570 54. Schwartz JD, Schwartz M, Goldman J, et al. Bevacizumab in hepatocellular carcinoma (HCC) for patients without metastasis and without invasion of the portal vein. 2005 Gastrointestinal Cancer Symposium, No. 134. 55. Zhu AX, Blaskowsky LS, Ryan DP, et al. Phase II study of gemcitabine and oxaliplatin in combination with bevacizumab in patients with advanced hepatocellular carcinoma. J Clin Oncol 2006; 24: 1898–903. 56. Sun W, Haller DG, Mykulowycz K, et al. Combination of capecitabine, oxaliplatin with bevacizumab in treatment of advanced hepatocellular carcinoma (HCC): a phase II study. J Clin Oncol 2007; ASCO Annual Meeting Proceedings Part I. Vol 25, No. 18S (June 20 Suppl): 4574. 57. Hsu C, Yang T, Hsu C, et al. Modified-dose capecitabine + bevacizumab for the treatment of advanced/metastatic hepatocellular carcinoma (HCC): a phase II, single-arm study. J Clin Oncol 2007; ASCO Annual Meeting Proceedings Part I. Vol 25, No. 18S (June 20 Suppl): 15190. 58. Thomas MB, Morris JS, Chadha R, et al. Phase II trial of the combination of bevacizumab and erlotinib in patients who have advanced hepatocellular carcinoma. J Clin Oncol 2009; 27: 843–50. 59. Zhu AX, Sahani DV, Duda DG, et al. Efficacy, safety, and potential biomarkers of sunitinib monotherapy in advanced hepatocellular carcinoma: a phase II study. J Clin Oncol 2009; 27: 3027–35. 60. Faivre S, Raymond E, Boucher E, et al. Safety and efficacy of sunitinib in patients with advanced hepatocellular carcinoma: an open-label, multicentre, phase II study. Lancet Oncol 2009; 10: 743–4. 61. Abou-Alfa GK, Johnson P, Knox J, et al. Final results from a phase II (PhII), randomized, double-blind study of sorafenib plus doxorubicin (S+D) versus placebo plus doxorubicin (P+D) in patients (pts) with advanced hepatocellular carcinoma (AHCC). 2008 Gastrointestinal Cancers Symposium, No. 128 62. Gupta S, Kobayashi S, Phongkitkarun S, et al. Effect of transcatheter hepatic arterial embolization on angiogenesis in an animal model. Invest Radiol 2006 Jun; 41(6): 516–21. 63. Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology 1999; 30: 1434–40.

222

24 Resection of intrahepatic cholangiocarcinoma
Intrahepatic cholangiocarcinoma (ICC) is a rare (1,2) and poorly understood biliary malignancy that is increasing in incidence and therefore, in importance. The literature concerning this disease is mostly limited to retrospective reviews of patients from specialized hepatobiliary centers; moreover, only a few case series have included 50 or more patients (3–6). Surgical resection has offered the only chance of cure in patients since the earliest era (7). However, no consensus or guidelines have been established on the optimum procedure in accordance with the stage of the disease. This article is a review of the current knowledge on intrahepatic cholangiocarcinoma, especially from the viewpoint of experienced liver surgeons who treat it.

Junichi Arita, Norihiro Kokudo, and Masatoshi Makuuchi
a better prognosis than the other types (15–20). The reason for the better prognosis of the latter type of ICC is that this type can be detected in the early stages based on positive radiologic signs or by the detection of liver dysfunction because of biliary obstruction. IG-ICC is often diagnosed before tumor extension beyond the bile ducts and patients with this type of ICC present with smaller tumor sizes than patients with the other types of ICC (20).

epidemiology and risk factors
Over 80% of patients with ICC are reported from the developing countries of Asia and Africa, and moreover half are from China (21). ICC accounts for only 1% of all liver malignancies in the United States (22), whereas the reported proportion from Japan is much higher at around 4% (23). The incidence and mortality have been reported to be increasing in most countries where wide disease surveillance has been undertaken (24,25). Several risk factors for ICC have been suggested, however, in most patients with ICC, no such risk factors can be identified (26). Conditions pre-disposing to ICC include infestation with parasitic liver flukes, such as Opisthorchis viverrini and Clonorchis sinensis (27,28); exposure to Thorotrast (29); hepatitis C virus infection (30–33); smoking (27,33); and HIV infection (32). Some common diseases co-existing with ICC have been reported, including primary sclerosing cholangitis and associated ulcerative colitis (26,34), choledochal cyst, Caroli’s syndrome (35), congenital hepatic fibrosis, alcoholic liver disease, and diabetes mellitus (32,33). There are three reports from the United States and Denmark of the results of surveys of large populations conducted to determine the risk factors of ICC (32,33,36). Common risk factors identified in at least two of these were cholangitis, gallbladder stones, choledocholithiasis, alcoholic liver disease, non-specific cirrhosis, hepatitis C virus infection, diabetes mellitus, inflammatory bowel disease, and smoking (32,33,36). Interestingly, although correlation with hepatitis C virus infection has often been reported, involvement of hepatitis B virus infection has never been reported.

terminology and classification
The term “cholangiocarcinoma” originally referred to primary tumors of the intrahepatic bile ducts, but not extrahepatic bile ducts (8); however, in its current usage, it includes intrahepatic, perihilar, and distal extrahepatic tumors of the bile ducts (9,10). Therefore, use of the sole word “cholangiocarcinoma” as a disease category is confusing and not recommended. ICC is a carcinoma arising from the mucosa of the intrahepatic bile ducts. The separation between “intrahepatic” and “extrahepatic” bile ducts occurs at the second-order branches of the biliary tree according to the Japanese classification (11), and extrahepatic bile ducts includes the hilar and common bile duct. ICC accounts for approximately 10% of all carcinomas arising from the biliary system (12). In addition to the above-mentioned confusion, “peripheral cholangiocarcinoma” and “cholangiocellular carcinoma” are used almost synonymously with ICC. Attempts have been made to morphologically classify ICC. The most popular classification, advocated by the Liver Cancer Study Group of Japan, is the mass-forming type, periductalinfiltrating type, and intraductal-growth type (11) (Fig. 24.1). The mass-forming type of ICC (MF-ICC) is a localized nodular tumor in the liver parenchyma, with a distinct border and primarily grows expansively, although it sometimes shows hilar invasion (3,13). The periductal-infiltrating type of ICC invades the connective tissue within Glisson’s sheath in an infiltrative rather than expansive fashion. The intraductalgrowth type of ICC (IG-ICC) is characterized by papillary or nodular growth within the lumen of the bile ducts. Some of these tumors exhibit superficial mucosal spread with minimal changes on radiological images, while others may occlude the intrahepatic bile duct to induce macroscopic bile duct dilatation (13). Among the three morphologic types, MF-ICC is the most prevalent, so that much of the description in the following sections refers to this type of the tumor. Originally, MF-ICC was thought to be a distinct category with a poorer prognosis compared with the other types (3,14). However, it has been elucidated that IG-ICC is associated with

pre-operative diagnosis
Because previous surveys of ICC have failed to define highrisk groups (37), unlike the case for hepatocellular carcinoma, early detection of ICC patients remains difficult. Moreover, absence of early symptoms in most patients with ICC leads to delays in diagnosis. Some of the possible symptoms associated with ICC include abdominal pain, anemia, and weight loss (38). In contrast to patients with extrahepatic bile duct carcinoma, only 0% to 28% of patients with ICC present with obstructive jaundice (5,6,38). There are no definite tumor markers for ICC, although carcinoembryonic antigen (CEA) and carbohydrate antigen

223

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C) Figure 24.1 Morphologic classification of intrahepatic cholangiocarcinoma according to Liver Cancer Study Group of Japan. (A) Mass-forming type. (B) Periductalinfiltrating type. (C) Intraductal-growth type. Source : From Ref. 11.

19–9 (CA19–9) are frequently elevated and so used for both helping to confirm the diagnosis and post-operative monitoring for tumor recurrence. The sensitivity and specificity of CA19–9 for the diagnosis of ICC have been reported to be in the range of 70% to over 90% in patients with primary sclerosing cholangitis, using various cut-off points (39–42), although a much lower sensitivity (53%) was reported in a population without primary sclerosing cholangitis (43). Low sensitivities for the diagnosis of ICC have also been reported for other serum markers that are sometimes used to identify extrahepatic bile duct carcinoma, for example, the serum levels of liver transaminases, alkaline phosphatase, and total bilirubin (5). Because ICCs are essentially adenocarcinomas, they often exhibit imaging findings similar to those of metastatic liver tumors from carcinomas of the gastrointestinal tract. Typically, the tumors are partially spherical in shape, with an irregular surface, and are almost hypovascular, with peripheral enhancement on contrast-enhanced CT, MRI, and ultrasound (44–46). Gastroscopy, colonoscopy, CT with optional positron emission tomography (PET) of the thorax and abdomen should be done to differentiate ICCs from metastatic tumors originating from the gastrointestinal tract. Percutaneous tumor biopsy is not recommended if surgery is considered, because it may cause tumor seeding and complicate the surgical procedure (47). The diagnostic ability of newer imaging modalities, including FDG-PET, for ICC still remains under debate. One study reported high SUV values in all of the 10 cases examined (48), and another reported that the diagnostic accuracy of PET for lymph node metastasis from this type of tumor was 86%; figures that were higher than those reported for CT or MRI; also, higher SUV values were reported to be correlated with a poorer prognosis (49).

surgical strategy
As with hepatocellular carcinoma, the counterpart of ICC as a primary liver cancer, the only treatment that may offer a

chance of long survival in patients with ICC is surgery. In the natural history of unresected ICC, the median survival after diagnosis is less than 1 year (50,51) and most patients present at an advanced stage when they are no longer suitable candidates for curative resection. A study from the United States reported that complete tumor resection was possible in only 29 of 44 patients who were considered for curative resection (5). The prognosis after palliative resection is poor, with median reported survivals of between 2 and 3 months (50,51). No consensus has been established with regard to the criteria for surgery, because of the small number of patients in each reported series. However, based on the literature, it would seem that surgery should be offered to all patients with potentially resectable ICC, regardless of tumor stage. Some have claimed that positive lymph node metastasis detected on examination of frozen sections at laparotomy is a contraindication for tumor resection (3), whereas two reports have recommended surgery for patients with lymph node metastasis, given a solitary hepatic lesion (52) or less than two lymph node metastases (53). Surgery affords no survival benefit in patients with ICC with any extent of distant metastasis (15). Although multiple hepatic lesions have been shown to be correlated with a poor prognosis (6,54–56), the indication for surgery in relation to the number of tumors remains unclear at present. There is also no established consensus with regard to the optimal surgical procedure for patients with ICC. A positive surgical margin has been shown to be associated with a poor prognosis (3,5,6,54,55) and tumor invasion along Glisson’s sheath is often observed (57,58), therefore extended anatomical hepatic resections would seem to be a more rational surgical procedure as compared with non-anatomical limited hepatic resections. However, the superiority of such a procedure has not yet been demonstrated. A recent study reported that under certain conditions, tumor resection with a minimal surgical margin may be reasonable (59). In patients with ICC

224

RESECTION OF INTRAHEPATIC CHOLANGIOCARCINOMA
100 90 80 Survival rate (%) 70 60 50 40 30 20 10 0 0 12 24 36 48 60 72 84 Survival period (month) 96 108 120 132 144 N1 495 N0 1,028

Figure 24.2 Impact of lymph node metastasis on the overall survival rate of patients with ICC who underwent tumor resection. N0 indicates patients without lymph node metastasis and N1 indicates those with lymph node metastasis. Source : From Ref. 11.

Table 24.1 Outcome After Liver Resection for Intrahepatic Cholangiocarcinoma in Literature
First author Kawarada (68) Cherqui (47) Schlinkert (66) Yamamoto (31) Pichlmayr (63) Nakeeb (9) Jan (67) Casavilla (54) Madariaga (55) Chen (38) Valverde (65) Inoue (3) Okabayashi (4) Weber (56) DeOliveira (5) Shimada (59) Paik (6) Year 1990 1995 1992 1992 1995 1996 1996 1997 1998 1999 1999 2000 2001 2001 2001 2007 2008 Country Japan France USA Japan Germany USA China USA USA Taiwan France Japan Japan USA USA Japan Korea n 11 14 6 10 32 9 41 34 34 138 30 52 60 33 44 47 97 1-year survival, % – – 59.3 – – 53.7 64 67 33 86 63 68 – – – 74.9 3-year survival, % 34 – – 44.4 – – 36.6 34 40 17 22 36 35 – – 45 51.8 5-year survival, % 34 – 33 44.4 – 44 26.8 26 35 14 – 36 29 – 40 40 31.1 Median survival, mo – 14 – – 12.8 22 12 – 19 – 28 – – 37.4 23 – –

with apparent invasion of the major portal pedicles or major hepatic veins, major hepatic resection should be considered. The decision on the appropriate surgical procedure must be made with reference to the functional liver reserve; an algorithm suggested by Makuuchi and colleagues (60) has been widely adopted for this purpose. Pre-operative portal vein embolization should be considered when the optimal surgical procedure is impossible based on the functional liver reserve (61). Some reports have recommended extrahepatic bile duct resection for MF-ICC, because the majority of this type of ICC shows biliary invasion (14,57,62). The reported incidence, from studies including at least 30 patients, of lymph node metastasis in surgically treated patients with ICC is as high as 18% to 40% (3,54,55,63–65). Patients with lymph node metastasis have a poor prognosis; a

nation-wide surveillance conducted in Japan (n = 495) revealed 1-, 3-, and 5-year survival rates of 52.4%, 23.1%, and 15.6%, respectively (Fig. 24.2) (23). The prognostic benefit of lymph node dissection remains controversial. Some authors have demonstrated its usefulness in cases where the tumor is solitary (52,53) and when a limited number of lymph nodes are metastatic (53). On the other hand, Shimada and colleagues (16) contradicted the survival benefit of lymph node dissection, because it was often seen in patients treated without lymph node dissection that recurrent lymph node metastasis occurred not at the hilar site but at distant sites.

prognosis after surgery
The prognosis after surgery in patients with ICC is slightly worse than that seen in patients with hepatocellular carcinoma.

225

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
100 90 80 Survival rate (%) 70 60 50 40 30 20 10 0 0 12 24 36 48 60 72 84 Survival period (month) 96 108 120 132 144 3,499

Figure 24.3 Overall survival rates of all patients with ICC who underwent tumor resection. Source : From Ref. 11.

Table 24.2 Prognostic Factors After Surgery for Intrahepatic Cholangiocarcinoma
First author Nakeeb (9) Jan (67) Chou (69) Casavilla (54) Madariaga (55) Valverde (65) Inoue (3) Weber (56) DeOliveira (5) Paik (6) Year 1996 1996 1997 1997 1998 1999 2000 2001 2001 2008 n 9 41 27 54 34 30 52 33 44 97 Lymph node – – U U, M U, M U U, M – U U Vascular invasion – U – U U – U, M U, M – – Positive margin U U, M – U, M – – U, M U U, M U, M Tumor size – – – – U – – – – U Multiple tumors – – – U, M U, M U – U – U, M Bilobar disease – – U U – – – – – Others – Lymphatic invasion – TNM stage – – Left lobe – – Morphologic type

U, significant by univariate analysis for overall survival; M, significant by multivariate analysis for overall survival.

According to a nation-wide survey conducted in Japan, the 1-, 3-, and 5-year survival rates in 1626 patients who underwent tumor resection for ICC were 70.5%, 43.8%, and 32.7%, respectively (Fig. 24.3) (23). The results of surgical treatment for ICC, most cases of which are accounted for by MF-ICC, are summarized in (Table 24.1 (3–6,9,31,38,47,54–56,59,63,65– 68). Median survival time ranged from 12 to 37 months, and the 5-year survival rate ranged from 14% to 44%; the wide range was probably caused by the small number of patients included in the studies. According to some studies, the pre-operative mortality rates ranged between 2% and 5%. The most common site of recurrence was the liver, followed in frequency by lungs, bones, peritoneum, adrenal glands, and kidneys (54,56). The prognostic factors in patients undergoing surgery for ICC proposed in the literature are listed in (Table 24.2 (3,5,6,9,54– 56,65,67,69). The indicators of poorer prognosis proposed in these multiple reports include positive lymph node metastasis, positive surgical margin, multiple hepatic lesions, presence of vascular invasion, and a large tumor size. While a large tumor size and bilobar tumor location were identified as prognostic factors by univariate analysis in a number of studies, these could

not be confirmed as prognostic factors by multivariate analysis. Despite these suggested prognostic factors, the contraindications to surgery have not yet been fully elucidated.

adjuvant therapy
Because the prognosis after tumor resection alone is far from satisfaction, adjuvant therapy may be considered in some patients. There have been only a few reports of long-term survival with such therapy among patients with ICC (70,71). Moreover, there have been no reported randomized prospective trials of adjuvant therapy in these patients, although the CRUK BILCAP study of surgery versus surgery plus adjuvant gemcitabine continues to recruit in the United Kingdom. There have also been only a few anecdotal reports of chemotherapy for patients with recurrent or unresectable ICC, using 5-FU, cisplatin, epirubicin, gemcitabine, capecitabine, and cetuximab (72–76). Further therapeutic trials are therefore warranted using recently developed treatment modalities or those that are developed in the future, including radiotherapy, immunotherapy, and chemotherapy with or without molecule-targeted drugs.

226

RESECTION OF INTRAHEPATIC CHOLANGIOCARCINOMA

references
1. SEER Cancer Statistics Review 1973–2005 [database on the Internet]. U.S. National Institute of Health. [cited 2009/02/08]. Available from: http:// www.seer.cancer.gov/ 2. Cancer Information Service [database on the Internet]. National Cancer Center. [cited 2009/02/08]. Available from: http://ganjoho.jp/ 3. Inoue K, Makuuchi M, Takayama T, et al. Long-term survival and prognostic factors in the surgical treatment of mass-forming type cholangiocarcinoma. Surgery 2000; 127: 498–505. 4. Okabayashi T, Yamamoto J, Kosuge T, et al. A new staging system for mass-forming intrahepatic cholangiocarcinoma: analysis of preoperative and postoperative variables. Cancer 2001; 92: 2374–83. 5. DeOliveira ML, Cunningham SC, Cameron JL, et al. Cholangiocarcinoma: thirty-one-year experience with 564 patients at a single institution. Ann Surg 2007; 245: 755–62. 6. Paik KY, Jung JC, Heo JS, et al. What prognostic factors are important for resected intrahepatic cholangiocarcinoma? J Gastroenterol Hepatol 2008; 23: 766–70. 7. Hammill CW, Wong LL. Intrahepatic cholangiocarcinoma: a malignancy of increasing importance. J Am Coll Surg 2008; 207: 594–603. 8. Kokudo N, Makuuchi M. Extent of resection and outcome after curative resection for intrahepatic cholangiocarcinoma. Surg Oncol Clin N Am 2002; 11: 969–83. 9. Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg 1996; 224: 463–73; discussion 73–5. 10. de Groen PC, Gores GJ, LaRusso NF, et al. Biliary tract cancers. N Engl J Med 1999; 341: 1368–78. 11. Liver Cancer Study Group of Japan. The General Rules for the Clinical and Pathological Study of Primary Liver Cancer, 5th ed. Tokyo: Kanehara, 2008. 12. Malhi H, Gores GJ. Cholangiocarcinoma: modern advances in understanding a deadly old disease. J Hepatol 2006; 45: 856–67. 13. Sano T, Kamiya J, Nagino M, et al. Macroscopic classification and preoperative diagnosis of intrahepatic cholangiocarcinoma in Japan. J Hepatobiliary Pancreat Surg 1999; 6: 101–7. 14. Yamanaka N, Okamoto E, Ando T, et al. Clinicopathologic spectrum of resected extraductal mass-forming intrahepatic cholangiocarcinoma. Cancer 1995; 76: 2449–56. 15. Isaji S, Kawarada Y, Taoka H, et al. Clinicopathological features and outcome of hepatic resection for intrahepatic cholangiocarcinoma in Japan. J Hepatobiliary Pancreat Surg 1999; 6: 108–16. 16. Shimada M, Yamashita Y, Aishima S, et al. Value of lymph node dissection during resection of intrahepatic cholangiocarcinoma. Br J Surg 2001; 88: 1463–6. 17. Ohtsuka M, Ito H, Kimura F, et al. Results of surgical treatment for intrahepatic cholangiocarcinoma and clinicopathological factors influencing survival. Br J Surg 2002; 89:1525–31. 18. Suh KS, Chang SH, Lee HJ, et al. Clinical outcomes and apomucin expression of intrahepatic cholangiocarcinoma according to gross morphology. J Am Coll Surg 2002; 195: 782–9. 19. Tajima Y, Kuroki T, Fukuda K, et al. An intraductal papillary component is associated with prolonged survival after hepatic resection for intrahepatic cholangiocarcinoma. Br J Surg 2004; 91: 99–104. 20. Yeh CN, Jan YY, Yeh TS, et al. Hepatic resection of the intraductal papillary type of peripheral cholangiocarcinoma. Ann Surg Oncol 2004; 11: 606–11. 21. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55: 74–108. 22. Martin R, Jarnagin W. Intrahepatic cholangiocarcinoma. Current management. Minerva Chir 2003; 58: 469–78. 23. Liver Cancer Study Group of Japan. National Surveillance of Primary Liver Cancer in Japan, 17th Report. Kyoto: Media Planning, 2006. 24. Patel T. Worldwide trends in mortality from biliary tract malignancies. BMC Cancer 2002; 2: 10. 25. Shaib YH, Davila JA, McGlynn K, et al. Rising incidence of intrahepatic cholangiocarcinoma in the United States: a true increase? J Hepatol 2004; 40: 472–7. 26. Chapman RW. Risk factors for biliary tract carcinogenesis. Ann Oncol 1999; 10 (Suppl 4): 308–11.

27. Parkin DM, Srivatanakul P, Khlat M, et al. Liver cancer in Thailand. I. A case-control study of cholangiocarcinoma. Int J Cancer 1991; 48: 323–8. 28. Watanapa P, Watanapa WB. Liver fluke-associated cholangiocarcinoma. Br J Surg 2002; 89: 962–70. 29. Khan SA, Davidson BR, Goldin R, et al. Guidelines for the diagnosis and treatment of cholangiocarcinoma: consensus document. Gut 2002; 51 Suppl 6: VI1-9. 30. Donato F, Gelatti U, Tagger A, et al. Intrahepatic cholangiocarcinoma and hepatitis C and B virus infection, alcohol intake, and hepatolithiasis: a case-control study in Italy. Cancer Causes Control 2001; 12: 959–64. 31. Yamamoto S, Kubo S, Hai S, et al. Hepatitis C virus infection as a likely etiology of intrahepatic cholangiocarcinoma. Cancer Sci 2004; 95: 592–5. 32. Shaib YH, El-Serag HB, Nooka AK, et al. Risk factors for intrahepatic and extrahepatic cholangiocarcinoma: a hospital-based case-control study. Am J Gastroenterol 2007; 102: 1016–21. 33. Welzel TM, Graubard BI, El-Serag HB, et al. Risk factors for intrahepatic and extrahepatic cholangiocarcinoma in the United States: a populationbased case-control study. Clin Gastroenterol Hepatol 2007; 5: 1221–8. 34. Bergquist A, Ekbom A, Olsson R, et al. Hepatic and extrahepatic malignancies in primary sclerosing cholangitis. J Hepatol 2002; 36: 321–7. 35. Dayton MT, Longmire WP, Jr., Tompkins RK. Caroli’s disease: a premalignant condition? Am J Surg 1983; 145: 41–8. 36. Welzel TM, Mellemkjaer L, Gloria G, et al. Risk factors for intrahepatic cholangiocarcinoma in a low-risk population: a nationwide case-control study. Int J Cancer 2007; 120: 638–41. 37. Yamanaka N, Takaya Y, Oriyama T, et al. Hepatoprotective effect of a nonselective endothelin receptor antagonist (TAK-044) in the transplanted liver. J Surg Res 1997; 70: 156–60. 38. Chen MF, Jan YY, Jeng LB, et al. Intrahepatic cholangiocarcinoma in Taiwan. J Hepatobiliary Pancreat Surg 1999; 6: 136–41. 39. Nichols JC, Gores GJ, LaRusso NF, et al. Diagnostic role of serum CA 19–9 for cholangiocarcinoma in patients with primary sclerosing cholangitis. Mayo Clin Proc 1993; 68: 874–9. 40. Chalasani N, Baluyut A, Ismail A, et al. Cholangiocarcinoma in patients with primary sclerosing cholangitis: a multicenter case-control study. Hepatology 2000; 31: 7–11. 41. Levy C, Lymp J, Angulo P, et al. The value of serum CA 19-9 in predicting cholangiocarcinomas in patients with primary sclerosing cholangitis. Dig Dis Sci 2005; 50: 1734–40. 42. Charatcharoenwitthaya P, Enders FB, Halling KC, et al. Utility of serum tumor markers, imaging, and biliary cytology for detecting cholangiocarcinoma in primary sclerosing cholangitis. Hepatology 2008; 48: 1106–17. 43. Patel AH, Harnois DM, Klee GG, et al. The utility of CA 19-9 in the diagnoses of cholangiocarcinoma in patients without primary sclerosing cholangitis. Am J Gastroenterol 2000; 95: 204–7. 44. Miura F, Okazumi S, Takayama W, et al. Hemodynamics of intrahepatic cholangiocarcinoma: evaluation with single-level dynamic CT during hepatic arteriography. Abdom Imaging 2004; 29: 467–71. 45. Kim SJ, Lee JM, Han JK, et al. Peripheral mass-forming cholangiocarcinoma in cirrhotic liver. AJR 2007; 189: 1428–34. 46. Chen LD, Xu HX, Xie XY, et al. Enhancement patterns of intrahepatic cholangiocarcinoma: comparison between contrast-enhanced ultrasound and contrast-enhanced CT. Br J Radiol 2008; 81: 881–9. 47. Cherqui D, Tantawi B, Alon R, et al. Intrahepatic cholangiocarcinoma. Results of aggressive surgical management. Arch Surg 1995; 130: 1073–8. 48. Kim YJ, Yun M, Lee WJ, et al. Usefulness of 18F-FDG PET in intrahepatic cholangiocarcinoma. Eur J Nucl Med Mol Imaging 2003; 30: 1467–72. 49. Seo S, Hatano E, Higashi T, et al. Fluorine-18 fluorodeoxyglucose positron emission tomography predicts lymph node metastasis, P-glycoprotein expression, and recurrence after resection in mass-forming intrahepatic cholangiocarcinoma. Surgery 2008; 143: 769–77. 50. Chu KM, Fan ST. Intrahepatic cholangiocarcinoma in Hong Kong. J Hepatobiliary Pancreat Surg 1999; 6: 149–53. 51. Kim HJ, Yun SS, Jung KH, et al. Intrahepatic cholangiocarcinoma in Korea. J Hepatobiliary Pancreat Surg 1999; 6: 142–8. 52. Uenishi T, Kubo S, Yamazaki O, et al. Indications for surgical treatment of intrahepatic cholangiocarcinoma with lymph node metastases. J Hepatobiliary Pancreat Surg 2008; 15: 417–22.

227

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
53. Nakagawa T, Kamiyama T, Kurauchi N, et al. Number of lymph node metastases is a significant prognostic factor in intrahepatic cholangiocarcinoma. World J Surg 2005; 29: 728–33. 54. Casavilla FA, Marsh JW, Iwatsuki S, et al. Hepatic resection and transplantation for peripheral cholangiocarcinoma. J Am Coll Surg 1997; 185: 429–36. 55. Madariaga JR, Iwatsuki S, Todo S, et al. Liver resection for hilar and peripheral cholangiocarcinomas: a study of 62 cases. Ann Surg 1998; 227: 70–9. 56. Weber SM, Jarnagin WR, Klimstra D, et al. Intrahepatic cholangiocarcinoma: resectability, recurrence pattern, and outcomes. J Am Coll Surg 2001; 193: 384–91. 57. Sasaki A, Aramaki M, Kawano K, et al. Intrahepatic peripheral cholangiocarcinoma: mode of spread and choice of surgical treatment. Br J Surg 1998; 85: 1206–9. 58. Shirabe K, Shimada M, Harimoto N, et al. Intrahepatic cholangiocarcinoma: its mode of spreading and therapeutic modalities. Surgery 2002; 131: S159–64. 59. Shimada K, Sano T, Sakamoto Y, et al. Clinical impact of the surgical margin status in hepatectomy for solitary mass-forming type intrahepatic cholangiocarcinoma without lymph node metastases. J Surg Oncol 2007; 96: 160–5. 60. Makuuchi M, Kosuge T, Takayama T, et al. Surgery for small liver cancers. Semin Surg Oncol 1993; 9: 298–304. 61. Kubota K, Makuuchi M, Kusaka K, et al. Measurement of liver volume and hepatic functional reserve as a guide to decision-making in resectional surgery for hepatic tumors. Hepatology 1997; 26: 1176–81. 62. Suzuki S, Sakaguchi T, Yokoi Y, et al. Clinicopathological prognostic factors and impact of surgical treatment of mass-forming intrahepatic cholangiocarcinoma. World J Surg 2002; 26: 687–93. 63. Pichlmayr R, Lamesch P, Weimann A, et al. Surgical treatment of cholangiocellular carcinoma. World J Surg 1995; 19: 83–8. 64. Chu KM, Lai EC, Al-Hadeedi S, et al. Intrahepatic cholangiocarcinoma. World J Surg 1997; 21: 301–5; discussion 5–6. 65. Valverde A, Bonhomme N, Farges O, et al. Resection of intrahepatic cholangiocarcinoma: a Western experience. J Hepatobiliary Pancreat Surg 1999; 6: 122–7. 66. Schlinkert RT, Nagorney DM, Van Heerden JA, et al. Intrahepatic cholangiocarcinoma: clinical aspects, pathology and treatment. HPB Surg 1992; 5: 95–101; discussion –2. 67. Jan YY, Jeng LB, Hwang TL, et al. Factors influencing survival after hepatectomy for peripheral cholangiocarcinoma. Hepatogastroenterology 1996; 43: 614–9. 68. Kawarada Y, Yamagiwa K, Das BC. Analysis of the relationships between clinicopathologic factors and survival time in intrahepatic cholangiocarcinoma. Am J Surg 2002; 183: 679–85. 69. Chou FF, Sheen-Chen SM, Chen YS, et al. Surgical treatment of cholangiocarcinoma. Hepatogastroenterology 1997; 44: 760–5. 70. Izumi R, Shimizu K, Kiriyama M, et al. Long-term survival of peripheral intrahepatic cholangiocarcinoma with distant metastasis. Am J Gastroenterol 1995; 90: 505–7. 71. Thongprasert S. The role of chemotherapy in cholangiocarcinoma. Ann Oncol 2005; 16 (Suppl 2): ii93–6. 72. Knox JJ, Hedley D, Oza A, et al. Combining gemcitabine and capecitabine in patients with advanced biliary cancer: a phase II trial. J Clin Oncol 2005; 23: 2332–8. 73. Park SH, Park YH, Lee JN, et al. Phase II study of epirubicin, cisplatin, and capecitabine for advanced biliary tract adenocarcinoma. Cancer 2006; 106: 361–5. 74. Paule B, Herelle MO, Rage E, et al. Cetuximab plus gemcitabine-oxaliplatin (GEMOX) in patients with refractory advanced intrahepatic cholangiocarcinomas. Oncology 2007; 72: 105–10. 75. Kim MJ, Oh DY, Lee SH, et al. Gemcitabine-based versus fluoropyrimidinebased chemotherapy with or without platinum in unresectable biliary tract cancer: a retrospective study. BMC Cancer 2008; 8: 374. 76. Takezako Y, Okusaka T, Ueno H, et al. Phase II study of cisplatin, epirubicin and continuous infusion of 5-fluorouracil in patients with advanced intrahepatic cholangiocellular carcinoma (ICC). Hepatogastroenterology 2008; 55: 1380–4.

228

25

Transplantation for hilar cholangiocarcinoma Julie K. Heimbach, Charles B. Rosen, and David M. Nagorney
desmoplastic tumor which leads to the inability to obtain an adequate number of cells to make a definitive diagnosis. Fluorescent in situ hybridization (FISH) is a cytologic technique which allows for the detection of aneuploidy in epithelial cells. In particular, FISH polysomy (two or more chromosomes duplicated) identifies another 20% of patients with CCA missed by conventional cytology while maintaining a specificity of 100% (13). Despite the use of advances cytologic techniques, the ability to make a timely and accurate diagnosis remains sub-optimal. The primary therapy for hilar CCA is surgical resection. There is no effective systemic therapy, though radiation may provide palliation and in exceptional cases, prolonged survival. Outcomes for resection of CCA are limited by the inability to obtain an accurate early diagnosis as many patients present with unresectable lesions. An R0 resection is possible in 70% to 80% of attempted resections and in most series, 5-year survival is approximately 25% to 30% (14–19). Recent resectional approaches for CCA have included techniques such as pre-operative portal vein embolization to induce hypertrophy in the anticipated-remaining hepatic remnant for prevention of liver failure and techniques of vascular reconstruction to expand the potential for R0 resectability. Yokoyama et al. have described an extensive experience with pre-operative portal vein embolization (PVE) in 240 patients, in which selective arterial embolization has been used to improve response to PVE (20). Though the authors report a decrease in operative mortality from 21.9% to most recently (after 2001) 1.6% following PVE, they do not describe the impact of PVE on long-term survival following resection for CCA. A large series from Johns Hopkins also published in 2008 review the outcomes of 564 patients with CCA (50% hilar CCA) and reports on predictors of improved outcome following resection (21). Favorable prognostic factors included patients resected in a more recent time period (after 1995), negative margins, well-differentiated tumors, and negative lymph nodes. For patients with hilar CCA, the median survival was 30 months with 30% 5-year survival. Three other recent single-center series have been reported in the last year and all report similar survival outcomes of 30% to 40% at 5 years and included the use of extended hepatic resection with portal vein reconstruction (22–24). When analysis is restricted to only to those patients with negative resection margins with or without portal vein reconstruction, a 5-year survival of 45% was reported by Hemming et al. (24). Similar improved outcomes with advanced techniques have also been reported by Neuhas et al. (25). All of these data are single-center non-randomized reports with outcomes compared to historical controls.

background
Cholangiocarcinoma (CCA) is a malignancy arising from the bile duct epithelium which has a very poor prognosis. The incidence in the United States is approximately 1.2 in 100,000, though it is far higher in Eastern Europe and Asia, and it appears to be increasing worldwide (1–3). Although CCA can occur anywhere along the intra- or extra-hepatic bile ducts, the most common location is at the hilar region, involving the confluence of the right and left ducts (60%) (4,5). While most patients develop CCA de novo, risk factors for the development of CCA relate to chronic inflammation of the biliary tree. In Western countries CCA is most often associated with Primary Sclerosing Cholangitis (PSC), though other risk factors include choledochal cysts, hepaticolithiasis, and parasitic infections including Clonorchis sinensis and Opisthorchis viverrini. The prevalence of CCA in patients with PSC is 5% to 15% (6). Establishing a timely diagnosis of hilar CCA is challenging. The tumor is a desmoplastic lesion which typically grows along the bile duct rather than in a radial diameter which limits early detection by cross-sectional imaging. Endoscopic brushing and biopsy are often negative even in of the presence well-established disease, and the tropism for bile producing growth in a longitudinal fashion often leads to the lack of a mass lesion, even in the face of biliary obstruction (7). Eventually, tumor growth becomes independent of bile and mass lesions develop, which are demonstrable by cross sectional imaging. Direct cholangiography (endoscopic retrograde cholangiography and percutaneous transhepatic cholangiogram) best characterizes the site, extent, and configuration of ductal CCA. However, the use of imaging modalities, including high-resolution CT and MRI, are also important for the diagnosis of CCA by providing additional information regarding adjacent structures which differentiate CCA from benign counterparts. While contrast-enhanced MRI and magnetic resonance cholangio-pancreatography have improved the diagnostic sensitivity and specificity of imaging, due to the biologic principles of the tumor, early diagnosis remains a significant problem (8–10). A promising technique in obtaining a tissue diagnosis may be the use of Spyglass cholangioscopy, which allows biopsy if bile duct lesions under direct endoscopic visualization. Although this technique may prove to be useful for diagnostic purposes, it will not enhance accuracy of pre-operative tumor staging (11). Thus far, no molecular markers in bile have proven reliable in establishing a more timely diagnosis. The serum marker CA 19–9 has not been shown to be accurate enough for screening. However, for patients with PSC and a malignant appearing stricture, serum CA 19–9 >130 U/mL has a sensitivity of 79% and a specificity of 98% (12). Conventional cytology has a sensitivity of only 20% to 40%, though the specificity is 100% (13). The limitation of conventional cytology is the paucicellular nature of a

neoadjuvant chemoradiotherapy followed by liver transplantation
Liver transplantation (OLT) was initially proposed as the ideal solution to the problem of achieving a complete resection.

229

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Complete hepatectomy followed by transplantation addressed all relevant margins of resection—hepatic, vascular, and ductal—and provided equivalent, if not more complete regional lymphadenectomy. Unfortunately, early experience with OLT was dismal, as recipients demonstrated a high recurrence rate and very poor survival (26–33). Moreover, the initial application of OLT for CCA did not discriminate between intra-hepatic and extra-hepatic origin. Currently OLT alone is contra-indicated in patients with CCA However, based on the observation that radiation provided effective palliation, and in some cases prolonged survival, a neoadjuvant protocol combining External Beam Radiotherapy (EBRT) and brachytherapy with 5-FU and oral capecitabine followed by OLT was started at the Mayo Clinic in 1993 (34). At the same time, a similar protocol was started in Nebraska employing higher dose of brachytherapy without EBRT (35). The Mayo Clinic protocol is restricted to patients with unresectable, non-metastatic hilar CCA. Diagnosis is established by a positive endoscopic biopsy or brushings, in the presence of a mass lesion (<3 cm in radial diameter), polysomy at multiple chromosomes by FISH and CA 19–9 >100, in the presence of a radiographically malignant stricture. Patients with resectable disease undergo resection, but those with bilobar segmental ductal extension which precludes R0 resection; patients with bilobar hilar vascular encasement or those with underlying PSC are considered for the protocol. Patients are staged by cross-sectional imaging and endoscopic ultra-sound prior to enrollment, and are excluded if they have evidence of metastatic disease, have undergone prior attempted resection with violation of the tumor plane or percutaneous (or transluminal biopsy) of the primary lesion or have medical contra-indications precluding OLT. Enrolled patients receive EBRT administered with a target dose of 4500 cGY in 30 fractions delivered over a 3-week period, with 5-FU given in 500 mg/m2 daily bolus for 3 days at the initiation of EBRT (see Fig. 25.1). Two to three weeks after completion of EBRT, a transluminal boost of radiation is delivered using transcatheter iridum192 seeds with target dose of 2000 to 3000 cGy. Patients remain on oral capecitabine while awaiting transplantation. Prior to undergoing transplantation, which is performed either with a deceased donor or living donor allograft, patients undergo formal operative staging (either open or more recently, using hand-assisted laparoscopy) with routine sampling of common hepatic artery and peri-choledochal lymph nodes in addition to biopsy of any other suspicious nodules. Our initial experience with neoadjuvant chemoradiotherapy followed by transplantation was encouraging (36,37) and was followed by a comparison of all 38 patients undergoing OLT for unresectable hilar CCA from 1993 to 2002 versus all patients who underwent resection of hilar CCA at the same institution during the same time period (14). A comparison between liver transplantation with neoadjuvant therapy and resection is imperfect because the OLT protocol was restricted to patients with unresectable disease. Nevertheless, transplantation provided superior outcomes, with 82% 5-year survival versus 21% 5-year survival after resection. In this study, a survival benefit for transplantation versus resection was seen both in patients with PSC and in those with de novo CCA. Our most recent published survival data from December of 2006 includes 65 patients with a mean follow-up of 32 months. Patient survival was 91% at 1 year and 76% at 5 years (38). To date, 167 patients have enrolled in this protocol. There have been 111 patients who have undergone OLT with 1- and 5-year survival of 96% and 72%, respectively (see Figs. 25.2 and 25.3). Though patient survival following the combined protocol for treatment of hilar CCA is similar to other indications for liver transplantation, the toxicity of the neoadjuvant therapy is significant. Common complications include cholangitis, liver abscess, duodenal ulceration, malnutrition, liver decompensation, and disease progression (14,36–38). Approximately 20% of patients have evidence of metastatic disease at operative staging and are therefore not eligible for OLT (see Fig. 25.2). Additionally, despite the aggressive pre-treatment protocol, 11 of 65 (17%) in the most recent published analysis still developed disease recurrence following liver transplantation (38). A retrospective analysis of these 11 patients who developed disease recurrence determined that older patients, those with

Hilar cholangiocarcinoma Mayo clinic protocol Selected patients with unresectable hilar cholangiocarcinoma (began in 1993) Neoadjuvant radiation and chemotherapy External beam radiotherapy with bolus 5-FU Brachytherapy with protracted capcitabine

Cholangiocarcinoma treatment protocol results (September 1, 2008)

167 patients
Irradiation + 5-FU 12 deaths/disease progression 2 transplanted elsewhere 10 receiving neoadjuvant therapy 27 (19%) staged positive 2 awaiting transplantation 2 transplanted elsewhere 1 death 75 decease donor 35 living donor 1 domino donor

Staging operation to rule-out metastases or local extension of tumor precluding complete resection of tumor

143 staging operation

Orthotopic liver transplantation Figure 25.1 Combined chemoradiotherapy followed by liver transplantation protocol for patients with unresectable hilar CCA.

111 liver transplantation

Figure 25.2 Results of a combined chemoradiotherapy followed by liver transplantation protocol for hilar CCA.

230

TRANSPLANTATION FOR HILAR CHOLANGIOCARCINOMA
larger tumors, a prior cholecystectomy, and those with an increased CA 19–9 at transplant (though not at enrollment) to be at an increased risk of recurrence, with a median time to recurrence of 22 months. Those with unfavorable tumor characteristics, such as residual tumor >2 cm in the explanted liver and perineural invasion were also at an increased risk of recurrence. In this series, three patients developed local-regional recurrence while eight had distant metastasis. Recently, Becker et al. analyzed the United Network for Organ Sharing/Organ Procurement and Transplantation Network (UNOS/OPTN) database for patients with CCA who underwent transplant from 1987 to 2005 (39). There were 280 patients who had a diagnosis of CCA either at listing, at discharge, or both, though it is not known whether the patients received neoadjuvant therapy, pre-transplant staging, or any standardized inclusion criteria. Survival for patients who underwent OLT from 1994 to 2005 with a listing diagnosis of CCA (n = 102) experienced a 5-year actuarial survival of 68%. Given these patients underwent OLT in an era when CCA was a known contra-indication to OLT, it is presumed that most of these patients received neoadjuvant therapy, though this assumption is uncertain. Those transplanted in the same era without a pretransplant diagnosis of CCA (presumably incidental CCA and therefore without neoadjuvant therapy, n = 77) remained with a 5-year survival of 20%, which is similar to the historically poor survival associated with OLT for CCA. The conclusion of this multi-center analysis is that transplant for early stage hilar CCA (unresectable or arising in the setting of PSC) is beneficial based on the survival rate of 68% for those with a known diagnosis of CCA, though the study is hindered by the lack of essential details, such as the use of neoadjuvant therapies. With the potential for broader application of the combined neoadjuvant chemo-radiotherapy and OLT protocol, attention must be focused on the recognition and management of anticipated complications following this aggressive treatment. In addition to pre-operative morbidity such as cholangitis and duodenal ulceration, intra-operative challenges attributable to the neoadjuvant therapy include severe inflammatory changes and dense fibrosis leading to difficulty in identifying and
Patient survival after transplantation 1993–2008 96 ± 2% n = 111 83 ± 4% 72 ± 6%

separating hilar structures. Late effects of tissue injury from radiation therapy, which may be broadly considered a result of fibrosis and chronic ischemic injury, are also of concern. A recent retrospective analysis of 68 patients who underwent neoadjuvant chemoradiotherapy followed by OLT showed increased incidence of late vascular strictures (arterial and portal vein) when compared to controls undergoing OLT for other indications (40). The late vascular complication rate was 40%, though patient and graft survival did not differ between the groups. In most patients, these complications were manageable with a percutaneous endovascular approach. The key findings from this analysis are that arterial interposition grafts should be used to replace the irradiated native artery in all cases of deceased donor transplantation. Additionally, late strictures of the portal vein (and non-replaced arteries as in the case of a living donor transplant) should be anticipated and may be amenable to endovascular interventions.

organ allocation
The findings from the Mayo Clinic and Nebraska protocols as well as the historical data for OLT without adjuvant therapy were carefully considered by the Model for End Stage Liver Disease (MELD) Exception Study Group, which was assembled to consider standardization of diagnoses which commonly led to MELD exception scores (41). The conclusion of this group was to recommend national standardization of MELD score exceptions for patients with early stage, unresectable hilar CCA provided they are enrolled in an approved protocol which includes neoadjuvant therapy, and standard inclusion (i.e., diagnostic), and exclusion criteria. The group proposed an allocation policy applied nationally similar to that currently granted for HCC, which is based on a 10% risk of mortality subject to readjustment every three months. Currently, organ allocation in the United States for hilar CCA, as well as other diagnoses which require MELD exception, varies considerably by each region. It is possible that a national policy for MELD exceptions for common diagnoses may be considered in the future.

conclusions
Excellent outcomes can be achieved with neoadjuvant chemoradiotherapy followed by OLT for selected patients with early stage hilar CCA. Obtaining a timely diagnosis remains a key challenge regardless of whether resection or transplantation is being considered. As with hepatocellular carcinoma, it is necessary to determine the risk for disease progression following neoadjuvant therapy for patients awaiting transplantation. Given the severe donor organ shortage, continued analysis of outcomes in order to identify patients not likely to gain benefit from this aggressive protocol is necessary. In light of the results achievable with the combined neoadjuvant chemoradiotherapy protocol, however, a standardized approach to organ allocation which is applied nationally, as for HCC, is necessary and will hopefully be forthcoming. Finally, with application of the protocol across multiple centers, the key principles, such as multi-disciplinary approach, pre-transplant staging to ensure inclusion of only those without metastasis, replacement of irradiated vessels when possible as well as monitoring for postoperative vascular complications, must be highlighted.

100 90 80 70 60 50 40 30 20 10 0 0

1

2 3 Years after transplantation

4

5

Figure 25.3 Kaplan–Meier analysis of survival for patients undergoing transplantation after completion of neoadjuvant therapy at Mayo Clinic.

231

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

key points
1. The diagnosis of hilar CCA remains a challenge (level II evidence). 2. Based on non-randomized analysis (level II evidence), neoadjuvant chemoradiotherapy followed by liver transplantation for patients with early stage unresectable hilar CCA offers excellent survival. 3. Percutaneous, open, or transgastric biopsy of the primary tumor should not be performed due to the risk of tumor seeding into the peritoneal cavity (level II). 4. Radiation injury may lead to hepatic artery thrombosis and late vascular strictures. Native hepatic arteries should be replaced by arterial conduits in deceased donor transplantation to reduce early thrombosis. Late strictures may be amenable to endovascular treatment (level III). 5. Toxicity of the neoadjuvant therapy is significant, and a multi-disciplinary approach including commitment from radiology, radiation oncology, medical oncology, hepatology, and hepatobiliary/ transplantation surgery is necessary.

references
1. Shaib Y, el-Serag HB. The epidemiology of cholangiocarcinoma. Semin Liver Dis 2004; 24: 115–25. 2. Sripa B, Pairojkul C. Cholangiocarcinoma: lessons from Thailand. Curr Opin Gastro 2008; 24: 349–56. 3. Maggs JR, Chapman RW. An update on primary sclerosing cholangitis. Curr Opin Gastro 2008, 24: 377–83. 4. Bismuth H, Castaing D. Hepatobiliary malignancy. London: Edward Arnold, 1994. 5. De Groen PC, Gores GJ, LaRusso NF, et al. Biliary tract cancers. N Engl J Med 1999; 341: 1368–79. 6. Knan Sa, Davidson BR, Goldin R, et al. Guidelines for the diagnosis and treatment of cholangiocarcinoma: consensus document. Gut 2002; 51 (Suppl 6): VI 1–9. 7. Gores GJ, Nagorney DM, Rosen CB. Cholangiocarcinoma: is transplantation an option? For whom? J Hepatol 2007; 47: 454–75. 8. Manfredi R, Barbaro B, Masselli G, et al. Magnetic resonance imaging of cholangiocarcinoma. Semin Liv Dis 2004; 24(2): 155–64. 9. Braga HJ, Imam K, Bluemke DA. MR imaging of intrahepatic cholangiocarcinoma: use of ferumoxides for lesion localization and extension. AJR 2001; 177: 111–14. 10. Yeh TS, Jan YY, Tseng JH, et al. Malignant perihilar biliary obstruction: magnetic resonance cholangiopancreaticographic findings. Am J Gastro 2000; 95: 432–40. 11. Chen YK, Pleaskow DK. SpyGlass single-operator peroral cholangiopancreatoscopy system for the diagnosis and therapy of bile-duct disorder: a clinical feasibility study. Gastrointest Endosc 2007; 65(6): 832–41. 12. Levy C, Lymp J, Angulo P, et al. The value of serum Ca 19-9 in predicting cholangiocarcinoma in patients with primary sclerosing cholangitis. Dig Dis Sci 2005; 50: 1734–40. 13. Moreno Luna LE, Kipp B, Halling et al. Advanced cytologic techniques for the detection of malignant pancreatobiliary strictures. Gastroenterology 2006; 131: 1064–72. 14. Rea D, Heimbach JK, Rosen CB, et al. Liver transplantation with neoadjuvant chemoradiation is more effective than resection for hilar cholangiocarcinoma. Ann Surg 2005; 3: 451–61. 15. Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg 1996; 224: 463–73; discussion 473–5.

16. Pichlmayr R, Weimann A, Klempnauer J, et al. Surgical treatment in proximal bile duct cancer. A single-center experience. Ann Surg 1996; 224: 628–38. 17. Kosuge T, Yamamoto J, Shimada K, et al. Improved surgical results for hilar cholangiocarcinoma with procedures including major hepatic resection. Ann Surg 1999; 230: 663–71. 18. Washburn WK, Lewis WD, Jenkins RL. Aggressive surgical resection for cholangiocarcinoma. Arch Surg 2001; 234: 270–6. 19. Jarnagin WR, Fong Y, DeMatteo RP, et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma. Ann Surg 2001; 234: 507–17; discussion 517–9. 20. Yokoyama Y, Nagina M, Nishio H, et al. Recent advances in the treatment of hilar cholangiocarcinoma: portal vein embolization. J Hepatobiliary Pancreat Surg 2007; 14: 447–54. 21. DeOliveira ML, Cunningham SC, Cameron JL, et al. Cholangiocarcinoma thirty-one-year experience with 564 patients at a single institution. Ann Surg 2007; 5: 755–62. 22. Miyazaki M, Kato A, Ito H, et al. Combined vascular resection in operative resection for hilar cholangiocarcinoma: does it work or not? Surgery 2007: 5: 581–8. 23. Hasegawa S, Ikai I, Fujii H, et al. Surgical resection of hilar cholangiocarcinoma: analysis of survival and postoperative complications. World J Surg 2007; 31: 1256–63. 24. Hemming AW, Kim RD, Mekeel KL, et al. Portal vein resection for hilar cholangiocarcinoma. Am Surg 2006; 72: 599–605. 25. Neuhaus P, Joans S. Surgery for hilar cholangiocarcinoma—the German experience. J Hepatobiliary Pancreat Surg 2000; 7(2):142–7. 26. Goldstein RM, Stone M, Tillery GW, et al. Is liver transplantation indicated for cholangiocarcinoma? Am J Surg 1993; 166: 768–71; discussion 771–2. 27. Jonas S, Kling N, Guckelberger O, et al. Orthotopic liver transplantation after extended bile duct resection as treatment of hilar cholangiocarcinoma. First long-terms results. Transpl Int 1998; 11 (Suppl 1): S206–8. 28. Loinaz C, Abradelo M, Gomez R, et al. Liver transplantation and incidental primary liver tumors. Transplant Proc 1998; 30: 3301–2. 29. Iwatsuki S, Todo S, Marsh JW, et al. Treatment of hilar cholangiocarcinoma (Klatskin Tumors) with hepatic resection or transplantation. J Am Coll Surg 1998;187: 358–64. 30. Meyer CG, Penn I, James L. Liver transplantation for cholangiocarcinoma: results in 207 patients. Transplantation 2000; 69: 1633–7. 31. Shimoda M, Farmer DG, Colquhoun SD, et al. Liver transplantation for cholangiocellular carcinoma: analysis of a single-center experience and review of the literature. Liver Transpl 2001; 12: 1023–33. 32. Robles R, Figueras J, Turrion VA, et al. Liver transplantation for hilar cholangiocarcinoma: Spanish experience. Transpl Proceedings 2003; 35: 1821–2. 33. Ghali P, Marotta PJ, Yoshida EM, et al. Liver transplantation for incidental cholangiocarcinoma: analysis of the Canadian experience. Liver Transpl 2005; 11: 1412–6. 34. De Vreede I, Steers JL, Burch PA, et al. Prolonged disease-free survival after orthotopic liver transplantation plus adjuvant chemoirradiation for cholangiocarcinoma. Liver Transpl 2000; 6: 309–16. 35. Sudan D, DeRoover A, Chinnakotia S, et al. Radiochemotherapy and transplantation allow long-term survival for nonresectable hilar cholangiocarcinoma. Am J Transpl 2002; 2: 774–9. 36. Hasson Z, Gores GJ, Rosen CB, et al. Preliminary experience with liver transplantation in selected patients with unresectable hilar cholangiocarcinoma. Surg Oncol Clin N Am 2002; 11: 909–21. 37. Heimbach JK, Gores GJ, Haddock MG, et al. Liver transplantation for unresectable perihilar cholangiocarcinoma. Semin Liver Dis 2004; 24(2): 201–7. 38. Heimbach JK, Gores GJ, Haddock MG, et al. Predictors of disease recurrence following neoadjuvant chemoradiotherapy and liver transplantation for unresectable perihilar cholangiocarcinoma. Transplantation 2006; 12: 1703–7. 39. Becker NS, Rodriguez JA, Barshes NR, et al. Outcomes analysis for 280 patients with cholangiocarcinoma treated with liver transplantation over an 18-year period. J Gastrointest Surg 2008; 12(1): 117–22. 40. Mantel HTJ, Rosen CB, Heimbach JK, et al. Vascular complications after orthotopic liver transplantation after neoadjuvant therapy for hilar cholangiocarcinoma. Liver Transpl 2007; 12: 1372–81. 41. Gores GJ, Gish RG, Sudan D, et al. MELD Exception Study Group. Model for end-stage liver disease (MELD) exception for cholangiocarcinoma or biliary dysplasia. Liver Transpl 2006; 12: S95–7.

232

26 Rare vascular liver tumors
introduction

Jan P. Lerut, Eliano Bonaccorsi-Riani, Giuseppe Orlando, Vincent Karam, René Adam, and the ELITA–ELTR Registry a
with intracellular vascular lumens containing red blood cells (Fig. 26.2A and B). In contrast to HAS, the hepatic acinar landmarks are better preserved. HEHE originates from endothelial cells which explains positive immunohistochemistry (IHC) for factor VIII-related antigen and for the endothelial markers CD31 and CD34. Ultrastructural examination also shows characteristic features of endothelial cells, such as a basal lamina, pinocytic vesicles, and Weibel–Palade bodies (2–4). The clinical manifestations of HEHE are unspecific, varying from totally asymptomatic to hepatic failure (Table 26.1). The malignant potential of HEHE often remains unclear in each individual patient. Based on the analysis of the ELITA–ELTR study, the most frequent symptoms are non-specific upper abdominal or epigastric discomfort or pain, weakness, general malaise, and jaundice. About 20% of patients are asymptomatic and 10% present with pulmonary symptoms (Table 26.1). Hepato-splenomegaly (50%) and weight loss (20%) are the most frequent clinical signs; portal hypertension may be caused by tumor compression or venous infiltration (3–6). Anicteric cholestasis and cytolytic activity are frequently present (60% and 40%, respectively). Serum tumor markers are normal in the absence of accompanying liver disease. Radiological investigation identifies two different patterns of HEHE: the early peripheral and nodular, usually bilobar type (peripheral pattern), and the later confluent type (diffuse pattern) with eventual invasion of the greater vessels (Fig. 26.3A). Focal calcifications are present in 20% of tumors. Angiography reveals only moderate vascularization with displacement of marginal vessels (Fig. 26.3B and C). Scintigraphic and FDG–PET imaging plays an important role in the staging of the disease, especially in view of later LT and also in view of the early detection of recurrent disease (7–8). Complete assessment of these patients is mandatory to exclude other, especially thoracic, disease localization. In doubtful cases thorascopic lung and/or pleural biopsy should be performed in order to exclude the often frequent thoracic involvement. Definitive diagnosis, frequently based on a high degree of suspicion, can be made by associating radiological and clinical features, such as occurrence in a young adult (usually female), the contrast between numerous intrahepatic (often calcified) tumors, the overall condition of the patient, and finally the longstanding clinical history. The diagnosis can only be confirmed by pathological examination of appropriate, surgically obtained biopsy material. The treatment algorithm of HEHE was till recently far from standardization. The literature review, including 13 very small

Vascular hepatic tumors form a continuum going from the completely benign hemangioma (HA) to the very aggressive hemangiosarcoma (HAS). Sometimes due to their difficult differential diagnosis and the rarity of the more malignant varieties, and also to the limited worldwide experience with liver resection and transplantation (LT) for these tumors, these vascular lesions are often neglected or misdiagnosed and their therapeutic management algorithm is unclear. The rarity of malignant vascular liver lesions is well demonstrated by the data of the European Liver Transplant Registry (ELTR). During the period January 1988 to June 2007, 12% (8278) of all European LT were performed because of hepatobiliary tumors, but of these, only 125 (0.1%) were done because of malignant vascular diseases. In this chapter hepatic epithelioid hemangioendothelioma (HEHE), infantile hemangioendothelioma (HIHE), and hemangiosarcoma (HAS) will be discussed, based on a recent literature review and of the experience gathered by the audited ELITA–ELTR during the period 1988 to 2004 (1). The information obtained from both sources allows a more clear insight into diagnosis and treatment of these rare liver diseases. For the sake of completeness, benign Nodular Regenerative Hyperplasia (NRH) is also included in this chapter because of its vascular etiopathogenesis. Hemangioma and lymphangioma are discussed in a separate chapter dealing with benign liver tumors.

hepatic epithelioid hemangioendothelioma ₍hehe₎
HEHE is a rare (<1 per million population), low-grade malignancy which has a malignant behavior potential between HA and HAS (2). This vascular tumor was first recognized in 1982 by Weiss and Enzinger in soft tissues, later on in the head and neck region, then in bones and finally in many other organs, such as lungs and bronchi. Liver involvement occurs most often as a primary tumor. HEHE is more frequent in young adult women; with a peak incidence during the fourth decade. This tumor has been rarely reported in children less than 15 years old. No definitive etiological factor has been clearly identified (2). Macroscopically, HEHE appears as multifocal fibrous masses (Fig. 26.1A and B). Microscopically, HEHE is characterized by pleiomorphic cells, both of medium and of large size, that are epithelioid in appearance and that spread within sinusoids and small veins at the periphery of the lesion, whereas the center is fibrous and studded with elongated tumor cells, sometimes
a

The European Liver Transplant Registry is a service of the European Liver and Intestinal Transplant Association (ELITA); see www.eltr.org.

233

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

Figure 26.1 Intraoperative view of 7.6 kg heavy HEHE responsible for IVC syndrome and supine dyspnea (A). Characteristic gross appearance of a HEHE hepatectomy specimen showing multiple fibrous masses with zoning phenomenon (B).

(A)

(B)

Figure 26.2 Histological features of HEHE at the periphery (A) and in the center (B) of the lesion. Source: C. Sempoux, Dept. of Pathology.

Table 26.1 Clinical presentation of HEHE in the ELITA–ELTR series
Age Gender Symptoms Upper abdominal pain/ discomfort Weakness/fatigue Asymptomatic Anorexia Nausea Dyspnea Signs Hepatomegaly Weight loss Jaundice Ascites Hemangioma Portal hypertension Hepatopulmonary syndrome 39 ± 15 years (range 0.4–65) – 2 children 43 F and 16 M 59% 20% 19% 15% 14% 10% 49% 20% 10% 9% 9% 7% 2%

(≥5 pts) series and three reviews, lacks detailed data analysis, and most importantly, long-term follow-up. The published experience does not allow comparison of results of untreated and of surgically and medically treated patients. The place of LT has especially been questioned in view of well-documented spontaneous resolution, long-term survival, the high incidence of extra-hepatic disease (up to 45%) (Table 26.2), the lack of predictive clinical or histological criteria, and finally the high incidence (up to 33%) of, sometimes even very late, recurrent allograft disease (6). The largest institutional series, coming from Pittsburgh and containing 16 patients, showed that LT offers 5-year overall and disease-free survival rates of 71% and 60%, respectively (9). The Mehrabi review indicates that the treatment of choice of HEHE is total hepatic resection (3). In this literature review, the 5-year survival rates of LT, partial liver resection (reported in only very few patients), local or systemic chemo- and radiotherapy (CHTH) and treatment abstention were 55%, 75%, 30%, and 0%, respectively. Partial liver resection is not a logical treatment because HEHE is nearly always a multinodular and bilobar disease. Indeed the examination of the total hepatectomy specimens analyzed in the ELITA–ELTR study showed

234

RARE VASCULAR LIVER TUMORS

(A)

(B)

(C)

Figure 26.3 CT-scan of HEHE showing the initial peripheral nodular (A) and the later confluent patterns of the disease (B). Angiography confirming the avascular nature and vascular displacement (C).

Table 26.2 Histological presentation of HEHE in the ELITA–ELTR series
Bilobar involvement Number nodules >15 Microvascular involvement Macrovascular involvement Hilar lymph node positivity 96% 86% 44.4% 9.4% 33% (18/54p) 4/18 pts (22%) patients only positive on IHC 4/10 pts with extrahepatic EHE localization were lymph node positive 17% (10 pts) 4 1 1 1 1 1

Extrahepatic localization Lung Mediastinum Peritoneum Spleen Femoral vein Brain and bone Omentum

bilobar disease in 96% of cases; 86% of the specimen contained more than 15 tumor nodules (Table 26.2). Moreover, several reports documented disease recurrence, or LT because of recurrent HEHE after partial resection; fortunately, pre-transplant liver surgery doesn’t seem to impair survival after LT (6). The assessment of non-surgical treatments, such as radiotherapy, local tumor destruction, hormonotherapy, systemic or locoregional CHTH, trans-arterial embolization, and chemoembolization, is even more difficult because of the lack of uniform treatment modalities and of long-term follow-up (3). The ELITA–ELTR study is the largest worldwide, detailed long-term study in relation to HEHE (1). This study collected 59 European patients from 32 centers. The follow-up from diagnosis is 104 ± 72 months (median of 93 months) and from LT 83 ± 55 months (median of 79 months). The study validated the place of LT in the treatment of this disease and 5 and 10 years post-transplant survival rates are 83% and 74%, respectively, with 5 and 10 years recurrence-free survival rates of 82% and 64%, respectively (Fig. 26.4A and B). This study confirms that medical and/or surgical pre-LT treatment (present in 30% of pts), invasion of regional lymph node (present in 33% of pts) and of (limited) extrahepatic disease (present in 17% of pts) are not formal contraindications to LT. Combined micro- and macrovascular invasion (present in

49% of pts) is the only parameter which significantly influences outcome after LT. Mitotic index and cellular pleiomorphism are other major histological prognostic criteria reflecting more aggressive tumor behavior (10). Their influence on patient survival could not be analyzed in the ELITA– ELTR study due to the impossibility of obtaining central pathology recording. The recently published UNOS data reporting 128 patients with a median follow-up of 24 months go in the same direction; 1- and 5-year patient survival rates were 80% and 64%, respectively (11). Recurrent disease after (partial and total) hepatectomy should be treated aggressively as prolonged, sometimes even disease-free, survival can be achieved (1,3). The role of re-LT in the treatment of recurrent allograft disease is unclear as only one single case has been reported (5). It is conceivable that in view of the high incidences of extrahepatic disease localization, and of recurrence in- and outside the allograft (22% in the ELITA–ELTR study) (Table 26.3), a more radical approach combining total hepatectomy and antiangiogenic therapy, such as use of VEGF-antibodies, α-interferon and rapamycin could be of value in the treatment of HEHE (1,12). Rapamycin is of particular interest in this context as this drug is nowadays frequently used not only as a renal sparing but also as an anti-tumoral immunosuppressive drug in liver transplant recipients. Such an approach, combined with a better study of the molecular and genetic markers based on tumor biology, will be of great assistance in further improving outcome, monitoring efficacy of emerging neo- and adjuvant treatments and in recognizing the aggressive subtypes of HEHE (3,5,13–15).

hepatic infantile hemangioendothelioma ₍hihe₎
HIHE, the most common mesenchymal tumor of the liver in infants (<3 years), is always diagnosed nearly during the first 6 months of life. The tumor is also more frequent in females and presents with (a)symptomatic hepato-splenomegaly, failure to thrive, congestive cardiac failure (15%) due to intratumoral arteriovenous shunting, cutaneous hemangiomas (20–40%), consumptive coagulopathy due to Kasabac–Merritt syndrome, and medically resistant hypothyroidism (2,16). In the majority of the cases, HIHE appears to be a histological benign tumor that may regress spontaneously in 5% to 10% of cases; however its outcome is usually much poorer because of its severe complications (2,16–18).

235

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
HIHE can occur either as a single tumor (Fig. 26.5A) or it can be multifocal and diffuse (Fig. 26.5B). The lesions are soft, spongy, red-tan nodules, sometimes more firm and greywhite or brown with areas of necrosis and hemorrhage. Histology identifies two types of HIHE. Type 1 the most frequent is made of intercommunicating small vascular channels lined by a single layer of regular endothelial cells (Fig. 26.5C). There may be areas of cavernous changes and extramedullary hematopoiesis is often found. Type 2 HIHE, a more pleiomorphic HIHE with nuclear atypia, multi-layering, and papillary projections (Fig. 26.5D), is now considered as a form of HAS. IHC examination is of importance in relation to the differentiation of HIHE from hepatic vascular malformations associated with capillary proliferation; immunoreactivity for GLUT-1 can be of help for the diagnosis of HIHE (19). The difficulty with these tumors lies in the fact that HIHE may harbor foci of HAS. Data from the ELITA–ELTR showed that children with HIHE present with a rapid deterioration of their condition or presenting with acute liver failure and/or Budd–Chiari syndrome, indeed also presented with (foci of) HAS (see further). HEHE can be differentiated from HIHE as both have different, age-related, clinical, and pathological features. The natural history of HIHE is variable, but up to two thirds of symptomatic patients die of tumor complications, for example, cardiac failure. Treatment modalities of HIHE vary from medical

Patient survival after liver transplantation for epithelioid hemangioendothelioma: 59 patients 100 90 80 70 60 50 40 30 20 10 0 0 1 2 3 yrs 85% 3 4 yrs 83% 4 5 yrs 83% 5 6 yrs 80% 7 yrs 75% 6 8 yrs 72% 7 9 yrs 72% 8 10 yrs 72% 9 10 Survival % 2 yrs 1 yr 86% 93%

Number of exposed patients Total 1 yrs 2 yrs 3 yrs 59 55 51 46

4 yrs 41

5 yrs 36

6 yrs 31

7 yrs 26

8 yrs 18

9 yrs 14

10 yrs 13

Disease free survival after liver transplantation for epithelioid hemangioendothelioma: 52 patients 100 90 80 70 60 50 40 30 20 10 0 0 1 2 3 yrs 85% 3 4 yrs 85% 4 5 yrs 82% 4 yrs 36 5 6 yrs 79% 5 yrs 31 7 yrs 76% 6 yrs 27 6 8 yrs 68% 7 yrs 23 7 9 yrs 64% 8 yrs 16 8 10 yrs 64% 9 yrs 11 10 yrs 11 9 10 Survival % 1 yr 2 yrs 90% 87%

Number of exposed patients Total 1 yr 2 yrs 3 yrs 52 47 45 40

Figure 26.4 Overall patient survival of HEHE after liver transplantation in the ELITA-ELTR study (A). Recurrent-free survival in the same patient cohort of 59 patients (B).

236

RARE VASCULAR LIVER TUMORS
anti-angiogenic therapies using or combining high dose steroids, interferon, CHTH or radiotherapy, other interventional radiological or surgical therapies (2,16–18). Recently the Boston group presented a clearly defined algorithm for the treatment of this paediatric condition. Partial hepatectomy is indicated if lesions are confined to one liver lobe; diffuse HIHE disease, resistant to steroid therapy, should be treated using LT (20). all primary liver tumors. Peak age of incidence is in the sixth and seventh decades of life, with a male to female ratio of 3/1. HAS has rarely been reported in children and, if so, is regarded as a distinct entity (2,21). HAS has received much recent attention because of its frequent association with several environmental carcinogens, such as thorotrast, vinyl chloride, monomer, radium, pesticides, external radiation, cyclophosphamide, arsenical compounds, use of androgenic/anabolic steroids and iron overload, such as seen in hemochromatosis (2). A clear etiological cause of HAS is however not found in 70% of patients. The diagnosis, especially of diffuse HAS, can be extremely difficult. Liver biopsy has been reported as dangerous and non-diagnostic (22). Macroscopically, the HAS appears as illdefined spongic hemorrhagic nodule(s) that involve(s) the whole liver. HAS can present four different types of growth patterns: multiple nodular, large dominant mass, mixed patterns of multiple nodules and dominant mass, and more rarely diffusely infiltrating macro-nodular tumor (Fig. 26.6). At the time of diagnosis, extra-hepatic metastases are present in 20% to 40% of patients; the most common sites of metastases are lung, spleen, bone, and adrenals (2,21). The most characteristic histological pattern of HAS is a sinusoidal and tectorial growth of malignant endothelial spindle cells on the surface of liver cell plates leading to their atrophy, and associated with formation of larger vascular channels and cavernous spaces with papillary projections into their lumen (Fig. 26.7A–C). Tumoral cells have bizarre and

hepatic hemangiosarcoma ₍has₎
Although relatively rare, HAS is the most common primary sarcoma occurring in the liver. It may account for up to 2% of

Table 26.3 Recurrence of HEHE after liver transplantation in the ELITA–ELTR series
Tumor recurrence in 14 (23.7%) patients: 5 patients still alive Delay posttransplantation Lung 45 ± 35 mo 5 49 mo (range 6–98) AWD 59 (ChTh), 84 (Res) mo DWD 23,41,73 mo AWD 112 mo (Res, RF, ChTh) 211 and 212 mo DWD 15,73 (Res, ChTh), 79 mo (Burkitt) DWD 9 mo DWD 4 mo (ChTh) DWD 19, 20 mo

Liver

5

Liver and omentum Liver and bone Not precised

1 1 2

(A)

(B)

(C)

(D)

Figure 26.5 Macro- and microscopic features of HIHE. The lesion can be either solitary (A) or multiple (B). Type 1 HIHE is made of multiple vascular channels with a single layer of regular endothelial cells (C) whereas in type 2 HIHE more atypia are observed (D). Source: S. Gosseye, Dept. of Pathology.

237

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
hyperchromatic nuclei with numerous mitoses and they exhibit IHC positivity for endothelial markers (Factor VIII, CD31, and CD 34). HAS differs from HEHE by the fact that it has more atypical cells and it is much more destructive with obliteration of the normal liver (21). Histological diagnosis on adequate tissue sampling is of utmost importance in order to avoid futile, especially surgical interventions. Although the diagnosis of HAS is nowadays easier following the introduction of nuclear magnetic imaging, the diagnosis still frequently remains difficult. This difficulty is exemplified in the ELITA–ELTR study collecting 22 patients of whom 6 were children. Correct pre-LT histological diagnosis of HAS was only made in one-third of 16 biopsied patients, and half of 22 patients were transplanted because of HIHE or HEHE. At the time of LT, 15% of patients already had distant metastases All four HAS presenting as acute Budd–Chiari syndrome were seen in children with a supposed diagnosis of HIHE (Table 26.4). Pathologic examination of the hepatectomy specimen showed massive and diffuse bilobar involvement in all cases. Biochemical abnormalities in HAS are non-specific; serum alkaline phosphatase is elevated in 70% of patients and tumor markers are negative in the absence of accompanying liver disease. In the early stages HAS may present with hepatosplenomegaly, ascites, jaundice, signs of portal hypertension, weight loss, and muscle wasting; later on, pain, peripheral edema, acute Budd–Chiari syndrome, acute abdomen due to tumor rupture, and thrombocytopenia may follow. In contrast to HEHE, evolution of these patients is much more rapid and worse. All patients in the ELITA–ELTR study died due to tumor recurrence after a median of 6 months (Fig. 26.8). Similarly, very poor results were reported recently by the Cincinnati transplant tumor registry in six patients and by the Memorial Sloan-Kettering group in five patients (23,24). Embryonal sarcoma is the only liver sarcoma that can nowadays be treated successfully by (partial or more seldom necessary) hepatectomy (23). Both European and American experiences confirm that HAS is an absolute contra-indication to LT. In cases of difficult differential diagnosis between HIHE/HEHE and HAS, it is wise to respect a waiting period of 6 months on the transplant list in order to avoid wasting scarce organ resources. Indeed the aggressive evolution of the liver tumor during this time span will allow the establishment of the correct diagnosis. The outcome for patients with HAS will only be improved by the development of a more effective interdisciplinary oncological approach (24).

Table 26.4 Indication for LT in 22 HAS patients of the ELITA–ELTR series
Age Gender Invalidating tumor (9 HEHE–6 HAS) Unclear diagnosis of liver failure (22%) Cryptogenic cirrhosis Focal nodular hyperplasia Hemochromatosis Subacute failure Budd–Chiari syndromea Solitary hepatocellular cancer Budd–Chiari syndrome,b HEHE HAS HAS
a b

35.3 ± 22.6 years Children 4.4 years (1–12) 14 males and 8 females Six children <15 years 15 5 1 1 1 1 1 1 3 2 1 1

Figure 26.6 CT-scan of HAS showing a diffuse nodular infiltration of the liver.

Budd–Chiari syndrome: 18%. All children.

(A)

(B)

(C)

Figure 26.7 HAS. Macroscopy showing a diffuse spongy tumoral mass (A); microscopy showing the development of cavitary space (B) because of the sinusoidal progressive growing of tumoral cells destroying the liver cell plates (C). Source: J. Rahier, Dept. of Pathology.

238

RARE VASCULAR LIVER TUMORS

nodular regenerative hyperplasia ₍nrh₎
NRH of the liver is a rare condition that is most often presents with the stigmata of portal hypertension and its sequelae (25). Its prevalence is estimated between 0.70% and 2.6% in autopsy series. NRH is the major cause of non-cirrhotic portal hypertension. Current theories favor a primary vasculopathy with alterations in blood flow as the causative agent. It is supposed that the presence of a portal venopathy (occlusion or terminal portal venules) with or without accompanying hepatic arterial disorders causes diffuse microvascular transformation, finally leading to NRH. Vasculopathy results in ischemia leading to atrophy with compensatory hyperplasia in adjacent unaffected areas. When a critical proportion of the portal venules is affected, portal hypertension ensues. NRH is characterized by the formation of usually small, up to 1 cm, non-fibrotic parenchymal nodules. This absence of fibrosis allows the establishment of the differential diagnosis with micronodular cirrhosis. Histologically, the regenerative nodules are composed by large hyperplastic hepatocytes which compress the adjacent liver cell plates that then become atrophic (Fig. 26.9A), a feature nicely underlined by reticulin stain (Fig. 26.9B). Hemodynamic

Survival function 1.0 0.8 Cum survival 0.6 0.4 0.2 0.0 0 5 10 15 Postoperative months 20 25

investigation shows pre-sinusoidal portal hypertension in association with a patent portal vein. Various systemic diseases are known to occur in association with NRH, such as primary biliary cirrhosis, Budd–Chiari syndrome, hemorraghic hereditary teleangiectasia or Rendu– Osler–Weber disease, lympho- and myeloproliferative disorders, collagen-vascular diseases, congenital and acquired hepatic macrovascular abnormalities, as well as exposure to toxins, such as azathioprine and some chemotherapeutic agents. Familial cases of NRH occurring without underlying or associated systemic disease have been described (26). These forms have a poor clinical course and are often associated with progressive renal failure. In relation to LT, it should be noted that recurrent allograft NRH has been reported, that NRH has been reported as a complication of chronic immunosuppression using azathioprine, leading even to re-LT, and NRH has been described in the context of living donor liver transplantation using smallfor-size grafts. Also, NRH and obliterative portal venopathy have been documented in recipients transplanted for biliary tract disease complicated with severe non-cirrhotic portal hypertension (27–29). Six patients with NRH have been reported to the ELTR. Four had favorable outcomes; two died in 20 and 42 months, respectively, post-transplantation due to cardiac failure. Three patients had to be treated prior to transplantation because of bleeding esophageal varices; three presented with cholestatic cirrhosis. One patient had a previous kidney transplant. Four larger series and some case reports describing successful LT for this condition have been reported in literature (26,30).

conclusion
Due to their scarcity, difficulties in diagnosis, as well as lack of consensus around treatment of vascular tumors of the liver, there is no universally accepted standard of care for these patients. The recent review of the literature and of the audited European Liver Transplant Registry data have permitted some clarification of the therapeutic algorithms for these lesions. One should always have a high degree of suspicion in relation to these lesions which may have an extremely divergent clinical

Figure 26.8 Overall patient survival of HAS after liver transplantation in the ELITA–ELTR study.

(A)

(B)

Figure 26.9 Microscopic features of NRH on a HE section (A) and reticulin stain (B). Source: C. Sempoux, Dept. of Pathology.

239

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
course. Aggressive diagnostic work-up, using adequate tissue sampling is mandatory especially in order to differentiate the potentially curable HEHE or HIHE from the incurable HAS. HEHE should be treated aggressively by LT. Indeed these lesions represent an excellent indication for LT allowing longterm (disease-free) survival even in the presence of (limited) extrahepatic disease localization at the time of transplantation. Combining LT and anti-angiogenic drugs during the post-transplant follow-up could be of interest to further improve the outcomes. HIHE has a highly variable disease outcome ranging from spontaneous disease regressions to fatal outcomes. Diffuse HIHE resistant to steroid therapy should be treated using LT. The appearance of acute liver failure and/or Budd–Chiari syndrome may be an indicator toward the development of HAS. The outlook for HAS remains extremely dismal to whatever therapy is applied. In HAS, LT is absolutely contra-indicated due to the very rapid and aggressive disease recurrence. In case of unclear diagnosis, futile liver transplantation can be avoided placing the patient on the waiting list for a minimum period of 6 months. This time span confers two advantages; first, it does not interfere with the outcome of liver transplantation in case of HIHE or HEHE, and second, it allows observation of the natural, always fatal, evolution of HAS. Severe, symptomatic, non-cirrhotic portal hypertension caused by nodular regenerative hyperplasia can also be cured by liver transplantation if the therapeutic armamentarium fails to control its severe complications.

references
1. Lerut JP, Orlando G, Adam R, et al. The place of liver transplantation in the treatment of hepatic epitheloid hemangioendothelioma: report of the European liver transplant registry. Ann Surg 2007; 246: 949–57. 2. Ishak K, Goodman Z, Stocker J. Tumors of the liver and intrahepatic bile ducts. Atlas of Tumor Pathology. Third series, fascicle 31 edn. Washington: Armed Forces Institute of Pathology, 1999: 282–307. 3. Mehrabi A, Kashfi A, Fonouni H, et al. Primary malignant hepatic epithelioid hemangioendothelioma: a comprehensive review of the literature with emphasis on the surgical therapy. Cancer 2006; 107: 2108–21. 4. Makhlouf HR, Ishak KG, Goodman ZD. Epithelioid hemangioendothelioma of the liver: a clinicopathologic study of 137 cases. Cancer 1999; 85: 562–82. 5. Demetris AJ, Minervini M, Raikow RB, et al. Hepatic epithelioid haemangioendothelioma – biological questions based on pattern of recurrence in an allograft and tumor immunophenotype. Am J Surg Pathol 1997; 21: 263–70. 6. Lerut JP, Orlando G, Sempoux C, et al. Hepatic haemangioendothelioma in adults: excellent outcome following liver transplantation. Transpl Int 2004; 17: 202–07. 7. Nguyen BD. Epithelioid haemangioendothelioma of the liver with F-18 FDG PET imaging. Clin Nucl Med 2004; 29: 828–30. 8. Lin E, Agoff N. Recurrent hepatic epithelioid hemangioendothelioma: detection by FDG PET/CT. Clin Nucl Med. 2007; 32: 949–51. 9. Madariaga JR, Marino IR, Karavia DD, et al. Long-term results after liver transplantation for primary hepatic epithelioid haemangioendothelioma. Ann Surg Oncol 1995; 2: 483–87. 10. Kelleher MB, Iwatsuki S, Sheahan Dg. Epitheloid haemangioendothelioma of the liver. Clinicopathological correlation of 10 cases treated by orthotopic liver transplantation. Am J Surg Pathol 1988; 89: 999–1008. 11. Rodriguez JA, Becker NS, O’Mahony CA, Goss JA, Aloia TA. Long-term outcomes following liver transplantation for hepatic hemangioendothelioma: the UNOS experience from 1987 to 2005. J Gastrointest Surg 2008; 12: 110–6. 12. Kayler LK, Merion RM, Arenas JD, et al. Epithelioid hemangioendothelioma of the liver disseminated to the peritoneum treated with liver transplantation and interferon alpha-2B. Transplantation 2002; 74: 128–30. 13. Theurillat JP, Vavricka SR, Went P, et al. Morphologic changes and altered gene expression in an epithelioid hemangioendothelioma during a tenyear course of disease. Pathol Res Pract 2003; 199: 165–70. 14. Calabrò L, Di Giacomo AM, Altomonte M, et al. Primary hepatic epithelioid hemangioendothelioma progressively responsive to interferonalpha: is there room for novel anti-angiogenetic treatments? J Exp Clin Cancer Res 2007; 26: 145–50. 15. Miller MA, Sandler AD. Elevated plasma vascular endothelial growth factor levels in 2 patients with hemangioendothelioma. J Pediatr Surg 2005; 40: 17–9. 16. Lopez-Terrada DL, Finegold MJ. Liver tumors. In: Walker AW, Durie PR, Hamilton JR, Walker-Smith JA, Watkins J, eds. Pediatric Gastrointestinal Disease: Pathophysiology, Diagnosis, Management, 5th edn. St. Louis: CV Mosby, 2008: 911–26. 17. Selby DM, Stocker J Th, Waclawiw MA, et al. Infantile hemangioendothelioma of the liver. Hepatology 1994; 20: 39–45. 18. Daller JA, Bueno J, Gutierrez J, et al. Hepatic hemangioendothelioma: clinical experience and management strategy. J Pediatr Surg 1999; 34: 98–106. 19. Mo JQ, Dimashkieh HH, Bove KE. GLUT1 endothelial reactivity distinguishes hepatic infantile hemangioma from congenital hepatic vascular malformation with associated capillary proliferation. Hum Pathol 2004; 35: 200–9. 20. Christison-Lagay ER, Burrows PE, Alomari A, et al. Hepatic hemangiomas: subtype classification and development of a clinical practice algorithm and registry. J Pediatr Surg 2007; 42: 62–8. 21. Maluf D, Cotterell A, Clark B, et al. Hepatic angiosarcoma and liver transplantation: case report and literature review. Transplant Proc 2005; 37: 2195–99. 22. Saleh HA, Tao LC. Hepatic angiosarcoma:aspiration biopsy cytology and immunocytochemical contribution. Diagn Cytopathol 1998; 18: 208–11. 23. Husted TL, Neff G, Thomas MJ, Gross TG, et al. Liver transplantation for primary or metastatic sarcoma to the liver. Am J Transplant 2006; 6: 392–7. 24. Weitz J, Klimstra DS, Cymes K, et al. Management of primary liver sarcomas. Cancer 2007; 109: 1391–6.

appendix
Recommended grading of categories of evidence Ia: Ib: IIa: IIb: III: evidence from meta-analysis of randomized controlled trials evidence from at least one randomized controlled trial evidence from at least one controlled study without randomization evidence from at least one other type of quasiexperimental study evidence from non-experimental descriptive studies, such as comparative studies, correlation studies and case–control studies evidence from expert committee reports or opinions and/or clinical experience of respected authorities

IV:

Recommended strengths of management recommendation A. directly based on category I evidence B. directly based on category II evidence or extrapolated recommendation from category I evidence C. directly based on category III evidence or extrapolated recommendation from category I or II evidence D. directly based on category IV evidence or extrapolated recommendation from category I, II, or III evidence

240

RARE VASCULAR LIVER TUMORS
25. Wanless IR. Micronodular transformation (nodular regenerative hyperplasia) of the liver. A report of 64 cases among 2500 autopsies and a new classification of benign hepatocellular nodules. Hepatology 1990; 11: 787–97. 26. Dumortier J, Boillot O, Chevallier M, et al. Familial occurrence of nodular regenerative hyperplasia of the liver: a report on three families. Gut 1999; 45: 289–94. 27. Breen DP, Marinaki AM, Arenas M, et al. Pharmacogenetic association with adverse drug reactions to azathioprine immunosuppressive therapy following liver transplantation. Liver Transpl 2005; 11: 826–33. 28. Gane E, Portmann B, Saxena R, et al. Nodular regenerative hyperplasia of the liver graft after liver transplantation. Hepatology 1994; 20: 88–94. 29. Demetris AJ, Kelly DM, Eghtesad B, et al. Pathophysiologic observations and histopathologic recognition of the portal hyperperfusion or smallfor-size syndrome. Am J Surg Pathol 2006; 30: 986–93. 30. Krasinskas AM, Eghtesad B, Kamath PS, et al. Liver transplantation for severe intrahepatic noncirrhotic portal hypertension. Liver Transpl 2005; 11: 627–34.

241

27

Management of recurrent pyogenic cholangitis W. Y. Lau and C. K. Leow
infection. In the acute stage of RPC, Ong reported that 39.5% of the studied cases had a positive portal blood culture, while the positive supraduodenal lymph node culture rate was 38.1% (12). The rate of infected bile in patients with pigmented stone compared to those with cholesterol stone is correspondingly much higher in the former. In comparing patients with pigmented stones against those with cholesterol stones, Maki demonstrated that 88.3% of the bile in patients with pigmented stone was infected and E. coli was isolated from all cases. This compared to 43.5% in patients with cholesterol stone and of these only 70% of cases had E. coli isolated (13). However, bacteria excreted into the bile within a non-obstructed biliary system will not usually give rise to infection and an attack of cholangitis. Thus obstruction by parasites, such as Clonorchis sinensis (Fig. 27.1) and Ascaris lumbricoides can initiate the sequence of events which eventually lead to the formation of intrahepatic pigment stones (2). Furthermore, the egg or carcass of the parasite can act as a nidus for the deposition of calcium bilirubinate (13). However, only a proportion of patients with RPC have positive ova in stools indicating parasitic infestation, while in some patients the remains of parasites can be identified within the stones recovered (12,13). Clonorchis infection does not occur in the Sarawak state of West Malaysia due to the absence of the snail, Bithynia, which is the first intermediate host in the life cycle of Clonorchis sinensis (12). Yet RPC is common among the Chinese and the indigenous race, the Dyaks. Thus parasitic infestation is only one of the etiological factors for RPC. Maki suggested that the migration of roundworms through the ampulla of Vater leads to papillitis and secondary dyskinesia of the Common Bile Duct (CBD). This leads to increased intrahepatic ductal pressure, dilatation of the CBD and poor drainage of the biliary system (13). A poorly draining biliary system contributing to the formation of intrahepatic stones and colonization of the bile by bacteria was experimentally shown in rabbits by Ong in 1962. Rabbits who had the CBD constricted by a linen thread prior to a single intraportal injection of an E. coli suspension developed ductal dilatation and stone formation in the CBD and intrahepatic ducts. On culture, the bile from the gallbladder and bile ducts was positive for E. coli (12). In the mid 1950s in Japan, the proportion of pigmented to cholesterol stones found in professionals was almost equal. However, almost 90% of the gallstones found in farmers were of pigmented calcium stone. As the farmers were economically less well off, they could only afford a diet which was deficient in fat and protein. It was postulated that the deficient diet may be a factor for the development of pigmented stone (13). Matsushiro et al. have demonstrated that a diet low in protein and fat leads to lower levels of glucaro-1:4-lactone, a powerful inhibitor of β-glucuronidase, in bile (14). The reduced level of glucaro-1:4-lactone in bile thus permits increased hydrolysis

introduction
While practicing in Hong Kong in 1930, Digby drew attention to a condition which was subsequently known as recurrent pyogenic cholangitis by reporting on eight cases of “common duct stones of liver origin” (1). The term Recurrent Pyogenic Cholangitis (RPC) was used by Cook et al. in 1954 when they reported their experience with the condition in a series of 90 patients (2). The synonyms associated with this condition include Asiatic cholangiohepatitis, oriental cholangiohepatitis, Hong Kong Disease (3), Chinese biliary obstruction syndrome (4), and primary cholangitis (5). This condition is commonly seen in Chinese living in Canton and Hong Kong but is not restricted to the Chinese in the Orient since it also occurs in Chinese immigrants in Malaysia, Singapore, North America, and Australia (6–8). RPC is also common in Japanese in Japan and Taiwanese in Taiwan. Although rare, RPC has also been reported to afflict occidentals (9,10).

pathogenesis
In RPC the gallstones found within the biliary system are calcium bilirubinate stones or pigmented calcium stones. Calcium bilirubinate stones are prevalent in Asia and are very rare in Europe and the United States. In addition to the presence of these friable concretions of various shapes and sizes within the biliary tree, the bile is often muddy in consistency and contains numerous fine particles of calcium bilirubinate. Biochemical analysis of these stones revealed a bilirubin content of 40.2% to 57.1% and a cholesterol content of 2.9% to 25.6%. This differs greatly from cholesterol stones, which are common in Europe and the United States, which contain >96% of cholesterol in pure cholesterol stone, and >71.3% in mixed cholesterol stone but the bilirubin content is only 0.02% to 5.0% (11). The peculiarity of the formation of calcium bilirubinate stones in RPC has been ascribed to the high incidence of bile being infected with Escherichia coli (E. coli). In man, the major portion of bilirubin is excreted in bile as bilirubin glucuronide. In the presence of β-glucuronidase, bilirubin glucuronide is hydrolyzed into free bilirubin and glucuronic acid. Normally, calcium is secreted into bile and when it combines with the carboxyl radical of free bilirubin, insoluble calcium bilirubinate is formed. Normal bile is free of β-glucuronidase activity, whereas bile infected with E. coli has intense β-glucuronidase activity. Bile calcium content increases in the presence of biliary tract inflammation and this coupled with the increased hydrolysis of bilirubin glucuronide by the β-glucuronidase from E. coli gives rise to the multiple stones formation classically seen in RPC (11). There are two types of pigmented stones, black, and brown. The infected type seen in RPC is the brown pigment stone. The postulated port of entry for the micro-organisms of bowel origin is via the portal vein from an attack of enteric

242

MANAGEMENT OF RECURRENT PYOGENIC CHOLANGITIS

(A)

(B)

Figure 27.1 (A) Magnified view of Clonorchis sinensis (×12.5). (B) Numerous Clonorchis removed from the CBD of one patient.

of bilirubin glucuronide to free bilirubin and glucuronic acid by the bacterial β-glucuronidase present in infected bile. The free bilirubin then conjugates with calcium in the bile to form the typical calcium bilirubinate stones of RPC. In Hong Kong, RPC is no longer seen in the younger generation born and bred in modern-day Hong Kong. Young patients, in their 30s, who present to our institution with RPC are invariably immigrants from China. We suspect that the much better social and economic conditions of modern-day Hong Kong have played a role in eradicating the condition (15). The reduced incidence of gastroenteritis, the inabilility of the enteric organism, which gained entry via the portal blood, to establish itself within the liver parenchyma due to better host defense from an improved high protein, low carbohydrate diet and possibly the fact that less Chinese herbal medicine is being consumed by the modern generation of youngsters may all have contributed to the demise of this condition.

pathology
Macroscopically, due to the repeated attacks of biliary sepsis, it is common to find adhesions between the liver surface and the surrounding parietal peritoneum, especially the diaphragmatic surface, at operation. The liver surface is scarred and prominent dilated ducts may be obvious. The affected lobe of the liver, usually the left, is normally atrophic with compensatory hypertrophy of the remaining lobe. On palpation, the stones within the dilated biliary ducts are easily palpable. Occasionally, the underlying lobe can be so destroyed by the repeated attacks of cholangiohepatitis that what remains is a cavernous biliary sac with minimal surrounding liver parenchyma (Fig. 27.2). Within the sac is a soup of biliary mud and

stones. The brown pigment stones are soft stones which crumble when squeezed between fingers or forceps. The size variation goes from fine grains to stones of 4 to 5cm in diameter. The stones are irregular, can take up the shape of the biliary duct or become faceted when the stones are packed (Fig. 27.3). Apart from the stones, the bile duct is filled with biliary mud. This is a broth of mucus, altered bile products, microcalculi, desquamated epithelium, parasites, and pus. The pathological hallmark of RPC is the steadily progressive, recurrent cholangiohepatitis with periportal fibrosis. Histologically, in the early acute stage of an attack of cholangiohepatitis, it is similar to that of bacterial cholangitis associated with cholecystitis and calculus obstruction seen in the Western world (16), while the histological picture of the acute, chronic, and advanced stage of the disease is not dissimilar to that seen in sclerosing cholangitis (17). In the early lesions the lumen of the small biliary ducts is filled with pus, with rapid extension into the surrounding tissue. There is marked dissociation of the liver cells by polymorphonuclear infiltration of the sinusoids together with Kupffer cell hyperplasia. In the lobules around the affected duct there is a varying degree of cellular necrosis. Resolution of the underlying inflammation leads to dense round-cell infiltration, which is then replaced by fibrous tissue. In the larger intrahepatic ducts, the duct wall becomes inflamed, ulcerated, and destroyed together with the formation of cholangitic abscesses. Resolution results in intense fibrosis which accounts for the undue prominence of the duct wall seen on sectioning a liver affected by RPC. During the acute episode, these larger ducts can become irregular in caliber and short segments of relative stricture can occur at intervals along the duct. The duct proximal to the stenosis dilates.

243

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 27.2 A totally destroyed left liver lobe consisting of a cavernous biliary sac with negligible liver parenchymal tissue (f = falciform ligament).

Figure 27.3 Large number of pigment stones removed from one patient with RPC.

Recurrent attacks of infection and resolution lead to permanent damage of the duct wall and the ducts remain dilated. The relative stricture then becomes a true localized stricture. These strictures are most frequently encountered at the site of ductal confluence. One of the main concerns of these inflammatory strictures is malignant transformation into cholangiocarcinoma. Ohta et al., from their autopsy studies, suggested that repeated inflammatory damage to the ductal epithelium from the attacks of cholangitis can lead to atypical epithelial hyperplasia, dysplasia, and eventually cholangiocarcinoma (18). The left lobe of the liver is preferentially affected. The exact reason for this is unknown. One possible explanation may be the selective distribution of portal blood within the liver. Two studies suggested that the left lobe of the liver receives blood from the colon and the left lobe will be the first port of call for enteric organisms such as E. coli which have entered the portal venous system (19,20). Colonization of the bile with E. coli will lead to the production of β-glucuronidase in the biliary system. Another explanation is that the more oblique course of the left hepatic duct results in poorer drainage of the left ductal system as compared to the right hepatic duct, thus leading to increased incidence of stone formation. If stones are found in the right hepatic duct, almost invariably stones are found in the left duct. In one study of 115 patients with hepatolithiasis, the ratio of stones found in the left and right hepatic ducts was 6:1 (5). The stones form in the dilated ducts proximal to the stricture site. These strictures can be multiple and bilobar in distribution and commonly occur at the origin of the right and left hepatic ducts. Stones within the common bile duct are usually lodged at the supraduodenal portion of the duct or at the ampulla. At ERCP, a patulous ampulla of Vater (probably a result of repeated passage of stones) is not an uncommon finding in patients with RPC. The bile in patients with hepatolithiasis is usually infected with enteric organisms. The two most common organisms isolated are E. coli and Klebsiella species. The overall positive bile culture rate has been reported to be as high as 87% and the incidence of positive culture in patients requiring surgical

intervention and those which settled on conservative measures is similar (90% vs. 85%) (21). The gallbladders in these patients are usually thin-walled, large, and distended. The majority of them do not contain gallstones. While the incidence of CBD stones and biliary mud varied from 60% to 90%, the incidence of associated gallstones in the gallbladder was only 15% to 40% (4,7). Macroscopically, the gallbladder looks normal but histological examination invariably shows features compatible with low-grade chronic cholecystitis. Along the gastrohepatic omentum gross lymphadenitis with enlarged lymph nodes is commonly encountered.

clinical presentation
Patients with RPC tend to be younger (third and fourth decade) than those affected by cholesterol stone disease, which is much more prevalent in older women, in the Western world. Although the condition does not have particular sex prevalence, those afflicted are almost invariably from the lower socio-economic classes. The usual presentation consists of the classical Charcot’s triad of abdominal pain, fever (with or without chills and rigors), and jaundice which signifies an attack of cholangitis. The patient may not notice the jaundice but a history of tea-colored urine is usual. The jaundice is usually not severe since cholangitis secondary to a completely obstructed biliary system will rapidly progress to acute suppurative obstructive cholangitis with septicemia. In addition to the triad of symptoms, these patients also develop mental confusion and shock, which is referred to as Reynaud’s pentad. The epigastric or right upper quadrant pain is usually described as a constant and gnawing or cutting (12) pain, which may radiate to the back. Vomiting is not a constant feature. Patients have spiking fever, not unlike that seen with an underlying abscess or collection, which normally resolves rapidly when the conservative treatment has been effective. On examination, the patient looks unwell and restless with a tinge of jaundice. The associated jaundice is typically mild and clinically can be just discernible. Abdominal examination may reveal the telltale signs of surgical scars from previous

244

MANAGEMENT OF RECURRENT PYOGENIC CHOLANGITIS
operations. Tenderness with varying degree of guarding is noted in the epigastrium or right upper quadrant. A tender hepatomegaly may be present. Marked tenderness may imply the presence of an underlying abscess. Deterioration in the abdominal signs (increasing and generalized tenderness) and/ or the development of worsening hemodynamic parameters (persistent hypotension, tachycardia, poor urine output despite adequate resuscitation) argues for emergency surgical intervention to decompress the biliary system. Those patients who present or develop shock have a flushed facie, warm periphery, bounding peripheral pulse, and hypotension. The massive vasodilatation and reduced cardiac contractility secondary to the endotoxemia adequately explain the state of shock. filling defects are of higher attenuation than bile, but have a lower attenuation than contrast-enhanced liver parenchyma (23). Although uncommon, the amorphous stones which completely fill the ducts, and which are isodense with the hepatic parenchyma could be missed on CT. Unlike USG, there is no difficulty in distinguishing pneumobilia from stones on CT and visualization of the extrahepatic ducts is not limited by overlying bowel gas. USG and CT do not provide sufficient details of the ductal anatomy. Cholangiography, in the form of Endoscopic Retrograde Cholangiopancreatography (ERCP) and/or Percutaneous Transhepatic Cholangiography (PTC), is essential for the detail delineation of the entire biliary tract. Older methods of delineating the biliary tracts, such as oral cholecystography and intravenous cholangiography are no longer used because they provide suboptimal visualization of the ductal system. In addition to the potential complication of allergic reaction to the contrast injected, there is no place for intravenous cholangiography in the presence of biliary obstruction or cholangitis. Our first line of investigation is ERCP since it is both diagnostic and therapeutic. The typical cholangiogram will show dilated extrahepatic ducts in more than half the cases. The intrahepatic ducts have the classical “truncated tree” pattern where the “tree” has been trimmed back to its “main branches”. The terminal end of these “branches” is tapered, resembling an arrow or a spear head (Fig. 27.4). It is more common to see a dilated left ductal system containing calculi than an affected right system. There may be an accompanying relative or true stricture distal to the dilated ducts. When a stone or stricture prevents the filling of the intrahepatic ducts, or when it is technically impossible to perform an ERCP due to previous biliary–enteric bypass surgery, PTC under USG guidance is performed. As a result of the stones and stricture(s), the biliary anatomy can be very complicated. Once an obstructed biliary duct is punctured during PTC, the obstructed system must be drained to avoid cholangitis and/or bile leak. Underfilling during cholangiography can lead to missed segmental ducts. More importantly, the paucity of intrahepatic ducts shown on cholangiograms should prompt the surgeon to count the number of segmental ducts present in order not to miss the diagnosis of undrained segment(s). Cholangiography complements USG and CT and their findings should be considered as a whole and not in isolation. We have shown that Magnetic Resonance (MR) cholangiography is comparable to ERCP in diagnosing choledocholithiasis (24). Apart from being non-invasive, MR cholangiography can delineate biliary strictures which may be difficult to show or missed on ERCP due to technical reasons (Fig. 27.5). In a recent study by Park et al. MR cholangiography has been shown to be better than ERCP/PTC (25). Occasionally, a radioisotope scan is performed to demonstrate the presence of undrained or hypo-functioning liver segments. The radiographic features of certain conditions can simulate RPC. Sclerosing cholangitis can lead to biliary tract strictures. However, these are usually more peripherally located and there is a lack of the marked proximal dilatation and stones seen in RPC. Although the common bile duct is massively dilated in choledochal cysts, in most cases, there is an abrupt transition

investigations
Both the hematological and biochemical tests do not differentiate patients with RPC from those with other causes of biliary obstruction and infection. Full blood count will reveal an underlying leucocytosis with neutrophilia and mild thrombocytopenia in some patients. A number of patients will also have a concomitant mild derangement of the clotting profile with a prolonged prothrombin time. The deranged liver function test is compatible with an obstructive picture with a moderately raised level of bilirubin and a high serum alkaline phosphatase level. The level of γ-glutamyltranspeptidase is elevated. The slightly elevated alanine transaminase level in some patients is a reflection of the parenchymal damage secondary to the underlying infection within the biliary system. Other than showing the presence of pneumobilia, in some cases a plain abdominal radiograph is not helpful. The calcium bilirubinate stones are radiolucent because of the low calcium and high bilirubin content. The least expensive and most helpful investigation is ultrasonography (USG). It can demonstrate the presence of stones within the dilated intrahepatic and common bile ducts, the presence or absence of an underlying liver abscess and occasionally the presence of a solid liver mass secondary to malignancy complicating a benign stricture. Also it can reveal the presence or absence of gallstone(s) within the gallbladder. Intra- and extrahepatic ductal stones are present in the majority of patients, but cholelithiasis is much less common and is seen in the minority cases. Although intrahepatic stones normally cast sonic shadows on USG, the presence of air within the bile ducts (either spontaneously or secondary to previous biliary drainage procedures) can give rise to highly reflective echoes with posterior shadows, thus confusing and misleading the radiologist in diagnosing the presence of stones within the bile ducts. In about 3% of RPC, pneumobilia is present (22). In some cases the amorphous and small stones can form a cast of the biliary tree and, under such circumstances, highly reflective echoes and posterior acoustic shadowing on USG may be absent. It may then be difficult to identify dilated bile ducts and the ducts can appear as soft tissue masses on USG (7). USG images and its interpretations are operator dependent. Computed tomography (CT) removes this bias and can provide images of the dilated intra- and extrahepatic ducts, even if they are filled with sludge or pus. On CT scanning these

245

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
merely help to define the extent and severity of the underlying disease and guide the management plan.

acute management
RPC patients have repeated attacks of acute cholangitis which would settle on conservative measures in the majority of cases. The need for urgent therapeutic interventional procedures only applies to a minority of cases, such as those with signs of peritonitis secondary to perforated gangrenous gallbladder, ruptured liver abscess, or those with septicemic shock despite conservative measures. The role of definitive procedures for most patients who settled on conservative measures depends on the frequency and severity of each attack, presence of biliary strictures (which may be malignant) and the presence of any existing co-morbid medical conditions. The initial approach to any acute attack is to control the underlying infection with the commencement of intravenous fluid infusion, antibiotic treatment after blood culture, prescription of adequate analgesia and keeping the patient nil per oral. Our standard first line antibiotic regimen is cefuroxime. Metronidazole is sometimes prescribed to cover the anaerobe Bacteroides fragilis, which is present in a minor proportion of patients who have had previous biliary tract surgery and/or complicated anatomy due to stones and strictures. An urgent ultrasound scan of the liver is performed to identify the extent of lithiasis within the biliary tree, the presence of a liver mass which can be an abscess or an underlying cholangiocarcinoma. Those patients who fail to respond or have evidence of a severe attack of cholangitis with or without shock undergo an urgent ERCP. The smallest amount of contrast feasible is used during ERCP as increased biliary pressure from excessive contrast injection will result in cholangio-venous reflux, which can lead to septicemia. No attempt is made to perform a full cholangiogram or to remove all calculi from the biliary system. In the procedure, the system is decompressed with a nasobiliary drain. Only when the patient’s condition has improved and stabilized would a check cholangiogram with endoscopic removal of stones be performed. If part of the biliary tree cannot be decompressed adequately because of an obstructing distal stone or stricture, then endoscopic drainage alone may not be adequate and successful. As such, percutaneous transhepatic biliary drainage of the obstructed biliary ducts will be of use. However, these drainage tubes are small and can be easily blocked by the tenacious biliary mud. If a liver abscess is present, the abscess is drained percutaneously under ultrasound guidance. The patient is monitored closely after admission for signs of deterioration. Those responding to the conservative treatment will have a reduction in abdominal pain, a fall in temperature toward normal and the disappearance of tachycardia over the first 24 to 48 hours. If there is no obvious improvement after 48 hours, the possibility of undrained biliary system or individual liver segments due to impacted stones or underlying strictures must be considered and the need for urgent surgical intervention entertained. At any time during conservative management, the presence of increasing abdominal pain coupled with shock and peritoneal signs mandate urgent surgical treatment.

Figure 27.4 Typical truncated biliary tree appearance together with the arrow or spear head sign (arrow) on ERCP. A large stone(s) in the left duct.

Figure 27.5 MR Cholangiogram demonstrating a stricture (arrow) at the confluence of the right and left hepatic ducts.

to normal or slightly dilated proximal ducts (26). Patients with Caroli disease (cavernous ectasia of the biliary tract) have dilated intrahepatic ducts and calculi but the extrahepatic ducts are disproportionately small (27). The condition is often associated with renal cystic disease (28) which, on CT, helps to distinguish it from RPC. In almost all cases, given the clinical and investigation findings, the diagnosis of RPC is seldom in doubt. The investigations

246

MANAGEMENT OF RECURRENT PYOGENIC CHOLANGITIS
Those patients who are present in shock must be actively resuscitated. Those who respond quickly can be treated conservatively, while those who fail must undergo therapeutic intervention. It is unclear why conservative measures work for some but not others. In a series of 88 RPC patients presenting with acute cholangitis, 17% required therapeutic intervention for septicemic shock. A pulse rate greater than 100/min and a platelet count of more than 150 × 109/l within 24 hours of presentation were the only two independent factors that predicted the need for therapeutic intervention (21). The sole aim of an urgent therapeutic intervention during an acute attack is to decompress the obstructed biliary system. Non-operative interventional procedures using the endoscopic or percutaneous routes are preferred to open surgical route (29). Occasionally, open surgery is required to deal with peritonitis as a result of gangrenous cholecystitis or ruptured liver abscess. The most expedient means to decompress the CBD is insertion of a large T-tube. No attempt is made to perform definitive surgery. When the usually enlarged and thickwalled CBD is opened, thick-infected biliary mud and bile will be gushed out. The biliary mud and friable bilirubinate stones are scooped out. Following a gentle saline flush of the CBD, a bougie is passed down the CBD into the duodenum to check for patency of the lower end of the CBD. The way an impacted stone in the lower end, which cannot be removed during CBD exploration, is dealt with depends on whether there is a concomitant attack of pancreatitis. In the absence of acute pancreatitis, the stone can be dealt with percutaneously via the T-tube tract when the acute episode is over. In the presence of acute pancreatitis, a transduodenal sphincteroplasty is performed to remove the stone. An alternative is the use of electrohydraulic lithotripsy to fragment the impacted stone, thus avoiding a transduodenal sphincteroplasty (30). The patency of the right and left hepatic ducts is checked to ensure there is free flow of bile into the CHD and CBD. Any strictures found are dilated with graduated sounds to release the infected bile and mud dammed up behind the stricture(s). Gentle irrigation of the hepatic ducts with saline is performed. Irrigation or flushing at high pressure via a syringe must be resisted as this can initiate or aggravate a septicemic state. When bile flow from both hepatic ducts is established, a large bore T-tube is placed, the choledochotomy closed with catgut and the operation is terminated. The T-tube not only decompresses the system but also affords a percutaneous route for endoscopic intervention when the patient has recovered. Large palpable liver abscesses are drained intraoperatively. Multiple small abscesses will respond to appropriate antibiotics after the biliary system is decompressed. A cholecystectomy is performed only when it is grossly distended or there is evidence of cystic duct obstruction, empyema or gangrene of the gallbladder. During the emergency CBD exploration, an otherwise non-inflamed gallbladder, with or without stone(s) in situ, is left behind because of the added risk of performing a cholecystectomy in an ill patient. establish adequate drainage to the biliary system, and to resect non-functioning liver segments which harbor bacteria and serve as foci of infection. If properly performed, definitive interventional procedures decrease the episodes and severity of future attacks of cholangitis. In some patients cure is possible.

minimal access approach
Once the acute episode has settled, more definitive treatment via the endoscope or under radiological guidance can be performed. Those treated initially by nasobiliary drainage have a check cholangiogram to delineate the extent of lithiasis and the existence of ductal stricture(s). Stones within the CBD and CHD can be removed with a dormia basket and large stones can be crushed with the mechanical lithotripter or fragmented by laser prior to their removal. For those RPC patients with stones confined to the CBD, endoscopic sphincterotomy with stone extraction only is safe and effective. The medium-term result of endoscopic sphincterotomy is comparable to surgical sphincteroplasty. In 118 patients who underwent endoscopic treatment, 95.8% remained symptom-free after a median follow-up of 2.3 years (8), compared to 83.4% who had a good outcome after surgical sphincteroplasty at a mean follow-up of 7.3 years (32). In the absence of ductal strictures, small intrahepatic calculi not retrieved by ERCP can be shattered by Extracorporeal Shock Wave Lithotripsy (ESWL) and the fragments allowed to fall into the bowel through a widely patent sphincteroplasty. Intrahepatic calculi within dilated biliary ducts usually lie proximal to a site of relative or true stricture. The stricture can be dilated sufficiently to allow complete removal of the stones endoscopically (Fig. 27.6). When the stricture is confined to one lobe of the liver which is atrophic, and the contralateral liver lobe is normal or relatively unaffected, hepatic resection should be performed unless the patient is medically unfit to undergo liver resection. In the presence of multiple strictures, a more conservative approach with repeated dilatation can be successful in achieving stone clearance and control of disease. Balloon dilatation of intrahepatic biliary strictures prior to stone removal has been reported to be highly successful. The immediate overall success rate of complete stone clearance with balloon dilatation in 57 patients was 94.5% (33). Long segmental strictures which are likely to restenose can be successfully stented. The main complications of dilatation therapy include septicemia, hemobilia, mild diarrhea, and restenosis. The cumulative probability of stricture recurrence after dilatation is 4% at 2 years and 8% at 3 years (33). The true long-term patency rate following dilatation alone is still unknown since benign strictures surgically treated can recur 10 or more years later, as partial obstructions can remain completely asymptomatic for long periods. Apart from the problem of restricturing, it is difficult to rule out the presence of a malignant stricture with certainty. In patients with a Percutaneous Transhepatic Biliary Drainage (PTBD) catheter in situ, the tract can be dilated to allow dormia basket stone retrieval under fluoroscopic screening or to allow passage of a flexible twin channel choledochoscope. Under direct vision the stone(s) can be fragmented using the

definitive management
The definitive management of RPC is to use a multidisciplinary approach (31), aiming to remove all biliary stones, to

247

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C) Figure 27.6 (A) ERCP demonstrating the presence of stones proximal to a relative stricture (arrow) in the left hepatic duct, (B) the stricture is dilated with a balloon prior to stone retrieval, and (C) the dilated left duct is cleared of stones.

electrohydraulic lithotripter and the fragments removed with a basket. Intrahepatic strictures can be dilated or stented. Instillation of stone dissolving agents directly into the affected biliary duct has been advocated by some, but we do not practise this approach as it is often painful, time-consuming, ineffective, and can lead to ascending cholangitis and sepsis. In patients where the initial endoscopic approach has failed, the established percutaneous route can be combined with endoscopy subsequently to achieve stone clearance or stricture dilatation. In those patients who received acute surgical intervention and T-tube decompression of the biliary system, the tract is allowed to mature. After 6 weeks, any stones present can be removed through the tract with a choledochoscope or under radiological control.

definitive surgery
Since RPC can and does affect the biliary tract at different sites with varying degrees of severity, the aim of the surgery is to provide adequate biliary drainage for bile and debris. This encompasses stone extraction, stricturoplasty or excision of

stricture, resecting non-functioning liver segments and creating a bilio–enteric bypass with a permanent percutaneous access loop to the biliary tract to allow subsequent access to the biliary system for stone extraction and dilatation of stricture(s). Before embarking on definitive surgery, it is mandatory to have a complete knowledge of the location of calculi and stricture(s), and the uni- or bilobar extent of disease with or without concomitant liver atrophy. In the presence of predominantly extrahepatic disease, simple exploration of the common bile duct with intraoperative choledochoscopy will suffice. In the absence of extrahepatic ductal stricture, the incisions used for the exploration of the CBD and hepatic ducts will depend on the location of the stones (Fig. 27.7A–D). We routinely remove the gallbladder in these patients since, histologically, it shows underlying evidence of low grade inflammation. Furthermore, if the sphincter of Oddi has been previously destroyed, the gallbladder will permanently be in a collapsed state. The placement of a large T-tube following the exploration will allow post-operative imaging of the biliary tracts and any residual stones found can

248

MANAGEMENT OF RECURRENT PYOGENIC CHOLANGITIS
stones from the intrahepatic ducts. Once the effluent is relatively clear, a repeat choledochoscopy is performed. Any residual stones can be retrieved with a dormia basket. Segmental ductal stricture is dilated prior to stone retrieval. Any impacted or large stones can be shattered with the electrohydraulic lithotripter introduced through the working channel of the flexible choledochoscope. It is not always possible to clear the intrahepatic ducts of stones completely. Once the large stones and strictures are dealt with, small stones or fragments can be left to “fall out”, provided an adequate biliary drainage procedure has been performed. The thick, dilated CBD/CHD with an inelastic wall behaves more like a cavernous sac which does not drain adequately, even in the presence of a sphinteroplasty. Although it is reasonable to perform a supraduodenal choledochoduodenostomy as a biliary drainage procedure, this has the disadvantage that patients may develop “sump syndrome”, developing symptoms of pain and fever, which is thought to be a result of debris being lodged in the diseased distal CBD. While the sump syndrome can be treated by performing an endoscopic sphincterotomy, supraduodenal choledochoduodenostomy is contra-indicated in the presence of a proximal biliary tract stricture or when the CBD is not wide enough. Under such circumstances, we routinely perform an end-to-side hepaticojejunostomy with a retrocolic Roux-en-Y loop. This provides a widely patent anastomosis for biliary drainage and for small stones to fall freely into the loop of bowel. Furthermore, the closed end of the Roux loop can provide a permanent access route to the biliary tract. With the hepaticojejunostomy it is crucial that the access loop is short and has a straight course from the anterolateral abdominal wall to the anastomosis. A poorly constructed jejunal loop with redundant length makes subsequent choledochoscopy and access to the intrahepatic ducts difficult. The access loop can either be placed intraperitoneally, to be accessed percutaneously under ultrasound guidance, or it can be placed under the skin as a hepaticocutaneous jejunostomy. We prefer to leave the loop intraperitoneally. In those cases where immediate access to the biliary tract is not necessary, the end of the loop is tacked to the peritoneal surface of the anterolateral abdominal wall with catgut. The staples from the GIA stapler used to transect the jejunum during Roux loop formation or the ligaclips placed around the blind end of the loop at the end of the operation allow the interventional radiologist subsequently to identify and access the loop percutaneously. Once the loop is punctured, the tract can be gradually dilated to admit a twin channel choledochoscope for endoscopic manipulation (15). For those cases where access to the biliary tract is required soon after the operation, a 24 Fr catheter is introduced through the anterolateral abdominal wall into the access loop at the end of the operation. The tract is allowed to mature for 4 to 6 weeks before instrumentation. Hepaticocutaneous jejunostomy has been used by others (36). Although the biliary tract can be accessed sooner through the cutaneous stoma compared to our technique, the cutaneous stoma is not without its complication. Fifteen percent of patients with hepaticocutaneous jejunostomy, performed either with the cutaneous stoma formed at the initial operation

(A)

(B)

(C)

(D)

Figure 27.7 (A) Longitudinal incision for the exploration of CBD only, (B) separate incisions for exploration of the CBD and the hepatic ducts, (C) incision for the combined exploration of the CBD and one of the hepatic ducts (left side is shown), (D) horizontal incision over the hepaticojejunostomy (---- = incision).

be easily removed under radiological control or with a flexible choledochoscope. When there is stenosis of the ampulla of Vater or distal CBD, or impacted stone(s) in the lower CBD, a transduodenal sphincteroplasty is performed. In patients who have had multiple operations on the CBD, the standard approach can be difficult. Under such circumstances, Ong et al. have described an extraperitoneal approach to the duodenum, which is located by its anterior position to the right kidney. A transduodenal sphincteroplasty is performed and the CBD is explored from below (34,35). However, this approach is seldom necessary as the result of endoscopic sphincterotomy has been shown to be comparable (8). In the presence of extra- and intrahepatic calculi, stone extraction can be difficult if they are impacted, situated behind relative or true ductal strictures or, present within angulated ducts, such as the right posterior or left medial segmental ducts. Although direct hepatotomy can be performed to remove the stones, it can be very bloody if the stones are deepseated. Following the removal of all the stones within the CBD and CHD, after a choledochoscopic examination to rule out any strictures in the right and left hepatic ducts, the right and left ducts are flushed with saline. The right and left lobes of the liver are gently massaged in-between the flushing. This maneuvre normally helps to discharge more biliary mud and small

249

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
or subsequently created from its subcutaneous site, did experience complications. These complications include wound infection of the closed stoma wound, development of cutaneous fistula after stoma closure, difficulty in reconstructing the stoma for subsequent use and development of parajejunostomy hernia. Repeated therapeutic intervention of the biliary tract is inevitable due to the chronic relapsing nature of the condition. The formation of a hepaticojejunostomy with an access loop is a satisfactory and adequate way to manage further attacks of stones, strictures and cholangitic attacks. On a median follow-up time of 27 months, symptoms recurred only in 12 patients (29%). Only one patient required a reoperation for stricture while the others were adequately treated through the access jejunal loop (36). RPC predisposes to the formation of inflammatory, noniatrogenic strictures in the biliary tract. Strictures in the subsegmental and peripheral ducts are difficult to treat. Fortunately, obstruction of these minor ducts does not produce jaundice and the cholangitic attacks may pass unnoticed and the infection subsides without any intervention. However, strictures in the major bile ducts do lead to unrelenting cholangitis, stones and liver abscess formation, septicemic episodes, and death. Over a 10-year period, we treated 57 patients with major bile duct strictures (37). Forty-four of the 60 strictures involved the left hepatic duct (23) or the left lateral segmental ducts (21). Strictures in the CBD or CHD can be treated by the formation of a choledochojejunostomy or a hepaticojejunostomy. Stricture involving the confluence of the hepatic ducts can be treated by hepatotomy and Y–V plasty, but the procedure is technically difficult and bleeding, bile leakage and restenosis are potential serious complications (5,38). We have stopped performing Y–V plasty for such strictures due to our poor results. Instead, we treat such strictures by performing bilateral hepaticojejunostomy. In a relatively normal left lobe of the liver with a left duct stricture at its origin, a left duct approach as described by Hepp and Couinaud (39) for a side-to-side anastomosis between the left duct and a Roux loop can be performed without the need for a left hepatectomy. The biliary drainage procedures performed for these inflammatory strictures are occasionally combined with liver resection, or liver resection alone is performed to deal with these strictures. Before embarking on hepatic resection, the severity of symptoms, status of the remaining biliary tract, parenchymal functional reserve, and alternative procedures have to be considered. Hepatic resection is only performed for those with recurrent, troublesome and localized severe disease. Disease can be confined to the left lateral segment which is atrophic and contains large number of calculi within cavernous bile ducts. In more extensive disease, the medial segment of the left lobe can also be affected due to a tight stricture in the left hepatic duct. Hepatectomy is performed to remove not only the source of symptoms and sepsis, but also the underlying stricture which has the potential to turn malignant. Liver resection for rightsided disease is unusual. By the time resection is necessary, the right lobe is usually destroyed, with compensatory left lobe hypertrophy. Consequently, a right hepatectomy should lead to little functional disturbance. In one series of 172 patients with hepatolithiasis, liver resection was necessary in 37% of patients, of which left lateral segmentectomy and left hepatectomy accounted for most of the resections (90.5%). Right hepatectomy was only performed in one patient (40). Troublesome adhesions between the diseased liver and the diaphragm and adjacent viscera can make liver resection difficult. Severe adhesions and fibrosis around the left hepatic vein and the inferior vena cava can make dissection in the region difficult and severe bleeding from these structures due to injudicious dissection can be a problem. The overall operative mortality in hepatic resection for hepatolithiasis is low (<2%), but the morbidity, such as wound infection, subphrenic collections, and biliary fistulae, from operating on an underlying septic condition is correspondingly high (approximately 30%).

complications
In RPC the biliary mud and stones within the common bile duct can lead to acute pancreatitis. In 1971, Ong et al. reported that approximately half of all patients with acute pancreatitis, in Hong Kong, were associated with RPC (41). Another report claims that about 20% of RPC patients had high serum amylase levels but were clinically asymptomatic (42). In some patients, the big common bile duct stones can lead to the formation of a choledochoduodenal fistula. Liver abscesses complicating RPC can present with rupture into the peritoneal cavity or adjacent viscera. A left lobe liver abscess can rupture into the pericardial cavity and cause cardiac tamponade (43), while a right lobe abscess can lead to the formation of a pleurobiliary or bronchobiliary fistula (44). A chronic abscess can, on clinical and radiological grounds, be indistinguishable from an underlying cholangiocarcinoma. The final diagnosis can only be certain after histological examination of the resected specimen. Cholangiocarcinoma complicating RPC has been reported. The higher incidence of cholangiocarcinoma in areas where RPC is also prevalent has been attributed to the presence of Clonorchis sinensis infestation (45). In a necropsy study of 50 cases of cholangiocarcinoma, 92% of the cases were associated with clonorchiasis and intrahepatic stones were found in 20% of the cases (45). In a huge series of 1105 cases of hepatolithiasis studied over the period 1978 to 1990, Chen et al. (46) reported that the incidence of cholangiocarcinoma in these patients increased from 2.4% (between 1978 and 1987) to 13.7% (between 1988 and 1990), despite a decreasing incidence of clonorchiasis in the population. Thrombophlebitis of the branches of the portal vein due to the underlying periductal inflammation can lead to portal venous thrombosis with an enlarged spleen (6). Occasionally, septic emboli to the pulmonary tree can lead to the development of lung abscesses and significant pulmonary hypertension (6,47). Despite multiple operations, RPC patients with long-standing severe disease can develop secondary biliary cirrhosis and liver failure (48). When cirrhosis sets in, portal hypertension and bleeding esophageal varices ensue, thus making further corrective surgery for the underlying stricture(s) more hazardous. In these patients the only available option is liver transplantation.

250

MANAGEMENT OF RECURRENT PYOGENIC CHOLANGITIS

conclusions
As the name implies, RPC runs a recurrent and unrelenting course with variable frequencies of attacks of cholangitis. Medical therapy is ineffective and surgical treatment is not entirely satisfactory. Despite surgery, stones and strictures can return. In Hong Kong and Taiwan, the peak age incidence of RPC has changed over the years from the third to the fifth decade in the 1950s and 1970s to the seventh decade in the early 1990s (49). This is due to an increasing proportion of patients who have survived previous surgery, only to have RPC recur again in later life. In Hong Kong, young patients in their third and fourth decade of life who present to us with RPC are invariably immigrants from mainland China. RPC is a dying disease in Hong Kong, but is still common in China. Although there are various theories on the pathogenesis of RPC, we believe the condition is closely linked to the level of social and economic conditions of a community. Surgery merely deals with the consequences of the condition, but does not address its roots. With better living conditions and public hygiene, perhaps RPC can be eradicated in this millennium. Until then, a judicious choice of a mixture of treatment, both medical and surgical, is necessary to achieve a satisfactory and long-lasting solution to a recurrent inflammatory condition.

key points
● ● ●





Calculi are predominantly calcium bilirubinate. Probably secondary to E. coli infection of bile. Presentation: Third and fourth decade Recurrent attacks of cholangitis Obstructive jaundice Preferentially affects left lobe. Investigations: Plain abdominal radiography (AXR) CT ERCP B1 PTC. Complications: Acute pancreatitis Liver abscess Cholangiocarcinoma Portal thrombophlebitis Secondary biliary cirrhosis.

references
1. Digby KH. Common-duct stones of liver origin. Br J Surg 1930; 17: 578–91. 2. Cook J, Hou PC, Ho HC, et al. Recurrent pyogenic cholangitis. Br J Surg 1954; 42: 188–203. 3. Mage S, Morel AS. Surgical experience with cholangiohepatitis (Hong Kong Disease) in Canton Chinese. Ann Surg 1965; 162: 187–90. 4. Harrison-Levy A. The biliary obstruction syndrome of the Chinese. Br J Surg 1962; 49: 674–85. 5. Choi TK, Wong J, Ong GB. The surgical management of primary intrahepatic stones. Br J Surg 1982; 69: 86–90. 6. Chou ST, Chan CW. Recurrent pyogenic cholangitis: a necropsy study. Pathology 1980; 12: 415–28. 7. Federle MP, Cello JP, Laing FC, et al. Recurrent pyogenic cholangitis in Asian immigrants. Use of ultrasonography, computed tomography, and cholangiography. Radiology 1982; 143: 151–6. 8. Lam SK. A study of endoscopic sphincterotomy in recurrent pyogenic cholangitis. Br J Surg 1984; 71: 262–6.

9. Wilson MK, Stephen MS, Mathur M, et al. Recurrent pyogenic cholangitis or ‘oriental cholangiohepatitis’ in occidentals: case reports of four patients. Aust N Z J Surg 1996; 66: 649–52. 10. Menu Y, Lorphelin JM, Scherrer A, et al. Sonographic and computed tomographic evaluation of intrahepatic calculi. AJR 1985; 145: 579–83. 11. Maki T. Pathogenesis of calcium bilirubinate gallstone: role of E. coli, β-glucuronidase and coagulation by inorganic ions, polyelectrolytes and agitation. Ann Surg 1966; 164: 90–100. 12. Ong GB. A study of recurrent pyogenic cholangitis. Arch Surg 1962; 84: 63–89. 13. Maki T. Cholelithiasis in the Japanese. Arch Surg 1961; 82: 599–612. 14. Matsushiro T, Suzuki N, Sato T, et al. Effects of diet on glucaric acid concentration in bile and the formation of calcium bilirubinate gallstones. Gastroenterology 1977; 72: 630–3. 15. Leow CK, Lau WY. Biliary access procedure in the management of oriental cholangiohepatitis. Am Surg 1998; 64: 99. 16. Flinn WR, Olson DF, Oyasu R, et al. Biliary bacteria and hepatic histologic changes in gallstone disease. Ann Surg 1977; 185: 593–7. 17. Thorpe MEC, Scheuer PJ, Sherlock S. Primary sclerosing cholangitis, the biliary tree and ulcerative cholangitis. Gut 1967; 8: 435–48. 18. Ohta G, Nakanuma Y, Terada T. Pathology of hepatolithiasis: cholangitis and cholangiocarcinoma. Prog Clin Biol Res 1984; 152: 91–113. 19. Copher GH, Dick BM. ‘Streamline’ phenomena in portal vein and selective distribution of portal blood in liver. Arch Surg 1928; 17: 408–19. 20. Hahn PF, Donald WD, Grier RC Jr. Physiological bilaterality of the portal circulation; streamline flow of blood into liver as shown by radioactive phosphorus. Am J Physiol 1945; 143: 105–7. 21. Fan ST, Lai ECS, Mok FPT, et al. Acute cholangitis secondary to hepatolithiasis. Arch Surg 1991; 126: 1027–31. 22. Wastie ML, Cunningham IGE. Roentgenologic findings in recurrent pyogenic cholangitis. AJR 1973; 119: 71–7. 23. Itai Y, Araki T, Furui S, et al. Computed tomography and ultrasound in the diagnosis of intrahepatic calculi. Radiology 1980; 136: 399–405. 24. Chan YL, Chan ACW, Lam WWM, et al. Choledocholithiasis: comparison of MR cholangiography and endoscopic retrograde cholangiography. Radiology 1996; 200: 85–9. 25. Park MS, Yu JS, Kim KW, et al. Recurrent pyogenic cholangitis: comparison between MR cholangiography and direct cholangiography. Radiology 2001; 220: 677–82. 26. Araki T, Itai Y, Tasaka A. CT of choledochal cyst. AJR 1980; 135: 729–34. 27. Kaiser JA, Mall JC, Salmen BJ, et al. Diagnosis of Caroli disease by computed tomography: report of two cases. Radiology 1979; 132: 661–4. 28. Mujahed Z, Glenn F, Evans JA. Communicating cavernous ectasia of the intrahepatic ducts (Caroli’s disease). AJR 1971; 113: 21–6. 29. A0Lillemoe KD. Surgical treatment of biliary tract infection. Am Surg 2000; 66: 138–44. 30. Fan ST. Transduodenal sphincteroplasty for impacted stone made unnecessary by electrohydraulic lithotripsy. Surg Gynecol Obstet 1989; 169: 363–4. 31. A0Cosenza CA, Durazo F, Stain SC, et al. Current management of recurrent pyogenic cholangitis. Am Surg 1999; 68: 939–43. 32. Choi TK, Wong J, Lam KH et al. Late result of sphincteroplasty in the treatment of primary cholangitis. Arch Surg 1981; 116: 1173–5. 33. Jeng KS, Yang FS, Ohta I, et al. Dilatation of intrahepatic biliary strictures in patients with hepatolithiasis. World J Surg 1990; 14: 587–93. 34. Ong GB, Kwong KH, Cheng FCY. Extraperitoneal transduodenal choledocho-duodenostomy for removal of overlooked common bile duct stones. Aust N Z J Surg 1970; 40: 166–70. 35. Choi TK, Lee NW, Wong J, et al. Extraperitoneal sphincteroplasty for residual stones: an update. Ann Surg 1982; 196: 26–9. 36. Fan ST, Mok F, Zheng SS, et al. Appraisal of hepaticocutaneous jejunostomy in the management of hepatolithiasis. Am J Surg 1993; 165: 332–5. 37. Lau WY, Chan KL, Li AKC. Surgical treatment of inflammatory strictures of the major bile ducts. Asian J Surg 1990; 13: 51–4. 38. Lau WY, Fan ST, Yip WC, et al. Surgical management of strictures of the major bile ducts in recurrent pyogenic cholangitis. Br J Surg 1987; 74: 1100–2. 39. Hepp J, Couinaud C. L’abord et l’utilisation du canal hepatique gauche dans les reparations de la voie biliaire principale. Presse Med 1956; 64: 947–8.

251

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
40. Fan ST, Lai ECS, Wong J. Hepatic resection for hepatolithiasis. Arch Surg 1993; 128: 1070–4. 41. Ong GB, Adiseshiah M, Leong CH. Acute pancreatitis associated with recurrent pyogenic cholangitis. Br J Surg 1971; 58: 891–4. 42. Wong J, Choi TK. Recurrent pyogenic cholangitis. Prog Clin Biol Res 1984; 152: 175–92. 43. Choi TK, Wong J. Recurrent pyogenic cholangitis. In: Schwartz SI, Ellis H, Husser WC, eds. Maingot’s Abdominal Operations. Volume 2, 9th edn. New Jersey: Prentice-Hall International, 1990: 1519–31. 44. Wei WI, Choi TK, Wong J, et al. Bronchobiliary fistula due to stones in the biliary tree: report of two cases. World J Surg 1982; 6: 782–5. 45. Chou ST, Chan CW. Mucin-producing cholangiocarcinoma: an autopsy study in Hong Kong. Pathology 1976; 8: 321–8. 46. Chen MF, Jan YY, Wang CS, et al. A reappraisal of cholangiocarcinoma in patient with hepatolithiasis. Cancer 1993; 71: 2461–5. 47. Lai KS, McFadzean AJS, Young RTT. Microemboli pulmonary hypertension in pyogenic cholangitis. Br Med J 1968; 1: 22–4. 48. Jeng KS, Shih SC, Chiang HJ, et al. Secondary biliary cirrhosis. A limiting factor in the treatment of hepatolithiasis. Arch Surg 1989; 124: 1301–5. 49. Fan ST, Choi TK, Lo CM, et al. Treatment of hepatolithiasis: improvement of result by a systemic approach. Surgery 1991; 109: 474–80.

252

28

Liver abscess: amebic, pyogenic, and fungal Purvi Y. Parikh and Henry A. Pitt
Pathogenesis Infection occurs after fecal–oral transmission of E. histolytica, which exists in either an immobile cyst form or the invasive trophozoite form. The main sources of infection are from asymptomatic carriers who transmit the cysts from water to vegetables contaminated with feces, food contaminated by fertilizers, or the hands of infected food handlers (9). The cystic form is swallowed and passes unaffected through the stomach into the intestine where the outer cyst wall is then digested by pancreatic enzymes. The trophozoite form is released into the intestine where it lives and multiplies. In most patients, the trophozoite asymptomatically colonizes the intestine, but some patients may develop amebic dysentery. The trophozoite invades through the intestinal mucosa, enters the mesenteric venules and lymphatics, travels to hepatic venules by the portal vein, and aggregates in the liver parenchyma (7). Accumulation of enough trophozoites leads to thrombosis, and necrosis and formation of an abscess. The outer rim of the abscess wall is infested by E. histolytica. The liquefied hepatic parenchyma has the appearance of “anchovy paste” (10). More than 80% of patients with amebic liver abscess have excess alcohol intake and thus a vulnerable liver to Entamoeba (11). Higher rates of tissue invasion are found in patients with Human Immunodeficiency Virus (HIV), splenectomy, and corticosteroid administration. Diagnosis The clinical presentation of amebic liver abscess is significantly different from pyogenic (Table 28.1). More than 90% of cases occur in men who usually live in or have a travel history to an endemic area in the last 2 to 5 months (5). Typical symptoms include fever, chills, anorexia, right upper quadrant pain, abdominal tenderness, and hepatomegaly with point tenderness over the liver or subcostal region. Diaphragmatic involvement causes right-sided pleural pain or pain referred to the shoulder (12). Patients with amebic liver abscess rarely have concurrent amebic dysentery, but 10% to 35% of patients have gastrointestinal symptoms that include nausea, vomiting, abdominal cramping, or distention (12). Jaundice, septic shock or a palpable mass may rarely be seen. Patients with an amebic liver abscess usually have a mild leukocytosis without eosinophilia (8). Mild anemia with moderate elevation of alkaline phosphatase and other liver function tests also may be present. The most common abnormality is an increased prothrombin time (12,13). Young males who are alcoholics may have normal or increased hemoglobin. If concurrent colitis is present, the wet mount prep will contain trophozoites 70% of the time if three separate stool samples are examined (11). If the patient has no colitis and a solitary liver abscess, stool samples are positive in 40% to 50% of cases (6). The definitive diagnosis of amebic liver abscess is by the detection of E. histolytica trophozoites in the abscess and by

introduction
Hepatic abscess is an uncommon disorder that continues to be a challenge to identify and treat. The three major types of hepatic abscesses are (i) pyogenic, most often polymicrobial, due to aerobic and anaerobic bacteria, (ii) amebic, due to Entamoeba histolytica, and (iii) fungal, principally due to Candida species. Pyogenic hepatic abscesses are seen most commonly in temperate climates, and in the United States they account for approximately 80% of the cases. Amebic abscesses account for about 10% of U.S. patients but are relatively more common in semitropical and tropical climates (1). Fungal abscesses are seen in 10% of U.S. cases and are increasing in frequency secondary to more immunocompromised patients as well as those having invasive hepatic procedures with prolonged antibiotic exposure (2,3). This chapter focuses on the pathogens, diagnosis, and current management of liver abscesses.

amebic abscess
Amebic liver abscess was not recognized as a separate entity until the nineteenth century. Several European investigators had suggested a relationship between dysentery and liver abscess; however, in 1875 Feder Losch discovered that Entamoeba histolytica was the causal agent. In 1890, Sir William Osler described the first North American case of amebic liver abscess when he found amoeba in a patient’s liver and stool sample (4). Open surgical drainage was the treatment of choice of amebic abscess until the 1930s, when aspiration and emetine became the standard of therapy. During this time, mortality decreased remarkedly from 57% to 14% by using therapeutic aspiration combined with amebicidal therapy (5). Further advancement in management happened with the introduction of serologic tests, improved radiologic imaging, and new amebicidal agents. Epidemiology Amebiasis is a widespread parasitic disease that affects people in developed and underdeveloped countries in tropical climates. Mexico, India, East and South Africa as well as Central and South America have the highest epidemic activity of E. histolytica (6). Worldwide, an estimated 500 million people are carriers of E. histolytica, and 50 million people worldwide develop amebic colitis or amebic liver abscesses resulting in 50,000 to 100,000 deaths each year (7). High-risk groups in the United States include immigrants, tourists who travel to endemic areas, institutionalized people, and the homosexual population. Amebic liver abscess is much more common in men, with a male to female ratio of 10:1. Low socio-economic status and unsanitary conditions are significant risk factors. Other associated risk factors include patients with heavy alcohol consumption, impaired immunity, malnutrition, and chronic infections (8).

253

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
finding serum antibodies to the amoeba. Serum antibodies are positive in 85% of patients with invasive colitis, and 99% of patients with liver abscesses (14). However, in endemic countries positive serologies can also be found in asymptomatic carriers. Biopsies of the abscess wall also can reveal trophozoites with periodic acid-Schiff stain (11). Chest radiographs typically show elevation of the right dome of the diaphragm, pleural effusion, and atelectasis. A plain abdominal film will usually show an enlarged hepatic shadow and may demonstrate air in the biliary tree if a communication to the pleural cavity or hollow viscous is present. However, ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI) are all excellent at detecting and characterizing hepatic abscesses (13). Ultrasonography is simple, inexpensive, and quick to perform with a diagnostic accuracy of 90% (15). In an acute setting, CT scan does not add to the diagnostic accuracy of US unless it is needed for evaluation of rupture of abscesses. However, the CT scan does show better delineation of other organs in relationship with the liver (13). MRI is not superior to CT for diagnosis of amebic liver abscess. A nuclear hepatic scan with gallium may differentiate a “cold” amebic abscess from a “hot” pyogenic abscess because of the presence of neutrophils (16). Performing a diagnostic aspiration is usually unnecessary if there is a high clinical suspicion of an amebic liver abscess (13). Treatment Treatment for uncomplicated amebic liver abscess is generally amecidal drugs. Selective patients may require further treatment such as aspiration and percutaneous or surgical drainage because diagnosis is questionable or when complications occur. The mainstay of therapy is metronidazole that is highly effective against both intestinal and extraintestinal infection. The standard dose is 500 mg IV q6 hr with oral administration of 750 mg q8 hr for a total of 7 to 10 days (6). Most patients have rapid clinical improvement in 72 hours, and 90% are cured with metronidazole. Patients who have a persistent abscess and do not respond to metronidazole within 5 days may require the addition of chloroquine (17). Approximately 10% of patients have a recurrence of their abscesses if the intestinal colonization is not treated (18). The majority of patients do not require percutaneous aspiration of the amebic abscess because they respond to amebicidal therapy. Blessmann and colleagues reported that therapeutic aspiration had minor benefit and was insufficient to justify routine needle aspirations (19). A 2009 Cochrane Review also found that simple aspiration of uncomplicated amebic liver abscesses did not help patients (20). However, simple aspiration may be beneficial in patients with failure to respond to drug therapy within 5 to 7 days. Aspiration should also be considered in patients with abscesses greater than 5 cm in size because of the risk of rupture or in patients with abscesses in the left hepatic lobe that have increased frequency of peritoneal leak or rupture into the pericardium and a higher mortality (Fig. 28.1) (21). Aspiration may be indicated if a pyogenic cause needs to be ruled out or for treatment of pulmonary, peritoneal, and pericardial complications. The procedure can be performed safely with US- or CT-guided aspiration. The fluid of an amebic abscess is odorless, and Gram stain and cultures are negative. Amoeba are recovered

Table 28.1 Comparison of Amebic and Pyogenic Abscess
Features Age (mean) Male: female ratio Abscess number Travel/emigration history Diabetes mellitus Alcohol abuse Jaundice Pruritus Abdominal pain Diarrhea Elevated bilirubin Elevated alkaline phosphate Elevated transaminase Positive blood culture Positive amebic serology Amebic abscess 20–40 years >10:1 1 in ≥80% Yes Uncommon Common Uncommon Uncommon Common Common Uncommon Common Uncommon No Yes Pyogenic abscess >50 years 1:1 ≥1 in 50% No
2

1

Common Common Common Common Uncommon Uncommon Common Common Common

3

8

A 4

B 9

7

Yes No

5

6

Figure 28.1 Paths of extension of amebic liver abscesses located within (A) the right hepatic (labels 1–7) and (B) the left hepatic lobe (8,9).

254

LIVER ABSCESS: AMEBIC, PYOGENIC, AND FUNGAL
in 33% to 90% of aspirates with a higher yield in patients with wall scrapings (13). Open surgical drainage of amebic liver abscess is rarely performed because most patients respond to amecidal drug therapy with or without aspiration. Most experts recommend avoidance of open surgical drainage because of concern about secondary bacterial colonization; however, they agree that surgical drainage is often necessary in patients with rupture into the peritoneal cavity, the pleura, lung or pericardium. Thus, open surgical drainage is reserved for patients with complications from rupture, failure to respond to medical therapy, and inadequate aspiration. Outcomes Complications of amebic liver abscesses occur secondary to rupture of the abscess into the peritoneum, pleural cavity, or pericardium. Ruptured amebic liver abscesses occur in 2% to 17% of patients with mortality between 12% and 50% (18). Free rupture into the peritoneal cavity is rare and occurs in patients with poor nutrition or with simultaneous amebic colitis. Peritonitis may occur secondary to rupture or secondary to necrotizing or perforated amebic colitis. Most often, the ruptured liver abscess adheres to the diaphragm or omentum that tends to wall it off. Thoracic amebiasis causing empyema, bronchohepatic fistulas, and pleuropulmonary diseases is the most common complication followed by pericardial amebiasis manifested by acute pericarditis with tamponade (6). When tamponade develops, aspiration of the pericardium, drainage of the liver abscess and anti-amebic drugs are indicated (18). Severely ill patients with sepsis secondary to amebiasis also may develop cerebral amebiasis and experience seizures (6). The overall mortality rates of patients with amebic liver abscess are from 0% to 18%, and high mortality rates suggest delayed diagnosis, secondary bacterial infection or complications (11). Factors that are associated with a poor prognosis are jaundice, advanced age, encephalopathy, decreased albumin, large abscess cavity, or complications such as rupture into the peritoneum or pericardium (22). Patients will have clinical improvements with amecidal therapy much more rapidly than radiologic resolution. Complete radiologic resolution may take up to 9 months. Follow-up imaging studies in patients with resolution of symptoms after treatment for amebic liver abscess is advised. Epidemiology Pyogenic liver abscess is the most common form of liver abscess accounting for approximately 80% of all liver abscesses in the United States (1). The disease process has significantly increased over the past 30 years, presenting in about 15 cases per 100,000 admissions. Seeto and Rocky reported an incidence nearly double that of earlier reports (22 per 100,000) due to a more aggressive management approach to hepatobiliary and pancreatic diseases (27). The predominant etiology of pyogenic liver abscess has changed from an intra-abdominal septic source to a biliary origin with more reports of cryptogenic liver abscesses in recent years. Advanced imaging techniques, improved antibiotics, and numerous treatment modalities have decreased morbidity and mortality. Pathogenesis The liver is a sterile organ that has Kupffer cells that filter out microorganisms. Abscesses form when the normal liver fails to clear the organisms, and the infection causes parenchymal necrosis. Potential sources of these microorganisms are (i) biliary, secondary to ascending cholangitis or biliary malignancy, (ii) portal, pyemia secondary to appendicitis, diverticulitis, or other intra-abdominal source, (iii) arterial, from generalized septicemia, (iv) direct extension, (v) trauma, and (vi) cryptogenic (Table 28.2). Ascending infection of the biliary tree causing obstruction now accounts for 35% to 40% of all pyogenic liver abscesses (1). The etiology of the biliary obstruction has some geographic differences. In Western countries, 40% of pyogenic liver abscesses have underlying malignant disease (28). In Asian countries, hepatolithiasis and associated biliary strictures predominate as the cause of pyogenic liver abscess (29). Percutaneous or endoscopic manipulation or stenting can introduce infection into the biliary tree causing cholangitis, which has the propensity to cause pyogenic liver abscess. Other causes include Caroli’s disease, invasion of the biliary tree by parasites and/or a previous biliary-enteric anastomosis. In the past, the most common source of pyogenic liver abscess was appendicitis. In Ochsner’s study, appendicitis accounted for 43% of all abscesses seeded via the portal vein (23). Today, even though intestinal pathology is responsible for 20% of all pyogenic liver abscesses, appendicitis accounts for only 2% of all pyogenic liver abscesses. Diverticulitis and perforated colon cancers remain more common causes of pyogenic liver abscesses. Hematogenous spread of infection from sources like endocarditis, dental abscess, and intravenous drug abuse are examples of generalized septicemia that may lead to a pyogenic liver abscess. Transarterial embolization or ablative therapies for liver tumors are also associated with liver abscess formation (30,31). Direct extension of empyema of the gallbladder, perforated gastric or duodenal ulcers, and subphrenic abscess also may cause a pyogenic liver abscess. Liver trauma from penetrating or blunt injury may cause parenchymal necrosis, hemorrhage, or intrahepatic biloma that may become infected and develop into pyogenic liver abscess (32). Hepatic artery thrombosis after liver transplantation or ligation after a pancreatobiliary operation may lead to a liver abscess especially following obstructive jaundice. In some

pyogenic liver abscess
Liver abscess was recognized in ancient times, and the first description is credited to Hippocrates in 4000 BC. No new information was identified until the landmark paper by Ochsner and Debakey in 1938 (23). They described the disease as having a 62% survival rate for patients undergoing surgical drainage. With improvements in antibiotics, the combination of surgical drainage and antibiotics became the standard of care. In 1953, McFadzean and associates first reported percutaneous drainage (24). By the mid-1980s, the development of US and CT scans led to numerous reports of patients successfully treated by percutaneous drainage and antibiotics (25,26).

255

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
patients, no identifiable cause of pyogenic liver abscess can be found; they are labeled cryptogenic abscesses and occur in 10% to 55% of patients (33,34). Cryptogenic abscess tend to be solitary, monomicrobial, and associated with diabetes, immunosuppression, or malignancy. Cryptogenic abscesses usually have a normal bilirubin and no duct dilation on imaging (27). Microbiology With the various routes of infection, identification of microorganisms recovered from blood culture or aspiration of pus from the pyogenic liver abscess is imperative for treatment. Patients with cryptogenic liver abscess are more like to have negative blood cultures. However, liver abscesses from biliary tract disorders are more likely to have positive cultures from blood and pus. The most common organisms cultured are Escherichia coli and Klebsiella pneumoniae (Table 28.3) (35). Other frequently encountered aerobes include enterococci and Streptococci viridians. An abscess secondary to biliary tract disease or originating from a gastrointestinal source is more likely to be polymicrobial with aerobic and anaerobic organisms. Bacteroides is the most common isolated anaerobic organism, followed by Fusobacterium (10). The increased use of indwelling biliary stents has resulted in an increased incidence of Klebsiella, streptococcal, staphylococcal, and pseudomonal species in liver abscesses (28). Fifteen percent of patients have a negative culture that may be attributed to poor anaerobic culture technique or the use of broad-spectrum antibiotics prior to abscess drainage. This factor also is thought to be responsible for the increased incidence of fungal abscesses (6). In patients with Acquired Immune Deficiency Syndrome (AIDS) the most common infecting organism for hepatic abscesses is Mycobacterium tuberculosis (36). Diagnosis The clinical presentation of pyogenic liver abscess is nonspecific and includes malaise, nausea, anorexia, weight loss, headaches, and myalgias. These symptoms may be present for many weeks before the appearance of more specific symptoms such as fever, chills and abdominal pain in the right upper quadrant (RUQ) (Table 28.1). The classic triad of fever, jaundice, and RUQ tenderness is present in less than 10% patients. Diarrhea also occurs in 10% of patients (37). On physical examination, some patients may also have a tender and enlarged abdomen. Abscesses adjacent to the diaphragm may cause pleuritic chest pain, cough, and dyspnea. Septic shock may be present in patients with cholangitis or peritonitis after rupture into the peritoneal cavity. Laboratory investigations demonstrate leukocytosis (70% to 90%), hyperbilirubinemia (50% to 67%), and elevated alkaline phosphatase (80%) in a majority of patients. Many patients may also have anemia, hypoalbuminemia, and prolonged prothrombin time (34,37,38). Patients with pyogenic abscesses are also more likely than those with an amebic abscess to have diabetes mellitus. Table 28.3 Spectrum of Microorganisms That Cause Pyogenic Liver Abscess
Gram-positive aerobes Streptococcus milleri Staphylococcus aureus Enterococcus spp. Others Gram-Negative Aerobes Escherichia coli Klebsiella pneumoniae Pseudomonas aeruginosa Proteus spp. Enterobacter cloacae Citrobacter freundii Others Gram-positive anaerobes Clostridium spp. Peptostreptococcus spp. Gram-negative anaerobes Bacteroides spp. Fusobacterium spp.

Table 28.2 Most Frequent Causes of Pyogenic Liver Abscess
Hepatobiliary Benign Choledocholithiasis Hepatolithiasis Biliary-enteric anastomosis Endoscopic biliary procedure Percutaneous biliary procedures Malignant Biliary Gallbladder Ampulla Head of pancreas Portal Benign Diverticulitis Anorectal abscess Pelvic abscess Post-operative abscess Intestinal perforation Pancreatic abscess Appendicitis Inflammatory bowel disease Malignant Colon cancer Gastric cancer Arterial Endocarditis Ear, throat, nose infection Dental infection Direct extension Empyema of the gallbladder Perforated ulcer Traumatic Benign Open or closed abdominal trauma Transarterial embolization Percutaneous ethanol injection Ablation Cryptogenic

256

LIVER ABSCESS: AMEBIC, PYOGENIC, AND FUNGAL
Imaging studies are essential in establishing the diagnosis of hepatic abscess. Plain films of the abdomen and chest show abnormalities in 50% of patients (39). The films will show rightsided atelectasis, pleural effusion, and hemi-diaphragm on chest x-ray and hepatomegaly, possible portal venous gas, and occassional air/fluid levels. Ultrasound and CT scans have a greater sensitivity in visualizing the liver abscess and providing imageguided therapy. The US has a sensitivity as high as 75% to 95%; however, US has difficulty detecting multiple abscesses or those on the dome of the right lobe (27). A CT scan is more accurate than US in differentiating liver lesions from abscesses and visualizing multiple abscesses throughout the liver (Fig. 28.2). CT also has a sensitivity of 95%, but the advantage of abdominal CT scans is that they may help to identify other intra-abdominal pathology and abscesses as small as 0.5 cm (27). Magnetic resonance imaging is also very accurate but has no advantage over ultrasounds or CT scans in the diagnosis of pyogenic liver abscess. Treatment The principles of treatment for liver pyogenic liver abscesses are to start appropriate antibiotics, drain the pus, and treat the underlying cause, if identified. Although antibiotics alone may be curative, these patients have a higher risk of recurrence and complications such as abscess rupture. Ultrasound and CT scans have allowed for earlier diagnosis and facilitated treatment by percutaneous aspiration and drainage (Fig. 28.2). The combination of antibiotic therapy and percutaneous drainage has become the first line treatment for most pyogenic liver abscesses (Table 28.4). Initial antibiotic therapy should be started to cover gramnegative and gram-positive aerobes and anaerobes. Multiple single antibiotics such as pipacillin/tazobactam, ticarcillin/clavulanate, imipenem/cilastin, or meropenem may be started initially if a biliary source is suspected. Alternative therapies including a third-generation cephalosporin or a fluroquinolone

(A)

(B)

(C)

(D)

Figure 28.2 (A) CT demonstrating pyogenic abscess with air/fluid level in segment VI of the liver. (B) CT scan showing residual abscess after initial percutaneous drainage. (C) Further percutaneous drainage catheters. (D) CT demonstrating resolution of abscess.

257

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 28.4 Comparison of Liver Abscess: Diagnosis and Treatment
Liver abscess Amebic Diagnosis Clinical Suspicion US or CT scan Positive amebic serology Treatment Amebicidal therapy Lumenal agent Drainage of complicated abscess Systemic antibiotics Drainage of abscess Systemic antifungal Drainage of abscess Systemic antibiotics for polymicrobial abscess

Pyogenic

Fungal

Clinical Suspicion US or CT scan ± Aspiration Immunocompromised patient US or CT scan ± Aspiration

Abbreviations: CT, computed tomography; US, ultrasound.

with metronidazole may be administered if the bowel is thought to be the source. Parenteral administration of antibiotics for the first 10 to 14 days followed by oral agents for a total of 4 to 6 weeks is the classic recommendation. However, recent studies suggest that only 2 weeks of antibiotic therapy may be required (40). A combination of an aminoglycoside with either an extended spectrum beta-lactam, such as piperacillin, or a third generation cephalosporin is preferred for patients with K. pneumoniae liver abscesses (41). Multiple abscesses <1.5 cm in size and no other surgical disease may be treated with antibiotics alone. However, patients with multiple small abscesses usually have biliary tract disease and require biliary drainage. Matoba and coworkes recently reported that when percutaneous drainage cannot be performed and intravenous administration of antibiotics are ineffective, intermittent hepatic artery infusion therapy may be a useful alternative (42). Clinicians have questioned when the primary treatment for pyogenic liver abscesses should be aspiration alone or placement of a percutaneous catheter for drainage. Giorgio and associates treated 115 patients with percutaneous needle aspiration with a 98% success rate (43). Half the patients required only one aspiration whereas the remainder needed two to three aspirations. In 2004, C.H. Yu and associates also found that aspiration and catheter drainage had a similar mortality, success rate, and hospital stay (44). On the other hand, Rajak and associates performed a prospective randomized controlled trial comparing aspiration and catheter drainage showing a 100% success rate for catheter drainage compared to only 60% for aspiration (45). In summary, needle aspiration is less invasive and avoids all of the complications associated with catheter care. However, recurrence rates and the requirement for surgical intervention may be increased for patients who undergo aspiration alone. Although highly successful, percutaneous catheter drainage fails in approximately 10% of patients. Incomplete or unsuccessful drainage may result from the catheter being too small, highly viscous material, malposition of the catheter, or premature removal of the catheter. Furthermore, contradictions of percutaneous drainage requiring surgical drainage include patients with (i) large

or multiple abscess, (ii) abscesses of unknown etiology, (iii) ascites, (iv) a known intra-abdominal source that requires surgery, and (v) abscesses that require transpleural drainage (1). Operative intervention is recommended when complications occur following percutaneous catheter drainage or failure of non-operative treatment. Open abdominal surgical exploration involves localization of the abscess, identification of additional lesions via ultrasound, evacuation of the abscess, insertion a large-bore drainage tube, and treatment of the inciting pathology (Fig. 28.3). Post-operative lavage of the cavity through the drainage tube has been shown to be advantageous. The role of hepatic resection for the treatment of pyogenic liver abscesses remains controversial. Chou and associates published a series of 483 patients with pyogenic liver abscesses where 27 patients had a liver resection with 1 death (3.7%) (46). They concluded that when single or multiple liver abscesses have caused severe hepatic destruction, resection should be considered. Patients with significant liver atrophy and multiple pyogenic liver abscesses secondary to a long term biliary obstruction from hepatolithiasis or intrahepatic biliary stricture may also be best treated by hepatic resection (46). Outcomes Pyogenic liver abscess may be fatal if left untreated. Forty percent of patients with pyogenic liver abscesses develop complications that include pleural effusions, empyema, pneumonia, and generalized sepsis. Pyogenic liver abscesses may also cause hemobilia and hepatic vein thrombosis. Deaths in patients with pyogenic liver abscesses have been attributed to the intra-peritoneal rupture of the abscess. In one analysis, the causative bacteria in 72% of the cases of failed medical treatment were S. milleri and K. pneumoniae (47). Bacteremia, disseminated intravascular coagulopathy (DIC), septic pulmonary emboli, and acute renal failure are common complications of liver abscesses. Since the introduction of broad-spectrum antibiotics and the advancements in image-guided drainage, the mortality from pyogenic liver abscess has drastically decreased. However, the mortality in patients with multiple abscesses remains higher than in those with a solitary abscess. Liver abscesses are now diagnosed earlier and may be treated non-operatively thus reducing the mortality to less than 20% (46). Seeto and Rockey showed in their series a mortality of only 2% (27). Risk factors for mortality include septic shock, jaundice, hypoalbuminemia, malignancy, disseminated intravascular coagulopathy, APACHE II score >10, multiple abscesses, and intra-peritoneal rupture (48).

fungal liver abscesses
Bacteria and parasites cause the vast majority of hepatic abscesses; however, fungal liver abscesses account for about 10% of hepatic abscesses (2,3). Most monomicrobial fungal abscesses occur in immunocompromised patients including patients receiving chemotherapy and patients with HIV infections. Patients with biliary malignancies, indwelling stents, and frequent courses of antibiotics tend to develop a polymicrobial fungal abscess in conjunction with pyogenic abscesses. Candida albicans and other Candida species are the predominate organism in about 80% of cases (3). Aspergillus, Cryptococcus and mixed infections make up the remaining fungal pathogens.

258

LIVER ABSCESS: AMEBIC, PYOGENIC, AND FUNGAL

Ultrasonic probe Aerobic culture

Site of unilocular abscess Deep multiloculated abscess Gallbladder Liver Loculations within abscess

T-extension

Anaerobic culture Normal liver parenchyma Abscess

Liver

Antibiotic containing saline solution

(A)

(B)

Figure 28.3 (A) Intra-operative ultrasound, aspiration, and anaerobic culture of abscess, and liver incision for pyogenic abscess. (B) Aerobic culture, manual disruption of loculations, and irrigation of the pyogenic abscess. Source: Reproduced from Ref. (51).

Treatment Fungal liver abscesses are treated with intravenous antifungal therapy as well as drainage of the abscess by simple aspiration, percutaneous drainage or open surgical drainage (Table 28.4). Caspofungin is the agent of choice for these patients (49,50). Patients with mixed bacterial and fungal abscesses should also be treated with appropriate systemic antibiotics for the isolated bacteria. An adequate course of caspofungin should be employed as an earlier analysis suggested that inadequate treatment with amphoteracin B was associated with a high mortality. Outcomes Very few patients develop pure fungal abscesses. Most patients have polymicrobial fungal and bacterial abscesses. The overall mortality rate is 50% because patients who do not receive antifungal therapy in early stage develop systemic fungemia (12). In addition, the underlying disease in the majority of these patients is associated with a high mortality. Moreover, the majority of these abscesses are multiple. As a result, patient survival is not influenced by different drainage procedures.

references
1. Pitt HA. Surgical management of hepatic abscess. World J Surg 1990; 14:498–504. 2. Simpfendorfer CH, Henderson JM. Hepatic abscess. In: Cameron J, ed. Current Surgical Therapy, 9th edn. Philadelphia, PA: Elsevier-Mosby, 2008: 317–21. 3. Lipsett PA, Huang CJ, Lillemoe KD, et al. Fungal hepatic abscess: characterization and management. J Gastrointestinal Surg 1997; 1:78–84. 4. Martinez BM. Historical introduction. In: Martinez-Palomo A, ed. Amebiasis. Human Parasitic Diseases, Vol. 2. Amsterdam, Holland: Elsevier; 1986: 1–9. 5. Hughes MA, Petri WA Jr. Amebic liver abscess. Infect Dis Clin N Am 2000; 14: 565–82.

6. Wells CD, Aruqedas M. Amebic liver abscess. S Med J 2004; 97: 673–82. 7. Santi-Rocca J, Rigothier MC, Guillen N. Host-microbe interactions and defense mechanisms in the development of amoebic liver abscesses. Clin Microbiol Rev 2009; 22: 65–75. 8. Tanyuksel M, Petri WA. Laboratory diagnosis of amebiasis. Clin Microbiol Rev 2003; 16:713–29. 9. Salles JM, Moraes LM, Salles MC. Hepatic amebiasis. Braz J Infect Dis 2003; 7:96–110. 10. Leslie DB, Dunn DL. Hepatic abscess. In: Cameron J, ed. Current Surgical Therapy, 8th edn. Philadelphia, PA: Elsevier Mosby; 2004: 298–303. 11. Petri WA Jr, Singh U. Diagnosis and management of amebiasis. Clin Infect Dis 1999; 29: 1117–25. 12. Haque R, Huston CD, Hughes M, et al. Current concepts: amebiasis. N Engl J Med. 2003; 48: 1565–73. 13. Barnes S, Lillemoe K. Liver abscess and hydatid cyst disease. In: Zinner M, Schwartz S, Ellis H, Ashley S, McFadden D, ed. Maingot’s Abdominal Operations, 10th edn. Stamford, CT: Appleton & Lange; 1997: 1513–45. 14. Haque R, Mollah NU, Ali IKM, et al. Diagnosis of amebic liver abscess and intestinal infection with the techLab Entamoeba histolytica II antigen detection and antibody tests. J Clin Microbiol 2000; 38: 3235–9. 15. Sepulveda B, Manzo NTG. Clinical manifestations and diagnosis of amebiasis. In: Martinez-Palomeo A, ed. Amebiasis: Human Parasitic Diseases, No. 2 Amsterdam: Elsevier, 1986; 169–87. 16. Stanley SL Jr. Amebiasis. Lancet 2003; 361: 1025–34. 17. Lodhi S, Sarwari AR, Muzammil M, et al. Features distinguishing amoebic from pyogenic liver abscess: a review of 577 adult cases. Trop Med Int Health 2004; 9: 718–23. 18. Thomas PG, Ravindra KV. Amebiasis and biliary infection. In: Blumgart LH, Fong Y, ed. Surgery of the Liver and Biliary Tract, 3rd edn. London: WB Saunders; 2001: 1147–65. 19. Blessmann J, Binh HD, Hung DM, et al. Treatment of amebic liver abscess with metronidazole alone or in combination with ultrasound-guided needle aspiration: a comparative, prospective and randomized study. Trop Med Int Health 2003; 8: 1030–4. 20. Chavez-Tapia NC, Hernandez-Calleros J, Tellez-Avia FI, et al. Imageguided percutaneous procedure plus metronidazole versus metronidazole alone for uncomplicated amoebic liver abscess (Review). The Cochrane Library 2009 Issue 2.

259

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
21. Weinke T, Grobusch MP, Buthoff W. Amebic liver abscess: rare need for percutaneous treatment modalities. Eur J Med Res 2002; 7:25–9. 22. Sharma MP, Dasarthy S, Verma N, et al. Prognostic markers in amebic liver abscess: a prospective study. Am J Gastro 1996; 91: 2584–8. 23. Ochsner A. Pyogenic abscess of the liver. Am J Surg 1938; 40: 292–353. 24. McFadzean AJS, Chang EPS, Wong CL. Solitary abscess of the liver treated by closed aspiration and antibiotics: a report of the 14 consecutive cases with recovery. Br J Surg 1953; 41: 141–52. 25. Bertel C. Treatment of pyogenic hepatic abscess: surgical versus percutaneous drainage. Arch Surg 1986; 121: 554–8. 26. Gerzof SG, Johnson WC, Robbins AH, et al. Intrahepatic pyogenic abscess: Treatment by percutaneous drainage. Am J Surg 1985; 149: 487–94. 27. Seeto R, Rockey D. Pyogenic liver abscess: changes in etiology, management and outcomes. Medicine 1996; 75: 99–113. 28. Huang CJ, Pitt HA, Lipsett PA, et al. Pyogenic liver abscess: changing trends over 42 years. Ann Surg 1996; 223: 600–9. 29. Kurland JE, Brann OS. Pyogenic and amebic liver abscesses. Curr Gastro Rep 2004; 6: 273–9. 30. Chen C, Chen PJ, Yang PM, et al. Clinical and microbiological features of liver abscess after transarterial embolization for hepatocellular carcinoma. Am J Gastro 1997; 92: 2257–9. 31. de Baère T, Roche A, Amenabar JM, et al. Liver abscess formation after local treatment of liver tumors. Hepatology 1996; 23: 1436–40. 32. Strong RW. Pyogenic liver abscess. In: Blumgart LH, ed. Surgery of the Liver, Biliary Tract and Pancreas, 4th edn. Philadelphia: Saunders/Elsevier, 2007; 1: 927–33. 33. Christians KK, Pitt HA. Hepatic abscess and cystic disease of the liver. In: Zinner M, Ashley S, ed. Maingot’s Abdominal Operations, 11th edn. Stamford, CT: Appleton & Lange 2007; 757–81. 34. Chu KM, Fan ST, Lai ECS, et al. Pyogenic liver abscess: an audit of experience over the past decade. Arch Surg 1996; 131: 148–52. 35. Alvarez Pèrez J, Gonzáles JJ, Baldonedo RF. Clinical course, treatment and multivariate analysis of risk factors for pyogenic liver abscess. Am J Surg 2001; 181: 177–86. 36. Pope IM, Poston GJ. Pyogenic liver abscess. In: Blumgart LH, Fong Y, ed. Surgery of the Liver and Biliary Tract, 3rd edn. London: WB Saunders, 2001: 1135–45. 37. Branum GD, Tyson GS, Branum MA, et al. Hepatic abscess: changes in etiology, diagnosis and management. Ann Surg 1990; 212: 655–62. 38. Leslie DB, Dunn DL. Hepatic abscess. In: Cameron J, ed. Current Surgical Therapy, 8th edn. Philadelphia, PA: Elsevier Mosby 2004; 298–303. 39. Saini S. Imaging of the hepatobiliary tract. New Engl J Med 1997; 336:1889–94. 40. Ng FH, Wong WM, Wong BC, et al. Sequential intravenous/oral antibiotics vs. continuous intravenous antibiotics in the treatment of pyogenic liver abscess. Aliment Pharmacol Ther 2002; 16: 1083–90. 41. Lederman ER, Crum NF. Pyogenic liver abscess with a focus in Klebsiella pneumoniae as a primary pathogen: an emerging disease with unique clinical characteristics. Am J Gastro 2005; 100: 322–31. 42. Matoba M, Tonami H, Kuginuki M, et al. Intermittent hepatic artery antibiotic infusion for pyogenic hepatic abscess. Acta Radiol 2004; 45: 13–17. 43. Giorgio A, Tarantino L, Mariniello N, et al. Pyogenic liver abscess: 13 years of experience in percutaneous needle aspiration with US guidance. Radiology 1995, 195: 122–4. 44. Ch Yu S, Hg Lo R, Kan PS, et al. Single and multiple pyogenic liver abscess: treatment with needle aspiration. Clin Radiol 1997; 52: 912–6. 45. Rajak C, Gupta S, Jain S, et al. Percutaneous treatment of liver abscess: needle aspiration vs. cathetar drainage. AJR 1998; 170: 1035–9. 46. Chou FF, Sheen-Chen SM, Chen YS, et al. Single and multiple pyogenic liver abscesses: clinical course, etiology and results of treatment. World J Surg 1997, 21: 384–9. 47. Lublin M, Bartlett DL, Danforth DN, et al. Hepatic abscesses in patients with chronic granulomatous disease. Ann Surg 2002; 235: 383–91. 48. Christein JD, Kendrick ML, Que FG. What affects mortality after the operative management of hepatic abscess? HPB 2006; 8:175–8. 49. Lai CH, Chen HP, Chen TL. Candidal liver abscesses and cholecystitis in a 37-year-old patient without underlying malignancy. World J Gastro 2005; 11: 1725–7. 50. Pappas PG, Rex JH, Sobel JD, et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004; 38: 161–89. 51. Cameron JL, et al. Liver abscess. In: Atlas of Surgery. Philadelphia, PA: BC Decker; 1990.

260

29 Benign solid tumors of the adult liver
Mark Duxbury and O. James Garden
introduction
As diagnostic radiological imaging techniques have improved and become more widely applied, benign solid lesions of the liver have become an increasingly common incidental finding. These lesions are often detected in asymptomatic patients without significant liver disease and there may be an increasing trend to resect such lesions with the introduction of laparoscopic liver surgery, an issue that remains contentious (1–3). Although the etiology of these lesions is not well understood, benign liver lesions can be classified according to their cellular origin (Table 29.1). Benign solid liver lesions can arise from hepatocytes, biliary epithelium, surrounding mesenchymal structures, or mixtures thereof. Some of these lesions are sufficiently rare that they can be regarded as medical curiosities, although they may present genuine diagnostic and management challenges. A comprehensive understanding of the clinical and pathological features of benign solid liver lesions will facilitate differential diagnosis and allow rational management. The principal issues to be resolved in the management of this group of patients are 1. Confirmation that the lesion is benign. 2. Determining whether the lesion requires surgical treatment. Detailed history addressing relevant risk factors and relevant clinical examination will guide further investigation. While commonly asymptomatic, benign liver lesions may present as surgical emergencies with acute local or systemic symptoms and signs due to hemorrhage, rupture, vascular thrombosis, or necrosis. Commonly, patients present with symptoms that are unrelated to an incidentally detected liver lesion. These patients frequently require further assessment to exclude an alternative cause for their symptoms. Particular aspects of the history may increase the pre-investigation probability of a particular diagnosis, for example, pain, weight loss—malignancy; known gallstones—biliary colic; previous cancer—metastasis; primary sclerosing cholangitis or ulcerative colitis— intrahepatic cholangiocarcinoma; oral contraceptive use: adenoma; cirrhosis, viral hepatitis, inborn errors of metabolism: hepatocellular carcinoma. While liver biochemistry is usually normal, this finding can be helpful in supporting the diagnosis of a benign lesion. Abnormal results may reflect a complication such as hemorrhage, infarction, or the presence of neoplasia. Systemic phenomena, more commonly associated with hepatic malignancies, are exceptional with benign liver lesions. Advances in imaging have meant an increasing majority of benign liver lesions are diagnosed solely on radiological features, particularly in the case of hemangiomata, focal nodular hyperplasia, and confounding focal fatty change. Percutaneous biopsy or surgical resection may be helpful in cases of diagnostic uncertainty, symptomatic lesions, or potentially premalignant conditions. Our institutional practice is to reserve biopsy for non-surgical candidates and only in cases where the biopsy result will influence subsequent management (recommendation strength: D). The role of percutaneous needle biopsy remains controversial, potential problems including bleeding from vascular lesions and a risk of tumor seeding. Needle biopsy and cytology are also subject to sampling errors which can lead to misdiagnosis. Current challenges for clinicians managing this group of patients include










Establishing the diagnosis of a radiologically detected solid liver lesion with a satisfactory degree of certainty, a persisting problem being the differentiation of FNH from adenoma and confirming the nature of focal lesions in the cirrhotic liver. Managing patients with lesions identified incidentally at operation Managing patients following complete or partial resection of a lesion Managing patients with an incidental histological diagnosis following resection or biopsy Determining optimal follow-up of these patients based on the natural history of these lesions

In a recent clinical database review, surgery for benign disease accounted for 5% of patients undergoing resectional liver surgery (4). Advances in preoperative assessment, perioperative care, anesthesia, and surgical technique, including the increasing application of minimally invasive laparoscopic approaches, have the potential to decrease the morbidity and mortality rates associated with resectional liver surgery (5,6). A recent European study reported 87 patients treated laparoscopically with zero mortality and a low-postoperative complication rate (5%) (7). These improvements have altered the analysis of clinical risk versus benefit required when managing a patient with a diagnosis of a solid benign liver lesion but should not increase the prevalence of unnecessary liver resection.

imaging
Ultrasound (US) is often helpful in confirming an intrahepatic mass and will differentiate most cystic from solid parenchymal lesions. However, axial imaging in the form of contrast-enhanced computerized tomography (CT) or magnetic resonance imaging (MRI) is usually required for adequate lesion characterization. Recently developed MRI contrast agents (superparamagnetic iron oxide, SPIO; gadobenate dimeglumine, Gd-BOPTA) have improved the ability to differentiate between solid benign lesions on MRI. Improvements in these modalities mean that angiography is

261

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 29.1 A Classification of Benign Solid Tumors of the Adult Liver
Hepatocellular origin Hepatocellular adenoma Hepatocellular adenomatosis Hepatocellular hyperplasia: Focal nodular hyperplasia Nodular regenerative hyperplasia Regenerative nodule Dysplastic nodules Cholangiocellular origin Biliary hamartoma Biliary adenoma Biliary cystadenoma Congenital hepatic fibrosis Mesenchymal origin Vascular Hemangioma Adult hemangioendothelioma Peliosis hepatis Hereditary hemorrhagic telangiectasia Lymphangioma Adipose Focal fatty changes Lipoma Angiomyolipoma Myelolipoma Smooth muscle Leiomyoma Fibroma Mesothelioma Mixed epithelial and mesenchymal Mesenchymal hamartoma Teratoma Inflammatory pseudotumor Heterotopic tissue Adrenal rests Pancreatic heterotopia

form of hemangioendothelioma with intermediate malignant potential has been reported (9). Hemangiomata are of mesenchymal origin and probably represent a congenital, hamartomatous proliferation of vascular endothelial cells. Approximately 90% are solitary and structurally, 80% are cavernous hemangiomata, the reminder being capillary hemangiomata (10). Over 90% of hemangiomata are less than 5 cm in diameter, which can occasionally make radiological characterization difficult. Current evidence indicates that hemangiomata have no malignant potential.

capillary hemangioma
In comparison to the more common cavernous form, capillary hemangiomata are generally smaller in size and are more frequently multiple. These smaller lesions are a common incidental finding in an asymptomatic patient. Once the diagnosis is established, no further treatment is required (recommendation strength: D).

cavernous hemangioma
Pathology Cavernous hemangiomata occur in up to 8% of autopsy studies. These lesions are the second most common solid hepatic tumor (10). Cavernous hemangiomata occur principally in women (F:M = 5:1) and are more common in the right liver. Large peripheral lesions may become pedunculated. The mean age at diagnosis is 50 years with most being detected between the third and fifth decade. Their prevalence is greatest in those with higher parity (11,12). Although there is no proven association with oral contraceptive use, the relationship remains controversial. Hemangiomata are generally asymptomatic until they reach diameters of over 10 cm. Symptoms include non-specific abdominal pain, pressure symptoms, and fever. Pain may be due to capsular stretching by larger lesions. Jaundice and rupture are rare. Cavernous hemangiomata may reach massive proportions, with reported lesions reaching several kilograms in weight. Giant hemangiomata are defined as those measuring 5 cm and above in diameter (13) and are multiple in 10% of cases (10). Hepatic cavernous hemangiomata may be associated with similar lesions in other organs (14). Other associated conditions, for example, liver cysts, gallbladder disease, gastroduodenal ulcers, or hiatus hernia are reported in 42% (15). In the rare Kasabach–Meritt syndrome intravascular coagulopathy may progress to systemic fibrinolysis and thrombocytopenia. The condition has a reported mortality rate of 20% to 30% (16). Macroscopically, cavernous hemangiomata appear purplish in color and may collapse on sectioning. A plane can generally be identified between the hemangioma and surrounding liver parenchyma. Thrombosis, fibrosis, and calcification are common features (Fig. 29.1). Cavernous hemangiomata may undergo complete fibrous transformation. Hemorrhage is remarkably rare and rupture exceptional. This lesion probably represents a congenital hamartoma with endothelium-lined spaces and fibrous septa being typical microscopic features. Lesions enlarge through ectasia rather than hypertrophy or hyperplasia.

now less commonly a part of initial patient assessment. Positron emission tomography (PET) is undergoing further evaluation as a diagnostic adjunct and laparoscopy is used increasingly to allow direct visualization of liver lesions and can be enhanced by intraoperative US. Nuclear medicine techniques such as single positron emission computerized tomography (SPECT) and tagged RBC scans, although less commonly applied, may facilitate differentiation of hemangioma from hepatocellular carcinoma. 99mTc-sulfur colloid scanning may help differentiate focal nodular hyperplasia from hepatocellular adenoma.

hemangioma ₍adult₎
Hemangiomata are the most common benign solid tumor of the liver. In terms of focal liver lesions they are second only to simple cysts in frequency. Autopsy studies have shown a prevalence of up to 20% (8). Adult hemangioma differs in both presentation and histology from infantile hemangioendothelioma, which is considered elsewhere. An extremely rare adult

262

BENIGN SOLID TUMORS OF THE ADULT LIVER
accepted, except as a temporizing measure to facilitate transfer of an emergency patient. The application of radiotherapy and corticosteroids is also reported but is not well supported by evidence.

focal nodular hyperplasia ₍fnh₎ and fnh-like lesions
Pathology FNH occurs in up to 3% of the population and is the second most common solid benign liver lesion after hemangioma (21,22). Ninety percent of patients are females in their second and third decades, although cases have been reported in males. Less than 10% of FNH is symptomatic and, in contrast to hepatocellular adenomata, hemorrhage and rupture are rare (23). Presence of hemorrhage raises the possibility that a hepatocellular adenoma has been misdiagnosed as FNH. The incidence of FNH is increasing, although this increase appears unrelated to oral contraceptive introduction. The effects of female sex hormones on the natural history of FNH remain controversial. However, discontinuing oral contraceptive use may be associated with a decrease in size of FNH (24,25). There is insufficient evidence to recommend avoiding pregnancy in the presence of FNH. FNH usually forms a firm lobulated lesion in otherwise normal liver (Fig. 29.2) and is more common in the right liver (26). Lesions are typically well circumscribed but lack a definite capsule. Differentiating FNH from other focal liver lesions, particularly hepatocellular adenoma, remains a challenge. The “classical” finding of a central scar (Fig. 29.3) may be absent in approximately 40% of FNH cases (27). Twenty percent of patients exhibit multifocal FNH and up to 20% of FNH coexist with hemangiomata (28). FNH and adenomata may also coexist. A histological diagnosis of FNH does not mean coexisting lesions will also be FNH. Furthermore, co-existing malignancy is reported in 6% of patients with an indeterminate presumed benign liver lesion (29). NH has a microscopic appearance similar to that of cirrhosis, exhibiting regenerative nodules. Normal hepatocytes with intervening connective tissue septa and bile ductules are typical. FNH is thought to be a hyperplastic response to abnormal vascularization, rather than a neoplastic process and is not regarded as having malignant potential (23). Moderate lymphocytic infiltration and mild cholestasis are common. It is unusual for FNH to undergo rapid changes in size and this should call the diagnosis into question. Calcification is a less common finding, occurring in approximately 1.4% of cases (30), and should alert the clinician to the possibility of fibrolamellar hepatocellular carcinoma. Telangiectatic FNH Telangiectatic FNH (TFNH) accounts for 10% to 15% of FNH cases and usually presents with multiple soft nodules without a central scar (31). Intra-lesion bleeding is more common in TFNH than in FNH. TFNH lesions commonly exhibit high levels of angiopoietin 2 gene expression (32). Although the origin of TFNH remains controversial, it is regarded by some as a neoplastic lesion and may in fact represent a variant of HA rather than a subtype of FNH.

Figure 29.1 Typical macroscopic appearances of hemangioma on sectioning of resection specimen.

Imaging Features Hemangiomata are usually hyperechoic on US, although US is often inadequate for differentiating between solid liver lesions. Color-flow Doppler shows peripheral vessel filling. Axial imaging with CT or MRI is usually sufficient to confirm the diagnosis. The principle challenge in managing small (<3 cm) hemangiomata is differentiating them from other lesions, particularly in cases with atypical enhancement and significant arterio-venous shunting. Lesions tend to be hypodense on non-contrast CT and show peripheral followed by central enhancement. Isoenhancement with the arteries is typical. Delayed scans show persisting contrast enhancement and features such as corkscrewing and “cotton wool” appearance reflect the abnormal vessels within the lesion. Globular enhancement isodense with the aorta has been shown to be 67% sensitive and 100% specific in differentiating hemangiomata from metastases (17). Hemangiomata are hypointense on T1-weighted sequences. They appear very bright on T2-weighted sequences (the “light bulb sign”). Peripheral nodular enhancement on dynamic gadolinium-enhanced images and a non-intact ring appearance immediately after contrast with centripetal enhancement (maximal at 90s) are typical. Hemangiomata do not take up SPIO or manganese dipyridoxyl ethylenediamine diacetate (bis)phosphate (Mn-DPDP) as they do not contain Kupffer cells or hepatocytes. 99mTc-SPECT is effective in assessing hemangiomata but appears inferior to MRI (18). Management Biopsy of hemangiomata is generally contraindicated given the risk of hemorrhage and samples obtained from hemangiomata may resemble fibrosis. Resection may be justified rarely for symptomatic lesions and in cases where the diagnosis cannot be established despite investigation. Surgical resection or enucleation remains the principal effective therapies (19). Use of liver transplantation has been reported for unresectable disease in association with Kasabach–Merritt (20). The use of arterial ligation and embolization has not been widely

263

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 29.2 Characteristic lobulated appearance of focal nodular hyperplasia.

Figure 29.3 Classical central scar of focal nodular hyperplasia.

FNH-Like lesions FNH-like lesions have close histological similarity to FNH. Although due to a variety of processes, the common etiological factor appears to be abnormal liver vascularity, specifically increased arterial flow and decreased venous flow. Investigators have noted that FNH and hepatic adenomata occur more often in patients who have coexisting vascular tumors, portal venous thrombosis or occlusion, and those with significant portohepatic venous shunts (33,34). Other conditions associated with FNH-like lesions include Budd–Chiari syndrome, hereditary hemorrhagic telangectasia, and congenital hepatic fibrosis. FNH-like lesions have a tendency to increase in number or size, unlike “classical” FNH. Imaging Features Radiological diagnosis of FNH can be challenging and a multimodality approach is often required. CT has a reported 82% sensitivity and 97% specificity (35). A well-defined hypervascular lesion with a single central artery and spoke wheel centrifugal vessel filling is typical. FNH and FNH-like lesions are hyperintense T1-weighted MRI and hypointense on T2-weighted series. Strong arterial enhancement is often observed. MRI has a reported sensitivity of 70% and a specificity of 98% (36). In a recent study, differentiation between HA and FNH was not possible on the basis of precontrast or dynamic phase images alone. However, images acquired 1 to 3 hours after gadobenate dimeglumine enhancement, allowed differentiation between FNH and HA/adenomatosis with a specificity, PPV, NPV, and overall accuracy of 96.9%, 100%, 100%, 96.4%, and 98.3%, respectively (37). Management Management of patients with FNH and FNH-like lesions depends on the level of certainty of diagnosis. Once the diagnosis of FNH is established with confidence, no further intervention is required. In cases of diagnostic uncertainty, laparoscopic or open biopsy may be helpful and may be preferable to percutaneous needle biopsy. Although observation may be more appropriate for some patients, resection or

enucleation is preferred in most cases if surgery can be achieved safely (recommendation strength: D).

hepatocellular adenoma
Pathology Hepatocellular adenoma is usually identified as a solitary lesion in an otherwise normal liver. Adenomata are well defined but lack a capsule. Approximately 90% occur in women, most being diagnosed in the third to fifth decade. The lesions occur more commonly in the right liver. Adenomata are significantly less common than both hemangiomata and FNH. Ninety percent of patients report oral contraceptive use. The incidence is 3–4/100,000/year if oral contraceptive use has exceeded 2 years and their diagnosis has increased following the introduction of the oral contraceptive (38). Higher dose and longer duration of oral contraceptive use appear to promote adenoma development. Anabolic steroid use is also a risk factor for the development of a hepatocellular adenoma. Patients may present with pain due to hemorrhage into or rupture of an adenoma. The risk of bleeding is increased in larger and rapidly growing adenomata, and in patients using OCP, particularly when use is prolonged (39). Hepatocellular adenomata are associated with abnormalities of carbohydrate metabolism. Adenomata are observed more frequently in glycogen storage disease type 1 (glucose-6phosphatase deficiency), type 3 (glycogen debrancher deficiency), galactosemia, and iron overload. In these patients, adenomata develop at an earlier age and tend to show a male preponderance (2:1) (40,41). Sectioning reveals a yellow or pale brown tumor with surrounding normal tissue and a variable degree of encapsulation (Fig. 29.4). There may be evidence of hemorrhage (Fig. 29.5). Adenomata are generally uniform masses consisting of benignappearing hepatocytes without ducts or portal triads. Hepatocytes are pale due to increased levels of glycogen or fat. Venous lakes (peliosis hepatis, vide infra) may be present. Initially, Kupffer cells were thought to be absent from adenomata but molecular techniques have confirmed their presence in these lesions. However, adenomata generally do not sequester 99m Tc-sulfur colloid which is taken up preferentially by Kupffer

264

BENIGN SOLID TUMORS OF THE ADULT LIVER

(A)

(B)

Figure 29.4 (A) Intraoperative image of pedunculated peripheral hepatocellular adenoma. (B) Section through resected hepatocellular adenoma.

PET appears to be useful in differentiating benign from malignant liver lesions (43) and 11C-acetate PET may have additional benefit in detecting HCC. In a study by Ho et al., benign tumors, such as adenomata and hemangiomata, were not 11C-acetate-avid. FNH showed only mild 11C-acetate uptake (44). Management In cases of acute hemorrhage, resuscitation may be combined with hepatic arterial embolization or laparotomy and packing to facilitate transfer to a specialist center. Formal hepatic resection represents optimal treatment for this group of acute patients, although this may be deferred in a stable patient with confined hemorrhage (recommendation strength: D). Discontinuing oral contraceptive use is probably advisable since there have been reports of lesion regression on their cessation, although there is little compelling evidence to support this position for all patients (recommendation strength: D). Although adenomata may regress following discontinuation of the oral contraceptive (45,46), malignant transformation has also been reported despite its discontinuation and in contraceptive-naive patients (47). Adenomata are currently thought to be premalignant with an approximate transformation rate of 10% (48). Transformation should be suspected in adenomata that increase in size, particularly in conjunction with increasing alpha-fetoprotein levels. The risk of malignant transformation is higher in males and in patients with large lesions. Difficulties in differentiating adenomata from FNH or HCC are commonly encountered in young female patients. Because of the abnormal vasculature typical of these lesions, biopsy is associated with a significant risk of hemorrhage and is generally contraindicated. In summary, if a focal liver lesion is suspected to be an adenoma, surgical resection should be considered, especially for symptomatic or large (5 cm or more in diameter) adenomata, provided resection can be accomplished safely (49) (recommendation strength: C).

Figure 29.5 Recent hemorrhage into hepatocellular adenoma.

cells. Adenomata fall into three molecular pathological subgroups: (1) those with inactivated hepatocyte nuclear factor 1-alpha (HNF-1alpha, chromosome 12q)-, (2) those with betacatenin activation, and (3) those with inflammatory changes. These subtypes reportedly display differing characteristics on MRI (42). Imaging Features Hepatocellular adenomata are usually heterogeneous in appearance on US. CT appearances are iso/hypodense precontrast with variable enhancement. There may be evidence of recent hemorrhage or infarction. MRI demonstrates a well-defined fatty lesion which is iso/ hyperintense on T1 sequences and mildly hyperintense on T2 MRI. Gadolinium-enhanced studies show hypervascularity in the arterial phase. Enhancement is often heterogeneous. Angiography demonstrates a peripherally supplied hypervascular lesion with areas of hypovascularity due to hemorrhage or infarction. Isotope scanning may help differentiate hepatocellular adenoma from FNH, an adenoma appearing as a filling defect (“black hole sign”).

hepatocellular adenomatosis
Pathology Hepatocellular adenomatosis is a rare condition, first recognized by Flejou et al. in 1985 (40). The condition is defined

265

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
arbitrarily as the presence of 10 or more adenomata (50). Adenomatosis appears to be a different pathological entity from the solitary adenoma, having a more equal sex distribution and no association with oral contraceptive use (48). Hepatocellular adenomatosis is associated with maturity-onset diabetes of the young (MODY). Adenomatous hyperplasia is often observed between lesions. Intervening cells are smaller with cytoplasm that appears lighter than normal. Mutations in the transcription factor 1 gene (TCF1) occur in MODY (the most common form, termed MODY3) and biallelic inactivation of TCF1 has been reported in hepatic adenomas. In contrast, TCF1 inactivation is not a feature of FNH. Management Patients with hepatic adenomatosis and a TCF1 mutation should be monitored for future onset of diabetes mellitus, if not already diagnosed. Family members of patients with the TCF1 mutation should be considered for genetic screening for such mutations or liver imaging. Liver adenomatosis also carries a risk of malignant transformation, similar to that of the solitary adenoma. Close follow-up in asymptomatic patients with regular liver imaging and serum alpha-fetoprotein estimation is advised to monitor for the potential complications of hemorrhage and malignant transformation (recommendation strength: D). In rare situations, orthotopic liver transplantation has been employed to treat extensive disease that progresses to liver failure, for malignancy, or for unresectable solitary lesions (51,52). Regenerative Changes Regenerative Nodules The cirrhotic liver is defined by the presence of fibrosis and nodules, which include regenerative nodules, dysplastic nodules, and hepatocellular carcinoma. Terms such as macroregenerative nodule and adenomatous hyperplasia have been superseded by a new classification of regenerative nodule, with or without low- or high-grade dysplasia (Terminology of nodular hepatocellular lesions, International Working Party (53)). Currently, agreed nomenclature envisages a progression of carcinogenesis from regenerative nodule to low-grade dysplastic nodule to high-grade dysplastic nodule to small then large hepatocellular carcinoma. Macroscopically, regenerative and dysplastic nodules appear similar although they may differ in color. In addition to the cirrhotic liver, these lesions may develop in the context of Budd–Chiari syndrome. These lesions are generally 0.8 to 3 cm in diameter. Dysplastic nodules may show signs of nuclear atypia, increased cytoplasmic fat or glycogen, and focally decreased reticulin staining is common. Differentiating dysplastic nodules from hepatocellular carcinoma by imaging or histopathology remains problematic. Imaging Features and Management US features are variable. CT may show high attenuation on pre-contrast CT with low attenuation on arterial phase although regenerative nodules in the context of Budd–Chiari are often hypervascular and multiple. MRI: T1 variable, T2 low intensity. Reliable differentiation of regenerative nodules from dysplastic nodules is beyond the capacity of current MRI technology (54), although high-grade dysplastic nodules may increase in size, display high intensity on T1-weighted MRI, and show increased vascularity (55). Any nodule within the cirrhotic liver that increases in size, shows altered signal or increased enhancement warrants correlation with alpha-fetoprotein and early serial imaging and should be regarded as suspicious for hepatocellular carcinoma (recommendation strength: D). Nodular Regenerative Hyperplasia This often asymptomatic condition usually affects patients over 50 years of age and is reported to occur in 2% autopsy livers (56,57). The etiology of NRH remains unclear, although it is associated with a range of lymphoproliferative and rheumatological disorders and may occur post-transplantation. Associations with a range of therapeutic drugs are reported. NRH is a benign diffuse nodular transformation and differs from the regenerative nodules found in cirrhosis and FNH. The hepatic parenchyma is entirely replaced by nodules between 0.1 and 4 cm in diameter. Hepatocyte hyperplasia separated by atrophic parenchyma with reticulin collapse and sinusoidal dilatation is typical. The diagnosis should be considered in patients with portal hypertension and mild liver function abnormalities without cirrhosis on liver biopsy. Portal hypertension and cholestasis may occur due to intrahepatic compression. Laparoscopic or open biopsy is usually required for diagnosis but NRH may be difficult to differentiate from hepatocellular adenoma, which forms a discrete lesion as opposed to the diffuse changes of NRH, on isolated biopsy. Needle biopsy may also not uncommonly yield normal liver tissue. Imaging Features and Management Imaging is often non-specific or normal, although NRH may cause pseudotumor formation. The lesions appear hypo/ isoechoic on US and display variable enhancement on CT. Lesions appear hyperintense on T1-weighted MRI and iso/ hypointense on T2-weighted sequences. NRH may rarely undergo malignant transformation and monitoring is therefore recommended (recommendation strength: D). Treatment involves the management of associated portal hypertension. NRH may result in liver failure that may necessitate liver transplantation. Biliary Hamartoma (Von Meyenburg complex) Biliary hamartomata are benign developmental abnormalities of the liver. Although biliary hamartomata occur in over 5% of the population, they are principally of clinical interest due to the frequency of their submission for intraoperative frozen section to differentiate them from metastases. They are usually less than 5 mm in diameter and multiple lesions are common. Imaging Features and Management Such lesions are often not apparent on preoperative imaging because of their size and US features are non-specific. CT usually shows a hypodense lesion without contrast enhancement. MR features are hypointense on T1-weighted images and strongly hyperintense on T2-weighted MRI. MRI contrast

266

BENIGN SOLID TUMORS OF THE ADULT LIVER
enhancement is variable although rim enhancement may occur due to compression of adjacent parenchyma. Microscopic features include bile ductules with enlarged lumina, inflammatory infiltrate, and fibrosis. Association with adult polycystic disease has been reported. No treatment is required (recommendation strength: D). Biliary Adenoma This benign cholangioma is a generally well defined and often subcapsular lesion, usually less than 1 cm in diameter. Microscopically a proliferation of non-cystic biliary structures with fibrous stroma is observed. The presence of mucin, but not bile, is typical. Although histological differentiation from hamartomata may be challenging, the practical significance of this is limited. Imaging Features and Management Early nodular enhancement, which is prolonged is usual on CT. Bile duct adenomata have no malignant potential but may require excisional biopsy to allow differentiation from the bile duct proliferations found in FNH as well as from some poorly differentiated adenocarcinoma of the biliary tract type. Once diagnosed no treatment is required (recommendation strength: D). with obesity, diabetes, alcoholism, steroids, total parenteral nutrition, chemotherapy, and antiretroviral therapy. Imaging Features and Management These fatty lesions are hyperechoic on US and hypodense on CT. Their fat content results in a hyperintense appearance on T1-weighted MRI. Radiological diagnosis usually suffices. If multiple areas of focal fatty change occur, pseudotumoral steatosis may result which can be mistaken for other tumors, and may require confirmatory biopsy. Angiomyolipoma This rare tumor consisting of fat, epithelioid, and smooth muscle cells with thick-walled blood vessels and may occur in association with tuberous sclerosis. Ten percent of patients with tuberous sclerosis and renal angiolipomata have either angiomyolipomata or lipomata (63). Women are predominantly affected. While these lesions often become large and may cause compressive effects requiring resection, malignant transformation has been reported, although this is extremely rare (64). Imaging Features and Management The reliability of diagnostic imaging is impaired by the variable proportion of fat (10–90%) and other constituents the angiomyolipomata contain. Lesions with low-fat content may mimic hepatocellular carcinoma. US examination usually demonstrates a hyperechoic lesion. The lesion is hypodense with arterial enhancement and central opacification on CT. Macroaneurysms may be observed. On MRI, the T1-weighted signal is high. Fat suppressed MR may help confirm the diagnosis. Management consists of observation with resection reserved for patients with symptomatic lesions (recommendation strength: D). Lipoma Hepatic lipomata are less common than angiomyolipomata. Lipomata are generally well defined and homogenous. Imaging Features and Management CT shows a hypodense, non-enhancing lesion. MRI features are hyperintense on T1-weighted series and moderately hyperintense on T2-weighted series. Decreased signal with fat suppression is typical. These lesions consist of well-differentiated adipose tissue and require no treatment (recommendation strength: D).

congenital hepatic fibrosis
This rare autosomal recessive fibropolycystic condition results from malformation of the ductal plate and may lie on a continuum with Caroli’s disease, the latter involving larger ducts, in contrast to CHF. This is a multisystem disorder (58), the kidneys being most commonly affected (usually autosomal recessive polycystic kidney disease). CHF is associated with overexpression of transforming growth factor-beta1 and thrombospondin-1 (59). Presentation is usually in adolescence, although presentations in the neonatal period and late adulthood are reported. Intrahepatic portal hypertension is typical, but patients may present a cholangitic or mixed picture. Focal solid liver lesions are reported (60,61). The diagnosis can be often established on axial imaging (62). Management of the manifestations of portal hypertension is the mainstay of treatment. Focal Fatty Variants Steatosis of the liver is usually a diffuse process but fat distribution may be heterogeneous leading to a focal lesion. The diagnosis should be apparent from imaging alone and particularly MRI. Focal Fatty Sparing This diagnosis is usually made radiologically and is most commonly encountered in the posterior aspect of segment 4. These lesions are hypoechoic on US and hyperdense on CT. Hypointensity on T1-weighted MRI is usual. Focal fatty sparing is of no pathological or surgical significance other than requiring differentiation from other focal liver lesions. Focal Fatty Change This abnormality of parenchymal fat distribution usually occurs adjacent to the falciform ligament and may be associated

pseudolipoma
These unusual lesions consist of well-differentiated subcapsular adipose tissue and may occur if an adherent fatty structure becomes detached and incorporated into the liver parenchyma. Pseudolipomata may require differentiation form metastasis, although they require no specific treatment. Peliosis Hepatis Peliosis hepatis refers to the development of multiple abnormal vascular channels with secondary fibrosis. This condition is associated with anabolic steroid use as well as with tuberculosis and, in some cases, other tumors.

267

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Imaging Features and Management The lesion is usually well defined. Arterial enhancement is observed on CT. Peliosis hepatis is hyperintense on T2-weighted MRI and contrast enhancement is observed. Management priorities include withdrawal of potential causative agents and treatment of primary/secondary infection. Malignant transformation is not a feature of peliosis hepatis. Hereditary Hemorrhagic Telangiectasia Hepatic involvement in HHT is recognized (65) and presents as a spectrum from small telangiectasia to large volume lesions, which may be associated with arteriovenous shunting. These lesions are usually asymptomatic and do not generally warrant treatment, although hepato-hepatic fistulas can induce highoutput cardiac failure and hepato-portal fistulas may induce portal hypertension (66). Heterotopic Tissue A choristoma, formed by the abnormal development of tissue of a type not normally found at that site, is an uncommon cause of a focal liver lesion. Intrahepatic heterotopic tissue formation can often be diagnosed radiologically but may require resection or biopsy for definitive diagnosis. Adrenal tissue tends to occur in the subcapsular region and may mimic adenoma. Heterotopic pancreatic tissue is an occasional finding. The origin of these lesions is uncertain although they are probably congenital. Acquired splenic choristoma formation can occur post-splenectomy through autotransplantation of splenic tissue. Nuclear medicine is specific in diagnosing this abnormality. In each of these cases, once the diagnosis of heterotopic tissue is established, no specific treatment is required (recommendation strength: D). Inflammatory Pseudotumor These inflammatory myofibroblastic lesions may exhibit radiological features suggestive of neoplasia. A male preponderance is reported and the average age at diagnosis is 35 years (67). The history is often suggestive of an infective etiology. Pseudotumor formation may occur secondary to thrombosis or infarction, may represent an immune reaction, or may occur as an abscess resolves. Imaging Features and Management The lesion is typically well defined, hyper, or hypoechoic on US hypodense on CT and hyperintense on T2-weighted MRI. Angiography usually confirms a hypervascular lesion. In some cases, pseudotumors can be structurally diffuse and locally destructive. Histopathological examination reveals myofibroblasts, immunocytes, and fibrous tissue. These lesions usually resolve spontaneously and are generally managed non-operatively, although resection may be indicated in some cases to prevent reactivation of infection (recommendation strength: D). Miscellaneous Rare Benign Solid Liver Lesions These lesions are individually very uncommon but as a group represent a significant proportion of benign solid liver lesions. Mesenchymal hamartomas are probably congenital, usually occur in infants but are reported in adults. Histologically a myxoid background with cellular mesenchymal components, hepatocytes, bile ducts, and cystic changes are characteristic. Liver function may be compromised through enlargement and resection may be required. Other rare tumors include myxoma, mesothelioma, leiomyoma, and fibroma. Benign teratoma of the liver is reported although this generally occurs in the pediatric population.

key points
● ●



















Benign liver tumors are increasingly detected due to greater use of improved diagnostic imaging There may be an increasing trend to resect benign lesions with the introduction of laparoscopic surgery Successful management of patients with benign liver lesions requires accurate diagnosis and an understanding of the natural history Most patients with benign liver tumors do not have associated liver disease Benign liver tumors rarely require surgery although, when necessary, centralized expert management minimizes associated morbidity and mortality Inappropriate investigation can lead to morbidity and compromises definitive treatment Hemangioma is the most common solid benign liver tumor Hepatocellular adenomata are rare but are associated with significant complications, are viewed as premalignant and generally warrant resection Hepatocellular adenomata are associated with oral contraceptive use. While controversial, no established links exist for hemangiomata or focal nodular hyperplasia A focal lesion in a cirrhotic liver should be regarded as malignant until proven otherwise Progressive symptoms, an enlarging tumor, complications or raising tumor markers are relative indications for resection

appendix
Recommended Grading of Categories of Evidence Ia: Ib: IIa: IIb: III: evidence from meta-analysis of randomized controlled trials evidence from at least one randomized controlled trial evidence from at least one controlled study without randomization evidence from at least one other type of quasiexperimental study evidence from non-experimental descriptive studies, such as comparative studies, correlation studies and case–control studies evidence from expert committee reports or opinions and/or clinical experience of respected authorities

IV:

268

BENIGN SOLID TUMORS OF THE ADULT LIVER
Recommended Strengths of Management Recommendation A: B: directly based on category I evidence directly based on category II evidence or extrapolated recommendation from category I evidence directly based on category III evidence or extrapolated recommendation from category I or II evidence directly based on category IV evidence or extrapolated recommendation from category I, II, or III evidence
21. Wanless IR, Mawdsley C, Adams R. On the pathogenesis of focal nodular hyperplasia of the liver. Hepatology 1985; 5(6): 1194–200. 22. Karhunen PJ, Penttila A, Liesto K, Mannikko A, Mottonen MM. Occurrence of benign hepatocellular tumors in alcoholic men. Acta Pathol Microbiol Immunol Scand [A] 1986; 94(2): 141–7. 23. Reddy KR, Kligerman S, Levi J, et al. Benign and solid tumors of the liver: relationship to sex, age, size of tumors, and outcome. Am Surg 2001; 67(2): 173–8. 24. Cote C. Regression of focal nodular hyperplasia of the liver after oral contraceptive discontinuation. Clin Nucl Med 1997; 22(9): 587–90. 25. Pain JA, Gimson AE, Williams R, Howard ER. Focal nodular hyperplasia of the liver: results of treatment and options in management. Gut 1991; 32(5): 524–7. 26. Nguyen BN, Flejou JF, Terris B, Belghiti J, Degott C. Focal nodular hyperplasia of the liver: a comprehensive pathologic study of 305 lesions and recognition of new histologic forms. Am J Surg Pathol 1999; 23(12): 1441–54. 27. Kerlin P, Davis GL, McGill DB, et al. Hepatic adenoma and focal nodular hyperplasia: clinical, pathologic, and radiologic features. Gastroenterology 1983; 84(5 Pt 1): 994–1002. 28. Vilgrain V, Uzan F, Brancatelli G, et al. Prevalence of hepatic hemangioma in patients with focal nodular hyperplasia: MR imaging analysis. Radiology 2003; 229(1): 75–79. 29. Belghiti J, Pateron D, Panis Y, et al. Resection of presumed benign liver tumours. Br J Surg 1993; 80(3): 380–383. 30. Caseiro-Alves F, Zins M, Mahfouz A-E, et al. Calcification in focal nodular hyperplasia: a new problem for differentiation from fibrolamellar hepatocellular carcinoma. Radiology 1996; 198(3): 889–892. 31. Bioulac-Sage P, Rebouissou S, Sa CA et al. Clinical, morphologic, and molecular features defining so-called telangiectatic focal nodular hyperplasias of the liver. Gastroenterology 2005; 128(5): 1211–1218. 32. Bioulac-Sage P, Rebouissou S, Sa CA, et al. Clinical, morphologic, and molecular features defining so-called telangiectatic focal nodular hyperplasias of the liver. Gastroenterology 2005; 128(5): 1211–1218. 33. Ishak KG. Benign tumors and pseudotumors of the liver. Appl Pathol 1988; 6(2): 82–104. 34. Grazioli L, Federle MP, Ichikawa T, et al. Liver adenomatosis: clinical, histopathologic, and imaging findings in 15 patients. Radiology 2000; 216(2): 395–402. 35. Weimann A, Ringe B, Klempnauer J, et al. Benign liver tumors: differential diagnosis and indications for surgery. World J Surg 1997; 21(9): 983–990. 36. Cherqui D, Rahmouni A, Charlotte F, et al. Management of focal nodular hyperplasia and hepatocellular adenoma in young women: a series of 41 patients with clinical, radiological, and pathological correlations. Hepatology 1995; 22(6): 1674–1681. 37. Grazioli L, Morana G, Kirchin MA, Schneider G. Accurate differentiation of focal nodular hyperplasia from hepatic adenoma at gadobenate dimeglumine-enhanced MR imaging: prospective study. Radiology 2005; 236(1): 166–177. 38. Baum JK, Bookstein JJ, Holtz F, Klein EW. Possible association between benign hepatomas and oral contraceptives. Lancet 1973; 2(7835): 926–929. 39. Mays ET, Christopherson W. Hepatic tumors induced by sex steroids. Semin Liver Dis 1984; 4(2): 147–157. 40. Flejou JF, Barge J, Menu Y, et al. Liver adenomatosis. An entity distinct from liver adenoma? Gastroenterology 1985; 89(5): 1132–1138. 41. Ronald M, Woodfield J, McCall J, Koea J. Hepatic adenomas in male patients. HPB (Oxford) 2004; 6(1): 25–27. 42. Laumonier H, Bioulac-Sage P, Laurent C, et al. Hepatocellular adenomas: magnetic resonance imaging features as a function of molecular pathological classification. Hepatology 2008; 48(3): 808–818. 43. Delbeke D, Martin WH, Sandler MP, et al. Evaluation of benign vs malignant hepatic lesions with positron emission tomography. Arch Surg 1998; 133(5): 510–515. 44. Ho CL, Yu SC, Yeung DW. 11C-acetate PET imaging in hepatocellular carcinoma and other liver masses. J Nucl Med 2003; 44(2): 213–221. 45. Tesluk H, Lawrie J. Hepatocellular adenoma. Its transformation to carcinoma in a user of oral contraceptives. Arch Pathol Lab Med 1981; 105(6): 296–299.

C:

D:

references
1. Morino M, Morra I, Rosso E, Miglietta C, Garrone C. Laparoscopic vs open hepatic resection: a comparative study. Surg Endosc 2003; 17(12): 1914–8. 2. Koffron AJ, Auffenberg G, Kung R, Abecassis M. Evaluation of 300 minimally invasive liver resections at a single institution: less is more. Ann Surg 2007; 246(3): 385–92. 3. Koffron A, Geller D, Gamblin TC, Abecassis M. Laparoscopic liver surgery: Shifting the management of liver tumors. Hepatology 2006; 44(6): 1694–1700. 4. Dimick JB, Cowan JA Jr., Knol JA, Upchurch GR Jr. Hepatic resection in the United States: indications, outcomes, and hospital procedural volumes from a nationally representative database. Arch Surg 2003; 138(2): 185–91. 5. Buell JF, Thomas MT, Rudich S, et al. Experience with more than 500 minimally invasive hepatic procedures. Ann Surg 2008; 248(3): 475–86. 6. Polignano FM, Quyn AJ, de Figueiredo RS, et al. Laparoscopic versus open liver segmentectomy: prospective, case-matched, intention-to-treat analysis of clinical outcomes and cost effectiveness. Surg Endosc 2008; 22(12): 2564–70. 7. Descottes B, Glineur D, Lachachi F, et al. Laparoscopic liver resection of benign liver tumors. Surg Endosc 2003; 17(1): 23–30. 8. Karhunen PJ. Benign hepatic tumours and tumour like conditions in men. J Clin Pathol 1986; 39(2): 183–8. 9. Frider B, Bruno A, Selser J, et al. Kasabach-Merrit syndrome and adult hepatic epithelioid hemangioendothelioma an unusual association. J Hepatol 2005; 42(2): 282–3. 10. Ishak KG, Rabin L. Benign tumors of the liver. Med Clin North Am 1975; 59(4): 995–1013. 11. Sewell JH, Weiss K. Spontaneous rupture of hemangioma of the liver. A review of the literature and presentation of illustrative case. Arch Surg 1961; 83: 729–33. 12. Schwartz SI, Husser WC. Cavernous hemangioma of the liver. A single institution report of 16 resections. Ann Surg 1987; 205(5): 456–465. 13. Adam YG, Huvos AG, Fortner JG. Giant hemangiomas of the liver. Ann Surg 1970; 172(2): 239–245. 14. Erdogan B, Sen O, Aydin VM, et al. Multi-organ cavernous hemangiomas: case report. Neurol Res 2003; 25(1): 92–94. 15. Farges O, Daradkeh S, Bismuth H. Cavernous hemangiomas of the liver: are there any indications for resection? World J Surg 1995; 19(1): 19–24. 16. el Dessouky M, Azmy AF, Raine PA, Young DG. Kasabach-Merritt syndrome. J Pediatr Surg 1988; 23(2): 109–11. 17. Leslie DF, Johnson CD, Johnson CM, Ilstrup DM, Harmsen WS. Distinction between cavernous hemangiomas of the liver and hepatic metastases on CT: value of contrast enhancement patterns. AJR 1995; 164(3): 625–9. 18. Krause T, Hauenstein K, Studier-Fischer B, Schuemichen C, Moser E. Improved evaluation of technetium–99m-red blood cell SPECT in hemangioma of the liver. J Nucl Med 1993; 34(3): 375–80. 19. Lerner SM, Hiatt JR, Salamandra J, et al. Giant cavernous liver hemangiomas: effect of operative approach on outcome. Arch Surg 2004; 139(8): 818–21. 20. Longeville JH, de la HP, Dolan P, et al. Treatment of a giant haemangioma of the liver with Kasabach-Merritt syndrome by orthotopic liver transplant a case report. HPB Surg 1997; 10(3): 159–62.

269

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
46. Steinbrecher UP, Lisbona R, Huang SN, Mishkin S. Complete regression of hepatocellular adenoma after withdrawal of oral contraceptives. Dig Dis Sci 1981; 26(11): 1045–1050. 47. Colovic R, Grubor N, Micev M, Radak V. Hepatocellular adenoma with malignant alteration. Hepatogastroenterology 2007; 54(74): 386–388. 48. Barthelmes L, Tait IS. Liver cell adenoma and liver cell adenomatosis. HPB (Oxford) 2005; 7(3): 186–196. 49. Cho SW, Marsh JW, Steel J, et al. Surgical management of hepatocellular adenoma: take it or leave it? Ann Surg Oncol 2008; 15(10): 2795–2803. 50. Chiche L, Dao T, Salame E, et al. Liver adenomatosis: reappraisal, diagnosis, and surgical management: eight new cases and review of the literature. Ann Surg 2000; 231(1): 74–81. 51. Weimann A, Ringe B, Klempnauer J, et al. Benign liver tumors: differential diagnosis and indications for surgery. World J Surg 1997; 21(9): 983–990. 52. Tepetes K, Selby R, Webb M, et al. Orthotopic liver transplantation for benign hepatic neoplasms. Arch Surg 1995; 130(2): 153–156. 53. Terminology of nodular hepatocellular lesions. International Working Party. Hepatology 1995; 22(3): 983–993. 54. Efremidis SC, Hytiroglou P. The multistep process of hepatocarcinogenesis in cirrhosis with imaging correlation. Eur Radiol 2002; 12(4): 753–764. 55. Earls JP, Theise ND, Weinreb JC, et al. Dysplastic nodules and hepatocellular carcinoma: thin-section MR imaging of explanted cirrhotic livers with pathologic correlation. Radiology 1996; 201(1): 207–214. 56. Wanless IR. Micronodular transformation (nodular regenerative hyperplasia) of the liver: a report of 64 cases among 2,500 autopsies and a new classification of benign hepatocellular nodules. Hepatology 1990; 11(5): 787–797. 57. Nakanuma Y. Nodular regenerative hyperplasia of the liver: retrospective survey in autopsy series. J Clin Gastroenterol 1990; 12(4): 460–465. 58. Yonem O, Ozkayar N, Balkanci F, et al. Is congenital hepatic fibrosis a pure liver disease? Am J Gastroenterol 2006; 101(6): 1253–1259. 59. El Youssef M, Mu Y, Huang L, Stellmach V, Crawford SE. Increased expression of transforming growth factor-beta1 and thrombospondin-1 in congenital hepatic fibrosis: possible role of the hepatic stellate cell. J Pediatr Gastroenterol Nutr 1999; 28(4): 386–392. 60. Zeitoun D, Brancatelli G, Colombat M, et al. Congenital hepatic fibrosis: CT findings in 18 adults. Radiology 2004; 231(1): 109–116. 61. Hausner RJ, Alexander RW. Localized congenital hepatic fibrosis presenting as an abdominal mass. Hum Pathol 1978; 9(4): 473–476. 62. Akhan O, Karaosmanoglu AD, Ergen B. Imaging findings in congenital hepatic fibrosis. Eur J Radiol 2007; 61(1): 18–24. 63. Ros PR. Hepatic angiomyolipoma: is fat in the liver friend or foe? Abdom Imaging 1994; 19(6): 552–553. 64. Deng YF, Lin Q, Zhang SH, et al. Malignant angiomyolipoma in the liver: A case report with pathological and molecular analysis. Pathol Res Pract 2008; 204(12): 911–918. 65. Sabba C, Pompili M. Review article: the hepatic manifestations of hereditary haemorrhagic telangiectasia. Aliment Pharmacol Ther 2008; 28(5): 523–533. 66. Buscarini E, Danesino C, Olivieri C, Lupinacci G, Zambelli A. Liver involvement in hereditary haemorrhagic telangiectasia or Rendu-OslerWeber disease. Dig Liver Dis 2005; 37(9): 635–645. 67. Koea JB, Broadhurst GW, Rodgers MS, McCall JL. Inflammatory pseudotumor of the liver: demographics, diagnosis, and the case for nonoperative management. J Am Coll Surg 2003; 196(2): 226–235.

270

30 Liver trauma
introduction

Timothy G. John, Myrddin Rees, and Fenella K. Welsh
OIS as determined by pre-operative CT can be fallible and instances of both understaging and overstaging by CT have been documented (3). A recent attempt to validate the AAST OIS identified 14,919 isolated hepatic injuries, and 21,532 with associated injuries, in the American College of Surgeons National Trauma Databank between 2000 and 2004 (5). A significant increase in outcomes including mortality, organ-specific operative rate, and hospital charges were associated with increasing OIS grades.

The size and location of the liver account for its high susceptibility to both blunt and penetrating trauma. The type and severity of liver injuries vary greatly and associated organ injuries are common. The leading cause of death is hemorrhage, with road traffic accidents accounting for an ever increasing proportion of blunt liver injuries. It also presents a large target for penetrating injuries to the trunk and these, including gunshot wounds, can be difficult to manage. The intimate connections between the liver capsule, the major hepatic veins, and the retrohepatic inferior vena cava anchor the liver within the upper abdominal cavity and most of the venous return from the lower body flows within these thin-walled veins. Injuries to these vulnerable, high-volume-flow “juxtahepatic” vessels pose a uniquely challenging aspect of severe liver trauma. Dramatic advances in the management of liver trauma have been reported in recent years and algorithms of management continue to evolve. The last two decades have witnessed a major paradigm shift in the management of liver trauma, with Non-Operative Management of Liver Injury (NOMLI) now firmly established as the standard of care for the majority of patients. For those undergoing surgery, the philosophy of damage control at abbreviated laparotomy prevails and the technique of perihepatic packing has come to serve a dominant role. Multidisciplinary interventions are indispensible in managing both the inevitable consequences of NOMLI and as adjuncts to surgical intervention. Unsurprisingly, little or no Grade I–II evidence exists from randomized or controlled studies of liver trauma, and recommendations for management are based on large descriptive studies, including multiinstitutional prospective cohort studies, and the clinical experience of respected authorities in the field.

initial assessment and diagnosis of liver trauma
Initial management of the patient with suspected liver trauma follows Advanced Trauma Life Support (ATLS) principles (6) with rapid identification of life-threatening injuries and aggressive fluid resuscitation. The importance of preventing systemic hypothermia during resuscitation and transfer deserve particular emphasis (7). The concept of the “Focused Assessment for the Sonographic examination of the Trauma patient” (FAST) has gained widespread acceptance in the initial rapid screening of the trauma patient for the presence of hemoperitoneum (8). A landmark study of surgeon-performed FAST in 1540 patients with blunt and penetrating truncal injuries reported high sensitivity (83.3%) and specificity (99.7%) for the technique, most pronounced following blunt injury (9). In this way, the invasive, overly sensitive, and non- organ-specific technique of diagnostic peritoneal lavage has become obsolete in the initial assessment of the trauma patient with suspected liver injury. Immediate laparotomy to control intra-abdominal bleeding remains the standard of care in the persistently hemodynamically unstable patient. The Western Trauma Association’s study of death in the operating room following major trauma at eight North American Level I trauma centers between 1985 and 1992 (10) reported uncontrolled hemorrhage as the primary cause of death in 82% of patients, while delays in transfer to the operating room and inadequate perioperative resuscitation were implicated in half the intra-operative deaths judged to have been preventable. Thus, shocked patients not responding to aggressive fluid resuscitation require immediate operation to attempt arrest of intra-abdominal hemorrhage. Alternatively, and more controversially, it has been suggested that selective hepatic arterial embolization may be effective in arresting hepatic arterial bleeding in the hemodynamically unstable patient, especially those who respond to initial fluid resuscitation, and its role extended as an alternative to surgery (11,12) (Fig. 30.1). Ciraulo and colleagues reported an encouraging experience with “resuscitative” angioembolization in “bridging” operative and non-operative management (13). Seven out of 11 patients with CT-determined Grade IV–V blunt hepatic injuries and requiring continuous resuscitation

classification of liver trauma
The importance of a standardized method of classifying liver trauma is fundamental to meaningful comparisons of treatment outcomes. The modern system of injury grading and standardization of care is based on the central role of computerized tomography (CT) in the assessment of liver trauma. The classification of liver injuries described by Moore and coworkers in 1989 (1), revised in 1994 (2), has now been adopted as the “industry standard” by the American Association for the Surgery of Trauma (AAST) (Table 30.1). Early attempts to validate the AAST Organ Injury Scale (OIS) noted good correlation between operative score severity on one hand and transfusional requirements and methods of operative management necessitated on the other (3). Similar conclusions were drawn from a study of 170 liver injury patients (90% blunt trauma) where blood transfusions, surgical interventions, and liver-related morbidity and mortality all correlated well with the grade of injury (4). However, the liver

271

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 30.1 AAST Liver Injury Scale (1994 Revision) (2)
Gradea I. II. III. Hematoma Laceration Hematoma Laceration Hematoma Injury descriptionb Subcapsular, non-expanding, <10% surface area Capsular tear, non-bleeding, <1 cm parenchymal depth Subcapsular, 10–50% surface area; Intra-parenchymal, <10 cm in diameter 1–3 cm parenchymal depth, <10 cm in length Subcapsular, >50% surface area or expanding Ruptured subcapsular or parenchymal hematoma Intra-parenchymal hematoma >10 cm or expanding >3 cm parenchymal depth Parenchymal disruption involving 25–75% of hepatic lobe or 1–3 Couinaud’s segments within a single lobe Parenchymal disruption involving >75% of hepatic lobe or >3 Couinaud’s segments within a single lobe Juxtahepatic venous injuries; i.e., retrohepatic vena cava / central major hepatic veins Hepatic avulsion

IV. V. VI.
a b

Laceration Laceration Laceration Vascular Vascular

Advance one grade for multiple injuries to the same organ. Based on most accurate assessment at autopsy, laparotomy, or radiological study.

hemoperitoneum (17) and demonstration of associated intraperitoneal and retroperitoneal injuries (Fig. 30.2). Technical refinements such as the development of rapid spiral CT scanners, protocols optimizing vascular contrast enhancement, and CT-scanning suites adjacent to the emergency department and equipped for critical care monitoring have established the central role of CT in the management of all but the most unstable cases of liver trauma (16). Despite concerns regarding the reliability of CT in the detection of hollow vicious injuries (18), the incidence of missed injuries following non-operative management based on clinical and CT findings is reported to be as low as 0.2% (7). Although reliable predictors of failure of non-operative management are lacking, contrast extravasations (“pooling” or a “blush”) merits emphasis as a cardinal sign of active hemorrhage mandating prompt (angiographic and / or surgical) intervention (7,15,16,19,20) (Fig. 30.3). Fang and colleagues reported contrast pooling in eight out of 150 stable patients treated non-operatively, six of whom (75%) developed hemodynamic instability requiring liver-related laparotomy (20).
Figure 30.1 Selective hepatic angiography (right hepatic artery arising from superior mesenteric artery) demonstrates contrast extravasation (arrows). Angioembolization was performed.

were successfully managed by hepatic embolization as definitive therapy. The “alternative” management of hemorrhage from Grade V juxtahepatic venous lacerations by percutaneous hepatic venous stenting has also been reported (14). Clearly much depends on the timing of such intervention and the prompt availability of appropriate expertise (with an operating room on standby) (12,15). Abdominal CT is established as the primary screening modality for the hemodynamically stable patient with suspected blunt liver trauma (16). Detailed cross-sectional imaging of the abdominal organs permits precise delineation of the type and extent of the liver injury, estimation of the volume of

Non-operative Management of Blunt Liver Trauma (NOMLI) Non-operative management of the majority of patients presenting with blunt liver injury (NOMLI) represents a major paradigm shift in liver trauma management during the last two decades (Fig. 30.2). Analysis of 35,510 hepatic injuries documented in the American College of Surgeons National Trauma Databank revealed a highly significant increase in NOMLI from 75% to 87% between 1994 and 2003 (95.1% for blunt liver injuries) (21). Factors heralding this change include (i) the recognition that as many as two-thirds of patients undergoing surgery on the basis of positive diagnostic peritoneal lavage were found to have relatively trivial injuries during non-therapeutic laparotomy (22–24), (ii) the improved imaging with abdominal CT and less concern regarding missed injuries (15,25), (iii) the precedents for successful nonoperative management of solid organ injuries documented in

272

LIVER TRAUMA

Figure 30.2 Grade V blunt liver trauma associated with right renal hematoma in a hemodynamically unstable patient responding to fluid resuscitation and angioembolization.

Figure 30.3 Abdominal CT performed after perihepatic packing for blunt liver trauma showing intra-parenchymal contrast extravasation (arrow). The patient was transferred from the referring hospital and underwent day 1 relaparotomy because of deteriorating LFTs and evidence of the abdominal compartment syndrome.

the pediatric literature (26–30), (iv) a better understanding of the pathophysiology and natural history of the injured and healing liver (31), and (v) the availability of adjunctive interventional techniques to deal effectively with the inevitable consequences of the non-operated liver injury (15). High success rates with NOMLI have been reported. Pachter and Hofstetter’s multi-center study of 404 patients in 1995 reported NOMLI in 98.5% of patients with just two liverrelated deaths (32). A collective review of 16 published series comprising 609 adult patients with liver trauma managed non-operatively between 1988 and 1997 reported success rates of 84% to100%, mean hospital stays of 11.5 to 16.6 days and mean transfusional requirements of 1 to 4 units (19). Others confirm that NOMLI may be undertaken in 85% to 95% of all adult patients with blunt liver trauma (15,23,24,30,33,34). It is generally accepted that the adoption of non-operative management, in place of the pre-1990s practice of liberal operative intervention, has heralded a substantial reduction in the mortality historically, associated with severe blunt liver injuries (7,15). Although it remains difficult to predict in whom NOMLI will fail, increased experience and confidence has lowered the threshold for attempting non-operative treatment to include all hemodynamically stable patients (i.e., systolic blood pressure >90 mmHg) without the signs of peritonitis, regardless of age, injury grade, or associated injuries (35). It should be emphasized that it is hemodynamic status and not the grade of liver injury which determines decision making. The patient typically at risk of failure of NOMLI, most often because of intra-abdominal bleeding, has been characterized as having a high-grade liver injury and remaining dependent on ongoingfluid resuscitation with persisting acidosis (35). Complications of Non-operative Management Contamination of the peritoneal cavity with blood and bile is an inevitable consequence of non-operative management of the fractured liver and both localized peritoneal signs and a

systemic inflammatory response may be expected. The anticipation and management of specific complications is integral to the successful non-operative management of severe liver injuries. Typically these include arterio-venous fistulas, bile leaks, intra- or peri-hepatic abscesses, and vascular-biliary communications (bilhemia and hemobilia). Such complications should be regarded as virtually obligatory consequences of non-operative management (36). This re-defines the philosophy of surgical abstention recognizing that delayed surgery and/or interventional procedures should be deemed an inherent part of the overall management plan rather than treatment failure. Carillo and colleagues reported that 32 patients out of 135 (24%) with severe blunt liver trauma managed non-operatively during 1995 to 1997 developed such complications (15). Strategies employed in their treatment included angioembolization (37%), CT-guided drainage of collections (31%), ERCP (25%), and laparoscopic drainage of collections (7%). A more recent multicenter study of 453 patients with Grades 3 to 5 blunt liver trauma managed by NOMLI at seven urban level 1 American trauma centers between 2000 and 2003 revealed 87 complications in 61 patients (13%)(37). Similarly, these comprised bleeding, biliary complications, abdominal compartment syndrome, and infective complications, which required 86 multimodality interventions. Hemorrhage Ongoing or recurrent bleeding in the non-operatively managed patient is typically recognized by hemodynamic instability with a gradually declining hematocrit and the requirement for repeated transfusions at an early stage. Angioembolization is usually effective for the relatively small proportion of patients who exhibit the signs of early ongoing hemorrhage or late re-bleeding, though this is unusual beyond 3 days postinjury (37). Liver enzyme derangement is common in the

273

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
aftermath of severe blunt liver trauma, though escalating serum transaminases reflect ongoing hepatocellular damage and may prompt further intervention. An algorithm for management of delayed hemorrhage following blunt liver injury instituted by Carillo and colleagues (15,19) is shown in Fig. 30.6. Biliary Complications (Algorithm from Wahl et al. (38)) Clinically significant biliary complications following NOMLI affect less than 5% of patients (7,34,39) and are typically heralded by elevated serum bilirubin and worsening abdominal pain. Bile collections are amenable to percutaneous drainage that, alone, may lead to resolution in 70% of cases (40). Biliary decompression by ERCP, sphincterotomy, and endobiliary stent insertion is indicated if a bile leak persists (15,41,42). Biliary complications tend to occur at a later stage than bleeding (mean 12 days post-injury (37)) and it has been suggested that screening with HIDA scans may increase detection and facilitate earlier intervention, especially in patients who have undergone angioembolization (38). Bile peritonitis may require laparotomy, although laparoscopic peritoneal washout of the old blood and bile with drain insertion has obvious advantages and has become mainstream therapy (15,36–38,43). Biliary stricture formation following intra-hepatic bile duct disruption is a less well understood aspect of the pathophysiology of blunt liver injury and seems to be rare. Disruption of the gall bladder or extra-hepatic bile ducts is also rare and usually follows more severe trauma in association with hepatic parenchymal or pancreatic injuries (44). Interventional procedures are unlikely to suffice following major central biliary injuries and surgery is usually required. Cholecystectomy, operative cholangiography, temporary external drainage, and delayed bilio-enteric reconstruction are the mainstays of treatment. Hemobilia and bilhemia are uncommon consequences of traumatic vascular–biliary communication. Intra-hepatic pseudoaneurysm formation is implicated in hemobilia which is characterized by pain, deranged liver function tests, jaundice, anemia, and overt gastrointestinal bleeding. Angioembolization, with or without ERCP (for evacuation of blood from the biliary tree) fulfill both diagnostic and therapeutic roles in the management of traumatic hemobilia (15,45–48). Bilhemia occurs because of hepatic venous-biliary disruption and is characterized by jaundice with a massive rise in serum bilirubin (>2000 μmol/L) disproportionate to any abnormality of liver enzymes. Successful resolution has been reported following endoscopic biliary decompression, hepatic venous embolization, and/or occlusion of the fistula with selective endobiliary tissue glue injection (49,50). Intra-abdominal Sepsis The prime role of CT in the diagnosis and guided drainage of delayed intra-hepatic abscesses or perihepatic collections following NOMLI is well established (15), although some patients may still require operative drainage. Abscesses typically complicate liver necrosis in patients with high grade liver injury managed non-operatively, particularly following angioembolization (35), and also following hepatic artery ligation in the operated patient where liver resection may be required (37). Abdominal Compartment Syndrome It has been recognized relatively recently that Abdominal Compartment Syndrome (ACS) may complicate NOMLI, as well as the better recognized scenario following laparotomy. Intra-abdominal hypertension (pressure > 25 mmHg) complicating NOMLI has been associated with high volume resuscitative infusions of fluid and blood and in patients managed by angioembolization in particular (51,52). Intra-abdominal pressure monitoring, the recognition of the clinical manifestations of ACS and expedient decompression by laparotomy (35), laparoscopy (51), or percutaneous drainage (52) are required to drain the responsible blood and/or bile-stained intra-abdominal fluid. Consequently, respiratory, renal, and gastrointestinal complications are usually resolved without necessarily compromising the tamponade effect on the underlying liver injury. Follow-up of Liver Injuries In practice, sequential CT scanning is commonly performed, particularly in those with higher grade liver injuries (53). However, there is little evidence to support routine repetition of scans. Cox and colleagues reviewed the outcomes of NOMLI in 530 patients and reported a change of management (angiography or percutaneous drainage) based on repeat CT findings in just three cases in whom clinical signs were anyway apparent (54). Finally, although early mobilization may seem counterintuitive during NOMLI, a large retrospective study recently reported that routine mobilization within 72 hours of admission seemed not to contribute to the risk of delayed hemorrhage (55). Gunshot Wounds to the Liver—The Role of Non-operative Management Liver gunshot wounds (LGSWs) present unique problems because of extensive hepatic parenchymal fragmentation together with a high incidence of associated organ and vascular injuries. Operative management has traditionally been regarded as mandatory. Non-operative management of LGSWs is therefore controversial and its role has tend to be restricted to very carefully selected patients in an environment of intensive monitoring (35,56). The challenge of managing LGSWs nonoperatively was highlighted by Demetriades and colleagues who reported that, of nearly 1000 patients presenting with abdominal gunshot wounds to a level I trauma center, successful non-operative management was accomplished in just 11 out of 16 selected cases (7% of all liver injuries or 21% of isolated liver injuries) (57). Important pre-requisites for non-operative management of LGSWs include: hemodynamic stability without blood transfusion, minimal, localized abdominal signs, and the absence of associated injuries which preclude adequate serial physical examination, and abdominal CT scanning plays a key role in this regard (58,59). Accordingly, a report from Cape Town, South Africa, describes successful non-operative management of LGSWs in 31 out of 33 selected patients regardless of liver injury grade (59). A recent update from the same group cites successful NOMLI for LGSWs in 58 out of 63 patients (92%) without mortality (60).

274

LIVER TRAUMA
Angioembolization plays an important role in this group of patients for treatment of false aneurysms or active bleeding (56,57). It remains to be seen whether diagnostic laparoscopy will be increasingly adopted to identify the isolated non-bleeding LGSW as suggested by some (61–64). Early Decision Making During Laparotomy Arrest of hemorrhage is the priority at initial operation and rapid decision making is required of the surgeon who encounters major liver injury at laparotomy. Most liver injuries are relatively minor and can be dealt with by simple maneuvers such as bimanual compression of the adjacent parenchyma, diathermy or suture ligation of visible bleeding points. Ongoing bleeding which is not easily controlled in this way requires temporary vascular inflow control by the Pringle maneuver, which serves both therapeutic and diagnostic roles. Grade IV–V injuries with hepatic venous/caval lacerations may not respond to inflow occlusion and any attempts to mobilize the liver or inspect the retrohepatic space may be met with profuse venous bleeding. Consideration of perihepatic packing as the mainstay of the Damage Control Laparotomy (DCL) should now occur (65–67). Perihepatic Packing Perihepatic packing is acknowledged as one of the most important factors in reducing the mortality following liver trauma in recent years (7). Temporary resuscitative packing may be distinguished from definitive therapeutic packing. The former aims to achieve hemodynamic stability, allows the surgeon to regain his composure, repair other priority vascular injuries, await more experienced surgical assistance, and/or facilitate transfer to a major trauma or hepatobiliary center for definitive treatment (67,68). Pack tamponade is the key maneuver underpinning the philosophy of “surgical restraint” and “damage control” during abbreviated laparotomy (69). Its judicious use may pre-empt the rapid cascade of events leading to refractory hypotension, dilutional coagulopathy, hypothermia, acidosis, and metabolic failure. A decision can then be taken either to attempt definitive control of bleeding or to proceed with therapeutic packing with abdominal closure and staged reoperation. The timing of the decision to pack is critical and the avoidance of packing as a desperate last resort when all other measures have failed should be emphasized (70). Indeed, a multicenter study identified failure to recognize the value of timely packing as the most common scenario in which patients died from fatal exsanguinations in the operating room (10). Furthermore, initially effective packing may engender a false sense of security when a “pack and peek” sequence develops and it is important to avoid the vicious cycle of repeated packing, resuscitation, unsuccessful attempts at definitive hemostasis, and repeated re-bleeding into shock (71). Perihepatic packing requires careful insertion of large, folded, dry gauze laparotomy packs around the diaphragmatic surfaces of liver and aims to restore its external contours (66,67) (Fig. 30.3). Sufficient packs must be inserted to provide adequate external counterpressure to achieve tamponade (without causing abdominal compartment syndrome), most of this external force being provided by the body wall and rib cage following wound closure. In this regard, Krige and colleagues describe a “six-pack technique”(72). The patient is later returned to the operating room for re-laparotomy and pack removal according to the progress of metabolic recovery. The Cape Town experience supports delay of liver pack removal beyond 48 hours without increased risk of septic complications or bile leaks, whereas early re-look laparotomy performed within 24 hours was associated with re-bleeding (73). A refinement in therapeutic perihepatic packing is the “mesh hepatorrhaphy” technique which employs a synthetic (polyglycolic acid or polygalactin) absorbable mesh and obviates the need for re-laparotomy (74). The technique seems logical for extensive lobar stellate lacerations, but unsurprisingly is ineffective in instances of juxtahepatic or caval injury and does not seem to have achieved widespread acceptance. Similarly, the use of a non-permeable “liver bag” following failure of conventional packing has been described (75). Refractory Bleeding Following Perihepatic Packing Continued bleeding through the packs that ceases with the Pringle manouevre and recurs with its release indicates hepatic arterial bleeding. Selective hepatic artery ligation remains an option although is regarded by some authorities as obsolete (76,77) and of historical interest only (78). Rather, postoperative transfer to the angiography suite for selective angioembolization has been identified as a major advance in this scenario (42,79). Asensio and colleagues reported early angioembolization as an adjunct to surgical intervention in 23 out of 57 survivors (40%) with grades IV–V liver trauma, and the use of angiography in this way was identified as a significant independent predictor of outcome associated with decreased mortality (80). Persistent bleeding despite inflow occlusion suggests retrohepatic caval or hepatic venous injury and critical decisions must be made regarding the next level of intervention and whether to attempt definitive control of parenchymal / vascular bleeding with or without debridement of devascularized parenchyma (Fig. 30.4). Definitive Surgical Procedures in Complex Liver Trauma The latter option requires the use of one or more recognized manouevres. Although these techniques all have their proponents and detractors, they are nevertheless not mutually exclusive. No controlled trials exist to provide an evidence base for the superiority of one particular method over the other and much depends on anecdote, local expertise, and individual experience. Hepatotomy and Selective Vascular Ligation Rapid exposure and selective vascular suture/ligation of deepseated, actively bleeding, intra-hepatic vessels under hepatic inflow control have long been advocated as a mainstay in some centers (7). In the series of 107 patients with Grade III–IV liver injuries reported from New York, finger-fracture hepatotomy technique was successful in controlling hemorrhage in no less than 100 cases (93.5%) for a cumulative mortality of 15% (81). It should be noted, however, that 83% of these patients had penetrating injuries. Similarly, Beal reported the use of

275

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
always achieving complete hemostasis and hemodynamic stability, nevertheless permitted patient transfer and avoided early death from exsanguinations (76). Type B juxtahepatic injuries implicate extra-hepatic venous avulsion with uncontained hemorrhage around the liver, and are likely to require additional operative measures when the severity of hemorrhage defies control by perihepatic packing (86). Such injuries are more likely to require exposure and direct vascular control. Total hepatic vascular isolation, with cross-clamping of both suprahepatic and infrahepatic segments of the IVC (87), may seem a daunting prospect to the uninitiated in the face of heavy bleeding, and the hypovolaemic patient may not tolerate trial caval occlusion without circulatory collapse. Also, reperfusion with gut endotoxins may cause tachyarrhythmias following clamp removal. However, good results have been reported for direct repair of penetrating juxtahepatic IVC injuries when total vascular isolation techniques were employed (88). Khaneja and co-workers reported operative survival in nine out of ten such patients presenting with Grade V stab and gunshot wounds between 1988 and 1996 for an overall survival of 70% (88). Others have reported more modest survival rates of 42% following direct venous repair (89). The concept of atrio-caval shunting (84,85,90) seems logical but in reality outcomes have been poor with >90% mortality and the technique is generally regarded as obsolete (77). Anatomical Hepatic Resection The decision to embark upon an anatomical liver resection for complex liver trauma must be regarded as one of the most difficult and controversial as most authorities emphasize a policy of conservative surgery and damage limitation. Detractors of formal resection cite prohibitive mortalities of 40% to 60% (71,91), and observe its dwindling usage (2% to 4% incidence) in the trauma centers of North America (7). Others recommend limiting hepatic resection to the management of the rare instance of major burst injuries with extensive lobar devitalization (72). The rationale for hepatic resection is nevertheless logical in fulfilling the dual role of eradicating both the site of hemorrhage and source of necrosis (Figs. 30.5–30.7). Although there is a paucity of evidence for anatomical resection in this scenario, the principles of surgical restraint observed in most North American trauma centers have also been challenged by surgeons in Japan (92) and France (93). Data supporting an enhanced role for anatomic resection in severe liver trauma from Brisbane comprised 37 such patients (76). Right hepatectomy was the most commonly performed procedure (27 cases (73%)), and Grade V juxtahepatic venous injuries were encountered in 11 patients (30%). There were no on-table deaths, three post-operative deaths, the overall mortality was 8% and re-exploration was performed in seven instances (three of which were for removal of post resection packs) (76). This experience should be considered in the context of a specialist hepatobiliary team with experience in elective liver resection and transplant techniques. Initial laparotomy had previously occurred at the referring hospital in two thirds of cases, and in this regard the life-saving role of perihepatic packing prior to transfer was emphasized.

Figure 30.4 Laparotomy and perihepatic packing was performed for postangioembolization hemodynamic instability. The patient continued to bleed and re-laparotomy with hepatic segmentectomy 7/8 and repair of a middle hepatic vein laceration was performed.

hepatotomy in 53 out of 121 patients (44%) with complex liver trauma, citing success in 87% of cases (71). The popularity of the technique in large North American trauma centers during the 1980s is evidenced by its reported use in approximately 43% of nearly 3000 collected cases of complex hepatic trauma (71,82,83), and in 28 out of 85 (33%), such cases reported from Cape Town (72). Non-anatomic Resection In the nomenclature of non-anatomic liver resection for trauma, Strong and colleagues emphasized that such procedures are appropriately defined as either partial resections, where devascularized liver peripheral to the fracture line is removed, or resectional debridement, which involves limited removal of non-viable liver bordering the injury (76). Such atypical liver resections may be performed during the index operation after hemostasis has been achieved, or during subsequent staged re-operations following therapeutic perihepatic packing. The limited removal of non-viable tissue in this way has gained popularity as part of the philosophy of surgical restraint in the management of severe liver trauma, as distinct from the more aggressive approach of definitive anatomical hepatic resection. Specific Manouevres for Control of Hemorrhage from Juxtahepatic/Caval Injuries Liver injuries involving the retrohepatic vena cava and hepatic veins are the most difficult and deadly, and are associated with mortality rates of 50% to 80% (81,83–86). Buckman and colleagues have classified two patterns of juxtahepatic injury: Type A intra-parenchymal hepatic venous injuries, and Type B extraparenchymal venous wounds (86). Type A injuries are probably more common, with predominant bleeding through the associated disrupted hepatic parenchyma and capsule. Restoration of containment by perihepatic packing may achieve tamponade of hemorrhage from high flow or low pressure intra-parenchymal veins in such cases. Thus, Beal reported success with packing in 20 patients with Grade 5 venous injuries (71). Similarly, Strong and colleagues observed that this manouevre, while not

276

LIVER TRAUMA

Figure 30.5 Hemodynamically unstable patient with severe blunt liver trauma underwent laparotomy, common hepatic artery ligation, liver suture, perihepatic packing, and was transferred for definitive management. Re-laparotomy and pack removal on day 1 revealed that active bleeding had stopped but the gall bladder was necrotic and there was a major bile leak from the sutured right liver laceration.

Figure 30.7 Abdominal CT performed 2 weeks later shows satisfactory hypertrophy of the remnant left hemiliver and no complications.

tion was still <1%, and the availability of experienced hepatobiliary and liver transplant surgeons at this institution was again a significant factor (95). It remains to be seen whether such results are reproducible widely, or whether the role of major resections in this context will remain the preserve of the surgical virtuoso. For deep penetrating liver injuries, including central bilobar LGSWs, balloon tamponade is an innovation which offers the chance of salvage (96–98). Total hepatectomy is the ultimate strategy in desperate circumstances when hemorrhage from a shattered liver cannot be controlled, or when a Grade VI avulsion injury renders the liver remnant non-viable. However, the logistical and ethical dilemmas presented by this scenario are daunting as the future survival of the patient depends at the very least on the availability of a suitable donor organ. While an increasing number of reports have testified to the feasibility of this approach (99,100), the reality was illustrated by the Hannover experience in which six out of eight such patients died (101). The authors emphasize the importance of the (early) timing of hepatectomy, but stress that this approach can only be justified in exceptional circumstances.

references
1. Moore EE, Shackford SR, Pachter HL, et al. Organ injury scaling: spleen, liver, and kidney. J Trauma 1989; 29: 1664–6. 2. Moore EE, Cogbill TH, Jurkovich GJ, et al. Organ injury scaling: spleen and liver (1994 revision). J Trauma 1995; 38: 323–4. 3. Croce MA, Fabian TC, Kudsk KA, et al. AAST organ injury scale: correlation of CT-graded liver injuries and operative findings. J Trauma 1991; 31: 806–12. 4. Rizoli SB, Brenneman FD, Hanna SS, et al. Classification of liver trauma. HPB Surgery 1996; 9: 235–8. 5. Tinkoff G, Esposito TJ, Reed J, et al. American Association for the Surgery of Trauma Organ Injury Scale I: spleen, liver, and kidney, validation based on the National Trauma Data Bank. J Am Coll Surg 2008; 207: 646–55. 6. Alexander RH, Proctor HJ. Advanced Trauma Life Support Program for Physicians. 5th edn. Chicago: American College of Surgeons, 1993.

Figure 30.6 Formal right hepatectomy was performed and the patient made an uncomplicated post-operative recovery.

Subsequent reports by Tsugawa and colleagues (94), and the University of Pittsburgh group (95), support further the role of hepatic resection in selected patients with severe liver trauma requiring operative intervention. A mortality of 9% and hepatic-related morbidity of 30% were achieved in the latter series. However, the overall rate of formal major hepatic resec-

277

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
7. Pachter HL, Feliciano DV. Complex hepatic injuries. Surg Clin N Am 1996; 76: 763–82. 8. Körner M, Krötz MM, Degenhart C et al. Current role of emergency us in patients with major trauma. Radiographics 2008; 28: 225–44. 9. Rozycki GS, Ballard RB, Feliciano DV, et al. Surgeon-performed ultrasound for the assessment of truncal injuries. Ann Surg1998; 228: 557–67. 10. Hoyt DB, Bulger EM, Knudson MM, et al. Death in the operating room: an analysis of a multi-center experience. J Trauma 1994; 37: 426–32. 11. Hagiwara A, Yukioka T, Ohta S, et al. Nonsurgical management of patients with blunt hepatic injury: efficacy of transcatheter arterial embolization. AJR 1997; 169: 1151–6. 12. Monnin V, Sengel C, Thony F, et al. Place of arterial embolization in severe blunt hepatic trauma: a multidisciplinary approach. Cardiovasc Intervent Radiol 2008; 31: 875–82. 13. Ciraulo DL, Luk S, Palter M, et al. Selective hepatic arterial embolization of grade IV and V blunt hepatic injuries: an extension of resuscitation in the non-operative management of traumatic hepatic injuries. J Trauma 1998; 45: 353–9. 14. Denton JR, Moore EE, Coldwell DM. Multimodality treatment for grade V hepatic injuries: perihepatic packing, arterial embolization, and venous stenting. J Trauma 1997; 42: 964–8. 15. Carrillo EH, Spain DA, Wohltmann CD, et al. Interventional techniques are useful adjuncts in nonoperative management of hepatic injuries. J Trauma 1999; 46: 619–24. 16. Shanmuganathan K, Mirvis SE. CT scan evaluation of blunt hepatic trauma. Radiol Clin N Am 1998; 36: 399–411. 17. Knudson MM, Lim RC, Oakes DD, et al. Nonoperative management of blunt liver injuries: the need for continued surveillance. J Trauma 1990; 30: 1494–1500. 18. Shankar KR, Lloyd DA, Kitteringham L, et al. Oral contrast with computed tomography in the evaluation of blunt abdominal trauma in children. Br J Surg1999; 86: 1073–7. 19. Carrillo EH, Platz A, Miller FB, et al. Non-operative management of blunt hepatic trauma. Br J Surg1998; 85: 461–8. 20. Fang J-F, Chen R-J, Wong Y-C, et al. Pooling of contrast material on computed tomography mandates aggressive management of blunt hepatic injury. Am J Surg 1998; 176: 315–9. 21. Hurtuk M, Reed RLn, Esposito TJ, et al. Trauma surgeons practice what they preach: the NTDB story on solid organ injury management. J Trauma 2006; 61: 243–55. 22. John TG, Greig JD, Johnstone AJ, et al. Liver trauma: a ten year experience. Br J Surg 1992; 79: 1352–6. 23. Meredith JW, Young JS, Bowling J, et al. Nonoperative management of blunt hepatic trauma: the exception or the rule. J Trauma 1994; 36: 529–35. 24. Croce MA, Fabian TC, Menke PG, et al. Nonoperative management of blunt hepatic trauma is the treatment of choice for hemodynamically stable patients. Results of a prospective trial. Ann Surg 1995; 221: 744–55. 25. Brasel KJ, DeLisleC.M., Olson CJ, et al. Trends in the management of hepatic injury. Am J Surg 1997; 174: 674–7. 26. Karp MP, Cooney DR, Pros GA, et al. The non-operative management of pediatric hepatic trauma. J Pediatric Surg 1983; 18: 512. 27. Cywes S, Rode H, Miller AJ. Blunt liver trauma in children: non-operative management. J Pediat Surg 1985; 20: 14. 28. Oldham KT, Guice KS, Ryckman F, et al. Blunt liver injury in childhood: evolution of therapy and current perspective. Surgery 1986; 100: 542–9. 29. Bond SJ, Eichelberger MR, Gotschall CS, et al. Nonoperative management of blunt hepatic and splenic injury in children. Ann Surg 1996; 223: 286–9. 30. Malhotra AK, Fabian TC, Croce MA, et al. Blunt hepatic injury: a paradigm shift from operative to nonoperative management in the 1990s. Ann Surg 2000; 231: 804–13. 31. Richardson JD. Changes in the management of injuries to the liver and spleen. J Am Coll Surg 2005; 200: 648–69. 32. Pachter HL, Hofstetter SR. The current status of nonoperative management of adult blunt hepatic injuries. Am J Surg1995; 169: 442–54. 33. Farnell MB, Spencer MP, Thompson E, et al. Nonoperative management of blunt hepatic trauma in adults. Surgery 1988; 104: 748–56. 34. Christmas AB, Wilson AK, Manning B, et al. Selective management of blunt hepatic injuries including nonoperative management is a safe and effective strategy. Surgery 2005; 138: 606–11. 35. Franklin GA, Casos SR. Current advances in the surgical approach to abdominal trauma. Injury 2007; 37: 1143–56. 36. Carrillo EH, Richardson JD. Delayed surgery and interventional procedures in complex liver injuries. J Trauma 1999; 46: 978. 37. Kozar RA, Moore FA, Cothren CC, et al. Risk factors for hepatic morbidity following non-operative management. Arch Surg 2006; 141: 451–9. 38. Wahl WL, Brandt M-M, Hemmila MR, Arbabi S. Diagnosis and management of bile leaks after blunt liver injury. Surgery 2005; 138: 742–8. 39. Durham RM, Buckley J, Keegan M, et al. Management of blunt hepatic injuries. Am J Surg1992; 164: 477–81. 40. Fingerhut A, Trunkey D. Surgical management of liver injuries in adults – current indications and pitfalls of operative and non-operative policies: a review. Eur J Surg 2000; 166: 676–86. 41. Harrell DJ, Vitale GC, Larson GM. Selective role for endoscopic retrograde cholangiopancreatography in abdominal trauma. Surg Endosc 1998; 12: 400–4. 42. Asensio JA, Demetriades D, Chahwan S, et al. Approach to the management of complex hepatic injuries. J Trauma 2000; 48: 66–9. 43. Franklin GA, Richardson JD, Brown AL, et al. Prevention of bile peritonitis by laparoscopic evacuation and lavage after nonoperative treatment of liver injuries. Am Surg 2007; 73: 611–6. 44. Burgess P, Fulton RL. Gallbladder and extrahepatic biliary duct injury following abdominal trauma. Injury 1992; 23: 413–4. 45. Clancy TE, Warren RL. Endoscopic treatment of biliary colic resulting from hemobilia after nonoperative management of blunt hepatic injury: case report and review of the literature. J Trauma 1997; 43: 527–9. 46. Velmahos GC, Toutouzas K, Radin R, et al. High success with non-operative management of blunt hepatic trauma: the liver is a sturdy organ. Arch Surg 2003; 138: 475–81. 47. Srivastava DN, Sharma S, Pal S, et al. Transcatheter arterial embolization in the management of hemobilia. Abdom Imaging 2006; 31: 439–48. 48. Forlee MV, Krige JEJ, Welman CJ, et al. Hemobilia after penetrating and blunt liver injury: treatment with selective hepatic artery embolisation. Injury 2004; 35: 23–8. 49. Glaser K, Wetscher G, Pointner R, et al. Traumatic bilhemia. Surgery 1994; 116: 24–7. 50. Sattawattamrong Y, Janssens AR, Alleman MJA, et al. Endoscopic treatment of bilhaemia following percutaneous liver biopsy. HPB 1999; 1: 33–5. 51. Chen RJ, Fang JF, Chen MF. Intra-abdominal pressure monitoring as a guideline in the nonoperative management of blunt hepatic trauma. J Trauma 2001; 51: 44–50. 52. Yang EY, Marder SR, Hastings G, Knudson MM. The abdominal compartment syndrome complicating nonoperative management of major blunt liver injuries: recognition and treatment using multimodality therapy. J Trauma 2002; 52: 982–6. 53. Lee SK, Carrillo EH. Advances and changes in the management of liver injuries. Am Surg 2007; 73: 201–6. 54. Cox JC, Fabian TC, Maish GO, et al. Routine follow-up imaging is unnecessary in the management of blunt hepatic injury. J Trauma 2005; 59: 1175–80. 55. London JA, Parry L, Galante J, et al. Safety of early mobilization of patients with blunt solid organ injuries. Arch Surg 2008; 143: 972–6. 56. Moore EE. When is nonoperative management of a gunshot wound to the liver appropriate. J Am Coll Surg 1999; 188: 427–8. 57. Demetriades D, Gomez H, Chahwan S, et al. Gunshot injuries to the liver: the role of selective nonoperative management. J Am Coll Surg 1999; 188: 343–8. 58. Velmahos GC, Constantinou C, Tillou A, et al. Abdominal computed tomographic scan for patients with gunshot wounds to the abdomen selected for nonoperative management. J Trauma 2005; 59: 1155–61. 59. Omoshoro-Jones JAO, Nicol JA, Navsaria PH, et al. Selective non-operative management of liver gunshot injuries. Br J Surg 2005; 92: 890–5. 60. Navsaria PH, Nicol JA, Krige JEJ, et al. Selective non-operative management of liver gunshot wounds. Ann Surg 2009; 249: 653–6. 61. Sosa JL, Markley M, Sleeman D, et al. Laparoscopy in abdominal gunshot wounds. Surg Laparosc Endosc 1993; 3: 417–9. 62. Ivatury RR, Simon RJ, Stahl WM. A critical evaluation of laparoscopy in penetrating abdominal trauma. J Trauma 1993; 34: 822–8.

278

LIVER TRAUMA
63. Simon RJ, Ivatury RR. Current concepts in the use of cavitary endoscopy in the evaluation and treatment of blunt and penetrating truncal injuries. Surg Clin N Am 1995; 75: 157–74. 64. Ditmars ML, Bongard F. Laparoscopy for triage of penetrating trauma: the decision to explore. J Laparoendosc S 1996; 6: 285–91. 65. Feliciano DV, Mattox KL, Burch JM, et al. Packing for control of hepatic haemorrhage. J Trauma 1986; 26: 738–43. 66. Krige JEJ. Perihepatic packing in the management of liver trauma. HPB Surg 1991; 3: 141–4. 67. Krige JEJ, Bornman PC, Terblanche J. Therapeutic perihepatic packing in complex liver trauma. Br J Surg 1992; 79: 43–6. 68. Watson CJE, Calne RY, Padhani AR, et al. Surgical restraint in the management of liver trauma. Br J Surg 1991; 78: 1071–5. 69. Rotondo MF, Schwab CW, McGonigal MD, et al. ‘Damage control’: an approach for improved survival in exsanguinating penetrating abdominal injury. J Trauma 1993; 35: 375–83. 70. Garrison JR, Richardson JD, Hilakos AS, et al. Predicting the need to pack early for severe intra-abdominal hemorrhage. J Trauma 1996; 40: 923–9. 71. Beal SL. Fatal hepatic hemorrhage: an unresolved problem in the management of complex liver injuries. J Trauma 1990; 30: 163–9. 72. Krige JEJ, Bornman PC, Terblanche J. Liver trauma in 446 patients. S Afr J Surg1997; 35: 10–5. 73. Nicol JA, Hommes M, Primrose R, et al. Packing for control of haemorrhage in major liver trauma. World J Surg 2007; 31: 569–74. 74. Reed RL, Merrell RC, Meyers WC, et al. Continuing evolution in the approach to severe liver trauma. Ann Surg 1992; 216: 524–38. 75. Sattler S, Gentilello LM. The liver bag: report of a new technique for treating severe, exsanguinating hepatic injuries. J Trauma 2004; 57: 884–6. 76. Strong RW, Lynch SV, Wall DR, et al. Anatomic resection for severe liver trauma. Surgery 1998; 123: 251–7. 77. Krige JEJ, Nicol JA. Treating major liver injuries. SAJS 2006; 44: 128–30. 78. Mays ET. Hepatic trauma. Curr Probl Surg 1976; 13: 6–73. 79. Fang J-F, Chen R-J, Lin B-C, et al. Blunt hepatic injury: minimal intervention is the policy of treatment. J Trauma 2000; 49: 722–8. 80. Asensio JA, Roldan G, Petrone P, et al. Operative management and outcomes in 103 AAST-OIS grades IV and V complex hepatic injuries: trauma surgeons still need to operate, but angioembolization helps. J Trauma 2003; 54: 647–54. 81. Pachter HL, Spencer FC, Hofstetter SR, et al. Significant trends in the treatment of hepatic trauma. Experience with 411 injuries. Ann Surg 1992; 215: 492–502. 82. Feliciano DV, Mattox KL, Jordan GL, et al. Management of 1000 consecutive cases of hepatic trauma (1979–1984). Ann Surg 1986; 204: 438–42. 83. Cogbill TH, Moore EE, Jurkovich GJ, et al. Severe hepatic trauma: a multicenter experience with 1,335 liver injuries. J Trauma 1988; 28: 1433–8. 84. Rovito PF. Atrial caval shunting in blunt hepatic vascular injury. Ann Surg 1987; 205: 318–21. 85. Burch JM, Feliciano DV, Mattox KL. The atriocaval shunt. Facts and fiction. Ann Surg 1988; 207: 555–68. 86. Buckman RF, Miraliakbari R, Badellino MM. Juxtahepatic venous injuries: a critical review of reported management strategies. J Trauma 2000; 48(5): 978–84. 87. Bismuth H, Castaing D, Garden OJ. Major hepatic resection under total vascular exclusion. Ann Surg 1989; 210: 13–9. 88. Khaneja SC, Pizzi WF, Barie PS, Ahmed N. Management of penetrating juxtahepatic inferior vena cava injuries under total vascular occlusion. J Am Coll Surg 1997; 184: 469–74. 89. Richardson JD, Franklin GA, Lukan JK, et al. Evolution in the management of hepatic trauma: a 25 year perspective. Ann Surg 2000; 232: 324–30. 90. Chen R-J, Fang J-F, Lin B-C, et al. Surgical management of juxtahepatic venous injuries in blunt hepatic trauma. J Trauma 1995; 38: 886–90. 91. Moore FA, Moore EE, Seagraves A. Nonresectional management of major hepatic trauma. An evolving concept. Am J Surg 1985; 150: 725–9. 92. Kasai T, Kobayashi K. Searching for the best operative modality for severe hepatic injuries. Surg Gynecol Obstet 1993; 77: 551–5. 93. Menegaux F, Langlois P, Chigot J-P. Severe blunt trauma of the liver: a study of mortality factors. J Trauma 1993; 35: 865–9. 94. Tsugawa K, Koyanagi N, Hashizume M, et al. Anatomic resection for severe blunt liver trauma in 100 patients. World J Surg 2002; 26: 544–9. 95. Polanco P, Leon S, Pineda J, et al. Hepatic resection in the management of complex injury to the liver. J Trauma 2008; 65: 1264–70. 96. Poggetti RS, Moore EE, Moore FA, et al. Balloon tamponade for bilobar transfixing hepatic gunshot wounds. J Trauma 1992; 33: 694–7. 97. Seligman JY, Egan M. Balloon tamponade: an alternative in the treatment of liver trauma. Am Surg 1997; 63: 1022–3. 98. Demetriades D. Balloon tamponade for bleeding control in penetrating liver injuries. J Trauma 1998; 44: 538–9. 99. Jeng LB, Hsu CH, Wang CS, et al. Emergent liver transplantation to salvage a hepatic avulsion injury with a disrupted suprahepatic vena cava. Arch Surg 1993; 128: 1075–7. 100. Demirbas A, Fragulidis GP, Karatzas T, et al. Role of liver transplantation in the management of liver trauma. Transpl Proc 1997; 29: 2848. 101. Ringe B, Pichlmayr R. Total hepatectomy and liver transplantation: a life-saving procedure in patients with severe hepatic trauma. Br J Surg 1995; 82: 837–9.

279

31

Portal hypertension Michael D. Johnson and J. Michael Henderson
with endothelins (8), norepinephrine, angiotensin II, vasopressin whose levels are elevated in cirrhotic patients plays a role (9). Insufficiency of local vasodilators, such as nitric oxide and carbon monoxide has been implicated, and while levels of Nitric Oxide Synthase (NOS) are normal, NOS activity is depressed, partly due to increased expression of caveolin (10). Splanchnic vasodilation and the development of portosystemic collateral are influenced by Vascular Endothelial Growth Factor (VEGF) and Platelet Derived Growth Factor (PDGF). VEGF receptor inhibition has been shown to prevent the formation of portosystemic collaterals in animal models (11,12) and reverse the adverse hemodynamic consequences of established portal hypertension (13). These neurohumoral changes lead to increased splanchnic hyperemia and a hyperdynamic systemic circulation with a low arterial blood pressure and increased cardiac output.

introduction
Portal hypertension has undergone major changes in understanding its pathophysiology, investigation, and management over the past 3 decades. It has moved from a dominantly surgical syndrome to a predominantly medical one, but in day-today practice includes a broad group of disorders that require a multidisciplinary team for evaluation and management. Basic science has defined the pathophysiology, allowing the introduction of pharmacologic therapies for treating many of the complications of portal hypertension. Technology advances in endoscopic therapies and endovascular stenting have changed the treatment options for variceal bleeding. The coming of age of liver transplantation has dramatically altered the management of patients with end stage liver disease and portal hypertension, and is now the dominant surgical option for suitable patients. This chapter will address these changes and present the evidence supporting management choices for the main clinical presentations of portal hypertension.

clinical manifestations
Variceal Bleeding Variceal bleeding is one of the major complications of portal hypertension requiring medical and surgical therapies (14). Screening endoscopy in cirrhotics shows that patients with more advanced liver disease and a lower platelet count are more likely to have varices (15). Epidemiologic studies have shown linear progression of varices over time (16). All patients with cirrhosis should undergo an initial screening upper endoscopy for varices: if none are present this should be repeated every 2 years. If present, intervention with prophylactic therapy to prevent bleeding (primary prophylaxis) should be considered. Ascites Ascites is a marker of advanced liver disease that develops later than varices. Refractory ascites which does not respond to salt restriction and diuretics and has consistently been linked to increased mortality (17). In addition, refractory ascites is potentially complicated by fluid and electrolyte imbalances, spontaneous bacterial peritonitis, and a higher incidence of abdominal wall hernias (18). Hypersplenism Portal hypertension can lead to splenomegaly and hypersplenism, with anemia, leukopenia, and thrombocytopenia. Blood is sequestered and destroyed in the spleen due to morphologic changes (19). Hypersplenism improves with decompression of the portal hypertension and rarely is splenectomy needed. Partial splenic embolization has been advocated by some (20), but is less effective. Pulmonary Syndromes Hepatopulmonary syndrome (HPS) is characterized by hypoxemia with the development of right-to-left intrapulmonary

etiology
Portal hypertension can be divided into prehepatic, intrahepatic, and posthepatic causes based on the location of obstruction to portal blood flow—Table 31.1. Most of the prehepatic causes have a normal liver which improves overall prognosis for this group. Portal vein thrombosis should prompt a workup for hypercoagulable states (1). Rarely, a procedural induced arteriovenous fistula or functional fistulae associated with Osler Weber Rendu disease (Hereditary hemorrhagic telangiectasia) (2,3), can cause portal hypertension. Intrahepatic portal hypertension is usually secondary to cirrhosis with its multiple causes. The severity of the liver disease is the most important factor in determining prognosis. Intrahepatic presinusoidal obstruction occurs in congenital hepatic fibrosis and schistosomiasis, and liver function is well preserved. Schistosomiasis accounts for 200 million cases of portal hypertension worldwide (4,5). Posthepatic portal hypertension is rare and includes Budd– Chiari Syndrome (BCS) and veno-occlusive disease. BCS results from obstruction of either the hepatic veins or suprahepatic vena cava, and is often associated with a myeloproliferative or hypercoagulable disorder. Outcome is determined by the extent of hepatic vein involvement, the rapidity of development, and the underlying liver function (6,7).

pathophysiology
Portal hypertension occurs when portal venous pressure rises above the normal pressure of 8 mmHg and becomes clinically important above 10 to 12 mmHg. The cascade of events that follow portal flow obstruction are illustrated in Figure 31.1. As portal pressure rises, additional dynamic factors come into play as a result of neurohormonal changes. Vasoconstriction

280

PORTAL HYPERTENSION
shunts and a widened alveolar-arterial oxygen gradient (21). Treatment is with liver transplantation which usually reverses the syndrome with survival rates similar to non-HPS cohorts (22). Pulmonary hypertension is also seen in patients with portal hypertension and carries a more sinister prognosis than HPS. Unless pulmonary arterial pressure can be reduced to normal ranges, liver transplant is contraindicated in these patients. Hepatocellular Carcinoma (HCC) Hepatocellular carcinoma is a common cause of death among patients with cirrhosis and has increased in importance as other causes of mortality have declined (23). The most common conditions leading to HCC are hepatitis B and C and alcoholism. Screening is recommended for patients with cirrhosis using hepatic ultrasound. However effectiveness of this approach has been questioned (24). Early detection of HCC in patients with well-compensated cirrhosis who are eligible for resection or can be prioritized for transplantation is the goal. Radiologic Imaging includes ultrasound as a versatile, low cost, low-risk tool for imaging of the liver parenchyma and the portal and hepatic venous systems. The addition of Doppler allows portal venous velocity assessment. Ultrasound can aid to guide percutaneous liver biopsy and paracentesis, screen for hepatocellular carcinoma, and define vessels. Computed tomography (CT) provides better anatomic detail, especially with newer multidetector array scanners and digital reconstruction of images in different planes and in three dimensions. Magnetic resonance imaging (MRI) is an imaging modality that may be preferred by some over CT scan, and can provide more information about the bile and pancreatic ducts. Visceral angiography and hepatic venography are occasionally indicated. The Hepatic Venous Pressure Gradient (HVPG) is calculated by measuring the hepatic venous wedge pressure with a balloon occlusion catheter minus the free hepatic vein pressure. In patients treated with pharmacotherapy, it has been shown that variceal bleeding rarely occurs when the HVPG is below 10 mmHg and if the HVPG can be reduced to below 12 mmHg or by 20% from baseline after an initial variceal bleed (29,30).

evaluation
The three essential components in evaluating patients with portal hypertension are (i) assessment of liver function, (ii) endoscopic study, and (iii) radiologic imaging. Liver Function is assessed from clinical and laboratory parameters. Ascites, jaundice, muscle wasting, and encephalopathy are markers for advanced liver disease. Biochemical and hematologic lab profiles include a complete blood count, prothrombin time, serum electrolytes, and liver chemistries. Specific markers of liver disease include hepatitis panels, antinuclear antibody, antimitochondrial antibody, iron and copper levels, alpha 1 antitrypsin, and alpha fetoprotein for HCC screening. The Child-Pugh and MELD scoring systems combine clinical and laboratory variables to determine prognosis in chronic liver disease. Endoscopy focuses on the presence, extent and size of esophageal and/or gastric varices and Portal Hypertensive Gastropathy (PHG). Grading of varices, using, for example, the North Italian Endoscopic Club (NIEC) system, risk-stratify based on variceal size, severity of red wale markings. When combined with Child–Pugh class, such grading can help to guide therapy (25,26). Similar grading systems have been developed and validated for PHG and gastric varices (27,28).

management of variceal bleeding
Primary Prophylaxis Figure 31.2 presents a management algorithm for primary prophylaxis to prevent an initial variceal bleed. This is based on data from multiple randomized controlled trials (Grade 1a evidence). If small, low-risk varices or no varices are discovered repeat endoscopy should be performed in 2 to 3 years. If small

Obstruction to portal flow (cirrhosis/PVT/etc.)

Increased portal venous pressure

Increased production vasoconstrictors Increased hepatic vascular tone Increased hepatic vascular resistance

Increased production vasodilators

Table 31.1 Causes of Portal Hypertension
Prehepatic Portal or splenic vein trombosis Arteriovenous fistula Extrinsic portal vein compression Pre-sinusoidal: Schistosomiasis Hepatic fibrosis Sarcoidosis Early primary biliary cirrhosis Sinusoidal—Alcoholic liver disease Most causes of Cirrosis Hepatic vein thrombosis Veno occlusive disease Caval and hepatic vein webs

Peripheral, ↓ BP

Splanchnic hyperemia

Activate neurohumoral

Intrahepatic

Na and H2O retention Increased collateral flow

Increased C.O. Portal hypertension

Post hepatic

Copyright 2007 by Saunders, an imprint of Elsevier Inc. Figure 31.1 Pathophysiology of portal hypertension demonstrating complex vascular and neurohormonal responses. BP, blood pressure; CO, cardiac output; PVT, portal vein thrombosis. Source: From Ref. (97).

Source: Adapted from (77).

281

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
varices (<5 mm) are found, but there are other risk factors (Child B/C or presence of red wale marks), non-selective betablockers should be started. The presence of medium or large esophageal varices is an indication for starting a noncardioselective beta-blocker (propranolol or nadolol). Treatment of the latter two populations reduces the risk of first variceal hemorrhage from 30% to 14% (31). Endoscopic Variceal Ligation (EVL) can be used in patients that are intolerant to beta blocker therapy or have large high-risk varices. In a meta-analysis of 12 studies, EVL was slightly better at preventing a first variceal bleed than beta-blocker therapy without any improvement in mortality (32). Beta-blocker therapy is first-line therapy in primary prophylaxis and reduction of the HVPG to less than 12 mmHg or by 20% from baseline is the goal of therapy (33–36). Acute Variceal Hemorrhage Accurate diagnosis of acute upper GI bleeding in patients with cirrhosis requires differentiation of variceal bleeding from Mallory Weiss lesions, portal hypertensive gastropathy, and peptic ulcer disease. Figure 31.3 shows a management approach for acute variceal bleeding. Management starts with conservative resuscitation in a closely monitored setting with large bore IV access in the setting of massive bleeding, recognizing that orotracheal intubation may be required. Blood product transfusion should be individualized based on hemoglobin, platelet count, INR, and overall hemodynamic profile. Over-resuscitation should be avoided as this may exacerbate bleeding by increasing portal venous pressure. In patients known to have cirrhosis, pharmacologic therapy with octreotide (Sandostatin—Novartis Pharmaceuticals) or terlipressin (Glypressin—Ferring Pharmaceuticals), and the
Algorithm for prophylaxis of variceal bleeding Cirrhosis

administration of intravenous antibiotics should precede endoscopy (37,38) (Grade 1a evidence). Terlipressin, a longacting synthetic vasopressin, in a meta-analysis of seven randomized controlled trials afforded a 34% relative risk reduction in all-cause mortality compared to placebo (39). Octreotide, was compared to terlipressin in two randomized controlled trials and both agents were found to be similarly effective (40,41). Endoscopy should be performed early for diagnosis and treatment. EVL and injection sclerotherapy are effective at controlling acute bleeding, especially when combined with pharmacologic treatment (42). When compared to sclerotherapy, banding is associated with a lower rate of rebleeding with fewer complications and endoscopy sessions (43) (Grade 1b evidence). In the 5% to 10% of patients not responding to the above measures, balloon tamponade may be necessary as a rescue measure until emergency Transjugular Intrahepatic Portosystemic Shunt (TIPS) can be performed. In-hospital mortality from acute variceal hemorrhage has steadily declined over the last couple of decades to 14.5%, thanks in large part to better resuscitation strategies, lower rates of rebleeding, and prevention of infectious complications (44). Prevention of Recurrent Variceal Bleed A first variceal bleed changes the odds of subsequent bleeding. Without specific therapy, approximately 60% of patients will rebleed within 1 to 2 years (31,45). This emphasizes the need for optimal strategies to prevent rebleeding and at the same time not accelerate the rate of progression of any underlying liver disease (46). Figure 31.4 illustrates a management algorithm. Pharmacologic and Endoscopic Therapy The best results for prevention of rebleeding have been obtained using a combination of pharmacologic therapy with non-selective beta-blockers (sometimes in combination with

Endoscopy

Suspected variceal bleed

No varices: f/u endoscopy 2 yrs

Varices (moderate or large)

Small varices (<5 mm) f/u endoscopy 1 yr

1. Somatostatin/octreotide 2. Conservative resuscitation 3. Antibiotics

Non-cardioselective β-blocker (Propanolol or Nadolol)

Endoscopy

-Diagnostic -Therapeutic: sclerotherapy or ligation

Intolerance to β-blockers or high-risk varices Continued bleed or rebleed Band ligation Figure 31.2 Primary prophylaxis to prevent an initial variceal bleed. Management algorithm based on variceal size. Source: From Ref. (97). -Balloon tamponade -TIPS

Copyright 2007 by Saunders, an imprint of Elsevier Inc. Figure 31.3 Acute variceal bleed. A management algorithm for diagnosis and management of acute variceal bleeding. Source: From Ref. (97).

282

PORTAL HYPERTENSION
nitrates) and EVL (Grade 1b evidence). The latter should be performed in 2 to 4 sessions, 7 to 10 days apart until variceal obliteration is achieved. Combination therapy has succeeded in lowering rebleeding rates to as low as 14% at 2 years (47). The lowest rates of rebleeding are found in patients deemed “responders” based on HVPG measurements as mentioned earlier. Patients with a HVPG < 12 mmHg or a 20% decrease in HVPG have a 10% rebleeding rate (48–50). Some have argued that pharmacologic therapy should be tried first, with EVL added for “non-responders.” In a meta-analysis of 12 randomized controlled trials comparing TIPS to endoscopic therapy for preventing variceal rebleeding, TIPS performed better in the prevention of rebleeding and deaths due to rebleeding, albeit with a higher rate of encephalopathy (51) (Grade 1a evidence). However, half of the studies used sclerotherapy alone for secondary prophylaxis. Subgroup analysis showed EVL was statistically equivalent to TIPS in terms of rebleeding, mortality, and encephalopathy. The data suggest TIPS should be used only once the combination of pharmacologic and endoscopic therapies has failed. Variceal Decompression Variceal decompression, either with a radiologic shunt (TIPS) or a surgical shunt, is indicated in patients that continue to bleed despite adequate medical and endoscopic therapy. TIPS has largely replaced surgical shunts both for patients with good hepatic reserve, and as a bridge to transplantation. TIPS Since its introduction in the 1990s TIPS has undergone a number of modifications, allowing it to become the primary means of portal decompression with a rebleeding rate at 1 year of about 13% (52). It is functionally an intrahepatic side-toside portocaval shunt which lowers the portal venous pressure and sinusoidal pressure. Initially, the main limitation of TIPS was stenosis, with the need for frequent surveillance and reintervention in the first year (50 % to 80%). The use of polytetrafluoroethylene (PTFE)-covered stents has markedly reduced the need for reintervention to <25%. Two randomized trials demonstrated improved primary patency rates with no impact on rate of encephalopathy or survival (53,54) (Grade 1b evidence). The other main side effect of TIPS is the development or worsening of encephalopathy which occurs in approximately 30% of patients (52). This is often amenable to medical management or modification of the TIPS itself. TIPS Procedure Venous access is usually obtained through the right internal jugular vein into the right or middle hepatic veins. The main right portal vein is identified by a retrograde hepatic venogram, and the portal venous system is entered above the bifurcation. After a contrast portal venogram confirms correct positioning, the TIPS is placed with a gentle curve to the prosthesis to prevent kinking. It is also important to choose a stent of the correct length, not extending too far into either the hepatic vein or the portal vein. At the same time, the entire hepatic vein segment should be covered to prevent the development of a stenosis at this end. Portal venous pressures are measured before and after the TIPS is deployed, and a completion venogram is obtained. The goal is a portal vein to right atrial gradient of <10 mmHg. Recurrence of variceal bleeding often heralds TIPS dysfunction and warrants recatheterization of the shunt to assess for patency and measure the gradient. Surgical Shunts Surgical shunts can be classified as total, partial, or selective. Total shunts divert all portal venous blood flow either a direct side-to-side portocaval shunt or an H-graft shunt using a 10-mm or greater PTFE graft. Portal decompression and resolution of variceal bleeding are excellent at greater than 90%; however, encephalopathy develops in up to 45% (55). Partial shunts use an 8 mm graft to partially decompress the portal venous system while preserving some hepatopedal blood flow. Prevention of variceal rebleeding remains greater than 90% with lower rates of encephalopathy compared to total shunts (56,57) (Grade 3 evidence). Selective shunts, such as the Distal Splenorenal Shunt (DSRS), selectively decompress gastroesophageal varices while maintaining portal hypertension (Fig. 31.5). These became the most widely used surgical shunts from 1980s to 1990s. Most series demonstrated rebleeding rates from 5% to7% and encephalopathy in 5% to 19% range. An NIH-funded prospective, randomized trial (1997 to 2003) compared DSRS to TIPS in patients with Child’s class A and B cirrhosis (58) (Grade 1b evidence). At the close of this study (mean follow-up of 46 ± 26 months with a range 24 to 96 months), the survival rates were not different at 2 and

Acute bleed controlled with banding

Evaluation

β-blocker and course of banding

Varices obliterated

Persistent high-risk varices or rebleeding

β-blocker

-Repeat banding -or decompress TIPS DSRS End-stage liver disease

Transplant Copyright 2007 by Saunders, an imprint of Elsevier Inc. Figure 31.4 Prevention of recurrent variceal bleeding with cascading therapy options. TIPS, transjugular intrahepatic portosystemic shunt; DSRS, distal splenorenal shunt. Source: From Ref. (97).

283

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
5 years (DSRS 81% and 62%; TIPS 88% and 61%). The overall rebleeding rates were not significantly different (5.5% in the DSRS group, 10.5% in the TIPS group), and the occurrence of a first episode of any degree of encephalopathy was 50% in each arm. The reintervention rate was significantly higher (p < 0.001) at 82% in the TIPS group versus 11% in the DSRS group. This study was conducted before covered stents were available for TIPS and the use of covered stents has lowered the shunt dysfunction rate at 1 year to 13% (53,54). Devascularization Procedures These series of operations were devised to decrease variceal inflow, while preserving portal hypertension and therefore decreasing the rate of encephalopathy. These have been used much more extensively in the Mid and Far East on patients with non-alcoholic liver disease with good results. In their original series of 276 patients, Sugiura and Futagawa reported an operative mortality of 4.3% and 2.3% rate of variceal recurrence (59). Various modifications have been made to the procedure that generally consists of splenectomy, devascularization of the upper two-thirds of the greater and lesser curvature of the stomach and 7 cm of distal thoracic esophagus, esophageal transection and reanastomosis, and pyloroplasty (60). Outside Japan, experience with gastroesophageal devascularization has not been quite as impressive. In the largest series in the Western hemisphere, Orozco et al. reported an operative mortality of 22% and rebleeding rate of 10% but included patients with Child’s class C cirrhosis (61). Encephalopathy rates have generally been less than 10% in most series. Currently, devascularization procedures are most appropriate for patients with extensive thrombosis of the portal venous system with recurrent variceal bleeding and preserved liver function who lack “shuntable” veins (Grade 3 evidence). Transplantation Liver transplantation constitutes the ultimate “shunt” with normalization of portal venous hemodynamics and restoration of liver synthetic function. Ongoing improvements in operative techniques and immunosuppression are responsible for a modern 5-year patient survival rates approaching 75% (62). The major shortcoming of transplantation remains a shortage of available donor organs. Strategies to deal with this have included “split liver” and living-related transplantation.

ascites
This is the most common complication of cirrhosis and portal hypertension and is an important source of morbidity and mortality. Figure 31.6 presents a management algorithm. Ascites develops as a result of sinusoidal portal hypertension and the concomitant vasodilatation within the splanchnic circulation with an imbalance between hepatic lymph production and return to the systemic circulation. A relative hypovolemia is sensed by the kidneys due to reflex systemic vasoconstriction that, in turn, activates the renin–angiotensin system leading to sodium and water retention. Paracentesis is used in the diagnosis and management of ascites. For diagnosis, a Serum-Ascites Albumin Gradient

Treatment of ascites Ascites

Mild/moderate

1. Diet, sodium restriction (2 gm/day) 2. Spironolactone (100 mg/day) 3. May add Lasix (40 mg/day)

Refractory Persistent ascites Large volume paracentesis ± albumin infusion Increased frequency TIPS

? Transplant Figure 31.5 Distal splenorenal shunt selectively decompresses gastroesophageal varices while maintaining portal perfusion through the superior mesenteric and portal veins. Source: From Ref. (97). Figure 31.6 Ascites. A management algorithm based on severity and response to therapy. TIPS, transjugular intrahepatic portosystemic shunt. Source: From Ref. (97).

284

PORTAL HYPERTENSION
(SAAG) greater than 1.1 is highly suggestive of cirrhosis. Other causes of ascites (malignancy, infection, pancreatic ascites) usually have a SAAG less than 1.1. In the cirrhotic patient with ascites, abdominal pain and signs of infection, paracentesis should be performed to evaluate for possible spontaneous bacterial peritonitis, which carries a mortality of around 37% (63). Diagnosis of SBP requires the presence of at least 250 polymorphonuclear cells per cubic millimeter of ascitic fluid (64). As with gastroesophageal varices, a systematic approach should be taken in the management of ascites. The cornerstones of treatment of mild to moderate ascites are sodium restriction, to less than 2 g/d and diuretics (65). Spironolactone is the first-line diuretic acting as an aldosterone antagonist. Therapy starts with 100 mg daily and can be titrated up to a maximum dose of 400 mg daily. Furosemide can be added to assist with natriuresis and reduction of peripheral edema, but caution must be exercised to prevent overdiuresis. Refractory ascites fails to respond to maximum medical management and carries a poor prognosis. It interferes significantly with quality of life and requires a more aggressive approach that may include Large Volume Paracentesis (LVP), TIPS, or transplant. LVP removes 5 or more liters of ascites, with or without concomitant albumin infusion. A metaanalysis of four randomized trials comparing LVP to TIPS, showed that TIPS was superior to LVP in terms of control of ascites and transplant-free survival but carried a higher risk of hepatic encephalopathy (66) (Grade 1a evidence). Patients with refractory ascites are best served with transplantation for long-term survival. be made with the use of contrast-enhanced transthoracic echocardiography with agitated saline to produce microbubbles. In the presence of the abnormally dilated pulmonary vascular bed typical of HPS, microbubbles can be seen in the left atrium within 3 to 6 cardiac cycles. Alternatively, a more sensitive method involves the injection of technetium-99m-labeled microaggregated albumin into a peripheral vein with quantitative uptake in the brain. Liver transplantation is the only known definitive treatment for HPS with a 5-year survival of 76% in one series compared to 23% for patients not undergoing transplant (22) (Grade 3 evidence). Given the progressive nature of the disease and inferior outcome after transplant for those with severe hypoxemia (PaO2 < 50 mmHg), patients diagnosed with HPS should be prioritized for transplantation. Portopulmonary hypertension is defined by the presence of the following in the setting of portal hypertension: elevated pulmonary artery pressures (>25 mmHg), elevated pulmonary vascular resistance (>240 dyne s–1 cm–5), and pulmonary artery occlusion pressures <15 mmHg (68). As with HPS, the presence or severity of PPH does not seem to correlate with the degree of liver disease or portal hypertension. The hyperdynamic circulation that accompanies portal hypertension leads to increased sheer stress and vascular remodeling in the pulmonary vasculature which, in turn, leads to elevated resistance. The most common symptom is dyspnea on exertion but fatigue, syncope, palpitations, and chest pain can also be seen. Transthoracic echocardiography is a good initial screening test to perform if PPH is suspected with right-heart catheterization used to confirm the diagnosis along with its severity. Milder cases of PPH may be reversible with liver transplant, but severe PPH is associated with high post-transplant mortality (69) (Grade 3 evidence). Aggressive medical therapy with prostanoids and other pulmonary vasodilators is mandatory prior to considering patients for transplantation.

portopulmonary syndromes
Lung dysfunction may develop for a variety of reasons in patients with cirrhosis and portal hypertension, with the two main clinical entities of hepatopulmonary syndrome (HPS) and portopulmonary hypertension (PPH). (Figure 31.7 summarizes key features of these two syndromes. HPS is characterized by a defect in arterial oxygenation in the setting of liver disease due to arteriovenous shunting in the pulmonary vascular bed (21). HPS manifests most commonly with dyspnea, digital clubbing, cyanosis, and hypoxemia. Diagnosis of this syndrome requires (1) advanced liver disease, (2) arterial hypoxemia (PaO2 < 80 mmHg or alveolar-arterial oxygen gradient > or = 15 mmHg), and (3) pulmonary vascular dilatation (67). The precise etiology for the pulmonary vascular changes that take place are not entirely known, but are thought related to elevated pulmonary nitric oxide levels. Diagnosis can

portal hypertension and the general surgeon
Although the role of surgical shunts has diminished greatly over the last couple of decades, general surgeons should be familiar with the pathophysiology and surgical risks in patients with cirrhosis and portal hypertension. Surgery in these patients is associated with higher morbidity and mortality in a wide range of surgical procedures (70–73) (Grade 3 evidence). The Child–Turcotte–Pugh (CTP) scoring system has been used to predict postoperative morbidity and mortality for 3 decades. Recently the Model for End-Stage liver Disease

Variables Prevalence Pulmonary vascular changes Contributors Transplant?

Hepatopulmonary Syndrome (pO2 < 70) 8–20% Vasodilatation Liver dysfunction portal hypertension Curative

Portopulmonary Hypertension (RVSP > 40) 3–12% Vasoconstriction Portal hypertension Contraindicated

Figure 31.7 Pulmonary syndromes in liver disease: characteristics and management. RVSP, right ventricular systolic pressure. Source: From Ref. (97).

285

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
(MELD) score, developed to predict TIPS outcome, and used as the basis for the allocation of cadaveric livers for transplantation, has become a useful prognostic tool for predicting postoperative outcomes in non-transplant surgery. CTP involves subjective grading of ascites and encephalopathy, but the MELD score utilizes objective, continuous variables derived from a linear regression model. Northrup et al. found MELD score to be the only statistically significant predictor of 30-day mortality among patients undergoing a wide range of surgical procedures (74). Likewise, in a larger retrospective series, Teh et al. identified MELD score, age, and American Society of Anesthesiologists (ASA) class as the only significant predictors of mortality in multivariate analysis (75). General surgery in patients with cirrhosis is complicated by higher rates of usual postoperative complications, and unique complications, such as decompensation of liver function, ascites leak, and impaired wound healing. Weighing risk versus benefit of operation is particularly difficult in these patients. When an elective procedure is indicated, measures taken to lessen the chances of perioperative complications include nutritional optimization, TIPS, and aggressive management of ascites (76). Surgeons must also take into account the potential conversion of an elective procedure, such as hernia repair, into an emergent procedure down the road. Emergent surgery in this patient group has been associated with significantly higher morbidity and mortality than in the elective setting, and may warrant accepting the lower, albeit elevated, risk of an elective procedure (70). Along with the surgeon, there are a number of other specialists in the multidisciplinary team charged with caring for patients with portal hypertension. Hepatologists perform an essential “quarterback” role by diagnosing and managing the underlying liver pathology and directing patients to other providers. Endoscopists diagnose gastroesophageal varices and manage bleeding in the emergent setting. Interventional radiologists have assumed an increasingly prominent role through the performance of TIPS, as well as portal venography and pressure measurements to guide medical therapy. Lastly, critical care physicians come into play during acute episodes of variceal bleeding and other events to help stabilize patients until more definitive treatment can be delivered. The general surgeon should never operate on a patient with portal hypertension without involving some or all of this team, and the importance of communication between all members of team in the management of these complex patients cannot be overemphasized.
6. Garcia-Pagan JC, Heydtmann M, Raffa S, et al. TIPS for Budd-Chiari syndrome; long-term results and prognostic factors in 124 patients. Gastroenterology 2008; 135(3):808–15. 7. Sharma S, Texeira A, Texeira P, et al. Pharmacologic thrombolysis in Budd Chiari syndrome: a single centre experience and review of the literature. J Hepatol 2004; 40(1): 172–80. 8. Garcia-Pagan JC, Bosch J, Rodes J. The role of vasoactive mediators in portal hypertension. Semin Gastrointest Dis 1995; 6(3): 140–7. 9. Schiff ER, Sorrell MF, Maddrey WC, editors. Schiff ’s diseases of the liver. 10th edn. Philadelphia: Lippincott Williams and Wilkins; 2007. 10. Garcia-Cardena G, Martasek P, Masters BS, et al. Dissecting the interaction between nitric oxide synthase (NOS) and caveolin. Functional significance of the nos caveolin binding domain in vivo. J Biol Chem 1997; 272(41): 25437–40. 11. Fernandez M, Vizzutti F, Garcia-Pagan JC, et al. Anti-VEGF receptor2 monoclonal antibody prevents portal-systemic collateral vessel formation in portal hypertensive mice. Gastroenterology 2004; 126 (3): 886–94. 12. Fernandez M, Mejias M, Angermayr B, et al. Inhibition of VEGF receptor-2 decreases the development of hyperdynamic splanchnic circulation and portal-systemic collateral vessels in portal hypertensive rats. J Hepatol 2005; 43(1): 98–103. 13. Fernandez M, Mejias M, Garcia-Pras E, et al. Reversal of portal hypertension and hyperdynamic splanchnic circulation by combined vascular ednothelial growth factor and platelet-derived growth factor blockade in rats. Hepatology 2007; 46(4): 1208–17. 14. Park DK, Um SO, Lee JW, et al. Clinical significance of variceal hemorrhage in recent years in patients with liver cirrhosis and esophageal varices. J Gastroenterol Hepatol 2004; 19(9): 1042–51. 15. Zaman A, Becker T, Lapidus J, et al. Risk factors for the presence of varices in cirrhotic patients without a history of variceal hemorrhage. Arch Int Med 2001; 161(21): 2564–70. 16. Merli M, Nicolini G, Angeloni S, et al. Incidence and natural history of small esophageal varices in cirrhotic patients. J Hepatol 2003; 38(3): 361–3. 17. Salerno F, Borroni G, Moser P, et al. Survival and prognostic factors of cirrhotic patients with ascites: a study of 134 outpatients. Am J Gastroenterol 1993; 88(4): 514–9. 18. Belghiti J, Durand F. Abdominal wall hernias in the setting of cirrhosis. Semin Liver Dis 1997; 17(3): 219–26. 19. Li ZF, Zhang S, Huang Y, et al. Morphologic changes of blood spleen barrier in portal hypertensive spleen. Chin Med J 2008; 121(6): 561–5. 20. N’Kontchou G, Seror O, Bourcier V, et al. Partial splenic embolization in patients with cirrhosis: efficacy, tolerance, and long-term outcome in 32 patients. Eur J Gastroenterol Hepatol 2005; 17(2): 179–84. 21. Rodriguez-Roisin R, Krowka MJ. Hepatopulmonary syndrome- a liverinduced lung vascular disorder. N Engl J Med 2008; 358(22): 2378–87. 22. Swanson KL, Wiesner RH, Krowka MJ. Natural history of hepatopulmonary syndrome: impact of liver transplantation. Hepatology 2005; 41(5): 1122–9. 23. Fattovich G, Stroffolini T, Zagni I, et al. Hepatocellular carcinoma in cirrhosis: incidence and risk factors. Gastroenterology 2004; 127: S35–50. 24. De Masi S, Tosti ME, Mele A. Screening for hepatocellular carcinoma. Dig Liver Dis 2005; 37: 260–8. 25. Northern Italian Endoscopy Club for the Study and Treatment of Esophageal Varices. N Engl J Med 1988; 319(15): 983–9. 26. Merkel C, Zoli M, Siringo S, et al. Prognostic indicators of risk for first variceal bleeding in cirrhosis: a multicenter study in 711 patients to validate and improve the North Italian Endoscopic Club (NIEC) index. Am J Gastroenterol 2000; 95(10): 2915–20. 27. Stewart CA, Sanyal AJ. Grading portal gastropathy: validation of a gastropathy scoring system. Am J Gastroenterol 2003; 98(8): 1758–65. 28. Lee CH, Lee JH, Choi YS, et al. Natural history of gastric varices and risk factors for bleeding. Korean J Hepatol 2008; 14(3): 331–41. 29. Albillos A, Banares R, Gonzales M, et al. Value of the hepatic venous pressure gradient to monitor drug therapy for portal hipertensión: a metaanalysis. Am J Gastroenterol 2007; 102(5): 1116–26. 30. D’Amico G, Garcia-Pagan JC, Luca A, et al. Hepatic vein pressure gradient reduction and prevention of variceal bleeding in cirrosis: a systematic review. Gastroenterology 2006; 131(5): 1611–24. 31. D’Amico G, Pagliaro L, Bosch J. Pharmacological treatment of portal hypertension: an evidence-based approach. Semin Liver Dis. 1999; 19(4): 475–505.

references
1. Valla DC, Condat B, Lebrec D. Spectrum of portal vein thrombosis in the West. J Gastroenterol Hepatol 2002; 17 (Suppl 3): S224–7. 2. Matsumoto S, Mori H, Yamada Y, et al. Intrahepatic porto-hepatic venous shunts in Rendu-Osler-Weber disease: imaging demonstration. Eur Radiol 2004; 14: 592–6. 3. Gabriel S, Maroney TP, Ringe BH. Hepatic artery-portal vein fistula formation after percutaneous liver biopsy in a living liver donor. Transplant Proc 2007; 39(5): 1707–9. 4. Souza MR, Toledo CF, Borges DR. Thrombocytemia as a predictor of portal hypertension in schistosomiasis. Dig Dis Sci 2000; 45(10): 1964–70. 5. Rodrigues de Araujo Souza M, Fischer de Toledo C, Borges DR. Thrombocytemia as a predictor of portal hypertension in schistosomiasis. Dig Dis Sci 2000; 45 (10): 1964–70.

286

PORTAL HYPERTENSION
32. Garcia-Pagan JC, Bosch J. Endoscopic band ligation in the treatment of portal hypertension. Nat Clin Pract Gastroenterol Hepatol 2005; 2(11): 526–35. 33. Groszmann RJ, Bosch J, Grace ND, et al. Hemodynamic events in a prospective randomized trial of propranolol versus placebo in the prevention of a first variceal hemorrhage. Gastroenterology 1990; 99(5): 1401–7. 34. Feu F, Garcia-Pagan JC, Bosch J, et al. Relation between portal pressure response to pharmacotherapy and risk of recurrent varicela haemorrhage in patients with cirrhosis. Lancet 1995; 346(8982): 1056–9. 35. Escorsell A, Bordas JM, Castaneda B, et al. Predictive value of the variceal pressure response to continued pharmacological therapy in patients with cirrhosis and portal hypertension. Hepatology 2000; 31(5): 1061–7. 36. Merkel C, Bolognesi M, Sacerdote D, et al. The hemodynamic response to medical treatment of portal hypertension as a predictor of clinical effectives in the primary prophylaxis of varicela bleeding in cirrhosis. Hepatology 2000; 32(5): 930–4. 37. Morales GF, Pereira Lima JC, Hornos AP, et al. Octreotide for esophageal variceal bleeding treated with endoscopic sclerotherapy: a randomized, placebo-controlled trial. Hepatogastroenterology. 2007; 54(73):195–200. 38. Soares-Weiser K, Brezis M, Tur-Kaspa R, et al. Antibiotic prophylaxis for cirrhotic patients with gastrointestinal bleeding. Cochrane Database Syst Rev 2002; (2): CD002907. 39. Ioannou G, Doust J, Rockey DC. Terlipressin for acute esophageal varicela hemorrhage. Cochrane Database Syst Rev 2003; (1): CD002147. 40. Feu F, Ruiz del Arbol L, Banares R, et al. Double-blind randomized controlled trial comparing terlipressin and somatostatin for acute variceal hemorrhage. Variceal Bleeding Study Group. Gastroenterology 1996; 111(5): 1291–9. 41. Walker S, Kreichgauer HP, Bode JC. Terlipressin (glypressin) versus somatostatin in the treatment of bleeding esophageal varices—final report of a placebo—controlled, double-blind study. Z Gastroenterol 1996; 34(10): 692–8. 42. Krige JE, Kotze UK, Bornman PC, et al. Variceal recurrence, rebleeding, and survival after endoscopic injection sclerotherapy in 287 alcoholic cirrhotic patients with bleeding esophageal varices. Ann Surg 2006; 244(5):764–70. 43. Lo GH, Lai KH, Cheng JS, et al. Emergency Banding ligation versus sclerotherapy for the control of active bleeding from esophageal varices. Hepatology 1997; 25(5): 1101–4. 44. Carbonell N, Pauwels A, Serfaty L, et al. Improved survival alter varicela bleeding in patients with cirrhosis over the past two decades. Hepatology 2004; 40(3): 652–9. 45. Yang MT, Chen HS, Lee HC, et al. Risk factors and survival of early bleeding after esophageal variceal ligation. Hepatogastroenterology 2007; 54(78):1705–9. 46. Kravetz D. Prevention of recurrent esophageal variceal hemorrhage: review and current recommendations. J Clin Gastro 2007; 41 (Suppl 3): S318–22. 47. de la Pena J, Brullet E, Sanchez-Hernandez E, et al. Variceal ligation plus nadolol compared with ligation for prophylaxis of variceal rebleeding: a multicenter trial. Hepatology 2005; 41(3): 572–8. 48. Bosch J, Garcia-Pagan JC. Prevention of variceal bleeding. Lancet 2003; 361(9361): 952–4. 49. Turnes J, Garcia-Pagan JC, Abraldes JG, et al. Pharmacological reduction in portal pressure and long-term risk of first variceal bleeding in patients with cirrhosis. Amer J Gastroenterol 2006; 101(3): 506–12. 50. Abraldes JG, Villanueva C, Banares R, et al. Hepatic venous pressure gradient and prognosis in patients with acute variceal bleeding treated with pharmacologic and endoscopic therapy. J Hepatol. 2008; 48(2): 229–36. 51. Zheng M, Chen Y, Bai J, et al. Transjugular intrahepatic portosystemic shunt versus endoscopic therapy in the secondary prophylaxis of variceal rebleeding in cirrhotic patients: meta-analysis update. J Clin Gastroenterol 2008; 42(5): 507–16. 52. Tripathi D, Helmy A, Macbeth K, et al. Ten-years’ follow-up of 472 patients following transjugular intrahepatic portosystemic stent-shunt insertion at a single centre. Eur J Gastroenterol Hepatol 2004; 16(1): 9–18. 53. Bureau C, Garcia-Pagan JC, Otal P, et al. Improved clinical outcome using polytetrafluoroethylene-coated stents for TIPS: results of a randomized study. Gastroenterology 2004; 126(2): 469–75. 54. Bureau C, Garcia-Pagan JC, Layrargues GP, et al. Patency of stents covered with polytetrafluoroethylene in patients treated by transjugular intrahepatic portosystemic shunts: long-term results of a randomized multicentre study. Liver Int 2007; 27(6): 742–7. 55. Stipa S, Balducci G, Ziparo V. Total shunting and elective management of variceal bleeding. World J Surg 1994; 18(2): 200–4. 56. Collins JC, Ong J, Rypins EB, et al. Partial portocaval shunt for variceal hemorrhage; longitudinal analysis of effectiveness. Arch Surg 1998; 133: 590–3. 57. Rosemurgy AS, Serafini FM, Zweibel BR, et al. Transjugular intrahepatic portosystemic shunt vs. small-diameter prosthetic H-graft portocaval shunt: extended follow-up of an expanded randomized prospective trial. J Gastrointest Surg 2000; 4: 589–97. 58. Henderson JM, Boyer TD, Kutner MH, et al. Distal splenorenal shunt versus transjugular intrahepatic portosystemic shunt for variceal bleeding: a randomized trial. Gastroenterology 2006; 130(6):1643–51. 59. Sugiura M, Futagawa S. A new technique for treating esophageal varices. J Thorac Cardiovasc Surg 1973; 66(5): 677–85. 60. Selzner M, Tuttle-Newhall JE, Dahm F, et al. Current indication of a modified sugiura procedure in the management of variceal bleeding. J Am Coll Surg 2001; 193(2): 166–73. 61. Orozco H, Mercado MA, Takahashi T, et al. Elective treatment of bleeding varices with the Sugiura operation over 10 years. Am J Surg 1992; 163(6): 585–9. 62. 2007.Annual Report of the U.S. Organ Procurement and Transplantation Network and the Scientific Registry of Transplant Recipients: Transplant Data 1997-2006. Health Resources and Services Administration, Healthcare Systems Bureau, Division of Transplantation, Rockville, MD. Available from: URL: http://www.ustransplant.org/annual_reports/current/914a_cv_li.htm 63. Nobre SR, Cabral JE, Gomes JJ, et al. In-hospital mortality in spontaneous bacterial peritonitis: a new predictive model. Eur J Gastroenterol Hepatol. 2008; 20(12): 1176–81. 64. Rimola A, Garcia-Tsao G, Navasa M, et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. J Hepatol 2000; 32(1): 142–53. 65. Sandhu BS, Sanyal AJ. Management of ascites in cirrhosis. Clin Liver Dis 2005; 9(4): 715–32. 66. Salerno F, Camma C, Enea M, et al. Transjugular intrahepatic portosystemic shunt for refractory ascites: a meta-analysis of individual patient data. Gastroenterology 2007; 133 (3): 825–34. 67. Krowka MJ. Hepatopulmonary syndrome versus portopulmonary hypertension: distinctions and dilemmas. Hepatology 1997; 25(5): 1282–4. 68. Hoeper MM, Krowka MJ, and Strassburg CP. Portopulmonary hypertension and hepatopulmonary syndrome. Lancet 2004; 363: 1461–8. 69. Krowka MJ, Plevak DJ, Findlay JY, et al. Pulmonary hemodynamics and perioperative cardiopulmonary-related mortality in patients with portopulmonary hypertension undergoing liver transplantation. Liver Transpl 2000; 6: 443–50. 70. Mansour A, Watson W, Shayani V, et al. Abdominal operations in patients with cirrhosis: still a major surgical challenge. Surgery 1997; 122(4): 730–5. 71. del Olmo JA, Flor-Lorente B, Flor-Civera B, et al. Risk factors for nonhepatic surgery in patients with cirrosis. World J Surg 2003; 27(6): 647–52. 72. Puggioni A, Wong L. A metaanalysis of laparoscopic cholecystectomy in patients with cirrhosis. J Am Coll Surg 2003; 197(6): 921–6. 73. Meunier K, Mucci S, Quentin V, et al. Colorectal surgery in cirrhotic patients: assessment of operative morbidity and mortality. Dis Colon Rectum 2008; 51(8): 1225–31. 74. Northrup PG, Wanamaker RC, Lee VD, et al. Model for end-stage liver disease (MELD) predicts nontransplant surgical mortality in patients with cirrhosis. Ann Surg. 2005; 242(2): 244–51. 75. Teh SH, Nagorney DM, Stevens SR, et al. Risk factors for mortality after surgery in patients with cirrhosis. Gastroenterology 2007; 132(4):1261–9. 76. Slakey DP, Benz CC, Joshi S, et al. Umbilical hernia repair in cirrhotic patients: utility of temporary peritoneal dialysis catheter. Am Surg 2005; 71(1): 58–61. 77. Henderson JM. Multidisciplinary approach to the management of portal hypertension. In: Yeo CJ, Dempsey DT, Klein AS, Pemberton JH, Peters JH, eds. Shackelford’s Surgery of the Alimentary Tract, 6th edn. Philadelphia: Saunders, 2007: 1751–70.

287

32

Liver transplantation for acute and chronic liver failure Vincent Kah Hume Wong and J. Peter A. Lodge
UKELD score of > 49 needs to be achieved which predicts a 1-year mortality risk of greater than 9% (7). Exceptions to this rule are patients with HCC (discussed in Chapters 20 and 21), and variant syndromes consisting of diuretic-resistant ascites, intractable pruritus, hepatopulmonary syndrome, chronic hepatic encephalopathy, familial amyloidosis, primary hyperlipidemias, and polycystic liver disease (7). Patients with Acute Liver Failure Etiology (5,15), grade of encephalopathy (16,17), serum pH (18), and serum lactate (19–21) are among the important prognostic indicators for acute liver failure (ALF) and in the United Kingdom, these factors are incorporated into the selection criteria for super-urgent LT (Table 32.2) (7). Timing of LT for ALF patients is vital as clinical deterioration can occur rapidly such that LT may be too risky. Therefore, patients listed on the super-urgent liver transplant waiting list are prioritized to be offered a deceased donor liver from any region within the United Kingdom, increasing the chances of receiving a deceased donor liver. Similarly, patients with ALF or primary nonfunction graft requiring retransplantation in the United States are exempted from MELD allocation and are listed with highest priority for available organs (22). Patients with Viral Hepatitis Hepatitis B virus (HBV)-associated liver cirrhosis was a relative contraindication for LT in the past but the introduction of potent antiviral agents and hepatitis B surface antigen antibody has resulted in a markedly reduced rate of recurrence post-LT (recurrence rate 5–8% in 5 years) (23,24) with concomitant improved patient and graft survival outcome comparable to that of non-viral hepatitis LT (25). In contrast, outcomes for hepatitis C virus (HCV)-related LT remain poorer in comparison with LT for other non-viral causes (26). Unlike HBV, there is no medication as yet for HCV able to provide excellent control of the virus and eventual recurrence of HCV in graft is inevitable. Risk factors for HCV recurrence include donor age >40 years (27), high viral load pre- and early post-LT, cytomegalovirus (CMV) and human immunodeficiency virus (HIV) co-infection (28), HCV genotype 1b (29), and over-immunosuppression or abrupt changes to the immunosuppression (30). Patients with HIV The improved prognosis of HIV (31) in the recent years with the advent of highly active anti-retroviral therapy (HAART), combined with effective prophylaxis of HBV re-infection, and a deeper understanding of HCV recurrence in LT, have made LT possible for HIV patients. HIV itself does not directly damage the liver parenchyma but rather the co-infection with HBV and HCV, HAART-related hepatotoxicity, and malignancies

introduction
Since the first successful liver transplant in 1963, liver transplantation (LT) has become an established form of therapy for patients with acute and chronic liver failure. With reported 5and 10-year survival rates for patients with liver transplant of 70% and 60%, respectively (1), indications for LT have expanded resulting in an increasing demand in this era of donor shortage. To augment the donor pool, expanded donor criteria and novel LT techniques are utilized increasingly. Artificial liver systems and hepatocyte cell transplantation, while still much in development, are exciting potential therapies to reduce the organ shortage burden. This chapter will look at the latest development in LT for acute and chronic liver failure and their indications.

selection of the recipient
According to the European Liver Transplant Registry (1), the major indications for LT in adults are cirrhosis (58%), cancers (13%), cholestatic disease (11%), acute liver failure (9%), metabolic disease (6%), and others (3%). The leading cause for chronic liver disease in the West is alcohol-induced liver disease (2) while acetaminophen overdose (AOD) is the commonest cause of acute liver failure (ALF) in the West, accounting for approximately 40% of cases in the United Kingdom (3,4) and United States (5,6). The widening discrepancy between the number of patients who would benefit from LT and the availability of deceased donor livers in the United Kingdom have led to overburdening of the system: approximately 14% of all patients on the waiting list die before a graft becomes available (7). Therefore, clear, evidence-based patient selection criteria are necessary to allow those on the list to expect a LT within a reasonable timeframe. This is based primarily on risk of death without transplantation and secondarily on the ability of transplant to improve quality of life (7). Patients with End-Stage Liver Disease The current 1-year post-liver transplant mortality risk is 9% in the United Kingdom and as a minimum listing criteria, patients with end-stage liver disease (ESLD) must have predicted 1-year mortality greater than 9% (7). In the past, the Child–Turcotte–Pugh score (8,9) was used to assess the prognosis of patients with ESLD (Table 32.1). Increasingly, the model of end-stage liver disease (MELD) (10) scoring system, based on patient’s serum bilirubin, creatinine, and INR, are adopted by regulatory bodies, UNOS and Eurotransplant to stratify mortality risk in ESLD patients (11). In the United Kingdom, a modified MELD scoring system with the addition of serum sodium (12,13), the UKELD score (United Kingdom Model for end-stage liver disease), is used (7,14). For ESLD patients to be listed in the liver transplant waiting list, a

288

LIVER TRANSPLANTATION FOR ACUTE AND CHRONIC LIVER FAILURE
Table 32.1a Child–Turcotte–Pugh Scoring System
Measure Bilirubin (total) Serum albumin INR Ascites Hepatic encephalopathy 1 point <34 (<2) >35 <1.7 None None 2 points 34–50 (2–3) 28–35 1.71–2.20 Suppressed with medication Grade I–II (or suppressed with medication) 3 points >50 (>3) <28 >2.20 Refractory Grade III–IV (or refractory) units µmol/l (mg/dl) g/l No unit No unit No unit

Table 32.1b Stratification of Mortality Risk According to Child–Turcotte–Pugh Score
Points 5–6 7–9 10–15 Class A B C 1-yr survival (%) 100 81 45 2-yr survival (%) 85 57 35

account for the majority of the LT indications in HIV patients. In addition to the selection criteria mentioned previously, criteria specific to HIV in the United Kingdom (32) include 1. CD4 counts >200 cells/µl or >100 cells/µl in the presence of portal hypertension 2. Absence of viremia 3. Absence of AIDS defining illness after immune reconstitution 4. Anti-retroviral therapeutic options available if HIV disease reactivates.

Table 32.2 Current UK Blood and Transplant Criteria for Listing as a Super-Urgent Transplant
Category 1: Etiology: AOD: pH <7.25 more than 24 hrs after overdose and after fluid resuscitation Category 2: Etiology: AOD: Co-existing PT >100 sec or INR >6.5, and serum creatinine >300 µmol/l or anuria, and grade 3–4 encephalopathy Category 3: Etiology: AOD: Serum lactate more than 24 h after overdose >3.5 mmol/l on admission or >3.0 mmol/l after fluid resuscitation Category 4: Etiology: AOD: Two of the three criteria from category 2 with clinical evidence of deterioration (eg, increased ICP, FiO2 >50%, increasing inotrope requirements) in the absence of clinical sepsis Category 5: Etiology: Seronegative hepatitis, hepatitis A, or hepatitis B or an idiosyncratic drug reaction. PT >100 sec or INR >6.5, and any grade of encephalopathy Category 6: Etiology: Seronegative hepatitis, hepatitis A, or hepatitis B, or an idiosyncratic drug reaction. Any grade of encephalopathy and any three from the following: unfavorable etiology (idiosyncratic drug reaction, seronegative hepatitis), age >40 yrs, jaundice to encephalopathy time >7 days, serum bilirubin >300 µmol/l, PT >50 sec or INR >3.5 Category 7: Etiology: Acute presentation of Wilson’s disease, or Budd–Chiari syndrome. A combination of coagulopathy and any grade of encephalopathy Category 8: Hepatic artery thrombosis on days 0–21 after liver transplantation Category 9: Early graft dysfunction on days 0–7 after liver transplantation with at least two of the following: AST >10,000 IU/l, INR >3.0, serum lactate >3 mmol/l, absence of bile production Category 10: Any patient who has been a live liver donor who develops severe liver failure within 4 wks of the donor operation
AOD – Acetaminophen overdose, PT – Prothrombin time, INR – International Normalized Ratio, ICP – Intracranial Pressure, FiO2 – inspired oxygen concentration

selection of donor
The “ideal” deceased donor profile is as follows: age <40 years, trauma as the cause of death, donation after brain death, hemodynamic stability at the time of procurement, no steatosis or any other underlying chronic liver lesions, and no transmissible disease (33). This implies a very low risk of initial poor graft function or primary graft failure resulting in death or retransplantation. However, the profile of deceased donors is changing and the past decade has seen an increasing proportion of donors >50 years of age with cerebrovascular disease as cause of death (34,35). Expanded Criteria Donor The impact of changing deceased donor characteristics and widening gap between the donor pool and the waiting list means that deceased donors with features deviating from the donor profile are increasingly utilized (36,37). The term “extended or expanded criteria donor” (ECD) has been coined for such donors (Table 32.3), which suggests a higher risk of graft failure and decreased survival. Rather than a clear “good or bad” liver graft, ECD represents a spectrum of cumulative donor risks which should be taken into consideration (33,38,39). Steatosis and Abnormal Liver Function A recent international consensus meeting on ECD grafts (40) recommended against using liver allografts with severe steatosis (>60%) and from elderly donors in HCV-infected recipients. Abnormal liver function tests are not contraindications but careful assessment of other donor factors is essential, especially if there is a marked rise in gamma glutamyl transpeptidase level (>200 UI/L). Elderly Donors Patients with liver transplant from elderly donors have shown comparable survival, provided that there are no additional risk factors (41,42). While there is no clear age limit in utilizing an

289

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 32.3 Definition for Expanded Criteria Liver Donors (ECD)a
Elevated risk for transmission of a disease (viral, eg, Hepatitis C or B) or bacterial infection (especially recovery after sepsis with bacteriemia or donor malignancy, etc.) Acute hemodynamic deterioration with the risk of organ loss Donor age >65 yrs Donor BMI >30 kg/m2 Bilirubin (total) >3 mg/dl or 51 µmol/l ASAT (GOT) or ALAT (GPT) above three times the upper reference threshold Serum sodium >165 mmol/l Hospitalization in ICU >7 days Hepatic steatosis >40%
a There is no internationally agreed definition of ECD as yet. Above is the German Medical Association definition of ECD donors for liver donation (154).

Donors with Hepatitis C Infection Livers from donors infected with HCV represent 2% to 6% of the donor pool. As the risk of HCV transmission is high, HCVinfected liver grafts should be limited to HCV-positive recipients only. Ideally, a biopsy of the HCV-positive graft should be obtained prior to transplantation to assess fibrosis (47). Using HCV-positive grafts with minimal inflammation and fibrosis does not negatively impact outcome. Several studies (55–58) have shown comparable 5-year graft and patient survival between HCV-positive and HCV-negative liver grafts in HCVinfected patients. Donors with Malignancy As the age of donors increases, the risk of cadaveric donors with previous malignancy increases. Currently, the incidence of donors having either an active or treated malignancy is 3% (59) and the risk of transmitted cancer is between 0.02% (60) and 0.2% (61). The most common cancer in an organ donor is skin malignancy followed by brain tumor (59). The risk of donor transmitted malignancy increases with tumor grade (Table 32.4) and any donors with a history of metastatic cancer should be excluded. Split-Liver Split-liver transplantation (SLT) in adults is associated with increased graft failure and recipient morbidity (33,62,63). Current UNOS criteria for potential splittable donors are donor age <40 years, on not more than a single vasopressor, rise in transaminases not more than three times of normal range, and BMI ≤28 (64). A recent study (65) interrogated the UNOS/OPTN data base and suggested two further risk criteria for considering donors for SLT: donor weight ≤40 kg and history of cardiac arrest after declaration of death. In our unit, criteria for splitting a deceased donor liver are (i) donor age <50 years, (ii) body weight >40 kg, (iii) ICU stay <5 days, (iv) minimal inotropic support, (v) good liver function, (vi) minimal hepatic steatosis, and (vii) suitable anatomy. Donation after Cardiac Death In donation after cardiac death (DCD), also called nonheart-beating donation, grafts can be procured either in a controlled setting, after a planned withdrawal of life support, or in an uncontrolled situation with the onset of sudden cardiac arrest. This can be classified into five categories, according to the Maastricht criteria (66) ( Table 32.5). DCD grafts are associated with a significantly increased risk of graft failure (67,68) and biliary complications (69–72), although recent large studies (73,74) have shown that with judicious selection of DCD donors and recipients, comparable results to donation after brain death (DBD) liver allografts can be achieved. In the United Kingdom, guidelines on the use of solid organ transplants from DCD donors were drawn up by the British Transplantation Society (Table 32.6) (75). Our unit practices controlled DCD with planned withdrawal treatment of patient in the ICU, with regular reassessment of success rates. Thus, our current criteria are donor age less than 50 years, asystole

elderly donor liver, there are two caveats: (i) liver grafts of advanced age have reduced regenerative capacity and synthetic function, and are more susceptible to cold ischemic injury and (ii) elderly liver grafts should not be used for HCV-positive patients because of the high risk of severe HCV recurrence (43) and reduced graft and patient survival (44). Donors with Infection The use of donors with bacteremia or bacterial meningitis seems safe as the risk transmission is low (∼4%) (40,45), especially when appropriate prophylactic antibiotic therapy is instituted (46). However, donors with systemic sepsis, multiorgan failure, tuberculosis, or infected with multi-resistant organisms should be avoided (46,47). Donors with Hepatitis B Core Antibody Positive The use of hepatitis B core antibody (anti-HBc) positive donor livers and post-operative management remain controversial. Recipients who receive anti-HBc-positive livers are at a greater risk of developing de novo HBV with reported incidences of 72% (48) to 78% (49), compared to candidates who had antiHBc-negative livers (0.5%) (49). This is due to detectable levels of HBV DNA in serum and/or liver tissues of anti-HBc-positive donors, despite no other serological markers of HBV being present (50,51). Interestingly, studies have shown that recipients who are anti-HBc and/or anti-HBs-positive are protected with a much lower risk of de novo HBV from an anti-HBcpositive liver (48,52). Management strategies to prevent de novo HBV in recipients of anti-HBc-positive livers involve using lamivudine and/ or hepatitis B immunoglobulin (HBIG) depending on the serological status of the donor and recipient. Recently, adefovir dipivoxil was proposed as an alternative to HBIG (23). There is no consensus as yet on the prophylactic strategies with some units preferring HBIG monotherapy (53) and others using lamivudine only (54). Our unit’s protocol comprises of long-term lamivudine for recipients of anti-HBc-positive livers. For patients receiving hepatitis B surface antigen-positive livers, long-term lamivudine and adefovir regimens are used.

290

LIVER TRANSPLANTATION FOR ACUTE AND CHRONIC LIVER FAILURE
Table 32.4 Risk of Transmission of Malignancy from Donor to Recipient and Recommendations for Use of Organs
Tumor Type Skin SCC Melanoma Central nervous System malignancy Grade I/II Grade III/IV Lung Colon Breast Renal cell carcinoma Choriocarcinoma Risk of Transmission (%) 0 81 Level of Risk Low Extreme Recommendations Use with caution Reject

0 40 39 19 29 61 93

Low High High Intermediate Intermediate Intermediate Extreme

Use with caution Reject Reject Dependent on tumor stage and interval from diagnosis. Dependent on tumor stage and interval from diagnosis. Use with caution as transmission has been often limited to the kidney graft Reject

Table 32.5 First International Workshop on Donation after Cardiac Death, Masstricht, 1994
Categories of Donation After Cardiac Death (categories 1 and 2 for uncontrolled DCD, 3 and 4 for controlled DCD) Dead on arrival Unsuccessful resuscitation Awaiting cardiac arrest-ventilator switch off Cardiac arrest while brain-dead Unexpected cardiac arrest while in ITU/ or critical care unita
a Category 5 is an addition to the original four categories of the Maastricht criteria.(155)

cell-mediated rejection. Early reports (77,78) showed much reduced graft survival for ABO-incompatible (ABO-I) LT with 1-, 2-, and 5-year graft survival of 66%, 30%, and 20% but recent reports (79–81) have demonstrated that comparable short- and long-term patient and graft survival with blood-matched LT is achievable. Various treatment strategies to reduce risk of rejections have been advocated including plasmapheresis, immunoadsorption, splenectomy, quadruple immunosuppression, rituximab, and hepatic artery or portal vein infusion of prostaglandin E1 (80,82).

retrieval of deceased donor liver graft
Standard Retrieval Prior to retrieval, the surgical team will need to ensure that brainstem death is confirmed, check the consent from next of kin and confirm the identity of the potential donor. A midline laparotomy is performed and a thorough inspection of the abdominal organs is carried out to rule out any malignancy. The liver is assessed visually for suitability. The aorta is isolated at the bifurcation, in preparation for systemic perfusion. If there is no retrieval of the pancreas, inferior mesenteric vein (IMV) is isolated for portal vein perfusion. Hepatic arterial anatomy is assessed. Cannulas for perfusion are inserted into the aorta and IMV and secured with ligation of aortic bifurcation. Heparin is given intravenously, followed by aortic clamping applied either at the subdiaphragmatic level or in the thoracic cavity, in coordination with the donor coordinator, other retrieval teams and anesthetist. The IVC is vented to allow drainage of venous blood and rapid cooling of liver is achieved by perfusion of cold UW preservation fluid and ice slush in the abdominal cavity. Typically, 4 to 6 l of preservation fluid is used. Once there is adequate perfusion of liver, the liver is then mobilized, with careful dissection of the vascular and anatomy and bile duct. Once retrieved, the liver is assessed at the backbench and further perfusion of the portal vein, hepatic artery, and common bile duct is performed. The liver with the preservation fluid is then double bagged and kept at 4°C in an ice box for transport to the implanting center.

Table 32.6 British Transplantation Society Selection Criteria for Potential DCD Donor for Liver Allografts (75)
Age < 70 yrs Warm ischemia timea ≤ 20 min No history of renal impairment No uncontrolled hypertension or complicated insulin dependent diabetes No uncontrolled systemic sepsis or malignancy using the same criteria as for potential DBD donors
a This time is defined as starting when there is hypotension below a systolic BP of 55 mmHg and is measured up to the point of the cold perfusion of the organ.

within 45 minutes after treatment withdrawal, and traveling time less than 2 hours. A 10 minutes “stand off ” period is instituted from time of asystole to confirm cardiac death. Failure of rapid asystole results procedure abandonment and a resultant increased unit workload without impacting the transplant rate. ABO-Incompatible Liver LT across the ABO barrier remains rare in the West and is used only in the setting of super-urgent need or as a bridging transplantation (76) because of the risk of antibody and

291

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Donation After Cardiac Death Retrieval As mentioned earlier, there is a “stand off ” period of 10 minutes from asystole prior to retrieval. Once cardiac death is confirmed, a midline laparotomy is performed and rapid cooling of abdominal organs is achieved via aortic perfusion of cold preservation solution with heparin or streptokinase and ice slush in the abdominal cavity. The IVC is vented and the aorta is clamped at the subdiaphragmatic level or in the thoracic cavity. The rest of the retrieval procedure is performed as per standard retrieval. The senior author has used VVB only once in the last 100 liver transplants and has only once used a portocaval shunt. Operative Technique of Orthotopic Liver Transplantation Hepatectomy of Recipient’s Native Liver OLT can be performed using a variety of incisions, most commonly the “Mercedes” incision in the literature, but the senior author prefers an inverted L, gentle rooftop or midline incision in order to offer a better cosmetic result. A mechanical retraction system is used to allow adequate exposure to the liver. The common bile duct (CBD) and hepatic artery are ligated and divided, preserving length to increase implantation options. The portal vein is then isolated. Hepatectomy begins with mobilization of the liver from the abdominal wall and IVC with ligation of retrohepatic and caudate lobe veins. The portal vein is clamped and divided and then the hepatic veins are stapled or suture-ligated and divided, allowing completion of the hepatectomy. Implantation of Deceased Donor Liver In the authors’ center, IVC anastomosis is by triangular cavocavostomy or self-triangulating cavocavostomy in most cases (90). Both surgical techniques have the advantage of having no posterior suture line and by triangulating the cavocavostomy, the outflow tract is widened, reducing the risk of venous outflow obstruction (90). The self-triangulating cavocavostomy is worthy of description. A side-biting clamp is placed on the IVC, below the level of the hepatic veins, to allow a 6 to 8 cm cavotomy, without occluding the IVC. The donor IVC is bisected from the top posteriorly for 4 to 6 cm, trimming the excess IVC from the split to create a 5 mm cuff of IVC around the caudate lobe. Two 3-0 polypropylene sutures are used for caval anastomosis, one from the top end and the other from the bottom end of the cavotomy of recipient IVC. The graft is flushed through the portal vein with 500 ml of 4.5% albumin (as UW is high in potassium and adenosine content), and the lower end of the donor IVC closed using a vascular stapler. End-to-end portal vein anastomosis is performed 5-0 polypropylene and the liver is reperfused. Removal of the IVC clamp creates the triangulation as the cuff of donor IVC will open the anastomosis transversely. Hepatic artery reconstruction is performed onto the recipient’s hepatic artery using 7-0 or 8-0 polypropylene sutures. CBD anastomosis is performed either end-to-end with recipient’s CBD or hepaticojejunostomy if there is a significant donor–recipient CBD size discrepancy or if recipient’s liver disease involves biliary strictures (e.g., primary sclerosing cholangitis—PSC). Once satisfactory hemostasis is achieved, drains are placed in the subhepatic regions and abdominal closure is performed under minimal tension. Modern liver transplant surgery takes on average 3 to 4 hours, although complex cases may take considerably longer. Post-operative Management and Immunosuppression Patients are normally nursed in the intensive care unit initially, although most will be extubated and returned to the general ward within 24 hours. Unit protocols vary but we use an intravenous infusions of heparin (40 units/kg/24 hours) and N-acetylcysteine (10 g over 24 h) for the first 4 to 7 days.

splitting of deceased donor liver graft
Splitting of a deceased donor liver graft can be performed at the donor institution (in situ) or at the liver transplant center (ex situ). While some argue that in situ splitting offers superior results with reduction of the cold ischemia time, better identification of vascular and biliary structures, and reduced bleeding from the cut surface (83), recent studies (65,84) suggest that there is no significant difference in terms of graft survival between in situ and ex situ surgical techniques in experienced centers. Our preference is ex situ splitting because in situ splitting will require extensive logistical coordination at the donor institution and between various different organ retrieval teams, highly specialized skills, and a prolonged donor operating time. Surgical Technique of Splitting Deceased Donor Liver to an Extended Right Lobe and a Left Lateral Segment Grafts The portal triad and hepatic venous anatomy are assessed prior to the decision to split. We do not routinely performed on-table cholangiography, although some centers advocate this. For splitting the liver into an extended right lobe (right trisectional graft—segments I and IV–VII) and left lateral segment (left lateral section graft—segment II and III), parenchymal transection at approximately 1 cm to the right of the falciform ligament is carried out usually using bipolar diathermy at a high setting. All ductal and vascular branches during hepatic transaction are ligated or sutured. The left lateral segment contains the left hepatic vein, left or segmental hepatic ducts, left portal vein, and the hepatic artery, while the inferior vena cava, common bile duct, portal vein, and right hepatic artery are left with the extended right lobe. Full right–full left splitting is also carried out but is less commonly reported.

orthotopic liver transplantation
The surgical technique for orthotopic LT (OLT) has evolved since its first conception in the 1960s (85). Refinement in surgical techniques with focus on intraoperative hemodynamic stability, vascular and biliary anastomosis, and hemostasis has led to novel surgical variations including veno-venous bypass (VVB) (86) during the anhepatic phase, IVC preservation using the “piggyback” technique (87), and temporary portocaval shunt (88,89). Up until fairly recently, VVB was used routinely during OLT in our institution with low VVB-related mortality and morbidity (90). However, the lack of evidence-based benefit of VVB over IVC preservation (91,92) and its potential complications (93) have resulted in a change of policy to selective VVB.

292

LIVER TRANSPLANTATION FOR ACUTE AND CHRONIC LIVER FAILURE
Prophylactic broad spectrum antibiotics are given in the first 48 hours post-surgery. Antiviral therapy (oral valganciclovir) is initiated if there is a mismatch between recipient and donor’s cytomegalovirus (CMV) serological status (recipient CMVnegative—donor CMV-positive) (94). Antifungal therapy is prescribed as outlined in Table 32.7. In general, in our unit, immunosuppression consists of a calcineurin inhibitor (tacrolimus or cyclosporine), azathioprine or mycophenolate mofetil and steroid but this varies, depending on the etiology of liver failure and clinical condition of patient. Patients with preoperative renal impairment, high blood transfusion requirements, or post-operative oliguria are given basiliximab, in addition with delayed introduction of the calcineurin inhibitor and tailoring the doses according to patient’s renal function. Immunosuppression regimen for OLT in HCV-infected recipients remains unclear with no standardized treatment (95). Studies have shown conflicting results regarding different immunosuppressions (96–98) and use of steroid (99–101) and the risk of HCV recurrence and liver fibrosis. A recent meta-analysis (102) of randomized trials on steroid avoidance in OLT demonstrated a lower risk of HCV recurrence with steroid avoidance (RR 0.90, p = 0.03). Others (103) have found that slow steroid tapering, rather than rapid withdrawal, and avoidance of steroid boluses were associated with less severe recurrent disease. Currently, we favor gradual reduction of steroids over 3 months post-operatively. As HAART medication for HIV can affect recipients’ immunosuppression levels, tacrolimus and cyclosporine blood concentration should be carefully monitored. For HCV/HIV co-infected recipients, a similar immunosuppression strategy as for HCV monoinfected recipients should be adopted with slow withdrawal of steroids and avoiding steroid boluses. Outcomes in Orthotopic Liver Transplantation The improvement in surgical techniques and anesthesia, a deeper understanding of disease pathology and immunotherapy, better donor and recipient selection, and the introduction of new immunosuppression, antibiotics and antiviral drugs have led to favorable short- and long-term survival rates: a 15-year survival rate of 64% has been reported recently (104). Analysis of a large cohort of OLT patients demonstrated that the long-term survival outcome is dependent on multiple factors including donor and recipient characteristics, etiology of liver disease, and surgical team experience (104). Despite the shift of donor profile to older donors and increasing use of ECD grafts, both super-urgent and elective patient survivals have improved in the most recent years (104,105). Among the elective OLT group, primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC), and HBV have superior 5-year survival rates at 82%, 78%, and 76%, respectively. Results for OLT for ALF remain inferior at about 66% at 5 years (105). While patient survival for HCV recipients is comparable with HBV at 1 year, 5-year survival remains poorer at 69% (105). Chronic HCV infection occurs in 75% to 90% of HCV recipient post-OLT (106) and approximately 20% to 30% of recurrent HCV patients will develop cirrhosis within 5 years (30,107). Pegylated interferon and ribavarin have been shown to improve long-term outcomes for recurrent HCV (108) but antiviral therapy is poorly tolerated (109) and the sustained virology response is low at between 33% and 42% (30). Retransplantation for recurrent HCV-cirrhosis is contentious, although similar 1- and 3-year survival rates after retransplantation between HCV retransplantation and nonHCV retransplantation groups have been reported (110). Overall 1- and 3-year patient survival rates post-OLT for HIV-recipients are reported to be between 83% to 91% and 58% to 64%, respectively (31,111). HCV/HIV co-infections have a significantly worse prognosis compared with HBV/HIV or HIV alone, with frequent development of aggressive HCV recurrence post-OLT (28,112). Recurrence of PSC is one of the leading causes of graft failure in the long term (113) with risk of recurrence at 1-, 5-, and

Table 32.7 Post-operative Prophylactic Antifungal Therapy
Type of Patient Routine elective patients High-risk patients defined as below: ● Acute liver failure ● Peri-operative blood transfusion ≥ 8 units ● Renal failure requiring renal replacement therapy ● Re-transplantation High-risk patients requiring amphotericin defined as below: ● Prolonged ITU stay >5 days ● Re-admission to ITU ● Re-laparotomy while in ICU Antifungal Therapy 5 days of 200 mg fluconazole per day (reduced dose in renal impaired recipients, according to serum creatinine) 14–28 days fluconazole (reduced dose in renal impaired recipients, according to serum creatinine)

If normal renal function: 1 mg per kg amphotericin maximum 50 mg standard preparation (Change to liposomal amphotericin if renal function deteriorates). If established renal failure on renal dialysis: 1 mg per kg amphotericin to maximum 50 mg or liposomal amphotericin at the team’s discretion. If impaired renal function but not requiring dialysis: 1 mg per kg liposomal amphotericin – rounded off to nearest ampoule All patients receiving amphotericin prophylaxis should have regular fungal cultures including weekly aspergillus antigen.

293

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
10-year estimated at 2%, 12%, and 20%, respectively (114). Following re-emergence of disease, PSC patients may require a second liver with a retransplantation rate reported at 12% (115). In comparison, the cumulative recurrent rate for PBC is 22%, 37%, and 43% at 5, 10, and 15 years (116) but the retransplantation rate for recurrent PBC is significantly lower at 8.5% (115). This may be due to the fact that recurrent PBC in the graft does not have a significant adverse influence on graft function and patient survival (116,117). constructed from donor iliac artery anastomosed to the recipient common/ external iliac artery d. Donor biliary diversion via a Roux-en-Y hepaticojejunostomy to avoid potential damage to the native biliary tree Recipient Selection Criteria for AWOLT The following criteria were applied for patient selection for AWOLT: 1. King’s College Hospital Criteria (18) for emergency LT for acetaminophen overdose. 2. No previous evidence of chronic liver disease and no macroscopic evidence of cirrhosis at laporatomy. 3. Age ≤ 50 years. 4. No prolonged circulatory arrest pre-operatively. 5. Cerebral perfusion pressure (mean arterial blood pressure—intracranial pressure) ≥40 mmHg, except for momentary surges in intracranial pressure (ICP). Surgical Technique of AWOLT Subtotal Hepatectomy by Right Trisectionectomy A midline laporatomy incision (with a right transverse extension if needed) and a mechanical retraction system (OmniTract Surgical, Division of Minnesota Scientific Inc., Minneapolis, MN) are used to allow adequate access to the liver and to the aorto-iliac system. The right portal structures are delineated extrahepatically, avoiding liver mobilization at this stage as this can cause a rise in ICP. The right hepatic artery is divided, allowing access to the right portal vein which is then stapled using a vascular stapler (TA30, Autosuture, United States Surgical, Tyco Healthcare Group LP, Norwalk, CT, USA.) and divided, leaving an adequate stump for subsequent anastomosis. The right liver is then mobilized off the IVC by dividing the multiple short hepatic veins with ligaclips (Autosuture, United States Surgical, Tyco Healthcare Group LP, Norwalk, CT, USA) and the right hepatic vein using a TA30 stapler. No attempt is made to control the middle hepatic vein before parenchymal dissection or to isolate the portal structures to segment 4 extrahepatically, as this may cause hemorrhage. The IVC flow is preserved during the entire procedure, negating the need for VVB. In contrast to the low central venous pressure (0–5 mmHg) used for hepatic resection (127,128), we have had to accept a central venous pressure of 10 to 25 mmHg in these cases. Parenchymal transaction is carried out using a Cavi-Pulse Ultrasonic Surgical Aspirator (CUSA, Model 200T, Valleylab, Boulder, CO, USA) in a plane 5 to 10 mm to the right of the falciform ligament. Pringle’s maneuver is applied during the hepatic resection because of the high central venous pressure and coagulopathy in these cases. The portal pedicles to segment 4 are delineated, ligated, and divided intraparenchymally, as is the middle hepatic vein. The right hepatic bile duct is ligated along with segmental portal and hepatic venous structures on the cut surface of the liver. Hemostasis of the cut edge of the liver is achieved with suture ligation and Argon beam coagulation (Erbe, ICC 350 INT UK, Elektromedizin GmbH, Tübingen, Germany) and by packing with oxidized cellulose absorbable hemostatic pads

auxiliary liver transplantation
First proposed in the 1960s (118), auxiliary liver transplantation (ALT) was developed because it is recognized that a significant group of patients with ALF who fulfill the transplant criteria can have a complete recovery of their native liver with no long-term sequelae (119–121). In particular is the subset of ALF patients secondary to acetaminophen overdose (AOD) (122). ALT consists of implantation of partial or whole liver allograft, either orthotopically or heterotopically while leaving all or part of the native liver. The attractiveness of ALT is the possibility of eventual immunosuppression withdrawal once the native liver has regenerated sufficiently. In the United Kingdom (3) and the United States (15,123), AOD is the leading cause of ALF, accounting for approximately 40% of all ALF cases. In our unit, patients with AOD ALF make up 4.4% of liver transplant activity (124). In the 1980s and 1990s our results of patients with AOD who had orthotopic liver transplantation (OLT) were disappointing, with a chronic rejection rate of 36.3%, and the 1- and 5-year survival rate of 52.9% and 47.1%, respectively (124). Patients with AOD ALF represent a highly vulnerable and potentially noncompliant group of patients. It is thought that our high chronic rejection rate was attributed to poor patient compliance. Psychiatric assessment prior to LT is not always possible as patients are often encephalopathic and rapidly develop coma by the time they are transferred to our unit or ICU. Auxiliary Whole Orthotopic Liver Transplantation (AWOLT) While most ALT currently performed are auxiliary partial orthotopic liver transplantation (APOLT), our unit utilizes a novel procedure for AOD patients fulfilling transplant criteria (125). It consists of a subtotal hepatectomy with auxiliary whole orthotopic liver transplantation (AWOLT) as a temporary hepatic support (125). The rationale and principles for this technique are as follows: 1. Reduction of the “toxic liver syndrome” (126) and thereby aiding hemodynamic and metabolic stability, by subtotal hepatectomy to remove approximately 75% of liver volume (right trisectionectomy— resection of hepatic segments 4–8). 2. Transplantation of whole donor LT for a maximal liver volume to aid recovery 3. Orthotopic positioning of the graft: a. IVC reconstruction by a piggy-back technique b. Anastomosis of the donor portal vein to the recipient right portal vein c. Hepatic artery anastomosis to the recipient right hepatic artery or to an arterial conduit

294

LIVER TRANSPLANTATION FOR ACUTE AND CHRONIC LIVER FAILURE
(Surgical Nu-Knit, Johnson & Johnson, Arlington, TX, USA) during the implantation phase. Orthotopic Implantation of the Donor Liver A side clamp is applied to the recipient IVC including the stapled right hepatic vein. The right hepatic vein stump is excised with a further extension by a cavotomy to accommodate the upper end of the donor IVC. The caval anastomosis is with 3-0 polypropylene. The graft is flushed through the portal vein with 500 ml of 4.5% albumin, and the lower end of the donor IVC closed using a TA30 vascular stapler. The donor portal vein is anastomosed end-to-end with the recipient right portal vein using 5-0 polypropylene and the liver is reperfused. Hepatic artery reconstruction is performed onto the recipient’s right hepatic artery using 7-0 or 8-0 polypropylene sutures. Alternatively, a donor aorto-iliac conduit can be used for arterialization of the graft with end-to-side anastomosis to the right common or external iliac artery. Donor Biliary Diversion and Considerations in Abdominal Closure The recipient left hepatic duct, common hepatic duct and common bile duct are left undisturbed to reduce the potential risk of long-term biliary structuring. A 50-cm Roux-en-Y hepaticojejunostomy is created with anastomosis of the donor common bile duct to the Roux using 10 interrupted 5-0 polydioxine sutures. Abdominal closure should be performed under minimal tension to avoid a rapid rise in ICP, difficulty with mechanical ventilation and compartment syndrome. This may necessitate undermining the abdominal wall and the use of a polypropylene mesh. Post-operative Management, Immunosuppression Withdrawal, and Follow-up Continuous hemofiltration established pre-operatively should be continued intra-operatively to create intravascular space for transfusion of blood products. Post-operative hemodialysis is guided by the patient’s clinical condition and biochemical results. To reduce the risk of cerebral edema, the patient is nursed in the reverse Trendelenberg position with about 45-degree head up in the operating room and for about 48 hours post-operatively. Full immunosuppression (as above) is initiated for 3 months. Withdrawal of immunosuppression is staggered, beginning with discontinuation of corticosteroids within 12 weeks from transplantation for all patients. A hepatic iminodiacetic acid (HIDA) and computerized tomography (CT) scans are used to assess recovery of native liver at 3 months post-operative. If radiological appearances are satisfactory, azathioprine/mycophenolate mofetil are withdrawn. The calcineurin inhibitors are reduced by one-third the dose every month with the aim to discontinue all Immunosuppression by 6 to 12 months. Outcome of Auxiliary Liver Transplantation In general, the results of ALT for ALF are mixed (see Table 32.8), reflecting the heterogeneity of causation, type of ALT procedure and experiences in different centers. The key issue for ALF is for recipients to be immunosuppression-free which is dependent on the ability of native liver to recover. AOD- and viral hepatitis-related ALF have a higher likelihood of native liver regeneration post-ALT and immunosuppression-free compared to seronegative or autoimmune ALF (122,124,129). A recent study (122) analyzing the histopathology of native liver regeneration in ALF patients post-ALT showed two distinct patterns of liver insults: “diffuse” or “map-like” appearance. Diffuse liver pattern injury, seen mainly in AOD ALF, was associated with rapid native liver regeneration whereas the outcome for “map-like” liver injury, observed in AIH- and seronegative hepatitis-related ALF, was unpredictable (122). Heterotopic ALT (HALT) consists of graft transplantation in a non-anatomical position, often on the infrahepatic vena cava while the native liver is left in situ. Results of HALT have been discouraging with a higher incidence of vascular complications (130), primary non-function (130), and delayed native liver regeneration (131). This has led to the abandoning of HALT currently (131). In contrast, orthotopic ALT, both APOLT and AWOLT, have shown comparable results with OLT for ALF with recent studies (122,124,129,132,133) reporting 1-year survival rates ranging from 57% to 80% (Table 32.8). The rate of immunosuppression cessation ranges from 25% (133) to 100% (124,132). A recent Japanese study (134), however, reported dismal outcomes with APOLT in ALF patients with no survivors. This may be due to patient selection (five unknown, one HBV; median jaundice–encephalopathy interval 42 days) and the use of portal vein diversion which can impair native liver regeneration (134). Portal vein diversion in ALT for ALF remains a controversial issue. It was first proposed because of the concern of functional competition between the graft and the native liver for shared portal blood flow. We do not think that portal vein diversion is indicated for ALT in ALF patients. We agree with Shaw’s (135) observations that the dynamics of portal flow between both livers benefit the recipient; with portal flow directed to the donor graft in the early stage because of low volume and high portal pressure in the injured native liver and reversal of portal flow on immunosuppressant withdrawal and native liver recovery. Two caveats remain in using APOLT for ALF. First, the graft volume should be carefully assessed to avoid graft insufficiency (small-for-size syndrome). Second, APOLT continues to have a higher incidence of morbidity compared to OLT, which may relate to graft insufficiency and the presence of a larger volume of remaining necrotic native liver (130,133). The AWOLT procedure overcomes the concern of small-for-size (136), providing a maximal functional hepatocyte mass to meet the metabolic demands and recovery of AOD ALF patients; a subtotal hepatectomy allows a maximal removal of necrotic liver contributing to “Toxic Liver Syndrome” and at the same time, leaving behind enough native liver to regenerate.

future perspectives on liver transplantation
Liver Support Devices Limited organ availability and potential spontaneous recovery in ALF patients have resulted in innovative attempts to develop extracorporeal liver support systems as a bridge to LT or to

295

296
Native Liver Operation Retransplant 3 4 11 80 4 69 Mortality 1-yr Survival R Trix 13 NR 2 NR 1 Whole 13 2 0 2 Donor Liver Transplant Vascular Complication Biliary Complication Primary Nonfunction ALT Off IS (%) 100 53 NR 1 1 0 2 6 0 0 NR Whole 5 R Tri 16 RL 7 LL 4 LLS 8 (only 40 recorded op) LLS 2 LL 3 RL 1 LL 6 RL 9 1 0 1 2 5 67 60 0 3 5 1 0 0 0 3 1 4 66 66 80 25 L Hep 2 R Hep 4 LLSx 1 L Hep 4 R Hep 7 LL 2 RL 4 LLS 1 LL 4 RL 7 LLSx 17 L Hep 11 R Trix 2 R trix + caudate 1 R Hep 4 11 APOLT 6 HALT 5 NR 6 APOLT 3 HALT 3 7 18 APOLT 10 HALT 8 62 APOLT 71 HALT 33 38

Table 32.8 Recent Published Studies of ALT for ALF

Studies

Patients

Indications

13

AOD 13

Leeds (124) 2008 London (122)a 2008

49

Non-ABC 24 AOD 15 HBV 4 Drug 3 AIH 2 Mushroom 1

Kyoto (134)ab 2005

6

HBV 1 Unknown 5

Strasborg (120)a 2002

15

6

HAV 3 HBV 3 Drugs 4 Others 5 HBV 6

Clichy (129) 2002 Villejuif (133)a 2001

12

EURALT (130)a 1999

47 (includes 12 HALT)

HAV 1 HBV 2 Drug 1 Unknown 7 Other 1 Viral 18 Nonviral 29 (AOD 5)

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Omaha (156)a 1997 L Hep 5 L Trix 2

7

Varicella 1 Non-ABC 4 HAV 2

Whole 5 Whole – Caudate 2 LL 20 LLS 8 R Tri 6 R Tri + Caudate 1 RL 5 LL 2 LLS 2 Whole 2 0

4

1

1

3

57

100

a

Contains both adult and children recipients; bLiving donors; Abbreviations: LLSx indicates left lateral sectionectomy; R Hep, Right hepatectomy; L Hep, Left hepatectomy; R Trix, Right trisectionectomy; L Trix, Left trisectionectomy; LLS, Left lateral section; LL, Left lobe; RL, Right lobe; R Tri, Right Trisectional graft; AOD, Acetominophen overdose; AIH, Autoimmune hepatitis; HAV, Hepatitis A virus; HBV, Hepatitis B virus; EBV, Epstein-Barr virus; HALT, Heterotropic auxiliary liver transplantation; APOLT, Auxiliary partial orthotopic liver transplantation; Whole, Auxiliary whole orthotopic liver graft transplantation; NR, Not recorded. Source: From Ref. 157.

LIVER TRANSPLANTATION FOR ACUTE AND CHRONIC LIVER FAILURE
provide support until the native liver recovers. Extracorporeal liver support devices can be divided into two categories (i) dialysis-based therapies and (ii) bioartificial liver (BAL) devices. Dialysis-based therapies include the molecular absorbent recirculation system (MARS), single pass albumin dialysis (SPAD), and a fractionated plasma separation and adsorption system (Prometheus). While MARS, SPAD, and Prometheus have been shown to improve biochemical markers, no reduction in mortality has been demonstrated (5,137) and further robust randomized controlled studies are needed. BAL devices utilize cultured hepatocytes (porcine or human), with a semi-permeable membrane to separate the hepatocytes from the patient’s circulation (5). It works on the principle of uptake and processing of toxins and nutrient by the hepatocytes while processed metabolites and synthesized proteins by the hepatocytes are passed back into the patient’s circulation (138). A prospective, randomized, multi-center, controlled trial (139) with BAL devices for ALF and primary non-function patients demonstrated that overall 30-day survival was not significantly different between the BAL group and the control arm. When survival was analyzed just for the ALF subset, the ALF subgroup treated with BAL had significantly higher 30-day survival. However, despite studies (140–142) demonstrating benefit in using BAL systems as a bridge to LT, none has demonstrated successful bridging to recovery as yet. Hepatocyte Transplantation Hepatocyte transplantation (HT) involves infusion of cryopreserved or freshly isolated human hepatocytes to repopulate a diseased liver. HT in animal models has been promising and human clinical application was started about 10 years ago (143). HT has been performed in recipients with metabolic liver disorders, ALF, acute-on-chronic liver diseases, and decompensated cirrhosis (143). The most promising results are seen in metabolic liver disease recipients with a beneficial effect of HT maintained up to 2 years (144,145). Active research is still ongoing to improve engraftment and efficacy of HT, development of alternative sources for hepatocytes, optimal immunosuppression regimens, and innovation of efficient means of monitoring presence and function of HT (5,143). Immune Tolerance The liver is unique among transplanted organs in that it is an immune privileged organ. LT recipients require less immunosuppression than other organ recipients and LT confers a degree of immune protection on other organ grafts during simultaneous transplantation. It is known that in some LT animal models, spontaneous tolerance to liver allografts can occur (146,147). Documented reports (148,149) on patients maintaining a healthy graft while immunosuppression-free has supported this observation and studies (150–152) have shown that approximately 20% of LT patients may be successfully withdrawn from immunosuppression. The prospect of identifying potential recipients in whom all immunosuppression can be discontinued along with their concomitant side effects is exciting. This prospect is aided by a recent discovery of a set of biomarkers which can identify tolerant LT recipients (153).

references
1. European Liver Transplant Registry. (Accessed 19th December, 2008, at http://www.eltr.org.) 2. Berlakovich G. Wasting your organ with your lifestyle and receiving a new one? Ann Transplant 2005; 10: 38–43. 3. Bernal W. Changing patterns of causation and the use of transplantation in the United Kingdom. Semin Liver Dis 2003; 23: 227–37. 4. O’Grady JG. Acute liver failure. Postgrad Med J 2005; 81: 148–54. 5. Lee WM, Squires RH, Jr., Nyberg SL, Doo E, Hoofnagle JH. Acute liver failure: summary of a workshop. Hepatology 2008; 47: 1401–15. 6. Norris W, Paredes AH, Lewis JH. Drug-induced liver injury in 2007. Curr Opin Gastroenterol 2008; 24: 287–97. 7. Neuberger J, Gimson A, Davies M, et al. Selection of patients for liver transplantation and allocation of donated livers in the UK. Gut 2008; 57: 252–7. 8. Child CG, Turcotte JG. Surgery and portal hypertension. In: Child CG, ed. The Liver and Portal Hypertension. Philadelphia: Saunders; 1964: 50–64. 9. Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973; 60: 646–9. 10. Santori G, Andorno E, Antonucci A, et al. Potential predictive value of the MELD score for short-term mortality after liver transplantation. Transplant Proc 2004; 36: 533–4. 11. Broering DC, Walter J, Braun F, Rogiers X. Current status of hepatic transplantation. Anatomical basis for liver transplantation. Curr Probl Surg 2008; 45: 587–661. 12. Ruf AE, Kremers WK, Chavez LL, et al. Addition of serum sodium into the MELD score predicts waiting list mortality better than MELD alone. Liver Transpl 2005; 11: 336–43. 13. Biggins SW, Kim WR, Terrault NA, et al. Evidence-based incorporation of serum sodium concentration into MELD. Gastroenterology 2006; 130: 1652–60. 14. Huo TI, Lee SD, Lin HC. Selecting an optimal prognostic system for liver cirrhosis: the model for end-stage liver disease and beyond. Liver Int 2008; 28: 606–13. 15. Ostapowicz G, Fontana RJ, Schiodt FV, et al. Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States. Ann Intern Med 2002; 137: 947–54. 16. Polson J. Assessment of prognosis in acute liver failure. Semin Liver Dis 2008; 28: 218–25. 17. Dhiman RK, Jain S, Maheshwari U, et al. Early indicators of prognosis in fulminant hepatic failure: an assessment of the Model for End-Stage Liver Disease (MELD) and King’s College Hospital criteria. Liver Transpl 2007; 13: 814–21. 18. O’Grady JG, Alexander GJ, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989; 97: 439–45. 19. Bernal W, Donaldson N, Wyncoll D, Wendon J. Blood lactate as an early predictor of outcome in paracetamol-induced acute liver failure: a cohort study. Lancet 2002; 359: 558–63. 20. Cholongitas E, O’Beirne J, Betrossian A, et al. Prognostic impact of lactate in acute liver failure. Liver Transpl 2008; 14: 121–2; author reply 3. 21. Hadem J, Stiefel P, Bahr MJ, et al. Prognostic implications of lactate, bilirubin, and etiology in German patients with acute liver failure. Clin Gastroenterol Hepatol 2008; 6: 339–45. 22. Ahmed A, Keeffe EB. Current indications and contraindications for liver transplantation. Clin Liver Dis 2007; 11: 227–47. 23. Angus PW, Patterson SJ, Strasser SI, McCaughan GW, Gane E. A randomized study of adefovir dipivoxil in place of HBIG in combination with lamivudine as post-liver transplantation hepatitis B prophylaxis. Hepatology 2008; 48: 1460–6. 24. Degertekin B, Lok AS. What is the optimal regimen for preventing hepatitis B recurrence after liver transplantation? Nat Clin Pract Gastroenterol Hepatol 2009; 6: 68–9 25. Gaglio P, Singh S, Degertekin B, et al. Impact of the hepatitis B virus genotype on pre- and post-liver transplantation outcomes. Liver Transpl 2008; 14: 1420–7. 26. Hong Z, Smart G, Dawood M, et al. Hepatitis C infection and survivals of liver transplant patients in Canada, 1997–2003. Transplant Proc 2008; 40: 1466–70.

297

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
27. Roche B, Sebagh M, Canfora ML, et al. Hepatitis C virus therapy in liver transplant recipients: Response predictors, effect on fibrosis progression, and importance of the initial stage of fibrosis. Liver Transpl 2008; 14: 1766–77. 28. Duclos-Vallee JC, Feray C, Sebagh M, et al. Survival and recurrence of hepatitis C after liver transplantation in patients coinfected with human immunodeficiency virus and hepatitis C virus. Hepatology 2008; 47: 407–17. 29. Pradat P, Tillmann HL, Sauleda S, et al. Long-term follow-up of the hepatitis C HENCORE cohort: response to therapy and occurrence of liver-related complications. J Viral Hepat 2007; 14: 556–63. 30. Mukherjee S, Sorrell MF. Controversies in liver transplantation for hepatitis C. Gastroenterology 2008; 134: 1777–88. 31. Roland ME, Barin B, Carlson L, et al. HIV-infected liver and kidney transplant recipients: 1- and 3-year outcomes. Am J Transplant 2008; 8: 355–65. 32. O’Grady J, Taylor C, Brook G. Guidelines for liver transplantation in patients with HIV infection (2005). HIV Med 2005; 6(Suppl 2): 149–53. 33. Feng S, Goodrich NP, Bragg-Gresham JL, et al. Characteristics associated with liver graft failure: the concept of a donor risk index. Am J Transplant 2006; 6: 783–90. 34. Merion RM, Goodrich NP, Feng S. How can we define expanded criteria for liver donors? J Hepatol 2006; 45: 484–8. 35. Bruna Esteban M, Lopez Andujar R, et al. [Donors yesterday and today: have the characteristics of liver donors changed over the last 15 years?]. Cir Esp 2008; 83: 194–8. 36. Barshes NR, Horwitz IB, Franzini L, Vierling JM, Goss JA. Waitlist mortality decreases with increased use of extended criteria donor liver grafts at adult liver transplant centers. Am J Transplant 2007; 7: 1265–70. 37. Tector AJ, Mangus RS, Chestovich P, et al. Use of extended criteria livers decreases wait time for liver transplantation without adversely impacting posttransplant survival. Ann Surg 2006; 244: 439–50. 38. Afonso RC, Hidalgo R, Paes AT, et al. Impact of cumulative risk factors for expanded criteria donors on early survival after liver transplantation. Transplant Proc 2008; 40: 800–1. 39. Fischer-Frohlich CL, Lauchart W. Expanded criteria liver donors (ECD): effect of cumulative risks. Ann Transplant 2006; 11: 38–42. 40. Durand F, Renz JF, Alkofer B, et al. Report of the Paris consensus meeting on expanded criteria donors in liver transplantation. Liver Transpl 2008; 14: 1694–707. 41. Alghamdi HM, Crawford MD, Gallagher JP, McCaughan GW, Strasser SI, Verran DJ. Cadaveric liver transplant from older donors. Saudi Med J 2008; 29: 533–8. 42. Cescon M, Grazi GL, Cucchetti A, et al. Improving the outcome of liver transplantation with very old donors with updated selection and management criteria. Liver Transpl 2008; 14: 672–9. 43. Petridis I, Gruttadauria S, Nadalin S, et al. Liver transplantation using donors older than 80 years: a single-center experience. Transplant Proc 2008; 40: 1976–8. 44. Perez-Daga JA, Ramirez-Plaza C, Suarez MA, et al. Impact of donor age on the results of liver transplantation in hepatitis C virus-positive recipients. Transplant Proc 2008; 40: 2959–61. 45. Cerutti E, Stratta C, Romagnoli R, et al. Bacterial- and fungal-positive cultures in organ donors: clinical impact in liver transplantation. Liver Transpl 2006; 12: 1253–9. 46. Gonzalez-Segura C, Pascual M, Garcia Huete L, et al. Donors with positive blood culture: could they transmit infections to the recipients? Transplant Proc 2005; 37: 3664–6. 47. Mueller NJ, Fishman JA. How should we evaluate organ donors with active or prior infections? J Hepatol 2006; 45: 507–13. 48. Dodson SF, Issa S, Araya V, et al. Infectivity of hepatic allografts with antibodies to hepatitis B virus. Transplantation 1997; 64: 1582–4. 49. Dickson RC, Everhart JE, Lake JR, et al. Transmission of hepatitis B by transplantation of livers from donors positive for antibody to hepatitis B core antigen. The National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database. Gastroenterology 1997; 113: 1668–74. 50. Van Thiel DH, De Maria N, Colantoni A, Friedlander L. Can hepatitis B core antibody positive livers be used safely for transplantation: hepatitis B virus detection in the liver of individuals who are hepatitis B core antibody positive. Transplantation 1999; 68: 519–22. Chazouilleres O, Mamish D, Kim M, et al. “Occult” hepatitis B virus as source of infection in liver transplant recipients. Lancet 1994; 343: 142–6. De Feo TM, Poli F, Mozzi F, Moretti MP, Scalamogna M. Risk of transmission of hepatitis B virus from anti–HBC positive cadaveric organ donors: a collaborative study. Transplant Proc 2005; 37: 1238–9. Takemura N, Sugawara Y, Tamura S, Makuuchi M. Liver transplantation using hepatitis B core antibody-positive grafts: review and university of Tokyo experience. Dig Dis Sci 2007; 52: 2472–7. Celebi Kobak A, Karasu Z, Kilic M, et al. Living donor liver transplantation from hepatitis B core antibody positive donors. Transplant Proc 2007; 39: 1488–90. Testa G, Goldstein RM, Netto G, et al. Long-term outcome of patients transplanted with livers from hepatitis C-positive donors. Transplantation 1998; 65: 925–9. Vargas HE, Laskus T, Wang LF, et al. Outcome of liver transplantation in hepatitis C virus-infected patients who received hepatitis C virusinfected grafts. Gastroenterology 1999; 117: 149–53. Velidedeoglu E, Desai NM, Campos L, et al. The outcome of liver grafts procured from hepatitis C-positive donors. Transplantation 2002; 73: 582–7. Ricchiuti A, Brunati A, Mirabella S, et al. Use of hepatitis C virus-positive grafts in liver transplantation: a single-centre experience. Transplant Proc 2005; 37: 2569–70. Buell JF, Alloway RR, Steve Woodle E. How can donors with a previous malignancy be evaluated? J Hepatol 2006; 45: 503–7. Myron Kauffman H, McBride MA, Cherikh WS, Spain PC, Marks WH, Roza AM. Transplant tumor registry: donor related malignancies. Transplantation 2002; 74: 358–62. Birkeland SA, Storm HH. Risk for tumor and other disease transmission by transplantation: a population-based study of unrecognized malignancies and other diseases in organ donors. Transplantation 2002; 74: 1409–13. Bonney GK, Aldouri A, Attia M, et al. Outcomes in right liver lobe transplantation: a matched pair analysis. Transpl Int 2008; 21: 1045–51. Burroughs AK, Sabin CA, Rolles K, et al. 3-month and 12-month mortality after first liver transplant in adults in Europe: predictive models for outcome. Lancet 2006; 367: 225–32. UNOS Policy 3.6.11 Organ distribution: Allocation of livers for segmental transplantation. 2007. (Accessed 23rd December, 2008, at http: //www.unos.org/PoliciesandBylaws2/policies/pdfs/policy_8.pdf.) Lee KW, Cameron AM, Maley WR, Segev DL, Montgomery RA. Factors affecting graft survival after adult/child split-liver transplantation: analysis of the UNOS/OPTN data base. Am J Transplant 2008; 8: 1186–96. Kootstra G, Daemen JH, Oomen AP. Categories of non-heart-beating donors. Transplant Proc 1995; 27: 2893–4. Merion RM, Pelletier SJ, Goodrich N, Englesbe MJ, Delmonico FL. Donation after cardiac death as a strategy to increase deceased donor liver availability. Ann Surg 2006; 244: 555–62. Bernat JL, D’Alessandro AM, Port FK, et al. Report of a National Conference on Donation after cardiac death. Am J Transplant 2006; 6: 281–91. Maheshwari A, Maley W, Li Z, Thuluvath PJ. Biliary complications and outcomes of liver transplantation from donors after cardiac death. Liver Transpl 2007; 13: 1645–53. Dezza MC, Berrevoet F, Sainz-Barriga M, et al. The choice of recipient does not have a bearing on early outcome in liver transplant patients receiving grafts from non-heart-beating donors: a reappraisal? Transplant Proc 2007; 39: 2675–7. Foley DP, Fernandez LA, Leverson G, et al. Donation after cardiac death: the University of Wisconsin experience with liver transplantation. Ann Surg 2005; 242: 724–31. D’Alessandro AM, Fernandez LA, Chin LT, et al. Donation after cardiac death: the University of Wisconsin experience. Ann Transplant 2004; 9: 68–71. Lee KW, Simpkins CE, Montgomery RA, Locke JE, Segev DL, Maley WR. Factors affecting graft survival after liver transplantation from donation after cardiac death donors. Transplantation 2006; 82: 1683–8.

51.

52.

53.

54.

55.

56.

57.

58.

59. 60.

61.

62. 63.

64.

65.

66. 67.

68. 69.

70.

71.

72.

73.

298

LIVER TRANSPLANTATION FOR ACUTE AND CHRONIC LIVER FAILURE
74. Mateo R, Cho Y, Singh G, et al. Risk factors for graft survival after liver transplantation from donation after cardiac death donors: an analysis of OPTN/UNOS data. Am J Transplant 2006; 6: 791–6. 75. Guidelines relating to solid organ transplant from non-heart beating donors. British Transplantation Society, 2004. (Accessed 9th December, 2008, at www.bts.org.uk/Forms/Guidelines_Nov_2004.pdf ) 76. Reddy MS, Wilson C, Torpey N, Manas DM. ABO incompatible liver transplantation: a case of immediate need. Transpl Int 2007; 20: 904–5. 77. Gugenheim J, Samuel D, Reynes M, Bismuth H. Liver transplantation across ABO blood group barriers. Lancet 1990; 336: 519–23. 78. Farges O, Kalil AN, Samuel D, et al. The use of ABO-incompatible grafts in liver transplantation: a life-saving procedure in highly selected patients. Transplantation 1995; 59: 1124–33. 79. Toso C, Al-Qahtani M, Alsaif FA, et al. ABO-incompatible liver transplantation for critically ill adult patients. Transpl Int 2007; 20: 675–81. 80. Egawa H, Ohdan H, Haga H, et al. Current status of liver transplantation across ABO blood-type barrier. J Hepatobiliary Pancreat Surg 2008; 15: 131–8. 81. Egawa H, Teramukai S, Haga H, et al. Present status of ABO-incompatible living donor liver transplantation in Japan.[see comment]. Hepatology 2008; 47: 143–52. 82. Troisi R, Noens L, Montalti R, et al. ABO-mismatch adult living donor liver transplantation using antigen-specific immunoadsorption and quadruple immunosuppression without splenectomy. Liver Transpl 2006; 12: 1412–7. 83. Malago M, Hertl M, Testa G, Rogiers X, Broelsch CE. Split-liver transplantation: future use of scarce donor organs. World J Surg 2002; 26: 275–82. 84. Noujaim HM, Gunson B, Mayer DA, et al. Worth continuing doing ex situ liver graft splitting? A single-center analysis. Am J Transplant 2003; 3: 318–23. 85. Starzl TE, Marchioro TL, Vonkaulla KN, et al. Homotransplantation of the liver in humans. Surg Gynecol Obstet 1963; 117: 659–76. 86. Griffith BP, Shaw BW Jr., Hardesty RL, et al. Veno-venous bypass without systemic anticoagulation for transplantation of the human liver. Surg Gynecol Obstet 1985; 160: 270–2. 87. Tzakis A, Todo S, Starzl TE. Orthotopic liver transplantation with preservation of the inferior vena cava. Ann Surg 1989; 210: 649–52. 88. Tzakis AG, Reyes J, Nour B, et al. Temporary end to side portacaval shunt in orthotopic hepatic transplantation in humans. Surg Gynecol Obstet 1993; 176: 180–2. 89. Belghiti J, Noun R, Sauvanet A. Temporary portocaval anastomosis with preservation of caval flow during orthotopic liver transplantation. Am J Surg 1995; 169: 277–9. 90. Dasgupta D, Sharpe J, Prasad KR, et al. Triangular and self-triangulating cavocavostomy for orthotopic liver transplantation without posterior suture lines: a modified surgical technique. Transpl Int 2006; 19: 117–21. 91. Hilmi IA, Planinsic RM. Con: venovenous bypass should not be used in orthotopic liver transplantation. J Cardiothorac Vasc Anesth 2006; 20: 744–7. 92. Reddy K, Mallett S, Peachey T. Venovenous bypass in orthotopic liver transplantation: time for a rethink? Liver Transpl 2005; 11: 741–9. 93. Budd JM, Isaac JL, Bennett J, Freeman JW. Morbidity and mortality associated with large-bore percutaneous venovenous bypass cannulation for 312 orthotopic liver transplantations. Liver Transpl 2001; 7: 359–62. 94. Guidelines for the prevention and management of cytomegalovirus disease after solid organ transplantation. British Transplantation Society, 2004. (Accessed 16th December, 2008, at http: //www.bts.org.uk/ Forms/4164_BTS%20Standards%20AW.pdf.) 95. Gedaly R, Clifford TM, McHugh PP, et al. Prevalent immunosuppressive strategies in liver transplantation for hepatitis C: results of a multicenter international survey. Transpl Int 2008; 21: 867–72. 96. Oton E, Barcena R, Castillo M, et al. Hepatitis C virus recurrence after liver transplantation: influence of immunosuppressive regimens on viral load and liver histology. Transplant Proc 2006; 38: 2499–501. 97. O’Grady JG, Hardy P, Burroughs AK, Elbourne D. Randomized controlled trial of tacrolimus versus microemulsified cyclosporin (TMC) in liver transplantation: poststudy surveillance to 3 years. Am J Transplant 2007; 7: 137–41. 98. Berenguer M, Royuela A, Zamora J. Immunosuppression with calcineurin inhibitors with respect to the outcome of HCV recurrence after liver transplantation: results of a meta-analysis. Liver Transpl 2007; 13: 21–9. 99. Kato T, Gaynor JJ, Yoshida H, et al. Randomized trial of steroid-free induction versus corticosteroid maintenance among orthotopic liver transplant recipients with hepatitis C virus: impact on hepatic fibrosis progression at one year. Transplantation 2007; 84: 829–35. 100. Klintmalm GB, Washburn WK, Rudich SM, et al. Corticosteroid-free immunosuppression with daclizumab in HCV(+) liver transplant recipients: 1-year interim results of the HCV-3 study. Liver Transpl 2007; 13: 1521–31. 101. Llado L, Fabregat J, Castellote J, et al. Impact of immunosuppression without steroids on rejection and hepatitis C virus evolution after liver transplantation: Results of a prospective randomized study. Liver Transpl 2008; 14: 1752–60. 102. Segev DL, Sozio SM, Shin EJ, et al. Steroid avoidance in liver transplantation: meta-analysis and meta-regression of randomized trials. Liver Transpl 2008; 14: 512–25. 103. Vivarelli M, Burra P, La Barba G, et al. Influence of steroids on HCV recurrence after liver transplantation: A prospective study. J Hepatol 2007; 47: 793–8. 104. Busuttil RW, Farmer DG, Yersiz H, et al. Analysis of long-term outcomes of 3200 liver transplantations over two decades: a single-center experience. Ann Surg 2005; 241: 905–16; discussion 16–8. 105. van der Meulen JH, Lewsey JD, Dawwas MF, Copley LP. Adult orthotopic liver transplantation in the United Kingdom and Ireland between 1994 and 2005. Transplantation 2007; 84: 572–9. 106. Roche B, Samuel D. Aspects of hepatitis C virus infection relating to liver transplantation. Eur J Gastroenterol Hepatol 2006; 18: 313–20. 107. Roche B, Samuel D. Risk factors for hepatitis C recurrence after liver transplantation. J Viral Hepat 2007; 14(Suppl 1): 89–96. 108. Berenguer M, Palau A, Aguilera V, et al. Clinical benefits of antiviral therapy in patients with recurrent hepatitis C following liver transplantation. Am J Transplant 2008; 8: 679–87. 109. Mukherjee S, Lyden E. Impact of pegylated interferon alpha-2B and ribavirin on hepatic fibrosis in liver transplant patients with recurrent hepatitis C: an open-label series. Liver Int 2006; 26: 529–35. 110. McCashland T, Watt K, Lyden E, et al. Retransplantation for hepatitis C: results of a U.S. multicenter retransplant study. Liver Transpl 2007; 13: 1246–53. 111. Vennarecci G, Ettorre GM, Antonini M, et al. Liver transplantation in HIV-positive patients. Transplant Proc 2007; 39: 1936–8. 112. de Vera ME, Dvorchik I, Tom K, et al. Survival of liver transplant patients coinfected with HIV and HCV is adversely impacted by recurrent hepatitis C. Am J Transplant 2006; 6: 2983–93. 113. Alexander J, Lord JD, Yeh MM, et al. Risk factors for recurrence of primary sclerosing cholangitis after liver transplantation. Liver Transpl 2008; 14: 245–51. 114. Campsen J, Zimmerman MA, Trotter JF, et al. Clinically recurrent primary sclerosing cholangitis following liver transplantation: a time course. Liver Transpl 2008; 14: 181–5. 115. Maheshwari A, Yoo HY, Thuluvath PJ. Long-term outcome of liver transplantation in patients with PSC: a comparative analysis with PBC. Am J Gastroenterol 2004; 99: 538–42. 116. Charatcharoenwitthaya P, Pimentel S, Talwalkar JA, et al. Long-term survival and impact of ursodeoxycholic acid treatment for recurrent primary biliary cirrhosis after liver transplantation. Liver Transpl 2007; 13: 1236–45. 117. Jacob DA, Neumann UP, Bahra M, et al. Long-term follow-up after recurrence of primary biliary cirrhosis after liver transplantation in 100 patients. Clin Transplant 2006; 20: 211–20. 118. Absolon KB, Hagihara PF, Griffen WO Jr., Lillehei RC. Experimental and clinical heterotopic liver homotransplantation. Rev Int Hepatol 1965; 15: 1481–90. 119. Chenard-Neu MP, Boudjema K, Bernuau J, et al. Auxiliary liver transplantation: regeneration of the native liver and outcome in 30 patients with fulminant hepatic failure--a multicenter European study. Hepatology 1996; 23: 1119–27.

299

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
120. Jaeck D, Boudjema K, Audet M, et al. Auxiliary partial orthotopic liver transplantation (APOLT) in the treatment of acute liver failure. J Gastroenterol 2002; 37(Suppl 13): 88–91. 121. Pereira SP, McCarthy M, Ellis AJ, et al. Auxiliary partial orthotopic liver transplantation for acute liver failure. J Hepatol 1997; 26: 1010–7. 122. Quaglia A, Portmann BC, Knisely AS, et al. Auxiliary transplantation for acute liver failure: histopathological study of native liver regeneration. Liver Transpl 2008; 14: 1437–48. 123. Larson AM, Polson J, Fontana RJ, et al. Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology 2005; 42: 1364–72. 124. Lodge JP, Dasgupta D, Prasad KR, et al. Emergency subtotal hepatectomy: a new concept for acetaminophen-induced acute liver failure: temporary hepatic support by auxiliary orthotopic liver transplantation enables long-term success. Ann Surg 2008; 247: 238–49. 125. Lodge JP, Prasad KR, Toogood GJ, et al. Auxiliary orthotopic liver transplantation: new technique and results in toxic liver injury. Transplant Proc 2001; 33: 1403–4. 126. Ringe B, Lubbe N, Kuse E, Frei U, Pichlmayr R. Total hepatectomy and liver transplantation as two-stage procedure. Ann Surg 1993; 218: 3–9. 127. Halazun KJ, Al-Mukhtar A, Aldouri A, et al. Right hepatic trisectionectomy for hepatobiliary diseases: results and an appraisal of its current role. Ann Surg 2007; 246: 1065–74. 128. Nishio H, Hidalgo E, Hamady ZZ, et al. Left hepatic trisectionectomy for hepatobiliary malignancy: results and an appraisal of its current role. Ann Surg 2005; 242: 267–75. 129. Durand F, Belghiti J, Handra-Luca A, et al. Auxiliary liver transplantation for fulminant hepatitis B: results from a series of six patients with special emphasis on regeneration and recurrence of hepatitis B. Liver Transpl 2002; 8: 701–7. 130. van Hoek B, de Boer J, Boudjema K, Williams R, Corsmit O, Terpstra OT. Auxiliary versus orthotopic liver transplantation for acute liver failure. EURALT Study Group. European Auxiliary Liver Transplant Registry. J Hepatol 1999; 30: 699–705. 131. Jaeck D, Pessaux P, Wolf P. Which types of graft to use in patients with acute liver failure? (A) Auxiliary liver transplant (B) Living donor liver transplantation (C) The whole liver. (A) I prefer auxiliary liver transplant. Journal of Hepatology 2007; 46: 570–3. 132. Sudan DL, Langnas AN, Shaw BW, Jr. Long-term follow-up of auxiliary liver transplantation for fulminant hepatic failure. Transplant Proc 1997; 29: 485–6. 133. Azoulay D, Samuel D, Ichai P, et al. Auxiliary partial orthotopic versus standard orthotopic whole liver transplantation for acute liver failure: a reappraisal from a single center by a case-control study. Ann Surg 2001; 234: 723–31. 134. Kasahara M, Takada Y, Egawa H, et al. Auxiliary partial orthotopic living donor liver transplantation: Kyoto University experience. Am J Transplant 2005; 5: 558–65. 135. Shaw BW, Jr. Auxiliary liver transplantation for acute liver failure. Liver Transpl Surg 1995; 1: 194–200. 136. Heaton N. Small-for-size liver syndrome after auxiliary and split liver transplantation: donor selection. Liver Transpl 2003; 9: S26–8. 137. Karvellas CJ, Gibney N, Kutsogiannis D, Wendon J, Bain VG. Bench-tobedside review: current evidence for extracorporeal albumin dialysis systems in liver failure. Crit Care 2007; 11: 215. 138. Singhal A, Neuberger J. Acute liver failure: bridging to transplant or recovery--are we there yet? J Hepatol 2007; 46: 557–64. 139. Demetriou AA, Brown RS Jr., Busuttil RW, et al. Prospective, randomized, multicenter, controlled trial of a bioartificial liver in treating acute liver failure. Ann Surg 2004; 239: 660–7; discussion 7–70. 140. Ellis AJ, Hughes RD, Wendon JA, et al. Pilot-controlled trial of the extracorporeal liver assist device in acute liver failure. Hepatology 1996; 24: 1446–51. 141. van de Kerkhove MP, Di Florio E, Scuderi V, et al. Phase I clinical trial with the AMC-bioartificial liver. Int J Artif Organs 2002; 25: 950–9. 142. Patzer JF, 2nd, Mazariegos GV, Lopez R. Preclinical evaluation of the Excorp Medical, Inc, Bioartificial Liver Support System. J Am Coll Surg 2002; 195: 299–310. 143. Smets F, Najimi M, Sokal EM. Cell transplantation in the treatment of liver diseases. Pediatr Transplant 2008; 12: 6–13. 144. Grossman M, Rader DJ, Muller DW, et al. A pilot study of ex vivo gene therapy for homozygous familial hypercholesterolaemia. Nat Med 1995; 1: 1148–54. 145. Dhawan A, Mitry RR, Hughes RD. Hepatocyte transplantation for liver-based metabolic disorders. J Inherit Metab Dis 2006; 29: 431–5. 146. Calne RY, Sells RA, Pena JR, et al. Induction of immunological tolerance by porcine liver allografts. Nature 1969; 223: 472–6. 147. Murase N, Starzl TE, Tanabe M, et al. Variable chimerism, graft-versushost disease, and tolerance after different kinds of cell and whole organ transplantation from Lewis to brown Norway rats. Transplantation 1995; 60: 158–71. 148. Starzl TE, Demetris AJ, Trucco M, et al. Cell migration and chimerism after whole-organ transplantation: the basis of graft acceptance. Hepatology 1993; 17: 1127–52. 149. Alexander SI, Smith N, Hu M, et al. Chimerism and tolerance in a recipient of a deceased-donor liver transplant. N Engl J Med 2008; 358: 369–74. 150. Koshiba T, Li Y, Takemura M, et al. Clinical, immunological, and pathological aspects of operational tolerance after pediatric living-donor liver transplantation. Transpl Immunol 2007; 17: 94–7. 151. Devlin J, Doherty D, Thomson L, et al. Defining the outcome of immunosuppression withdrawal after liver transplantation. Hepatology 1998; 27: 926–33. 152. Mazariegos GV, Reyes J, Marino IR, et al. Weaning of immunosuppression in liver transplant recipients. Transplantation 1997; 63: 243–9. 153. Martinez-Llordella M, Lozano JJ, Puig-Pey I, et al. Using transcriptional profiling to develop a diagnostic test of operational tolerance in liver transplant recipients. J Clin Invest 2008; 118: 2845–57. 154. Bundesaerztekammer (German Medical Association) ed.: Richtlinien zur Organtransplantation nach §16 TPG. Aenderung; Deutsches Aerzteblatt 2004; 101: A246–A7. 155. Sanchez-Fructuoso AI, Prats D, Torrente J, et al. Renal transplantation from non-heart beating donors: a promising alternative to enlarge the donor pool. J Am Soc Nephrol 2000; 11: 350–8. 156. Sudan DL, Shaw BW Jr., Fox IJ, Langnas AN. Long-term follow-up of auxiliary orthotopic liver transplantation for the treatment of fulminant hepatic failure. Surgery 1997; 122: 771–7; discussion 7–8. 157. Lodge JP, Dasgupta D, Prasad KR, et al. Emergency subtotal hepatectomy: a new concept for acetaminophen-induced acute liver failure: temporary hepatic support by auxiliary orthotopic liver transplantation enables long-term success. Ann Surg 2008; 247: 238–49.

300

33

Benign cystic disease of the liver Stephen W. Fenwick and Dowmitra Dasgupta
1.6% and 18% (5–9). Females are affected more than males (6). The frequency of diagnosis increases with age with the peak incidence occurring between the ages of 50 and 60 years (10,11). Typically, patients with simple hepatic cysts have no symptoms. However, larger cysts may exert a mass effect and cause upper abdominal discomfort and early satiety. Symptoms are more common with right sided cysts (10). Complications are rare but include intra-cystic hemorrhage (12), biliary obstruction due to compression of the biliary tree (13,14), vascular compression (15), cyst rupture (16), and cyst torsion. Bacterial infection can occur within a cyst, particularly when there is communication with the biliary tree (17).

introduction
Benign cystic lesions of the liver were historically an uncommon clinical entity, presenting only when symptoms or complications occurred. However, with the wider availability of sophisticated radiological techniques, these lesions are being recognized more commonly. In spite of this they sometimes represent a diagnostic challenge. Common cystic lesions range from the single, simple, small liver cyst to the florid appearance of polycystic liver disease. More uncommon lesions are mesenchymal hamartomas (in pediatric patients), biliary hamartomas, and ciliated foregut hepatic cysts. Once a firm diagnosis is made, management is usually expectant unless symptoms worsen.

simple cysts
Simple biliary hepatic cysts are congenital lesions which are thought to result from progressive dilatation of biliary microhamartomas, otherwise known as von Meyenberg’s complexes. They are lined by flattened biliary epithelium which rests on a thin underlying rim of fibrous stroma, without a distinct separation from adjacent hepatic parenchyma. They may be solitary or multiple, and do not normally communicate with the biliary tree (1). The cysts contain serous fluid which is continuously excreted by the lining epithelial cells.

treatment
The vast majority of simple hepatic cysts are incidental and, once the diagnosis is established, require no treatment. Even for those patients with abdominal symptoms the possibility of another underlying cause must always be considered and investigated.

aspiration
Ultrasound-guided percutaneous cyst aspiration has little role in the treatment of symptomatic simple cysts as the recurrence rate is high (78–100%) (18–20), sometimes as rapid as 2 weeks (18). However, cyst aspiration can be used as a trial of therapy. If symptoms persist after needle aspiration then the cyst is unlikely to be the cause, and other pathology must be sought. Conversely, if symptoms abate after cyst aspiration, and return with recurrence of the cyst, then a definitive treatment of the cyst is indicated.

clinical presentation
Most simple hepatic cysts are asymptomatic and are detected as an incidental finding during imaging of the abdomen for other indications. Ultrasound scanning demonstrates a rounded, anechoic intra-hepatic mass with good-through transmission and an imperceptible wall (2). On unenhanced computed tomography (CT) imaging a simple cyst appears as a homogenous lesion of low attenuation, with no enhancement of the wall or content following the administration of contrast (3) (Fig. 33.1). With magnetic resonance (MR) scanning, simple cysts show low attenuation on T1-weighted images (Fig. 33.2A) and homogeneous, very high signal intensity on T2-weighted images (Fig. 33.2B). The differential diagnosis includes multiple cysts arising as a result of polycystic liver disease, juvenile hydatid cysts, parasitic liver cysts, and biliary cystadenomas or cystadenocarcinomas. The differentiation between these lesions can largely be made on imaging characteristics. Hepatic metastases can occasionally appear cystic, particularly neuroendocrine tumors and sarcomas.

aspiration sclerotherapy
Aspiration sclerotherapy is a well-tested therapeutic technique for the treatment of simple hepatic cysts. The procedure involves the aspiration of cyst fluid followed by the instillation of a sclerosant. A number of sclerosing agents have been used including tetracycline (21,22), minocycline (23,24), pantopaque (25), and alcohol (11,26–29). The procedure is performed under ultrasound or CT guidance. A pigtail catheter is inserted into the cyst, and the cyst fluid aspirated. The fluid is usually clear and should be sent for cytology to exclude malignancy, culture to exclude infection, and microscopy to look for hydatid scolices. In addition the fluid should be assessed for the tumor marker CA19–9 which, if elevated, suggests an underlying diagnosis of cystadenoma or cystadenocarcinoma (30). Any bile staining of the fluid would suggest a communication with the biliary tree and would mandate abandoning the procedure and further assessment with cholangiography. The consequence of inadvertent injection of sclerosant into the biliary tree is devastating, and following cyst drainage a contrast cystogram should always be performed (28).

natural history
An early series, based on findings at laparotomy, estimated a low prevalence of simple hepatic cysts of 0.17%, although many small cysts were probably missed (4). More recent data based on imaging studies suggests the prevalence is between

301

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
The most commonly used sclerosant is alcohol. Reported techniques vary widely. Alcohol strengths of 95% (11,26), 96% (28), and 99% (27) have all been reported with favorable results. Prior to instillation of the alcohol into the cyst, it is imperative to remove as much of the cyst fluid as possible to prevent dilution of the alcohol. An alcohol volume of approximately 25% of the cyst aspirate is instilled (26) for up to 2 hours (11,31), during which time the patient’s position is altered frequently to maximize contact between the cyst wall and the alcohol. The alcohol fixes the epithelial cells lining the cyst cavity. Most authors advocate a single procedure with withdrawal of the catheter on completion. However, if the cyst is large (>400 ml) then multiple procedures can be attempted. The catheter should be left in place and the procedure repeated according to daily cyst drainage. Complications from the procedure are mild pain, transient hyperpyrexia, nausea and vomiting. To reduce these symptoms, patients can be premedicated with an opiate analgesic and an anti-emetic, and a local anesthetic, such as lignocaine can be instilled into the cyst prior to the injection of the alcohol. Severe pain can be a sign of alcohol leaking into the peritoneal cavity. If this occurs the procedure should be abandoned and repeated at a later date. Significant but uncommon complications relate to the needle puncture. These include pleural effusion and hemothorax (32). There are also reports of a transient elevation of blood alcohol level, and for this reason patients should be advised not to drive or operate machinery following the procedure. Aspiration sclerotherapy does have several advantages over other more invasive approaches. It is a relatively simple technique which does not require general anesthesia and can be performed on an out-patient basis. However, it does require experienced radiologists with expertise in the interpretation of hepatobiliary imaging and with competence in interventional hepatobiliary procedures. For these reasons the procedure should only be performed where such expertise is available. The published results of aspiration sclerotherapy appear encouraging. Recurrence rates are reported at between 0 and 30% (11,26–28,31–37), although there is significant disparity between series. The number of patients is often small; the indications are variable with polycystic patients often enrolled alongside patients with simple cysts. The techniques reported vary in terms of sclerosant concentration, volume administered, duration of sclerosis, and whether single or multiple procedures were performed. The length of patient follow-up is variable and reported outcome measures include symptomatic relief or a reduction in cyst volume or the complete ablation of the cyst. Recently, one group has reported similar results when comparing prolonged negative-pressure catheter drainage with single session alcohol sclerotherapy (38). Clearly further studies with greater patient numbers and standardized outcome measures, including quality of life assessment, are required to fully assess this technique.

surgical treatment
Figure 33.1 Multidetector computed tomography (MDCT) image with iodinated intravenous contrast media. A large simple cyst is seen in segments 6 and 7. A small cyst is noted on the periphery of segment 2.

There are three surgical techniques that have been employed in the management of simple benign liver cysts. These are fenestration (deroofing of the cyst), local excision of the cyst (cystectomy), and anatomic or non-anatomic liver resection.

(A)

(B)

Figure 33.2 (A) T1-weighted MRI image through the upper abdomen showing multiple cysts throughout the liver. Fluid is dark on T1-weighting (B) T2-weighted coronal view of the abdomen of the same patient as in Figure 33.2A. The cysts are clearly seen as bright throughout the liver. This section is taken through the vascular pedicle of the liver.

302

BENIGN CYSTIC DISEASE OF THE LIVER
Fenestration has emerged as the most popular technique with fewer complications than the more radical procedures. The concept of cyst fenestration was first described by Lin et al. in 1968 (39). The principle is to form a permanent communication between the cyst cavity and the peritoneum so that any fluid subsequently produced by the cyst lining can be re-absorbed by the peritoneum. The technique involves aspiration of the cyst content, which can be inspected and sent for cytology, microbiological culture, and tumor marker analysis, followed by a wide excision of the extra-hepatic portion of the cyst wall. The cavity can be inspected and biopsies taken from any area of concern. The excised cyst wall should be subjected to histopathological assessment. Hemostasis can be achieved by oversewing the edge of the cyst wall. The results of laparoscopic cyst fenestration are encouraging though most reported studies have few patients with short follow-up periods. Three recent studies have reported outcomes in patients treated with laparoscopic cyst fenestration with a mean follow-up greater than 4 years (40,42,43). There is no reported mortality, with asymptomatic recurrence seen in up to 50% (40) but symptomatic recurrence noted in only 4.5% (43).

open surgery
With the development of the laparoscopic technique, the procedure of open cyst fenestration for benign simple cysts should be reserved for those patients with recurrent disease or with extensive abdominal adhesions that preclude the laparoscopic technique. More radical procedures, including cystectomy and liver resection, carry a greater morbidity and mortality than cyst fenestration (44). Cystectomy can be particularly dangerous as the adjacent parenchyma is compressed and major vascular structures are often within the walls of the cyst. The main indication for resection is when there is suspicion that the cyst may be neoplastic in nature.

laparoscopic surgery
In recent years, the technique of laparoscopic cyst fenestration has developed and should now be the first-line surgical approach for patients with symptomatic simple cysts. Laparoscopic cyst fenestration has all the advantages of minimally invasive surgery including reduced postoperative pain, shorter hospital stay, and quicker recovery. The laparoscopic procedure was initially recommended for cysts in segments II, III, IVb, and V. However, with increasing experience, location should not be considered a contraindication to laparoscopic surgery (40). A standard 4 port laparoscopic technique is used with the patient in a supine position, although for right-sided posteriorly placed cysts a left lateral position can give superior access. An angled laparoscope (30 or 45 degrees) gives superior views of the cyst cavity. The cyst can be punctured by insertion of a trocar through the exposed wall. The contents are sent for analysis. A wide excision of the cyst wall is then made, taking great care not to venture into the hepatic parenchyma. Various techniques have been employed to achieve hemostasis of the remnant cyst wall including diathermy, over sewing of the remnant cyst wall edge and using a laparoscopic linear cutting vascular stapler to excise the cyst wall. The authors’ preference is to use harmonic shears. The resected cyst wall is removed in an endoscopic retrieval bag, and the remnant cyst wall is inspected. Although unusual, if a bile leak is identified it must be controlled either with a suture or with a laparoscopic clip. Some authors advocate obliteration of the residual cyst wall with electrocautery. If this is attempted it should be done without breaching the cyst wall as major vascular structures, distorted by the cyst, can lie just underneath. The argon beam coagulator is probably best suited to the task (41) as the burn is very superficial. There is the potential risk of gas embolus and careful control of intraperitoneal pressure must be maintained. To prevent recurrence of the cyst, particularly when the exposed cyst cavity will come to lie against the abdominal wall, an omentoplasty should be performed. A pedicle of omentum is mobilized from the transverse colon and advanced into the cyst cavity. It can be secured either with sutures or with laparoscopic clips. A cholecystectomy is not routinely performed unless there is evidence of cholecystolithiasis, or where the cyst drainage leaves the gallbladder excessively mobile and at risk of subsequent torsion.

polycystic liver disease
Adult Polycystic Liver Disease (PCLD) is a hereditary condition characterized by the development of multiple macroscopic and microscopic cysts throughout the liver. They are histopathologically similar to simple biliary cysts. There are two main forms of the disease and both show an autosomal dominant pattern of inheritance. The most common form of PCLD develops in association with autosomal-dominant polycystic kidney disease (ADPKD) (Figs. 33.3 and 33.4). This is one of the most common inherited diseases with a prevalence of 1 in 400 to 1 in 1000 and accounts for 8% to 10% of all cases of end stage renal failure (45). Factors associated with more extensive hepatic involvement are increasing age, female gender, severity of renal disease, and severity of renal dysfunction (46). The prevalence of liver cysts in ADPKD rises from 20% in the third to 70% in the seventh decade of life (47). The severity of disease in females correlates with the number of pregnancies and the use of exogenous female hormones (46), and may be due to the stimulatory effects of estrogen (48). The much rarer hereditary form of PCLD, known as Autosomal Dominant Polycystic Liver Disease (ADPLD), occurs with liver-only involvement (49).

clinical presentation
The majority of patients with PCLD are asymptomatic and, as with simple cysts, the diagnosis is made during routine investigation (50). Laboratory tests of liver function including bilirubin, alkaline phosphatase, alanine aminotransferase, and prothrombin time are usually normal. Symptoms tend to occur in patients with longstanding disease and are related to liver enlargement and compression of adjacent organs. Patients may complain of an increase in abdominal girth, upper abdominal pain, early satiety, nausea, respiratory compromise, and limitation in physical ability. More significant complications include hemorrhage into a cyst, infection within a cyst and compression of vascular and

303

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 33.1 Gigot Classification of Adult Polycystic Liver Disease
Classification Type I Type II Features Limited number (<10) of large cysts with large areas of non-cystic parenchyma. Diffuse involvement of liver with multiple medium sized cysts with remaining large areas of non-cystic parenchyma. Massive diffuse involvement of all areas of liver by small and medium-sized cysts with only a few areas of normal parenchyma between cysts.

Type III

Figure 33.3 Portal venous phase MDCT image showing multiple large cysts throughout the liver. Cysts are also noted within the kidneys.

due to PCLD. The aim of treatment should be to reduce the size of the cystic component of the liver, whilst preserving parenchymal liver function, leading to long-term relief of symptoms.

aspiration sclerotherapy
The techniques of serial cyst aspiration (55) and serial aspiration sclerotherapy (56) have been used with success in the treatment of PCLD. However, there is a higher rate of recurrence in PCLD than seen in the treatment of simple hepatic cysts, one explanation being that the more rigid hepatic parenchyma prevents complete collapse of the cysts (27). As a result, this technique is probably best suited to patients with Gigot type I disease in whom more aggressive surgical options would be contraindicated. One exception would be in treating the life-threatening complication of cyst infection where a combination of antibiotic therapy and cyst drainage is required to reduce mortality (52).

surgical treatment
Surgical options for the treatment of PCLD include cyst fenestration, liver resection, a combination of cyst fenestration and liver resection, and liver transplantation. The procedure of cyst fenestration, either by an open or by a laparoscopic approach, involves the progressive deroofing of liver cysts, starting superficially and working through to cysts placed deeper within the liver parenchyma. This approach is best suited to patients with Gigot type I disease. The resulting decrease in liver volume can lead to a significant improvement in symptoms for the majority of patients (57,58). For patients with more severe parenchymal involvement, classified as Gigot types II and III, fenestration alone is rarely successful. A combination of hepatic resection with fenestration may, however, offer alleviation of symptoms through a reduction in liver volume and mass (59–61). The procedure typically involves a non-anatomical resection of either the right or left hemi-liver with fenestration of accessible cysts on the remnant side. The preservation of hepatic parenchyma is crucial when planning the resection. The procedure can be technically challenging as the hepatic anatomy is distorted by the cysts. The absence of landmarks predisposes to injury of the remnant liver inflow or outflow. The major vasculo-biliary structures are often compressed between adjacent cyst walls,

Figure 33.4 T2-weighted MRI image of the upper abdomen showing the appearance of multiple fluid filled simple cysts both in the liver and in the left kidney of a patient with PCLD. Fluid is bright on T2-weighting.

biliary structures. Infection within a cyst is a rare but serious complication and the patient will usually present with abdominal pain and raised inflammatory markers. Awareness of the diagnosis along with prompt treatment with antibiotics and a drainage procedure are necessary to reduce morbidity and mortality (51,52). Jaundice can occur in PCLD due to direct compression of the extrahepatic biliary tree by cysts (53). Most patients with PCLD have well-preserved parenchyma and hepatic failure is rarely reported. Gigot et al. produced a useful classification of adult PCLD according to the number and size of liver cysts and the amount of remaining liver parenchyma (54) (Table 33.1). This classification of PCLD is of use in determining treatment algorithms for patients.

treatment
Therapeutic intervention should only be considered for those patients with significant symptoms or complications

304

BENIGN CYSTIC DISEASE OF THE LIVER
and during resection major bleeding can be encountered. Whilst hepatic pedicle inflow control is usually possible, venous outflow control can be difficult to secure. For these reasons the procedure should only be undertaken by experienced hepatobiliary surgeons. One recent study of resection-fenestration in 21 PCLD patients reported symptom resolution in 100% of cases (62). However, procedure-related morbidity was high (76.2%), with the main complications being ascites (71%), pleural effusion (43%), and bile leak (10%). Transfusion requirements were high (mean 4.5 units per patient) and hospital stay was long (mean 15.5 days). This high morbidity rate has been reported in earlier series (59–61). Therefore, although this procedure can offer long-term relief of symptoms for patients with advanced PCLD, the benefits must be weighted against the significant morbidity related to the surgery. While the majority of patients with PCLD have wellpreserved liver function, cadaveric (63–66) and live donor (67–70) orthotopic liver transplantation have been successfully used in the treatment of symptomatic patients. It can be argued that such an approach is overly aggressive. However, for patients with numerous truly diffuse small cysts there may be no other effective treatment. In those patients who are dialysis dependent the procedure can also be combined with a kidney transplant using a graft from the same donor. The first successful series of liver transplantation for PCLD was reported by Starzl and colleagues for patients with what they described as the syndrome of “lethal exhaustion” (71). These patients are cachectic and have severe functional limitation due to the weight of their enlarging livers. They fatigue easily, have intractable pain, and are usually opiate dependent. The outcome of liver transplantation for PCLD has been reported from a number of centres. The largest is a series of 36 patients transplanted over a period of 13 years (63). Twenty one received a liver-only graft, with 15 receiving combined liver and kidney grafts. The 1- and 5- year patient survival was 86%. Five deaths occurred in the series, all within 2 months. Four deaths were attributed to sepsis and one to a myocardial infarction. Perhaps most importantly, measures of post transplant quality of life were assessed using a combination of questionnaires. Of the 74% of patients who responded, 91% reported feeling “much better” or “better,” with only 9% feeling “worse” than before. Fatigue, physical fitness, loss of appetite, and vomiting improved significantly after transplantation, and 78% of patients said they would opt for transplantation again. In summary, liver transplantation offers a complete and definitive treatment for a minority of patients with end stage PCLD. The benefits of transplantation, however, must be balanced by the risks of the procedure and the need for life-long immunosuppression. of patients with PCLD with the immunosuppressive agent sirolimus was associated with decreased polycystic liver volume. The mechanism for this effect is not clear, but may be through preventing aberrant activation of mTOR in the cyst epithelial cells (73). Recently, percutaneous transcatheter hepatic artery embolization has been reported in the treatment of polycystic liver disease (74,75). Polyvinyl alcohol particles and micro-coils are used to selectively embolize the segmental hepatic arteries of the most affected segments of the liver. Initial results are encouraging, with a reduction in both total liver and cyst volumes, and an improvement in symptoms. Further studies will be required to investigate long term effectiveness, but the procedure may have a role in treating patients unsuitable for more invasive therapy.

rare benign cystic lesions of the liver
Mesenchymal hamartomas are the second commonest benign liver lesion in children. They account for 5% of pediatric liver tumors (76). Presentation is usually with progressive abdominal distension. Radiologically a large, multicystic liver mass is seen, more commonly in the right lobe than the left. Histologically they are composed of bile ducts, immature mesenchymal cells and hepatocytes (77). Surgical resection is usually necessary. Ciliated foregut cysts are rare and usually solitary. Microscopically they have an inner ciliated pseudostratified columnar epithelium, a smooth muscle layer and a fibrous capsule (78). They are commonly located in segment IV (77) and can rarely undergo malignant transformation (squamous metaplasia) (79). Surgical treatment is required if the diagnosis is uncertain or if there is compression of the vasculobiliary tree. Bile duct hamartomas are small, usually less than 1 cm in size. They are composed of dilated bile ducts with a dense collagenous stroma. They do not communicate with the biliary system (80). Since this lesion can appear to have solid elements, it can be difficult to differentiate from a metastasis or a cyst adenocarcinoma (81). T2-weighted MRI images are very useful in the diagnosis since biliary hamartomas are hyperintense (82). Diagnostic uncertainty can occasionally lead to a percutaneous or surgical biopsy.

summary
Benign cystic lesions of the liver are being increasingly recognized due to improvements in and wider availability of radiological techniques. They only require treatment if they produce symptoms or if there is a diagnostic uncertainty. Treatment is usually in the form of percutaneous aspiration, aspiration sclerotherapy or surgery. Surgical treatment ranges from fenestration of the cyst wall, cystectomy, or liver resection. These can be performed as open or laparoscopic procedures according to the surgeon’s expertise. PCLD may rarely require liver transplantation. As yet there are no randomized or case control studies comparing different treatment options. Therefore, treatment decisions have to be based on level 3 evidence (the case report or small case series) and the experience of the treating physician.

novel treatments
Medical treatments have in the past had little role in the treatment of PCLD. There are reports of a reduction in cyst volume using the somatostatin analog octreotide. The effect is thought to be due to a direct reduction of fluid secretion by cholangiocytes (72). In one recent clinical trial, the treatment

305

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

key points
The vast majority of benign cystic lesions of the liver do not require treatment. Rarer forms of these lesions may have an element of diagnostic uncertainty. Indications for treatment are
● ● ●

● ● ● ●

Increase in size in surveillance scans Pain affecting quality of life Symptoms due to compression of the stomach, duodenum, biliary tree, portal venous system or inferior vena cava Intracystic hemorrhage Spontaneous/traumatic rupture Evidence of infection in the cyst Diagnostic uncertainty

acknowledgments
The authors would like to acknowledge and thank Dr David White, Consultant Radiologist, University Hospital Aintree, Liverpool, for contributing the images reproduced in this chapter.

references
1. Dhumeaux D. Congenital cystic diseases of the intra and extrahepatic bile ducts. Gastroenterol Clin Biol 2005; 29: 878–82. 2. Mergo PJ, Ros PR. Benign lesions of the liver. Radiol Clin N Am 1998; 36: 319–31. 3. Mortelé KJ, Ros PR Cystic focal liver lesions in the adult: differential CT and MR imaging features. Radiographics 2001; 21: 895–910. 4. Sanfelippo PM, Beahrs OH, Weiland LH. Cystic disease of the liver. Ann Surg 1974; 179: 922–5. 5. Chang CS, Lin KC, Hwang SL, et al. Nonparasitic hepatic cysts detected in ultrasonographic examination: analysis of 95 cases. Taiwan Yi Xue Hui Za Zhi 1989; 88: 394–9. 6. Gaines PA, Sampson MA. The prevalence and characterization of simple hepatic cysts by ultrasound examination. Br J Radiol 1989; 62: 335–7. 7. Caremani M, Vincenti A, Benci A, et al. Ecographic epidemiology of nonparasitic hepatic cysts. J Clin Ultrasound 1993; 21(2):115–8. 8. Carrim ZI, Murchison JT. The prevalence of simple renal and hepatic cysts detected by spiral computed tomography. Clin Radiol 2003; 58: 626–9. 9. Larssen TB, Rørvik J, Hoff SR, et al. The occurrence of asymptomatic and symptomatic simple hepatic cysts. A prospective, hospital-based study. Clin Radiol 2005; 60: 1026–9. 10. Sanchez H, Gagner M, Rossi RL, et al. Surgical management of nonparasitic cystic liver disease. Am J Surg 1991; 161: 113–8. 11. Erdogan D, van Delden OM, Rauws EA, et al. Results of percutaneous sclerotherapy and surgical treatment in patients with symptomatic simple liver cysts and polycystic liver disease. World J Gastroenterol 2007; 13: 3095–100. 12. Takahashi G, Yoshida H, Mamada Y, et al. Intracystic hemorrhage of a large simple hepatic cyst. J Nippon Med Sch 2008; 75: 302–5. 13. Ishikawa H, Uchida S, Yokokura Y, et al. Nonparasitic solitary huge liver cysts causing intracystic hemorrhage or obstructive jaundice. J Hepatobiliary Pancreat Surg 2002; 9: 764–8. 14. Kanai T, Kenmochi T, Takabayashi T, et al. Obstructive jaundice caused by a huge liver cyst riding on the hilum: report of a case. Surg Today 1999; 29: 791–4. 15. England RA, Wells IP, Gutteridge CM. Benign external compression of the inferior vena cava associated with thrombus formation. Br J Radiol 2005; 78: 553–7. 16. Poggi G, Gatti C, Delmonte A, et al. Spontaneous rupture of non-parasitic hepatic cyst. Int J Clin Pract 2006; 60: 99–103.

17. Klingler PJ, Gadenstätter M, Schmid T, et al. Treatment of hepatic cysts in the era of laparoscopic surgery. Br J Surg 1997; 84: 438–44. 18. Saini S, Mueller PR, Ferrucci JT Jr, et al. Percutaneous aspiration of hepatic cysts does not provide definitive therapy. Am J Roentgenol 1983; 141: 559–60. 19. Koperna T, Vogl S, Satzinger U, et al. Nonparasitic cysts of the liver: results and options of surgical treatment. World J Surg 1997; 21: 850–4. 20. Regev A, Reddy KR, Berho M, et al. Large cystic lesions of the liver in adults: a 15-year experience in a tertiary center. J Am Coll Surg 2001; 193: 36–45. 21. Lopes HM, Portela FA, e Silva Pontes JM, et al. Treatment of benign hepatic cysts by instillation of tetracycline hydrochloride. Hepatogastroenterology 1998; 45: 496–9. 22. McFarlane ME, Venugopal R, McDonald A, Ewing, et al. Management of hepatic cysts by percutaneous drainage and instillation of tetracycline hydrochloride. Indian Med J 2001; 50: 230–3. 23. Cellier C, Cuenod CA, Deslandes P, et al. Symptomatic hepatic cysts: treatment with single-shot injection of minocycline hydrochloride. Radiology 1998; 206: 205–9. 24. Yoshida H, Onda M, Tajiri T, et al. Long-term results of multiple minocycline hydrochloride injections for the treatment of symptomatic solitary hepatic cyst. J Gastroenterol Hepatol 2003; 18: 595–8. 25. Goldstein HM, Carlyle DR, Nelson RS. Treatment of symptomatic hepatic cyst by percutaneous instillation of Pantopaque. Am J Roentgenol 1976; 127: 850–3. 26. Bean WJ, Rodan BA. Hepatic cysts: treatment with alcohol. Am J Roentgenol 1985; 144: 237–41. 27. Kairaluoma MI, Leinonen A, Ståhlberg M, et al. Percutaneous aspiration and alcohol sclerotherapy for symptomatic hepatic cysts. An alternative to surgical intervention. Ann Surg 1989; 210: 208–15. 28. Larssen TB, Jensen DK, Viste A, et al. Single-session alcohol sclerotherapy in symptomatic benign hepatic cysts. Long-term results. Acta Radiol 1999; 40: 636–8. 29. Blonski WC, Campbell MS, Faust T, et al. Successful aspiration and ethanol sclerosis of a large, symptomatic, simple liver cyst: case presentation and review of the literature. World J Gastroenterol 2006; 12: 2949–54. 30. Horsmans Y, Laka A, Gigot JF, et al. Serum and cystic fluid CA 19-9 determinations as a diagnostic help in liver cysts of uncertain nature. Liver 1996; 16: 255–7. 31. Yang CF, Liang HL, Pan HB, et al. Single-session prolonged alcohol-retention sclerotherapy for large hepatic cysts. Am J Roentgenol 2006; 187: 940–3. 32. vanSonnenberg E, Wroblicka JT, D’Agostino HB, et al. Symptomatic hepatic cysts: percutaneous drainage and sclerosis. Radiology 1994; 190: 387–92. 33. Andersson R, Jeppsson B, Lunderquist A, et al. Alcohol sclerotherapy of non-parasitic cysts of the liver. Br J Surg 1989; 76254–5. 34. McCullough KM. Alcohol sclerotherapy of simple parenchymal liver cysts. Austr Radiol 1993; 37: 177–81. 35. Simonetti G, Profili S, Sergiacomi GL, et al. Percutaneous treatment of hepatic cysts by aspiration and sclerotherapy. Cardiovasc Intervent Radiol 1993; 16: 81–4. 36. Montorsi M, Torzilli G, Fumagalli U, et al. Percutaneous alcohol sclerotherapy of simple hepatic cysts. Results from a multicentre survey in Italy. HPB Surg 1994; 8: 89–94. 37. Tikkakoski T, Mäkelä JT, Leinonen S, et al. Treatment of symptomatic congenital hepatic cysts with single-session percutaneous drainage and ethanol sclerosis: technique and outcome. J Vasc Interv Radiol 1996; 7: 235–9. 38. Zerem E, Imamovic ´ G, Omerovic ´ S. Percutaneous treatment of symptomatic non-parasitic benign liver cysts: single-session alcohol sclerotherapy versus prolonged catheter drainage with negative pressure. Eur Radiol 2008; 18: 400–6. 39. Lin TY, Chen CC, Wang SM. Treatment of non-parasitic cystic disease of the liver: a new approach to therapy with polycystic liver. Ann Surg 1968; 168: 921–7. 40. Fabiani P, Iannelli A, Chevallier P, et al. Long-term outcome after laparoscopic fenestration of symptomatic simple cysts of the liver. Br J Surg 2005; 92596–7. 41. Tan YM, Chung A, Mack P, et al. Role of fenestration and resection for symptomatic solitary liver cysts. ANZ J Surg 2005; 75: 577–80.

306

BENIGN CYSTIC DISEASE OF THE LIVER
42. Palanivelu C, Jani K, Malladi V. Laparoscopic management of benign nonparasitic hepatic cysts: a prospective nonrandomized study. South Med J 2006; 99: 1063–7. 43. Bai XL, Liang TB, Yu J, et al. Long-term results of laparoscopic fenestration for patients with congenital liver cysts. Hepatobiliary Pancreat Dis Int 2007; 6: 600–3. 44. Gigot JF, Hubert C, Banice R, et al. Laparoscopic management of benign liver diseases: where are we? HPB 2004; 6: 197–212. 45. Gabow PA. Autosomal dominant polycystic kidney disease. N Engl J Med 1993; 329: 332–42. 46. Gabow PA, Johnson AM, Kaehny WD, et al. Risk factors for the development of hepatic cysts in autosomal dominant polycystic kidney disease. Hepatology 1990; 11: 1033–7. 47. Chauveau D, Fakhouri F, Grunfeld JP. Liver involvement in autosomaldominant polycystic kidney disease: therapeutic dilemma. J Am Soc Nephrol 2000; 11: 1767–75. 48. Sherstha R, McKinley C, Russ P, et al. Postmenopausal estrogen therapy selectively stimulates hepatic enlargement in women with autosomal dominant polycystic kidney disease. Hepatology 1997; 26: 1282–6. 49. Pirson Y, Lannoy N, Peters D, et al. Isolated polycystic liver disease as a distinct genetic disease, unlinked to polycystic kidney disease 1 and polycystic kidney disease 2. Hepatology 1996; 23: 249–52. 50. Russell RT, Pinson CW. Surgical management of polycystic liver disease. World J Gastroenterol 2007; 13: 5052–9. 51. Abascal J, Moya M, Martin F. Infection of hepatic cysts in polycystic disease. World J Surg 1984; 8: 424–5. 52. Telenti, A, Torres, VE, Gross, JB, et al. Hepatic cyst infection in autosomal dominant polycystic kidney disease. Mayo Clin Proc 1990; 65: 933–42. 53. Dmitrewski J, Olliff S, Buckels JA. Obstructive jaundice associated with polycystic liver disease. HPB Surg 1996; 10: 117–20. 54. Gigot JF, Jadoul P, Que F, et al. Adult polycystic liver disease: is fenestration the most adequate operation for long-term management? Ann Surg 1997; 225: 286–94. 55. Arrivé L, Tubiana JM. Serial needle aspiration in polycystic liver disease. Am J Roentgenol 2003; 181: 1717. 56. Ferris JV. Serial ethanol ablation of multiple hepatic cysts as an alternative to liver transplantation. AJR 2003; 180: 472–4. 57. Konstadoulakis MM, Gomatos IP, Albanopoulos K, et al. Laparoscopic fenestration for the treatment of patients with severe adult polycystic liver disease. Am J Surg 2005; 189: 71–5. 58. Robinson TN, Stiegmann GV, Everson GT. Laparoscopic palliation of polycystic liver disease. Surg Endosc 2005; 19: 130–2. 59. Vauthey JN, Maddern GJ, Blumgart LH. Adult polycystic disease of the liver. Br J Surg 1991; 78: 524–7. 60. Que F, Nagorney DM, Gross JB Jr, et al. Liver resection and cyst fenestration in the treatment of severe polycystic liver disease. Gastroenterology 1995; 108: 487–94. 61. Yang GS, Li QG, Lu JH, et al. Combined hepatic resection with fenestration for highly symptomatic polycystic liver disease: a report on seven patients. World J Gastroenterol 2004; 10: 2598–601. 62. Li TJ, Zhang HB, Lu JH, et al. Treatment of polycystic liver disease with resection-fenestration and a new classification. World J Gastroenterol 2008; 14: 5066–72. 63. Kirchner GI, Rifai K, Cantz T, et al. Outcome and quality of life in patients with polycystic liver disease after liver or combined liver-kidney transplantation. Liver Transpl 2006; 12: 1268–77. 64. Ueno T, Barri YM, Netto GJ, et al. Liver and kidney transplantation for polycystic liver and kidney-renal function and outcome. Transplantation 2006; 82: 501–7. 65. Rossi M, Spoletini G, Bussotti A, et al. Combined liver-kidney transplantation in polycystic disease: case reports. Transplant Proc 2008; 40: 2075–6. 66. Schnelldorfer T, Torres VE, Zakaria S, et al. Polycystic liver disease: a critical appraisal of hepatic resection, cyst fenestration, and liver transplantation. Ann Surg 2009; 250: 112–8. 67. Takegoshi K, Tanaka K, Nomura H, et al. Successful living donor liver transplantation for polycystic liver in a patient with autosomal-dominant polycystic kidney disease. J Clin Gastroenterol 2001; 33: 229–31. 68. Koyama I, Fuchinoue S, Urashima Y, et al. Living related liver transplantation for polycystic liver disease. Transpl Int 2002; 15: 578–80. 69. Ueda M, Egawa H, Oike F, et al. Living-donor liver transplantation for polycystic liver disease. Transplantation 2004; 77: 480–1. 70. Mekeel KL, Moss AA, Reddy KS, et al. Living donor liver transplantation in polycystic liver disease. Liver Transpl 2008; 14: 680–3. 71. Starzl TE, Reyes J, Tzakis A, et al. Liver transplantation for polycystic liver disease. Arch Surg 1990; 125: 575–7. 72. van Keimpema L, de Man RA, Drenth JP. Somatostatin analogues reduce liver volume in polycystic liver disease. Gut 2008; 57: 1338–9. 73. Qian Q, Du H, King BF, et al. Sirolimus reduces polycystic liver volume in ADPKD patients. J Am Soc Nephrol 2008; 19: 631–8. 74. Takei R, Ubara Y, Hoshino J, et al. Percutaneous transcatheter hepatic artery embolization for liver cysts in autosomal dominant polycystic kidney disease. Am J Kidney Dis 2007; 49: 744–52. 75. Park HC, Kim CW, Ro H, et al. Transcatheter arterial embolization therapy for a massive polycystic liver in autosomal dominant polycystic kidney disease patients. J Korean Med Sci 2009; 24: 57–61. 76. D’ Angelica M, Fong Y. The Liver. In: Townsend CM, Beauchamp RD, Evers BM, Mattox KL, eds. Sabiston Textbook of Surgery, 17th edn. Philadelphia: Elsevier, 2004: 1513–75. 77. Del Poggio P, Buonocore M. Cystic tumours of the liver: a practical approach. World J Gastroenterol 2008; 14: 3616–20. 78. Bogner B, Hegedus G. Ciliated hepatic foregut cyst. Pathol Oncol Res 2002; 8: 278–9. 79. Shaw JM, Krige JE, Beningfield SJ, et al. Ciliated hepatic foregut cyst: a rare cystic liver lesion. J Gastrointest Surg 2008; 12: 1304–6. 80. Precetti S, Gandon Y, Vilgrain V. Imaging of cystic liver diseases. J Radiol 2007; 88: 1061–72. 81. Karahan OI, Kahriman G, Soyuer I, et al. Hepatic von Meyenburg complex simulating biliary cystadenocarcinoma. Clin Imaging 2007; 31: 50–3. 82. Maher MM, Dervan P, Keogh B, et al. Bile duct hamartomas (von Meyenburg complexes): value of MR imaging in diagnosis. Abdom Imaging 1999; 24: 171–3.

307

34

Management of hydatid disease of the liver Adriano Tocchi
layer for impaired vital exchanges with the host and decreased endocyst pressure (1). Among the various causes suggested, pericyst thickening and the penetration of bile between the pericyst and cyst wall and inside it are of concern for the surgeon. The pericyst, initially composed of very thin connective lamina, subsequently tends to become thicker (up to 1 cm or more), sclerose, and calcify. The process of cyst expansion causes compression of hepatic parenchymal structures, in turn engulfed into the pericyst. Large vessels are compressed and displaced while remaining, however, patent for a long time. Similarly, bile ducts remain patent and may open into the pericyst, between it, and the parasite wall. This phenomenon occurs frequently, unlike the rare frank rupture of the cyst with effusion of the cyst contents into a large duct and the main bile duct. This is of the utmost importance for surgery, since on the one hand its appearance causes major changes in the cyst and pericyst development, and on the other it favors the development of postoperative complications such as biliary fistulas. Bile filtration in the virtual interstitium between the pericyst and the chitinous membrane can form a perivesicular biloma with loss of direct contact of the cyst with the pericyst, a decrease in the mother cyst pressure and membrane rupture. At the same time, the appearance of bile is preliminary to cyst infection. Whatever the cause, endogenous vesiculation should be considered as an initial attempt at survival by the primary parasite. Subsequently the neoformed hydatid material packed into the cystic cavity tends to show signs of stress and to degenerate extensively with varying consistencies akin to: fruit jelly, putty, plaster, dry clay, or pus (Fig. 34.2). At the same time, the fibrous pericyst becomes thicker and calcium deposits appear as increasingly extended and confluent granules and laminae, forming in some cases a continuous thick shell. These degenerative aspects have been considered as corresponding to the parasite’s death, which is not. Viable hydatids are most often found within this degenerate material (2). Protoscoleces and brood vesicles generated by the germinal layer can penetrate the chitinous membrane through fissures and then tend to advance into the pericyst (3). Alternatively, there may be germinal islets trapped between the lamellae of the syncytial layer. Once the germinal elements penetrate the pericyst, they may grow inside and then project toward the liver parenchyma as diverticular protrusions surrounded by their own thin pericyst: exogenous vesiculation (Fig. 34.3). In their cavity, they contain growing cysts, favored by easy exchanges through the thin neoformed pericyst and behave as the mother cyst. As there is no connection with the inner surface of primary pericyst they cannot be detected or even suspected with the most careful examination after emptying (Fig. 34.4). The exogenous cyst, while growing, can pull away from the mother cyst and this results in the commonly

biological and pathological basis of modern surgery
The echinococcus, or hydatid cyst, represents the larval stage of Echinococcus granulosus, a 2 to 6 mm long tapeworm. The adult tapeworm consists of a head (scolex) and three following segments (proglottides). The scolex has four suckers and a prominent rostellum armed with a double row of 30 to 36 hooks. Sexual, mature organs and countless eggs are contained in the more distal of the three proglottides. Each egg consists of a shell containing six hook armed embyo hexacant (six hooks). In the adult stage, the tapeworm lives in the gut of the dog, the definitive host. The intermediate animal hosts, where the parasite lives and develops at the larval stage, are sheep, cattle, pigs, and man (considered an “accidental” intermediate host). There are also “sylvatic cycles” of echinococcus which occur in Canada, Alaska, Australia, and other countries with different definitive and intermediate hosts according to the local prevalence of animal species. Human infection is direct or indirect from the dog through the parasite eggs. Once ingested, they hatch and liberate the hexacant embryo. This attaches to and crosses the intestinal mucosa and via the portal system migrates to the liver where the parasite develops into the larval stage which is the hydatid cyst. However, the parasite may cross the portal network and reach the lung, where it may lodge or continue beyond and into the vascular network, toward the various organs by way of systemic arterial vessels.

structure of the cyst
The echinococcus cyst is composed of the wall and contents. The parasite-derived endocyst, namely the wall of the true vesicular metacestode, may consist of either one or two layers. The outer one, laminated or chitinous layer, is a totally acellular membrane, composed of concentric hyaline laminae, permeable to water and electrolytes, which protects the cyst from host enzymes, bile, and bacteria. The inner layer consists of a thin (10–25 mm) germinal or parenchymal layer, which represents the living element of the parasite, composed of an outer basal syncytial layer and an inner nucleated cell layer. Invaginations of the germinal layer form brood capsules each containing 5 to 10 protoscolex. Cyst’s growth leads to the formation in the surrounding parenchyma, in the case of man the liver, of a connective lamina ectocyst or pericyst, able to ensure nutritional exchanges for a long time (Fig. 34.1). Cysts provided of the sole laminated layer are sterile cysts also called univesicular or clear cysts (exempt from vesiculation!) whereas cysts provided with both laminated and germinal layers are fertile or multivesicular cysts. When brood capsules open, protoscoleces are released into the cystic fluid giving raise to daughter cysts: endogenic vesiculation. The origin and cause of daughter cyst formation are not well understood, apart from a non-specific stress of the germinal

308

MANAGEMENT OF HYDATID DISEASE OF THE LIVER
observed pattern of two or more adjacent cysts, or “satellite cysts” (4), separated by a parenchymal septum usually rich in vascular and ductal structures already displaced by the mother cyst. In other frequently observed instances, the exogenous cyst remains in contact with the primary cyst separated by a
L P C G

BC P DC

Figure 34.1 Structure of the liver hydatid cyst. Abbreviations: L, liver; P, pericyst; C, chitinous layer; G, germinal layer; BC, brood capsules or vesicles; P, protoscoleces; DC, daughter cysts (similar to mother cyst).

thin residual septum (“sand-glass-like cyst”), or with the collapse of separating septum, the two cavities communicate through a more or less wide operculum (“sacculations”). As for the presence and frequency of ectogenic vesiculation, the phenomenon is either ignored or largely underestimated (5–9), because of the preference for conservative operations which do not allow their identification, and because in all reported series the recurrence rates of clear and multivesicular cysts are calculated together, thus leading to an under-reporting of the real incidence of recurrence in multivesicular cysts. However, ectogenic vesiculation is recognized in about 30% of radical operations for multivesicular cysts (10,11). Once this phenomenon was identified and quantitatively assessed, its importance was recognized beyond just biological and pathological interest. Consequently, in a large number of patients in whom procedures performed including no removal of the pericyst, this could not be considered effective: actually, only the cyst was resected. Viable, vital parasite foci remained, bound to lead to disease progression. This was incorrectly considered a recurrence attributed to implantation from accidental dissemination of the operating field or reinfection. The latter interpretations, already unconvincing, have lost credibility, based on the observation that the findings of ectogenic vesiculation, compared with the incidence of recurrence in series of conservative surgery, interestingly enough, were similar, at about 30%. This was furthermore confirmed by the fact that the so-called recurrences were practically absent in series of radical surgery (12–14).

(A)

(B)

(C)

(D)

Figure 34.2 Cyst features (total pericystectomy specimens). (A) Multivesicular cyst; (B) yellow-colored cystic membrane for biliary infiltration; (C) calcific cyst of jelly-like necrotic contents and intense biliary infiltration; (D) calcific coarctate cyst of chalk-colored contents and dry clay consistency.

309

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C)

(D)

Figure 34.3 Exogenous vesiculation: microscopic appearance. (A) Brood capsules and protoscoleces contained in a protrusion of germinal layer within the cuticle; (B and C) intrapericyst exogenous vesiculation hydatid membranes within the pericyst of primary cyst; (D) extrapericyst exogenous vesiculation encircled by a new pericyst and protruding in the liver parenchyma adjacent to the mother cyst.

(A)

(B)

(C)

(D)

Figure 34.4 Intra- and extrapericyst exogenous vesiculation. Macroscopic appearance in four total pericystectomy specimens. (A and B) Within the pericyst of open cysts viable daughter cysts are observed, separated from the mother cyst cavity; (C and D) clusters of pedunculated pseudodiverticula, non-communicating with the mother cyst cavity, covered with a thin pericyst and containing daughter cysts.

310

MANAGEMENT OF HYDATID DISEASE OF THE LIVER

diagnosis
Hydatid cyst of the liver may be asymptomatic for years, sometimes for decades. Diagnosis may be accidental, based on an incidental clinical exam that detects swelling when the cyst is located in a palpable abdominal area or, in the case of hepatomegaly, subsequently assessed with other examinations. Liver hydatidosis may be an incidental finding in a plain radiograph of the hepatic region when the cyst is calcified, during a chest radiograph for a raised hemidiaphragm or during US examination performed for other reasons such as gallstones. In children, large hepatic swellings from hydatid cysts are accompanied by evident deformities of the chest involving the last ribs and arches. Apart from a sensation of pressure, a cyst of the liver may cause deep-seated pain at the chest base because of diaphragmatic pleural or peritoneal reactive process. Dyspepsia, possibly from reflexes originating in the periductal nervous network, is not unusual. Cholestasis from major bile duct compression may be responsible for fever, also of high grade. Liver function tests remain normal for a long time. Diagnosis is established using several investigations. Conventional radiology may show a raised hemidiaphragm. In calcific cysts, high-density roundish shadows are readily visualized (Fig. 34.5). Diagnostic Imaging Ultrasonography (US) Ultrasound is the preferred first-line imaging method for hydatid liver cysts. It is non-invasive, low-cost and reproducible, thus suitable for postoperative follow-up or during medical therapy. US supplies precise information on the size, number, location, and vascular and biliary relationships of the cyst as well as on its structure. Cysts are staged according to the content patterns. Based on several studies and classifications (15–18), liver hydatid cysts can be divided into six types:












CE 1 type, a concentric hyperechogenic halo around the cyst containing free-floating hyperechogenic foci corresponding to “hydatid sand” CE 2 type includes multivesicular cysts that have the most characteristic appearance with daughter cysts identified by “honeycomb,” “rosette,” ”spoke wheel,” or “cluster” images (Fig. 34.6) CE 3 type is characterized by a partial or total detachment of the chitinous layer showing the “dual wall,” “water-lily,” and “water snake” signs. CE 4 type includes cysts containing cystic and solid components without visible daughter cysts. CE 5 includes cysts with a matrix or amorphous mass with a solid or semisolid appearance. Calcification in the rim of the host adventitial tissue is common.

Staging is important to allow more objective comparison of different management strategies. US is also useful in postoperative follow-up. Computed Tomography (CT) CT is the procedure of choice when considering radical surgery. Besides precise information on the cyst features, similar to those acquired by US (Fig. 34.7), CT is fundamental in identifying vascular relationships, number, site, and type of the cysts: dual, sand-glass like, with vesiculations (Fig. 34.8). CT is invaluable for the diagnosis of recurring patterns. Spiral CT is presently the gold standard investigation (19). Magnetic Resonance Imaging (MRI) A low-signal intensity rim on T2-weighted magnetic resonance imaging is a characteristic sign of hydatid liver disease that represents the outer collagen-rich laminated membrane of the cyst. When present, daughter cysts are seen as cystic structures attached to the germinal layer that are hypointense relative to intracystic fluid on T1-weighted images and hyperintense on T2-weighted images (20). MRI cholangiography can provide good visualization of the intrahepatic and extrahepatic biliary tree and its relationship with the hydatid cysts and cystobiliary communication (21,22) (Fig. 34.9).

CL type (univesicular cyst), a well circumscribed, round liquid, anechogenic image with a clearly defined wall.

(A)

(B)

Figure 34.5 Plain radiographs. (A) Partial “en brioche” image of diaphragm profile; (B) calcific image pathognomonic of hydatid cyst.

311

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

Figure 34.6 US image. (A) Total detachment of parasite membrane from pericyst; (B) multivesicular hydatid cyst: “rosette” sign.

(A)

(B) Figure 34.7 CT image. (A) Univesicular cyst; (B) “water-snake” sign of membrane detachment.

(A)

(B)

Figure 34.8 CT image. (A) Multivesicular cyst: “honeycomb” or “rosette” sign; (B) calcific cyst of segment VII in contact with the caval vein and causing intrahepatic duct dilation stasis.

312

MANAGEMENT OF HYDATID DISEASE OF THE LIVER

(A)

(B)

Figure 34.9 (A) MRI coronal T1-weighted sequence after DTPA gadolinium injection with visualization of a cyst, about 2 cm in diameter, of segment IV at the level of portal vein bifurcation (in the same patient a bulky hydatid cyst of segments VII, VIII, and V is present); (B) same technique in another patient. Inferior caval vein compression with marked stenosis caused by a bulky cyst of right hemiliver.

Angiography Hepatic artery angiography, inferior caval vein, and hepatic vein venography have been progressively abandoned following the introduction of US and CT-angiography. They may be still of some use in the case of huge cysts to define their relationships with the main parenchymal vessels. Percutaneous cholangiography is contraindicated in liver hydatidosis because of the risk of perforation of the cystic wall and dissemination of hydatid contents. Endoscopic retrograde cholangiopancreatography (ERCP) can be considered the most suitable procedure for the characterization of the common bile duct and the biliary relationships and communication of the cyst. It allows pre- or postoperative papillotomy and bile duct clearing (23–25). Perioperative cholangiography through the remnant of the cystic duct may be of great use to detect the site of cystobiliary fistulas. Radioisotope Imaging Scintigraphic imaging of the liver has been practically abandoned as a preoperative exam. Its use is presently restricted to monitor postoperative liver function. Serology The hydatid cyst secretes and exposes numerous immunomodulatory molecules to the host’s immune system. These molecules modulate both the innate and the adaptive arms of the immune response and appear to target cellular and humoral response. Several techniques for biologic diagnosis and follow-up of human cystic hydatidos are used and there are significant differences in specificity and sensitivity among various tests (26). The diagnostic value of hydatidosis with immunoelectrophoresis ranges between 91% and 94%. Immunoelectrophoresis can be used for post-treatment followup (27). Sensitivities for enzyme-linked immunosorbent assay (ELISA) vary from 64% to 100% depending on the antigens

used (28,29). This test is not suitable for post-treatment follow-up because of its longstanding positivity after successful treatment (30). Western blotting proved to be highly useful in the diagnosis and postsurgical monitoring of hydatidosis patients. Purification of antigens strongly affects the diagnostic value of the test which, however, when using purified fractions enriched in antigens 5 and B and glycoprotein, yield sensitivity and specificity close to 100% (31,32).

complications
During its development, hydatid cysts of the liver may undergo a number of complications, some of them clinically dramatic. Infection Infection of the cystic cavity and its contents is an ill defined and frequently asymptomatic complication (33). It is most likely caused by the penetration of bile into the cavity. Together with the contents, the mother membrane can be destroyed and consequently the altered escavated pericystic wall loses its function of delimiting the infectious process which reaches the hepatic parenchyma. Clinical course and related treatment are, in this case, those of a hepatic abscess (34,35). Rupture Frank rupture (major communication) into the bile ducts should be distinguished from simple biliary communication (minor communication). Minor communications are usually asymptomatic, revealed intraoperatively by a yellowish staining of the cystic content and the finding of biliary leakage in the residual cavity. The true incidence of minor communication is ill defined; and the reported rates range between 28.6% and 70% (36,37). The risk of biliary cyst communication has been reported to be higher in patients with multiple cysts, in patients to multilocular and degenerated cysts, and in patients with cysts near the biliary bifurcation (38). Cyst diameter

313

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
greater than 10 cm has been considered as an independent clinical predictor for the presence of intrabiliary rupture (39). Major biliary communications are defined by a fistula greater than 5 mm diameter or by a cystic communication into the main bile duct (40). The pattern of symptoms usually includes colicky pain accompanied and preceded by jaundice and cholangitic fever (41). Ultrasound and CT are considered first-line diagnostic tools to detect frank cystic rupture into the biliary tree (42). ERCP has proved to be an invaluable tool for the diagnosis and treatment of frank rupture into the biliary tract (43). Intraperitoneal rupture of hydatid cyst is a serious but uncommon complication of liver hydatidosis (44). Rupture may occur spontaneously, but in most cases, the determining cause is blunt trauma. A further cause of peritoneal effusion of hydatid contents is iatrogenic from percutaneous puncture for diagnosis or emptying, or it may occur during surgery. Although this complication may be clinically silent (45), abdominal pain, vomiting, and anaphylactic reactions of varying degree to severe shock, characterized by intense dyspnea, tachycardia, marked hypotension, and urticaria (46–48). The reaction is due to the abrupt release of allergens reabsorbed from the peritoneal serosa and conveyed to the circulation in a sensitized subject. The cyst rupture may be followed by bile peritonitis with either a well defined or insidious clinical pattern. The release of brood cystic fluid into the peritoneal cavity leads to multiple cysts throughout the peritoneal cavity (49). Consequences include occupancy, compression, and displacement phenomena of organs and structures with extremely severe and complex clinical patterns and corresponding general impairment. Benzoimidazole therapy has represented a marked improvement in the treatment of rupture of a cyst into the peritoneal cavity (50). In cases of manifest, diffuse and inoperable peritoneal hydatidosis, ultrasonography-guided emptying of huge, packed cysts is useful to relieve the most severe clinical patterns of compression and dysfunction. Bile-stained cysts located in segments VII and VIII may induce inflammatory adhesions. Necrosis of the latter results in cyst communication with the pulmonary parenchyma at the lung base. Pulmonary inflammation, together with the necrotizing action of bile, cause erosion into a peripheral bronchus with consequent passage of hydatid material and bile into the bronchial tree. Rupture of the cyst into the bronchial tree may be dramatic with abundant expectoration of bile and hydatid material. Daily bile effusion is persistent and increasing, resulting in an extremely severe clinical pattern characterized by cough, abundant expectoration of up to 1000 ml of bile and hydatid contents, fever, and very poor general condition. Bronchopulmonary involvement tends to involve several segments (fatal necrotizing bronchitis) leading to serious injury to the bronchial tree (51,52). Hydatid broncho-biliary fistula is a very severe complication which imposes early, demanding surgery requiring the simultaneous radical treatment of all pathological aspects of broncho-biliary fistula (53).

cyst’s topography ₍location₎
According to location and size, cysts can be divided into parenchymal or superficial and vasculobiliary or deep. In turn the distinction may be based on the predominant vascular relationship. Obviously, the validity of the topographic definition according to hemilivers, sectors, segments, or subsegments adopted by the most reliable classifications is confirmed (54,55). However, since hydatid cysts are spherical and often huge, and since they can be removed sparing the healthy tissue, they do not fit properly into the anatomical distribution usually adopted for cancer surgery. While no intrahepatic expanding neoplasm is free from vasculobiliary contacts, especially the hepatic veins, superficial cysts have vascular relationships limited to minor peripheral structures. Vasculobiliary or deep cysts represent about 75% of cysts that come to surgery and are those with relationships to first-, second-, and third-order branches of hilar elements, the hepatic veins, and the inferior caval vein in both its supraretro and subhepatic segments (Fig. 34.10) (10). First- and second-order portal branches may be involved when deep cysts are located close to the hilum. They are bulky, thus their dissection is difficult both in the case of hemihepatectomy or

(A)

(B)

Figure 34.10 Residual surfaces after removal of deep or vasculobiliary cysts: dissected and preserved vascular and biliary elements are indispensable for the survival and function of parenchymal structures adjacent to the cyst. (A) The caval vein (c) and right hepatic vein with branchings are well visualized; (B) vasculobiliary network of hilar origin distribution.

314

MANAGEMENT OF HYDATID DISEASE OF THE LIVER
the more frequent total pericystectomy. Usually, the cysts of segments VII and VIII, on the right, and segment II on the left, have relationships with the major hepatic veins. In these cases dissection of the cyst from the cava and involved hepatic vein is mandatory. The latter might be ligated and sectioned or more frequently dissected and preserved with adjacent lateral sutures. As for bile ducts, their adhesion to the pericyst is very dense and dissection is difficult. If there is a communication, this requires very careful dissection for effective repair. With reference to the hepatic segments of the Couinaud classification, it is suitable to distinguish vasculobiliary or deep cysts according to the topographical denominations immediately indicative of their predominant relationship with hilar vascular and/or hepatic venous or caval vein structures:
● ●

parasite’s biology and interrelationship between the cyst and liver represent the scientific basis for a rational approach to surgery for hydatid liver disease. The choice of the procedure should not be at the surgeon’s discretion but determined by the cyst’s characteristics. For the two different types of sterile and fertile cysts, indications for surgery are as follows:




Univesicular (sterile), clear cysts, lacking of the pericyst, can be safely treated by conservative surgery. Multivesicular (fertile) cysts at different developmental stage should be treated by radical surgery because of the risk of exogenous vesiculation and because of high rates of postoperative complications.

● ●

Hepatic venous cysts Right or caval intermediate cysts (segments VII, VIII, VI, V) Hilar cysts Central cysts (interportohepatic) (VIII, IV and V) (55).

With the concept of radicality, surgery of liver hydatidosis becomes demanding and therefore selective surgical experience is required as the only means of ensuring a good chance of recovery.

approaches
Access must be generous for two main reasons: first, because of the frequent presence of adhesions of the protruding cyst to adjacent structures and organs, in particular the diaphragm. Second, because of the need for extended liver mobilization to control the vessels and exploit the liver flexibility to reduce the cavities or residual surfaces after pericyst removal. The bilateral subcostal approach, possibly extended depending on the location and size of the cyst(s), to the right as far as the midaxillary line, to the left as far as the lateral end of the rectus muscle, and medially upward as far as the xiphoid process of the sternum (mercedes incision) is the classic approach for liver surgery and hence for the surgical treatment of liver hydatidosis. Median laparotomy may be adequate for cysts located in the left lobe, when the right liver is known to be unaffected. The thoraco-abdominal approach is a no longer used, except in case of hydatid cysts involving the right lung base. Intraoperative General Concerns Protection of the operating field is mandatory before the planned operation on the cyst or before the cyst is emptied. A cautious approach is to apply protection before liver dissection, and when the cyst is protruding from the liver surface and adhering to the adjacent structures and organs. Isolation of the peritoneal and/or pleural cavity to limit the access to the operative field is achieved with dry gauze, preferred by the authors, or soaked in a parasiticidal or hypertonic saline solution, not unanimously considered harmless. During prior emptying of the cyst the gauze pads should be placed around the site of puncture by the trocar. When emptying the cyst, a large caliber trocar is connected with an aspirator by a similarly large non-collapsible tube. As soon as the cyst pressure is relieved and the protruding pericyst tends to collapse, two of its folds are grasped and raised with Allis or ovum forceps. The amount emptied depends on the contents: it will be practically complete in univesicular cysts, more or less partial in multilocular cysts where the hydatid material is abundant and dense. If the pericyst wall is opened with electric cautery, direct emptying is completed through a large tube with a frontal

In turn, hepatic venous cysts are divided into right (VII and VIII), median (IV), left (II); hilar cysts are divided into right (V), anterior median (IV), and posterior (I), left (II and/ or III) (Fig. 34.11). Clearly, there may be some overlap between locations. For example, right hepatic cysts may extend to the midright lobe and left hepatic cysts may be located across the hilum. Obviously, these possibilities do not affect the principles on which the classification is based.

treatment
The desired goals of treatment of liver hydatidosis include complete elimination of the parasite, prevention of recurrent disease, achieved with minimum mortality. In spite of other proposed techniques, surgery remains the first-line treatment. There is, however, considerable disagreement about the surgical technique to be adopted. The major issue of debate is whether complete removal of the pericyst is necessary for the proper care of the disease. The focused concepts about the

1

4

8

7 2 5 3 6 9

Figure 34.11 Topography of vasculobiliary hydatid cysts. 1, 4, 8: Right, median, left hepatic cysts; 3, 5, 6, 9: right, anterior median, posterior median, left hilar cysts; 2: intermediate cyst; 7: interportohepatic cyst.

315

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
opening, connected to a powerful aspirator. If possible, two alternate, separate systems are more suitable because of the unavoidable tube blockage. Now and then paraffin oil aspiration is useful to facilitate the flow of material in the tube. Aspiration is eased by mobilizing with a long ovum forceps the material attached to the endocyst walls or removing big daughter cysts. Clearing of possible communicating sacculations is also mandatory. In the case of adjacent cysts, separated by a relatively thin septum, emptying is performed with the trocar introduced into the cavity through the septum to minimize the risk of dissemination. Several parasiticidal substances have been proposed for sterilization by injection of the cyst before emptying: 2% to 10% formalin solution, 33% hypertonic saline, 0.5% silver nitrate, 10 vol hydrogen dioxide, 1% iodide alcohol solution, and 0.1% cetrimide (56–59). This method should be proscribed for two reasons. First, it appears deceptive to pretend that the entire contents of a multivesicular cyst could be reached by the substance in a few minutes, undergoing the supposed parasiticidal effect (60). In any case, it would be ineffective against vital parasitic elements trapped into the pericyst or developed externally as exogenous vesiculations. Second, practically all solutions markedly damage bile ducts, the cause of severe sclerosing cholangitis, even if in the absence of an open communication (61–66). Consequently, apart from the actual efficacy, the injection of parasiticidal substances, leaving them in the cavity for some time, should be abandoned (59). Some, such as iodine solution, can be applied after emptying of the cyst. Benzoimidazole drugs have been used preoperatively for cyst sterilization; however, surgery must be delayed and this is not attractive to patients and surgeons because of the indefinite outcome of these drugs (67,68). Conservative Procedures Operations not involving the removal of pericyst adhering to the hepatic parenchyma are considered conservative or nonradical. They were devised many years ago and continued because there were no alternatives, with unsatisfactory results, high morbidity, biliary complications in particular, and a high incidence of recurrence (2). At present conservative operations should be limited to the treatment of clear univesicular cysts possibly extended to CE1 cysts. Radical procedures are unnecessary in the former cases because of the absence of the pericyst in clear cysts, while in CE1 cysts the pericyst is so thin and elastic making it possible to exhaustively inspect the underlying parenchyma through it. Exogenous vesiculation or biliary fissurations cannot be missed; the latter evidenced by the color of cystic fluid. However, the large size and high pressure of these cysts may exceptionally be responsible for biliary wall impairment which results, after the cyst removal, in postoperative bile leakage. In these rare cases, bile will be seen from the drainage tube for no longer than 15 to 20 days, which is not following conservative operations on cysts with thick or calcific pericyst that hinders the reduction of the residual cavity and the closure of the cystobiliary communication. After emptying and clearing the cyst cavity, its size and penetration in the depth of liver, especially toward the hilum, the hepatic veins, and the retrohepatic caval vein, are assessed. The relationships with the hepatic ducts and biliary communication should be carefully evaluated. In fact, it becomes evident only after emptying and removal of membrane. The extent of pericyst resection is based on how much of it is protruding or how close it is to the hilar and adjacent major vasculobiliary structures. A pericyst margin adequate for subsequent suturing should be considered. Resection is performed with electric cautery and the help of Allis forceps on the residual margin; particular attention should be paid to adjacent structures at risk (Fig. 34.12). The analogy of the method with the “dome saillant” resection proposed by Lagrot in the 1950s (69) and still used (70) is only apparent because here it is limited to clear cysts with a thin and soft pericyst after wide-field exposure, complete liver mobilization, control of hilar and caval structures. All these measures enable the resection of ample pericyst surfaces up to two-thirds of it, while suturing of residual margins exploits the flexibility of the liver once free of its ligaments. The control of bile leaks is very important. Bile transudation may occur on the resection margins from small orifices, which must be sutured to prevent bile collection within the cavity. As for the search for major communications, the induction of biliary hypertension by compression of the distal hepatic pedicle over the duodenum, squeezing the gallbladder at the same time, may be useful. Then even minimal communication is evidenced by a drop or flow of bile. The use of dyes must be considered pointless because the site of the leak is always recognizable after having cleansed the residual cavity with a pad. However, these are not the typical patterns of univesicular cysts. Once the residual cystic wall has been controlled, closure

Figure 34.12 Resection of protruding pericyst during conservative surgery. The left-hand fingers protect the inferior caval vein.

316

MANAGEMENT OF HYDATID DISEASE OF THE LIVER
is performed by suturing the resection edges. Approximation can be longitudinal, transverse, or oblique, but traction should be prevented and circulation in the adjacent parenchyma should not be jeopardized. A double, continuous inverting suture is made with atraumatic 3/0 prolene. In many cases the procedure corresponds to the canalization of the cavity on a rubber tube, advanced outside the cavity in the most suitable position through a counter incision at the level of the hypochondrium or flank. The progress of postoperative cyst collapse until complete obliteration of the cavity can be documented by imaging. Drainage is preferred because in spite of the absence of obvious biliary communications, some days after surgery a bile leakage might occur which, if drained, heals with no further consequences. In theory, a cavity, reduced as described above, or even left in its original size, could be closed without drainage and heals with no complications. However, the actual possibility of bile collection with the risk of infection and abscess formation, argues in favor of the placement of drains. The use of omentum (71–73) to fill the cavity or cover the residual surface should be avoided. As for the omentum, an excellent barrier against infections, is not equally effective against bile and may even become necrotic. When packed into the cavity, it hinders the interpretation of US or CT images in long-term follow-up then possibly concealing the occurrence of recurrences (74,75). Radical Procedures Radical procedures include major hepatic resection and total pericistectomy. Hepatic resection since ever has not been considered the principal technique to resort to when treating liver hydatidosis. The notable development of liver resective surgery in more recent years has not changed this trend so that the rate of hepatic resections rarely exceeds 10% (76,77). Technically these resections are similar to those performed for other indications, except for the possible extension of the cyst beyond the anatomical limits of the hemiliver or sector to be removed and the biliary relationship on which the indication was based (Fig. 34.13). The first occurrence may involve difficult pericyst dissection from vasculobiliary structures supplying territories to be preserved. Furthermore, resections are adopted in case entire sectors, lobes, or hemilivers are destroyed by the cyst(s) and interruption of main bile ducts would make their repair questionable because of a very high risk of serious cholerrhagia. With the exception of these particular cases the general policy should be to avoid liver resection in order to spare as much health hepatic parenchyma as possible. Wedge resections and liver transplantation should be mentioned in passing. Actually while the former might find an occasional indication in case of minute marginal anterior cysts, the use of liver transplantation should be considered to the utmost anedoctical (78).

(A)

(B)

(C)

(D)

Figure 34.13 Right hepatectomy extended to segment IV and caudate lobe. (A) Bulky cyst mass; (B) the residual surface is sutured; (C) surgical specimen; (D) the gallbladder is packed with hydatid material.

317

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Total pericystectomy (or cystopericystectomy) is presently considered to be the ideal radical operation suitable for the requirements of radicality for the hydatid disease, with maximal preservation of hepatic tissue and complete early recovery. Its feasibility is similar to that of liver resections. The same basic training is required, enhanced by a specific experience. The operation was conceived and proposed by Napalkoff (79) in 1927 and again described in 1936 by Melnikoff (80). However, in subsequent experiences, results were unsatisfactory and even disastrous due to hemorrhage; thus, following unanimous disapproval, it was practically abandoned. Costantini applied it again in 1950s (81) and Yovanovitch in 1959 (82), with indications for peripheral cysts, distant to porta hepatis. Bourgeon, the most convinced supporter, advocated it in following years, even if, in many cases, he favored partial pericystectomy (83,84). Since the early 1970s it was understood that the persistence of pericyst, with its close biliary relationships, more or less wide fissurations and true ruptures, represented the main cause of frequent postoperative complications: abundant, prolonged bile leakage, infections, longstanding residual cavities, biliary fistulas (85,86). The correlation of persistent hydatid material with the frequent recurrences with which surgery of hydatidosis seemed to be inevitably burdened was noted only later on (33,87). For all these reasons and because of the development of hepatic surgery the operation has gained widespread acceptance (88–90). Pericystectomy can be performed either as a closed or as an open procedure. In the first case, en bloc removal of the pericyst and its contents is a safe operation with respect to the risk of contamination. The closed procedure is mainly indicated in case of superficial cysts and unilobar deep cysts. Open pericystectomy should be performed every time there is a risk of breaking the cyst wall, in case of cysts of the hepatic dome, strictly adherent to the hepatic veins and vena cava, and in case of deep intraparenchymal cysts with interporto-cava location. However, protective measures are necessary also before dissection of a closed cyst. The latter procedure is more elegant, rapid, and simpler, but in the case of bulky, deep cysts or those which have ruptured into the common bile duct, the procedure may be risky, thus operations on the open cyst are preferable and necessary. This enables the dissection of the pericyst from vasculobiliary structures, even in the most difficult conditions. When the cyst protrudes at any site of the liver surface, dissection is developed along the transition line, sometimes delineated by a groove, between the pericyst and the parenchyma. Concomitant, chronic, non-parasitic liver disease, or consequent to hepatic vein stasis caused by the cyst, a true Budd–Chiari syndrome, creates major difficulties. The smaller vascular branches entering the pericyst must be electrocoagulated or ligated and dissected. Step by step hemostasis is a determinant caution of the procedure, or else consequent bleeding would hinder the operatory field and cause an unnecessary and considerable blood loss. Dissection of large vessels from the pericyst should be centriperipheral and along the course of vascular and biliary structures, following the direction of its emerging or confluent branches (Fig. 34.14). In short, when entailing the hepatic veins the direction of the dissection should be from the apical liver convexity toward the free margin. Dissection of the hilar elements should be developed from the hilum toward the periphery. Therefore, in vessel dissection, the surface corresponding to the obtuse angle the branches describe when emerging or merging should be preferred. The large vessel is more readily dissected and longitudinal lacerations are prevented when scissors are trapped into the acute angle between it and its branch. The tight adhesion of large vessels to the pericyst might encourage the finger fracture procedure which would result, in these cases, incorrect and hazardous. The use of the ultrasound dissector, which favors the visualization of the vascular network, and hemostasis seems advantageous, while dissection of large vessels from pericyst is still feasible (91,92). Dissection is circumferential yet conditioned by the cyst location. The dissected parenchyma tends to flatten its spherical hollow surface, leaving a largely naked area previously adherent to the pericyst. The dissection of very deep and/or bulky cysts is hazardous because most of the operatory field is not under visual control. For this reason, and to prevent excessive manipulation of the cyst, its emptying is suitable and then open cyst pericystectomy can be safely performed. The same procedure should be adopted in cases of cysts with pedunculated protrusions from exogenous vesiculations. They have a very thin pericyst, which may breach specially at the pedicle level. As a general rule, whenever the risk of cyst rupture, prompt emptying of the cyst should be performed and the operation carried on as an open procedure. By folding the pericyst at the level of the dissection plane the procedure is facilitated, especially if the dissected pericyst is sectioned in strips, subjecting each to tension independently (Fig. 34.15). If the pericyst is not very calcific, a further very useful “trick” is

Figure 34.14 Total pericystectomy. Dissection of pericyst from vasculobiliary structures should be along the presumed centriperipheral direction (following the direction of emerging and merging branches).

318

MANAGEMENT OF HYDATID DISEASE OF THE LIVER
to carefully incise it with a scalpel on its internal surface to reach the adhering parenchyma. With a cross- or star-shaped incision, and by lifting backward each strip with Allis or Kocher forceps and dissecting from different sides, apparently unfavorable situations are resolved and dissection can be completed (Fig. 34.16). Access to mid-sized cysts through one of the hepatic fissures is another very effective and in some cases resolving technique. Once major vasculobiliary structures are reached with dissection, the corresponding fissure is identified, opened, and extended to the cyst wall. The cyst turns, this way, seems to become more superficial and then accessible from several sides (Fig. 34.17). Retraction of intersectorial surfaces resolves the problem of the difficult access to the deep hemisphere of the cyst. Therefore, operations on the closed cyst are facilitated with relevant vascular relationships as the interportal liver becomes accessible (Fig. 34.18). Despite all precautions, vascular lacerations may occur during pericystectomy, especially the hepatic veins. The dissection may be interrupted, in this case, and temporary hemostasis obtained by compressing the parenchyma against the pericyst while progressing with the operation on a different side. Also direct digital compression on the bleeding vessel will allow the surgeon to

Figure 34.15 Total open pericystectomy. Stripping of pericyst and traction on each strip by folding facilitates deep dissection of vasculobiliary structures also in case of calcific pericyst. Figure 34.17 Total pericystectomy. In deep, non-emerging cysts, opening of the corresponding fissure is very useful. The cyst becomes accessible from several sides, somewhat similar to more superficial cysts.

Figure 34.16 Total open pericystectomy. Cautious full depth incision of pericyst with a lancet on the cavity bottom allows access to vessel dissection from several directions, even in the case of very thick cysts.

Figure 34.18 Total pericystectomy of deep cysts with multiple vasculobiliary relationships such as interportohepatic cysts is facilitated by opening the median fissure. Venous stasis from compression and compensatory collaterals may be present.

319

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
continue the dissection and achieve a better exposure of the breach. Once adequately identified and exposed, the laceration is sutured, yet paying attention to maintain vessel patency. Pringle’s maneuver is not suitable for pericystectomy because of the long time usually required for pericyst dissection. However, during the critical phase of the procedure, intermittent clamping followed by brief period of reperfusion may come in very handy. Clamping of hepatic veins is rarely necessary. In particular cases preventive isolation and encircling of the subhepatic and suprahepatic vena cava above the confluence of hepatic veins may turn out to be a wise precaution. Dissection of the right hepatic vein and of the right edge of the vena cava is a cornerstone of the procedure, especially for cysts of segment VII and VIII. It should be kept in mind that a hepatic vein laceration proximal to the confluence is to be feared more for air embolism than for bleeding. Bleeding, possibly occurring during total pericystectomy, represented the principal cause of surgical aversion to this operation. According to the site of the cyst, 2 to 4 units of blood should be available for transfusion. Intraoperative recovery, obviously limited to the sterile phases of the operation, offers great advantage. In the course of pericystectomy the surgeon might decide to leave behind one or several areas of the pericyst in case their excision is felt to be too hazardous because of their adherence to vascular structures. The use of this fairly common solution will progressively lessen as the experience of each surgeon increases. Such a decision is anyway advisable in some crucial areas such as the confluence of hepatic veins into the caval vein and at hilar structures, in particular contralateral ones, and in case of cysts extending beyond the involved hemiliver. This is also the not rare case of the retrohepatic caval vein protruding in large cystic cavities of the right lobe. During pericystectomy, a limited part of the neighboring hepatic parenchyma may turn ischemic because of impairment of blood supply. In this case, the extension of the involved region should be carefully evaluated for a worthwhile and, if no recovery occurs, resected together with pericyst removal. Vascular impairment of a wider extension of hepatic parenchyma is, of course, an entirely different problem, but as a rule this should not be the case. The residual liver surfaces, after pericystectomy, are characterized by a pattern of protruding hepatic veins or Glissonian’s vessels. The closure of the residual cavity, even if easier after the excision of deep cysts, does neither entail main difficulties after the removal of superficial cysts nor imply any vascular impairment. As a matter the procedure does not correspond as much to the closure of a residual cavity as rather to the simple approximation of the residual, smooth liver surfaces. If approximation is complete and there is no reason to doubt for bile loss, drainage in the residual space may be omitted. If the pericyst has been completely removed and bile leakage excluded, then an omental flap can be used on residual surfaces to prevent adhesion of displaced intestinal loops. In the course of operations on hydatid cysts complementary surgery may be required for “parahydatid” biliary pathology such as cholelithiasis; en bloc cholecystectomy is routinary performed in case of cysts of segments IV and V. All patients with large biliary cyst communication should undergo operative cholangiography and exploration of the main duct is mandatory in cases of presence of common bile duct filling defects on cholangiography. T-tube drainage is usually added in case of the presence of hydatid debris in the common bile duct. Access to the common bile duct is through a choledochotomy and less commonly transduodenal (Fig. 34.19). In fact, surgical trans-duodenal papillosphincterotomy has been, more or less completely, replaced by endoscopic papillosphincterotomy. Laparoscopy Limited area of manipulation, difficulty in controlling spillage during puncture, difficulty in aspirating the thick, degenerated cyst contents, putting pressure by the pneumoperitoneum on the hydatid cyst and consequent increased risk of hydatid fluid contamination have been reported as the main disadvantages of the laparoscopic approach (93). However several authors have performed conservative procedures, mainly cistotomy and drainage, on superficial cysts located in the left lobe and in the anterior aspect of the right lobe (94,95). Laparoscopic pericystectomy performed on selected cases has been reported as a safe and effective procedure (96). PAIR Since 1985, Puncture Aspiration Injection Reaspiration (PAIR) has been proposed as an alternative to surgery (97). After percutaneous puncture under ultrasonographic guidance,

Figure 34.19 Exploration and cleansing of biliary tract. Through papillosphincterostomy the spoon for stones or a probe can be carefully advanced to identify the biliary breach and specify the type of communication, whether lateral or terminal, with the cyst cavity.

320

MANAGEMENT OF HYDATID DISEASE OF THE LIVER
aspiration of as much as possible of the cyst content is performed; the residual cavity is then filled with a protoscolicide, usually ethanol, reaspired 10 minutes later. Detailed practical guidelines have been defined after a careful evaluation of the technique by the WHO-IWGE (98). As in the PAIR technique, the daughter cysts and the germinal membrane would remain inside the cavity; this technique is advocated for uncomplicated univesicular cysts, but not for multivesicular, so-called mother and daughter cysts. In accordingly selected series, complete disappearance of the cyst has been reported in 48% of cases (99). PAIR is contraindicated if there is a communication between the cyst and the biliary tree because of the risk of sclerosing cholangitis. Vacuum aspiration and dissection of the endocyst non-drainable material through a stiff sheath, introduced into the cavity at the site of puncture after removal of the needle, is an alternative method to PAIR called PEVAC (100).Radiofrequency ablation, mainly focused on solid hydatid cysts, has been proposed as a further alternative to PAIR (101). Only preliminary results on the efficacy of this technique are so far available as only very few patients have been treated with this technique. Medical Therapy Benzimidazole carbamates (mebendazole and albendazole) are antihelminthic drugs that affect the parasite viability, mainly, by impairing its glucose uptake. Mebendazole was introduced first (102), but albendazole became the drug of choice because of its better absorption and better clinical results (103). These results are achieved after long treatment (104). Adverse events of this treatment have been reported in about 10% of patients treated (105). General complaints are headache, nausea, anorexia, vomiting, abdominal pain, and itching. A transient increase in liver enzymes may be observed in the first weeks of treatment. For clinical practice, albendazole should be administered in a dose of 10 mg/kg twice daily during a meal in four 1-month cycles with a 15-day rest, or 10 to 12 mg/kg/day continuously for 3 months. It should not be associated with drugs that reduce gastric acidity (106). Factors affecting the efficacy of benzimidazoles have not been well defined, but it is known that penetration of drug across the cyst walls depends on the nature of the cyst. Young cysts without thick, fibro calcified pericyst are more sensitive to drugs (107,108). It is difficult to understand how the drug could overcome the barrier of a dense fibrotic or calcific pericyst, up to 0.5 cm thick, and kill the hydatid material packed into the cavity. Moreover, it is hardly believable that exogenous vesiculations within the pericyst might be reached by the drug. Recurrence following albendazole therapy occurs in at least 20% to 30% of responsive cases (108,109), to which a further 20% of patients, considered negative in whom no change was visualized, should be added. Preoperative albendazole treatment has been suggested in order to lower cyst viability. However, contrasting outcomings have been reported possibly because viability tests on the surgical specimen cannot be considered conclusive as confirmed by the development of parasites from culture of cystic fluid shown to be negative on direct microscopy (110). Although no conclusive data on the efficacy of perioperative profilaxis are available, it is generally advised at least 2 days before surgery. Similarly, postoperative treatment is recommended for 6 months in case of intraoperative hydatid spillage (98).

key points






Complications of hepatic hydatidosis include: Metastatic hydatid Secondary bacterial infection Intrabiliary rupture Intraperitoneal rupture Bronchobiliary fistula Diagnosis of hepatic hydatidosis: Incidental finding (in patient from endemic region) Abdominal mass Calcified hepatic cyst on the plain abdominal photograph (AXR) Ultrasound/CT/MRI Hydatid serology/Casoni skin test. Preoperative management: Systemic albendazole/mebendazole ERCP (exclude cystobiliary fistula) Protection of operative field before surgical emptying of cyst contents Sterilization of cyst cavity

references
1. Akinoglu A, Arparslan, Kaza K. Hydatid disease of the liver: prevention of postoperative biliary fistula. Arch Hidatid 1991; 30: 649–55. 2. Chigot JP, Langlois P, Teboul F, et al. Le traitement des kystes hydatiques du foie. Ann Chir 1986; 40: 177–82. 3. Euzeby J. L’echinococcose larvaire. Rev Med Vet 1955 ; 106 : 456-468. 4. Kalovidouris A, Voros D, Gouliamos A, Vlachos L. Papavasiliou C. Extracapsular (satellite) hydatid cysts. Gastrointest Radiol 1992; 17: 353–6. 5. Bourgeon R, Catalano H, Guntz M. La pE9rikystectomie dans le traitement des kystes hydatiques du foie. J Chir 1961; 81: 153–74. 6. Belli L, Del Favero E, Marni A, Romani F. Resection versus pericystectomy in the treatment of hydatidosis of the liver. Am J Surg 1983; 145: 239–42. 7. Moreno Gonzales E, Jovez Navalon JM, Landa Garcia JI, et al. Surgical management of liver hydatidosis. 10-year experience with 269 patients. It J Surg Sci 1985; 15: 267–73. 8. Debesse B, Dujon A. La pE9rikystectomie au plus prE8s dans le traitement du kyste hydatique du foie. Ann Chir 1987; 41: 646–51. 9. Moreno Gonzales E, Rico Selas P, Bercedo Martinez J, et al. Results of surgical treatment of hepatic hydatidosis: current therapeutic modifications. World J Surg 1991; 15: 254–63. 10. Tagliacozzo S, Daniele GM, Pisano G. Pericistectomia totale per echinococco epatico. Arch Atti Soc It Chir 1979; 1: 657–712. 11. Tagliacozzo S, Daniele GM, Pisano G. Total pericystectomy for hydatid disease of the liver. 6B0 World Congr Coll Int Chir Dig, Lisbona 1980, abst SP5–03. 12. Magistrelli P, Masetti R, Coppola R, et al. Surgical treatment of hydatid disease of the liver. A 20 year experience. Arch Surg 1991; 126: 518–23. 13. Moreno Gonzales E, Rico Selas P, Martinez B, et al. Results of surgical treatment of hepatic hydatidosi: current therapeutic modifications. World J Surg 1991; 15: 254–63. 14. Aydin U, Yazici P, Onen Z, et al. The optimal treatment of hydatid cyst of the liver: radical surgery with a significant reduced risk of recurrence. Turk J Gastroenterol 2008; 19: 33–9.

321

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
15. Beggs I The radiological appearance of hydatid disease of tha liver. Clin Radiol 34: 555–63. 16. Gharbi HA, Hassine W, Brauner MW, Dupuch K. Ultrasound examination of the hydatic liver. Radiology 1981; 139: 459–63. 17. Hassine W, Dupuch K, Gharbi HA. Aspects E9chographiques de l’hydatidose. J Radiol 1979; 60: 660–6. 18. Lewall DB, McCorkell SJ. Hepatic echinococcal cysts: sonographic appearance and classification. Radiology 1985; 155: disease of the liver. Clin Radiol 198; 34: 555–63. 19. Bobek-Billewicz B, Szurowska E, Zapasnik A, et al. Localization of focal liver lesions to specific heapatic segments. Comparison of multiphase spiral CT and MR imaging. Folia Morphol (Wars) 2002; 61: 291–7. 20. Pedrosa I, Saiz A, Arrazola J, Ferreiros J, Pedrosa CS. Hydatid disease: radiologic and pathologic fatures and complications. Radiographics2000; 20: 795–817. 21. Basaran C, Karcaaltincaba M, Akata D, et al. Fat containing lesions of the liver: cross sectional imaging findings with emphasis on MRI. AJR 2005, 184: 1103–10 22. Little AF, Lee WK, Mathison K MR cholangiography in the evaluation of suspected intrabiliary rupture of hepatic hydatid cys. Abdom Imaging 2002; 27: 333–5. 23. Simsek H, Ozaslan E, Sayek I, et al. Diagnostic and therapeutic ERCP in hepatic hydatid disease. Gastrointest Endosco 2003; 58: 384–9. 24. Galati G, Sterpetti AV, Caputo M, et al. Endoscopic retrograde cholangiography for intrabiliary rupture of hydatid cyst. Am J Surg 2006; 191: 206–10. 25. Cicek B, Parlak E, Disibeyaz S, et al. Endoscopic therapy of hepatic hydatid cyst disease in preoperative and postoperative settings. Dig Dis Sci 2007; 52: 931–5. 26. Sbihi Y, Rmiqui A, Rodriguez-Cabezas MN, et al. Comparative sensitivity of six serological tests and diagnostic value of ELISA using pirified antigen in hydatidosis. J Clin Lab Anal 2001; 15: 14–8. 27. Varela-Diaz VM, Coltorti EA, de Zavaleta O, et al. Immunodiagnosis of human hydatid disease: applications and contributions to a control program in Argentina. Am J Trop Med Hyg 1983; 32: 1079–87. 28. Coltorti EA. Standardization and evaluation of an enzyme immunoassay as a screening test for the seroepidemiology of human hydatidosis. Am J Trop Med Hyg 1986; 35: 1000–5. 29. Sadjjadi SM, Abidi H, Sarkari B, Izadpanah A, Kazemian S. Evaluation of enzyme linked immunosorbent assay, utilizing native antigen B for serodiagnosis of human hydatidosis. Iran J Immunol 2007; 4: 167–72. 30. Zhang W, Li J, McManus DP. Concepts in immunology and diagnosis of hydatid disease. Clin Microbiol Rev 2003; 16: 18–36. 31. Sbihi Y, Jansse D, Osuna A. Serologic recognition of hydatid cyst antigens using different purification methods. Diagn Microbiol Infect Dis 1996; 24: 205–11. 32. Logar J, Soba B, Kotar T. Serological evidence for human cystic echinococcosis in Slovenia. BMC Infect Dis.2008; 9: 63–72. 33. Tagliacozzo S, Daniele GM, Pisano G. Total pericistectomy for hydatid of the liver. 6 (th) World Congress Coll Int Chir Dig. Lisbona, abst SP5–03. 34. Monterola C, Barroso M, Vial M, et al. Liver abscess of hydatid origin: clinical features and results of aggressive treatment. ANZ J Surg 2003; 73: 220–4. 35. Gaibatov SP, Gaibatova DS. Clinical finding and treatment of festered hepatic echinococcosis. Khirurgiia (Mosk) 2006; 6: 16–8. 36. Ozmen V, Igci A, Kebudi A, et al. Surgical treatment of hepatic hydatid disease. Can J Surg 1992, 35: 423–7. 37. Assadourian R, Bricot R, Djilalli G. Les kystes hydatiques du foie 1984. Rev Ass Fr Chir 1984; 33: 15–39. 38. Kayaalp C, Bostanci B, Yol S, Akoglu M. Distribution of hydatid cysts into the liver with reference to cystobiliary communications and cavityrelated complications. Am J Surg 2003; 185: 175–9. 39. Atli M, Kama NA, Yuksek YN, et al. Intrabiliary rupture of a hepatic hydatid cyst: associated clinical factors and proper management. Arch Surg 2001; 136: 1249–55. 40. Bourgeon R. L’ouverture des kyststes hydatiques aux voies biliares intrhepatiques. Lyon Chir 1985; 81: 161–4. 41. Lygidakis NJ. Diagnosis and treatment of intrabiliary rupture of hydatid cyst of the liver. Arch Surg 1983; 118: 1186–89. 42. Valle-Sanz YD, Lorente-Ramos RM. Sonograaphic and computer tomographic demonstration of hydatid cysts communication with the biliary tree. J Clin Ultrasound 2004; 32: 144–8. 43. Bilsel Y, Bulut T, Yamaner S, et al. ERCP in the diagnosis and management of complications after surgery for hepatic echinococcosis 2003; 57: 210–3. 44. Sozuer EM, Ok E, Arslan M. The perforation problem in hydatid disease. Am J Trop Med Hyg 2002; 66: 575–7. 45. Abdel Hameed AA, Abu Aisha H. Uneventful intraperitoneal rupture of a hepatic cyst: a case report Trop Geogr Med 1987, 39: 80–2. 46. Barros JL. Hydatid disease of the liver. Am J Surg 1978; 135: 597–600. 47. Marcos Sanchez F, Turabian Fernandez JL, Caballero Gomez F, et al. Reacciones alergicas-shock anafilactico como manifestacion de enfermidad hidatica hepatica. Rev Clin Esp 1984; 173: 49–52. 48. Giulekas D, Papacosta D, Papacostantinou C, Barbarousis D, Angel J. Recurrent anaphylactic shock as a manifestation of echinococcosis. Scand J Thorac Cardiovasc Surg 1986; 20: 175–7. 49. Lewall DB, McCorkell SJ. Rupture of echinococcal cysts: diagnosis, classification and clinical implications. AJR 1986; 146: 391–94. 50. Morris DL, Chinnery JB, Hardcaster JD. Can albendazole reduce the risk of implication of spilled protoscolices? An animal study. Trans R Soc Trop Med Hyg 1986; 80: 481–4. 51. Gerazounis M, Athanassiadi K, Metaxas E, Athanassiou M, Kalantzi N. Bronchobiliary fistulae due to echinococcosis. Eur J Cardiothorac Surg 2002; 22: 306–8. 52. Gomez R, Moreno E, Loinaz C, et al. Diaphragmatic or transdiaphragmatic thoracic involvement in hepatic disease: surgical trends and classification. World J Surg 1995; 19: 714–9. 53. Tocchi A, Mazzoni G, Miccini M, et al. Treatment of hydatid bronchobiliary fistulas: 30 years of experience. Liver Int 2007; 27: 209–14. 54. Couinaud C. Le foie. Etudes anatomiques et chirurgicales. Paris: Masson; 1957; 9 : 335–74. 55. Goldsmith NA, Woodburne RT. Surgical anatomy pertaining to liver resection. Surg Gynecol Obstet 1957; 105: 310–18. 56. Cirenei A, Bertoldi I. Evolution of surgery for liver hydatidosis from 1950 to to-day: analysis of personal experience. World J Surg 2001; 25: 87–92. 57. Agaoglu N, Turkilaz S, Arslan MK. Surgical treatment of hydatid cysts of the liver. British J Surg 2003; 90: 1536–41. 58. Meymarian E, Luttermoser GW, Frayha GJ, Schwabe DVM, Prescott B. Host parasite relationship in echinococcosis. Laboratory evaluations of chemical scolicides as adjuncts hydatid surgery. Ann Surg 1963; 158: 211–5. 59. Saidi F. A new approach to the surgical treatment of hydatid cyst. Ann R Coll Surg Engl 1977; 59: 115–28. 60. Frayha GJ, Bikhazi KJ, Kachachi TA. Treatment of hydatid cysts (Echinococcus granulosus) by Cetrimide (AE). Trans R Soc Trop Med Hyg 1981; 75: 447–50. 61. Morris DL. Surgical management of hepatic hydatid cyst. In: Morris DL, Richards KS, eds. Hydatid Disease. Current medical and surgical management. Philadelphia: Butterworth-Heinemann 1992: 57–75. 62. Little JM. Hydatid disease of the liver. In: Hadfield J, Hobsley M, Tressure T, eds. Current Surgical Practice, Vol 5. London: Edward Arnold, 1990: 146–61. 63. Khodadadi DJ, Kurgan A, Schmidt B. Sclerosing cholangitis following the treatment of echinococcosis of the liver. Int Surg 1981; 66: 361–2. 64. Cohen-Solal JL, Eroukmhanoff P, Desoutter P, et al. Cholangite sclerosante survE9nue aprE8s traitement chirurgical d’un kyste hydatique du foie. Sem HF4pitaux Paris 1983; 59: 1623–4. 65. Teres J, Gomez-Moli J, Bruguera M, et al. Sclerosing cholangitis after surgical treatment of hepatic echinococcal cysts. Report of three cases. Am J Surg 1984; 148: 694–7. 66. Belghiti J, Benhamou JP, Houry S, Grenier P, Huguier M, Fekete F. Caustic sclerosing cholangitis. A complication of the surgical treatment of hydatid disease of the liver. Arch Surg 1986; 121: 1162–5. 67. Stubbs RS. Management of hydatid disease: scolicidal agent instillation versus perioperative chemotherapy. HPB Surgery 1990; 2(Suppl): 155.

322

MANAGEMENT OF HYDATID DISEASE OF THE LIVER
68. Erzurumlu K, Ozdemir M, Mihmanli M, Cevikbas U. The effect of intraoperative mebendazole-albendazole applications on the hepatobiliary system. Eur Surg Res 1995; 27: 340–5. 69. Falagas M, Bliozotis I. Albendazole for the treatment of human echinococcosis: a review of comparative clinical trials. Am J Med Sci 2007; 334: 171–9. 70. Monterola M, Monsilla JA, Fonseca, F. Preoperative albendazole and scolices viability in patients with hepatic echinococcosis. World J Surg 2005; 29: 750–3. 71. Lagrot F, Coriat P. Justification et valeur de la resection du dome saillant dans les kystes hydatiques du foie. Lyon Chir 1959 ; 55: 826–37. 72. Moumen M, Elalauni ME, Mehane M, Jami D, Mokhtari M, El Fares. La rE9section du dF4me saillant du kyste hydatique du foie. A propos de 360 cas. J Chir 1990; 127: 83–6. 73. Goinard P, Note D, Girardot M. IIe: sur le traitement de KHF (kyste hydatique du foie). L’E9piploonplastie intracavitaire. Presse ME9d 1950; 58: 1203–5. 74. Xu M-Q. Diagnosis and management of the liver: hydatidosis complicated with biliary fistula. HPB Surgery 1990; 2 (Suppl): 153. 75. Safioleas M, Misiakos E, Manti C, Katsikas D, Skalkeas G. Diagnostic evaluation and surgical management of hydatid disease of the liver. World J Surg 1994; 18: 859–65. 76. Chaimoff C, Lubin E, Dintsman M. The postoperative appearance of the liver on scanning following omentopexy of the hydatid cyst. Int Surg 1980; 65: 331–3. 77. Beggs I, Walmsley K, Cowie AGA. The radiological appearances of the liver after surgical removal of hydatid cyst. Clin Radiol 1983; 34: 555–63. 78. Moreno-Gonzales E, Loinaz Segurola C, Garcia Urena A, et al. Liver transplantation for Echinococcus Granulosus hydatid disease. Transplantation 1994 ; 58 : 797–800. 79. Napalkoff N. A propos de la decortication des Kystes hydatiques. Rev Chir 1927; 65: 524–8. 80. Melnikoff AW. Sur la chirurgie des kystes hydatiques. J Chir 1936; 47: 197–219. 81. Costantini H, Bourgeon R, Pantin JP, Rives J. Des indications de l’ablation du sac pE9ri parasitaire dans le traitement des kystes hydatiques du foie. La kystectomie de routine des kystes suppures du foie. J Chir 1950; 66: 177–89. 82. Yovanovitch BY. Place de la kystectomie dans le traitement des kystes hydatique du foie. Ann Chir 1959; 13: 31–6. 83. Bourgeon R. Les bases physiopathologiques necessaires à la conduite therapeutique du kyste hydatique du foie, en particulier par perikystectomie. In: Cirenei A, Hess W, eds. Chirurgie du foie, des voies biliares et du pancreas. Padova: Piccin, 1977: 21–35. 84. Bourgeon R, Guntz M, Catalano H, Alexandre JH, Mouiel J. Incidence de la topographie sur le traitement des kystes hydatiques du foie. J Chir 1964; 88: 375–88. 85. Tagliacozzo S. Total pericystectomy and complementary papillostomy in the treatment of hepatic hydatid cyst. Proc XX Biennal World Congr Int Coll Surg, Atene, 1976: 630–1. 86. Tagliacozzo S. Typical and atypical (total pericystectomies) resections in the surgical treatment of hydatid disease of the liver. Surg It 1977; 7: 133–42. 87. Tagliacozzo S. Exogenous vesiculation and radical treatment of hepatic hydatid cyst. 30(th)Congr Soc Int Chir, Hamburg, 1983: 51. 88. Magistrelli P, Masetti R, Coppola R, et al. Surgical treatment of hydatid disease of the liver. A 20-year experience. Arch Surg 1991; 126: 518–23. 89. Moreno Gonzales E, Rico Selas P, Martinez B, et al. Results of surgical treatment of hepatic hydatidosis: current therapeutic modifications. World J Surg 1991; 15: 254–63. 90. Aydin U, Yazici P, Onen Z, et al. The optimal treatment of hydatid cyst of the liver: radical surgery with a significant reduced risk of recurrence. Turk J Gastroenterol 2008; 19: 33–9. 91. Little JM. Hydatid disease of the liver. In: Hadfield J, Hobley M, Tressure T, eds. Current Surgical Practice, Vol. 5. London: Edward Arnold 1990: 146–61. 92. Vicente E, Devesa JM, Nuno J, Fernandez JM, Angel V. Progress in the surgical treatment of hepatic hydatid cysts. 34th World Congr ISS/SIC, Stockholm, 1991, A139: 255. 93. Kayalp C. Hydatid cyst of the liver. In Blumgart LH, ed. Surgery of the Liver, Biliary Tract, and Pancreas, Vol. 2. Philadelphia: Saunders 2006: 963–4. 94. Baskaran V, Patnaik PK. Feasibility and safety of laparoscopic management of hydatid disease of the liver. JSLS 2004; 8: 359–63. 95. DervenisC, Delis S, Avgerinos C, Madariaga J Milicevic M. Changing concepts in the management of liver hydatid disease. J Gastrointest Surg 2005; 9: 869–77. 96. Busic Z, Lemac D, Stipancic I, et al. Surgical treatment of liver echinococcosis: the role of laparoscopy. Acta Chir Belg 2006; 106: 688–91. 97. Mueller PR, Dawson SL, Ferrucci JT, Nardi GL. Hepatic echinococcal cyst: successful percutaneous drainage. Radiology 1985; 155: 627–8. 98. WHO – Informal Working Group on Echinococcosis. Guidelines for treatment of cystic and alveolar echinococcosis in humans. Bull WHO 1996; 74: 231–42. 99. Giorgio A, de Stefano G, Esposito V, et al. Long-term results of percutaneous treatment of hydatid liver cysts: a single center 17 years experience. Infection 2008; 36: 256–61. 100. Saremi F, McNamara TO. Hydatid cysts of the liver: long term results of percutaneous treatment using a cutting instrument. AJR Am J Roent 1995; 165: 1163–7. 101. Brunetti E, Filice C. Radiofrequency treatment ablation of echinococcal liver cyst. Lancet 2001; 358: 1464. 102. Belghiti A, Scaaps JP, Capron M, et al. Treatment of hepatic hydatid disease with mebendazole: preliminary results in four cases. Br Med J 1977; 2: 1047–51. 103. Lacey E. Mode of action of benzoimidazole. Parasitol Today 1990; 6: 112–5. 104. Morris DL, Dykes PW, Dickson B, et al. Albendazole in hydatid disease. Br Med J 1983; 286: 103–4. 105. Schipper HG, Koopmans RP, Nagy J, et al. Effect of dose increase or cimetidine co-administration on albendazole bioavailability. Am J Trop Med Hyg 2002; 63: 270–3. 106. Teggi A, Lastilla MG, De Rosa F. Therapy of human hydatid disease with benzoimidazole carbamates. Arch Hidatid 1991; 30: 773–95. 107. Teggi A, Lastilla MG, De Rosa F. Therapy of human hydatid disease with mebendazole and albendazole. Antimicrob Agents Chemother 1993; 37: 1679–84. 108. Todorov T, Vutova K, Mechkov G, et al. Evaluation of response to chemotherapy of human cystic echinococcosis. Br J Radiol 1990; 63: 523–31. 109. Morris DL. Albendazole treatment of hydatid disease, follow up at five years. Trop Doc 1989; 19: 179–80. 110. Filice C, Trosselli M, Brunetti E, et al. P.A.I.R. (Puncture, Aspiration, Introduction, Reaspiration) with alcohol under US guidance of hydatid liver cysts. Arch Hydatid 1991; 30: 811–7.

323

35

Surgical management of primary sclerosing cholangitis Jason A. Breaux and Steven A. Ahrendt
IBD and because PSC patients may harbor a higher risk of colorectal cancer (7).

overview
Primary sclerosing cholangitis (PSC) is a chronic, progressive disease characterized by inflammation and fibrotic strictures of the biliary tree. Strictures are usually multi-focal, with 75% of patients demonstrating both intra- and extrahepatic duct involvement. Only 10% of patients have isolated extrahepatic duct involvement (1–3). The incidence of dominant strictures accounting for the majority of symptoms is approximately 45% and most (up to 80%) of these occur at or near the hilum (4). PSC is not only associated with inflammatory bowel disease (IBD) in 70% of cases, most commonly ulcerative colitis, but also occurs in the setting of autoimmune diseases such as ankylosing spondylitis, celiac sprue, autoimmune pancreatitis, thyroiditis, and others. The overall risk of developing PSC in patients with ulcerative colitis approaches 10%, for patients with Crohn’s disease the risk is lower, approximately 2%. As with inflammatory bowel disease there is an approximate 2:1 male to female predominance and most patients present in young adulthood or middle age (5,6).

natural history
The natural history of primary sclerosing cholangitis is variable but generally involves progression from cholestasis to biliary cirrhosis and ultimately hepatic failure leading to liver transplant or death. Many more patients are being diagnosed in the asymptomatic stage as screening for IBD patients with serum liver studies has become routine. PSC generally follows an insidious course and some patients remain asymptomatic for many years. However, others progress rapidly or present initially with high-grade obstruction, cirrhosis, or cholangiocarcinoma. The median time from diagnosis to death or need for liver transplantation is 12 to 18 years. Liver failure and cholangiocarcinoma are the leading causes of death, in order of frequency (6). The Mayo Clinic developed a mathematical model to stratify patients’ risk, which has recently been updated. It utilizes data on patient age, total bilirubin level, aspartate aminotransferase level, presence of variceal bleeding, and serum albumin level to predict survival and assign a Mayo Risk Score (8). This can be used to prioritize treatment plans for individual patients. Unfortunately, surgical treatment for inflammatory bowel disease such as proctocolectomy for ulcerative colitis does not alter the natural course of PSC and patients may even present years after successful treatment for IBD.

diagnosis
Patients with PSC usually present with signs of cholestasis including right upper quadrant abdominal pain, jaundice, pruritis, and/or abnormal serum liver studies. Some IBD patients are diagnosed during routine screening for liver disease revealing elevated liver function tests particularly an elevated serum alkaline phosphatase (2,6). In the absence of alternative etiologies of cholestatic liver disease, the diagnosis is usually confirmed with high-quality imaging of the biliary tree utilizing endoscopic retrograde cholangiography (ERC), or more recently, magnetic resonance cholangiography (MRC). Both methods effectively demonstrate the characteristic “beads-on-a string” strictures (Fig. 35.1) of PSC making diagnosis possible in over 95% of cases using image criteria alone (3). ERC offers the ability to perform concurrent intervention, with the disadvantage of the low (3–8%) but welldefined incidence of complications related to this invasive procedure. MRC is diagnostic only, but has the advantage of minimal risk. MRC is also superior for visualization of ductal anatomy proximal to dominant strictures, and the two modalities may be used in complimentary fashion. Percutaneous transhepatic cholangiography (PTC) and intervention can also be utilized where the expertise exists, but PTC is notoriously difficult in patients with the altered ductal anatomy inherent to PSC. Regardless of imaging modality, detailed knowledge of a patient’s ductal anatomy is essential prior to any surgical intervention. Liver biopsy should be obtained at the time of diagnosis to assess the hepatic parenchyma for the presence and degree of fibrosis, which can alter treatment plans. Colonoscopy should also be performed to identify patients with subclinical

cholangiocarcinoma
Cholangiocarcinoma (CCA) develops in 10% to 30% of patients with PSC, and the incidence of CCA increases with the length of follow-up (5–10 years). The incidence of 10% at 5 years correlates with an increased risk of 160-fold over the general population. It is an ominous finding as the majority of patients have unresectable disease at the time of diagnosis and an overall median survival of only 5 to 11 months. A high index of suspicion is warranted, as just over half of patients with CCA related to PSC are diagnosed concurrently or within 1 year of their initial presentation (9,10). The diagnosis of cholangiocarcinoma in PSC can be challenging. Imaging techniques such as CT, MR, and positron emission tomography have proven largely unreliable in distinguishing benign from malignant biliary strictures. The majority (approximately 80%) of CCAs in PSC occur at the liver hilum, corresponding with the most frequent location for dominant strictures in this disease. Brushings and/or biopsies for cytology taken at the time of ERC have only up to 43% sensitivity in detecting malignancy (11). Recently, advanced cytologic techniques that identify chromosomal abnormalities including digital image analysis (DIA) and fluorescence in situ hybridization (FISH) have shown promise, but further study is warranted before widespread application is possible (12).

324

SURGICAL MANAGEMENT OF PRIMARY SCLEROSING CHOLANGITIS

(A)

(B) Figure 35.1 Cholangiograms (A and B) demonstrating the characteristic “beads on a string” strictures of PSC.

Tumor markers are moderately helpful in distinguishing benign from malignant strictures in PSC. Serum CEA and CA19-9 levels may both be elevated in CCA. Of the two tests, CA19-9 is the most useful and is relatively sensitive and specific in the diagnosis of CCA (78% and 98%, respectively) when elevated greater than 129 U/ml in the setting of PSC, based on a recent Mayo Clinic study. However, the majority of patients diagnosed by an elevated serum CA 19-9 level have unresectable disease and the utility of CA19-9 as a screening tool to identify early-stage disease in PSC patients is lacking. The best approach to diagnosis in patients with suspicious history or imaging findings seems to include the combination of a high index of suspicion, high-quality imaging, endoscopic brushings/biopsy, and CA19-9 levels. However, the diagnosis often remains in question and surgical excision with histopathologic examination is the only option in many cases to definitively rule out malignancy. Margin-negative surgical resection, when possible, also offers the only chance for longterm survival as traditional chemotherapeutic agents and radiation have poor efficacy in cholangiocarcinoma (10–13). Recent trials indicating improved survival with newer agents such as gemcitabine given neoadjuvantly followed by resection or transplantation have shown promise, but these approaches await validation and can only be recommended in the protocol setting (14,15).

of high-dose ursodeoxycholic acid in PSC in an attempt to resolve this issue (18).

endoscopic management
Endoscopic treatment involving balloon dilation to relieve symptomatic obstruction has become first-line treatment for benign dominant biliary strictures in primary sclerosing cholangitis. Stents were routinely employed in the past but have fallen out of favor due to studies identifying an increased risk of bacterial cholangitis with their use (19,20). Dominant strictures are defined as common bile duct or common hepatic duct stricture with a cross-sectional diameter of <1.5 mm and/ or a hepatic duct stricture with a diameter <1.0 mm within 2 cm of the hepatic duct bifurcation (21). Dilation of dominant strictures relieves symptoms, improves serum liver studies, and can produce durable improvement in imaging findings. Most patients require multiple interventions for adequate biliary drainage (20). The effect of endoscopic treatment on disease progression and survival is controversial. Two studies by Stiehl et al. and Baluyut et al. demonstrated improved overall survival and transplant-free survival with repeated endoscopic dilation, compared to that which would be predicted using the Mayo mathematical model. The incidence of cholangiocarcinoma developing during the follow-up period in the two series was 3% and 8%, respectively (Table 35.1) (21,22).

medical treatment
The inflammation and strictures of primary sclerosing cholangitis are thought to be immune-mediated and various immunosuppressive medications have been used in an attempt to slow the progression of PSC. However, numerous prospective randomized trials have failed to identify an agent that slows progression or improves outcome in patients with PSC (6). The most extensively studied drug, ursodeoxycholic acid, has been shown anecdotally to improve bile acid transport and have immuno-modulatory effects. However, despite initial encouraging studies showing improvement in serum liver studies and histologic findings on liver biopsy in PSC patients, no definitive improvement in survival or outcome has been observed in two large randomized trials (16,17). The National Institute of Health has sponsored a multicenter trial

surgical management
Surgical resection for primary sclerosing cholangitis was the only therapeutic option for patients with dominant strictures before the development of advanced endoscopic techniques and liver transplantation (23–25). The fact that most of the dominant strictures in PSC occur at or near the hepatic bifurcation makes resection and biliary-enteric drainage feasible and effective therapy (26). However, morbidity and mortality are high in patients with advanced disease and cirrhosis. Liver transplant is the treatment of choice for these patients and PSC has become a leading indication for transplantation. However, although advancements in endoscopy and transplantation have made resection of dominant strictures in PSC less common, there is still a role for this approach in select patients.

325

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 35.1 Overall Survival Results and Incidence of Cholangiocarcinoma in Recent Studies on Endoscopic and Surgical Treatment of Dominant Strictures in Noncirrhotic Patients with Primary Sclerosing Cholangitis
1 year Endoscopic therapy Baluyut et al. n = 63 Stiehl et al. n = 106 Ahrendt et al. n = 35 Surgical therapy Ahrendt et al. n = 40 Pawlik et al. n = 67 97% –87% 3 year 87% 86% 74% 92% 89% 5 year 83% 77% 59% 85% 83% 10 year – – CCA 8% 3% 8% 0 0


– 60%

95% 95%

Abbreviation: CCA, cholangiocarcinoma incidence.

Table 35.2 Transplant-free Survival in Recent Studies on Endoscopic and Surgical Treatment of Dominant Strictures in Noncirrhotic Patients with Primary Sclerosing Cholangitis
1 year Endoscopic therapy Stiehl et al. Ahrendt et al. Surgical therapy Ahrendt et al. – 85% 95% 3 year 93% 59% 92% 5 year 89% 46% 82%

Table 35.3 Survival of Patients with Primary Sclerosing Cholangitis Without Cholangiocarcinoma Treated by Surgical Resection (Extrahepatic Bile Duct Resection) Versus Transplantation
3 year Surgical resection Overall Noncirrhotics Cirrhotics Transplantation Source: Adapted from Ref. (31). 85% 95% 60% 87% 5 year 76% 83% 36% 67% 10 year 52% 60% 12% 57%

The indications for surgical resection in noncirrhotic PSC patients include symptomatic dominant strictures not amenable or recalcitrant to endoscopic dilation, failure to rule out malignancy in suspicious lesions, the finding of dysplasia or atypia on endoscopic brushings/biopsies and when resectable cholangiocarcinoma is identified (27–29). Biliary strictures that persist or recur despite dilation raise suspicion for malignancy and resection should be considered. High-quality preoperative imaging to define biliary and hepatic vascular anatomy is essential for operative planning. Broad-spectrum antibiotics covering biliary-enteric organisms should be administered preoperatively and continued postoperatively for 24 to 48 hours or until postoperative fever and cholangitis resolve. Surgical treatment involves resection of the extrahepatic biliary tree including the biliary bifurcation at the hilum, cholecystectomy, division and oversewing of the distal common

bile duct at the pancreatic head, and reconstruction with Roux-en-Y bilateral hepaticojejunostomies. Bile cultures should be obtained intraoperatively to guide therapy if postoperative cholangitis persists. Bilateral percutaneous transhepatic biliary stents are usually placed preoperatively to aid in portal dissection, and these are exchanged intraoperatively for silastic stents which are placed across the anastamosis and remain in long term to provide drainage and access to the biliary tree (23–28,30–33). There is some debate on the duration of post-operative drainage, but most experts advocate removal of the stents at approximately 1 year if the biliary-enteric anastamosis is widely patent. Major hepatic resection may also be included, using standard techniques, to achieve margins in the case of cholangiocarcinoma or extensive hilar fibrosis (11,30). Postoperative mortality following extrahepatic biliary resection for PSC is low (2–4%) as reported by centers with a high volume of hepatobiliary surgical experience. Morbidity is approximately 35%, with the majority of postoperative complications being mild and related to cholangitis. The effect of surgical resection on long-term outcome remains controversial. Early series suggested an overall and transplant-free survival benefit in favor of resection, when compared with similar series on endoscopic management (Tables 35.1 and 35.2). The most extensive investigation into this subject to date has come from data collected and published from The Johns Hopkins Hospital. This was recently expanded and updated by Pawlik et al. to include data from multiple centers with an extended follow-up of 10 years (31). They observed overall survival in noncirrhotic PSC patients undergoing resection of 95%, 83%, and 60% at 1 year, 5year and 10 year follow-up, respectively. They also reported favorable results resection, especially in noncirrhotic patients, when compared with liver transplantation (Table 35.3). Only 4 of 66 patients (6%) in the resection group went on to require transplant in this series. They also reported a relatively low rate of readmission following resection for PSC, with over half of patients requiring no readmissions for PSC-related problems at 3-year follow-up. The most common indication for readmission was for stent change and/ or treatment of cholangitis. Most significantly, no patients developed cholangiocarcinoma during the median follow-up time of 10.5 years (Table 35.1). This is likely due to the removal of the most common source of malignancy with this approach,

326

SURGICAL MANAGEMENT OF PRIMARY SCLEROSING CHOLANGITIS
a concept supported by transplant data reporting that over 70% of incidental CCAs found in explanted livers of patients with PSC occur at the hepatic bifurcation (34). Finally, as expected, cirrhotic patients undergoing resection fair much worse when compared to liver transplant (Table 35.3), and this should be the preferred treatment in cases of cirrhosis. Orthotopic liver transplant is the surgical treatment of choice for PSC patients with cirrhosis and without evidence of cholangiocarcinoma preoperatively. Standard techniques are employed and previous biliary surgery does not seem to influence overall survival, based on recent large series. Overall survival and graft survival in PSC patients are similar to results for other indications for transplant (35–37). The risk for recurrent PSC in the transplanted liver is approximately 15% and may lead to retransplant (38).
7. Broome U, Berg quist A. Primary slerosing cholangitis, inflammatory bowel disease and colon cancer. Semin Liver Dis 2006; 26: 31–41. 8. Kim WR, Therneau TM, Wiesner RH, et al. A revised natural history model for primary sclerosing cholangitis. Mayo Clin Proc 2000; 75: 688–94. 9. Rosen CB, Nagorney DM, Wiesner RH, et al. Cholangiocarcinoma complicating primary sclerosing cholangitis. Ann Surg 1991; 213: 21–5. 10. Kaya M, de Groen PC, Angulo P, et al. Treatment of cholangiocarcinoma complicating primary sclerosing cholangitis: the Mayo Clinic experience. Am J Gastroenterol 2001; 96: 1164–9. 11. Ahrendt SA, Pitt HA, Nakeeb A, et al. Diagnosis and management of cholangiocarcinoma in primary sclerosing cholangitis. J Gastrointest Surg 1999; 3: 357–67. 12. Charatcharoenwitthaya P, Lindor KD. Primary sclerosing cholangitis: diagnosis and management. Current Gastroenterol Rep 2006; 8: 75–82. 13. Levy C, Lymp J, Anulo P, et al. The value of serum CA 19-9 in predicting cholangiocarcinomas in patients with primary sclerosing cholangitis. Dig Dis Sci 2005; 50: 1734–40. 14. Rea DJ, Heimbach JK, Rosen CB, et al. Liver transplantation with neoadjuvant chemoradiation is more effective than resection for hilar cholangiocarcinoma. Ann Surg 2005; 242: 451–61. 15. Heimbach JK, Gores GJ, Nagorney DM, et al. Liver transplantation for perihilar cholangiocarcinoma after aggressive neoadjuvant therapy: a new paradigm for liver and biliary malignancies? Surgery 2006; 140: 331–4. 16. Mitchell SA, Bansi DS, Hunt N, et al. A preliminary trial of high-dose ursodeoxycholic acid in primary sclerosing cholangitis. Gastroenterology 2001; 121: 900–7. 17. Olsson R, Boberg KM, de Muckadell OS, et al. High-dose ursodeoxycholic acid in primary sclerosing cholangitis: a 5-year multicenter, randomized, controlled study. Gastroenterology 2005; 129: 1464–72. 18. Hoofnagle JH. Primary sclerosing cholangitis. Hepatology 2005; 41: 955. 19. Kaya M, Petersen BT, Angulo P, et al. Balloon dilation compared to stenting of dominant strictures in primary sclerosing cholangitis. Am J Gastroenterol 2001; 96: 1059–66. 20. Linder S, Soderlund C. Endoscopic therapy in primary sclerosing cholangitis: outcome of treatment and risk of cancer. Hepatogastroenterology 2001; 48: 387–92. 21. Stiehl A, Rudolph G, Sauer P, et al. Efficacy of ursodeoxycholic acid treatment and endoscopic dilation of major duct stenoses in primary sclerosing cholangitis. An 8-year prospective study. J Hepatol 1997; 26: 560–6. 22. Baluyut AR, Sherman S, Lehman GA, et al. Impact of endoscopic therapy on the survival of patients with primary sclerosing cholangitis. Gastrointest Endosc 2001; 53: 308–12. 23. Cameron JL, Pitt HA, Zinner MJ, et al. Resection of hepatic duct bifurcation and transhepatic stenting for sclerosing cholangitis. Ann Surg 1988; 207: 614–22. 24. Pitt HA, Thompson HH, Tompkins RK, et al. Primary sclerosing cholangitis: results of an aggressive surgical approach. Ann Surg 1982; 196: 259–68. 25. Cameron JL, Gayler BW, Herlong HF, et al. Sclerosing cholangitis: biliary reconstruction with Silastic transhepatic stents. Surgery 1983; 94: 324–30. 26. Cameron JL, Gayler BW. Sclerosing cholangitis: anatomical distribution of obstructive lesions. Ann Surg 1984; 200: 54–60. 27. Martin FM, Rossi RL, Nugent FW, et al. Surgical aspects of sclerosing cholangitis. Ann Surg 1990; 212: 551–8. 28. Ahrendt SA, Pitt HA, Kalloo AN, et al. Primary sclerosing cholangitis: resect, dilate, or transplant? Ann Surg 1998; 227: 412–23. 29. Ahrendt SA, Domajnko B. Indications for non-transplant surgery in primary sclerosing cholangitis. HPB 2005; 7: 292–7. 30. Ahrendt SA. Surgical approaches to strictures in primary sclerosing cholangitis. J Gastrointest Surg 2008; 12: 423–5. 31. Pawlik TM, Olbrecht VA, Pitt HA, et al. Primary sclerosing cholangitis: role of extrahepatic biliary resection. J Am Coll Surg 2008; 206: 822–30.

summary
Primary sclerosing cholangitis is a chronic, stricturing disease of the bile ducts leading to progressive cholestatic liver disease and a dramatic increase in risk for cholangiocarcinoma. Dominant strictures are common and usually involve the hepatic bifurcation. Medical therapy is ineffective in preventing progression of disease or improving outcomes. Optimal management of symptomatic noncirrhotic patients with dominant strictures has been somewhat controversial. Endoscopic therapy palliates symptoms and is relatively low risk and outcomes seem to be improved over mathematical predictions. However, the risk of cholangiocarcinoma remains in endoscopically treated patients and close follow-up is necessary. Surgical resection of the extrahepatic biliary tree also improves survival over predicted, and the primary site for the development of cholangiocarcinoma is removed. Therefore, a treatment strategy for symptomatic, noncirrhotic patients with PSC should involve a multi-disciplinary approach. High-quality imaging and a high index of suspicion for cholangiocarcinoma with appropriate screening measures are essential after diagnosis. Initially, endoscopic dilation of seemingly benign dominant strictures should be undertaken. Recurrent or suspicious lesions should be strongly considered for resection due to the risk of CCA. The surgical approach should involve resection of the extrahepatic biliary tree including the hepatic bifurcation with hepaticojejunostomy. PSC patients that progress to cirrhosis are best treated with orthotopic liver transplant.

references
1. Lee YM, Kaplan MM. Primary sclerosing cholangitis. N Engl J Med 1995; 332: 924–33. 2. Zyromski NJ, Pitt HA. Primary sclerosing cholangitis. In: Cameron JL, ed. Current Surgical Therapy, 9th edn, Philadelphia: Mosby/Elsevier, 2008: 438–42. 3. LaRusso NF, Shneider BL. Primary sclerosing cholangitis: summary of a workshop. Hepatology 2006; 44: 746–64. 4. Bjornsson E, Lindquist-Ottosson J, Asztely M, et al. Dominant stricture in patients with primary sclerosing cholangitis. Am J Gastroenterol 2004; 99: 502–8. 5. Talwalkar JA, Lindor KD. Primary sclerosing cholangitis. Inflamm Bowel Dis 2005; 11: 62–72. 6. Silveria MG, Lindor KD. Clinical Features and management of primary sclerosing cholangitis. World J Gastroenterol 2008; 14(21): 3338–49.

327

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
32. Yamamoto T, Hirohashi K, Kubo S, et al. Surgery for segmental primary sclerosing cholangitis. Hepatogastroenterology 2004; 51: 668–71. 33. Hirai I, Ishiyama S, Fuse A, et al. Primary sclerosing cholangitis successfully treated by resection of the confluence of the hepatic duct. J Hepatobiliary Pancreat Surg 2001; 8: 169–73. 34. Abu-Elmagd KM, Selby R, Iwatsuki S, et al. Cholangiocarcinoma and sclerosing cholangitis: clinical characteristics and effect on survival after liver transplantation. Transplant Proc 1993; 25: 1124–25. 35. Goss JA, Shackleton CR, Farmer DG, et al. Orthotopic liver transplantation for primary sclerosing cholangitis. A 12-year single center experience. Ann Surg 1997; 225: 472–81. 36. Farges O, Malassagne B, Sebagh M, et al. Primary sclerosing cholangitis: liver transplantation or biliary surgery. Surgery 1995; 117: 146–55. 37. Abu-Elmagd KM, Malinchoc M, Dickson ER, et al. Efficacy of hepatic transplantation in patients with primary sclerosing cholangitis. Surg Gynecol Obstet 1993; 177: 335–44. 38. Oldakowsk-Jedynak U, Nowak M, Mucha K, et al. Recurrence of primary sclerosing cholangitis in patients after liver transplantation. Transplant Proc 2006; 38: 240–3.

328

36

Management of advanced gallbladder cancer Hiromichi Ito and William R. Jarnagin
advanced gallbladder cancer (10,12). Doppler ultrasound is helpful not only to identify the presence of hepatic arterial or portal venous invasion, but also to improve specificity of US by differentiating malignant tumor from benign lesions by measuring blood flow into the suspected lesions (13). Endoscopic retrograde cholangiopancreatography (ERCP) or percutaneous transhepatic cholangiography (PTC) is useful to identify the spread of gallbladder cancer into biliary tree. A mid bile duct stricture is a classic sign of gallbladder cancer involving bile duct (Fig. 36.1). For patients with jaundice, cholangiography is useful for localizing the obstruction and also facilitating stent placement and establishing a diagnosis of cancer via brush cytology (14). If gallbladder cancer is suspected, abdominal cross-sectional imaging (CT or MRI) is mandatory to evaluate for nodal or metastatic disease as well as to further define the local extent of disease (Fig. 36.2). Lymph nodes involved by cancer are usually >1 cm diameter and ring-shaped heterogeneous enhancement with IV contrast. Ohtani and his colleagues reported the positive predictive value of conventional CT scan for detecting involvement in various lymph node stations as 75% to 100% despite lower sensitivity as 17% to 78% (15). The same authors reported the sensitivity of CT scan to detect of tumor invasion into liver, bile duct, or other adjacent organs such as pancreas and transverse colon as 50% to 65% and the positive predictive value as 77% to 100% (16). The use of spiral CT provides a better diagnostic accuracy in both nodal spread as well as indepth invasion than conventional CT scan (17,18). In a report by Yoshimitsu and his colleagues, the sensitivity of detecting tumor invasion into liver or other adjacent organ was 80% to 100%. MRI is less frequently used for staging of gallbladder cancer, but sometimes the use of MR cholangiography (MRCP) or angiography (MRA) provides more information than US or CT. Schwartz and colleagues demonstrated in retrospective analysis of 34 patients with gallbladder cancer that combination of conventional MRI and MRCP achieved a sensitivity of 100% for liver invasion and 92% for lymph node involvement (19). Positron emission tomography (PET) using fluorine-18labeled fluoro-deoxyglucose (FDG) is an emerging imaging modality that may prove to be of clinical value in the preoperative work-up of patients with gallbladder cancer. Multiple studies have shown that PET scans reliably detect primary and metastatic gallbladder cancer (20,21) as well as residual tumor after cholecystectomy (22). Corvera and his colleagues demonstrated that PET added information and altered management in 23% of selected patients with gallbladder who were preoperatively staged using US/CT/MRI (23). Since PET is not routinely available and the data for real contribution to preoperative staging are relatively limited, the role of PET in the multimodality work-up of patients with suspected

Gallbladder cancer is uncommon disease, although it is not rare. Indeed, gallbladder cancer is the fifth most common gastrointestinal cancer and the most common biliary tract cancer in the United States. The incidence is 1.2 per 100,000 persons per year (1). It has historically been considered as an incurable malignancy with a dismal prognosis due to its propensity for early invasion to liver and dissemination to lymph nodes and peritoneal surfaces. Therefore, clinical attitudes toward gallbladder cancer were pervaded with pessimism and nihilism. Population data in United States from 1988 through 2003 suggested >95% of surgically resectable gallbladder cancer has been treated with only simple cholecystectomy (2). Although recent advances in surgical technique and perioperative management have allowed an increased role for radical surgery in appropriately selected cases, the outcomes of majority of patients with advanced gallbladder cancer remains poor. Patients with gallbladder cancer usually present in one of three ways: (1) advanced unresectable cancer; (2) detection of suspicious lesion preoperatively and resectable after staging work-up; (3) incidental finding of cancer during or after cholecystectomy for benign disease. In this chapter, we describe a contemporary approach to advanced gallbladder cancer in the former two scenarios. We define “advanced” cancer as tumor penetrating through gallbladder wall (T3 or greater), metastasizing to regional lymph node (N1) or distant organ (M1). In the AJCC staging system, this is staged as II or higher on 6th edition (3) and as III, IVa, or IVb on 5th edition system (4). Refer previous chapter for more detailed discussion for staging systems of gallbladder cancer.

clinical presentation and work-up
The symptoms associated with gallbladder cancer are in general vague and non-specific; most patients with gallbladder cancer present when the disease is at an advanced stage, and majority of patients are diagnosed when the disease is beyond the borders of resection (5–9). The most common symptoms at presentation are abdominal pain or biliary colic (5,8,9). Patients with advanced disease may also present with jaundice from tumor invasion of the biliary tree or with systemic signs such as malaise and weight loss. Jaundice is well recognized as predictor of worse outcomes. In the series from Memorial Sloan-Kettering from 1995 through 2005, one-third of patients presented with jaundice and only 7% had resectable disease (6). The diagnosis is often suspected on an ultrasound done to evaluate right upper quadrant abdominal pain. Echogenic or discontinuous gallbladder mucosa, submucosal echolucency, or a mass should lead one to suspect gallbladder cancer. The presence of gallstones trapped within the tumor during its growth is a useful sign of possible gallbladder cancer (10,11). Although the detection of early lesions is challenging, ultrasound has a sensitivity of 85% and accuracy of 80% to diagnose

329

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
gallbladder cancer is still being defined and its use should be individualized. The goal of resection should always be complete extirpation with microscopic negative margins. Tumors beyond T2 are not cured by simple cholecystectomy and as with most of early gallbladder cancer, hepatic resection is always required. The extent of liver resection required depends upon whether involvement of major hepatic vessels, varies from segmental resection of segments IVb and V, at minimum to formal right hemihepatectomy or even right trisectionectomy. The right portal pedicle is at particular risk for advanced tumor located at the neck of gallbladder, and when such involvement is suspected, right hepatectomy is required. Bile duct resection and reconstruction is also required if tumor involved in bile duct. However, bile duct resection is associated with increased perioperative morbidity (26) and it should be performed only if it is necessary to clear tumor; bile duct resection does not necessarily increase the lymph node yield. Because of its propensity to spread to regional lymph nodes at early stage, resection of the liver involved and regional lymph node should be included for definitive treatment. In fact, frequency of metastasis to regional lymph nodes (hilar, celiac, peripancreatic, periduodenal) is fairly high for advanced tumors; pT3/4 60% to 81% versus pT1/2 0 to 62 (27–30). The most common lymph nodes involved are pericholedochal (42%) and retropancreatic (37%). Other nodal stations including celiac, SMA, para-aortic are involved in 20% to 25% of patients (31). However, optimal extent of lymphadenectomy is ill defined. It is the authors’ practice to include extirpation of lymph nodes within the hepatoduodenal ligament but not retropancreatic or celiac nodes as patients with involvement in these nodal basin are unlikely to benefit from resection. Nodal metastasis beyond the hepatoduodenal ligament on exploration is associated extremely poor outcomes (24) and we generally do not proceed with operation if gross metastasis is discovered on exploration. In the other hand, direct involvement of colon, pancreas, or duodenum is not an absolute contraindication of surgery. Several authors have reported that en bloc resection of adjacent organs (26,32–34), such as duodenum or pancreas, can be

surgical management
Although, many studies have suggested improved survival in patients with early gallbladder cancer with radical surgery including en bloc resection of gallbladder fossa and regional lymphadenectomy, its role for those with advanced gallbladder cancer remains controversial. First, patients with more advanced disease often require more extensive resections than early stage tumors, and operative morbidity and mortality rates are higher (24). Second, the long-term outcomes after resection, in general, tend to be poorer; long-term survival after radical surgery has been reported only for patients with limited local and lymph node spread. Therefore, the indication of radical surgery should be limited to well-selected patients based on thorough preoperative and intra-operative staging and the extent of surgery should be determined based on the area of tumor involvement. Surgical resection is warranted only for those who with locoregional disease without distant spread. Because of the limited sensitivity of current imaging modalities to detect metastatic lesions of gallbladder cancer, staging laparoscopy prior to proceeding to laparotomy is very useful to assess the abdomen for evidence of discontinuous liver disease or peritoneal metastasis and to avoid unnecessary laparotomy. Weber et al. reported that 48% of patients with potentially resectable gallbladder cancer on preoperative imaging work-up were spared laparotomy by discovering unresectable disease by laparoscopy (25). Laparoscopic cholecystectomy should be avoided when a preoperative cancer is suspected because of the risk of violation of the plane between tumor and liver and the risk of port site seeding.

Figure 36.1 ERCP of an advanced gallbladder cancer showing mid-bile duct obstruction.

Figure 36.2 Axial, contrast-enhanced computed tomogram of an advanced gallbladder cancer showing invasion into the liver parenchyma (arrowhead) and involvement of the stomach and first portion of the duodenum (arrow).

330

MANAGEMENT OF ADVANCED GALLBLADDER CANCER
associated with prolonged survival. In a recent study from our institution, resection of adjacent organ was performed in 21 patients for presumable malignant involvement; the resected adjacent organ was histologically involved only in half of the cases and only 16 of 21 cases were node negative, emphasizing that the finding of adherent organs does not necessarily imply advanced disease. Most importantly, adjacent organ resection was not associated with changes in long-term survival of patients (26). reported significant improvement in 5-year overall survival rate (26% vs. 14.4%) with postoperative mitomycin C and 5-FU following surgery compared with surgery alone as well as improvement in 5-year disease-free survival rate (20.3% vs. 11.6%) (36). However, definitive conclusion from this trial is limited by the small numbers of patients and the inclusion of patients undergoing incomplete (i.e., R1) resections. Indeed, subgroup analysis of patients who underwent a complete resection showed no survival benefit with adjuvant treatment. Most other data for the use of adjuvant or neo-adjuvant therapy in patients with gallbladder cancer is derived from phase II trials, in which treated patients were compared with historical controls (37,38). Kreral and his colleagues reported a 64% 5-year survival rate of patients who received 5-FU and external beam radiation following surgical resection compared to 33% of those their historical control (38). In contrast, Houry and his colleagues reported no survival benefit from adjuvant chemoradiation therapy on patients who underwent curative resection (39). Unfortunately, no study has provided conclusive evidence for benefit of adjuvant chemo or chemoradiation treatment for gallbladder cancer.

outcomes
Although advances in surgical technique and improvement in perioperative care allow us to perform radical resection for patients with gallbladder cancer safely, the outcomes for those with advanced cancer remain disappointing. The 5-year survival rates for patients having radical surgery ranged from 0% to 51%, most of them fall in 20% to 30% (Table 36.1). Nodal status and histological margin have been reported as predictive factors of survival after radical resection for this group of patients throughout the literature. For example, Behari and his colleague reported that positive node was associated with incomplete resection and none of the patients with N1 disease survived beyond 5 years (30). Endo and his colleague reported in their analysis of 55 patients who underwent complete resection, a 77% 5-year survival for patients without nodal involvement, 33% for those with single lymph node involvement, and 0% for those with two or more lymph nodes involvement (35). These findings suggest that radical resection should not be performed for patients with gross lymph nodes involvement or extensive tumor infiltration to adjacent structure on perioperative evaluation, both of which make complete resection with histological negative margin unlikely.

palliative care
Most patients with gallbladder cancer present with advanced, incurable disease and many are not candidates for surgical resection. The median survival of patients with advanced gallbladder cancer who are deemed inoperable ranges between 2 and 4 months (6,9,40) and palliation of symptoms should be the primary goal. Symptoms and conditions associated with incurable gallbladder cancer include jaundice, cholangitis, pain, and gastrointestinal obstruction. For obstructive jaundice or gastrointestinal obstruction, palliative intervention may be required. The common procedure for biliary obstruction due to gallbladder cancer is a segment III bypass (41). In their series of 41 consecutive segment III bypass for patients with advanced gallbladder cancer, Kapoor and his colleagues reported 87% success rate with 12% mortality and 51% morbidity rate (42). Because of poor survival, biliary stent is a preferred option for most of the patients. It can be placed via either percutaneous transhepatic route or endoscopic approach with minimal morbidity. Intestinal bypass should be performed only in patients who have symptomatic obstruction. Systemic chemotherapy and radiation therapy have, in general, little impact on unresectable gallbladder cancer. Multiple

adjuvant therapy
Because of its propensity to spread to regional lymph nodes at early stage and high rate of locoregional recurrence, adjuvant chemotherapy and/or chemoradiation therapy seems a rational therapeutic option for gallbladder cancer. Traditionally 5-FU based chemotherapeutic regimen has been used with or without combination of chemoradiation. However, there are few data to support its efficacy. The rarity of gallbladder cancer and further limitation of patients who can undergo complete resection make the randomized trial difficult to conduct. To date, there is only one randomized trial examining the efficacy of adjuvant chemotherapy for gallbladder cancer. This study

Table 36.1 Outcomes of Radical Surgery for Advanced Gallbladder Cancer
Authors Fong et al. (5) Kondo et al. (47) Behari et al. (30) Shih et al. (40) Kayahara et al. (48) D’Angelica et al. (26)
a

Year 2000 2002 2003 2007 2008 2009

N 58 38 24 39 631 72

Stage III/IVa III/IVab III/IVaa IIc III/IVaa IIc
a

5-yr survival rate 28/25% 33/17% 28/0% 34% 39–51/22–24% 22%

Note

Multi-institutional study

AJCC 5th edn. UICC 5th edn. c AJCC 6th edn.
b

331

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
regimens have been tested including combinations of 5-FU, leucovorin, mitomycin C, doxorubicin, and methotrexate. However, the effects have been mostly disappointing with poor response rates of 10% to 20% (43). Recent phase II trials using combination of gemcitabine and oxaliplatin showed an improved response rate ranging from 40% to 50% (44–46), and large scale randomized trial is warranted.
23. Corvera CU, Blumgart LH, Akhurst T, et al. 18F-fluorodeoxyglucose positron emission tomography influences management decisions in patients with biliary cancer. J Am Coll Surg 2008; 206(1): 57–65. 24. Kondo S, Nimura Y, Hayakawa N, et al. Regional and para-aortic lymphadenectomy in radical surgery for advanced gallbladder carcinoma. Br J Surg 2000; 87(4): 418–22. 25. Weber SM, DeMatteo RP, Fong Y, et al. Staging laparoscopy in patients with extrahepatic biliary carcinoma. Analysis of 100 patients. Ann Surg 2002; 235(3): 392–9. 26. D’Angelica M, Dalal KM, Dematteo RP, et al. Analysis of the Extent of Resection for Adenocarcinoma of the Gallbladder. Ann Surg Oncol 2008. 27. Matsumoto Y, Fujii H, Aoyama H, et al. Surgical treatment of primary carcinoma of the gallbladder based on the histologic analysis of 48 surgical specimens. Am J Surg 1992; 163(2): 239–45. 28. Pawlik TM, Gleisner AL, Vigano L, et al. Incidence of finding residual disease for incidental gallbladder carcinoma: implications for re-resection. J Gastrointest Surg 2007; 11(11): 1478–86; discussion 1486–7. 29. You DD, Lee HG, Paik KY, et al. What is an adequate extent of resection for T1 gallbladder cancers? Ann Surg 2008; 247(5): 835–8. 30. Behari A, Sikora SS, Wagholikar GD, et al. Longterm survival after extended resections in patients with gallbladder cancer. J Am Coll Surg 2003; 196(1): 82–8. 31. Shimada H, Endo I, Togo S, et al. The role of lymph node dissection in the treatment of gallbladder carcinoma. Cancer 1997; 79(5): 892–9. 32. Doty JR, Cameron JL, Yeo CJ, et al. Cholecystectomy, liver resection, and pylorus-preserving pancreaticoduodenectomy for gallbladder cancer: report of five cases. J Gastrointest Surg 2002; 6(5): 776–80. 33. Miyazaki M, Itoh H, Ambiru S, et al. Radical surgery for advanced gallbladder carcinoma. Br J Surg 1996; 83(4): 478–81. 34. Shirai Y, Ohtani T, Tsukada K, et al. Combined pancreaticoduodenectomy and hepatectomy for patients with locally advanced gallbladder carcinoma: long term results. Cancer 1997; 80(10): 1904–9. 35. Endo I, Shimada H, Tanabe M, et al. Prognostic significance of the number of positive lymph nodes in gallbladder cancer. J Gastrointest Surg 2006; 10(7): 999–1007. 36. Takada T, Amano H, Yasuda H, et al. Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A phase III multicenter prospective randomized controlled trial in patients with resected pancreaticobiliary carcinoma. Cancer 2002; 95(8): 1685–95. 37. Czito BG, Hurwitz HI, Clough RW, et al. Adjuvant external-beam radiotherapy with concurrent chemotherapy after resection of primary gallbladder carcinoma: a 23-year experience. Int J Radiat Oncol Biol Phys 2005; 62(4): 1030–4. 38. Kresl JJ, Schild SE, Henning GT, et al. Adjuvant external beam radiation therapy with concurrent chemotherapy in the management of gallbladder carcinoma. Int J Radiat Oncol Biol Phys 2002; 52(1): 167–75. 39. Houry S, Schlienger M, Huguier M, et al. Gallbladder carcinoma: role of radiation therapy. Br J Surg 1989; 76(5): 448–50. 40. Shih SP, Schulick RD, Cameron JL, et al. Gallbladder cancer: the role of laparoscopy and radical resection. Ann Surg 2007; 245(6): 893–901. 41. Jarnagin WR, Burke E, Powers C, et al. Intrahepatic biliary enteric bypass provides effective palliation in selected patients with malignant obstruction at the hepatic duct confluence. Am J Surg 1998; 175(6): 453–60. 42. Kapoor VK, Pradeep R, Haribhakti SP, et al. Intrahepatic segment III cholangiojejunostomy in advanced carcinoma of the gallbladder. Br J Surg 1996; 83(12): 1709–11. 43. Hejna M, Pruckmayer M, Raderer M. The role of chemotherapy and radiation in the management of biliary cancer: a review of the literature. Eur J Cancer 1998; 34(7): 977–86. 44. Harder J, Riecken B, Kummer O, et al. Outpatient chemotherapy with gemcitabine and oxaliplatin in patients with biliary tract cancer. Br J Cancer 2006; 95(7): 848–52. 45. Andre T, Tournigand C, Rosmorduc O, et al. Gemcitabine combined with oxaliplatin (GEMOX) in advanced biliary tract adenocarcinoma: a GERCOR study. Ann Oncol 2004; 15(9): 1339–43. 46. Verderame F, Russo A, Di Leo R, et al. Gemcitabine and oxaliplatin combination chemotherapy in advanced biliary tract cancers. Ann Oncol 2006; 17(Suppl 7): vii68–72. 47. Kondo S, Nimura Y, Hayakawa N, et al. Extensive surgery for carcinoma of the gallbladder. Br J Surg 2002; 89(2): 179–84. 48. Kayahara M, Nagakawa T, Nakagawara H, et al. Prognostic factors for gallbladder cancer in Japan. Ann Surg 2008; 248(5): 807–14.

references
1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008; 58(2): 71–96. 2. Coburn NG, Cleary SP, Tan JC, et al. Surgery for gallbladder cancer: a population-based analysis. J Am Coll Surg 2008; 207(3): 371–82. 3. Greene F, Page D, Fleming I, et al. AJCC Cancer Staging Manual, 6th edn. New York: Springer-Verlag, 2002. 4. Fleming I, Cooper J, Henson D, et al. AJCC Cancer Staging Manual, 5th edn. Philadelphia: Lippincott-Raven, 1998. 5. Fong Y, Jarnagin W, Blumgart LH. Gallbladder cancer: comparison of patients presenting initially for definitive operation with those presenting after prior noncurative intervention. Ann Surg 2000; 232(4): 557–69. 6. Hawkins WG, DeMatteo RP, Jarnagin WR, et al. Jaundice predicts advanced disease and early mortality in patients with gallbladder cancer. Ann Surg Oncol 2004; 11(3): 310–5. 7. Chan SY, Poon RT, Lo CM, et al. Management of carcinoma of the gallbladder: a single-institution experience in 16 years. J Surg Oncol 2008; 97(2): 156–64. 8. Dixon E, Vollmer CM, Jr., Sahajpal A, et al. An aggressive surgical approach leads to improved survival in patients with gallbladder cancer: a 12-year study at a North American Center. Ann Surg 2005; 241(3): 385–94. 9. Ito H, Matros E, Brooks DC, et al. Treatment outcomes associated with surgery for gallbladder cancer: a 20-year experience. J Gastrointest Surg 2004; 8(2): 183–90. 10. Gandolfi L, Torresan F, Solmi L, et al. The role of ultrasound in biliary and pancreatic diseases. Eur J Ultrasound 2003; 16(3): 141–59. 11. Levy AD, Murakata LA, Rohrmann CA Jr. Gallbladder carcinoma: radiologic-pathologic correlation. Radiographics 2001; 21(2): 295–314; questionnaire, 549–55. 12. Onoyama H, Yamamoto M, Takada M, et al. Diagnostic imaging of early gallbladder cancer: retrospective study of 53 cases. World J Surg 1999; 23(7): 708–12. 13. Komatsuda T, Ishida H, Konno K, et al. Gallbladder carcinoma: color Doppler sonography. Abdom Imaging 2000; 25(2): 194–7. 14. Gourgiotis S, Kocher HM, Solaini L, et al. Gallbladder cancer. Am J Surg 2008; 196(2): 252–64. 15. Ohtani T, Shirai Y, Tsukada K, et al. Carcinoma of the gallbladder: CT evaluation of lymphatic spread. Radiology 1993; 189(3): 875–80. 16. Ohtani T, Shirai Y, Tsukada K, et al. Spread of gallbladder carcinoma: CT evaluation with pathologic correlation. Abdom Imaging 1996; 21(3): 195–201. 17. Kumaran V, Gulati S, Paul B, et al. The role of dual-phase helical CT in assessing resectability of carcinoma of the gallbladder. Eur Radiol 2002; 12(8): 1993–9. 18. Yoshimitsu K, Honda H, Shinozaki K, et al. Helical CT of the local spread of carcinoma of the gallbladder: evaluation according to the TNM system in patients who underwent surgical resection. AJR Am J Roentgenol 2002; 179(2): 423–8. 19. Schwartz LH, Black J, Fong Y, et al. Gallbladder carcinoma: findings at MR imaging with MR cholangiopancreatography. J Comput Assist Tomogr 2002; 26(3): 405–10. 20. Petrowsky H, Wildbrett P, Husarik DB, et al. Impact of integrated positron emission tomography and computed tomography on staging and management of gallbladder cancer and cholangiocarcinoma. J Hepatol 2006; 45(1): 43–50. 21. Rodriguez-Fernandez A, Gomez-Rio M, Llamas-Elvira JM, et al. Positronemission tomography with fluorine-18-fluoro-2-deoxy-D-glucose for gallbladder cancer diagnosis. Am J Surg 2004; 188(2): 171–5. 22. Anderson CD, Rice MH, Pinson CW, et al. Fluorodeoxyglucose PET imaging in the evaluation of gallbladder carcinoma and cholangiocarcinoma. J Gastrointest Surg 2004; 8(1): 90–7.

332

37

Extrahepatic cholangiocarcinoma Yuji Nimura
rendered (VR) images clarify the anatomical variations of the hepatic artery and portal vein, and possible vascular invasion can be diagnosed by combined axial, MPR, and VR images (13,14) (Fig. 37.5). According to the above information, the resection site of the liver can be determined. Also a site of biliary drainage, the right or left hepatic duct and/or the right anterior or right posterior sectional duct, can be recommended. Further possible portal vein embolization prior to major hepatectomy can be advised. MDCT should be taken prior to biliary drainage (BD) to prevent artifacts of the drainage catheter which can influence the precise diagnosis of cancer extension along the involved bile ducts. Peroral cholangioscopy (POC) with or without intraductal ultrasonography (IDUS) should be performed before placing an endoscopic nasobiliary drainage (ENBD) catheter which produces artifacts: inflammation with or without granulomatous hyperplasia. Those changes hinder endoscopic diagnosis of mucosal spread of cholangiocarcinoma (15,16). Biliary Drainage (BD) Most patients with distal cholangiocarcinoma can safely undergo pancreatoduodenectomy (PD) without preoperative BD; however, there have been several controversies about BD prior to hepatectomy for jaundiced patients with proximal cholangiocarcinoma (17). The incidence of contaminated bile increases after biliary stenting, which is higher after endoscopic BD than percutaneous BD (18). Some retrospective studies did not show any difference in postoperative mortality after major hepatectomy for patients with or without preoperative BD, but reported higher morbidity in jaundiced patients (19,20). Another retrospective study showed significantly higher rate of infectious complications after major hepatectomy for proximal bile duct cancer in patients following preoperative BD than those without BD (21). Although no randomized controlled trial (RCT) have been performed to clarify the value of preoperative BD for jaundiced patients undergoing major hepatectomy, most major centers prefer to use preoperative BD followed by portal vein embolization (PE) prior to major hepatectomy for such patients with hilar cholangiocarcinoma (22–28). Preoperative BD has another diagnostic advantage to take selective tube cholangiography through both endoscopic BD and percutaneous transhepatic biliary drainage (PTBD), which clarifies the complicated anatomy of the intrahepatic segmental ducts and provides precise information about longitudinal cancer extension along the involved extrahepatic bile duct and/or the intrahepatic segmental ducts in the future remnant liver. The resection and reconstruction line of the intrahepatic bile ducts can then be defined prior to surgery (Fig. 37.6). In cases of superficially spreading

surgical anatomy of the bile duct
Although the middle and distal bile ducts follow the simple anatomy of the duct (see chapter 1), the proximal bile duct anatomy is frequently complicated, especially at the hepatic hilus for which many variations have been described (1–3). In cases of hilar cholangiocarcinoma, the hepatic confluence is separated into multiple units and possible proximal extension of the cancer must be determined in each isolated sectional or segmental bile duct. Therefore a fundamental knowledge of surgical anatomy of the intrahepatic sectional, segmental, and subsegmental bile ducts is essential for hepatobiliary gastroenterologists, radiologist, and surgeons to diagnose the preoperative stage of the disease and to design the planned surgical procedure for each individual patient with complex hilar cholangiocarcinoma. The applied surgical anatomy of the intrahepatic bile duct and the hepatic hilus has been clinically modified (Fig. 37.1) (4–7). Surgical experiences with aggressive hepatobiliary resection for biliary malignancies have led to more precise investigation of surgical anatomy of the hepatic hilus, revealed important variations of the intrahepatic bile ducts, and developed comprehensive studies on the biliary tree and vascular systems at the hepatic hilus which are mandatory when designing more complicated surgical procedures for locally advanced cholangiocarcinoma (Figs. 37.2 and 37.3) (8–11). As described in the above studies, the preoperative investigation of normal and/or abnormal anatomy, usual or unusual variations of the segmental bile ducts and the type of the hepatic confluence are necessary not only to design difficult hepatobiliary resections and reconstructions but also to prevent postoperative biliary complications (12).

preoperative managements
Staging of Cholangiocarcinoma Recent developments in diagnostic modalities have changed the preoperative staging system, with invasive techniques being replaced by non-invasive diagnostic procedures. Extracorporeal ultrasonography (US) is first used to detect biliary dilation proximally to a possible biliary lesion, and magnetic resonance cholangiopancreatography (MRCP) is performed to demonstrate gross anatomy of the biliary tree and the variation of the intrahepatic bile ducts. Surgical anatomy and the extent of the cancer along the involved intrahepatic segmental ducts also have to be clarified in patients with hilar cholangiocarcinoma (Fig. 37.4A). Multi-detector row CT (MDCT) is helpful not only to assess the depth of invasion and longitudinal extension of cholangiocarcinoma but also to find lymph node and distant organ metastases. Furthermore multiplanar reformation (MPR) images provide more useful information about complex structures at the hepatic hilus and display the entire length of the involved bile duct, and show ductal thickening and intraductal masses (Fig. 37.4B). Also volume

333

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Supine position
8a 8b 8c 4b 4c 4c 2 7b 7a 6c 5c 5a 4a 4a 5a 3b 3a 3a 3b 5c 6c 2 7a 7b 4b

Right lateral position
8a 8b 8c

6b 5b

5b 6a 6a 6b

Figure 37.1 Cholangiographic anatomy of the intrahepatic subsegmental bile duct. Numerals refer to Couinaud’s segments. 3a, Superior branch; 3b, inferior branch; 4a, inferior branch; 4b, superior branch; 4c, dorsal branch; 5a, ventral branch; 5b, dorsal branch; 5c, lateral branch; 6a, ventral branch; 6b, dorsal branch; 6c, lateral branch; 7a, ventral branch, 7b, dorsal branch; 8a, ventral branch; 8b, lateral branch; 8c, dorsal branch.

curative hepatobiliary resection (29–36). These invasive diagnostic procedures are carried out during the preoperative period of biliary drainage.
1r 1r 1ls

2 A U P 4 3

P 1li

1c

Portal Vein Embolization At the final stage of preoperative BD, liver function tests including indocyanine green (ICG) test are performed. Also the functional capacity of each section of the liver is carefully estimated by both the CT volumetric study and the ICG test (37,38). PE is performed safely before extended hepatectomy to prevent postoperative liver failure for patients with marginal functional capacity of the future remnant liver and to increase the resectability rates for patients with advanced hilar cholangiocarcinoma (39,40). At a minimum of 2 weeks later, liver resection volume and functional capacity of the future remaining liver are estimated again by CT volumetry and ICG test to decide the timing of the definitive surgery (Fig.37.7). Clinical studies on PE offered revolutionary progress in hepatobiliary surgery and have actually increased resectability and the safety of major liver resection for locally advanced hilar cholangiocarcinoma (41,42). Synbiotics Treatments with Bile Replacement Obstructive jaundice is associated with an increased incidence of bacterial translocation and infectious complications after hepatectomy for biliary cancer patients still remain a major problem, although surgical techniques and perioperative care have been improved. External BD alone cannot re-establish the defense system against bacterial translocation and absence of intestinal bile plays an important role in the development of infectious complications related to biliary obstruction. On the contrary, internal BD prevents the loss of bile from the gastrointestinal tract, preserves the enterohepatic biliary circulation, and normalizes the enhanced intestinal permeability in obstructive jaundice. Therefore bile replacement should be carried out during external BD to

Figure 37.2 Surgical anatomy of the hepatic hilus, including the biliary and portal branches of the caudate lobe. U, Umbilical portion of the left portal vein; P, right posterior branch; A, right anterior branch; 1ls, superior branch of the left caudate lobe; 1li, inferior branch of the left caudate lobe; 1r, branch of the right caudate lobe; 1c, branch of the caudate process; 2, left lateral posterior branch; 3, left lateral anterior branch; 4, left medial branch.

cholangiocarcinoma, per-oral cholangioscopy (POC) or percutaneous transhepatic cholangioscopy (PTCS), followed by mapping biopsy is useful to detect minor mucosal changes and define the proximal mucosal extension of the cancer into the intrahepatic segmental ducts, and so design an appropriate

334

EXTRAHEPATIC CHOLANGIOCARCINOMA
Round ligament Round ligament Round ligament

S4 P4 B4

S3 P3 B3 B2 P2 P4 B4

S4

S3 P3 B3 B2 P2 P4 B4

S4

S3 B3a P3 B3b B2 P2

LHD LPV (A)

LHD LPV (B)

LHD LPV (C)

Figure 37.3 Schema of the infraportal variation of the anterior branch of the left lateral section (B3). (A) Normal anatomy, (B) infraportal B3 joining B4, (C) supraportal B3d (superior branch) and infraportal B3b (inferior branch).

(A)

(B)

(C)

Figure 37.4 (A) MRCP shows biliary stricture at the hepatic hilus. A possible diagnosis is hilar cholangiocarcinoma separating the hepatic confluence. (B) Coronal images of multiplanar reformation (MPR) show a soft tissue tumor at the hepatic confluence separating the right and left hepatic duct (arrow). (C) The soft tissue tumor separates the confluence of the right anterior and posterior sectional ducts (arrow).

(A)

(B)

Figure 37.5 Volume rendered (VR) images of the hepatic artery (A) and portal vein (B). 3D images can be obtained. (A). Irregular encasement is demonstrated on the right hepatic artery (arrow). (B) The left portal vein is obstructed and the right portal vein is involved (arrow). P, right posterior sectional branch; 7d, paracaval branch of the segment 7.

335

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
repair the physical damage to the intestinal mucosa and to restore the intestinal barrier function in patients with obstructive jaundice (43). In addition to bile replacement during external BD, perioperative synbiotic treatment is an useful measure to prevent bacterial translocation triggered by intestinal microbial imbalance and host immunodeficiency. Several RCTs reported that consecutive preoperative and postoperative synbiotic treatment could reduce postoperative infectious complications after high-risk hepatobiliary resection for patients with biliary tract carcinoma (44,45). These RCTs revealed that administration of synbiotics could enhance immune responses, attenuate systemic postoperative inflammatory responses, and improve intestinal microbial environment by increasing beneficial bacteria and decreasing harmful bacteria during recovery from major hepatobiliary surgery. Preoperative oral intake of synbiotics followed by postoperative administration through an enteral feeding tube is safe, simple, and convenient treatment of choice which also reduces medical costs in shortening the period of postoperative antibiotics treatment and hospital stay.

operative procedures
Pancreatoduodenectomy for Mid-third and Distal Cholangiocarcinoma PD and pylorus-preserving pancreatoduodenectomy (PPPD) have been used as the standard operation not only for pancreatic cancer but also for distal cholangiocarcinoma. Although the details of this surgical procedure are presented in the chapter of pancreatic cancer, an important part of the procedure as related to cholangiocarcinoma is presented to avoid an overlap of description. PD versus PPPD PPPD is preferably used in biliary tract cancer surgery to preserve the important organs as much as possible. Also, as the risk of peripyloric lymph node metastasis is low in extrahepatic

A

L B4b

B2

B1 B3 P B4a1
(A) (B)

B4a2

Figure 37.6 (A) PTBD tube cholangiography in a supine position. A tip of the PTBD catheter (arrow) is introduced from the right anterior sectional duct into the left hepatic duct across the hepatic confluence occupied by the tumor to drain bile from the future remnant hepatic lobe. Another PTBD catheter (arrow head) is placed in the right posterior sectional duct. A: right anterior sectional duct, P: right posterior sectional duct, L: left hepatic duct. (B) PTBD tube cholangiography in a right anterior and cranioanterior oblique position. Selective cholangiography of the left hepatic duct clearly demonstrates each segmental duct and the expected resection line can be defined at the confluence of the left medial segmental duct proximally to the confluence of the caudate lobe branch. B1: caudate lobe branch, B2: lateral posterior branch, B3: lateral anterior branch, B4a: medial inferior branch, B4b: medial superior branch.

Before

After

Figure 37.7 Volumetric changes of the liver sections before and after right trisectional portal vein embolization. Hypertrophy of the left lateral section is observed after portal vein embolization.

336

EXTRAHEPATIC CHOLANGIOCARCINOMA
cholangiocarcinoma, PPPD is advisable as appropriate surgery for distal cholangiocarcinoma. Extent of Lymph Node Dissection Lymph node metastasis, perineural invasion, surgical resection margins, and pancreatic invasion are prognostic factors after curative resection for middle and distal cholangiocarcinoma (46–51). Therefore regional lymph node dissection is necessary, including the nodes in the hepatoduodenal ligament and along the common hepatic and superior mesenteric arteries. Proximal Extension of Resection In cases of distal cholangiocarcinoma with proximal extension close to the hepatic confluence, the proximal bile duct is carefully dissected while detaching the portal bifurcation and the left hepatic duct is transected on the left extremity of the hilar plate along the right wall of the umbilical portion of the left portal vein (UP). At the resected margin of the left hepatic duct, the openings of the resected segmental ducts (B2, B3, and B4) are identified according to their anatomical variation. On the right extremity of the hilar plate, the right hepatic artery is carefully skeletonized in Rouviere’s sulcus and the right posterior sectoral duct is carefully divided with a negative margin. Next, the right anterior sectoral duct is divided. The caudate lobe branches are sometimes identified and divided according to their anatomical variation (Fig. 37.8). Hepaticojejunostomy After resecting the proximal bile duct, each individual sectional or segmental bile duct should be anatomically identified and sutured side by side to complete the hepaticoplasty and to minimize the number of hepaticojejunostomies with a Roux-en-Y jejunal limb. A running suture of 5-0 PDS is preferably used for both posterior and anterior wall anastomosis. Hepatobiliary Resection for Hilar Cholangiocarcinoma Most of hilar cholangiocarcinoma involving the hepatic confluence are indicated for liver and extrahepatic bile duct resection with caudate lobe resection, because the caudate lobe branches join the right and left hepatic ducts and/or their confluence and mostly be involved by carcinoma at the hepatic confluence (4–6). In this section, important parts of the surgical procedures for cholangiocarcinoma are described. Left Hepatectomy, Caudate Lobectomy, and Extrahepatic Bile Duct Resection Left-sided hepatectomy is indicated for hilar cholangiocarcinoma predominantly involving the left hepatic duct (51). After regional node and connective tissue dissection, the distal bile duct is resected in the pancreas with a histologically free margin. After dividing the vascular structures for the left liver and the caudate lobe, the left lateral section of the liver is mobilized toward the right anteriorly and the caudate lobe is also mobilized to the right anteriorly, while ligating and dividing all the short hepatic veins. Then the caudate lobe is completely detached from the inferior vena cava (IVC). Next, the liver is transected along the demarcation line and this dissection progressed toward the hepatic hilus to identify the right hepatic duct and the right anterior sectional duct crossing behind the middle hepatic vein (MHV). The dorsal aspect of liver dissection is aligned along the right lateral edge of the IVC and the caudate process is detached from the segment 7 to remove the entire caudate lobe. Then the isolated right anterior sectional duct or segmental ducts are divided with free margins. Next, the right posterior sectional duct is exposed cranially to the right portal vein and divided with a negative margin; and the left liver, caudate lobe, and extrahepatic duct are removed en bloc.

4 3 A P A 1 1 2

LH M HA

A

P

RH

A

Figure 37.8 Hilar bile duct resection. The right and left sectional or segmental ducts and caudate lobe branches are identified and divided with free margins. Numerals refer to Couinaud’s segments. A, anterior sectional duct; P, posterior sectional duct; RHA, right hepatic artery; MHA, middle hepatic artery; LHA, left hepatic artery.

337

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Right Hepatectomy, Caudate Lobectomy, and Extrahepatic Bile Duct Resection Right hepatectomy is indicated for hilar cholangiocarcinoma which predominantly involves the right hepatic duct. Regional node and connective tissue dissection is performed, and the right hepatic artery, the right portal vein, and the caudate lobe branches are divided. During this procedure, the proximal and distal ends of the Arantius canal are ligated and divided below the UP and close to the confluence of the left hepatic vein (LHV) or the IVC. As in most cases of this operation, right PE has been performed pre-operatively to increase the safety of major hepatectomy, careful attention should be paid to observe the lumen of the resected margin of the right portal vein and not to overlook possible remaining portal thrombus at the portal bifurcation which can precipitate postoperative portal thrombosis in the future remnant left liver. Next, the right liver is mobilized ventrally to the left and short hepatic veins are ligated and divided step by step. Finally, the right hepatic vein (RHV) is clamped, divided, and closed at the confluence of the IVC. Then the caudate lobe and the right liver are completely detached from the IVC. Liver dissection is started along the demarcation and progressed horizontally on the visceral aspect of the segment 4 about 1 cm above the hilar plate to keep a surgical resection margin and reaches to the left hepatic duct on the right of the UP. Then the left hepatic duct is divided with a free margin perpendicularly parallel to the UP, and the right liver, caudate lobe, and extrahepatic duct are removed en bloc. Openings of the segmental branches are identified on the resected margin of the left hepatic duct. Left Trisectionectomy Left trisectionectomy is indicated for hilar cholangiocarcinoma which predominantly involves the left hepatic duct and the right anterior sectional duct. Left trisectional PE and right posterior sectional BD should be carried out prior to definitive surgery. After skeletonization and dissection of the hepatoduodenal ligament and division of the distal bile duct in the pancreas, the left hepatic artery and the right anterior branch are ligated and divided. Next, the portal bifurcation is dissected and the caudate lobe branches are ligated and divided. The left portal vein and the right anterior branch are divided and closed after careful observation of the lumen of the divided portal vein to exclude portal thrombi. Next, the left liver is mobilized right anteriorly and the caudate lobe is also mobilized while all short hepatic veins are ligated and divided to detach the caudate lobe from the IVC. This procedure is progressed cranially and the common trunk of the LHV and MHV is divided and closed. Liver transection is carried out along the demarcation on the right portal fissure while exposing the RHV on the raw surface of the liver and progressed toward Rouviere’s sulcus to expose and encircle the right posterior sectional duct which should carefully be detached from the right posterior branches of the both hepatic artery and portal vein running behind the bile duct. Dorsal liver dissection is started between segment 7 and the caudate process and advanced cranially along the right edge of the IVC and reaches to the caudal aspect of the distal end of the RHV to remove the entire caudate lobe en bloc. Finally the right posterior sectional duct is divided with a free margin, and the liver and the extrahepatic bile duct are removed en bloc. At the resected margin of the right posterior sectional duct, single or double openings, which is a sectional or segmental ducts, are identified (Fig. 37.9). Right Trisectionectomy Right trisectionectomy is indicated for hilar cholangiocarcinoma which predominantly involves the right hepatic duct and the left medial sectional duct and has been considered as the most high-risk liver resection which sometimes is associated with serious liver failure. Therefore careful preoperative management is necessary and functional reserve of the liver should be carefully estimated before surgery to prevent postoperative hepatic failure. BD of the left lateral sectional duct with right trisectional PE should be performed prior to definitive surgery. CT volumetric study with ICG test before and after right trisectional PE is helpful to estimate the functional capacity of the future remnant left lateral section of the liver (7,8,40–42) (Fig. 37.7). After skeletonization and dissection of the hepatoduodenal ligament and resection of the distal bile duct in the pancreas, the right and middle hepatic arteries are ligated and divided. Then a demarcation appears on the umbilical fissure if the right trisectional PE has already been carried out. Next, the portal vein is dissected distally up to the bifurcation and the caudate lobe branches are ligated and divided. Also the proximal portion of the Arantius canal is ligated and divided to free the left portal vein from the surrounding hilar plate. Next, the right portal vein is divided at the bifurcation while the lumen is carefully inspected to exclude portal thrombi at the bifurcation, and the vein is closed transversely. Next, the right liver is mobilized to the left anteriorly and all short hepatic veins are ligated and divided to detach the

Figure 37.9 The right posterior segmental and subsegmental ducts are identified at the resected margin of the right posterior sectional duct. P6, posterior inferior portal branch; P7, posterior superior portal branch; B6a, ventral biliary branch; B6bc, dorsal and lateral biliary branch; B7, posterior superior biliary branch; RHV, right hepatic vein..

338

EXTRAHEPATIC CHOLANGIOCARCINOMA
caudate lobe from the IVC. At the cranial part of this procedure, the distal end of the RHV is divided and closed at the confluence of the IVC. Next, liver transection is started from the right edge of the attachment of the round ligament and progressed dorsally along the right edge of the UP and the portal branches for the left medial section (S4) are ligated and divided. During this procedure, the lumen of the divided portal branches must be carefully observed not to overlook portal thrombi at the bifurcation if the right trisectional PE has already been carried out. When medial branches of the left portal vein are divided, the demarcation is observed more clearly along the right edge of the falciform ligament. Then liver transection is advanced cranially along the demarcation and reaches the distal end of the MHV at the confluence of the LHV or the IVC. And the MHV is divided and closed at the confluence. Finally the left lateral sectional duct is exposed and isolated cranially close to the UP and divided with a free resection margin, and the right trisection of the liver, caudate lobe and the extrahepatic bile duct are removed en bloc. Anatomic Right Trisectionectomy Anatomic right trisectionectomy is a more extensive hepatectomy than traditional right trisectionectomy and this more aggressive procedure is indicated for more advanced hilar cholangiocarcinoma which involves not only the right hepatic duct and the left medial sectional duct but also the confluence of the left medial and lateral sectional ducts (52). The actual surgical procedure is slightly different from that of traditional right trisectionectomy in dividing all the portal branches for the left medial section (S4), including the main superior and inferior branches and the dorsal branches which ramify from the dorsal aspect of the UP. By progressing this procedure, the UP is gradually turned counterclockwise to expose the left lateral sectional duct more proximally on the left of the UP (Fig. 37.10). Then a clear demarcation appears on the umbilical fissure behind or left laterally to the falciform ligament. Then liver transection along the umbilical fissure reaches left laterally to the UP or just behind the UP, and the left lateral sectional duct or segmental ducts can be divided more proximally with free margins (Fig. 37.11). Intrahepatic Cholangiojejunostomy After hepatobiliary resection and removal of the tumor, the resected margins of the intrahepatic ducts must be identified anatomically and sutured side by side to complete the hepaticoplasty and to minimize the number of anastomotic orifices for intrahepatic cholangiojejunostomy with a Roux-en-Y jejunal limb lifted via the retrocolic–retrogastric route (12,53). Previously, single layer cholangiojejunostomy was performed using interrupted sutures of 5-0 or 6-0 PDS and all anastomosed bile ducts drained externally with 6 Fr. polyvinyl chloride tubes. However, a continuous suture of 5-0 or 6-0 PDS has recently been used for each anterior and posterior wall anastomosis of all cholangiojejunostomies and external biliary drainage tube(s) has not been used routinely. Thus, difficult and time-consuming hepatobiliary resectional and reconstructional surgery became simplified. Combined Liver and Portal Vein Resection for Advanced Cholangiocarcinoma Combined liver and portal vein resection is indicated for locally advanced cholangiocarcinoma, and several types of liver resection and combined portal vein resection and reconstruction have been reported as an aggressive surgical approach which provides both negative surgical margins and contributes to prolonged survival for resected patients compared to non-resected patients (54–58). The portal vein is usually resected at the final step of hepatobiliary resection and is reconstructed after removal of the

Figure 37.10 All left medial branches of the portal vein are divided to turn the umbilical portion of the vein and to expose the left lateral sectional duct more proximally and left laterally to the vein. B2: lateral posterior branch, B3: lateral anterior branch, B4, P4: medial branch, P4c: dorsal branch, RPV: right portal vein.

339

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
the extent of cholangiocarcinoma as described in the previous section (Fig. 37.11).

discussion
Recent improvement in preoperative staging system and managements of difficult biliary cancer patients, as well as technical developments in hepatobiliary surgery have increased the rate of potentially curative resection and decreased postoperative morbidity and mortality (13,14,36–40,42–44). Although preoperative BD has recently not been used prior to PD, the value of preoperative BD before cholestatic liver resection has not been clarified by RCTs (28). As the HPD is the ultimate surgery for the visceral organs, unexpectedly high morbidity and mortality have been encountered (63); however, recent progress in surgical techniques and perioperative management for biliary cancer patients with difficult preoperative complications improved the outcome of this surgery and an increasing number of operations and 5-year survivors have been reported not only from aggressive surgeons in Japan but also from the United States (64–68). It is expected that an increasing number of patients with locally advanced cholangiocarcinoma will be considered for difficult surgeries employing HPD with or without vascular resection, which should be justified by improved outcome (69–72). Recent retrospective analysis of the impact of future liver remnant (FLR) volume and preoperative BD on postoperative hepatic insufficiency and mortality rates revealed that preoperative BD did not appear to improve perioperative outcome in patients with FLR ≥30% and PVE is likely to offer little benefit (73). According to the continued and utmost efforts of aggressive hepatobiliary surgeons, favorable results of PD and hepatectomy have been reported and prognostic factors after curative resection of extrahepatic cholangiocarcinoma have been revealed (22–27,42,46–51). Although preoperative BD followed by PE has been used prior to major hepatectomy for biliary cancer patients with obstructive jaundice in many centers all over the world, it may be difficult to clarify the value of this strategy by RCTs.

Figure 37.11 The left lateral sectional or segmental ducts ( ) are resected left dorsally to the umbilical portion of the left portal vein. In this case, right hepatopancreatoduodenectomy with portal vein resection and reconstruction (arrow) are performed. P, pancreas; RL, round ligament.

tumor. However, in case of right-sided hepatectomy, the portal bifurcation can be resected and reconstructed prior to liver dissection using an end-to-end anastomosis between the left portal vein and the main trunk to establish the non-touch resection of hilar cholangiocarcinoma (59). Several techniques have been reported for portal vein reconstruction. In addition to the end-to-end anastomosis which is commonly used, a graft interposition using an external iliac vein (54,60), a hepatic vein segment (61), a saphenous vein (54), or a left renal venous patch (62) are used for difficult portal vein reconstruction after segmental or wedge resection of the portal bifurcation. Hepatopancreatoduodenectomy for Advanced Cholangiocarcinoma Hepatopancreatoduodenectomy (HPD) has been used as the most extensive surgery for advanced carcinoma of the biliary tract (63). Recent development of diagnostic modalities has increased the opportunity to demonstrate the extent of cholangiocarcinoma precisely by MDCT, selective cholangiography, and cholangioscopy, and increasing numbers of patients with cholangiocarcinoma have been considered for HPD. Distal or middle bile duct cancer with superficial spread to the intrahepatic bile duct or intrahepatic cholangiocarcinoma with superficial spread to the distal bile duct are indications for HPD (33,64,65). As HPD is composed of PD or PPPD for distal bile duct cancer and hepatectomy for proximal cholangiocarcinoma, PD or PPPD is carried out at the first step of this operation, and the hepatoduodenal ligament is dissected upward to skeletonize the hepatic arteries and portal bifurcation. As the next step, hemihepatectomy or hepatic sectionectomy is performed according to the preoperative diagnosis of

references
1. Healey JE Jr, Schroy PC. Anatomy of the biliary ducts within the human liver, analysis of the prevailing pattern of branchings and the major variations of the biliary ducts. AMA Arch Surg 1953; 66: 599–616. 2. Couinaud C. Surgical anatomy of the liver revisited. Paris: Couinaud C, 1989: 61–74. 3. Mizumoto R, Suzuki H. Surgical anatomy of the hepatic hilum with special reference to the caudate lobe. World J Surg 1988; 12: 2–10. 4. Nimura Y, Hayakawa N, Kamiya J, Kondo S, Shionoya S. Hepatic segmentectomy with caudate lobe resection for bile duct carcinoma of the hepatic hilus. World J Surg 1990; 14: 535–44.(Category III) (Grade C). 5. Kamiya J, Nimura Y, Hayakawa N, et al. Preoperative cholangiography of the caudate lobe: Surgical anatomy and staging for biliary carcinoma. J Hepatobiliary Pancreat Surg 1994; 1: 385–9. 6. Nimura Y, Hayakawa N, Kamiya J, et al. Hilar cholangiocarcinoma – surgical anatomy and curative resection. J Hepatobiliary Pancreat Surg 1995; 2: 239–48. (Category III) (Grade C). 7. Nimura Y. Surgical anatomy of the biliary ducts. In: Rossi P, ed. Biliary Tract Radiology. Berlin, Heidelberg, Springer-Verlag, 1997: 21–30. 8. Nimura Y. Living donor liver transplantation: importance of anatomical variations of biliary tree and vascular systems. Transpl Proc 2003; 35: 955.

340

EXTRAHEPATIC CHOLANGIOCARCINOMA
9. Ozden I, Kamiya J, Nagino M, et al. Clinicoanatomical study on the infraportal bile ducts of segment 3. World J Surg 2002; 26: 1441–5. 10. Ohkubo M, Nagino M, Kamiya J, et al. Surgical anatomy of the bile ducts at the hepatic hilum as applied to living donor liver transplantation. Ann Surg 2004; 239: 82–6. 11. Sugiura T, Nagino M, Kamiya J, et al. Infraportal bile duct of the caudate lobe: a troublesome anatomic variation in right-sided hepatectomy for perihilar cholangiocarcinoma. Ann Surg 2007; 246: 794–8. 12. Nagino M, Nishio H, Ebata T, et al. Intrahepatic cholangiojejunostomy following hepatobiliary resection. Br J Surg 2007; 94: 70–7. (Category III) (Grade C). 13. Choi JY, Lee JM, Lee JY, et al. Assessment of hilar and extrahepatic bile duct cancer using multidetector CT: value of adding multiplanar reformations to standard axial images. Eur Radiol 2007; 17: 3130–8. (Category III) (Grade C). 14. Senda Y, Nishio H, Oda K, et al. Value of multidetector row CT in the assessment of longitudinal extension of cholangiocarcinoma-correlation between MDCT and microscopic findings. World J Surg 2009; 33: 1459– 67. (Category III) (Grade C). 15. Fukuda Y, Tsuyuguchi T, Sakai Y, Tsuchiya S, Saisyo H. Diagnostic utility of peroral cholangioscopy for various bile-duct lesions. Gastrointest Endosc 2005; 62: 374–82. (Category IV) (Grade D). 16. Itoi T, Sofuni A, Itokawa F, et al. Peroral cholangioscopic diagnosis of biliary-tract diseases by using narrow-band imaging (with videos). Gastrointest Endosc 2007; 66: 730–6. (Category IV) (Grade D). 17. Nishio H, Hidalgo E, Hamady ZZ, et al. Left hepatic trisectionectomy for hepatobiliary malignancy: results and an appraisal of its current role. Ann Surg 2005; 242: 267–75. (Category III) (Grade C). 18. Hochwald SN, Burke EC, Jarnagin WR, Fong Y, Blumgart LH. Association of preoperative biliary stenting with increased postoperative infectious complications in proximal cholangiocarcinoma. Arch Surg 1999; 134: 261–6. (Category III) (Grade C). 19. Figueras J, Llado L, Valls C, et al. Changing strategies in diagnosis and management of hilar cholangiocarcinoma. Liver Transpl 2000; 6: 786–94. (Category III) (Grade C). 20. Cherqui D, Benoist S, Malassagne B, et al. Major liver resection for carcinoma in jaundiced patients without preoperative biliary drainage. Arch Surg 2000; 135: 302–8. (Category III) (Grade C). 21. Ferrero A, Lo Tesoriere R, Viganò L, et al. Preoperative biliary drainage increases infectious complications after hepatectomy for proximal bile duct tumor obstruction. World J Surg 2009; 33: 318–25. (Category III) (Grade C). 22. Nimura Y, Kamiya J, Kondo S, et al. Aggressive preoperative management and extended surgery for hilar cholangiocarcinoma: Nagoya experience. J Hepatobiliary Pancreat Surg 2000; 7: 155–62. (Category III) (Grade C). 23. Lee SG, Lee YJ, Park KM, Hwang S, Min PC. One hundred and eleven liver resections for hilar bile duct cancer. J Hepatobiliary Pancreat Surg 2000; 7: 135–41. (Category III) (Grade C). 24. Seyama Y, Kubota K, Sano K, et al. Long-term outcome of extended hemihepatectomy for hilar bile duct cancer with no mortality and high survival rate. Ann Surg 2003; 238: 73–83. (Category III) (Grade C). 25. Kawasaki S, Imamura H, Kobayashi A, et al. Results of surgical resection for patients with hilar bile duct cancer: application of extended hepatectomy after biliary drainage and hemihepatic portal vein embolization. Ann Surg 2003; 238: 84–92. (Category III) (Grade C). 26. Hemming AW, Reed AI, Fujita S, Foley DP, Howard RJ. Surgical management of hilar cholangiocarcinoma. Ann Surg 2005; 241: 693–702. (Category III) (Grade C). 27. Sano T, Shimada K, Sakamoto Y, et al. One hundred two consecutive hepatobiliary resections for perihilar cholangiocarcinoma with zero mortality. Ann Surg 2006; 244: 240–7. (Category III) (Grade C). 28. Nimura Y. Preoperative biliary drainage before resection for cholangiocarcinoma (Pro). HPB 2008; 10: 130–3. (Category III) (Grade C). 29. Nimura Y, Shionoya S, Hayakawa N, et al. Value of percutaneous transhepatic cholangioscopy (PTCS). Surg Endosc 1988; 2: 213–9. (Category III) (Grade C). 30. Nimura Y, Kamiya J, Hayakawa N, Shionoya S. Cholangioscopic differentiation of biliary strictures and polyps. Endoscopy 1989; 21(Suppl 1): 351–6. (Category IV) (Grade D). 31. Nimura Y. Staging of biliary carcinoma: cholangiography and cholangioscopy. Endoscopy 1993; 25: 76–80. (Category IV) (Grade D). 32. Seo DW, Lee SK, Yoo KS, et al. Cholangioscopic findings in bile duct tumors. Gastrointest Endosc 2000; 52: 630–4. (Category IV) (Grade D). 33. Kato M, Nimura Y, Kamiya J, et al. Carcinoma of the common bile duct with superficial spread to the intrahepatic segmental bile ducts: a case report. Am Surg 1997; 63: 943–7. 34. Itoi T, Shinohara Y, Takeda K, et al. Detection of telomerase activity in biopsy specimens for diagnosis of biliary tract cancers. Gastrointest Endosc 2000; 52: 380–6. (Category III) (Grade C). 35. Murata T, Nagasaka T, Kamiya J, et al. P53 labeling index in cholangioscopic biopsies is useful for determining spread of bile duct carcinomas. Gastrointest Endosc 2002; 56: 688–95. (Category III) (Grade C). 36. Nimura Y. Staging cholangiocarcinoma by cholangioscopy. HPB 2008; 10: 113–5. (Category IV) (Grade D). 37. Uesaka K, Nimura Y, Nagino M. Changes in hepatic lobar function after right portal vein embolization: an appraisal by biliary indocyanine green excretion. Ann Surg 1996; 1: 77–83. (Category III) (Grade C). 38. Nagino M, Nimura Y, Kamiya J, et al. Changes in hepatic lobe volume in biliary tract cancer patients after right portal vein embolization. Hepatology 1995; 21: 434–9. (Category III) (Grade C). 39. Makuuchi M, Thai BL, Takayasu K, et al. Preoperative portal embolization to increase safety of major hepatectomy for hilar bile duct carcinoma: a preliminary report. Surgery 1990; 107: 521–7. 40. Nagino M, Nimura Y, Kamiya J, Kondo S, Kanai M. Selective percutaneous transhepatic embolization of the portal vein in preparation for extensive liver resection: the ipsilateral approach. Radiology 1996; 200: 559–63. (Category III) (Grade C). 41. Nagino M, Kamiya J, Kanai M, et al. Right trisegment portal vein embolization for biliary tract carcinoma: technique and clinical utility. Surgery 2000; 127: 155–60. (Category III) (Grade C). 42. Nagino M, Kamiya J, Nishio H, et al. Two hundred forty consecutive portal vein embolizations before extended hepatectomy for biliary cancer: surgical outcome and long-term follow-up. Ann Surg 2006; 243: 364–72. (Category III) (Grade C). 43. Kamiya S, Nagino M, Kanazawa H, et al. The value of bile replacement during external biliary drainage: an analysis of intestinal permeability, integrity, and microflora. Ann Surg 2004; 239: 510–7. (Category Ib) (Grade A). 44. Kanazawa H, Nagino M, Kamiya S, et al. Synbiotics reduce postoperative infectious complications: a randomized controlled trial in biliary cancer patients undergoing hepatectomy. Langenbecks Arch Surg 2005; 390: 104–13. (Category Ib) (Grade A). 45. Sugawara G, Nagino M, Nishio H, et al. Perioperative synbiotic treatment to prevent postoperative infectious complications in biliary cancer surgery: a randomized controlled trial. Ann Surg 2006; 244: 706–14. (Category Ib) (Grade A). 46. Bhuiya MR, Nimura Y, Kamiya J, et al. Clinicopathologic studies on perineural invasion of bile duct carcinoma. Ann Surg 1992; 215: 344–9. (Category III) (Grade C). 47. Kayahara M, Nagakawa T, Ohta T, et al. Role of nodal involvement and the periductal soft-tissue margin in middle and distal bile duct cancer. Ann Surg 1999; 229: 76–83. (Category III) (Grade C). 48. Wakai T, Shirai Y, Moroda T, Yokoyama N, Hatakeyama K. Impact of ductal resection margin status on long-term survival in patients undergoing resection for extrahepatic cholangiocarcinoma. Cancer 2005; 103: 1210–6. (Category III) (Grade C). 49. Jang JY, Kim SW, Park DJ, et al. Actual long-term outcome of extrahepatic bile duct cancer after surgical resection. Ann Surg 2005; 241: 77–84. (Category III) (Grade C). 50. Cheng Q, Luo X, Zhang B, et al. Distal bile duct carcinoma: prognostic factors after curative surgery. A series of 112 cases. Ann Surg Oncol 2007; 14: 1212–9. (Category III) (Grade C). 51. Murakami Y, Uemura K, Hayashidani Y, et al. Pancreatoduodenectomy for distal cholangiocarcinoma: prognostic impact of lymph node metastasis. World J Surg 2007; 31: 337–42. (Category III) (Grade C). 52. Nagino M, Kamiya J, Arai T, et al. “Anatomic” right hepatic trisectionectomy (extended right hepatectomy) with caudate lobectomy for hilar cholangiocarcinoma. Ann Surg 2006; 243: 28–32. (Category III) (Grade C).

341

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
53. Nagino M, Kamiya J, Kanai M, et al. Hepaticojejunostomy using a Rouxen-Y jejunal limb via the retrocolic-retrogastric route. Langenbecks Arch Surg 2002; 387: 188–9. (Category III) (Grade C). 54. Nimura Y, Hayakawa N, Kamiya J, et al. Combined portal vein and liver resection for carcinoma of the biliary tract. Br J Surg 1991; 78: 727–31. (Category III) (Grade C). 55. Neuhaus P, Jonas S, Bechstein WO, et al. Extended resections for hilar cholangiocarcinoma. Ann Surg 1999; 230: 808–19. (Category III) (Grade C). 56. Muñoz L, Roayaie S, Maman D,et al Hilar cholangiocarcinoma involving the portal vein bifurcation: long-term results after resection. J Hepatobiliary Pancreat Surg 2002; 9: 237–41. (Category III) (Grade C). 57. Ebata T, Nagino M, Kamiya J, et al. Hepatectomy with portal vein resection for hilar cholangiocarcinoma: audit of 52 consecutive cases. Ann Surg 2003; 238: 720–7. (Category III) (Grade C). 58. Hemming AW, Kim RD, Mekeel KL, et al. Portal vein resection for hilar cholangiocarcinoma. Am Surg 2006; 72: 599–605. (Category III) (Grade C). 59. Kondo S, Katoh H, Hirano S, et al. Portal vein resection and reconstruction prior to hepatic dissection during right hepatectomy and caudate lobectomy for hepatobiliary cancer. Br J Surg 2003; 90: 694–7. (Category III) (Grade C). 60. Sakaguchi S, Nakamura S. Surgery of the portal vein in resection of cancer of the hepatic hilus. Surgery 1986; 99: 344–9. (Category III) (Grade C). 61. Lorf T, Hanack U, Ringe B. Portal vein replacement by hepatic vein transposition. Am J Surg 1997; 174: 353–4. (Category III) (Grade C). 62. Miyazaki M, Kato A, Ito H, et al. Combined vascular resection in operative resection for hilar cholangiocarcinoma: does it work or not? Surgery 2007; 141: 581–8. (Category III) (Grade C). 63. Nimura Y, Hayakawa N, Kamiya J, et al. Hepatopancreatoduodenectomy for advanced carcinoma of the biliary tract. Hepatogastroenterology 1991; 38: 170–5. (Category III) (Grade C). 64. Nakamura S, Suzuki S, Serizawa A, et al. Hepatopancreatoduodenectomy for superficially spreading bile duct carcinoma: a report of two 5 year survivals. Hepatogastroenterology 1996; 43: 138–42. 65. Kurosaki I, Hatakeyama K, Tsukada K, Date K, Tomiyama T. Major hepatectomy combined with pylorus-preserving pancreatoduodenectomy for middle bile duct cancer with multiple lymph node metastases: a case report of 5-year survival. Hepatogastroenterology 1999; 46: 1623–6. 66. Kaneoka Y, Yamaguchi A, Isogai M, Suzuki M. Longer than 3-year survival following hepato-ligamento-pancreatoduodenectomy for hilar cholangiocarcinoma with vascular involvement: report of a case. Surg Today 2003; 33: 772–6. (Category III) (Grade C). 67. Ebata T, Nagino M, Nishio H, Arai T, Nimura Y. Right hepatopancreatoduodenectomy: improvements over 23 years to attain acceptability. J Hepatobiliary Pancreat Surg 2007; 14: 131–5. (Category III) (Grade C). 68. D’Angelica M, Martin RC 2nd, Jarnagin WR, et al. Major hepatectomy with simultaneous pancreatectomy for advanced hepatobiliary cancer. J Am Coll Surg 2004; 198: 570–6. (Category III) (Grade C). 69. Nimura Y. Radical surgery: vascular and pancreatic resection for cholangiocarcinoma. HPB 2008; 10: 183–5. (Category III) (Grade C). 70. Yoshimi F, Asato Y, Amemiya R, Shioyama Y, Itabashi M. Comparison between pancreatoduodenectomy and hepatopancreatoduodenectomy for bile duct cancer. Hepatogastroenterology 2001; 48: 994–8. (Category III) (Grade C). 71. Miwa S, Kobayashi A, Akahane Y, et al. Is major hepatectomy with pancreatoduodenectomy justified for advanced biliary malignancy? J Hepatobiliary Pancreat Surg 2007; 14: 136–41. (Category III) (Grade C). 72. Kaneoka Y, Yamaguchi A, Isogai M. Hepatopancreatoduodenectomy: its suitability for bile duct cancer versus gallbladder cancer. J Hepatobiliary Pancreat Surg 2007; 14: 142–8. (Category III) (Grade C). 73. Kennedy TJ, Yopp A, Qin Y, et al. Role of preoperative biliary drainage of liver remnant prior to extended liver resection for hilar cholangiocarcinoma. HPB 2009; 11: 445–51. (Category III) (Grade C).

342

38

Endoscopic management of malignant biliary obstruction Nick Stern and Richard Sturgess
ultrasound
One of the first investigations important in the jaundiced patient is an ultrasound of the liver and biliary tree. This will determine whether the jaundice is obstructive or whether it is due to a parenchymal liver disease or pre-hepatic cause (10,11). Obstructive jaundice cannot be diagnosed without imaging of the biliary tree. It is worth noting that abnormal LFTs in a cholestatic pattern do not necessarily equate to biliary obstruction. Trans-abdominal ultrasound will be able to confirm whether the patient has gall bladder calculi, which can cause obstructive jaundice if they migrate to the bile ducts. It can also detect whether there are intra-hepatic lesions, such as liver metastases or primary liver malignancies (12,13). The main role in this context, however, is to detect whether there is biliary obstruction. A mildly dilated common bile duct is expected in patients with prior cholecystectomy. It is not abnormal for the common bile duct to increase in size slightly with age. Should there be dilatation of the intra-hepatic ducts, this almost always suggests pathology and biliary obstruction.

background
There are many aspects of diagnosis, staging, and therapy in biliary malignancies that are managed by the endoscopist. As different imaging modalities of the biliary system have advanced, there has been a move in recent years away from using endoscopic retrograde cholangio-pancreatography (ERCP) as a diagnostic tool. High-quality diagnostic images of the biliary tree can now be obtained with modern radiological techniques. Newer MRI scanners using an MRCP protocol (1–3) can provide highquality cholangiography and cross-sectional imaging. Endoscopic ultrasound (4,5) allows sonographic views from the lumen of the upper GI tract, particularly of the distal biliary system and pancreas. These, along with the improved quality of CT scanning (6–8) can provide safe diagnosis and staging of biliary malignancies, saving potentially risky ERCP for appropriate therapeutic procedures.

causes of malignant biliary obstruction
Common causes of malignant biliary obstruction include pancreatic carcinoma, primary biliary or liver tumors: cholangiocarcinoma or some hepatocellular carcinomas; or metastatic disease, including lymphadenopathy. This is often at the porta hepatis but can also occur distally (9). The majority of patients with these malignancies will not undergo surgical resection of the tumor due to either the advanced nature of the disease (inoperable disease) or the co-existing morbidities precluding surgery. There are therefore, a large proportion of patients that will require a non-operative intervention to manage their biliary obstruction that is often associated with debilitating symptoms. This is commonly offered in the form of procedures to palliate jaundice. These are normally performed endoscopically or via a percutaneous radiological approach. While these are normally palliative procedures, in some cases access to the biliary system provides the opportunity to deliver potentially survival-enhancing therapies which we will be described later in this chapter.

ct scanning
For detailed information about potential causes of obstructive jaundice, a high-quality, contrast-enhanced CT scan is often necessary. CT scanning can give important information about the pancreas (often obscured by bowel gas on ultrasound scanning). Heads of pancreas tumors, that often present with obstructive jaundice, are normally visible on targeted, enhanced CT scans of the pancreas (14,15). The degree of biliary dilatation and the site of any caliber change in the biliary tree can provide additive information about the tumor. Segmental intra-hepatic dilatation of the bile ducts can suggest cholangiocarcinoma. Lymphadenopathy can cause malignant biliary obstruction and when this occurs this tends to be at the porta hepatis. This is normally well seen on CT scans with the opportunity to get information about a possible site of primary tumor. Large colonic masses or widespread lymphadenopathy with splenomegaly may suggest a possible cause of the malignancy. The importance of CT scanning is that of diagnosing the primary tumor, as well as providing information for staging and operability. CT scanning should ideally be performed prior to any endoscopic intervention of the biliary tree (ERCP) as the complication of pancreatitis can interfere with accurate staging of the disease. While CT may give a good indication of the site of biliary obstruction and the cause, detailed views of the biliary tree are better obtained with MR scanning.

radiological diagnostic imaging
Following presentation with suspected biliary malignancy, radiological investigations are necessary to make a diagnosis and stage the disease. Radiology is particularly important to help clarify the appropriate management and to target and plan any therapy that will later be necessary for the patient. As with all patient management, assessment should begin with the taking of a good history and examination of the patient. Depending on the symptoms and mode of presentation, the pathway for investigation will vary. What we will describe will be assuming that most patients present with symptoms and signs of cholestatic jaundice due to their biliary obstruction.

mr scanning
Magnetic Resonance Imaging (MRI scanning) is increasingly important in providing good imaging of the biliary tree (1–3).

343

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
It can complement CT scanning in selected patients and has the advantage over CT of not exposing the patient to ionizing radiation. Newer MRI scanners with stronger magnets provide detailed cholangiograms without the need of intravenous contrast using an MRCP (magnetic resonance cholangio-pancreatography) protocol. These are obtained without needing endoscopic instrumentation of the bile ducts and therefore eliminate the risks of pancreatitis and other risks associated with ERCP. While MRCP provides important diagnostic information, it is purely used for diagnostic information and interventional methods (ERCP or PTC) are needed to provide therapy. MRCP is very useful in patients with biliary strictures caused by lesions not visible on CT scanning, such as small cholangiocarcinomas. It helps with the planning of therapy and in decision making about whether intervention may be best carried out endoscopically or percutaneously (Fig. 38.1). EUS is a very useful way of imaging the biliary tree. It is the most sensitive modality for detecting small bile duct calculi and in the context of malignancy can look intimately at the pancreas as well as direct visualization of the ampulla and ultrasonographic views of any lesions (Fig. 38.3). The sensitivity of EUS does diminish with more proximal lesions in the biliary tree and the liver itself. As well as the imaging advantages with EUS, it provides a useful and relatively safe way to sample tissue with EUS-guided fine needle aspiration cytology (16,17). This can be used to target pancreatic lesions, lymph nodes, ampullary lesions as well as non-biliary disease such as mediastinal lymphadenopathy. One of the current limitations to EUS as a modality is due to its availability. EUS (unlike CT or MRI scanners) is only available at certain specialized centers in the United Kingdom; however, most regional hepatobiliary units are likely to have an EUS service.

endoscopic assessment
Endoscopic Intervention The anatomy of the biliary tree allows good access to the biliary system with an endoscope. The bile ducts drain, via the common bile duct, through the ampulla of Vater and sphincter of Oddi into the second part of the duodenum. This allows an accessible port of access to the biliary system by use of an endoscope designed to sit opposite the ampulla within the duodenum. The use of endoscopy in the management of malignant biliary obstruction can be in the diagnosis, staging, and therapy of the disease. Endoscopic Ultrasound (Endosonography, EUS) A relatively new and very useful part of diagnosis and staging is the method of endoscopic ultrasound (EUS) (4). This involves an endoscope designed with an ultrasound probe at the tip (Fig. 38.2). The imaging can be either a “radial” EUS that provides a 360 degree image around the probe at the tip of the endoscope or a “linear” EUS that gives an ultrasound picture in the plane of the scope.

ercp
Endoscopic Retrograde Cholangio-Pancreatography (ERCP) is an endoscopic method of accessing the biliary tree. ERCP is performed using a side-viewing endoscope, passed normally to the second part of the duodenum, sitting opposite the ampulla of Vater (Fig. 38.4).

Figure 38.1 MRCP cholangio.

Figure 38.2 EUS stack.

344

ENDOSCOPIC MANAGEMENT OF MALIGNANT BILIARY OBSTRUCTION
When in position, the biliary tree (or pancreatic duct if appropriate) is accessed by using a hollow cannula with a 0.035″ diameter guide wire being passed through the working channel of the endoscope into the duct of choice under fluoroscopic screening. When in position, a radio-opaque contrast agent is injected through the cannula, opacifying the duct of interest on fluoroscopy. The cholangiogram (or pancreatogram) can then be used to confirm the diagnosis, often a filling defect such as a common bile duct calculus or a stricture that may be malignant or benign (Figs. 38.5 and 38.6). Further diagnostic information can be gained by tissue acquisition. This normally involves passing a cytology brush over the wire and brushing the stricture for cytological analysis. Depending on the pathology confirmed, therapy can be delivered as appropriate. Sphincterotomy can be performed using over the wire “sphincterotome” with diathermy current allowing better access to the duct. This can allow extraction of calculi with either balloon trawls or basket. If strictures are present causing jaundice, the stricture can be stented to enable good biliary drainage.

(A)

(B) Figure 38.3 (A) EUS cholangio. (B) A Eus Ca Panc (arrow indicates the pancreatic tumor).

Figure 38.4 ERCP stack.

Figure 38.5 ERCP distal stricture.

345

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Indications for ERCP Common Indications for ERCP Obstructive jaundice
● ●

this risk is much smaller than the risk of a non-draining, obstructed biliary system. ERCP Complications Pancreatitis Perforation Bleeding Cholangitis Drug reactions/effects Aspiration pneumonia

Definitive biliary drainage Pre-operative drainage

Confirmed/strongly suspected bile duct calculi (EUS/MRCP would often be used prior to ERCP) Severe gallstone pancreatitis (particularly with jaundice and cholangitis) Cholangitis Bile leak post cholecystectomy Cytological sampling of strictures Pancreatic duct dilatation/stenting Sphincter of Oddi manometry

direct cholangioscopy
A technique that has developed for the assessment of intrabiliary pathology is that of direct, per oral cholangioscopy. Whereas traditional ERCP provides detailed information about intra-biliary pathology, this is normally part of a contrastenhanced radiological cholangiogram performed with the assistance of an endoscope rather than direct vision. Strictures may be misclassified as malignant rather than benign, particularly with the relatively poor yield of biliary brushings (25,26). Direct cholangioscopy provides the user the opportunity to see directly into the biliary tree to the second- or third-order ducts. A clearer impression about the nature of a stricture, be it benign or malignant can be obtained and the cause of the stricture can be biopsied as well as brushed, and this can be done under direct vision, thereby improving the yield of diagnostic histology to about 80% (27,28). The original cholangioscopy via ERCP scopes, were the “mother and baby scopes” that required two operators to use with one controlling the duodenoscope (mother) and the second operating the smaller cholangioscope (baby). The baby scope passed through the working channel of the duodenoscope. As well as being clumsy, irrigation and visualization were poor and newer technologies have enabled improved and constant irrigation improving cholangioscopic views.

Given the potential complications of ERCP (see below), it is now almost totally used as a method of providing therapy to the biliary tree and pancreatic duct. In the context of malignant biliary obstruction, therapy is often targeted at the relief of jaundice. With malignancy, drainage of jaundice normally involves the insertion of one of the variety of stents. When planning to drain jaundice in the patient with malignancy, consideration needs to be given to the choice of stent type, the site of stenting – whether unilateral in hilar strictures, or bilateral and the method of approaching this. It is because of this reason that reviewing good quality imaging prior to the procedure and planning the procedure in advance is of paramount importance. Additional diagnostic information can be obtained using direct cholangioscopy. This is an emerging technology and its current availability is limited. It is being used for those strictures that can’t be classified with conventional imaging.

complications of ercp
The reason to limit the use of ERCP for therapy rather than diagnosis, unlike most other modalities of endoscopy, is the risk profile. Because of the anatomy of the biliary tree, and the distal portion of the common bile duct passing through the head of the pancreas, the main risk of ERCP is that of acute pancreatitis. A lot of effort has been put into reducing the risk as much as possible. This includes the improved training of ERCP endoscopists, as well as the development of guidance suggesting that a fewer number of endoscopists perform ERCP to increase the numbers performed by each individual. The technical changes of wire-guided cannulation are also thought to reduce pancreatitis risk as wire cannulation of the pancreatic duct should result in lower rates of ERCP-induced pancreatitis than opacification with contrast (18). Rates of pancreatitis have been quoted very variably as 1% to 30% (18–22); however, the recent large volume British Society of Gastroenterology (BSG) audit of ERCP practice in the United Kingdom reported pancreatitis rates of 1.5% with an overall complication rate of 5.1% (23). Other ERCP complications are those for standard endoscopy, namely bleeding and perforation. Another risk following stenting of the ampulla is the risk of cholangitis (24); however,

Figure 38.6 ERCP proximal stricture.

346

ENDOSCOPIC MANAGEMENT OF MALIGNANT BILIARY OBSTRUCTION
A new style of cholangioscope has been developed that can allow a single endoscopist the chance to visualize the bile ducts (Fig. 38.7). The SpyglassTM is a method of passing a 10 F catheter into the bile duct over a wire as normal, and a fiberoptic cable can be passed through one of four working channels to provide direct vision. Intra-biliary anatomy and pathology can be clearly identified and treatment can be targeted as necessary. Indeterminate strictures are able to be assessed as in conventional endoscopy and biopsies taken under direct vision. In the case of large stone disease, these areas can be closely evaluated with the additive benefit of a therapeutic channel to allow direct electro-hydrolic or laser lithotripsy when conventional methods have failed (Figs. 38.8 and 38.9). Endoscopic Therapy Indications The use of endoscopy in the management of biliary malignancy can be to aid diagnosis and staging, with the use of endoscopic ultrasound for assessment or histological sampling. A cholangiogram is a useful diagnostic adjunct, however as detailed previously can be obtained in a safer way with lower risk techniques. ERCP still has a major role in the therapeutic management of biliary malignancy. These therapies can broadly can be subdivided into palliating symptoms, namely jaundice (29–31), and as a method of providing disease modifying treatment to the biliary tree. Most cases of biliary malignancy requiring ERCP will be those presenting with jaundice, and the need for this to be treated either as a palliative modality or as a bridge to surgery to reduce peri-operative morbidity. Methods of Therapy Stent The use of stenting in the drainage of obstructive jaundice is well established (31–34). The decision that has to be made by the multi-discipliniary team managing the jaundiced patient with malignancy is that of the optimum way of accessing the biliary tree (endoscopic or percutaneous), the type of stent to be used (plastic or metal, which size, and if metal, covered or uncovered). The other area that needs to be assessed and decided upon, ideally prior to intervention, is the need for unilateral or bilateral stenting for patients with hilar obstruction – so-called Klatskin tumors (35). A variety of factors need to be considered when deciding on the appropriateness of stent and the type of stent to be used. The first of these is often whether this is to be used as a definitive palliative treatment, or as a bridge to surgery in the jaundiced patient (Figs. 38.10 and 38.11).

Figure 38.7 ERCP and spyglass stack.

Figure 38.8 Spyglass—single operator.

Figure 38.9 Spyglass in use—Stone disease.

347

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Percutaneous Drainage of Jaundice One of the methods available for the drainage of obstructive jaundice is via a per-cutaneous route (36). This involves the interventional radiologist accessing the biliary tree percutaneously via a trans-hepatic approach (37). Percutaneous trans-hepatic cholangiography (PTC) can be performed as a temporary measure, with an intra-biliary catheter left in situ and an external drain to relieve jaundice externally. Alternatively an “internal–external” drain can be sited that crosses the stricture, which also allows an external drain for the relief of jaundice, leaving a catheter in the duodenum to allow internal drainage. This can then be internalized with a combined approach involving a radiologist performing the percutaneous aspect and an endoscopist performing the ERCP (38). Permanent stents can be sited via the percutaneous approach as both 10 F plastic (39) and metal stents (40) can be introduced collapsed and expanded as would be done at ERCP. Operating on patients with jaundice does increase morbidity (48–51) and mortality (52). For those patients who require palliation this is normally very well achieved with non-operative stenting done as an endoscopic or radiological procedure that has a significantly lower morbidity than surgical palliative methods (37,39,45). In jaundiced patients with potentially curative disease, morbidity can be reduced by the use of pre-operative stenting (48), be it endoscopic or percutaneous (53). Recent data, however, do suggest that in those with pancreatic cancer with a bilirubin <250 µmol/l early surgery reduces the complication rate when compared to pre-operative plastic stenting (54). Due to the combination of the reasons listed above, as well as the reduced morbidity and shorter in-hospital stays, nonoperative management of malignant jaundice has the added advantage of being a cost-effective treatment (32,55). While there are advantages to an endoscopic approach to biliary stenting and drainage of jaundice, these are seen maximally in distal biliary strictures where ERCP has become the gold standard of treatment (56). This is due to the relatively easy access to the common bile duct and ability to site a stent that will cover the stricture and allow complete drainage from all areas of the liver. Morbidity is reduced in these patients when compared to the risk of bleed and bile leak that can occur with per-cutaneous drainage (57,58). It is worth noting that many studies and median survivals in regarding biliary malignancies include patients without histological proof of malignancy and so care must be taken in the interpretation of these (59).

ercp versus ptc versus surgery
When approaching the optimal way to drain a patients’ jaundice, the method of approach is important and needs to be carefully considered. Comparing ERCP to surgical bypass, both have been shown to be effective in the palliation of jaundice (41–43). Surgery, while effective does carry a higher complication rate (44,45) and a higher procedure-related mortality (46). With the increasing early literature on the feasibility of endoscopic and percutaneous (36) stenting, and the advanced safety profile compared to surgery this started to be the preferred method of treatment (47).

plastic versus metal
Early studies looking at the role of endoscopically sited biliary stents looked at narrow caliber stents generally 6 to 8 F (30). These often provided symptomatic relief of the patients’ jaundice; however, occlusion with further episodes of obstructive jaundice was common. This has improved with the use of therapeutic duodenoscopes able to site larger plastic stents, and now in patients

Figure 38.10 Metal stent—distal stricture.

Figure 38.11 Metal stents—proximal stricture.

348

ENDOSCOPIC MANAGEMENT OF MALIGNANT BILIARY OBSTRUCTION
with malignant obstructive jaundice, normal practice would be to site a stent of at least 10 F caliber (60). While these do provide longer periods of palliation for patients than the previously used narrow stents, there is still a significant rate of stent occlusion and plastic stents will often need to be replaced after about 4 to 6 months. Self-expanding metal stents (SEMS) have been developed for a variety of strictures and these have been shown to have longer stent patency than plastic stents (61). SEMS can be sited percutaneously (62–64) or endoscopically (40,65). They have been shown to be cost-effective in patients who are expected to survive more than 4 to 6 months (66–68) as they reduce hospital admissions and repeated procedures (69–72). Favorable increases in stent patency appeared to be related to the early expansion of the stent (73). SEMS are generally not removable after being sited, therefore these are often sited for palliative patients. In those patients having stents sited for drainage of jaundice prior to potentially curative resection, 10 F plastic stents are often the stent of choice. The insertion of a short SEMS needn’t be a contraindication to a curative resection (74) if placed with care. The development of fully covered metal stents does allow for the removal of SEMS following their deployment. In those patients that do not achieve adequate drainage with a unilobar stent, subsequent attempts at bilobar stenting are very problematic due to a track being patent into the drained lobe, and so hilar lesions are normally best managed at centers that perform high volumes with local policies regarding the relative role of ERCP and PTC. In this scenario, it is likely to be advantageous to attempt draining the un-drained lobe via the percutaneous approach that will enable direct targeting of the obstructed system and placement of stent. Disease Modifying Treatment While most of the endoscopic management concentrated on to this point has been in the palliation of jaundice due to biliary malignancies, there is a role developing for disease modifying therapy that can be delivered directly to the biliary tree at endoscopy. We concentrate on the experience with radiotherapy as well as photodynamic therapy.

radiotherapy
Brachytherapy Intraductal radiotherapy, brachytherapy has been used in a small number of studies in an attempt to improve survival in patients with nonresectable cholangiocarcinoma. A retrospective study comparing brachytherapy to stenting alone did show some survival benefit, that didn’t achieve statistical significance (p = 0.06) in patients with type II or III tumors treated with brachytherapy, but this was associated with more stent changes and longer hospital stays. The authors felt that the benefit was limited to those with type II or III tumors treated within 10 months of diagnosis (86). A more recent study looking at external beam radiotherapy in conjunction with expandable metal stents in patients with cholangiocarcinoma, showed longer survival in those than in stents alone (10.6 vs. 6.4 months, p < 0.05) and longer stent patency than metal stents alone (9.8 vs. 3.7 months, p < 0.001);

covered versus uncovered
Covered stents have been developed where by the SEMS has a thin plastic covering designed to reduce tumor in-growth (75–77). These have been developed to improve stent patency further and have been shown to have a longer median patency before occlusion than uncovered stents (78). When placing metal stents at the hilum, uncovered stents are preferable as covered stents are likely to occlude the bile flow from the un-stented lobe and tributaries. There is a risk that covered stents that occlude the cystic duct insertion in patients with a gall bladder in place may put the patients at risk of cholecystitis (79), and are therefore often avoided. Covered stents in distal biliary strictures, such as pancreatic carcinoma, can be useful and improve stent patency and median time to occlusion (78).

hilar strictures
Unilateral Versus Bilateral The more difficult scenario is in strictures that affect the bifurcation of the common hepatic duct, the hilar strictures. These are commonly caused by cholangiocarcinomas and if extend beyond the hilum a decision has to be made about the best approach to drainage (Fig. 38.12). While the optimal drainage is achieved by bi-lobar drainage, this can prove technically difficult, and the great danger in this group of patients is in failing to drain an opacified area. This has been shown in various series to increase morbidity and mortality due to cholangitis (80,81). Adequate palliation of jaundice is likely to be obtained with drainage of 30% of the liver volume and this can normally be achieved through unilobar stenting (82–85). Due to these problems, in some centers PTC is preferred to ERC as at PTC the obstructed area is targeted first and this can easily be drained.

Figure 38.12 Hilar stricture (cholangiocarcinoma).

349

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
however, this was with a shorter patency than previously described for metal stents (87). The variety of approaches described using radiotherapy can best be regarded as of unproven benefit. procedure to ensure continued biliary drainage following treatment. The major side effect of PDT is skin photosensitization. Patients are advised to stay indoors away from bright light for 3 to 4 days and then cautiously increase exposure to sunlight avoiding strong sunlight for 4 to 6 weeks. The contraindications to PDT include porphyria and decompensated cirrhosis. After a number of uncontrolled studies suggesting significant benefit of ERC-PDT for cholangiocarcinoma (89–93), a randomized study, looking at non-resectable cholangiocarcinomas randomized to stenting alone or stenting and PDT showed an increase in median survival from 98 to 493 days (p < 0.0001) with some of the PDT group still alive at analysis (94). This study was terminated prematurely because of the superior results from PDT made further randomization unethical. In addition this study excluded patients who had a successful stenting procedure previously. Thus the results may not be applicable to the group of patients who do have an initial successful stenting procedure. Further controlled studies have shown significant improvements in survival from 7 to 21 months with PDT (95). An uncontrolled study showed a median survival post-PDT of 276 days (89). A large retrospective study, looking at patients with hilar cholangiocarcinoma treated with either surgery, stenting alone, or stenting with PDT showed longer survival with PDT compared to stenting (p < 0.01) and similar survival to R1/R2 resection (96). Generally ERC-PDT was tolerated well with mild photosensitivity, the commonest problem, and cholangitis also causing significant problems in some studies. These studies suggested significant promise for ERC-PDT as a therapy for cholangiocarcinoma. In addition to the criticisms detailed above, however, the studies have been small with variability in the number and type of stent used. The number of PDT sessions have varied and in uncontrolled

photodynamic therapy
The greater majority of patients with cholangiocarcinoma do not undergo resection (88). The disease has a particularly poor prognosis, especially if the disease is advanced with a median survival time of 62 days if there is bilateral intrahepatic disease (81). Photodynamic therapy (PDT) has emerged as a promising new modality of treatment for patients who do not undergo resection. PDT uses the combination of two non-toxic moieties; light and a photosensitizing chemical, which when combined together produce a cytotoxic effect. The photosensitizer tends to accumulate in proliferating tissue. This tissue is then illuminated with light of a wavelength appropriate to the absorption spectrum of the photosensitizer. A photochemical reaction generates cytotoxic reactive oxygen species resulting in apoptosis or necrosis of tumor cells. PDT may also cause thrombosis in tumor blood vessels and induce a tumor-specific immune reaction with more distant effects from the site of illumination. Photofrin® (Axcan Pharma Inc.) has been the most widely used photosensitizer in this application and is a complex mix of compounds derived from hematoporphyrin. It is activated at a wavelength of 630 nm. The depth of necrosis obtained is 4 to 6 mm. The technique of ERC-PDT involves the prior administration of Photofrin® at a dose of 2 mg/kg, optimally 48 hours before the procedure. A cylindrical laser diffuser fiber is positioned across the malignant stricture through a standard cannula. Concurrent oxygen administration (4 l/min) optimizes the PDT effect. A light dose of 180 to 200 J/cm is delivered. Stents, preferably plastic, are inserted after the
Refer to gastroenterology/ hepatology

Jaundiced patient

Non-obstructed biliary system

Biliary ultrasound

Obstructed biliary system

Other imaging Hepatobiliary MDT meeting review Operable

Cross sectional imaging (CT/MR)

Inoperable

Life expectancy <4 months Resection ±perioperative biliary drainage

Life expectancy ≥4 months ERC/PTC and metal stent

ERC/PTC and plastic stent

Figure 38.13 Algorithm for management of malignant biliary obstruction.

350

ENDOSCOPIC MANAGEMENT OF MALIGNANT BILIARY OBSTRUCTION
studies, it is difficult to determine what benefit is due to PDT and that due to stenting. The Photostent-2 trial had been designed to look at this area, and despite the promising prior studies, this was terminated early due to an increased mortality in the PDT arm compared to the stent arm. Interestingly, this improvement in the non-treated arm involved a longer median survival than previously described in historical studies and in the pilot series. This may be due to improved biliary drainage in this group.
4. Rösch T, Lorenz R, Braig C, et al. Endoscopic ultrasound in pancreatic tumor diagnosis. Gastrointest Endosco 1991; 37(3): 347–52. 5. Saftoiu A, Vilmann P. Role of endoscopic ultrasound in the diagnosis and staging of pancreatic cancer. J Clin Ultrasound 2009; 37(1): 1–17. 6. Zandrino F, Benzi L, Ferretti M, et al. Multislice CT cholangiography without biliary contrast agent: technique and initial clinical results in the assessment of patients with biliary obstruction. Euro Radiol 2002; 12(5): 1155–61. 7. Ochotorena IJL, Kiyosue H, Hori Y, et al. The local spread of lower bile duct cancer: evaluation by thin-section helical CT. Euro Radiol 2000; 10(7): 1106–13. 8. Zech CJ, Schoenberg SO, Reiser M, Helmberger T. Cross-sectional imaging of biliary tumors: current clinical status and future developments. Euro Radiol 2004; 14(7): 1174–87. 9. Stern N, Sturgess R. Endoscopic therapy in the management of malignant biliary obstruction. Euro J Surg Oncol (EJSO) 2008; 34(3): 313–7. 10. K. J. W. Taylor DACVRM. Ultrasound and scintigraphy in the differential diagnosis of obstructive jaundice. J Clin Ultrasound 1974; 2(2): 105–16. 11. Gordon S. Perlmutter BBGGS. Ultrasonic evaluation of the common bile duct. J Clin Ultrasound 1976; 4(2): 107–11. 12. Taylor KJW, Carpenter DA, McCready VR. Grey scale echography in the diagnosis of intrahepatic disease. J Clin Ultrasound 1973; 1(4): 284–7. 13. Colin RM. Ultrasonic diagnosis of liver metastases. Journal of Clinical Ultrasound 1976; 4(4): 265–8. 14. Fuhrman GM, Charnsangavej C, Abbruzzese JL, et al. Thin-section contrast-enhanced computed tomography accurately predicts the resectability of malignant pancreatic neoplasms. Am J Surg1994; 167(1): 104–13. 15. Itai Y, Araki T, Tasaka A, Maruyama M. Computed tomographic appearance of resectable pancreatic carcinoma. Radiology 1982 June 1, 1982; 143(3): 719–26. 16. Vilmann P, Jacobsen GK, Henriksen FW, Hancke S. Endoscopic ultrasonography with guided fine needle aspiration biopsy in pancreatic disease. Gastrointest Endosc 1992; 38(2): 172–3. 17. Agarwal B, Abu-Hamda E, Molke KL, Correa AM, Ho L. Endoscopic ultrasound-guided fine needle aspiration and multidetector spiral CT in the diagnosis of pancreatic cancer. Am J Gastroenterol 2004; 99(5): 844–50. 18. Artifon ELA, Sakai P, Cunha JEM, et al. Guidewire cannulation reduces risk of post-ercp pancreatitis and facilitates bile duct cannulation. Am J Gastroenterol 2007; 102(10): 2147–53. 19. Freeman ML, Nelson DB, Sherman S, et al. Complications of endoscopic biliary sphincterotomy. N Engl J Med 1996 September 26, 1996; 335(13): 909–19. 20. Freeman ML, DiSario JA, Nelson DB, et al. Risk factors for post-ERCP pancreatitis: a prospective, multicenter study. Gastrointest Endosco 2001; 54(4): 425–34. 21. Freeman ML. Adverse outcomes of ERCP. Gastrointest Endosco 2002; 56(6, Suppl 1): S273–82. 22. Vandervoort J, Soetikno RM, Tham TCK, et al. Risk factors for complications after performance of ERCP. Gastrointest Endosco 2002; 56(5): 652–6. 23. Williams EJ, Taylor S, Fairclough P, et al. Are we meeting the standards set for endoscopy? Results of a large-scale prospective survey of endoscopic retrograde cholangio-pancreatograph practice. Gut 2007 June 1, 2007; 56(6): 821–9. 24. Motte S, Deviere J, Dumonceau J-M, et al. Risk factors for septicemia following endoscopic biliary stenting. Gastroenterology 1991; 101(5): 1374–81. 25. Angelo Paulo F, David RL, Adam S, Catherine C, David LC-L. Brush cytology during ERCP for the diagnosis of biliary and pancreatic malignancies. Gastrointest Endosc 1994; 40(2): 140–5. 26. Mahmoudi N, Enns R, Amar J, AlAli J, Lam E, Telford J. Biliary brush cytology: factors associated with positive yields on biliary brush cytology. World J Gastroenterol 2008 January 28; 14(4): 569–73. 27. Dwyer L, Polavarapu N, Hood S, Sheard J, Sturgess R. Per oral cholangioscopy: Systematic evaluation in clinical practice. Gut 2010; 59(Suppl 1): A88–9. 28. Aljabiri M, Lerman A, Kalaitzakis E, et al. Initial experience of Spyglass per-oral cholangioscopy in a UK centre. Gut 2010; 59(Suppl 1): A87. 29. Burcharth F, Jensen LI, Olesen K. Endoprosthesis for internal drainage of the biliary tract. Technique and results in 48 cases. Gastroenterology 1979; 77(1): 133–7.

current u.k. practice
In the United Kingdom, cancer care is delivered in local cancer centers. For rarer malignancies, these tend to be based within a regional center. Patients treated at a regional center for hepatobiliary malignancy will be discussed at the appropriate multidisciplinary team (MDT) meeting, comprising gastroenterologists/hepatologists, hepatobiliary surgeons, oncologists, GI radiologists, and interventional radiologists as well as palliative care representatives and specialist nurses. Following an MDT discussion, individualized treatment regimens will be offered to each patient when seen by a member of the team. This will include a review of the appropriate cross-sectional imaging and the detailed planning of any intervention necessary (Fig. 38.13).

conclusions
The management of biliary malignancy can be a complex issue. There are various treatment options that are appropriate and the choice of optimal treatment may depend on local expertise. We have found that in patients with proximal biliary neoplasia, the risks of cholangitis are higher than in other groups, particularly when the biliary system has been previously instrumented. Our group strongly advocates that patients with a proximal biliary malignancy should be transferred to the regional specialist center for a review of the imaging and an appropriate method of biliary decompression to be arranged and performed there. Local data suggest that this reduces the rate of cholangitis in this cohort of patients. We have found that the drainage of biliary disease can be optimized with the use of SEMS as opposed to 10 F plastic stents and that patients can still be operable following this. Carefully sited SEMS can allow biliary drainage and a welldecompressed system without interfering in the surgical field. We have also started using fully covered SEMS which have proved safe to remove up to 3 months following deployment.

references
1. Ishizaki Y, Wakayama T, Okada Y, Kobayashi T. Magnetic resonance cholangiography for evaluation of obstructive jaundice. Am J Gastroenterol 1993 Dec; 88(12): 2072–7. 2. Pavone P, Laghi A, Passariello R. MR cholangiopancreatography in malignant biliary obstruction. Semin Ultrasound CT MRI 1999 Oct; 20(5): 317–23. 3. Vaishali MD, Agarwal AK, Upadhyaya DN, et al. Magnetic resonance cholangiopancreatography in obstructive jaundice. J Clin Gastroenterol 2004; 38(10): 887–90.

351

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
30. Cotton PB. Duodenoscopic placement of biliary prostheses to relieve malignant obstructive jaundice. Br J Surg 1982; 69(9): 501–3. 31. Soehendra N, Reynders-Frederix V. Palliative bile duct drainage - a new endoscopic method of introducing a transpapillary drain. Endoscopy 1980; 12(01): 8–11. 32. Brandabur J, Kozarek R, Ball T, et al. Nonoperative versus operative treatment of obstructive jaundice in pancreatic cancer: cost and survival analysis. Am J Gastroenterol 1988 Oct; 83(10): 1132–9. 33. Walta DC, Fausel CS, Brant B. Endoscopic biliary stents and obstructive jaundice. Am J Surg 1987; 153(5): 444–7. 34. Earnshaw J, Hayter J, Teasdale C, Beckly D. Should endoscopic stenting be the initial treatment of malignant biliary obstruction? Ann R Coll Surg Engl 1992 Sep; 74(5): 338–41. 35. Klatskin G. Adenocarcinoma of the hepatic duct at its bifurcation within the porta hepatis. An unusual tumour with distinctive clinical and pathological features. Am J Med 1965 February; 38: 241–56. 36. Koven I, Steinhardt M, Reichstein B. Percutaneous antegrade biliary drainage: a nonoperative approach to biliary obstruction. Can J Surg 1981; 24(6): 591–3. 37. Gouma DJ, Wesdorp RIC, Oostenbroek RJ, Soeters PB, Greep JM. Percutaneous transhepatic drainage and insertion of an endoprosthesis for obstructive jaundice. Am J Surg 1983; 145(6): 763–8. 38. Robertson D, Ayres R, Hacking C, et al. Experience with a combined percutaneous and endoscopic approach to stent insertion in malignant obstructive jaundice. Lancet 1987 Dec 19; 2(8573): 1449–52. 39. Pereiras RV Jr., Rheingold OJ, Hutson D, et al. Relief of malignant obstructive jaundice by percutaneous insertion of a permanent prosthesis in the biliary tree. Ann Intern Med 1978 November 1, 1978; 89(5_Part_1): 589–93. 40. Pinol V, Castells A, Bordas JM, et al. Percutaneous self-expanding metal stents versus endoscopic polythene endoprostheses for treating malignant biliary obstruction: randomized clinical trial. Radiology 2002; 225(1): 27–34. 41. Lameris J, Stoker J, Dees J, et al. Non-surgical palliative treatment of patients with malignant biliary obstruction - the place of endoscopic and percutaneous drainage. Clin Radiol 1987 Nov; 38(6): 603–8. 42. Andersen JR, Sorensen SM, Kruse A, Rokkjaer M, Matzen P. Randomised trial of endoscopic endoprosthesis versus operative bypass in malignant obstructive jaundice. Gut 1989 August 1, 1989; 30(8): 1132–5. 43. Ozaki H, Hojo K, Miwa K. A safe method of intrahepatic biliary drainage in obstructive jaundice. Am J Surg 1977; 133(3): 379–82. 44. Sonnenfeld T, Gabrielsson N, Grangvist S, Perbeck L. Nonresectable malignant bile duct obstruction. Surgical bypass or endoprosthesis? Acta Chir Scand 1986 Apr; 152: 297–300. 45. Burcharth F, Efsen F, Christiansen L, et al. Nonsurgical internal biliary drainage by endoprosthesis. Surg Gynaecol Obstet 1981; 153(6): 857–60. 46. Smith A, Dowsett J, Russell R, Hatfield A, Cotton P. Randomised trial of endoscopic stenting versus surgical bypass in malignant low bileduct obstruction. Lancet 1994 December 17; 344: 1655–60. 47. Gholson C, Burton F. Obstructive jaundice. Nonsurgical options for ‘surgical’ jaundice. Postgrad Med 1991 Dec; 90(8): 107–10. 48. Denning DA, Ellison EC, Carey LC. Preoperative percutaneous transhepatic biliary decompression lowers operative morbidity in patients with obstructive jaundice. Am J Surg 1981; 141(1): 61–5. 49. Irvin TT, Vassilakis JS, Chattopadhyay DK, Greaney MG. Abdominal wound healing in jaundiced patients. Br J Surg 1978; 65(7): 521–2. 50. Greig JD, Krukowski ZH, Matheson NA. Surgical morbidity and mortality in one hundred and twenty-nine patients with obstructive jaundice. Br J Surg 1988; 75(3): 216–9. 51. Grande L, Garcia-Valdecasas JC, Fuster J, Visa J, Pera C. Obstructive jaundice and wound healing. Br J Surg 1990; 77(4): 440–2. 52. Pitt HA, Cameron JL, Postier RG, Gadacz TR. Factors affecting mortality in biliary tract surgery. Am J Surg 1981; 141(1): 66–72. 53. Rosen J, Young SC, Berman J, Magill T. Management of malignant obstructive jaundice. Journal of Surgical Oncology 1989; 40(4): 256–60. 54. van der Gaag NA, Rauws EAJ, van Eijck CHJ, et al. Preoperative Biliary Drainage for Cancer of the Head of the Pancreas. N Engl J Med 2010 January 14, 2010; 362(2): 129–37. 55. Martin Rn, Vitale G, Reed D, Larson G, Edwards M, McMasters K. Cost comparison of endoscopic stenting vs surgical treatment for unresectable cholangiocarcinoma. Surg Endosc 2002 Apr; 16(4): 667–70. 56. Soehendra N, Grimm H, Berger B, Nam VC. Malignant jaundice: Results of diagnostic and therapeutic endoscopy. World J Surg 1989; 13(2): 171–7. 57. Speer AG, Cotton PB, Russell RCG, et al. Randomised trial of endoscopic stenting versus percutaneous stent insertion in malignant obstructive jaundice. Lancet 1987 11 July; 1: 57–62. 58. Cowling MG, Adam AN. Internal stenting in malignant biliary obstruction. World J Surg 2001 March 2001; 25(3): 355–61. 59. Webster G, Pereira S. Mesh-metal stents for hilar cholangiocarcinoma. Gastrointest Endosco 2009; 70(4): 817–8. 60. Speer AG, Cotton PB, MacRae KD. Endoscopic management of malignant biliary obstruction: stents of 10 French guage are preferable to stents of 8 French gauge. Gastrointest Endosco 1988 Sep–Oct; 34(5): 412–7. 61. Lammer J, Hausegger KA, Fluckiger F, et al. Common bile duct obstruction due to malignancy: treatment with plastic versus metal stents. Radiology 1996; 201: 167–72. 62. Indar AA, Lobo DN, Gilliam AD, et al. Percutaneous biliary metal wall stenting in malignant obstructive jaundice. Euro J Gastroenterol Hepatol 2003; 15(8): 915–9. 63. Men S, Hekimoglu B, Kaderoglu H, et al. Palliation of malignant obstructive jaundice. Use of self-expandable metal stents. Acta Radiol 1996 May; 37(3 Pt 1): 259–66. 64. Rossi P, Bezzi M, Rossi M, et al. Metallic stents in malignant biliary obstruction: results of a multicentre European study of 240 patients. J Vasc Interv Radiol 1994 Mar–Apr; 5(2): 279–85. 65. Yoshioka T, Sakaguchi H, Yoshimura H, et al. Expandable metallic biliary endoprostheses: preliminary clinical evaluation. Radiology 1990 October 1, 1990; 177(1): 253–7. 66. Yeoh KG, Zimmerman MJ, Cunningham JT, Cotton PB. Comparative costs of metal versus plastic biliary stent strategies for malignant obstructive jaundice by decision analysis. Gastrointest Endosco 1999; 49(4): 466–71. 67. Soderlund C, Linder S. Covered metal versus plastic stents for malignant common bile duct stenosis: a prospective, randomized, controlled trial. Gastrointest Endosc 2006; 63(7): 986–95. 68. Arguedas MR, Heudebert GH, Stinnett AA, Wilcox CM. Biliary stents in malignant obstructive jaundice due to pancreatic carcinoma: a costeffectiveness analysis. Am J Gastroenterol 2002; 97(4): 898–904. 69. Kaassis M, Boyer J, Dumas R, et al. Plastic or metal stents for malignant stricture of the common bile duct? Results of a randomized prospective study. Gastrointest Endosc 2003; 57(2): 178–82. 70. Prat F, Chapat O, Ducot B, et al. A randomized trial of endoscopic drainage methods for inoperable malignant strictures of the common bile duct. Gastrointest Endosc 1998; 47(1): 1–7. 71. Wagner H, Knyrim K, Vakil N, Klose K. Plastic endoprostheses versus metal stents in the palliative treatment of malignant hilar biliary obstruction. A prospective and randomized trial. Endoscopy 1993; 25(3): 213–8. 72. Guo Y-X, Li Y-H, Chen Y, et al. Percutaneous transhepatic metal versus plastic biliary stent in treating malignant biliary obstruction: a multiple centre investigation. Hepatobiliary Pancreatic Dis Int 2003 November 2003; 2(4): 594–7. 73. Kim HS, Lee DK, Kim HG, et al. Features of malignant biliary obstruction affecting the patency of metallic stents: A multicenter study. Gastrointest Endosc 2002; 55(3): 359–65. 74. Mullen JT, Lee JH, Gomez HF, et al. Pancreaticoduodenectomy after placement of endobiliary metal stents. J Gastrointest Surg 2005; 9(8): 1094–105. 75. Hiroyuki Isayama YK, Takeshi T, Haruhiko Y, et al. Polyurethane-covered metal stent for management of distal malignant biliary disease. Gastrointest Endosc 2002; 55(3): 366–70. 76. Kahaleh M, Tokar J, Conaway MR, et al. Efficacy and complications of covered Wallstents in malignant distal biliary obstruction. Gastrointest Endosc 2005; 61(4): 528–33. 77. Yousuke Nakai HI, Yutaka K, Takeshi T, et al. Efficacy and safety of the covered Wallstent in patients with distal malignant biliary obstruction. Gastrointest Endosc 2005; 62(5): 742–8. 78. Isayama H, Komatsu Y, Tsujino T, et al. A prospective randomised study of “covered” versus “uncovered” diamond stents for the management of distal malignant biliary obstruction. Gut 2004 May 1, 2004; 53(5): 729–34. 79. Fumex F, Coumaros D, Napoleon B, Barthet M, Laugier R, Yzet T, et al. Similar performance but higher cholecystitis rate with covered biliary stents: results from a prospective multicenter evaluation. Endoscopy 2006; 38(08): 787–92.

352

ENDOSCOPIC MANAGEMENT OF MALIGNANT BILIARY OBSTRUCTION
80. Hochwald S, Burke E, Jarnagin W, Fong Y, Blumgart L. Association of preoperative biliary stenting with increased postoperative infectious complications in proximal cholangiocarcinoma. Arch Surg 1999 Mar; 134(3): 261–6. 81. Chang W-H, Kortan P, Haber GB. Outcome in patients with bifurcation tumors who undergo unilateral versus bilateral hepatic duct drainage. Gastrointest Endosc 1998; 47(5): 354–62. 82. Inal M, Akgul E, Aksungur E, Seydaoglu G. Percutaneous Placement of Biliary Metallic Stents in Patients with Malignant Hilar Obstruction: Unilobar versus Bilobar Drainage. J Vascul Interven Radiol 2003 November 2003; 14: 1409–16. 83. DePalma G, Pezzullo A, Rega M, et al. Unilateral placement of metallic stents for malignany hilar obstruction: a prospective study. Gastrointest Endosc 2003 Jul; 58(1): 50–3. 84. De Palma GD, Galloro G, Siciliano S, Iovino P, Catanzano C. Unilateral versus bilateral endoscopic hepatic duct drainage in patients with malignant hilar biliary obstruction: Results of a prospective, randomized, and controlled study. Gastrointest Endosc 2001; 53(6): 547–53. 85. Liu C-L, Lo C-M, Lai ECS, Fan S-T. Endoscopic Retrograde Cholangiopancreatography and endoscopic endoprosthesis insertion in patients with Klatskin tumors. Arch Surg 1998 March 1, 1998; 133(3): 293–6. 86. Bowling TE, Galbraith SM, Hatfield AR, Solano J, Spittle MF. A retrospective comparison of endoscopic stenting alone with stenting and radiotherapy in non-resectable cholangiocarcinoma. Gut 1996 December 1, 1996; 39(6): 852–5. 87. Shinchi H, Takao S, Nishida H, Aikou T. Length and quality of survival following external beam radiotherapy combined with expandable metallic stent for unresectable hilar cholangiocarcinoma. J Surg Oncol 2000; 75(2): 89–94. 88. Shaib YH, Davila JA, Henderson L, McGlynn KA, El-Serag HB. Endoscopic and surgical therapy for intrahepatic cholangiocarcinoma in the United States: a population based study. J Clin Gastroenterol 2007 November/ December; 41(10): 911–7. 89. Harewood GC, Baron TH, Rumalla A, Wang KK, Gores GJ, Stadheim LM, et al. Pilot study to assess patient outcomes following endoscopic application of photodynamic therapy for advanced cholangiocarcinoma. J Gastroenterol Hepatol 2005; 20(3): 415–20. 90. McCaughan JS Jr., Mertens BF, Cho C, Barabash RD, Payton HW. Photodynamic therapy to treat tumors of the extrahepatic biliary ducts. A case report. Arch Surg 1991 January 1, 1991; 126(1): 111–3. 91. Ortner M-AEJ, Liebetruth J, Schreiber S, et al. Photodynamic therapy of nonresectable cholangiocarcinoma. Gastroenterology 1998; 114(3): 536–42. 92. Rumalla A, Baron TH, Wang KK, et al. Endoscopic application of photodynamic therapy for cholangiocarcinoma. Gastrointest Endosc 2001; 53(4): 500–4. 93. Berr F, Wiedmann M, Tannapfel A, et al. Photodynamic therapy for advanced bile duct cancer: Evidence for improved palliation and extended survival. Hepatology 2000; 31(2): 291–8. 94. Ortner MEJ, Caca K, Berr F, et al. Successful photodynamic therapy for nonresectable cholangiocarcinoma: a randomized prospective study. Gastroenterology 2003; 125(5): 1355–63. 95. Zoepf T, Jakobs R, Arnold JC, Apel D, Riemann JF. Palliation of nonresectable bile duct cancer: improved survival after photodynamic therapy. Am J Gastroenterol 2005; 100(11): 2426–30. 96. Witzigmann H, Berr F, Ringel U, et al. Surgical and palliative management and outcome in 184 patients with hilar cholangiocarcinoma: palliative photodynamic therapy plus stenting is comparable to R1/R2 resection. Ann Surg 2006 August 2006; 244(2): 230–9.

353

39

Choledochal cyst detected in adulthood Bilal Al-Sarireh and Hassan Malik
type III choledochocele (dilatation of intra-duodenal CBD); type IVa intra- and extra-hepatic bile duct dilatation; type IVb multiple cysts (extra-hepatic only), and type V intra-hepatic cysts only (Fig. 39.1). Initially, these different types thought to represent a spectrum of the same disease. However, subsequently several authors believed that each type represents a unique disease with separate etiologies, natural course, and ideal treatment (23,24). The most common cyst is type I (79%), followed by type IV (13%), type III (4%), type II (2.6%), and type V (<1%) (24–26).

incidence and pathophysiology
Choledochal cysts are cystic dilatation of the biliary tree, which present as isolated or combined dilatation of both the extraand the intra-hepatic biliary tree. Its incidence, although as high as 1:1000 in Asian population, is only 1:100,000 to 1:150,000 in Western population, with a female preponderance of up to 4:1 (1,2). Nearly 60% of choledochal cysts are detected before the age of 10, but they are increasingly diagnosed in adult population and in nearly 20% the diagnosis is delayed until adulthood when presentation is different compared to childhood (3–5).

etiology and classification
The etiology of choledochal cysts remains speculative. It is largely believed that the majority of choledochal cysts are congenital lesions (6). The most widely accepted theory is the presence of anomalous pancreaticobiliary ductal confluence (APBDC), which is frequently associated with choledochal cysts (7,8). It is hypothesized that this anomaly predisposes to reflux of pancreatic secretion up to the biliary tree leading to mucosal break down and dilatation (9). The observation that biliary cysts are more common in the extra-hepatic biliary tree supports this hypothesis. Also, this hypothesis has been supported by several authors who reported the presence of a common pancreaticobiliary channel in as much as 70% of choledochal cysts (8,10,11). In such cases the biliary amylase is usually elevated (12). However, experimental model failed to confirm this finding (13). This theory also seems to be rejected by the reports of antenatal choledochal cysts at an age before the exocrine function of the pancreas has even begun (14). Therefore, it is unlikely that a common channel is the sole explanation for choledochal cysts as there are many of these lesions in which it is not present. Other authors contend that these cysts are congenital in nature, due to either distal aganglionosis and proximal dilatation or aberrancies in embryologic recanalization (15,16). Such theories are largely speculative, but are supported by antenatal observational studies and experimental studies on animals (17,18). More over, viral infection has been proposed as a possible cause for some of these lesions as has been demonstrated by Petersen et al. in his experiment on mice (19). Since cyst location determines the type of treatment, most authors opt for the 1997 classification of Tandoni et al. (20). Choledochal cysts were first described by Vater and Ezler in 1723 (21). However, it was not until 1959 when Alonzo-lej et al. gave a thorough description of the disease and classified choledochal cysts into three types (22). This classification was then modified by Tandoni et al. (20), who described eight different types: type 1 extra-hepatic bile duct cystic dilatation: (Ia) common type, (Ib) segmental dilatation, and (Ic) diffuse dilatation; type II common bile duct (CBD) diverticulum;

presentation
The clinical triad of abdominal pain, mass, and jaundice is seldom seen in adults (3,25). Clinical symptoms in most cases are intermittent and nonspecific resulting in delayed diagnosis. Also the majority of these patients do not have consistent abnormalities in liver function tests (LFTs) (27). Abdominal pain and recurrent cholangitis are the most common presentation (3,28–30). The abdominal pain usually mimics that of calculus cholecystitis and many patients do have gallstones either in the cyst and/or in the gallbladder. This might lead to misdiagnosis of calculus cholecystitis and or choledocholithiasis with biliary ductal dilatation. Subsequently, a number of these patients who genuinely have cystic dilatation of the biliary tree end up treated inadequately by simple cholecystectomy and or common bile duct (CBD) exploration or occasionally treated by biliary enteric bypass (31). In areas where incidence of choledochal cysts is relatively high, it is prudent to consider choledochal cyst in the differential diagnosis in patients who present with biliary colic, recurrent cholangitis, and evidence of dilated biliary tree (32,33). However, in areas where incidence of choledochal cysts is rare, this might represent a real-diagnostic dilemma requiring careful management of these patients at a specialist unit to ensure correct and adequate treatment tailored to each individual. Other presentations include those related to complications that might develop secondary to cystic dilatation of the biliary tree such as calculi, recurrent hepatic abscesses, recurrent pancreatitis, bleeding, cyst erosion into surrounding structure, cirrhosis, portal hypertension, and cholangiocarcinoma (6,34). Such presentations may obscure the primary problem and may increase the complexities of treatment (32,35). Indeed, studies in adults have shown that nearly 80% of them present with one or more of these conditions (5,6,27).

diagnosis
Correct diagnosis of true cystic dilatation of the biliary tree in adults depends mainly on the use of imaging studies and some times can be quite challenging. In indeterminate cases, careful evaluation for anomalous pancreaticobiliary ductal confluence may be useful.

354

CHOLEDOCHAL CYST DETECTED IN ADULTHOOD
Trans-abdominal ultrasonography is an extremely useful investigation which provides adequate information about the intra- and extra-hepatic biliary tree. However, anatomic delineation of the non-dilated biliary system can be quite limited (36,37). Computed tomography (CT) imaging provides adequate information about the size, extent, and characteristics of choledochal cysts as well as about its relationship to adjacent structures. More over, CT may reveal associated pathology within the parenchyma of related organs. CT scan is also
I II III

IVa

IVb

V

Figure 39.1 Tandoni classification of choledochal cysts: Type 1 extra-hepatic bile duct cystic dilatation, Type II common bile duct diverticulum, Type III choledochocele, Type IVa intra- and extra-hepatic bile duct dilatation, Type IVb multiple cysts (extra-hepatic only), and Type V intra-hepatic cysts only.

considered to be more accurate than ultrasonography in demonstrating the biliary tree anatomy especially when using the most recent spiral CT cholangiography technology which allows three-dimensional reconstructions (38,39). CT scan and ultrasonography can be sometimes of limited value in demonstrating the biliary origin of choledochal cyst. However, hepatobiliary scintigraphy with technetium-99 labeled iminodiacetic acid (IDA) derivatives will often show the cyst and in particular helpful at establishing that the cystic structure is an intrinsic part o the biliary tree. IDA scanning is also helpful at assessing hepatobiliary function and anastomotic patency postoperatively (33,36,40). Magnetic resonance imaging (MRI) is non-invasive and produces clear images of the biliary tree. Also, it may even demonstrate APBDC similar to endoscopic retrograde cholangiopancreaticograpgy (ERCP), but without the complications of ERCP (3,41–43). Several studies reported that MRI has an estimated diagnostic accuracy of 82% to 100% (43,44). Hence, MRI is widely held as the imaging modality of choice for choledochal cysts (Figs. 39.2A and 39.3A). ERCP defines the anatomy of the biliary tree accurately and reveals the presence of any associated intra-ductal pathology or an APBDC (Figs. 39.2B and 39.3B) (3,41). Also, ERCP allows dealing with coexisting ductal stones in patients present with jaundice and in the rare instance of type III cyst; it facilitates a therapeutic papillotomy simultaneously (45). However, this is invasive investigation with a significant complication rate and although it provides excellent anatomic delineation, it rarely produces information which alters the subsequent management of the choledochal cysts. More over cholangiography can always be performed intra-operatively by injecting contrast directly into the bile duct to display the ductal anatomy (6). Endoscopic ultrasound scan and cholangioscopy are relatively more recent investigations that can be of great use in the diagnosis of extra-hepatic biliary cystic dilatation and more important in the diagnosis of early malignant changes within the cyst wall where biopsies can also be taken to confirm abnormal tissue findings (46).

(A)

(B)

Figure 39.2 (A) MRCP image demonstrating segmental dilatation of the intra-hepatic biliary tree in the right lobe of the liver (Type V). (B) ERCP image demonstrating segmental dilatation of the intra-hepatic biliary tree in the right lobe of the liver (Type V).

355

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

complications
The most common complication associated with choledochal cysts is stones in the gallbladder, within the cysts or in the pancreatic duct (6). The calculi in the cysts are of the type seen in biliary stasis (47). In choledochal cysts the calculi can be mistaken for choledocholithiasis in a dilated bile duct secondary to obstruction. The main distinguishing features are nondilated proximal and intra-hepatic biliary radicles in the case of type I cysts and non-dilated intervening ducts in the case of type IV cysts. Biliary strictures are another recognized complication associated with adult choledochal cysts. This is most often related to earlier surgical intervention but also can occur secondary to chronic inflammation or rarely related to congenital etiology (47,48). Cirrhosis and liver abscesses are often associated with longstanding biliary obstruction and recurrent cholangitis (48,49). Some patients develop portal hypertension due to compression of the portal vein by a large choledochal cyst or more often secondary to biliary cirrhosis (50). Pancreatitis is a well-recognized complication related to choledochal cysts. Incidence (10–70%) has been reported in the literature by several authors (3,51,52). The pathophysiology of pancreatitis in these patients is not well understood but most authors believe that it is due to activation of pancreatic enzymes associated with biliary reflux especially in patients with APBDC (51,53). Choledochal cysts are associated with an increased risk of developing cholangiocarcinoma and gallbladder tumors (54–56). The risk is about 20 times higher than that in the general population ranging from 2.5% to 30% (3,56–58). The risk is age related: 0.7% in children under the age of 10, 6.8% in patients 11 to 20 years of age, and 14.3% in patients over the age of 20 years (57). Carcinoma occurs not only in the cyst wall but also in the remainder of the hepatobiliary and pancreatic tree (57,59). Therefore, long-term follow-up is indicated in all

patients with choledochal cysts even after complete excision of the cysts. The pathogenesis of malignant change in biliary cysts is unknown. It has been thought to be related to bile stasis and/or reflux of pancreatic juice, which give chronic irritation and metaplasia (59).

treatment
Management of choledochal cysts in adults should be performed at specialist unit and tailored according to each patient’s circumstances. Cyst type and more important patient’s general health, presenting symptoms and risk of developing complications related to the choledochal cysts are major factors that should be considered when deciding upon which type of treatment to be adapted. Total cyst excision when possible remains the gold standard treatment for choledochal cysts. This has been recommended by many investigators because of lower incidence of both early and late postoperative complications. Therefore, this approach has gained popularity worldwide (10,27,33,60,61). In choledochal cysts types I and IVb involve complete excision of the bile duct from the confluence of the hepatic ducts proximally up to the pancreaticobiliary junction distally. Cholecystectomy at the same time is a necessity, as the gallbladder usually arises from or adheres to the cyst wall and may contain stones or occasionally even tumor. Reconstruction of the biliary-enteric communication in these patients is usually achieved by Roux-en-Y hepaticojejunostomy or less commonly by hepaticoduodenectomy. For type IVa choledochal cyst, the extrahepatic component should be treated in the same way as for types I and IV cysts. The management of the intra-hepatic component depends on the severity of the disease and extent of involvement. This could range from percutaneous drainage through stricture dilatation and stone removal to partial hepatectomy (Fig. 39.4) or liver transplantation (3). More radical surgery is required if malignancy is confirmed on preoperative or fresh frozen biopsies. This might involve

(A)

(B)

Figure 39.3 (A) MRCP image demonstrating intra- and extra-hepatic bile duct dilatation (Type IVa). (B) ERCP image demonstrating intra- and extra-hepatic bile duct dilatation (Type IVa).

356

CHOLEDOCHAL CYST DETECTED IN ADULTHOOD
either pancreaticoduodenectomy or some form of hepatectomy according to the location of the tumor. Thus, the probable extent of surgery should be discussed with these patients beforehand and the use of fresh frozen biopsies in this situation could not be overemphasized bearing in mind its limitations. Removal of the entire cyst wall is considered the most effective prophylaxis against developing malignancy and other associated biliary tract pathology (33,61,62). Therefore, the strategy of complete excision of the choledochal cyst should be strictly adhered to whenever it is possible to do so (3,27,32). In situations where the intensity of fibrosis precludes safe dissection to allow total cyst excision, it is advisable to follow the approach described by lilly et al. (63). In this technique, the most densely adherent portion of the cyst wall is retained on the hepatoduodenal ligament, removing only the less adherent portion. The mucosal lining of the retained cyst wall should be ablated by diathermy to minimize the risk of malignant transformation in that segment of the cyst. Simple cystenterostomy, which was used to be commonly performed in the past, should no longer be considered a therapeutic option due to a high incidence (30–50%) of late complications such as cholangitis, anastomotic stricture, stone formation, biliary cirrhosis, and, worst of all, malignancy (5,33,56,60–62,64). Therefore, these patients with previous inadequate choledochal cyst excision should, if appropriate, undergo further revision surgery to reduce recurrent symptoms and the risk of developing malignancy. Treatment for type V cysts (Caroli’s disease) is still controversial. Hepatic resection is safe and effective for part of type V cysts (Fig. 39.5A,B) (10,65). For the type V cysts with diffuse intra-hepatic cholangiectasia and frequent recurrent cholangitis resulting in hepatic cirrhosis, liver resection is rarely feasible in which cases liver transplantation may be the only effective treatment (66). Type II and III cysts are rare lesions. Type II cysts may be treated by cyst excision and primary ductal reconstruction or reconstruction using Roux-en-Y hepaticojejunostomy while Type III cysts, also known as choledochocele, may be excised trans-duodenal following which both biliary and pancreatic ducts are anastomosed to duodenum (64). Small lesions (type III cysts) may be treated by sphincteroplasty or possibly by endoscopic sphincterotomy (67,68). All patients with choledochal cysts including those who had complete choledochal cyst excision and more important those

Figure 39.4 Intra-operative view following surgical resection of type IVa choledochal cyst seen in Figure 3A,B. Surgery involved left hepatectomy, total excision of the bile duct and cholecystectomy.

(A)

(B)

Figure 39.5 (A) Demarcation of right anterior sector (segments 5 and 8) following extra-hepatic division of the right anterior sectoral hepatic artery and portal vein. (B) Intra-operative view following segmental (central) liver resection (segments 5 and 8) for localized intra-hepatic segmental dilatation of the biliary tree seen in Figure 2A,B.

357

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
who had previous cyst drainage procedures should be kept under surveillance. Patients who had complete choledochal cyst excision are prone to develop biliary-enteric anastomotic stricture which can lead to cholangitis, jaundice, and biliary cirrhosis. Also, these patients are still at a slightly higher risk than the general population of developing malignancy in the rest of the biliary tree even after complete excision of the choledochal cysts (69).
4. Suita S, Shono K, Kinugasa Y, Kubota M, Matsuo S. Influence of age on the presentation and outcome of choledochal cyst. J Pediatr Surg 1999; 34(12): 1765–8. 5. Flanigan PD. Biliary cysts. Ann Surg 1975; 182(5): 635–43. 6. Yamaguchi M. Congenital choledochal cyst. Analysis of 1,433 patients in the Japanese literature. Am J Surg 1980; 140(5): 653–7. 7. Rattner DW, Schapiro RH, Warshaw AL. Abnormalities of the pancreatic and biliary ducts in adult patients with choledochal cysts. Arch Surg 1983; 118(9): 1068–73. 8. Wiedmeyer DA, Stewart ET, Dodds WJ, et al. Choledochal cyst: findings on cholangiopancreatography with emphasis on ectasia of the common channel. AJR Am J Roentgenol 1989; 153(5): 969–72. 9. Babbitt DP. Congenital choledochal cysts: new etiological concept based on anomalous relationships of the common bile duct and pancreatic bulb. Ann Radiol (Paris) 1969; 12(3): 231–40. 10. Liu YB, Wang JW, Devkota KR, et al. Congenital choledochal cysts in adults: twenty-five-year experience. Chin Med J (Engl) 2007; 120(16): 1404–7. 11. Gauthier F, Brunelle F, Valayer J. Common channel for bile and pancreatic ducts. Presentation of 12 cases and discussion. Chir Pediatr 1986; 27(3): 148–52. 12. Davenport M, Stringer MD, Howard ER. Biliary amylase and congenital choledochal dilatation. J Pediatr Surg 1995; 30(3): 474–7. 13. Benhidjeb T, Said S, Rudolph B, Siegmund E. Anomalous pancreaticobiliary junction--report of a new experimental model and review of the literature. J Pediatr Surg 1996; 31(12): 1670–4. 14. Schroeder D, Smith L, Prain HC. Antenatal diagnosis of choledochal cyst at 15 weeks’ gestation: etiologic implications and management. J Pediatr Surg 1989; 24(9): 936–8. 15. Schweizer P. Pathogenesis of choledochal cyst. Pediatr Surg Int 1995; 10(7): 475–7. 16. Davenport M, Basu R. Under pressure: choledochal malformation manometry. J Pediatr Surg 2005; 40(2): 331–5. 17. Wong KC, Lister J. Human fetal development of the hepato-pancreatic duct junction--a possible explanation of congenital dilatation of the biliary tract. J Pediatr Surg 1981; 16(2): 139–45. 18. Spitz L. Experimental production of cystic dilatation of the common bile duct in neonatal lambs. J Pediatr Surg 1977; 12(1): 39–42. 19. Petersen C, Biermanns D, Kuske M, et al. New aspects in a murine model for extrahepatic biliary atresia. J Pediatr Surg 1997; 32(8): 1190–5. 20. Todani T, Watanabe Y, Narusue M, Tabuchi K, Okajima K. Congenital bile duct cysts: classification, operative procedures, and review of thirty-seven cases including cancer arising from choledochal cyst. Am J Surg 1977; 134(2): 263–9. 21. Vater a, Ezler C. Dissertatio de scirrhis viserum occasione sections viri typanite. wittenburgae 1723; Pamphlers 88: 22. 22. Alonso-Lej F, Rever WB, Jr, Pessagno DJ. Congenital choledochal cyst, with a report of 2, and an analysis of 94, cases. Int Abstr Surg 1959; 108(1): 1–30. 23. Visser BC, Suh I, Way LW, Kang SM. Congenital choledochal cysts in adults. Arch Surg 2004; 139(8): 855–60; discussion 860–2. 24. Wiseman K, Buczkowski AK, Chung SW, et al. Epidemiology, presentation, diagnosis, and outcomes of choledochal cysts in adults in an urban environment. Am J Surg 2005; 189(5): 527–31; discussion 531. 25. Singham J, Schaeffer D, Yoshida E, Scudamore C. Choledochal cysts: analysis of disease pattern and optimal treatment in adult and paediatric patients. HPB (Oxford) 2007; 9(5): 383–7. 26. Rha SY, Stovroff MC, Glick PL, Allen JE, Ricketts RR. Choledochal cysts: a ten year experience. Am Surg 1996; 62(1): 30–4. 27. Liu CL, Fan ST, Lo CM, et al. Choledochal cysts in adults. Arch Surg 2002; 137(4): 465–8. 28. Sela-Herman S, Scharschmidt BF. Choledochal cyst, a disease for all ages. Lancet 1996; 347(9004): 779. 29. Soreide K, Korner H, Havnen J, Soreide JA. Bile duct cysts in adults. Br J Surg 2004; 91(12): 1538–48. 30. de Vries JS, de Vries S, Aronson DC et al. Choledochal cysts: age of presentation, symptoms, and late complications related to Todani’s classification. J Pediatr Surg 2002; 37(11): 1568–73. 31. Banerjee Jesudason SR, Ranjan Jesudason M, Paul Mukha R, et al. Management of adult choledochal cysts – a 15-year experience. HPB (Oxford) 2006; 8(4): 299–305.

summary
Diagnosis and treatment of adult choledochal cysts can be quite challenging. Therefore, management of these patients should be carried out at specialist units. Choledochal cysts rarely present for the first time in adulthood. However when they do, the presentation is variable and most often due to presence of associated disease. Also they can be totally asymptomatic. The appropriate treatment strategy should be selected based on the type of the cyst and more important on the individual’s general health and presenting symptoms. Total cyst excision and Roux-en-Y hepaticojejunostomy are the current recommended surgical options world wide for type I and IV choledochal cysts while for type V cyst (diffuse Caroli’s disease) with frequent recurrent cholangitis, liver transplantation should be considered. Patients with previous cyst drainage procedures should undergo revisional surgery if appropriate to reduce recurrent symptoms and the risk of developing malignancy. Conservative treatment might be the most appropriate strategy in asymptomatic individuals with poor general health and significant co-morbidities. All patients with choledochal cysts including those who had complete choledochal cyst excision should be kept under long-term surveillance.

key points
Choledochal cysts are increasingly diagnosed in adult population Etiology of choledochal cysts remains speculative Adults with choledochal cysts most commonly present with complications related to their choledochal cysts Choledochal cysts are associated with an increased risk of developing cholangiocarcinoma and gallbladder tumors MRI is widely held as the imaging modality of choice for choledochal cysts Patients with choledochal cysts should be managed at a specialist unit to ensure correct and adequate treatment tailored to each individual Total cyst excision when possible remains the gold standard treatment for choledochal cysts

references
1. Gigot JF, Nagorney D, Farnell M, Moir C, Ilstrup D. Bile duct cysts: a changing spectrum of disease. J Hepato-Biliary-Pancreatic Surg 1996; 3(4): 405–11. 2. Lenriot JP, Gigot JF, Segol P, et al. Bile duct cysts in adults: a multi-institutional retrospective study. French Associations for Surgical Research. Ann Surg 1998; 228(2): 159–66. 3. Lipsett PA, Pitt HA, Colombani PM, Boitnott JK, Cameron JL. Choledochal cyst disease. A changing pattern of presentation. Ann Surg 1994; 220(5): 644–52.

358

CHOLEDOCHAL CYST DETECTED IN ADULTHOOD
32. Hewitt PM, Krige JE, Bornman PC, Terblanche J. Choledochal cysts in adults. Br J Surg 1995; 82(3): 382–5. 33. Jesudason SR, Govil S, Mathai V, Kuruvilla R, Muthusami JC. Choledochal cysts in adults. Ann R Coll Surg Engl 1997; 79(6): 410–3. 34. Fieber SS, Nance FC. Choledochal cyst and neoplasm: a comprehensive review of 106 cases and presentation of two original cases. Am Surg 1997; 63(11): 982–7. 35. Chaudhary A, Dhar P, Sachdev A. Reoperative surgery for choledochal cysts. Br J Surg 1997; 84(6): 781–4. 36. Han BK, Babcock DS, Gelfand MH. Choledochal cyst with bile duct dilatation: sonography and 99mTc IDA cholescintigraphy. AJR Am J Roentgenol 1981; 136(6): 1075–9. 37. Young W, Blane C, White SJ, Polley TZ. Congenital biliary dilatation: a spectrum of disease detailed by ultrasound. Br J Radiol 1990; 63(749): 333–6. 38. Katyal D, Lees GM. Choledochal cysts: a retrospective review of 28 patients and a review of the literature. Can J Surg 1992; 35(6): 584–8. 39. Groebli Y, Sarraj A, Pfister L, Lopez J. Spiral-CT cholangiography with 3D reconstruction in the diagnosis of choledochocele. Eur Radiol 2000; 10(2): 395. 40. Grossman SA, Patel BA, Blend MJ. Use of CCK cholescintigraphy to differentiate choledochal cyst from gallbladder. Clin Nucl Med 1991; 16(4): 226–9. 41. Komi N, Takehara H, Kunitomo K, Miyoshi Y, Yagi T. Does the type of anomalous arrangement of pancreaticobiliary ducts influence the surgery and prognosis of choledochal cyst? J Pediatr Surg 1992; 27(6): 728–31. 42. Yamataka A, Ohshiro K, Okada Y, et al. Complications after cyst excision with hepaticoenterostomy for choledochal cysts and their surgical management in children versus adults. J Pediatr Surg 1997; 32(7): 1097–102. 43. Kim SH, Lim JH, Yoon HK, et al. Choledochal cyst: comparison of MR and conventional cholangiography. Clin Radiol 2000; 55(5): 378–83. 44. Dinsmore JE, Murphy JJ, Jamieson D. Pediatric surgical images: MRCP evaluation of choledochal cysts. J Pediatr Surg 2001; 36(5): 829–30. 45. Martin RF, Biber BP, Bosco JJ, Howell DA. Symptomatic choledochoceles in adults. Endoscopic retrograde cholangiopancreatography recognition and management. Arch Surg 1992; 127(5): 536–8; discussion 538–9. 46. Judah JR, Draganov PV. Intraductal biliary and pancreatic endoscopy: an expanding scope of possibility. World J Gastroenterol 2008; 14(20): 3129–36. 47. Uno K, Tsuchida Y, Kawarasaki H, Ohmiya H, Honna T. Development of intrahepatic cholelithiasis long after primary excision of choledochal cysts. J Am Coll Surg 1996; 183(6): 583–8. 48. Simici P, Ratiu O. Cystic dilatation of the common bile duct with cholangitis, jaundice and cholestatic cirrhosis. Rev Chir Oncol Radiol O R L Oftalmol Stomatol Chir 1980; 29(3): 201–6. 49. Nagorney DM, McIlrath DC, Adson MA. Choledochal cysts in adults: clinical management. Surgery 1984; 96(4): 656–63. 50. Nunez-Hoyo M, Lees CD, Hermann RE. Bile duct cysts. Experience with 15 patients. Am J Surg 1982; 144(3): 295–9. 51. Okada A, Nakamura T, Higaki J, et al. Congenital dilatation of the bile duct in 100 instances and its relationship with anomalous junction. Surg Gynecol Obstet 1990; 171(4): 291–8. 52. Swisher SG, Cates JA, Hunt KK, et al. Pancreatitis associated with adult choledochal cysts. Pancreas 1994; 9(5): 633–7. 53. Todani T, Urushihara N, Watanabe Y, et al. Pseudopancreatitis in choledochal cyst in children: intraoperative study of amylase levels in the serum. J Pediatr Surg 1990; 25(3): 303–6. 54. Todani T, Tabuchi K, Watanabe Y, Kobayashi T. Carcinoma arising in the wall of congenital bile duct cysts. Cancer 1979; 44(3): 1134–41. 55. Tsuchiya R, Harada N, Ito T, Furukawa M, Yoshihiro I. Malignant tumors in choledochal cysts. Ann Surg 1977; 186(1): 22–8. 56. Jan YY, Chen HM, Chen MF. Malignancy in choledochal cysts. Hepatogastroenterology 2000; 47(32): 337–40. 57. Voyles CR, Smadja C, Shands WC, Blumgart LH. Carcinoma in choledochal cysts. Age-related incidence. Arch Surg 1983; 118(8): 986–8. 58. Shi LB, Peng SY, Meng XK, et al. Diagnosis and treatment of congenital choledochal cyst: 20 years’ experience in China. World J Gastroenterol 2001; 7(5): 732–4. 59. Kimura K, Ohto M, Saisho H, et al. Association of gallbladder carcinoma and anomalous pancreaticobiliary ductal union. Gastroenterology 1985; 89(6): 1258–65. 60. Rush E, Podesta L, Norris M, et al. Late surgical complications of choledochal cystoenterostomy. Am Surg 1994; 60(8): 620–4. 61. Lee KF, Lai EC, Lai PB. Adult choledochal cyst. Asian J Surg 2005; 28(1): 29–33. 62. Tan SS, Tan NC, Ibrahim S, Tay KH. Management of adult choledochal cyst. Singapore Med J 2007; 48(6): 524–7. 63. Lilly JR. The surgical treatment of choledochal cyst. Surg Gynecol Obstet 1979; 149(1): 36–42. 64. Powell CS, Sawyers JL, Reynolds VH. Management of adult choledochal cysts. Ann Surg 1981; 193(5): 666–76. 65. Madariaga JR, Iwatsuki S, Starzl TE, et al. Hepatic resection for cystic lesions of the liver. Ann Surg 1993; 218(5): 610–4. 66. Sans M, Rimola A, Navasa M, et al. Liver transplantation in patients with Caroli’s disease and recurrent cholangitis. Transpl Int 1997; 10(3): 241–4. 67. Zheng LX, Jia HB, Wu DQ, et al. Experience of congenital choledochal cyst in adults: treatment, surgical procedures and clinical outcome in the Second Affiliated Hospital of Harbin Medical University. J Korean Med Sci 2004; 19(6): 842–7. 68. Akkiz H, Colakoglu SO, Ergun Y, et al. Endoscopic retrograde cholangiopancreatography in the diagnosis and management of choledochal cysts. HPB Surg 1997; 10(4): 211–8; discussion 218–9. 69. Watanabe Y, Toki A, Todani T. Bile duct cancer developed after cyst excision for choledochal cyst. J Hepatobiliary Pancreat Surg 1999; 6(3): 207–12.

359

40 Bile duct injuries and benign biliary strictures
Steven M. Strasberg
Benign biliary strictures are largely due to trauma or inflammation. Most traumatic biliary strictures result from iatrogenic operative injuries. In addition to causing strictures iatrogenic operative injuries may lead to other types of biliary abnormalities such as loss of continuity of the biliary tract and fistula. Inflammatory strictures are due to pancreatitis and less commonly sclerosing cholangitis, septic cholangitis, and inflammatory pseudotumors. Today it is usually possible to treat most inflammatory strictures by endoscopic stenting. Consequently the focus of this chapter will be on iatrogenic biliary injury. period when all cholecystectomies were performed by laparotomy to 0.47% in the latest period of the study 1996 to 2001 when most procedures were laparoscopic (9). To summarize it is likely that the incidence of biliary injury during laparoscopic cholecystectomy is higher than it was when cholecystectomies were performed by laparotomy but how much higher is uncertain. Clearly, ongoing good population studies defining the incidence of this problem are needed.

classification of biliary injuries
The Bismuth classification was the standard means of classifying biliary injuries in the era of open cholecystectomy (10). It classified strictures into five types based mainly on the upper level of injury, i.e., the lowest level at which healthy biliary mucosa is available for anastomosis (11). Several new classifications of laparoscopic biliary injury were proposed in the 1990s, including a classification by the Amsterdam group (12), a classification by Stewart and Way (13), and a classification introduced by our group in 1995 (14). The former two classifications divide injuries mainly on the basis of mechanism of injury and place strictures and complete transections in different categories. Our classification is a more detailed anatomic classification based on the level of the injury and essentially is an extension of the Bismuth classification applicable to injuries observed in the laparoscopic era (Fig. 40.1). For instance, Type A–D injuries are much more common in the laparoscopic era and need to be represented in such a classification (Fig. 40.2). Like that classification it is intended to help the surgeon or endoscopist choose the appropriate technique for the repair. Strictures and transections are in the same category and are separated within each category. Each classification has its advantages and disadvantages. Our classification is best suited for studies of surgical treatment of biliary injuries. Biliary injuries are often accompanied by vascular injuries. Vascular injuries are associated with a greater tendency for restricture of bile duct repairs (15), but apparently not when repairs are done after an interval in expert centers using the Hepp–Couinaud approach (16). Portal vein transection and traumatic thrombosis have also been reported. The vascular component may be come the predominant feature of the injury with necrosis of the intrahepatic biliary system or hepatic infarction. Infarction of the intrahepatic biliary tree requires transplantation, while hepatic infarction may lead to the need for hepatic resection or transplantation (17).

iatrogenic biliary injuries
Most traumatic injuries are due to operative trauma and more than 95% of biliary injuries occur during cholecystectomy. Biliary injury is the most severe common complication of cholecystectomy. The causes of injury are increasingly better understood and there have been improvements in strategies for preventing injury.

incidence of biliary injuries
Laparoscopic Versus Open Cholecystectomy in Randomized Controlled Trials A recent systematic review of randomized trials (evidence level 1a) of open versus laparoscopic surgery concluded that “laparoscopic cholecystectomy did not carry more bile duct injuries than open cholecystectomy” (1). Thirty trials randomizing 1914 patients were reviewed. The incidence, 0.2%, was the same for both operations. However, there were only four “high-quality” trials. It is questionable whether an analysis based on many small randomized trials of variable quality is capable of determining the true incidence of biliary injury during laparoscopic cholecystectomy or whether it is higher than in open cholecystectomy. Population-Based Studies Most registry results were published before 1998 (2–7) and may not reflect current results a decade later. They cite an incidence of major bile duct injury ranging from 0.15 (3) to 0.86 (4). Adamsen et al. described results in a rigorously maintained registry involving over 7000 Danish patients from 1991 to 1997. The major bile duct injury rate was 0.74% and the total injury rate including bile leaks was 2.8%. In a tightly controlled registry in the Department of Defense Hospitals in the United States from 1990 to 1992, the incidence was 0.57% (7). Flum et al., using an administrative database covering 1.5 million Medicare patients in the United States (population over 65 years) from 1992 to 1999 found a major bile duct injury rate of 0.5% (8). In a Swedish registry covering 150,000 cholecystectomies over the period from 1987 to 2001, the major bile duct injury rate increased from 0.40% in the earliest

pathogenesis of bile duct injuries
Patient-Related Factors Inflammation Acute cholecystitis: In population studies the incidence of biliary injury is higher when laparoscopic cholecystectomy is performed for acute cholecystitis than for elective indications (2,3),

360

BILE DUCT INJURIES AND BENIGN BILIARY STRICTURES

Figure 40.1 A classification of laparoscopic injuries to the biliary tract. The injuries Type A to E are illustrated. Type A injuries originate from small bile ducts in the liver bed that or from the cystic duct. Type B and C injuries almost always involve aberrant right hepatic ducts. The notations >2 cm and <2 cm in Type E1 and Type E2 indicate the length of common hepatic duct remaining. The corresponding Bismuth classification numbers are given when possible as B1-5. Note that Types A and D do not exist in the Bismuth classification, which is a classification of bile duct strictures rather than injuries. Note that B,C and E5 injuries would be classified as B5. They are separated in our classification because of the increase incidence of B and C injuries in the laparoscopic era.

Figure 40.2 Type A (left) and Type D injuries (right) are much more common in the laparoscopic era and were given separate categories in our classification. The Type A injury figure demonstrates extravasation of contrast from the liver bed of the gallbladder due to injury to a bile duct in that location. The Type D injury figure demonstrates an injury which was incompletely repaired over a t-tube (arrow). There is extravasation of contrast from the common bile duct when T tube is injected. A percutaneous drain (arrowhead) was inserted to drain postoperative bilomas. The injury healed without further treatment.

although this was not borne out in a Cochrane review of randomized trials (18) (evidence level 1a). However, in the latter, conclusions were drawn from a total of four bile duct injuries among 438 patients (0.9%) (18). Acute inflammation causes thickening of tissues, increases friability and vascularity, and promotes adhesions. These factors obscure normal anatomical relationships (Fig. 40.3A) and increase difficulty of exposure

and dissection. The inflammatory mass may effectively obliterate the triangle of Calot and hide the cystic duct (19) (Fig. 40.3B). Severe inflammation is more likely to be encountered when the time between onset of symptoms and surgery for acute cholecystitis is greater than 72 hours (20,21), when the white blood cell count is greater than 18,000 cells/cu mm (20,22), or if the patient’s age is over 60 years (20,23,24). The Tokyo Severity

361

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Grading for Acute Cholecystitis incorporates these findings in its recommendations regarding timing and expertise for cholecystectomy in this disease (25,26). Severe chronic inflammation with dense scarring: Repeated episodes of acute cholecystitis result in severe scarring of the gallbladder and adjacent tissues with the result that the gallbladder may contract markedly. Fibrotic retraction often binds the gallbladder to the common hepatic duct and right hepatic artery, effectively obliterating the triangle of Calot (Fig. 40.3C). Such inflammation makes dissection very difficult and contributes to visual deception when certain techniques are used (19). Congenital Abnormalities Aberrant Right Hepatic Ducts A low-lying aberrant right hepatic duct is present in about 2% of patients. This aberrancy is the most common type associated with biliary injury. These ducts may lie in or close to the triangle of Calot and are in danger of injury during dissection (27). The most perilous situation occurs when the cystic duct unites with the low-lying aberrant right duct, which then continues to a confluence with the common hepatic duct. The appearance of the junction of the aberrant duct with the hepatic duct is similar to that of the normal confluence of the cystic duct with the common hepatic duct and consequently there is great potential for injury (27). Aberrant right hepatic duct injuries are more common in laparoscopic than open cholecystectomy. Parallel Union Cystic Duct The parallel union cystic duct (Fig. 40.3D) occurs in about 20% of individuals. It was a well-described risk factor for biliary injury in the era of open cholecystectomy and continues to be a risk factor today. Aberrant Position of Cystic Duct Termination The cystic duct may insert into the biliary tree at any point from the right hepatic duct to the termination of the common bile duct. Insertion into the right hepatic duct is very uncommon but exposes this duct to injury. Congenital Adhesions Between the Gallbladder and Common Hepatic Duct Such adhesions are prominent in some individuals. They obscure the triangle of Calot and may fix the common hepatic duct to the side of the gallbladder. In some cases the Pouch of Hartmann may actually lie over and to the left of the common bile and common hepatic ducts (Fig. 40.3E).

Triangle of Calot

(A)

(B)

(C)

(D)

(E)

(F)

Figure 40.3 Conditions which predispose to the deception that the common bile duct is the cystic duct especially when the infundibular techniques is used. (A) Situation in which the infundibular technique is effective. There is mild inflammation. The triangle of Calot is open and the funnel shape of the junction between the gallbladder and cystic duct (heavy line) is readily apparent. (B–F) Conditions which may give a misleading funnel shape (heavy lines) due to concealment or obliteration of the triangle of Calot. In these cases the common bile duct is dissected by a surgeon who thinks it is the cystic duct widening to become the gallbladder. (B) Severe acute cholecystitis. (C) Severe chronic inflammation. (D) Parallel cystic duct insertion. (E) Congenital bands. (F) Large impacted stone.

362

BILE DUCT INJURIES AND BENIGN BILIARY STRICTURES
Abnormal Diameter or Length of Bile Ducts There is considerable variation of bile duct size from person to person. The internal diameter of the cystic duct is normally 2 to 3 mm which would make the external diameter about 3 to 5 mm (28). The normal supraduodenal common bile duct external diameter duct ranges from 4 to 13 mm (28) but the normal internal diameter measured by ultrasound is considered to be 3 to 8 mm with wall thickness accounting for the difference. Rarely duct size may be less than these norms and expose the ducts to injury especially when certain techniques of ductal identification are used. Intrahepatic Gallbladder An intrahepatic gallbladder is difficult to grasp and contributes indirectly to injury by making it difficult to expose the cystic duct (19). Hepatic Ducts Uniting with or Lying in Proximity to the Gallbladder The commonest duct of this type is a duct of Luschka, a 1 to 2 mm accessory duct that runs between an intrahepatic duct and the gallbladder. An injury to a duct of Luschka is difficult to recognize as the duct is tiny and hepatic bile is dilute and lightly straw-colored. In about 10% of individuals a right hepatic duct measuring 2 to 3 mm in diameter lies immediately deep to the cystic plate. In this location it is in danger of injury if the cystic plate is penetrated when dissecting the gallbladder off the plate (gallbladder bed). Isolated reports describe a major hepatic duct that enters directly into the gallbladder usually the right hepatic duct. To the author’s knowledge they have not been described in anatomical dissections of normal specimens. Therefore such ducts are most likely secondary to erosion of a stone into a major bile duct causing a fistula between the gallbladder and the bile duct akin to a Mirizzi syndrome but affecting only the right hepatic duct. Other Patient-Related Factors Large Impacted Gallstones These are not infrequently mentioned in operative notes of cholecystectomies in which biliary injuries have occurred. They tend to impair retraction and hide the cystic duct (19) (Fig. 40.3F). As noted they may efface the cystic duct thus shortening or obliterating it and in severe cases cause common bile duct compression or erosion and with severe pericholecystic inflammation (Mirizzi’s syndrome). Obesity, Body Habitus Obesity, a risk factor for cholelithiasis is common in patients having a cholecystectomy. Morbid obesity and large body size in general contribute to difficulty in operative exposure. The same is true of skeletal deformities. These may be contributing factors to biliary injury. Procedure-Related Factors Misidentification—A Concept Problem There are two main types of bile duct misidentification— misidentification of the common bile duct as the cystic duct (Fig. 40.4A–C) and misidentification of an aberrant right

(A)

(B)

(C)

(D)

(E)

(F)

Figure 40.4 Patterns of biliary injury due to misidentification. (A). The “classical” type E injury in which the common duct is divided between clips at point x. The ductal system is later divided again to remove the gallbladder either at point y1 producing E1 or E2 injuries, or at point y2 producing E3 or E4 injuries. (B) Variant of Type E injury which leads to bile leakage into the operative field and thereby an increased chance of recognition before the entire injury evolves. (C) Variant of Type E injury leading to clipping but not excision of the duct. This injury also causes intraoperative bile leakage, except when cystic and common bile ducts are both occluded, as shown in the inset. D,E, and F represent variants of injury to aberrant right hepatic duct, producing Type B or Type C injuries. The injuries shown in D, E, and F correspond to the injuries shown in A, B, and C, but affect the aberrant right duct.

363

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
hepatic duct as the cystic duct (Fig. 40.4D–F). This type of injury has been called the “classical injury” (29). The type of injury produced may be E1 to E4 and depends upon the level of the second transection. These injuries either result in bile duct obstruction or bile leak depending upon whether the biliary tree is clipped and cut and/or only cut. The second type of misidentification leads to injury of an aberrant right hepatic duct (B and C injuries). The section of the aberrant right hepatic duct, between entry of cystic duct and junction with the common hepatic is mistaken to be the cystic duct (Fig. 40.4D–F). The misidentified section is clipped and usually cut. To remove the gallbladder the aberrant duct must be cut again at a higher level. The key to understanding why misidentification occurs rests with examining the rationale for identification of the cystic structures during cholecystectomy. There are five techniques in general use. “Infundibular Technique” (Fig. 40.5) In this fallible method the putative cystic duct is traced to the gallbladder or the gallbladder down to the cystic duct, at which point, after circumferential dissection of these structures the infundibulum (funnel) is displayed. It is this flaring or widening that was believed to give reliable identification of the cystic duct. However, a flare may also be observed when the common bile duct is followed up to an inflammatory mass within which the cystic duct is hidden (“hidden cystic duct syndrome”) (19).This visual deception is most likely to occur when one or more factors described above are present—severe acute or chronic inflammation, a large stone in the pouch of Hartmann, adhesive bands, intrahepatic gallbladder, short cystic duct, etc. Intraoperative Cholangiography (IOC) IOC reduces the incidence of biliary injury (30,31). Operative cholangiography is best at detecting misidentification of the common bile duct as the cystic duct and will prevent excisional injuries of bile ducts, if the cholangiogram is correctly interpreted. Unfortunately, operative cholangiograms are sometime misinterpreted. The most common misinterpretation is the failure to recognize that when only the lower part of the biliary tree is seen, the common bile duct rather than the cystic duct has been incised and cannulated. IOC is not effective at detecting aberrant right ducts, which unite with the cystic duct before joining the common duct. The aberrant duct appears to be the cystic duct visually and on cholangiograms, especially since some nonaberrant right-sided ducts usually fill (Fig. 40.6). Since it is not unusual to obtain only partial filling of the right hepatic ducts by IOC this is taken as a normal pattern. Another drawback is that an incisional injury of the

Cystic plate

Figure 40.5 The “critical view of safety”. The triangle of Calot is dissected free of all tissue except for cystic duct and artery and the base of the liver bed is exposed. When this view is achieved, the two structures entering the gallbladder can only be the cystic duct and artery. It is not necessary to see the common bile duct.

(A)

(B)

Figure 40.6 (A) Intraoperative cholangiogram which was interpreted as “normal” with only a wisp of dye entering right hepatic ducts. Note stone at ampulla. (B) Postoperative ERCP in same patient done to remove stone. Note retrograde filling of aberrant right hepatic duct, made stenotic by clips. Note how the point of union of the aberrant duct with the common hepatic duct looks like a cystic duct- common duct junction. Problem successfully treated by stenting.

364

BILE DUCT INJURIES AND BENIGN BILIARY STRICTURES
common bile duct, made in order to perform IOC, may not be innocuous. However, the benefits of IOC in ductal identification far outweigh its disadvantages. Dissection of the Cystic Duct to the Confluence with the Common Hepatic Duct/Common Bile Duct This was a common and usually safe technique in performance of open cholecystectomy. There is reason to believe that its use during laparoscopic cholecystectomy has been associated with an increase in lateral injuries to the common hepatic duct (Type D), which have become much more common. The injury is more prone to occur when this technique is used in patients with a parallel union cystic duct. The “Critical View Technique” Introduced in 1992 and modified in 1995 this technique recommends clearing the triangle of Calot of fat and fibrous tissue and taking the gallbladder off the lowest part of its attachment to the gallbladder bed (cystic plate). Only two structures will be connected to the lower end of the gallbladder once this is done. Raising the gallbladder off the lower part of the cystic plate is an important step, equivalent in the open technique to taking the gallbladder off the gallbladder bed. No attempt is made to expose the common bile duct or common hepatic duct (Fig. 40.5). This view provides conclusive, i.e., convincing demonstration that the two structures entering the gallbladder are the cystic duct and artery. “Top-Down” Cholecystectomy In this technique the cholecystectomy is started at the fundus, taking the gallbladder off the gallbladder bed prior to any dissection or identification of structures in the triangle of Calot. While it may be an effective technique of identification in most instances, our experience is that it may lead to serious biliary and vascular injuries in the presence of severe inflammation and usually after conversion to open cholecystectomy. The problem is that the surgeon dissecting from above may “see” the inflammatory mass containing the gallbladder and critical vessels and bile ducts as the gallbladder alone and lacerate or divide one or more of these structures (Figs. 40.7 and 40.8). Also it has been recognized since the earliest days of laparoscopic cholecystectomy that the direction of traction of the gallbladder may contribute to the mistaken conclusion that the common bile duct is the cystic duct—and that this may lead to the misidentification injury. When the pouch of Hartmann is pulled superiorly rather than laterally, the cystic and common bile ducts appear to be a single continuous structure. Technical Problems Bile ducts may be injured in the course of dissection much in the same way that an enterotomy occurs in the course of dissecting adhesions. Inflammation, aberrant anatomy, small duct size, and large body habitus contribute to the likelihood of this occurrence. gallbladder. As noted above about 10% of patients have a sizable hepatic duct which lies immediately deep to the cystic plate and is therefore prone to injury. Failure to Obtain Secure Closure of the Cystic Duct The cystic duct is normally occluded with metallic clips. When the duct is thick, rigid or wide, clips may fail and their use should usually be avoided under these circumstances. Retained common duct stones may contribute to clip failure by raising bile duct pressure.

Common hepatic duct

Cystic duct Common bile duct

(A)

(B)

Figure 40.7 Cause of bile duct injury in “top down” cholecystectomy in the face of severe inflammation with obliteration of triangle of Calot which draws the side of the common hepatic duct to the side of the gallbladder. (A) Real anatomical situation. (B) Apparent anatomical situation is shown by heavy line. The surgeon sees the anatomy bounded by the heavy line as the gallbladder and the cystic duct and as the gallbladder is excised top-down (arrow) the common hepatic duct is transected. Vascular injuries are also common with this mechanism of injury.

365

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

III IV Sg 6,7 Sg 5,8 LHD II

Figure 40.8 Intraoperative photo (left) and diagram (right) showing transection of multiple hepatic ducts during open cholecystectomy by the fundus down technique in a very inflamed gallbladder. The right anterior sectional duct (Sg 5,8), the right posterior sectional duct (Sg 6,7) and the left hepatic duct (LHD) were transected in this high E4 injury.

Thermal Injuries (Fig. 40.9) Thermal injuries are more likely to occur in the presence of severe inflammation, because hemorrhage is more common when dissecting in the face of acute inflammation and higher power settings may be used to control hemorrhage. These injuries are often not recognized at surgery and usually result in bile duct stenosis rather than loss of continuity. Tenting Injuries The junction of the common bile duct and hepatic bile ducts may be occluded when clipping the cystic duct while pulling up forcefully on the gallbladder. This is a very uncommon laparoscopic injury perhaps due to the magnification afforded by laparoscopy. Surgeon/Hospital-Related Factors Learning Curve Effect Inexperience with laparoscopic cholecystectomy was a well documented cause of bile duct injuries in the 1990s. The likelihood of biliary injury was much greater during the early experience of a surgeon than subsequently. It is possible that inexperience in the procedure during acute cholecystitis is still contributing to injury. The Psychology of Human Error Hugh (32) and then Way et al. (33) described traits of human behavior that cause or contribute to biliary injury. Surgery is a complex task in which visual disorientation will occur occasionally and persistence in error due to the deadly mind set error is a common human failing. The mind set error is the tendency to interpret information incorrectly after one has first made a decision. The point of departure of the author with this view is that the visual disorientation is more likely to occur with certain methods of procedure and can be greatly diminished with the use of routine cholangiography or the critical view technique of identification.
Figure 40.9 T-tube cholangiogram in a patient 2 months after a thermal injury. The common hepatic duct (arrow) appears “shrink wrapped” over the T-tube. At the time of reconstruction the common hepatic duct was replaced by scar.

Equipment Laparoscopic equipment must be regularly maintained. Focal loss of insulation on cautery instruments can result in arcing and thermal injuries to bile ducts or bowel.

avoidance of biliary injuries
General Only surgeons trained and proctored in laparoscopic cholecystectomy should perform it. Since laparoscopic cholecystectomy

366

BILE DUCT INJURIES AND BENIGN BILIARY STRICTURES
for acute cholecystitis is more difficult and associated with a higher incidence of biliary injury, it should not be attempted until experience is gained. Special note should be taken of conditions that make surgery during acute cholecystitis particularly difficult and consideration given to percutaneous cholecystostomy as a temporizing measure (25). When inflammation is severe and mandates conversion the open procedure may also be very difficult especially for the surgeon inexperienced in difficult open cholecystectomy. Cholecystostomy or partial cholecystectomy with occlusion of the cystic duct using a pursestring suture placed from the interior of the gallbladder are excellent options under these conditions. Misidentification Injuries Although still in widespread use the author believes that the infundibular technique ought to be discarded as a sole means of ductal identification. It is an error trap, i.e., it works well in most circumstances and seems to be very reliable, but it is actually prone to fail under particular circumstances. Similarly dissection of the cystic duct down to the union with the major bile ducts ought to be discouraged as a routine method of ductal identification. The author favors identification of biliary anatomy by the “critical view of safety” technique since this method is good at identifying the cystic duct even when aberrant ducts are present. Studies attest to the effectiveness of this method and it has been adopted in the guidelines for performance of cholecystectomy in Holland (34,35) If this method is not used then routine use cholangiography is recommended.

presentation and investigation
About 1/3 of the more serious injuries are diagnosed during surgery. Most of the rest are identified in the first 30 days after surgery. Postoperative presentations are influenced by the type of injury and whether a drain has been left. The commonest presentations are pain and sepsis with or without jaundice, jaundice without other symptoms, and biliary fistula. Some patients present only with distension and malaise. The latter is usually due to bile ascites. It is a particularly insidious presentation that may lead to delay in diagnosis. Pain/Sepsis CT scan is performed first to identify fluid collections, which may then be aspirated to determine if they are bilious. Usually a drain is placed in the biloma and an ERCP follows. MRI with MRC has the potential to replace CT scan plus ERCP with a single study. Jaundice Jaundice is usually indicative of the more severe Type E injuries. If jaundice is the only symptom, duct occlusion alone, e.g., by clips, is most likely. Conversely, transections are often accompanied by pain and sepsis due to accumulation of bile in the peritoneal cavity and this is especially true if the injury is several days old. In either case ERCP is the first-line investigation. Next a CT scan is performed. In patients with complete occlusion of the bile duct(s) the bile ducts will be dilated and no biloma will be seen. Percutaneous transhepatic cholangiography (PTC) is performed next to delineate the proximal ducts and to provide external drainage of bile. In patients with transection of bile duct without occlusion the ducts will be decompressed and a biloma or bile ascites is usually present. Our approach for such patients is to drain the biloma and wait for several weeks to perform the PTC. Bile Fistula The first-line investigation is a fistulogram. Subsequent management depends upon anatomical findings. Other Symptoms Occasionally patients with bile ascites may complain only of vague symptoms such as malaise, constipation, or distension. This is because hepatic bile is relatively nonirritating. Hematobilia due to an arterial pseudoaneurysm is a rare but very dangerous presentation. The MRI or CT should be evaluated in all cases for the presence of vascular injury Some have advocated MRI as the best first investigation of a biliary injury since it can show bile ducts, blood vessels, and fluid collections in a single investigation. There is theoretical merit in this argument, but in our experience MRI often lacks the detail required to make detailed plans regarding reconstruction.

technical problems
Injury to a Bile Duct in the Course of Dissection Avoidance depends on the principles of careful dissection and experience, as well as recognition of circumstances in which the potential hazard in continued dissection may outweigh the benefit of completing a cholecystectomy. Also the author recommends not using the “top-down” technique in the face of severe inflammation either in open or laparoscopic surgery. Failure to Obtain Secure Closure of the Cystic Duct Tips of clips should be noted to project beyond the cystic duct. Pre-formed ligature loops should be used for closure of the cystic duct if the cystic duct is thick, rigid or wide. Two loops should be applied on the side of the cystic duct to be retained. Thermal Injuries Cautery should be used with great care in the porta hepatis. Low cautery settings are essential, characteristically 30 W or less. Attempts to stop hemorrhage by blind application of cautery clamps, or clips are very unwise. Brisk bleeding requires conversion. Adhesions should be divided sharply or with minimum application of power. Tenting Injuries The injury is avoided by not lifting the gallbladder forcefully when applying clips to the cystic duct. It is recommended that the surgeon sees that a length of cystic duct will remain below the clip closest to the common bile duct end of the cystic duct before applying that clip.

management of biliary injuries
Management of Injuries Recognized at the Initial Operation Intraoperative recognition of biliary injury is usually an indication for conversion. When appropriate expertise for repair of injury is not available, closed suction drains should be

367

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
placed in the right upper quadrant laparoscopically and the patient referred. Type A injuries, recognized at the time of surgery, are repaired by suture of the cystic duct and drainage. If the anatomy has been clearly demonstrated through dissection or cholangiography, laparoscopic repair by ligature loop or suture is sometimes possible. Type D injuries are repaired by closure of the defect using fine absorbable sutures over a T-tube and placement of a closed suction drain, in the vicinity of the repair. This usually requires conversion to an open procedure. When the Type D injury is thermal in origin, or when the injury involves more than 50% of the circumference of the duct the preferred treatment is Roux-en-Y hepaticojejunostomy, applying the principles of anastomosis given below. Type E injuries recognized intraoperatively should be repaired by hepaticojejunostomy. Choledocho-choledochotomy should be avoided because of considerations of blood supply and tension. Choledocho-duodenostomy has the theoretical disadvantage of tension on the anastomosis as does a loop hepaticojejunostomy. Management of Biliary Injuries Diagnosed Postoperatively The approach depends on type of injury, on type of initial management and its result, and on time elapsed since the initial operation or repair. Type A injuries: The treatment is endoscopic sphincterotomy with placement of a stent or a nasobiliary catheter. Type B and C Injuries: Symptomatic patients with B injuries require hepatico-jejunostomy or hepatic resection if biliaryenteric anastomosis is not possible. In asymptomatic patients, treatment is not recommended when the volume of liver affected is small or if the injury was remote and the isolated portion has atrophied. Type C injuries require drainage of the bile collections and biliary-enteric anastomosis, hepatic resection or ligation of the duct. Type D Injuries: Treatment by endoscopic sphincterotomy and stent is the treatment of choice in the postoperative period. Type E injuries: The best chance for lasting repair is the initial repair. Strictures and sometimes clip occlusions may be treated by dilatation and stents placed either by ERCP. In our experience non-surgical therapy is most likely to be successful when the strictures are mild, appear months to years after surgery, or are of short length. Lillimoe reported 64% success rate with interventional techniques (36). Failures tended to occur when E3 or E4 lesions were treated, when a fistula was present or when a stricture occurred shortly after a hepaticojejunostomy had been done. Non-surgical therapy is most likely to be successful in cases in which operative repair is rather easy and non-surgical treatment often requires multiple endoscopic procedures. The age and health of the patient as well as the likelihood of good long-term outcome should be considered when choosing therapy for a stricture. Timing of Surgery Factors favoring immediate repair are early referral, stable patient, lack of right upper quadrant bile collections, and simpler injuries, which can be rapidly diagnosed and are unlikely to involve vascular injury. Many patients are referred between 1 and 6 weeks after the primary operation when local inflammation may be expected to be great. In these patients percutaneous tubes are inserted to relieve obstruction from affected segments, to drain subhepatic collections and control sepsis. Repair is performed when inflammation has settled, usually about 3 months after the last operation. This delayed approach is sometimes used even when the patient is referred within the first week, especially in complex injuries and those in which either a thermal etiology or a concomitant ischemic injury is suspected (14,16). Immediate repair may also be undertaken when the injury is diagnosed months after surgery, for instance, after failure of stenting of a stenosis or late failure of a biliaryenteric anastomosis. Preoperative Preparation The complete extent of the injury must be diagnosed preoperatively. Failure to do so may result in exclusion of bile ducts from the repair. The percutaneous transhepatic tubes placed to assure biliary drainage from all liver segments also serve as guides to the position of the injured ducts at surgery (Fig. 40.10). Our policy is to perform conciliation between CT and PTC studies to be sure that all ducts in the liver are accounted for. Operative Technique Success depends upon fulfilling six principles of repair, i.e., to construct anastomoses, which are well vascularized, tension free, mucosa-to-mucosa, widely patent, precisely constructed, and that drain all parts of the liver. Most experts in this field recommend hepatico-jejunostomy in preference to either choledocho-choledochotomy or choledocho-duodenostomy, since a tension free anastomosis is always possible with hepaticojejunostomy. Whenever possible, we prefer to construct side-to-side anastomoses to avoid dissection behind bile ducts, which may affect their blood supply (37). Often the anastomosis

Figure 40.10 An E4 injury in which the confluence of the right and left hepatic ducts has been resected. Note the wide separation of the ducts indication a very high injury. Also note multiple clips. Preoperative placement of percutaneous transhepatic tubes facilitated identification of anatomy at surgery.

368

BILE DUCT INJURIES AND BENIGN BILIARY STRICTURES
is done to the extrahepatic portion of the left hepatic duct after it is lowered by dividing the hilar plate (Hepp-Couinaud approach) (38). This approach is particularly suitable for injuries at or just below the bifurcation (Types E2, E3). Right ducts do not lend themselves to this approach as well, since they have a short extrahepatic length. Sometimes the end of the right duct is used. We have described an approach to isolated right hepatic duct injuries (Fig. 40.11) (39). Dissection of the left duct provides a guide to the coronal plane in which the intrahepatic right hepatic ducts will be found and these may be exposed by removing liver tissue. Exposure is also facilitated by dividing the bridge of tissue between segments 3 and 4, by fully opening the gallbladder fossa which often collapses with adherence of it walls. If these maneuvers are not sufficient resecting part of segment 4b and/or 5 will open the upper porta hepatis as described by Mercado (40). When ductal reconstruction to a part of the liver is impossible then resection should be performed (41). Occasionally prior failure of reconstruction leads to secondary biliary cirrhosis and end-stage liver failure. Then liver transplantation is required (17,41). In almost all examples of this outcome, high reconstructions have been attempted by surgeons lacking experience in these difficult procedures or there has been a combined biliary/vascular injury. Treatment of failed repairs with metallic stents gives very poor results in the long term with 50% of treated patient suffering from repeated cholangitis. Outcome of Treatment Most surgical series of biliary reconstruction cite very good short-term results. However, it is well established from older literature describing ductal injury during open cholecystectomy that there is a progressive re-stenosis rate. Two-thirds of recurrences are diagnosed in the first two years after repair but re-stenosis has been described after 10 years. In reports of more than 50 repairs, the re-stenosis rate varies from 4% to 21% and the need for operative re-repair from 1% to 9.5% (13,16,40,42–51). Postoperative complications are common. The experience from our center in 113 repairs followed for a median of 4.9 years is a 4.5% rate of poor outcomes requiring interventional treatment but no requirement for operative re-repair (37). Comparison among surgical series is difficult because of lack of standard reporting variables and effect of differences in the severity of injuries treated in different series. Injuries above the confluence involving several bile ducts have a worse prognosis than injuries of the common hepatic duct and the proportion of severe injuries in a series will affect outcome. Reporting of treatment failure is not uniform. Length of follow-up is another obvious variable affecting outcome and is not uniform among series. Several authors have reported that quality of life (52–54) and lifespan (8) are adversely affected by a biliary injury.

post-transplantation biliary strictures
The incidence of biliary strictures is 5% to 15% after whole liver transplantation and higher after right-lobe live donor surgery. Technical problems that lead to anastomotic stricture, ischemia and rejection are the etiologic factors (55). Today management is by endoscopic means, which is usually successful in the anastomotic type of stricture (56) and less so in the other types. Avoidance strategies focus mainly on prevention of ischemia at the lower end of duct to duct anastomoses and at the hilum in live donor transplantation.

noniatrogenic bile duct strictures
Biliary Strictures Secondary to Pancreatitis Severe acute pancreatitis and chronic pancreatitis may produce benign biliary strictures in the intrapancreatic segment of the common bile duct. That associated with acute pancreatitis usually resolves with the attack. The pancreatitis may be diffuse or focal in the head of the pancreas. Alcoholic chronic pancreatitis that produces a dense scarring of the gland is a common type of this stricture however other types of chronic pancreatitis including what appears to be an autoimmune variant have been increasingly recognized as being etiologic. Sometimes the bile duct narrowing is due to compression by a pseudocyst rather than fibrosis. In these cases relief of

Step 1

Step 2
C. Core or lift liver off right Portal pedicle

Gallbladder plate

A. Identify left duct by HeppCouinaud technique D. Open bile duct on anterior surface

B. Find right portal pedicle and divide gallbladder plate

(A)

(B)

Figure 40.11 (A) Technique for identifying isolated right ducts. Step 1: Finding and dividing cystic plate to expose right pedicle. (B) Technique for identifying isolated right ducts. Step 2: Elevating right portal pedicle, identifying and incising right duct.

369

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
compression by treatment or more rarely by spontaneous resolution of the pseudocyst may be relieve the stenosis. Presenting symptoms are jaundice with or without pain. Diagnosis has classically been made by CAT scan combined with ERCP, but more recently MRI and MRCP has been increasingly used, as it is less invasive. ERCP permits brushings and these may diagnose malignancy. Characteristically, the stricture is of the long smooth “rat-tail” type. One of the continuing conundrums in pancreatic surgery is the differentiation between benign and malignant causes of intrapancreatic bile duct stricture. Focal pancreatitis can produce a mass in the head of the pancreas, cause jaundice and thereby mimic cancer. Cancer is more likely arising in the chronically inflamed gland. The cancers are often scirrhous making diagnosis by needle biopsy more difficult especially in an already chronically inflamed gland. EUS-guided FNA and core biopsies are very helpful in making the diagnosis and the CA19-9 level is also helpful. Occasionally laparoscopic US-guided biopsy may be useful. Treatment depends upon whether biliary stricture is an isolated problem or whether it is part of a more general problem such as unremitting pancreatic pain, whether a pseudocyst is present, or whether cancer is a serious consideration. For isolated strictures biliary enteric anastomosis was the standard treatment; however, dilatation with multiple endoscopically placed stents can be successful (57) and may be used first, surgery being reserved for patients who fail this treatment (58). When caused by pseudocyst compression treatment may simply require drainage of the cyst, which now can often be accomplished endoscopically. Treatment may be part of a procedure to treat chronic pancreatitis such as the Frey or Beger procedures. These relieve the obstruction by removal of the surrounding compressive scar or by adding biliary enteric anastomosis. Pylorus-preserving or standard Whipple procedures, which relieve the biliary obstruction by resection are sometimes used when cancer is suspected. Metallic stents should never be used in this or other benign strictures as they eventually clog and result in the need for a Whipple procedure. Stricture due to Noniatrogenic Bile Duct Injuries Noniatrogenic bile duct injuries are usually caused by nonpenetrating trauma and are often a part of wider injuries (59). Isolated bile duct injury due to penetrating trauma may occur but more often it is not isolated and is usually fatal as the hepatic artery and portal vein are also injured. The injury is often missed and diagnosis is delayed. The principles of complete diagnosis and repair enunciated above apply to these injuries as well. Combined biliary and arterial injuries may occur and the surgeon must be aware that the bile duct may be devascularized to a higher level than the laceration or transection Strictures due to Calculous Disease Oriental cholangiohepatitis is a disease associated with biliary parasite infestation leading to the development of intrahepatic and extrahepatic bile duct stones, usually of the brown pigment type. Recurrent pyogenic infections are common and these lead to strictures. Treatment is by a combination of therapies, including eradication of the parasites, removal of the stones, dilatation of the strictures, and occasionally local liver resection. Sometimes much of the therapy can be accomplished percutaneously. At other times operative removal of stones and hepaticojejunostomy is required; the Roux loop may be place in proximity to the skin to permit later percutaneous access. Choledocholithiasis occurring in western countries may lead to biliary strictures through repeated bouts of cholangitis or local stone ulceration and resultant stricture formation often near the lower end of the bile duct, although today this is very rare. Endoscopic sphincterotomy (ES) is the treatment for low strictures and biliary enteric anastomoses are used when is unsuccessful or not applicable. Large stones within the gallbladder may cause biliary obstruction by external compression (Mirizzi’s syndrome). Sclerosing Cholangitis Sclerosing cholangitis is an idiopathic disease, probably autoimmune in type, frequently associated with ulcerative colitis, that causes biliary strictures, which are usually both intrahepatic and extrahepatic and multiple. Stones are not usually present. Degeneration to cancer may occur and these cancers are difficult to diagnose early. Treatment of symptomatic localized strictures in the larger bile ducts is usually endoscopic (60). Resection of the confluence has been advocated when the major area of stricturing is localized to that area (61). Many of these patients require orthotopic liver transplantation as the definitive procedure surgical procedure and this is done when end-stage liver disease appears or when cancer is suspected. Prior surgery on the bile ducts may make OLT more difficult. Benign Inflammatory Pseudotumors This inflammatory process of unknown etiology was first described by Stamatakis et al. (62) and has several synonyms. The masses consist of chronic inflammatory cells and fibrosis. Benign inflammatory tumors appear to occur most frequently in extrahepatic upper ducts, but also occur intrahepatically (63), and less commonly in lower ducts. Benign inflammatory tumors mimic malignant biliary obstructions at the hilum, causing jaundice. Patients tend to be younger with a median age of 50 but the tumors cannot be differentiated from cancer on the basis of presentation or liver function tests. CA19-9 serum levels are normal or slightly elevated (64) unless cholangitis is present. A mass at the hilum is commonly present on axial imaging but vascular invasion/ encasement has not been described with this process. Currently they are treated as cholangiocarcinomas and resected.

references
1. Keus F dJJ, Gooszen HG, van Laarhoven CJHM. Laparoscopic versus open cholecystectomy for patients with symptomatic cholecystolithiasis. The Cochrane Library 2008, Issue 2 2008: 1–133. 2. Adamsen S, Hansen OH, Funch-Jensen P, Schulze S, Stage JG, Wara P. Bile duct injury during laparoscopic cholecystectomy: a prospective nationwide series. J Am Coll Surg 1997; 184: 571–8. 3. Russell JC, Walsh SJ, Mattie AS, Lynch JT. Bile duct injuries, 1989–1993. A statewide experience. Connecticut Laparoscopic Cholecystectomy Registry. Arch Surg 1996; 131: 382–8.

370

BILE DUCT INJURIES AND BENIGN BILIARY STRICTURES
4. Merrie AE, Booth MW, Shah A, Pettigrew RA, McCall JL. Bile duct imaging and injury: a regional audit of laparoscopic cholecystectomy. Austr NZ J Surg 1997; 67: 706–11. 5. Solheim K, Buanes T. Bile duct injury in laparoscopic cholecystectomy. Int Surg 1995; 80: 361–4. 6. Richardson MC, Bell G, Fullarton GM. Incidence and nature of bile duct injuries following laparoscopic cholecystectomy: an audit of 5913 cases. West of Scotland Laparoscopic Cholecystectomy Audit Group [see comments]. BrJ Surg 1996; 83: 1356–60. 7. Wherry DC, Rob CG, Marohn MR, Rich NM. An external audit of laparoscopic cholecystectomy performed in medical treatment facilities of the department of Defense. Ann Surg 1994; 220: 626–34. 8. Flum DR, Cheadle A, Prela C, et al. Bile duct injury during cholecystectomy and survival in medicare beneficiaries. JAMA 2003; 290: 2168–73. 9. Waage A, Nilsson M, Waage A, Nilsson M. Iatrogenic bile duct injury: a population-based study of 152 776 cholecystectomies in the Swedish Inpatient Registry. Arch Surg 2006; 141: 1207–13. 10. Bismuth H. Postoperative strictures of the bile duct. In: Blumgart LH, ed. The Biliary Tract. Edinburgh: Churchill Livingston; 1982: 209–18. 11. Bismuth H, Majno P. Biliary strictures: classification based on the principles of surgical treatment. World J Surg 2001; 25: 1241–4. 12. Boerma D, Rauws EA, Keulemans YC, et al. Impaired quality of life 5 years after bile duct injury during laparoscopic cholecystectomy: a prospective analysis.[see comment]. Ann Surg 2001; 234: 750–7. 13. Stewart L, Way LW. Bile duct injuries during laparoscopic cholecystectomy: Factors that influence the results of treatment. Arch Surg 1995; 1995: 1123–9. 14. Strasberg SM, Hertl M, Soper NJ. An analysis of the problem of biliary injury during laparoscopic cholecystectomy [see comments]. [Review] [111 refs]. J Am Coll Surg 1995; 180: 101–25. 15. Koffron A, Ferrario M, Parsons W, et al. Failed primary management of iatrogenic biliary injury: incidence and significance of concomitant hepatic arterial disruption. Surgery 2001; 130: 722–8; discussion 8–31. 16. Alves A, Farges O, Nicolet J, et al. Incidence and consequence of an hepatic artery injury in patients with postcholecystectomy bile duct strictures. Ann Surg 2003; 238: 93–6. 17. Thomson BN, Parks RW, Madhavan KK, Garden OJ, Thomson BNJ. Liver resection and transplantation in the management of iatrogenic biliary injury. World J Surg 2007; 31: 2363–9. 18. Gurusamy KS, Samraj K. Early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Cochrane Database Syst Rev 2006: CD005440. 19. Strasberg SM, Eagon CJ, Drebin JA. The “hidden cystic duct” syndrome and the infundibular technique of laparoscopic cholecystectomy—the danger of the false infundibulum. J Am Coll Surg 2000; 191: 661–7. 20. Brodsky A, Matter I, Sabo E, et al. Laparoscopic cholecystectomy for acute cholecystitis: can the need for conversion and the probability of complications be predicted? A prospective study. Surg Endosc 2000; 14: 755–60. 21. Hadad SM, Vaidya JS, Baker L, et al. Delay from symptom onset increases the conversion rate in laparoscopic cholecystectomy for acute cholecystitis. World J Surg 2007; 31: 1298–01; discussion 302–3. 22. Halachmi S, DiCastro N, Matter I, et al. Laparoscopic cholecystectomy for acute cholecystitis: how do fever and leucocytosis relate to conversion and complications? Euro J Surg 2000; 166: 136–40. 23. Lim KR, Ibrahim S, Tan NC, et al. Risk factors for conversion to open surgery in patients with acute cholecystitis undergoing interval laparoscopic cholecystectomy. Ann Acad Med, Singapore 2007; 36: 631–5. 24. Lo CM, Fan ST, Liu CL, Lai EC, Wong J. Early decision for conversion of laparoscopic to open cholecystectomy for treatment of acute cholecystitis. Am J Surg 1997; 173: 513–7. 25. Yamashita Y, Takada T, Kawarada Y, et al. Surgical treatment of patients with acute cholecystitis: Tokyo Guidelines. J Hepato-Biliary-Pancreatic Surg 2007; 14: 91–7. 26. Strasberg SM, Strasberg SM. Clinical practice. Acute calculous cholecystitis. New Engl J Med 2008; 358: 2804–11. 27. Meyers WC, Peterseim DS, Pappas TN, et al. Low insertion of hepatic segmental duct VII-VIII is an important cause of major biliary injury or misdiagnosis [see comment]. Am J Surg 1996; 171: 187–91. 28. Kune G, Sali A. The Practice of Biliary Surgery. 2nd edn. Oxford: Blackwell Scientific Publications; 1980. 29. Davidoff AM, Pappas TN, Murray EA, et al. Mechanisms of major biliary injury during laparoscopic cholecystectomy . Ann Surg 1992; 215: 196–202. 30. Fletcher DR, Hobbs MS, Tan P, et al. Complications of cholecystectomy: risks of the laparoscopic approach and protective effects of operative cholangiography: a population-based study. Ann Surg 1999; 229: 449–57. 31. Flum DR, Dellinger EP, Cheadle A, et al. Intraoperative cholangiography and risk of common bile duct injury during cholecystectomy. JAMA 2003; 289: 1639–44. 32. Hugh TB, Hugh TB. New strategies to prevent laparoscopic bile duct injury--surgeons can learn from pilots. Surgery 2002; 132: 826–35. 33. Way LW, Stewart L, Gantert W, et al. Causes and prevention of laparoscopic bile duct injuries: analysis of 252 cases from a human factors and cognitive psychology perspective [see comment]. Ann Surg 2003; 237: 460–9. 34. Heistermann HP, Tobusch A, Palmes D. [Prevention of bile duct injuries after laparoscopic cholecystectomy. “The critical view of safety”]. Zentralblatt fur Chirurgie 2006; 131: 460–5. 35. Yegiyants S, Collins JC, Yegiyants S, Collins JC. Operative strategy can reduce the incidence of major bile duct injury in laparoscopic cholecystectomy. Am Surg 2008; 74: 985–7. 36. Lillemoe KD, Martin SA, Cameron JL, et al. Major bile duct injuries during laparoscopic cholecystectomy. Follow-up after combined surgical and radiologic management. Ann Surg 1997; 225: 459–68; discussion 68–71. 37. Winslow ER, Fialkowski EA, Linehan DC, et al. “Sideways”: results of repair of biliary injuries using a policy of side-to-side hepatico-jejunostomy. Ann Surg 249: 426–34, 2009 38. Hepp J. Hepaticojejunostomy using the left biliary trunk for iatrogenic biliary lesions: the French connection. World J Surg 1985; 9: 507–11. 39. Strasberg SM, Picus DD, Drebin JA. Results of a new strategy for reconstruction of biliary injuries having an isolated right-sided component. J Gastrointest Surg 2001; 5: 266–74. 40. Mercado MA, Chan C, Orozco H, et al. Long-term evaluation of biliary reconstruction after partial resection of segments IV and V in iatrogenic injuries. J Gastrointest Surg 2006; 10: 77–82. 41. Laurent A, Sauvanet A, Farges O, et al. Major hepatectomy for the treatment of complex bile duct injury. Ann Surg 2008; 248: 77–83. 42. Kaman L, Sanyal S, Behera A, et al. Comparison of major bile duct injuries following laparoscopic cholecystectomy and open cholecystectomy. ANZ J Surg 2006; 76: 788–91. 43. Murr MM, Gigot JF, Nagorney DM, et al. Long-term results of biliary reconstruction after laparoscopic bile duct injuries. Arch Surg 1999; 134: 604–9; discussion 9–10. 44. Mercado MA, Chan C, Orozco H, et al. To stent or not to stent bilioenteric anastomosis after iatrogenic injury: a dilemma not answered? Arch Surg 2002; 137: 60–3. 45. Wudel LJ, Jr., Wright JK, Pinson CW, et al. Bile duct injury following laparoscopic cholecystectomy: a cause for continued concern. Am Surg 2001; 67: 557–63; discussion 63–4. 46. Tocchi A, Costa G, Lepre L, Liotta G, Mazzoni G, Sita A. The long-term outcome of hepaticojejunostomy in the treatment of benign bile duct strictures. Ann Surg 1996; 224: 162–7. 47. Walsh RM, Henderson JM, Vogt DP, et al. Long-term outcome of biliary reconstruction for bile duct injuries from laparoscopic cholecystectomies. Surgery 2007; 142: 450–6; discussion 6–7. 48. Chapman WC, Halevy A, Blumgart LH, Benjamin IS. Postcholecystectomy bile duct strictures. Management and outcome in 130 patients. Arch Surg 1995; 130: 597–602; discussion 4. 49. de Reuver PR, Grossmann I, Busch OR, et al. Referral pattern and timing of repair are risk factors for complications after reconstructive surgery for bile duct injury. Ann Surg 2007; 245: 763–70. 50. Lillemoe KD, Melton GB, Cameron JL, et al. Postoperative bile duct strictures: management and outcome in the 1990s. Ann Surg 2000; 232: 430–41. 51. Pottakkat B, Sikora SS, Kumar A, et al. Recurrent bile duct stricture: causes and long-term results of surgical management. J Hepato-Biliary-Pancreatic Surg 2007; 14: 171–6. 52. de Reuver PR, Sprangers MA, Rauws EA, et al. Impact of bile duct injury after laparoscopic cholecystectomy on quality of life: a longitudinal study after multidisciplinary treatment. Endoscopy 2008; 40: 637–43.

371

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
53. Moore DE, Feurer ID, Holzman MD, et al. Long-term detrimental effect of bile duct injury on health-related quality of life. Arch Surg 2004; 139: 476–81; discussion 81–2. 54. Melton GB, Lillemoe KD, Cameron JL, et al. Major bile duct injuries associated with laparoscopic cholecystectomy: effect of surgical repair on quality of life. Ann Surg 2002; 235: 888–95. 55. Sharma S, Gurakar A, Jabbour N, et al. Biliary strictures following liver transplantation: past, present and preventive strategies. Liver Transplantation 2008; 14: 759–69. 56. Barriga J, Thompson R, Shokouh-Amiri H, et al. Biliary strictures after liver transplantation. Predictive factors for response to endoscopic management and long-term outcome. Am J Med Sci 2008; 335: 439–43. 57. Pozsar J, Sahin P, Laszlo F, et al. Medium-term results of endoscopic treatment of common bile duct strictures in chronic calcifying pancreatitis with increasing numbers of stents.[see comment]. J Clin Gastroenterol 2004; 38: 118–23. 58. Judah JR, Draganov PV, Judah JR, Draganov PV. Endoscopic therapy of benign biliary strictures. World J Gastroenterol 2007; 13: 3531–9. 59. Sawaya DE Jr., Johnson LW, Sittig K, McDonald JC, Zibari GB. Iatrogenic and noniatrogenic extrahepatic biliary tract injuries: a multi-institutional review. Am Surg 2001; 67: 473–7. 60. Johnson GK, Saeian K, Geenen JE, et al. Primary sclerosing cholangitis treated by endoscopic biliary dilation: review and long-term follow-up evaluation. Curr Gastroenterol Rep 2006; 8: 147–55. 61. Pawlik TM, Olbrecht VA, Pitt HA, et al. Primary sclerosing cholangitis: role of extrahepatic biliary resection. J Am Coll Surg 2008; 206: 822–30; discussion 30–2. 62. Stamatakis JD, Howard ER, Williams R. Benign inflammatory tumour of the common bile duct. Br J Surg 1979; 66: 257–8. 63. Schmid A, Janig D, Bohuszlavizki A, Henne-Bruns D. Inflammatory pseudotumor of the liver presenting as incidentaloma: report of a case and review of the literature. Hepatogastroenterol 1996; 43: 1009–14. 64. Koea J, Holden A, Chau K, et al. Differential diagnosis of stenosing lesions at the hepatic hilus. World J Surg 2004; 28: 466–70.

372

41

Gallstones and common bile duct stones—surgical and non-surgical approaches Matthew P. Dearing and Michael Rhodes
with radiation to the right lower chest or the lower pole of the scapula. This can be explained by the origin of the gallbladder from the lower thoracic segments, and transmission of visceral sensation through splanchnic nerves to the lower thoracic spinal cord. Some pain afferents may travel within the right phrenic nerve and peritoneum below the right diaphragm, accounting for the radiation of pain to the right shoulder tip. The pain of biliary colic often has a slower periodicity than ureteric colic, lasting between 30 minutes and 2 hours, but sometimes persisting for up to 8 hours. Although it is a commonly held belief that the consumption of fatty food is likely to provoke an attack, there is little evidence to support this (6). The pain may resolve spontaneously but often requires parenteral analgesia and anti-emetics. Acute Cholecystitis This is characterized by the presence of right upper quadrant or epigastric pain lasting for more than 8 hours, accompanied by systemic signs of infection such as fever and leukocytosis. The condition arises after impaction of a stone in the neck of the gallbladder or cystic duct, leading to sustained high pressure in the gallbladder. This leads to a reduction in the blood flow to the mucosa with subsequent ischemic injury (6). As a result of this, a chemical cholecystitis develops with inflammatory infiltrate and edema of the gallbladder wall. In the first 48 hours it is unusual for bacterial infection to occur but the incidence of infection after 1 week is in the order of 70% (6). Traditionally patients were admitted to hospital for a period of intense medical therapy—analgesia, intra-venous fluid rehydration and broad-spectrum antibiotics, with cholecystectomy deferred for 6 to 12 weeks (12). This used to be common practice across the United Kingdom. In one survey, 88% of surgeons reported routinely managing patients in this manner (13). However, it has been shown that early surgery (laparoscopic or open) is preferable to delayed surgery and is not associated with any additional morbidity or mortality (12–14). Laparoscopic cholecystectomy, widely accepted as the gold standard of treatment, has been shown to be safe and effective in acute cholecystitis, with shortened length of hospital stay (15).

introduction
About 10% to 15% of the adult Western population has gallstones (1,2). This equates to 7.5 million Britons and 20 million Americans with gallstones, with between 1% and 4% a year developing symptoms (2,3). Cholecystectomy is not recommended in asymptomatic patients (4). However, symptomatic gallstone disease is one of the most common conditions in the United Kingdom requiring surgery (5,6). Up to 35% of patients with gallstones will ultimately become symptomatic, requiring cholecystectomy (7). In England, 49,077 cholecystectomies were performed between April 2005 and March 2006 (3). It is estimated that over half a million cholecystectomies are performed each year in the United States (5). Gallstones are seen in all age groups but incidence increases with age, with over one-third of the population over the age of 70 years affected (3,5). The development of gallstones is a multifactorial process, but has been shown to be associated with family history, pregnancy, obesity, rapid weight loss (such as after bariatric surgery), hemolytic anemias, parenteral nutrition, loss of bile salts (as seen in terminal ileitis and after terminal ileal resection), and diabetes mellitus (via the metabolic syndrome) (8).

classification of gallstones
There are three common types of gallstone. 1. Cholesterol (20%) 2. Bile Pigment (5%) 3. Mixed (75%) In western populations the consumption of a diet which is high in cholesterol and fat leads to the supersaturation of bile, with resultant precipitation and crystal growth (Fig. 41.1). Pure cholesterol stones are often solitary or exist as clusters of “mulberries” (9). Bile pigment stone are multiple, irregular, small, black, and fragile. They are seen in patients with chronic hemolysis (hereditary spherocytosis and sickle cell disease) and cirrhosis, where there is an increase in bilirubin (10). Mixed stones, composed of varying proportions of the above ingredients, are usually multiple.

presentations of gallstone disease
Gallstones are responsible for a wide range of clinical problems. The most common of these are biliary colic (56%) and acute cholecystitis (36%) (11). Biliary Colic This arises when a gallstone impacts in the neck of the gallbladder leading to obstruction of the cystic duct. Sustained gallbladder contraction, produces a rise in pressure within the gallbladder, leading to the pain perceived as biliary colic (6). This is typically centered in the right upper quadrant, or epigastrium,

empyema
The contents of the gallbladder become purulent, as a result of bacterial proliferation and exudation of neutrophils. The resultant collection of pus is termed an empyema. The adjacent omentum often becomes involved in the inflammatory process encasing the gallbladder and leading to the development of large tender mass in the right upper quadrant. The patient will be systemically unwell with a high swinging temperature and leukocytosis. Treatment involves controlling the sepsis with broad-spectrum antibiotics and percutaneous ultrasound-guided drainage, prior to definitive surgery. Most

373

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Decreased cholesterol 7 alpha-hydroxylase

Liver

XXX

Increased cholesterol Decreased bile salt secretion

These may demonstrate small bowel obstruction, a gallstone in the bowel lumen and gas in the biliary tree. Often the diagnosis is made at laparotomy. These conditions are rare but important causes of bowel obstruction, particularly in the elderly, where gallstones and cholecystenteric fistula are more common. Gallstones are implicated in 20% of cases of small bowel obstruction where there is no history of a hernia or previous abdominal surgery (17).

obstructive jaundice
Gall bladder Bile supersaturated with cholesterol

Nidus for gallstone formation

Cholesterol gallstone

Figure 41.1 Pathogenesis of formation of cholesterol stones.

patients can be managed in this way, although in severe cases the inflammatory process produces patchy necrosis of the gallbladder wall, which can result in perforation. Often the omentum is adherent around the gallbladder and this contains the perforation, leading to the formation of a peri-cholecystic abscess. In a small proportion of patients, usually the elderly, the perforation is not contained and generalized peritonitis develops with associated high morbidity and mortality (16).

Gallstones can migrate through the cystic duct into the common bile duct. Sometimes they will pass spontaneously causing no symptoms. The likelihood of stones passing spontaneously is related to their size (18). Stones up to 8 mm in diameter may pass without problems (19,20). When a stone impacts at the lower end of the common bile duct this can lead to obstructive jaundice. The impacted stone prevents the normal flow of bile resulting in biliary colic as the gallbladder contracts against the obstruction. The jaundice may be transient, resolving in 24 to 48 hours, as the stone either disimpacts and floats free in the bile duct or passes into the duodenum (6). If the stone remains impacted then the jaundice will persist and deepen. The diagnosis is based on analysis of the liver function tests, showing a characteristic rise in alkaline phosphatase and only minor derangement of transaminases, together with finding dilated extra-hepatic bile ducts on ultrasound. This can be confirmed with magnetic resonance cholangiopancreatography, endoscopic ultrasound, or ERCP (21). The prevalence of asymptomatic bile duct stones has been estimated between 5.2% and 12% (22–24). Common bile duct stones have been found in between 10% and 18% of patients undergoing cholecystectomy (25).

acute cholangitis
Acute cholangitis results when bacterial infection develops within a partially or completely obstructed biliary system. Stasis leads to an increase in the resident bacterial flora, which are often found when gallstones are present in the biliary tract. Classically the condition presents with Charcot’s Triad of symptoms—right upper quadrant pain, rigors, and jaundice. Ascending infection of the biliary tree can lead to septicemia and multiple hepatic abscesses. One-fifth of patients presenting with cholangitis have a bacteremia, with gram negative organisms mainly responsible (26). Treatment is with broad spectrum antibiotics, such as second generation cephalosporins or ciprofloxacin. ERCP is performed at an early stage allowing both confirmation of the diagnosis and decompression of the bile ducts. Biliary stents can be placed to encourage further drainage prior to definitive treatment.

mucocele
In this situation the obstructed gallbladder remains free of infection. Although the mucosa remains inflamed, it continues to function, absorbing water from bile and secreting mucus. The gallbladder fills with clear or bile-stained mucus. The patient will often report an episode of severe pain, consistent with cholecystitis. The symptoms may resolve partly, but there will be some persisting right upper quadrant pain and tenderness.

cholecystenteric fistula and gallstone ileus
A cholecystenteric fistula develops when a gallstones fistulates directly into the stomach, duodenum, or colon following adherence of the gallbladder to these structures. This occurs as a result of pressure necrosis. Whilst this can be seen after an episode of acute cholecystitis it often follows a period of silent inflammation (3). If the stones are small, as is often the case, they will not obstruct the bowel. However, larger stones can cause obstruction at certain sites. First, at the ileo-caecal valve causing small bowel obstruction, or within the duodenum causing gastroduodenal obstruction (Bouveret’s syndrome). The diagnosis can sometimes be made with plain radiographs.

acute pancreatitis
Gallstones are responsible for up to 60% of all cases of pancreatitis. Between 4% and 8% of patients with gallstones will develop acute pancreatitis due to migratory stones (27). Small stones, with mean diameter of 4 mm, are more likely to cause pancreatitis than larger stones (28). The condition develops when a small gallstone, traveling from gallbladder to the bowel, impacts in the biliopancreatic duct

374

GALLSTONES AND COMMON BILE DUCT STONES
(the common channel) of the duodenal ampulla. Transitory obstruction of the duct leads to reflux of duodenal and/or biliary fluid into the pancreatic duct. This initiates premature activation of enzymes in the pancreas, leading to pancreatitis (29,30). Patients with pancreatitis present with central or epigastric abdominal pain. Classically this is constant, radiating to the back and relieved by leaning forwards. Profuse vomiting is common. The diagnosis can be confirmed by significantly elevated serum amylase or lipase. There are several scoring systems used to predict the severity of pancreatitis, including the Ranson system, the modified Imrie system, Apache II score, and Balthazar grading system. They are based on organ dysfunction and local complications (31,32)). Most patients will have a self-limiting disease but in severe cases the mortality can be high. Overall mortality remains around 10% (33). with cholesterol stones (38). However, stones were found to recur in up to 70% of patients over 5-year follow-up (39). UCDA has a role in the prevention of gallstones, although it does not appear to be of use once stones have developed (40,41)). Miller and colleagues looked at the formation of gallstones in patients following obesity surgery (an established risk factor for developing stones). They found significantly lower gallstone formation in patients treated with UCDA compared to placebo (3% vs. 22% at 12 months and 8% vs. 30% at 24 months) (40). However, Venneman et al. found that UCDA was not beneficial in patients with symptomatic gallstones (41). Ezetimibe, a potent cholesterol absorption inhibitor, has recently been found to be effective in reducing biliary cholesterol content. This offers the potential for promising new strategies in the prevention and treatment of cholesterol gallstones (42). Percutaneous Cholecystostomy In critically ill or high-risk patients with biliary sepsis, percutaneous cholecystostomy allows decompression of the inflamed gallbladder, with resolution of sepsis. Patients can then undergo definitive surgery once their overall condition has improved. A retrospective review of 55 patients treated by percutaneous transhepatic cholecystostomy reported successful biliary drainage in 98%, with 95% of patients recovering well (43). Percutaneous cholecystostomy can be performed under ultrasound or fluoroscopic guidance. The gallbladder can be entered using the Seldinger technique with tract dilatation and catheter placement via guide wire or by means of the direct trocar technique (44). The choice of tract depends on the anatomy and whether stone extraction is planned. The transhepatic route is associated with less risk of bile leakage, but the transperitoneal approach is preferable for stone removal through a larger tract (43). Surgical Management of Gallstones Cholecystectomy Cholecystectomy was first performed by Langenbuch in 1882. It has been the accepted treatment for symptomatic gallstones for over a century. Laparoscopic cholecystectomy has revolutionized the surgical management of gallstones, replacing the standard “open” cholecystectomy as the gold standard of treatment (45). Open Cholecystectomy The main indication for open cholecystectomy is failure of the laparoscopic approach. This is often due to inflammation around the cystic duct or artery which makes dissection and definition of Calot’s triangle difficult. In some cases dense adhesions from previous abdominal surgery necessitate conversion to open surgery. Open cholecystectomy is also performed when there is a suspicion of gallbladder cancer as there is well-documented evidence of laparoscopic port site recurrence (46–49). Traditionally, open cholecystectomy was performed through an oblique right subcostal incision. The incision should be centered over the fundus of the gallbladder, allowing optimal access. The gallbladder should be removed using a fundus-first

mirrizi's syndrome
This occurs when a gallstone, impacted in the neck of the gallbladder, causes inflammation to the surrounding tissue thereby compressing the adjacent bile duct. The stone induces fibrosis leading to obliteration of the cystic duct. Ongoing inflammation results in adherence of the gallbladder to the common hepatic duct, causing partial obstruction and jaundice. Mirrizi’s syndrome has been classified into two types. In type 1, there is no communication between the gallbladder and the bile duct. In type 2, the gallstone has eroded into the bile duct, resulting in a cholecyst–choledochal fistula.

biliary dyspepsia
This represents a set of vague abdominal symptoms which have been attributed to gallstones. These include non-specific upper abdominal pain, nausea, belching, and food intolerance. These symptoms have been found to occur with equal frequency in patients with or without gallstones (34). In fact, symptoms other than typical biliary colic are rarely due to gallstones (6). Cholecystectomy in these situations is often ineffective with up to 30% of patients continuing to suffer dyspeptic symptoms (35,36).

management of gallstone disease
Non-surgical Management of Gallstones Analgesia for Biliary Colic/Cholecystitis In the acute setting the initial management involves analgesia. This is best achieved with diclofenac and opioids (morphine and pethidine). Antiemetics are often required as nausea and vomiting are prominent symptoms. Drug Dissolution Therapy This involves the use of medications, such as methyl tert butyl ether (MTBE) and ursodeoxycholic acid (UCDA) to dissolve cholesterol back into bile. They are of use in the treatment of cholesterol gallstones but are not able to achieve dissolution in calcified or pigment stones. This restricts their use to around 10% to 20% of patients. MTBE is directly instilled into the gallbladder via endoscopic transpapillary catheterization (37). The early results suggested this to be an effective treatment, achieving complete dissolution of stones in 80% to 90% of patients

375

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Figure 41.2 Laparoscopic cholecystectomy. The gallbladder fundus is grasped and pushed cephalad.

Figure 41.4 A window is created behind the cystic duct and artery.

Figure 41.3 Dissection of Calot’s triangle—identification of the cystic duct.

technique. During dissection, the gallbladder neck is retracted to the patient’s right. This stretches the peritoneum overlying the cystic duct and common duct, facilitating safe dissection. The cystic duct, cleared of peritoneum, can be cannulated in order to perform operative cholangiography. Following this the duct can be clipped and ligated. The cystic artery should be clipped and ligated similarly. The gallbladder can be removed from the liver bed using diathermy. The rectus sheath should be closed in 2 layers with continuous monofilament suture. Laparoscopic Cholecystectomy Laparoscopic cholecystectomy, which was first performed in 1985 (50), is well established as the preferred method of treatment (45). These procedures are routinely performed as day cases (51) and during the index admission for cholecystitis (12). This has been shown to be both safe and efficacious. The standard procedure involves four ports and will be described below. Operative Technique The patient is positioned supine on the operating table. The pneumoperitoneum is established via an open longitudinal

incision below the umbilicus, followed by insertion of a blunt 10 mm trocar and insufflation of carbon dioxide. This technique is preferable to the use of the Veress needle which has a higher incidence of bowel and vascular injuries (52). Further ports can then be inserted under direct vision. Two 5 mm ports are placed in the right flank (in the anterior axillary and midclavicular lines) and a further 10 mm port is placed in the midline about 5 cm below the xiphisternum. The usual setup requires the surgeon to stand on the patient’s left side with the monitor positioned on the right. The gallbladder fundus is grasped atraumatically and pushed cephalad (Fig. 41.2). The assistant is required to hold the grasper with the left hand, maintaining traction, and the camera with the right hand. Adhesions can be taken down with either blunt or sharp dissection. Another grasper, inserted via the medial 5 mm port, is used to apply lateral and caudal traction to Hartmann’s pouch. This allows Calot’s triangle to be identified. The peritoneum overlying Calot’s triangle is opened with diathermy. The cystic duct and artery are carefully dissected and identified (Fig. 41.3). This can be done with either blunt dissection or with cautious and sparing use of the diathermy hook. By moving the gallbladder medially the surgeon can continue the dissection, developing a window behind the cystic artery (Fig. 41.4). Having identified the cystic duct, cholangiography can be performed. Although this is not routinely performed in all cases it should be available, particularly if bile duct stones are suspected or there is difficulty in defining the biliary anatomy. Scissors are used to open the cystic duct and a 4 Fr catheter is inserted and held in place with a clip. Contrast media is injected via the catheter and real-time images are obtained using the image intensifier. A normal cholangiogram must demonstrate flow of contrast in the main bile duct and right and left hepatic ducts. There must be flow distally into the duodenum, and no stones present in the duct. After obtaining a cholangiogram and removing the catheter, the cystic duct and artery can be clipped and divided (Fig. 41.5). The gallbladder can then be removed with the diathermy hook. The gallbladder fossa is inspected thoroughly at the end

376

GALLSTONES AND COMMON BILE DUCT STONES
Endoscopic Removal of Bile Duct Stones/Biliary Stenting Endoscopic retrograde cholangiography (ERC) is highly sensitive in the detection of common bile duct stones, with rates of over 90% reported (28). It also permits therapeutic removal of stones at the same time. After cannulation of the papilla, spincterotomy can be performed with diathermy. This allows small stones to be extracted using a Dormia basket or balloon catheter. Endoscopic balloon dilatation of the sphincter can also be performed. However, balloon dilation of the papilla should be avoided in most patients because of the increased risk of post procedure pancreatitis, compared to biliary sphincterotomy (28). These techniques are effective in removing 90% to 95% of bile duct stones (28). Short-term biliary stenting can be used in the management of retained common bile duct stones, permitting adequate biliary drainage prior to further endoscopy or surgery. The use of biliary stents as the sole treatment of common bile duct stones is recommended only for high risk patients or those with limited life expectancy (72). Lithotripsy Mechanical lithotripsy can be used to break up larger stones. Several studies have reported on the efficacy of the technique, with bile duct clearance rates of between 68% and 98% (59,60). The wide variation in the success rates can be attributed to the size of the stone. Over 90% of stones with a diameter of 10 mm or less can be cleared, whereas only 68% of stones with a diameter over 28 mm can be removed (61). Laser lithotripsy can also be used in the treatment of retained bile duct stones, with reported success rates of between 64% and 97% (62–64). Extracorporeal shockwave lithotripsy and electrohydraulic lithotripsy have been used to treat large, retained bile duct stones and bile clearance rates of 83% (65–67) and 74% (68) have been reported. Surgical Management Open Choledocholithotomy The common bile duct is accessed via a supraduodenal approach. A cholangiogram can be performed via the cystic duct to delineate the anatomy and determine the site and number of stones. Following cholecystectomy and mobilization of the duodenum (kocher’s maneuver), the lower common bile duct can be accessed and choledochotomy performed. Gentle palpation of the duct permits stones to be milked into the choledochotomy. Choledochoscopy can then be performed, with removal of stones by Dormia basket or Fogarty catheter. The bile duct can be closed primarily or over a T-tube. T-tube drainage used to be standard practice after bile duct exploration, but a large retrospective audit published recently found that there was a lower biliary complication rate after primary closure of the duct (69). A T-tube may be inserted if there is evidence of cholangitis, multiple stones, or a large duct. A further cholangiogram can be done at this stage to check for retained stones and confirm that contrast can be seen flowing into the duodenum. A drain should be placed after choledochotomy for 24 to 48 hours. It can be removed if there is no bile drainage.

Figure 41.5 The cystic duct is clipped and divided.

of the procedure for bleeding and bile leaks, as are the cystic duct and artery clips. The area can be gently irrigated with saline and then suctioned. The camera is moved to the top port and the gallbladder is grasped via the umbilical port and delivered. The linea alba can be closed with absorbable sutures and the skin with steristrips or sutures. Future Developments Advances in surgical technology and innovation have made possible the advent of 2 port (53) and single port laparoscopic cholecystectomy (54) and the development of NOTES—natural orifice transluminal endoscopic surgery (55). This will allow the surgeon to undertake major intraperitoneal surgery without the need for skin incisions, with access to the peritoneal cavity via the mouth (via the stomach), the rectum and the vagina, using flexible endoscopes. There is, to date, no research published on resections in humans but the first transvaginal cholecystectomy in a porcine model was carried out in 2005 (55). Progress in this area is likely to be rapid since the set up of working groups, dedicated to advancing the concept (56).

management of common bile duct stones
Non-surgical Management Drug Dissolution Therapy As previously discussed, ursodeoxycholic acid (UCDA) has been used to dissolve cholesterol gallstones (28). However, at present there are no large randomized controlled trials to demonstrate that UCDA is of benefit in the treatment of bile duct stones (28). Much of the work done to evaluate the use of UCDA has involved patients with gallstones rather than bile duct stones (57). UCDA has been used, in combination with endoscopic retrograde cholangiography (ERC) and stent insertion, in the management of difficult to extract bile duct stones. In one study, Johnson and colleagues demonstrated that 90% of patients treated with both UCDA and stent insertion had ductal clearance at repeat ERC compared to none of the patients in the stent alone group (58). However, further work is needed to define the role of UCDA in the management of bile duct stones.

377

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Laparoscopic Common Bile Duct Stone Removal Between 70% and 95% of bile duct stones can be removed using the laparoscopic approach (70,71)). Laparoscopic access is achieved as described earlier. An operative cholangiogram can be performed via the cystic duct and small stones can be removed without the need to open the main bile duct. However, this approach will not be successful in all patients. The cystic duct may have a very tortuous anatomical course with prominent spiral valves or a small diameter which prevent stone extraction. In cases of multiple or large stones or where proximal duct stones have been identified on the cholangiogram, it may be necessary to perform a choledochotomy in the distal bile duct. Choledochoscopy can then be performed using a flexible choledochoscope. The stones can be extracted using a Dormia basket or Fogarty catheter. The choledochotomy can be primarily closed or a T-tube inserted if conditions are not favorable for primary closure.

references
1. NIH Consensus Statement on gallstones and laparoscopic cholecystectomy. National Institute of Health Consensus Development Conference Statement Sep 14–16, 1992. 2. Halldestam I, Enell EL, Kullman E, et al. Development of symptoms and complications in individuals with asymptomatic gallstones. Br J Surg 2004; 91(6): 734–8. 3. Sanders G, Kingsnorth AN. Gallstones. BMJ 2007; 335: 295–9. 4. Gurusamy KS, Samraj K. Cholecystectomy versus no cholecystectomy in patients with silent gallstones. Cochrane Database Syst Rev 2007. CD 006230. 5. Geibel J, Longo W. Cholelithiasis: evolving standards for diagnosis and management. World J Gastroenterol 2006 May 28; 12(20): 3162–7. 6. Johnson CD. Gallstones. Surg J 21: 5 May 2003. 7. Schirmer BD, Winters KL, Edlich RF. Cholelithiasis and cholecystitis. J Long Term Eff Med Implants 2005; 15: 329–38. 8. Shaffer EA. Gallstone disease: epidemiology of gallbladder stone disease. Best Pract Res Clin Gastrenterol 2006; 20: 981–96. 9. Ellis H, Calne R, Watson C. Lecture Notes on General Surgery, 9th edn. 256–64. 10. Kumar P, Clark M. Clinical Medicine, 5th edn. Edinburgh: W B Saunders, 2002: 981–6. 11. Glasgow RE, Cho M, Hatter MM, et al. The spectrum and cost of complicated gallstone disease in California. Arch Surg 2000; 135: 1021–5. 12. Papi C, Catarci M, D’Ambrosio L, et al. Timing of cholecystectomy for acute calculous cholecystitis: a meta-analysis. Am J of Gastroenterol 2004; 99(1): 147–55. 13. Cameron IC, Chadwick C, Phillips J, et al. Current practice in the management of acute cholecystitis. Br J Surg 2000; 87: 366–7. 14. Gurusamy KS, Samraj K. Cholecystectomy for acute cholecystitis. Cochrane Database Syst Rev 2006 CD 005440. 15. Keus F, deJong JA, Gooszen HG, et al. Laparoscopic versus open cholecystectomy for patients with symptomatic cholecystolithiasis. Cochrane Database Syst Rev 2006 CD 006231. 16. Stefandis D, Kirinek KR, Bingener J. Gallbladder perforation: risk factors and outcome. J Surg Res. 2006; 131(2): 204–8. 17. Clavien PA, Richon J. Gallstone ileus. Br J Surg 1990; 77(7): 737–42. 18. Fisher EJ, Sherman S. The interface of ERCP and laparoscopic cholecystecomy. Gastrointest Endosc Clin N Arm 1996; 6(1): 57–80. 19. Chung RS, Chad V, Eisenstat M. Choledocholithiasis treated with endoscopic stenting of the papilla followed by stent guided sphincterotomy. Gastrointest Endosc 1997; 45: 121–30. 20. Frossard JL, Hadengue A, Amouyal G, et al. Choledochlithiasis: a prospective study of common bile duct stone migration. Gastrointest Endosc 2000; 51(2): 175–9. 21. Beckingham IJ. MRCP versus ERCP in the diagnosis of choledochlithiasis. Eur J Gastro Hepatol 2003; 15(7): 809–13. 22. Murison MS, Gartell PC, McGinn FP. Does selective preoperative cholangiography result in missed bile duct stones? J R Coll Surg Edin 1993; 38(4): 220–4. 23. Rosseland AR, Stomsaker TB. Asymptomatic common bile duct stones. Eur J Gastr Hepatol 2000; 12(11): 1171–3. 24. Sarli L, Pietra N, Franze A, et al. Routine intravenous cholangiography; selective ERCP and endoscopic treatment of bile duct stones before laparoscopic cholecystectomy. Gastrointest Endosc 1999; 50(2): 200–8. 25. Martin DJ, Vernon DR, Toouli J. Cochrane Database of Systematic Review 2006 CD 003327. 26. Leung JW, Ling TK, Chan RL, et al. Antibiotics, biliary sepsis and bile duct stones. Gastrointest Endosc 1994; 40(6): 716–21. 27. Ayub K, Imada R, Slavis J. ERCP in gallstone associated pancreatitis. Cochrane Database of Systematic Review 2004; CD 003630. 28. Caddy GR, Tham TC. Gallstone disease: Symptoms, diagnosis and endoscopic management of common bile duct stones. Best Prac Res Clin Gastr 2006; 20(6): 1085–101. 29. Mayer AD, McMahon MJ, Benson EA, et al. Operations on the biliary tract in patients with acute pancreatitis; aims, indications and timing. Ann R Coll Surg Engl 1984; 66: 179–83. 30. Neoptolemos JP, Hall AW, Finlay DF, et al. The urgent diagnosis of gallstones in acute pancreatitis: a prospective study of 3 methods. Br J Surg 1984; 71: 230–3.

key points
10% to 15% of the adult Western population have gallstones. Between 1% and 4% of patients a year develop symptoms. Cholecystectomy is not recommended for asymptomatic gallstones (4)—Grade A. Cholecystectomy is recommended for all patients with symptomatic gallbladder stones and CBDS (except where surgery is considered inappropriate). Grade A. Laparoscopic cholecystectomy is the gold standard of treatment for symptomatic gallstone disease (45)—Grade A. Laparoscopic cholecystectomy is safe and effective in the management of acute cholecystitis (12,14))—Grade A. Cholecystectomy during the index admission for cholecystitis is recommended (12) – Grade A. Symptomatic patients with suspected ductal stones should undergo stone extraction where possible—Grade B. ERCP is not recommended solely as a diagnostic test, it should only be done in patients for whom an intervention is planned. Grade B Biliary sphincterotomy and endoscopic stone extraction are recommended for patients with bile duct stones post cholecystectomy. Biliary stenting should only be used as a sole treatment in patients with limited life expectancy or very high-surgical risk. Grade A. Patients with ductal stones undergoing laparoscopic cholecystectomy may be managed by laparoscopic common bile duct exploration at the time of surgery, or undergo peri-operative ERCP. Transcystic and transductal exploration of the common bile duct are both recommended approaches for removal of stones. Grade A. In patients with bile duct stones that have not been extracted, short term use of a biliary stent is recommended, followed by further endoscopy or surgery. Grade B. If duct clearance is not achieved by minimally invasive methods then open surgical exploration is recommended as an important treatment option. Grade B.

378

GALLSTONES AND COMMON BILE DUCT STONES
31. Banks PA. A new classification system for acute pancreatitis. Am J Gastroenterol 1994; 89(2): 151–2. 32. Bradley EL. A clinically based system for acute pancreatitis. Summary of the international symposium on acute pancreatitis. Arch Surg 1993; 128(5): 586–90.. 33. Toh SK, Phillips S, Johnson CD. A prospective audit against national standards of the presentation and management of acute pancreatitis in the South of England. Gut 2000; 46(2): 239–43. 34. Price WH. Gallbladder dyspepsia. Br Med J 1963 Jul 20; 2(5350) 138–41. 35. Black NA, Thompson E, Sanderson CF. Symptoms and health status before and 6 weeks after cholecystectomy. Gut 1994; 35(9): 1301–5. 36. Ferster LF, Lonborg R, Thirlby RE. Am J Surg 1995; 169(5): 533–8. 37. Uchida N, Nakatsu T, Hirabayashi S, et al. J Gastroenterol Aug 1994; 29: 4. 38. Thistle JU, May GR, Bender CE, et al. Dissolution of cholesterol gallstones b Methyl tert butyl ether administered by percutaneous transhepatic catheter. N Engl J Med 1989; 320: 633–9. 39. Hellstern A, Leuschner U, Benjaminov A, et al. Dissolution of gallbladder stones with methyl tert butyl ether and stone recurrence. A European Survey. Dig Dis Sci 1998; 43: 911. 40. Miller K, Hell E, Lang B, et al. Gallstone formation prophylaxis after gastric restriction procedures for weight loss, a randomized double-blind placebo controlled trial. Ann Surg 2003; 238: 697–702. 41. Venneman NG, Besselink MG, Keulemans VC, et al. Ursodeoxycholic acid exerts no beneficial effect in patients with symptomic gallstones awaiting cholecystectomy. Hepatology 2006; 43: 1276–83. 42. Wang HH, Portincasa P, Mendez-Sanchez N, et al. Effect of ezetimibe on the prevention and dissolution of cholesterol gallstones. Gastroenterology 2008 Jun; 134(7): 2101–10. 43. Spira RM, Nissan A, Zamir O, et al. Percutaneous transhepatic cholecystostomy and delayed cholecystectomy in critically ill patients with acute calculous cholecystitis. Am J Surg; 183: 62–6. 44. Borloff A, Chen MY, Ott DJ, et al. Gallbladder stones: imaging and intervention. Radiographics 2000; 20(3): 751–66. 45. Bittner R. The standard of laparoscopic cholecystectomy. Langenbecks Arch Surg 2000; 389: 157–63. 46. Clair DG, Lautz DB, Brooks DC. Rapid development of umbilical metastases after laparoscopic cholecystectomy for unsuspected gallbladder cancer. Surgery 1993; 113: 355–8. 47. Landen SM. Laparoscopic surgery and tumour seeding. Surgery 1993; 114: 131–2. 48. Fligelstone L, Rhodes M, Flook D, et al. Tumour inoculation during laparoscopy. Lancet 1993; 342: 368–9. 49. Kim HJ, Roy T. Unexpected gallbladder cancer with cutaneous seeding after laparoscopic cholecystectomy. South Med J 1994; 87: 817–20. 50. Muhe E. The first cholecystectomy through the laparoscope. Langenbecks Arch Surg 1986; 369: 804. 51. Gurusamy KS, Junnakar S, Farouk M, et al. Meta-analysis of randomised controlled trials on the safety and effectiveness of day-case laparoscopic cholecystectomy. Br J Surg 2008; 95: 161–8. 52. Osborne DA, Alexander G, Boe B, et al. Laparoscopic cholecystectomy – past, present and future. Surg Technol Int 2006; 15: 81–5. 53. Bonjer HJ, Hazebroek EJ, Kazemier G, et al. open versus closed establishment of pneumoperitoneum in laparoscopic surgery. Br J Surg 1997; 84: 599–602. 54. Romanelli JK, Mark L, Omotosho PA. Single port laparoscopic cholecystectomy with the Triport system, a case report. Surg Innov 2008; 15(3) 223–8. 55. Park PO, Bergstrom M, Ikeda K, et al. Experimental studies of transgastric gallbladder surgery: cholecystectomy and cholecystogastric anastomosis. Gastrointest Endosc 2005; 61: 601–6. 56. Rattner D, Kalloo A. ASGE/SAGES. Working group on natural orifice transluminal endoscopic surgery. Oct 2005. Surg Endosc 2006; 20: 329–33. 57. Danziger RG, Hoffman AF, Schoenfield P, et al. Dissolution of cholesterol gallstones by chendeoxycholic acid. N Engl J Med 1972; 286(1): 1–8. 58. Johnson GK, Geenen JE, Venu RP, et al. Treatment of non-extractable CBD stones with combination ursodeoxycholic acid and endoprosthesis. Gastrointest Endosc 1993; 39(4): 528–32. 59. Chung WH, Chu CH, Wang TE, et al. Outcome of simple use of mechanical lithotripsy of difficult common bile duct stones. World J Gastroenterol 2005; 11(4): 593–6. 60. Garg PK, Tandon RK, Akuja V, et al. Predictors of unsuccessful mechanica lithotripsy and endoscopic clearance of large bile duct stones. Gastrointest Endosc 2004; 59(6): 601–5. 61. Cipolletta C, Costamagna G, Bianco MA, et al. Endoscopic mechanical lithotripsy of difficult bile duct stones. Br J Surg1997; 84(10): 1407–9. 62. Brambs HJ, Duda SH, Rieber A, et al. Treatment of bile duct stones: value of laser lithotripsy delivered by percutaneous endoscopy. Eur J Radiol 1996; 6(5): 734–40. 63. Harris VJ, Sherman S, Trevotola S, et al. Complex biliary stones: treatment with a small choledochoscope and laser lithotripsy. Radiology 1996; 199(1): 71–7. 64. Born P, Neuhaus H. Laser lithotripsy of refrctory bile duct stones after failure of endoscopic shockwave lithotripsy. Gastroenterol 1995; 33(4): 202–8. 65. Sauerbruch T, Delius M, Paumgartner G, et al. Fragmentation of gallstones by endoscopic shockwaves. N Engl J Med 1986; 314: 818–22. 66. Sackmann M, Holl J, Sanffer G, et al. Endoscopic shockwave lithotripsy for clearance of bile duct stones. Gastrointest Endosc 2001; 53(1): 27–32. 67. Ellis RD, Jenkins AP, Thompson RP, et al. Clearance of refractory bile duct stones with endoscopic shockwave lithotripsy. Gut 2000; 47(5) 728–31. 68. Arya N, Nelles E, Haber GB, et al. EHL in 111 patients: a safe and effective therapy for difficult bile duct stones. Am J Gastroenterol 2004; 99(12): 2330–4. 69. Ahmed I, Pradhan C, Beckingham IJ, et al. Is T Tube necessary after common bile duct exploration. World J Surg 2008; 32: 1485–88. 70. Rhodes M, Sussman L, Cohen, et al. Randomized trial of laparoscopic exploration of common bile duct versus postoperative ERCP for common bile duct stones. Lancet 1998 Jan 17; 351: 159–61. 71. Cuschieri A, Croce E, Faggioni A, et al. EAES ductal stone study: preliminary findings of multi-centre prospective randomized trial comparing 2 stage versus single stage management. Surg Endosc 1996 Dec; 10(2): 1130–5. 72. Williams EJ, Green J, Beckingham I, et al. Guidelines on the management of common bile duct stones. British Society of Gastroenterology. Gut 2008 Jul; 57(7): 1004–21.

379

42 Adenocarcinoma of the pancreas
introduction
The most frequent type of pancreatic adenocarcinoma, pancreatic ductal adenocarcinoma (PDAC), accounts for more than 85% of all pancreatic malignancies (1). Other malignant pancreatic tumors include intraductal papillary mucinous carcinomas, mucinous cystadenocarcinomas, acinar cell cancers, as well as endocrine cancers. These rare tumors will not be discussed in this chapter. PDAC is characterized by late presentation, aggressive local tumor growth, early lymphogenic and hematogenic spread, and poor prognosis.

André L. Mihaljevic, Jörg Kleeff, and Helmut Friess
although an association between excessive drinking and pancreatic cancer cannot be ruled out (35). Furthermore, it should be mentioned that alcohol is one of the leading causes of sporadic chronic pancreatitis which increases the risk for pancreatic cancer 2.3- to 25-fold (36–40). Finally, diabetes mellitus poses an increased risk for the development of pancreatic cancer. In one population-based study, the risk was increased up to eightfold (41), although other studies showed smaller increases in the relative risk (13,42–46) Familial background is an important, non-modifiable risk factor. First-degree relatives of PDAC patients have an approximately two- to three-fold increased risk to develop the disease themselves depending on the age of the affected family member (47,48). Familial predisposition seems to account for approximately 1.9% to 5% of all PDAC cases (1,49,50) and occurs in three clinical settings: (1) familial pancreatic cancer syndrome (FPC), (2) hereditary pancreatitis and cystic fibrosis, and (3) hereditary tumor syndromes involving the pancreas such as the Peutz–Jeghers syndrome. In FPC, two or more first-degree relatives are affected by PDCA while other hereditary causes (familial tumor syndromes, cystic fibrosis, and hereditary pancreatitis) have been ruled out (51). FPC is believed to account for 70% of all hereditary pancreatic cancer cases (51) and the increase in risk is dependent on the number of affected individuals in a family (52). Individuals from FPC families with two affected first-degree relatives have a relative risk of about 6.4%, those with three or more affected first-degree relatives of 32% (48). Importantly, in FPC families, patients from younger generations are, on average, affected 10 years earlier than their parents (genetic anticipation) (53). The genetic alterations causing FPC remain elusive and up till now potential genetic defects have been identified in only a number of FPC families including mutations in the Palladin (54) or BRCA2 gene (55–57). Two hereditary syndromes causing chronic pancreatitis are also associated with an increased risk for PDAC: hereditary pancreatitis and cystic fibrosis. Hereditary pancreatitis is an autosomal dominant disorder characterized by recurrent episodes of acute pancreatitis at a young age eventually leading to chronic pancreatitis. It is associated with a 50- to 70-fold increased risk for PDAC (38,58). Approximately 70% of hereditary pancreatitis cases are caused by loss-of-function mutations in the protease serine 1 (PRSS1) (59) blocking the trypsine inactivation in pancreatic acinar cells and leading to autodigestion. Less frequently, mutations in the SPINK1 gene, coding for the serine protease inhibitor Kazal-type 1, were found to be associated with hereditary pancreatitis (60). Cystic fibrosis, the most frequent hereditary metabolic disorder in Caucasians, on the other hand, is an autosomal-recessive disease caused by mutations in the CFTR gene (cystic fibrosis transmembrane regulator) and is characterized by viscous

epidemiology
PDAC is the tenth most common type of cancer in the United States and the United Kingdom but accounts for the fourth and sixth most frequent type of cancer-related death in these countries (2–4), respectively. Due to its aggressive nature, the number of new cases per year (an estimated 37,680 in the U.S. in 2008 and 7,632 in the UK in 2005) almost equals the number of deaths per year (34,290 in the U.S. and 7315 in the UK) (3–6). PDAC has an overall median survival of less than 6 months and one of the lowest overall 5-year survival rates of any malignant disease with 0.4% to 5% (7,8). Furthermore, these dismal figures seem to have improved little compared to the 1950s (5). The incidence rate peaks between 65 and 75 years of age (9) and only 15% to 20% of patients initially present with localized disease suitable for potential curative surgical resection.

etiology and risk factors
Numerous risk factors have been implicated in the development of PDAC. Studies have focused not only on identifying modifiable risk factors like diet and life-style habits, but also tried to elucidate hereditary and genetic risk factors. Diet as a risk factor for pancreatic cancer has been investigated in multiple studies yielding contradicting results (10–16). Currently, no specific diet can be recommended to reduce the risk of PDAC (evidence grade 2b, recommendation C). An overview of the available data is given in Table 42.1. Obesity, however, seems to pose a risk factor for the development of pancreatic cancer (17,18). In a metaanalysis of six case–control and eight cohort studies, a body-mass index of 30 was associated with an increase in the relative risk of 1.19 compared to normal weighing subjects (19). Consequently, an increase in physical activity seems to lower the risk for pancreatic cancer (18,20,21). Another well-documented risk factor is smoking which seems to roughly double the risk for pancreatic cancer (22–26). Individual genetic factors appear to influence this association (27–29) and even passive smoking has been linked to an increased risk in one population-based study (30). Alcohol, albeit one of the leading causes of chronic pancreatitis, does not seem to increase the risk for pancreatic cancer (31–34),

380

ADENOCARCINOMA OF THE PANCREAS
Table 42.1 Influence of Dietary Factors on the Development of Pancreatic Cancer.
Dietary factor Dietary fiber Fruit/vegetables Vitamin C Low fat low cholesterol Grilled/smoked meat Low sugar Fish Tea Coffee Alcohol Influence on risk No proven benefit No proven benefit Potential benefit No proven benefit Potential risk No proven benefit No proven benefit No proven benefit No proven risk No proven risk Grade of evidence and recommendation 3, C 2b, C 3, D 2b, B 3, C 2b, B 2b, B 2b, B 2b, B 3, C References (12,248,249) (250) (12,251) (15,14) (252–254) (254–256) (15,14) (257,258) (25,31,259) (25,31,260,34)

mucus production in exocrine glands. Several studies have analyzed the association between cystic fibrosis and PDAC showing a 2.3- to 32-fold increase in relative risk (61,62). Finally, a number of hereditary tumor syndromes are associated with an increased risk for PDAC. Table 42.2 gives an overview of hereditary pancreatic cancer disorders.

symptoms and clinical presentation
Symptoms of PDAC include back pain, painless obstructive jaundice, weight loss, loss of appetite, and fatigue and are largely unspecific. Associated conditions include endocrine (new-onset diabetes mellitus) as well as exocrine pancreatic insufficiency (maldigestion, steatorrhea) or an unexplained attack of acute pancreatitis. Clinical features such as persistent back pain, ascites, supraclavicular lymphadenopathy, or a palpable abdominal mass may indicate advanced tumor disease. No data exist regarding which clinical symptom or combination of symptoms should trigger further diagnostic measures. New onset epigastric or back pain is the most frequent symptom of PDAC. Based on expert opinion (evidence grade 4, recommendation D), a clinical approach based on age and coexisting symptoms has been proposed for patients presenting with new onset back pain (Table 42.3) (63). Painless jaundice in PDAC patients is caused by obstruction of the intrapancreatic common bile duct. It is a classical feature of PDAC since 80% to 90% of tumors are located in the head of the gland (64). Painless jaundice should always trigger further diagnostic measures to rule out pancreatic or common bile duct malignancies since they account for over 20% of cases in patients over 60 years of age (evidence grade 2b, recommendation B) (65–67). Conversely, the less frequent tumors of the pancreatic tail rarely cause jaundice and are typically diagnosed at a later stage. Late-onset diabetes mellitus (>50 year of age) may be caused by PDAC (68). Conversely, diabetes mellitus seems to be a risk factor for pancreatic cancer (see Etiology and Risk Factors). Eighty percent of PDAC patients exhibit diabetes mellitus or an impaired glucose tolerance at the time of diagnosis (69). Furthermore, a population-based study suggests that PDAC will be detected in about 1% of diabetics over 50 years who had their diabetes diagnosed within 3 years of cancer diagnosis. Most (56%) of these patients had been diagnosed with diabetes within 6 months of cancer detection (70). Others have

argued that the lack of efficient diagnostic tools to detect pancreatic cancer early does not warrant further diagnostic measures in patients with late-onset diabetes as their sole symptom (63). Although PDAC contributes only marginally to the overall number of acute pancreatitis cases, idiopathic acute pancreatitis in patients over 50 of age justifies further diagnostic workup, since up to 5% of PDAC patients are present with idiopathic pancreatitis as their first clinical manifestation (63,71).

diagnosis and staging
The diagnostic work-up in patients with suspected pancreatic cancer should verify the diagnosis and determine surgical resectability. Pathological tumor staging follows the UICC staging system (Table 42.4); however, for further clinical decision making it suffices to group the tumor into one of the following simple categories: 1. Resectable tumors without distant metastasis 2. Locally advanced tumors, borderline resectable, no distant metastasis 3. Locally advanced tumors, unresectable, no distant metastasis 4. Tumors with distant metastasis The following questions must be answered during the diagnostic work-up in order to correctly classify the tumor: 1. Presence or absence of distant metastasis (hepatic and/or peritoneal) 2. Patency of the superior mesenteric vein (SMV), the splenic vein, the portal confluence, and the portal vein 3. Patency of the superior mesenteric artery (SMA), the coeliac axis, the hepatic artery, and the gastroduodenal artery 4. Presence of aberrant vascular anatomy 5. Tumor invasion into neighboring organs per continuitatem Table 42.5 gives an overview of how tumor infiltration translates into surgical resectability. The diagnostic work-up should follow a logical order, be as little time consuming as possible, and start with simple, inexpensive measures followed by more invasive procedures (Fig. 42.1).

381

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 42.2 Overview of Hereditary Pancreatic Cancer Disorders.
Clinical presentation 1. Familial pancreatic cancer Two or more first-degree relatives with PDAC Gene Not known, BRCA2 (6–17%, (57,56,55)) PRSS1 SPINK1 CFTR Cumulative risk for PDAC till the age of 70 (Ref. 268) 40%

2. Hereditary forms of CP Hereditary pancreatitis Cystic fibrosis

Recurrent acute pancreatitis in young patients Exocrine pancreas insufficiency DIOS COPD Biliary cirrhosis Reduced fertility Intestinal hamartomatous polyps Mucocutaneous pigment spots FAMM: multiple dysplastic naevi/melanoma in two or more first-degree relatives 25% of FAMM families show association with PDAC MPCS: melanoma and PDAC Breast cancer Ovarian cancer Early-onset right-sided colorectal cancer Carcinoma of the pancreas, hepatobiliary tract, endometrium, ovary, stomach, small bowel, brain, upper uroepithelial tract Multiple colonic polyps and colon cancer Duodenal tumors Desmoids and others Breast cancer, Sarcoma, leukemia Brain tumors and others

40% <5%

3. Hereditary tumor syndromes Peutz–Jeghers FAMM/MPCS

STK11 CDKN2A

36% 17%

FBOC HNPCC

BRCA1/2 MLH1, MSH2, MSH6, PMS2

3–8% <5%

FAP

APC

<5%

LiFraumeni

TP53

<5%

Familial pancreatic cancer syndrome (FPC), hereditary pancreatitis and cystic fibrosis as well as hereditary tumor syndromes with PDAC association are listed. The causative genes are listed as well as the cumulative risk to develop PDAC the age of 70. Abbreviations: DIOS, distal intestinal obstruction syndrome; COPD, chronic obstructive pulmonary disease; STK11, serine-threonine protein kinase 11; PDAC, pancreatic ductal adenocarcinoma; FPC, familial pancreatic cancer; BRCA1/2, breast-cancer gene 1/2; PRSS1, Protease Serine 1; SPINK1, serine protease inhibitor Kazal-type 1; CFTR, cystic fibrosis transmembrane regulator; FAMM, familial atypical multiple mole melanoma syndrome; MPCS, melanoma pancreatic cancer syndrome; CDKN2A, cyclin-dependent kinase inhibitor 2A; HNPCC, hereditary non-polyposis colon cancer; FAP, familial adenomatous polyposis; APC, adenomatous polyposis coli; FBOC, familial breast and ovarian cancer syndrome; MLH1, MutL homolog 1; MSH2/6, MutS homolog 2/6; PMS2, postmeiotic segregation 2; TP53, tumor protein 53.

Following the initial suspicion of a pancreatic malignancy after taking the patients’ history and conducting a thorough physical exam, a laboratory work-up and a transabdominal ultrasound are the first steps to be taken. Laboratory values should include amylase and lipase, cholestasis parameters (γGT, direct and indirect bilirubin, alkaline phosphatase), the transaminases ALT and AST, a blood count to evaluate tumor anemia, coagulation parameters, and the tumor marker carbohydrate antigen 19-9 (CA199). None of these parameters are specific for pancreatic cancer and none can be used as a screening tool. The tumor-associated antigen CA19-9 specifically has been tested in two large studies and found to be ineffective as a screening tool in asymptomatic patients because of its low positive predictive value (72,73). Furthermore, controversies exist regarding the correct cut-off value for CA19-9.

However, in the context of suspected pancreatic malignancy, CA19-9 remains a valuable adjunct since in combination with computed tomography (CT) a positive predictive value of 99% can be achieved when levels are over 120 U/ml (74). In addition, high levels of CA19-9 (over 150 U/ml) were shown to serve as an indicator for irresectability with a positive predicitive value of 88% and might justify further staging procedures (75). Finally, CA19-9 can be used as a parameter of therapeutic response and as an indicator of recurrence in patients with PDAC. Transabdominal ultrasonography is a fast and cost-effective measure and provides valuable information. The presence of ascites, hepatic metastasis as well as dilated intrahepatic and extrahepatic bile ducts can be evaluated with high accuracy, whereas the visualization of the primary tumor is achieved

382

ADENOCARCINOMA OF THE PANCREAS
Table 42.3 Age- and Symptom Oriented Approach to New Onset Epigastric or Back Pain Based on Expert Opinion (Evidence Grade 4, Recommendation D)
Level of suspicion Low Medium Age <50 <50 Symptoms Pain only
a

Work-up US if pain persists US, (CT)

High

>50 >50

Pain plus Loss of appetite/loss of weight/fatigue Pain onlya Pain plus Loss of appetite/loss of weight/fatigue

US, EUS, (CT)

a

Epigastric pain that radiates to the back and persists at night should trigger further investigations independent of age. Abbreviations: US, abdominal ultrasound; EUS, endoscopic ultrasound; CT, computed tomography.

Table 42.4 UICC Staging of Pancreatic Cancer (Seventh Edition, 2010)
M0 N0 Tis T1 T2 T3 T4 0 IA IB IIA III N1 ø IIB IIB IIB III M1 N0/N1 IV

Abbreviations: Tis: carcinoma in situ; T1, tumor limited to the pancreas <2 cm; T2, tumor limited to the pancreas >2 cm; T3: tumor extending beyond the pancreas but without involvement of the coeliac axis or the superior mesenteric artery; T4, tumor involving coeliac axis or superior mesenteric artery; N0, no regional lymph node metastasis; N1, regional lymph node metastasis; M0, no distant metastasis; M1, distant metastasis.

with less sensitivity (95% in tumors greater than 3 cm, 81% in tumors 1–3 cm, and 50% in tumors less than 1 cm) (76). Subsequently, a qualified imaging procedure should be performed to clarify local resectability and the presence of potential metastasis. CT is now widely regarded as the “gold standard” for this purpose although based on local experience and availability, magnetic resonance imaging (MRI) with magnetic resonance cholangiopancreaticography (MRCP) or endoluminal ultrasound (EUS) may be applied without compromising sensitivity and specificity (evidence grade 3, recommendation C). CT diagnostic should be performed with a multidetector row CT (MDCT) with arterial and portal venous phases of contrast enhancement and a maximum section thickness of 3 mm. In this setting, MDCT will accurately predict respectability in 80% to 90% of cases (77–79), but sensitivity drops below 80% when tumors are smaller than 2 cm (80). It should be mentioned, however, that PDAC induces a heavy desmoplastic reaction in the surrounding stroma tissue

(81) that is difficult to distinguish from the actual tumor on CT imaging which may lead to the false positive diagnosis of irresectability. Using a cut-off value of 180° of vascular involvement on CT yielded a sensitivity of 84% and a specificity of 98% for unresectable disease in one study (82), but recent data suggest that greater tumor involvement of arteries is necessary to accurately identify arterial invasion (83). Furthermore, enlargement of lymph nodes on CT is a poor indicator for metastasis and irresectability (84). Finally, the sensitivity of CT imaging to detect hepatic metastasis varies depending on the size of the lesion, but is considered to range between 38% and 73% (76). MRI should be performed with a field volume of at least 1.5 T, with sections of 5 mm and T1 and T2 weighted as well as MRCP images. In this setting, MRI/MRCP yields similar accuracy in diagnosis and staging of PDAC than CT imaging and might be superior in the diagnosis of hepatic lesions (85– 87). As with endoscopic retrograde cholangio-pancreaticography (ERCP), a double-duct sign (obstruction of both the pancreatic and the bile ducts) strongly suggests pancreatic malignancies (88). Recent reports foster the hope that special MRI sequences may be superior to other imaging techniques in detecting early PDAC lesions (89). EUS is highly sensitive in the diagnosis of pancreatic cancer as well as in the evaluation of vascular involvement (90). It has a sensitivity of 95%, a specificity of 80%, a positive predictive value of 95%, and a negative predictive value of 80% for pancreatic tumors in experienced hands (91–93). It was reported to be superior to CT imaging in detecting small tumors (93,94), but less effective in assessing vascular involvement and distant metastasis (91,95). Furthermore, EUS is a sensitive (over 90%) and highly specific (over 90%) method to obtain fine-needle aspirates (FNA) for pathological analysis (96–99). In most cases, however, a histological verification before surgery is not necessary since any potentially malignant lesions of the pancreas should be resected (evidence level 2; recommendation B) (63). Histological verification should, therefore, only be sought if the result would influence further treatment, which is particularly important in the palliative setting. In order to complete staging procedures, a chest X-ray or thoracic CT scan should be performed to rule out potential pulmonary metastasis. Further diagnostic procedures that may be employed in specific clinical settings include ERCP, positron emission tomography with CT (PET-CT), and diagnostic laparoscopy. PET-CT does not play a role in the routine evaluation of pancreatic cancers, although it may be used in the context of equivocal radiographic findings, detection of occult metastasis, or in the case of recurrent disease (100,101). Similarly, diagnostic laparoscopy may detect peritoneal carcinosis and occult organ metastasis not previously visible on imaging studies. Early studies demonstrated a 15% to 25% incidence of such lesions when compared with preoperative imaging studies (102–104) leading to the routine employment of diagnostic laparoscopy in some centers. Advances in the accuracy of imaging techniques, however, reduced the incidence of newly diagnosed lesions detected by diagnostic laparoscopy to 5% to 15% (105–107). In addition, diagnostic laparoscopy

383

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 42.5 Criteria for Categorizing PDAC According to Respectability into Resectable, Borderline and Locally Irresectable Tumors as Well as Tumors with Distant Metastases
Resectable Venous involvement Splenic vein Portal vein-SMV confluence Arterial involvement Superior mesenteric artery (SMA) Coeliac axis Common hepatic artery Gastroduodenal artery Involvement of neighboring organs Duodenum Stomach Colon Spleen Kidney / adrenal gland Spine Distant metastasis ✓✗ ✗ Borderline ✓✗ ✓ ✗ (short segment) Local irresectability ✓✗ ✓(long segment and/or major tumor thrombosis) ✓(circumferential/long segment) ✓ ✓(encasement) ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓ ✗ Distant metastasis ✓✗ ✓✗

✗ ✗ ✗ ✓✗ ✓✗ ✓ ✗ (pylorus or antrum) ✗ ✗ ✗ ✗ ✗

✓ ✗ (not circumferential) (✗) ✓ ✗ (short segment) ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✗ ✗

✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓✗ ✓

✗: must be absent. ✓ : must be present. ✓✗ : may or may not be present (in brackets the maximum extent of involvement). For locally irresectabe tumors: one ✓-criterion is enough to classify the tumor as locally irresectable. Abbreviations: SMV: superior mesenteric vein.

History physical exam

Labs transabd. US

MDCT or MRI/MRCP or EUS

(ERCP) (PET) (diag. laparoscopy)

Resectable

Borderline

Locally irresectable

Metastatic disease

Figure 42.1 Preoperative diagnosis pathway in patients with pancreatic cancer. The aim is to determine surgical respectability. Abbreviations: US, ultrasonography; MDCT, multidetector-computed tomography; MRI, magnetic resonance imaging; MRCP, magnetic resonance cholangiopancreaticography; EUS, endoluminal ultrasound; ERCP, endoscopic retrograde cholangiopancreaticography; PET, positron emission tomography.

384

ADENOCARCINOMA OF THE PANCREAS
cannot clarify the question of local resectability, which can ultimately be assessed only by laparotomy and exploration. Therefore, diagnostic laparoscopy should be confined to specific clinical settings in which tumor spread is likely but could not been verified by routine procedures (evidence grade IV, recommendation D). These include (1) tumors larger than 5 cm, (2) high levels of CA19-9 preoperatively (over 680 U/ml) (108), and (3) tumors of the body and tail of the pancreas (107). ERCP is not regularly used to diagnose pancreatic cancer since its accuracy does not exceed that of less invasive imaging techniques, its association with a significant complication rate of 5% to 10% (109,110) and its failure to visualize the tumor and its relation to surrounding organs. Furthermore, brush biopsies taken from suspect parts of the duct system for cytological analysis have a low sensitivity (59.8%), albeit high specificity (98.1%) (111), and are thus inferior to EUS-FNA (see above). ERCP should therefore be confined to clinical settings in which a direct visualization of the ampullary region is necessary (e.g., in suspected intraductal papillary mucinous neoplasm, IPMN, in which mucin excretion from the ampulla is almost diagnostic) or in which a preoperative decompression of the common bile duct is necessary (see below). Finally, a thorough assessment of any relevant co-morbidity (cardiac, pulmonary, renal, etc.) must precede any decision concerning further treatment. latter are difficult to extract during surgery and may harm the bile duct (Evidence grade IV; recommendation D).

management
Management of PDAC depends on the clinical tumor stage (Fig. 42.2): 1. Patients with resectable tumors should undergo surgery as soon as possible followed by adjuvant chemotherapy. 2. Patients with borderline tumors should undergo exploration and resection if possible. If resection is not possible the decision to perform a (double) bypass operation (gastrojejunostomy and/or hepaticojejunostomy) must be made. The patient can then either be enrolled in a neoadjuvant treatment protocol to achieve respectability or undergo palliative care. 3. Patients with locally advanced, unresectable tumors can be enrolled in a neoadjuvant treatment trial or be treated palliatively. 4. Patients with distant metastasis should be offered palliative treatment. The rational to perform surgical resection whenever possible is the vastly improved prognosis and quality of life following a macroscopic complete (R0/R1) removal of the tumor (see below). It should be mentioned that age should not be a contraindication for surgery since several studies have demonstrated comparable outcomes between young and old patient cohorts (evidence grade 4, recommendation D) (121–123).

preoperative biliary drainage
Whether or not preoperative biliary drainage (PBD) by ERCP and stenting or by percutaneous cholangio-drainage (PTCD) should be sought prior to resection remains a matter of debate. While some studies have reported increased morbidity rates following pancreaticoduodenectomy after PBD (112) others could not verify this association except an increase in the rate of postoperative wound infections (113–116). However, PBD was significantly associated with an increase in the rate of biliary infections (112,117), and this seemed to directly translate into an increase in overall morbidity (118). However, most of these studies have been retrospective. A metaanalysis of five randomized controlled trials and 18 cohort studies evaluating PBD versus no PBD in jaundiced patients undergoing surgical resection showed no difference in overall mortality, but a significantly reduced overall morbidity in the non-PBD group (114). Interestingly, in the randomized controlled trials postoperative morbidity was actually significantly lower in the PBD group, a difference, however, that was reversed by the significant number of preoperative complications associated with PBD (around 27%) (119,120). A recent multicenter, randomized controlled trial confirmed these findings insofar as the early-surgery group (no PBD) had significantly less serious complications than the PBD group while overall mortality did not significantly differ (269). In this study both pre- and postoperative morbidity was higher in the PBD group. PBD should therefore be avoided prior to surgery (evidence 1a, recommendation A). PBD may be used in specific clinical situations in which immediate surgery is not feasible (e.g., during neoadjuvant treatment) or in the case of cholangitis. Furthermore, most surgeons would recommend to have plastic rather than metal stents placed in these settings, since the

surgical resection
The objective of surgical management of PDAC is the macroscopic complete resection of the primary tumor, complete regional lymphadenectomy, and reconstruction of the gastrointestinal tract. Depending on the localization of the tumor this goal can be achieved either by left (distal) pancreatectomy, pancreatoduodenectomy (PD) or by total pancreatectomy. A standard PD (Kausch–Whipple operation) (124) involves resection of the pancreatic head, the duodenum, the distal common bile duct, the distal stomach, the gall bladder, and regional lymph nodes. A variant of this standard PD procedure preserves the stomach and is thus termed pylorus-preserving pancreatoduodenectomy (ppPD, Traverso-Longmire operation) (125). Resections as well as the extent of lymphadenectomy (standard and extended) have been defined in order to allow for better comparison of data (126,127). For details concerning the surgical resection procedures see the chapter on pancreatic resection in this book. Standard Versus Pylorus-Preserving Pancreatoduodenectomy The question whether standard or pylorus-preserving pancreatoduodenectomy (PD vs. ppPD) is superior in the treatment of PDAC has been a matter of debate for some time. The main areas of concern were the oncological completeness of the ppPD procedure as well as the potential physiological side

385

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
I Resectable II Borderline III Local irresectability IV Metastatic disease

Laparotomy

Histology
-CT/US/EUS: biopsy

Histology
-CT/US/EUS: biopsy

Resection
-PD -ppPD -Left resection -total pancreatecotmy

Irresectable

Neoadjuvant (R)CTx

Palliative CTx supportive care

Adjuvant CTx

(Double bypass)

Resectable

Irresectable

Neoadjuvant (R)CTx

Palliative CTx supportive care

Laparotomy

Figure 42.2 Treatment of PDAC dependent on clinical tumor stage. Abbreviations: CTx, chemotherapy; RCTx, radio-chemotherapy; EUS, endoluminal ultrasound; US, ultrasound; PD, pancreaticoduodenectomy; ppPD, pylorus-preserving pancreaticoduodenectomy.

effects of the distal gastrectomy performed during PD. A number of series have compared the two procedures; however, results were inconclusive or conflicting (128–131). A recent meta-analysis of seven randomized controlled trials including a total of 496 patients, however, found no difference between the two procedures concerning in-hospital mortality, overall survival, and overall morbidity. Merely intra-operative blood loss and operation time were significantly reduced in the ppPD procedure (132). Although the authors pointed out that the analyzed studies were of clinical and methodological heterogeneity, currently both procedures seem to be equally appropriate in treating tumors of the pancreatic head (evidence grade Ia, recommendation A). Pancreatic–Enteric Anastomosis One of the most common and feared complications of pancreatic surgery is the development of a pancreatic fistula which was historically associated with mortality rates of up to 40% (133–135). Mortality rates have dropped significantly with the advent of CT-guided drainage, modern antibiotic regimes, and nutritional support, but the question which anastomotic techniques are associated with the lowest possible leakage rate remains.

Several studies have investigated whether pancreaticogastrostomy might be superior to the traditional pancreaticojejunostomy. A recent meta-analysis of three randomized controlled trials and 13 nonrandomized observational studies found no significant differences between the two procedures regarding overall postoperative complications, pancreatic fistula, intra-abdominal fluid collection, or mortality when analyzing the randomized controlled trials (136). On the contrary, analysis of the 13 nonrandomized observational studies showed significant results in favor of pancreaticogastrostomy, a result which was most likely due to publication bias. Therefore, neither pancreaticogastrostomy nor pancreaticojejunostomy seems to be superior in reconstruction after PD (evidence grade Ia; recommendation A). Further areas of research concern technical aspects of the pancreaticojejunostomy. Specifically the question whether the anastomosis should be performed with invagination (end of the pancreas invaginated into either the end or the side of the jejunum) or whether a duct-to-mucosa approach leads to less pancreatic fistulas has been investigated. All have been proven to be safe and feasible with no clear advantage of one over the other (137,138). Similarly, the use of anastomotic stents has failed to affect fistula rates (137,139).

386

ADENOCARCINOMA OF THE PANCREAS
Most pancreatic surgeons would argue that the texture of the pancreatic gland remnant influences postoperative fistula rates with a soft pancreatic tissue promoting pancreatic leakage. PDAC with its strong desmoplastic reaction would therefore favor anastomotic stability in comparison to ampullary or primary duodenal neoplasms (evidence grade IV). Extended Lymphadenectomy Given the high incidence of lymph node metastasis in PDAC the idea of an extended lymphadenectomy to improve prognosis seems reasonable. Standard lymphadenectomy during PD or ppPD comprises of the lymph nodes on the right side of the hepatoduodenal ligament, along the hepatic artery, the portal vein, and the cranial part of the SMV. Figure 42.3 gives an overview of the lymph nodes involved in PDAC according to the Japanese Pancreatic Society classification (140). According to this classification, standard lymphadenectomy should remove the lymph node stations 12, 13, 17, as well as partly 14. Studies comparing standard lymphadenectomy to extended lymphadenectomy procedures have not been standardized in respect to the extent and localization of lymph node groups resected. Three randomized controlled trials have been published on this question (141–143) and all of them have failed to show a significant benefit on survival from extended lymphadenectomy. One study, however, reported a significantly higher postoperative morbidity rate, mainly dumping syndrome and debilitating diarrhoea, in the extended lymphadenectomy group (141). A recent long-term (5 year) follow-up of this study seems to confirm the initial data. A nonsignificant trend toward improved survival in the radical lymphadenectomy group can be accounted for by the higher incidence of microscopically margin positive resections in the standard resection group (144). These results were confirmed in a recent metanalysis (145). Therefore, extended lymphadenectomy cannot be recommended in the surgical treatment of PDAC (evidence grade Ia, recommendation A). Extended Resections Partial resection of the portal vein, the SMV, and/or the venous confluence is by now a well-established procedure in specialized centers. It is indicated if it results in complete macroscopic tumor resection (R0/R1) since R-status is an important prognostic factor (see below). In experienced hands it does not seem to increase operative morbidity or mortality (146–148) and has been shown to result in similar long-term survival rates compared to standard surgical resections (147,149–151). Technical options include primary end-to-end anastomosis, reconstruction with a vein patch, autologous interposition graft, or with an alloplastic vascular graft (152). The success of major venous resection has lead to even more aggressive resections involving the arterial system as well. Studies showing a benefit for such extended resections, however, are sparse and consists mostly of small single-center cohorts and are at present far from being an established treatment option (148,153–157). In these studies, mortality rates between 0% and 17% (i.e., higher than during standard pancreatic resections), with median survival rates between 12.2 and 22 months were reported. The improved median survival in these patients in comparison to palliative therapy has led some surgeons to advocate for arterial resection in selected cases. Reconstruction after hepatic artery resection can be achieved by reinsertion of the artery into the abdominal aorta. In the case of pancreatic left resections, reconstruction may not be necessary if the retrograde flow along the pancreaticoduodenal and gastroduodenal arteries is sufficient for liver perfusion. The SMA may be reconstructed using an endto-end anastomosis or a venous or synthetic graft. The complete resection of distant metastasis has been reported only in a limited number of highly selected patients at specialized centers and did not improve median survival compared to palliative therapy alone (median survival 5.9–13.8 months) (158,159). Improved survival rates in patients undergoing complete surgical resection have raised the interest in downstaging pancreatic tumors that seem irresectable at presentation by means of neoadjuvant therapy.

neoadjuvant treatment
Currently, no randomized controlled trial comparing neoadjuvant therapy (radiochemotherapy or chemotherapy alone) with direct resection for locally advanced pancreatic cancer exists. A number of phase I/II trials or retrospective studies, however, have demonstrated that neoadjuvant treatment is safe and may result in down-sizing in a number of cases resulting in resection rates as high as 51% for tumors that deemed unresectable at presentation (160–162). These patients might benefit from the improved prognosis of an R0/R1 resection, although data to confirm this notion are lacking. Due to the lack of evidence, however, neoadjuvant treatment for pancreatic cancer should be confined to clinical trials. In this context, it should be pointed out that specialized, high-volume centers have markedly improved resection rates compared to low-volume centers (over 50%), i.e., patients that might be classified as unresectable in one hospital might undergo successful resection at a specialized center (163,164).

mortality and morbidity
Mortality after pancreatic resection has decreased dramatically from over 30% in the 1970s to below 4% in specialized centers today (165–176). At the same time, the hospital stay decreased from a median of 16 days to currently 8 to 10 days at specialized units (177). Several studies have stressed the significant improvement in mortality in high-volume centers compared to low-volume hospitals (see Table 42.6 for a selected list). Long-term follow-up studies have demonstrated that this translates into an increased overall survival (171,178). This volume–outcome relationship seems to be proportional to the number of pancreatic resections performed (179). In highvolume centers more patients die from systemic than from surgical complications (180). Therefore, strong evidence exists that pancreatic surgery should be performed at specialized high-volume centers (evidence grade II, recommendation B). While mortality decreased significantly within the last years, morbidity rates remain high and range between 30% and 40% (181,182). As for mortality, morbidity rates seem to be significantly reduced in high-volume centers (183). Most

387

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Group 1 2 3 4 5 6 7 8 9 10 11 12 12h 12a1 12a2 12p1 12p2 12b1 12b2 12c 13 13a 13b 14 14a 14b 14c 14d 15 16 16a1 16a2 16b1 16b2 17 17a 17b 18

Lymph nodes Right cardial Left cardial Along the lesser curvature of the stomach Along the greater curvature of the stomach Suprapyloric Infrapyloric Along the left gastric artery Along the common hepatic artery Around the celiac artery Splenic hilum Along the splenic artery In the hepatoduodenal ligament Hepatic hilum Superior to hepatic artery Inferior to hepatic artery Superior to portal vein Inferior to portal vein Superior to bile duct Inferior to bile duct Around cystic duct Posterior pancreatoduodenal nodes Superior to ampulla of Vater Inferior to ampulla of Vater Along the superior mesenteric artery Origin of the SMA Origin of the pancreatoduodenal artery Origin of the middle colic artery Origin of the jejunal arteries Along the middle colic artery; Around the abdominal aorta Above the origin of the coeliac axis Between coeliac axis and left renal artery Between left renal artery and IMA Between IMA and aortic bifurcation On the anterior surface of the pancreatic head Superior to ampulla of Vater Inferior to ampulla of Vater Along the inferior margin of the pancreatic body/tail

Figure 42.3 Japanese Pancreatic Society lymph node groups (140). Only the lymph node groups relevant for pancreatic surgery are shown in the figure, while the table lists all abdominal lymph node groups. 13 and 17 are first-order lymph nodes. 6, 8, 12 and partly 9 are second-order lymph nodes. 10, 11, 15, 16, 18 and partly 9 are third-order lmyph nodes. Abbreviations: SMA, superior mesenteric artery; IMA, inferior mesenteric artery. Opaque numbers indicate posteriorly positioned lymph nodes.

388

ADENOCARCINOMA OF THE PANCREAS
Table 42.6 Comparison of In-Hospital Mortality Between High- and Low-Volume Centers. Criteria for High- and Low-Volume Centers Vary Between Different Studies and are Listed as Cases Per Year
Cases per year Year Netherlands USA USA California and Florida, USA New York, U.S. Ontario Maryland, USA USA Maryland, USA Maryland, USA 1994–1998 1994–1999 1992–1995 1988–1998 1984–1991 1998–1995 1984–1995 1984–1993 1988–1993 1990–1995 Low-volume <5 <1 <1 ≤1 <10 <3 <20 1–5 ≤1 <5 High-volume ≥25 >16 >5 ≥10 >81 >6 ≥20 >11 >4 >20 In-hospital mortality (%) Low-volume 16 16.3 16 9.5 21.8 14.4 14.2 12.9 19 19 High-volume 0.86 3.8 4 3.3 4 3.4 1.8 5.8 2.2 1 References (165) (170) (169) (172) (173) (174) (167) (175) (168) (176)

complications are surgical but frequently they can be handled by pharmaceutical, radiological, and/or endoscopic interventions. Complications that require reoperation are still associated with high-mortality rates between 23% and 67% (184–186). The most common surgical complications in order of frequency are (the percentages given indicate the frequencies reported at two specialized high-volume centers and should be regarded as minimum numbers) (181,187):
● ● ● ● ●

been reported to alleviate the symptoms of DGE in up to 37% of patients (192). Postoperative Pancreatic Fistula (POPF) As for DGE, the rate of POPF in reports varies depending on the definition used and the experience of the team, but is approximately 5% in specialized centers (181,187). POPF is a feared complication since it is associated with a significant mortality rate of up to 28% secondary to sepsis and hemorrhage (189,193,194). An early sign of POPF may be a persistently elevated CRP level after postoperative day 4 (187). Surgical risk factors for POPF have been discussed in section Pancreatic-Enteric Anastomosis. Recently, POPF have been defined and graded according to their clinical relevance ( Table 42.7) (195). It should be mentioned that imaging studies are not necessary for diagnosis, although helpful in deciding further treatment options. Treatment should be tailored to the grade of POPF (all recommendations are grade D): Grade A fistulas have little or no clinical impact. The patient does well and can continue to be fed orally. Neither antibiotics nor total parenteral nutrition or somatostatin analogs are indicated. Most surgeons would leave the intraoperatively placed drains in situ and remove them slowly. Grade B fistulas require an adaption of the clinical management. Pancreatic fluid should be drained effectively. If this cannot be achieved via the intraoperative drains visualization by radiologic imaging and effective percutaneous drainage of any pancreatic fluid collection should be sought. Frequently, the patient is kept with nil by mouth and has to be supported by parenteral or enteral nutrition. Frequently, grade B fistulas are associated with fever and leucocytosis in which case antibiotics should be applied. Grade C fistulas constitute a worrisome clinical setting and demand immediate action since clinical stability of the patient is often compromised. Patients should be transferred to an ICU or at least intermediate care unit for better observation. The patient is kept with nil by mouth and is supported by parenteral or enteral nutrition. Intravenous antibiotics as well as somatostatin analogs are usually instituted. Radiologic imaging should be sought quickly and any peripancreatic fluid collection

Delayed gastric emptying (DGE) 9% to 15% Postoperative pancreatic fistula (POPF) 5% Wound infection 3% to 8% Intraabdominal abscess 1% to 4% Postpancreatectomy hemorrhage (PPH) 1% to 3%

The reported incidence rates for these complications vary widely since up to recently standardized definitions for these entities were lacking. In order to facilitate the reporting and comparison of morbidity data between different institutions, consensus definitions and classifications have been achieved recently and should be used in all future reports ( Table 42.7). Delayed Gastric Emptying (DGE) DGE has been reported in 14% to 70% of cases without application of a standardized definition (186,188,180), but seems to be considerably lower in specialized centers (9–15%) (181,187). DGE is graded according to an international consensus definition ( Table 42.7). A number of studies have demonstrated an association between DGE and other postpancreatectomy complications like intraabdominal abscesses or pancreatic fistulas (186). A relationship between DGE and the pylorus preserving Whipple procedure reported in one trial (189) could not be verified in consecutive randomized trials (128,190). However, the route of reconstruction, i.e., antecolic versus retrocolic gastrojejunostomy is associated with a significant lower DGE rate (191). Furthermore, early initiation of enteral feeding via an intraoperatively placed jejunal feeding tube after Whipple resections resulted in significantly higher rates of DGE as compared to patients not receiving early enteral nutrition (267). Intravenous erythromycin has

389

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 42.7 Consensus definitions and grading of three common complications following pancreatic surgery
Delayed Gastric Emptying (DGE) (261) Grade A Nasogastric tube NPO till POD Nausea/vomiting 4–7 days or reinsertion > POD 3 7 Present or absent Grade B 8–14 days or einsertion > POD 7 14 Present Grade C >14 days or reinsertion >POD 14 21 Present

Postoperative Pancreatic Fistula (POPF) (195) Definition: Output via an operatively placed drain (or a subsequently placed, percutaneous drain) of any measurable volume of drain fluid on or after postoperative day 3, with an amylase content greater than three times the upper normal serum value Grade A Clinical condition Specific treatmenta CT/US findings Persistent drainage >3 weeks Reoperation Death related to POPF Signs of infection Sepsis Well No Negative No No No No No Grade B Often well Yes/no Negative/positive Usually yes No No Yes No Grade C Ill appearing/bad Yes Positive Yes Yes Possible Yes Yes

Postpancreatectomy Hemorrhage (PPH) (200) Time of onset (early <24 hr, late >24 hr) Location: intraluminal vs. extraluminal Severity: mild vs. severe b Grade A Time of onset, location, severity and clinical impact Clinical condition Diagnostic consequence Therapeutic consequence Early, mild, intra-/extraluminal Grade B Early, severe, intra-/extraluminal or late, mild, intra-/extraluminal Grade C Late, severe, intra-/extraluminal

Well Observation, blood count, US, if necessary CT None

Often well, intermediate, very rarely Severely impaired, life-threatening life-threatening abservation, blood count, US, CT, Angiography, CT, endoscopy, angiography, endoscopy transfusion, ICU, therap. endoscopy, Localization of bleeding, embolization, relaparotomy angiography and embolization, endoscopy, relaparotomy, ICU

a

Partial (peripheral) or total parenteral nutrition, antibiotics, enteral nutrition, somatostatin analog and/or minimal invasive drainage; bFor the definition of mild and severe PPH see text. Abbreviations: delayed gastric emptying (DGE) (261), postoperative pancreatic fistula (POPF) (195), and postpancreatectomy hemorrhage (PPH) (200). CT, computed tomography; US, ultrasound; POD, postoperative day; NPO, nothing by mouth.

should be adequately drained percutaneously. Deteriorating clinical status (e.g., sepsis and organ dysfunction) may require re-exploration in order to attempt either to repair the site of leakage, convert to alternative means of pancreatic–enteric anastomosis (e.g., conversion of pancreaticojejunostomy to pancreaticogastrostomy), or to resect the pancreatic remnant. Intraabdominal Abscess Intraabdominal abscesses have been reported in 1% to 12% of patients following pancreatic resection (186,188,180,196). The usual cause is a persistent leak from the pancreatojejunostomy (see pancreatic fistula) or any other anastomosis. Intraabdominal abscesses following pancreatic surgery typically present as subhepatic or left subdiaphragmatic collections (182) and can usually be managed by CT-guided percutaneous drainage and intravenous antibiotic application (197). Postpancreatectomy Hemorrhage (PPH) PPH occurs in 1% to 8% of pancreatic resections and accounts for approximately 11% to 38% of overall mortality

(189,198,199). Again the variation seems to be in part due to a lack of a uniform definition, which has recently been achieved (Table 42.7) (200). PPH should henceforth be classified according to: (1) time of onset, (2) location and cause, and (3) severity. Early onset PPH (<24 hours) seems to be due to insufficient intraoperative hemostasis or an underlying coagulopathy while late PPH (>24 hours) may occur from: (a) erosion of (peripancratic) vessels due to pancreatic fistulas or intraabdominal drains, (b) gastric or duodenal ulcers, (c) anastomotic suture lines, (d) the resected area, (e) intraabdominal abscesses, or (f) the formation and rupture of pseudoaneurysms in the peripancreatic vasculature (201–203). It is important to distinguish between intraluminal and extraluminal PPH with markedly different clinical presentation. While the former is characterized by melena, hematemesis, or blood flow from the nasogastric tube, the latter is more often characterized by hemorrhage from the abdominal drains. The severity of the PPH should be determined clinically.

390

ADENOCARCINOMA OF THE PANCREAS
Mild PPH is defined as:






Small or medium volume blood loss, decrease in hemoglobin level by <3 g/dl Mild clinical impairment of the patient, no therapeutic consequence, or at most the need for noninvasive treatment with volume resuscitation or blood transfusions (2–3 units packed cells within 24 hr of end of operation or 1–3 units if later than 24 hr after operation) No need for reoperation or interventional angiographic embolization; endoscopic treatment of anastomotic bleeding may occur provided the other conditions apply

the analysis was, however, limited by the lack of any standardized definition for pancreatic fistulas between different trials (see above). As a result a meaningful subgroup analysis was not possible to elucidate which patients benefit from prophylactic application. Therefore, currently, prophylactic use of somatostatin analogs cannot be recommended indiscriminately.

adjuvant treatment
Two recent randomized controlled trails (CONKO-1 (210); ESPAC-1 (211)) as well as one meta-analysis (212) have demonstrated a significant benefit of adjuvant chemotherapy on outcome following pancreatic cancer resection compared to observation. Therefore, all patients undergoing curative resection should receive adjuvant chemotherapy (evidence grade 1b, recommendation A). The current data suggest that tumor-specific risk factors like grading or T-stage as well as age (even >80) are no contraindication for adjuvant chemotherapy (210,211). Poor postoperative performance status should be regarded as only a relative contraindication since in the trial of Oettle et al. even patients with Karnofsky index ≥50 were included (210). Currently, both 5-FU/folic acid (211) and gemcitabine (210) chemotherapy schemes are accepted after curative resection. A direct comparison of gemcitabine with 5-FU/folinic acid was recently conducted in the randomized, multicenter ESPAC-3 trial (270). Both groups had comparable overall survival rates after a median of 34.2 months of follow-up. While the treatment-related serious adverse events were significantly more frequent in the 5-FU/folinic acid group compared to gemcitabine, this did not translate into a difference in overall quality of life, which was comparable between the two groups. Based on the data from the currently available studies, adjuvant chemotherapy should be initiated within 6 weeks of surgery and be applied for 6 months. Table 42.8 lists all available randomized controlled trials of adjuvant treatment after resection of pancreatic carcinoma. Furthermore, based on subgroup analysis of the CONKO-1 trial (210), patients receiving a gemcitabine-based adjuvant chemotherapy demonstrated a significantly longer disease-free survival compared to observation alone after R1 resection of pancreatic cancer. Therefore, even in the setting of an R1 resection postoperative chemotherapy has been proven beneficial (evidence grade 1b, recommendation A). In contrast, the evidence for a combination of radiochemotherapy with chemotherapy is less clear, although this treatment is common in the United States (213). Most adjuvant trials have been underpowered and the randomized controlled ESPAC-1 (211) trial as well as one meta-analysis (212) failed to show any benefit for radiochemotherapy ( Table 42.8). Radiochemotherapy actually seemed to have a detrimental, albeit nonsignificant, effect on outcome. Therefore, currently, radiochemotherapy cannot be recommended as adjuvant treatment for PDAC outside trials (evidence grade 1b, recommendation A). Shortly data of the randomized controlled CAPRI trial (combination of radio-, chemo- and immunotherapy) (214) should be available to see if the promising data from previous reports on this combination can be confirmed (215).

Severe PPH on the other hand is defined as:






Large volume blood loss (drop of hemoglobin level by >3 g/dl) Clinically significant impairment (e.g., tachycardia, hypotension, oliguria, hypovolemic shock), need for blood transfusion (>3 units packed cells) Need for invasive treatment (interventional angiographic embolization, or relaparotomy

Time of onset, localization, and severity allow an exact classification of PPH into three grades and thus define further diagnostic and therapeutic measures (Table 42.7).

postoperative treatment
In addition to the management of complications, standard postoperative treatment should include nutritional support which is vital since many patients suffer from tumor cachexia going into the surgery and the return of normal bowl function may be delayed because of postoperative complications like DGE. However, early total parenteral nutrition after pancreatic resection was associated with an increased risk of complication (204). Enteral feeding is recommended after pancreatic resection whenever possible, however, early initiation of enteral feeding via an intraoperatively placed jejunal feeding tube after Whipple resections resulted in significantly higher rates of DGE as compared to patients not receiving early enteral nutrition in one study (267). Another study evaluating continuous enteral feeding to cyclic enteral nutrition after ppPD showed significant benefits for the latter with earlier commencement of normal oral diet and shorter hospital stay (205). Increasingly, components of fast-track surgery are applied to pancreatectomized patients including early postoperative mobilisation, oral diet, and modern forms of analgesia (206). Intraoperatively placed drains for example can be removed on the second postoperative day if inconspicuous to avoid bacterial contamination of the abdominal cavity (207,208). The benefit of prophylactic application of somatostatin analogs to prevent pancreas-specific complications has been controversial with some studies arguing for and some against it. A recent meta-analysis of 10 randomized trials including 1918 patients showed no benefit of somatostatin analogs to reduce mortality, but did show a significant reduction in morbidity in the treatment group (209). The external validity of

391

392
Treatment 20.1 15.5 39.7 30.0 0.71 (0.55, 0.92) p = 0.009) Comparison Median survival (months) Conclusion Significant increase in survival for CT (P = 0.009) in 289 eligible patients Overall 2-year survival rate Hazard ratio (95% CI), p-value Ref. (211) 13.4 6.9 34% 20.5% estimated overall 3-year survival NN 23.0(fluorouracil) vs. 23.6 (gemcitabine) 48.1% vs. 49.1% 0.94 (0.81–1.08) p=0.39 Median disease-free survival was (210) 13.4 months in the gemcitabine group (95% CI, 11.4–15.3) and 6.9 months in the control group (95% CI, 6.1–7.8; p = 0.001, log-rank). Estimated diseasefree survival at 3 and 5 years was 23.5% and 16.5% in the gemcitabine group, and 7.5% and 5.5% in the control group. (270) No significant difference in overall survival between the two groups. Treatment-related serious adverse events were more frequent in the fluorouracil group. Both groups had comparable overall QoL. 24.2 29.6 1.18 (0.84, 1.66) p = 0.33 12.8 12.4

Table 42.8 Randomized Controlled Trials for Adjuvant Treatment after Resection for Pancreatic Cancer

Year

Chemotherapy ESPAC-1a

1994–2000

CONKO-1

1998–2004

CTx vs. no CT 2x2 factorial design 2x(20 Gy in 10 fractions+500 mg/m2 5FU/FA days 1–3) (20 mg/m2 FA+425 mg/m2 5FU days 1–5) x 6 cycles CTx vs. OBS 6 cycles of gemcitabine every 4 weeks. Each cycle consisted of 3 weekly infusions of i.v. gemcitabine 1000 mg/m2 followed by a 1-week pause.

ESPAC-3

2000–2007

Takade et al.

1986–1992

Fluorouracil (425 mg/m2 Fluorouracil/ folinic acid intravenous bolus injection vs. given 1–5 days every 28 days) gemcitabine plus folinic acid (20 mg/m2, intravenous bolus injection ) or gemcitabine (1000 mg/m2 intravenous infusion once a week for 3 of every 4 weeks) for 6 months 6 mg/m2 mytomycin C day1 CTx vs. OBS +310 mg/m2 5FU days 1–5 and days 15–20 followed by 100 mg/m2 oral 5FU daily to recurrence CTx vs. OBS 17.7 10.4 30.6% 24.3%

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Bakkevold et al.

1984–1987

AMF (40 mg/m2 doxorubicin, 6 mg/m2 mytomycin C,500 mg/m2 5FU) once every 3 weeks for six courses

0.80 (0.42, 1.53) p = 0.48

(262) Significant survival benefit in gallbladder cancer patients. No difference in 158 eligible pancreatic cancer patients. No difference in 48 eligible ampullary cancer patients (263) Significant increase in median survival (23 vs. 11 mo, P1/40.02) in 60 pancreatic and ampullary patients combined (Continued)

Table 42.8 (Continued) Randomized Controlled Trials for Adjuvant Treatment after Resection for Pancreatic Cancer
Treatment RCTx vs. No RCTx 15.9 17.9 28.5 41.4 1.28 (0.99, 1.66) p = 0.053 Comparison Median survival (months) Overall 2-year survival rate Hazard ratio (95% CI), p-value Conclusion Non-significant decrease in survival for RCTx (P=0.053) in 289 patients. Ref. (211)

Year

Radiochemotherapy 1994–2000 ESPAC-1a

2x2 factorial design 2x (20 Gy in 10 fractions+500 mg/m2 5FU/FA days 1–3) (20 mg/m2 FA+425 mg/m2 5FU days 1–5) x 6 cycles RCTx vs. No RCTx 17.5 12.6 35.7 23.2 0.70 (0.46, 1.06) p = 0.088

EORTC

1987–1995

2x (20 Gy in 10 fractions+25 mg/kg 5FU/FA days 1–5)

(264)

ADENOCARCINOMA OF THE PANCREAS
(265) 0.82 (0.65, 1.03) p = 0.09

RTOG 97-04 RCTx + gemcit. vs. RCTx + 5FU 20.5 16.9

1998–2002

31% 22% 3-year survival

GITSG RCTx vs. OBS

1974–1982

CTx with either 5FU (i.v. 250 mg/m2 per day) or gemcitabine (1000 mg/m2 once per week) for 3 weeks prior to and for 12 weeks after chemoradiation. Chemoradiation with 50.4 Gy + 5FU (250 mg/m2 per day) 2x (20 Gy in 10 fractions + 500 mg/m2 5FU days 1–3) + weekly 5FU to recurrence 20.0 10.9

42% 15%

0.54 (0.27, 1.05) p = 0.035

Non significant increase in median survival (25 vs. 19 mo, P=0.21) in 207 eligible patients Non significant increase in median survival in 114 eligible pancreatic (17 vs. 13 mo, P = 0.099) The addition of gemcitabine to adjuvant fluorouracil-based chemoradiation was associated with a survival benefit for patients with resected pancreatic cancer, although this improvement was not statistically significant Significant increase in median survival (20 vs. 11 mo, p = 0.035) in 43 eligible patients

(266)

Trials addressing chemotherapy (CTx) and radiochemotherapy (RCTx) are listed. aThe ESPAC-1 trial had a factorial 2x2 design, meaning that patients were initially randomized into one of the four groups: observation, chemotherapy, radiochemotherapy, or a combination of both. Comparisons were made between (1) patients receiving chemotherapy (chemotherapy alone or radiochemotherapy + chemotherapy) and those not receiving it (observation or radiochemotherapy only) or (2) patients receiving radiation therapy (radiochemotherapy or radiochemotherapy + chemotherapy) and those not receiving it (observation or chemotherapy alone). Abbreviations: OBS, observation; NN, not reported.

393

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

prognosis
Curative surgical resection in patients with PDAC is the single most important factor influencing survival and the quality of life (216). Advanced age (>80 years) is no contraindication for resection which can be performed with similar mortality and morbidity rates and offers the same advantages than in the young (217). Although mortality rates for pancreatic surgery decreased significantly in the last years (see above) the overall 5-year survival rate for all patients with PDAC remains dismal (around 5%) (7,8). This number can be explained in part by the low incidence of resectable PDAC at presentation, but might also reflect a failure of adequate treatment. A recent nationwide study in the United States including over 9,500 patients with early, potentially resectable (stage I, T1/2N0M0) PDAC demonstrated that 71% of these patients did not undergo resection, mostly because they were not offered surgery (218). Following curative resection and adjuvant chemotherapy, patients have a 5-year survival rate of 20% to 25% and a median survival of 17 to 28 months (181,201,210,211,217,219– 223). Even after 5 years, recurrence does occur meaning that true long-term survivors are rare (224,225). These numbers, however, compare favorable to the devastating prognosis of patients with inoperable, locally advanced, or metastatic disease that have a median survival of 8 to 12 and 3 to 6 months, respectively (226). In certain subgroups (R0 resection, negative lymph node status, and specialized center), the 5-year survival might actually be as high as 40% (227). Multiple pre-, intra-, and postoperative factors influencing prognosis have been studied, but consistently negative resection margins (R0), negative lymph node status (N0), favorable tumor grading, primary tumor size under 2 cm, and absence of perineural invasion have been identified as favorable prognostic markers. Tumor size under 2 cm seems to improve median survival dramatically from an average of 14 months to 35.5 months and significantly improves 5-year survival (201,228–232). Similarly, a positive lymph node status (N1) has been demonstrated to be a negative prognostic factor (181,233–235), a finding confirmed by data from the ESPAC-1 trial (211,212). Finally, poor tumor grading significantly worsens median as well as overall survival in a number of studies (181,223,236), as does perineural invasion (237–239), although in the latter case a significant association could not be verified in other studies (231,240). Similarly, the results concerning the influence of microscopic tumor-free resection margins (R0) on prognosis have been ambiguous. While a number of studies have demonstrated a significant improvement in median as well as 5-year survival following R0 resection as compared to R1 resection (181,227,236,241) others have failed to confirm this association (240,242–244). This observation is confirmed by data from the ESPAC-1 trial in which only tumor grade and lymph node status were identified as significant prognostic factors (245). The reason for this apparent disparity is that the pathological handling and reporting of pancreatic specimens vary widely between different institutions and guidelines concerning this matter have not been standardized (246). Consequently, the rate of R0 resections dropped from 86% to 24%

after implementation of a standardized pathological examination protocol (246,247).

conclusion
All patients with PDAC should be evaluated for potential surgical resection given the improved prognosis and quality of life following resection compared to palliative treatment only. In case of questionable local respectability determined by the radiologist, an experienced pancreatic surgeon should evaluate the radiographic images to determine whether a tumor is resectable or not. Surgery can nowadays be carried out with low mortality and acceptable morbidity rates at high-volume specialized centers. Extended resections are justified if this results in macroscopic complete resection of the tumor. Surgical resection as the mainstay of treatment should be complemented by adjuvant chemotherapy in order to improve survival in patients suffering from PDAC.

references
1. Alexakis N, Halloran C, Raraty M, et al. Current standards of surgery for pancreatic cancer. Br J Surg 2004; 91(11): 1410–27. 2. Jemal A, Siegel R, Ward E, et al. Cancer Statistics. CA Cancer J Clin 2007; 57(1): 43–66. 3. Cancer Research UK: UK cancer mortality statistics for males. 2005. [Available from: http://info.cancerresearchuk.org/cancerstats/mortality/ males/?a=5441] 4. Cancer Research UK: UK cancer mortality statistics for females. 2005. [Available from: http://info.cancerresearchuk.org/cancerstats/mortality/ females/?a=5441] 5. SEER Cancer Statistics Review 1975–2005. Available from: http://seer. cancer.gov/cgi-bin/csr/1975_2005/search.pl#results 6. Cancer Research UK: UK cancer incidence statistics by country. 2005. Available from: http://info.cancerresearchuk.org/cancerstats/incidence/ site/?a=5441 7. Bramhall SR, Allum WH, Jones AG, et al. Treatment and survival in 13,560 patients with pancreatic cancer, and incidence of the disease, in the West Midlands: an epidemiological study. Br J Surg 1995; 82(1): 111–5. 8. Hedberg M, Borgström A, Genell S, Janzon L. Survival following pancreatic carcinoma: a follow-up study of all cases recorded in Malmö, Sweden, 1977–1991. Br J Surg 1998; 85(12): 1641–4. 9. Lankisch PG, Assmus C, Maisonneuve P, Lowenfels AB. Epidemiology of pancreatic diseases in Lüneburg County. A study in a defined german population. Pancreatology 2002; 2(5): 469–77. 10. Glade MJ. Food, nutrition, and the prevention of cancer: a global perspective. American Institute for Cancer Research/World Cancer Research Fund, American Institute for Cancer Research, 1997. Nutrition 1999; 15(6): 523–6. 11. Soler M, Chatenoud L, La Vecchia C, et al. Diet, alcohol, coffee and pancreatic cancer: final results from an Italian study. Eur J Cancer Prev 1998; 7(6): 455–60. 12. Ji BT, Chow WH, Gridley G, et al. Dietary factors and the risk of pancreatic cancer: a case-control study in Shanghai China. Cancer Epidemiol Biomarkers Prev 1995; 4(8): 885–93. 13. Chow WH, Gridley G, Nyrén O, et al. Risk of pancreatic cancer following diabetes mellitus: a nationwide cohort study in Sweden. J Natl Cancer Inst 1995; 87(12): 930–1. 14. Nöthlings U, Wilkens LR, Murphy SP, et al. Meat and fat intake as risk factors for pancreatic cancer: the multiethnic cohort study. J Natl Cancer Inst 2005; 97(19): 1458–65. 15. Michaud DS, Giovannucci E, Willett WC, et al. Dietary meat, dairy products, fat, and cholesterol and pancreatic cancer risk in a prospective study. Am J Epidemiol 2003; 157(12): 1115–25. 16. Michaud DS, Skinner HG, Wu K, et al. Dietary patterns and pancreatic cancer risk in men and women. J Natl Cancer Inst 2005; 97(7): 518–24. 17. Patel AV, Rodriguez C, Bernstein L, et al. Obesity, recreational physical activity, and risk of pancreatic cancer in a large U.S. Cohort. Cancer Epidemiol Biomarkers Prev 2005; 14(2): 459–66.

394

ADENOCARCINOMA OF THE PANCREAS
18. Michaud DS, Giovannucci E, Willett WC, et al. Physical activity, obesity, height, and the risk of pancreatic cancer. JAMA 2001; 286(8): 921–9. 19. Berrington de Gonzalez A, Sweetland S, Spencer E. A meta-analysis of obesity and the risk of pancreatic cancer. Br J Cancer 2003; 89(3): 519–23. 20. Berrington de González A, Spencer EA, Bueno-de-Mesquita HB, et al. Anthropometry, physical activity, and the risk of pancreatic cancer in the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomarkers Prev 2006; 15(5): 879–85. 21. Stolzenberg-Solomon RZ, Pietinen P, Taylor PR, et al. prospective study of medical conditions, anthropometry, physical activity, and pancreatic cancer in male smokers (Finland). Cancer Causes Control 2002; 13(5): 417–26. 22. Qiu D, Kurosawa M, Lin Y, et al. Overview of the epidemiology of pancreatic cancer focusing on the JACC Study. J Epidemiol 2005; 15(Suppl 2): S157–67. 23. Larsson SC, Permert J, Håkansson N, et al. Overall obesity, abdominal adiposity, diabetes and cigarette smoking in relation to the risk of pancreatic cancer in two Swedish population-based cohorts. Br J Cancer 2005; 93(11): 1310–5. 24. Coughlin SS, Calle EE, Patel AV, Thun MJ. Predictors of pancreatic cancer mortality among a large cohort of United States adults. Cancer Causes Control 2000; 11(10): 915–23. 25. Lin Y, Tamakoshi A, Kawamura T, et al. A prospective cohort study of cigarette smoking and pancreatic cancer in Japan. Cancer Causes Control 2002; 13(3): 249–54. 26. Yun YH, Jung KW, Bae J, et al. Cigarette smoking and cancer incidence risk in adult men: National Health Insurance Corporation Study. Cancer Detect Prev 2005; 29(1): 15–24. 27. Duell EJ, Holly EA, Bracci PM, et al. A population-based, case-control study of polymorphisms in carcinogen-metabolizing genes, smoking, and pancreatic adenocarcinoma risk. J Natl Cancer Inst 2002; 94(4): 297–306. 28. Miyasaka K, Kawanami T, Shimokata H, et al. Inactive aldehyde dehydrogenase-2 increased the risk of pancreatic cancer among smokers in a Japanese male population. Pancreas 2005; 30(2): 95–8. 29. Wang L, Miao X, Tan W, et al. Genetic polymorphisms in methylenetetrahydrofolate reductase and thymidylate synthase and risk of pancreatic cancer. Clin Gastroenterol Hepatol 2005; 3(8): 743–51. 30. Villeneuve PJ, Johnson KC, Mao Y, Hanley AJ. Environmental tobacco smoke and the risk of pancreatic cancer: findings from a Canadian population-based case-control study. Can J Public Health 2004; 95(1): 32–7. 31. Michaud DS, Giovannucci E, Willett WC, et al. Coffee and alcohol consumption and the risk of pancreatic cancer in two prospective United States cohorts. Cancer Epidemiol Biomarkers Prev 2001; 10(5): 429–37. 32. Lin Y, Tamakoshi A, Kawamura T, et al. Risk of pancreatic cancer in relation to alcohol drinking, coffee consumption and medical history: findings from the Japan collaborative cohort study for evaluation of cancer risk. Int J Cancer 2002; 99(5): 742–6. 33. Silverman DT, Schiffman M, Everhart J, et al. Diabetes mellitus, other medical conditions and familial history of cancer as risk factors for pancreatic cancer. Br J Cancer 1999; 80(11): 1830–7. 34. Brown LM. Epidemiology of alcohol-associated cancers. Alcohol 2005; 35(3): 161–8. 35. Ye W, Lagergren J, Weiderpass E, et al. Alcohol abuse and the risk of pancreatic cancer. Gut 2002; 51(2): 236–9. 36. Bansal P, Sonnenberg A. Pancreatitis is a risk factor for pancreatic cancer. Gastroenterology 1995; 109(1): 247–51. 37. Karlson BM, Ekbom A, Josefsson S, et al. The risk of pancreatic cancer following pancreatitis: an association due to confounding? Gastroenterology 1997; 113(2): 587–92. 38. Lowenfels AB, Maisonneuve P. Risk factors for pancreatic cancer. J Cell Biochem 2005; 95(4): 649–56. 39. Talamini G, Falconi M, Bassi C, et al. Incidence of cancer in the course of chronic pancreatitis. Am J Gastroenterol 1999; 94(5): 1253–60. 40. Howes N, Neoptolemos JP. Risk of pancreatic ductal adenocarcinoma in chronic pancreatitis. Gut 2002; 51(6): 765–6. 41. Chari ST. Detecting early pancreatic cancer: problems and prospects. Semin Oncol 2007; 34(4): 284–94. 42. Coughlin SS, Calle EE, Teras LR, et al. Diabetes mellitus as a predictor of cancer mortality in a large cohort of US adults. Am J Epidemiol 2004; 159(12): 1160–7. 43. Huxley R, Ansary-Moghaddam A, Berrington de González A, et al. TypeII diabetes and pancreatic cancer: a meta-analysis of 36 studies. Br J Cancer 2005; 92(11): 2076–83. 44. Stolzenberg-Solomon RZ, Graubard BI, Chari S, et al. Insulin, glucose, insulin resistance, and pancreatic cancer in male smokers. JAMA 2005; 294(22): 2872–8. 45. Wideroff L, Gridley G, Mellemkjaer L, et al. Cancer incidence in a population-based cohort of patients hospitalized with diabetes mellitus in Denmark. J Natl Cancer Inst 1997; 89(18): 1360–5. 46. Everhart J, Wright D. Diabetes mellitus as a risk factor for pancreatic cancer. A meta-analysis. JAMA 1995; 273(20): 1605–9. 47. McWilliams RR, Rabe KG, Olswold C, et al. Risk of malignancy in firstdegree relatives of patients with pancreatic carcinoma. Cancer 2005; 104(2): 388–94. 48. Klein AP, Brune KA, Petersen GM, et al. Prospective risk of pancreatic cancer in familial pancreatic cancer kindreds. Cancer Res 2004; 64(7): 2634–8. 49. Bartsch DK, Kress R, Sina-Frey M, et al. Prevalence of familial pancreatic cancer in Germany. Int J Cancer 2004; 110(6): 902–6. 50. Hemminki K, Li X. Familial and second primary pancreatic cancers: a nationwide epidemiologic study from Sweden. Int J Cancer 2003; 103(4): 525–30. 51. Brand RE, Lerch MM, Rubinstein WS, et al. Advances in counselling and surveillance of patients at risk for pancreatic cancer. Gut 2007; 56(10): 1460–9. 52. Tersmette AC, Petersen GM, Offerhaus GJ, et al. Increased risk of incident pancreatic cancer among first-degree relatives of patients with familial pancreatic cancer. Clin Cancer Res 2001; 7(3): 738–44. 53. McFaul CD, Greenhalf W, Earl J, et al. Anticipation in familial pancreatic cancer. Gut 2006; 55(2): 252–8. 54. Pogue-Geile KL, Chen R, Bronner MP, et al. Palladin mutation causes familial pancreatic cancer and suggests a new cancer mechanism. PLoS Med 2006; 3(12): e516. 55. Murphy KM, Brune KA, Griffin C, et al. Evaluation of candidate genes MAP2K4, MADH4, ACVR1B, and BRCA2 in familial pancreatic cancer: deleterious BRCA2 mutations in 17%. Cancer Res 2002; 62(13): 3789–93. 56. Hahn SA, Greenhalf B, Ellis I, et al. BRCA2 germline mutations in familial pancreatic carcinoma. J Natl Cancer Inst 2003; 95(3): 214–21. 57. Couch FJ, Johnson MR, Rabe KG, et al. The prevalence of BRCA2 mutations in familial pancreatic cancer. Cancer Epidemiol Biomarkers Prev 2007; 16(2): 342–6. 58. Lowenfels AB, Maisonneuve P, Cavallini G, et al. Pancreatitis and the risk of pancreatic cancer. International Pancreatitis Study Group. N Engl J Med 1993; 328(20): 1433–7. 59. Whitcomb DC, Gorry MC, Preston RA, et al. Hereditary pancreatitis is caused by a mutation in the cationic trypsinogen gene. Nat Genet 1996; 14(2): 141–5. 60. Witt H, Luck W, Hennies HC, et al. Mutations in the gene encoding the serine protease inhibitor, Kazal type 1 are associated with chronic pancreatitis. Nat Genet 2000; 25(2): 213–6. 61. Maisonneuve P, FitzSimmons SC, Neglia JP, et al. Cancer risk in nontransplanted and transplanted cystic fibrosis patients: a 10-year study. J Natl Cancer Inst 2003; 95(5): 381–7. 62. Neglia JP, FitzSimmons SC, Maisonneuve P, et al. The risk of cancer among patients with cystic fibrosis. Cystic Fibrosis and Cancer Study Group. N Engl J Med 1995; 332(8): 494–9. 63. Adler G, Seufferlein T, Bischoff SC, et al. [S3-Guidelines “Exocrine pancreatic cancer” 2007]. Z Gastroenterol 2007; 45(6): 487–523. 64. Solcia E, Capella C, Kloppel G. Tumors of the Pancreas. Washington, DC: Armed Forces Institute of Pathology, 1997. 65. Björnsson E, Ismael S, Nejdet S, Kilander A. Severe jaundice in Sweden in the new millennium: causes, investigations, treatment and prognosis. Scand J Gastroenterol 2003; 38(1): 86–94. 66. Reisman Y, Gips CH, Lavelle SM, Wilson JH. Clinical presentation of (subclinical) jaundice--the Euricterus project in The Netherlands.

395

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
United Dutch Hospitals and Euricterus Project Management Group. Hepatogastroenterology 1996; 43(11): 1190–5. Watanabe I, Sasaki S, Konishi M, et al. Onset symptoms and tumor locations as prognostic factors of pancreatic cancer. Pancreas 2004; 28(2): 160–5. Pannala R, Basu A, Petersen GM, Chari ST. New-onset diabetes: a potential clue to the early diagnosis of pancreatic cancer. Lancet Oncol 2009; 10(1): 88–95. Wang F, Herrington M, Larsson J, Permert J. The relationship between diabetes and pancreatic cancer. Mol Cancer 2003; 2: 4. Chari ST, Leibson CL, Rabe KG, et al. Probability of pancreatic cancer following diabetes: a population-based study. Gastroenterology 2005; 129(2): 504–11. Gambill EE. Pancreatitis associated with pancreatic carcinoma: a study of 26 cases. Mayo Clin Proc 1971; 46(3): 174–7. Homma T, Tsuchiya R. The study of the mass screening of persons without symptoms and of the screening of outpatients with gastrointestinal complaints or icterus for pancreatic cancer in Japan, using CA19-9 and elastase-1 or ultrasonography. Int J Pancreatol 1991; 9: 119–24. Kim J, Lee KT, Lee JK, et al. Clinical usefulness of carbohydrate antigen 19-9 as a screening test for pancreatic cancer in an asymptomatic population. J Gastroenterol Hepatol 2004; 19(2): 182–6. Ritts RE, Nagorney DM, Jacobsen DJ, et al. Comparison of preoperative serum CA19-9 levels with results of diagnostic imaging modalities in patients undergoing laparotomy for suspected pancreatic or gallbladder disease. Pancreas 1994; 9(6): 707–16. Schlieman MG, Ho HS, Bold RJ. Utility of tumor markers in determining resectability of pancreatic cancer. Arch Surg 2003; 138(9): 951–5; discussion 955–6. Balci NC, Semelka RC. Radiologic diagnosis and staging of pancreatic ductal adenocarcinoma. Eur J Radiol 2001; 38(2): 105–12. McCarthy MJ, Evans J, Sagar G, Neoptolemos JP. Prediction of resectability of pancreatic malignancy by computed tomography. Br J Surg 1998; 85(3): 320–5. Chong M, Freeny PC, Schmiedl UP. Pancreatic arterial anatomy: depiction with dual-phase helical CT. Radiology 1998; 208(2): 537–42. Vedantham S, Lu DS, Reber HA, Kadell B. Small peripancreatic veins: improved assessment in pancreatic cancer patients using thin-section pancreatic phase helical CT. AJR 1998; 170(2): 377–83. Bronstein YL, Loyer EM, Kaur H, et al. Detection of small pancreatic tumors with multiphasic helical CT. AJR 2004; 182(3): 619–23. Erkan M, Michalski CW, Rieder S, et al. The activated stroma index is a novel and independent prognostic marker in pancreatic ductal adenocarcinoma. Clin Gastroenterol Hepatol 2008; 6(10): 1155–61. Lu DS, Reber HA, Krasny RM, et al. Local staging of pancreatic cancer: criteria for unresectability of major vessels as revealed by pancreaticphase, thin-section helical CT. AJR 1997; 168(6): 1439–43. Li H, Zeng MS, Zhou KR, et al. Pancreatic adenocarcinoma: the different CT criteria for peripancreatic major arterial and venous invasion. J Comput Assist Tomogr 2005; 29(2): 170–5. Roche CJ, Hughes ML, Garvey CJ, et al. CT and pathologic assessment of prospective nodal staging in patients with ductal adenocarcinoma of the head of the pancreas. AJR; 180(2): 475–80. Ichikawa T, Haradome H, Hachiya J, et al. Pancreatic ductal adenocarcinoma: preoperative assessment with helical CT versus dynamic MR imaging. Radiology 1997; 202(3): 655–62. Megibow AJ, Zhou XH, Rotterdam H, et al. Pancreatic adenocarcinoma: CT versus MR imaging in the evaluation of resectability— report of the Radiology Diagnostic Oncology Group. Radiology 1995; 195(2): 327–32. Arslan A, Buanes T, Geitung JT. Pancreatic carcinoma: MR, MR angiography and dynamic helical CT in the evaluation of vascular invasion. Eur J Radiol 2001; 38(2): 151–9. Fayad LM, Mitchell DG. Magnetic resonance imaging of pancreatic adenocarcinoma. Int J Gastrointest Cancer 2001; 30(1–2): 19–25. Reiser-Erkan C, Gaa J, Kleeff J. T1 pancreatic cancer with lymph node metastasis and perineural invasion of the celiac trunk. Clin Gastroenterol Hepatol 2008; 6(11): e41–2. 90. Rösch T, Lightdale CJ, Botet JF, et al. Localization of pancreatic endocrine tumors by endoscopic ultrasonography. N Engl J Med 1992; 326(26): 1721–6. 91. Rösch T, Dittler HJ, Strobel K, et al. Endoscopic ultrasound criteria for vascular invasion in the staging of cancer of the head of the pancreas: a blind reevaluation of videotapes. Gastrointest Endosc 2000; 52(4): 469–77. 92. Volmar KE, Vollmer RT, Routbort MJ, Creager AJ. Pancreatic and bile duct brushing cytology in 1000 cases: review of findings and comparison of preparation methods. Cancer 2006; 108(4): 231–8. 93. Mertz HR, Sechopoulos P, Delbeke D, Leach SD. EUS, PET, and CT scanning for evaluation of pancreatic adenocarcinoma. Gastrointest Endosc 2000; 52(3): 367–71. 94. Howard TJ, Chin AC, Streib EW, et al. Value of helical computed tomography, angiography, and endoscopic ultrasound in determining resectability of periampullary carcinoma. Am J Surg 1997; 174(3): 237–41. 95. Aslanian H, Salem R, Lee J, et al. EUS diagnosis of vascular invasion in pancreatic cancer: surgical and histologic correlates. Am J Gastroenterol 2005; 100(6): 1381–5. 96. Raut CP, Grau AM, Staerkel GA, et al. Diagnostic accuracy of endoscopic ultrasound-guided fine-needle aspiration in patients with presumed pancreatic cancer. J Gastrointest Surg 2003; 7(1): 118–26; discussion 127–8. 97. Agarwal B, Abu-Hamda E, Molke KL, et al. Endoscopic ultrasoundguided fine needle aspiration and multidetector spiral CT in the diagnosis of pancreatic cancer. Am J Gastroenterol 2004; 99(5): 844–50. 98. Varadarajulu S, Tamhane A, Eloubeidi MA. Yield of EUS-guided FNA of pancreatic masses in the presence or the absence of chronic pancreatitis. Gastrointest Endosc 2005; 62(5): 728–36; quiz 751, 753. 99. Klapman JB, Chang KJ, Lee JG, Nguyen P. Negative predictive value of endoscopic ultrasound in a large series of patients with a clinical suspicion of pancreatic cancer. Am J Gastroenterol 2005; 100(12): 2658–61. 100. Shreve PD. Focal fluorine-18 fluorodeoxyglucose accumulation in inflammatory pancreatic disease. Eur J Nucl Med 1998; 25(3): 259–64. 101. Sendler A, Avril N, Helmberger H, et al. Preoperative evaluation of pancreatic masses with positron emission tomography using 18F-fluorodeoxyglucose: diagnostic limitations. World J Surg 2000; 24(9): 1121–9. 102. Warshaw AL, Tepper JE, Shipley WU. Laparoscopy in the staging and planning of therapy for pancreatic cancer. Am J Surg 1986; 151(1): 76–80. 103. Warshaw AL, Gu ZY, Wittenberg J, Waltman AC. Preoperative staging and assessment of resectability of pancreatic cancer. Arch Surg 1990; 125(2): 230–3. 104. Conlon KC, Dougherty E, Klimstra DS, et al. The value of minimal access surgery in the staging of patients with potentially resectable peripancreatic malignancy. Ann Surg 1996; 223(2): 134–40. 105. Friess H, Kleeff J, Silva JC, et al. The role of diagnostic laparoscopy in pancreatic and periampullary malignancies. J Am Coll Surg 1998; 186(6): 675–82. 106. Barreiro CJ, Lillemoe KD, Koniaris LG, et al. Diagnostic laparoscopy for periampullary and pancreatic cancer: what is the true benefit? J Gastrointest Surg 2002; 6(1): 75–81. 107. Pisters PW, Lee JE, Vauthey JN, et al. Laparoscopy in the staging of pancreatic cancer. Br J Surg 2001; 88(3): 325–37. 108. Katz A, Hanlon A, Lanciano R, et al. Prognostic value of CA 19-9 levels in patients with carcinoma of the pancreas treated with radiotherapy. Int J Radiat Oncol Biol Phys 1998; 41(2): 393–6. 109. Mallery JS, Baron TH, Dominitz JA, et al. Complications of ERCP. Gastrointest Endosc 2003; 57(6): 633–8. 110. Freeman ML. Adverse outcomes of endoscopic retrograde cholangiopancreatography: avoidance and management. Gastrointest Endosc Clin N Am 2003; 13(4): 775–98, xi. 111. Stewart CJ, Mills PR, Carter R, et al. Brush cytology in the assessment of pancreatico-biliary strictures: a review of 406 cases. J Clin Pathol 2001; 54(6): 449–55. 112. Povoski SP, Karpeh MS, Conlon KC, et al. Association of preoperative biliary drainage with postoperative outcome following pancreaticoduodenectomy. Ann Surg 1999; 230(2): 131–42.

67.

68.

69. 70.

71. 72.

73.

74.

75.

76. 77.

78. 79.

80. 81.

82.

83.

84.

85.

86.

87.

88. 89.

396

ADENOCARCINOMA OF THE PANCREAS
113. Pisters PW, Hudec WA, Hess KR, et al. Effect of preoperative biliary decompression on pancreaticoduodenectomy-associated morbidity in 300 consecutive patients. Ann Surg 2001; 234(1): 47–55. 114. Sewnath ME, Karsten TM, Prins MH, et al. A meta-analysis on the efficacy of preoperative biliary drainage for tumors causing obstructive jaundice. Ann Surg 2002; 236(1): 17–27. 115. Sohn TA, Yeo CJ, Cameron JL, et al. Do preoperative biliary stents increase postpancreaticoduodenectomy complications? J Gastrointest Surg 2000; 4(3): 258–67; discussion 267–8. 116. Martignoni ME, Wagner M, Krähenbühl L, et al. Effect of preoperative biliary drainage on surgical outcome after pancreatoduodenectomy. Am J Surg 2001; 181(1): 52–9; discussion 87. 117. Schwarz RE. Technical considerations to maintain a low frequency of postoperative biliary stent-associated infections. J Hepatobiliary Pancreat Surg 2002; 9(1): 93–7. 118. Jagannath P, Dhir V, Shrikhande S, et al. Effect of preoperative biliary stenting on immediate outcome after pancreaticoduodenectomy. Br J Surg 2005; 92(3): 356–61. 119. Lai EC, Mok FP, Fan ST, et al. Preoperative endoscopic drainage for malignant obstructive jaundice. Br J Surg 1994; 81(8): 1195–8. 120. Pitt HA, Gomes AS, Lois JF, et al. Does preoperative percutaneous biliary drainage reduce operative risk or increase hospital cost? Ann Surg 1985; 201(5): 545–53. 121. Conlon KC, Klimstra DS, Brennan MF. Long-term survival after curative resection for pancreatic ductal adenocarcinoma. Clinicopathologic analysis of 5-year survivors. Ann Surg 1996; 223(3): 273–9. 122. Kim HJ, Czischke K, Brennan MF, Conlon KC. Does neoadjuvant chemoradiation downstage locally advanced pancreatic cancer? J Gastrointest Surg 2002; 6(5): 763–9. 123. Espat NJ, Brennan MF, Conlon KC. Patients with laparoscopically staged unresectable pancreatic adenocarcinoma do not require subsequent surgical biliary or gastric bypass. J Am Coll Surg 1999; 188(6): 649–55; discussion 655–7. 124. Whipple AO, Parsons WB, Mullins CR. Treatment of Carcinoma of the Ampulla of Vater. Ann Surg 1935; 102(4): 763–79. 125. Traverso LW, Longmire WP. Preservation of the pylorus in pancreaticoduodenectomy. Surg Gynecol Obstet 1978; 146(6): 959–62. 126. Jones L, Russell C, Mosca F, et al. Standard Kausch-Whipple pancreatoduodenectomy. Dig Surg 1999; 16(4): 297–304. 127. Pedrazzoli S, Beger HG, Obertop H, et al. A surgical and pathological based classification of resective treatment of pancreatic cancer. Summary of an international workshop on surgical procedures in pancreatic cancer. Dig Surg 1999; 16(4): 337–45. 128. Seiler CA, Wagner M, Sadowski C, et al. Randomized prospective trial of pylorus-preserving vs. Classic duodenopancreatectomy (Whipple procedure): initial clinical results. J Gastrointest Surg 2000; 4(5): 443–52. 129. Williamson RC, Bliouras N, Cooper MJ, Davies ER. Gastric emptying and enterogastric reflux after conservative and conventional pancreatoduodenectomy. Surgery 1993; 114(1): 82–6. 130. Zerbi A, Balzano G, Patuzzo R, et al. Comparison between pylorus-preserving and Whipple pancreatoduodenectomy. Br J Surg 1995; 82(7): 975–9. 131. Mosca F, Giulianotti PC, Balestracci T, et al. Long-term survival in pancreatic cancer: pylorus-preserving versus Whipple pancreatoduodenectomy. Surgery 1997; 122(3): 553–66. 132. Diener MK, Heukaufer C, Schwarzer G, et al. Pancreaticoduodenectomy (classic Whipple) versus pylorus-preserving pancreaticoduodenectomy (pp Whipple) for surgical treatment of periampullary and pancreatic carcinoma. Cochrane Database Syst Rev 2008; (2): CD006053. 133. Lerut JP, Gianello PR, Otte JB, Kestens PJ. Pancreaticoduodenal resection. Surgical experience and evaluation of risk factors in 103 patients. Ann Surg 1984; 199(4): 432–7. 134. Herter FP, Cooperman AM, Ahlborn TN, Antinori C. Surgical experience with pancreatic and periampullary cancer. Ann Surg 1982; 195(3): 274–81. 135. Aston SJ, Longmire WP. Pancreaticoduodenal resection. Twenty years’ experience. Arch Surg 1973; 106(6): 813–7. 136. Wente MN, Shrikhande SV, Müller MW, et al. Pancreaticojejunostomy versus pancreaticogastrostomy: systematic review and meta-analysis. Am J Surg 2007; 193(2): 171–83. 137. Sikora SS, Posner MC. Management of the pancreatic stump following pancreaticoduodenectomy. Br J Surg 1995; 82(12): 1590–7. 138. Bassi C, Falconi M, Molinari E, et al. Duct-to-mucosa versus end-to-side pancreaticojejunostomy reconstruction after pancreaticoduodenectomy: results of a prospective randomized trial. Surgery 2003; 134(5): 766–71. 139. Balcom JH, Rattner DW, Warshaw AL, et al. Ten-year experience with 733 pancreatic resections: changing indications, older patients, and decreasing length of hospitalization. Arch Surg 2001; 136(4): 391–8. 140. Japanese Pancreatic Society. Classification of Pancreatic Carcinoma, 2nd edn. Tokyo: Kanehara, 2003. 141. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma, part 2: randomized controlled trial evaluating survival, morbidity, and mortality. Ann Surg 2002; 236(3): 355–66; discussion 366–8. 142. Pedrazzoli S, DiCarlo V, Dionigi R, et al. Standard versus extended lymphadenectomy associated with pancreatoduodenectomy in the surgical treatment of adenocarcinoma of the head of the pancreas: a multicenter, prospective, randomized study. Lymphadenectomy Study Group. Ann Surg 1998; 228(4): 508–17. 143. Farnell MB, Pearson RK, Sarr MG, et al. A prospective randomized trial comparing standard pancreatoduodenectomy with pancreatoduodenectomy with extended lymphadenectomy in resectable pancreatic head adenocarcinoma. Surgery 2005; 138(4): 618–28; discussion 628–30. 144. Riall TS, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma—part 3: update on 5-year survival. J Gastrointest Surg 2005; 9(9): 1191–204; discussion 1204–6. 145. Michalski CW, Kleeff J, Wente MN, et al. Systematic review and metaanalysis of standard and extended lymphadenectomy in pancreaticoduodenectomy for pancreatic cancer. Br J Surg 2007; 94(3): 265–73. 146. Leach SD, Lee JE, Charnsangavej C, et al. Survival following pancreaticoduodenectomy with resection of the superior mesenteric-portal vein confluence for adenocarcinoma of the pancreatic head. Br J Surg 1998; 85(5): 611–7. 147. Bachellier P, Nakano H, Oussoultzoglou PD, et al. Is pancreaticoduodenectomy with mesentericoportal venous resection safe and worthwhile? Am J Surg 2001; 182(2): 120–9. 148. Nakao A, Takeda S, Inoue S, et al. Indications and techniques of extended resection for pancreatic cancer. World J Surg 2006; 30(6): 976–82; discussion 983–4. 149. Sasson AR, Hoffman JP, Ross EA, et al. En bloc resection for locally advanced cancer of the pancreas: is it worthwhile? J Gastrointest Surg 6(2): 147–57; discussion 157–8. 150. Nakagohri T, Kinoshita T, Konishi M, et al. Survival benefits of portal vein resection for pancreatic cancer. Am J Surg 2003; 186(2): 149–53. 151. van Geenen RC, ten Kate FJ, de Wit LT, et al. Segmental resection and wedge excision of the portal or superior mesenteric vein during pancreatoduodenectomy. Surgery 2001; 129(2): 158–63. 152. Weitz J, Kienle P, Schmidt J, et al. Portal vein resection for advanced pancreatic head cancer. J Am Coll Surg 2007; 204(4): 712–6. 153. Stitzenberg KB, Watson JC, Roberts A, et al. Survival after pancreatectomy with major arterial resection and reconstruction. Ann Surg Oncol 2008; 15(5): 1399–406. 154. Yekebas EF, Bogoevski D, Cataldegirmen G, et al. En bloc vascular resection for locally advanced pancreatic malignancies infiltrating major blood vessels: perioperative outcome and long-term survival in 136 patients. Ann Surg 2008; 247(2): 300–9. 155. Hirano S, Kondo S, Hara T, et al. Distal pancreatectomy with en bloc celiac axis resection for locally advanced pancreatic body cancer: longterm results. Ann Surg 2007; 246(1): 46–51. 156. Gagandeep S, Artinyan A, Jabbour N, et al. Extended pancreatectomy with resection of the celiac axis: the modified Appleby operation. Am J Surg 2006; 192(3): 330–5. 157. Kondo S, Katoh H, Hirano S, et al. Results of radical distal pancreatectomy with en bloc resection of the celiac artery for locally advanced cancer of the pancreatic body. Langenbecks Arch Surg 2003; 388(2): 101–6.

397

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
158. Shrikhande SV, Kleeff J, Reiser C, et al. Pancreatic resection for M1 pancreatic ductal adenocarcinoma. Ann Surg Oncol 2007; 14(1): 118–27. 159. Gleisner AL, Assumpcao L, Cameron JL, et al. Is resection of periampullary or pancreatic adenocarcinoma with synchronous hepatic metastasis justified? Cancer 2007; 110(11): 2484–92. 160. Allendorf JD, Lauerman M, Bill A, et al. Neoadjuvant chemotherapy and radiation for patients with locally unresectable pancreatic adenocarcinoma: feasibility, efficacy, and survival. J Gastrointest Surg 2008; 12(1): 91–100. 161. Krempien R, Muenter MW, Harms W, Debus J. Neoadjuvant chemoradiation in patients with pancreatic adenocarcinoma. HPB (Oxford) 2006; 8(1): 22–8. 162. Kleeff J, Friess H, Büchler MW. Neoadjuvant therapy for pancreatic cancer. Br J Surg 2007; 94(3): 261–2. 163. Sohn TA, Lillemoe KD, Cameron JL, et al. Reexploration for periampullary carcinoma: resectability, perioperative results, pathology, and longterm outcome. Ann Surg 1999; 229(3): 393–400. 164. Michalski CW, Kleeff J, Bachmann J, et al. Second-look operation for unresectable pancreatic ductal adenocarcinoma at a high-volume center. Ann Surg Oncol 2008; 15(1): 186–92. 165. Gouma DJ, van Geenen RC, van Gulik TM, et al. Rates of complications and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume. Ann Surg 2000; 232(6): 786–95. 166. Gordon TA, Bowman HM, Bass EB, et al. Complex gastrointestinal surgery: impact of provider experience on clinical and economic outcomes. J Am Coll Surg 1999; 189(1): 46–56. 167. Gordon TA, Bowman HM, Tielsch JM, et al. Statewide regionalization of pancreaticoduodenectomy and its effect on in-hospital mortality. Ann Surg 1998; 228(1): 71–8. 168. Gordon TA, Burleyson GP, Tielsch JM, Cameron JL. The effects of regionalization on cost and outcome for one general high-risk surgical procedure. Ann Surg 1995; 221(1): 43–9. 169. Birkmeyer JD, Finlayson SR, Tosteson AN, et al. Effect of hospital volume on in-hospital mortality with pancreaticoduodenectomy. Surgery 1999; 125(3): 250–6. 170. Birkmeyer JD, Siewers AE, Finlayson EVA, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002; 346(15): 1128–37. 171. Birkmeyer JD, Sun Y, Wong SL, Stukel TA. Hospital volume and late survival after cancer surgery. Ann Surg 2007; 245(5): 777–83. 172. Ho V, Heslin MJ. Effect of hospital volume and experience on in-hospital mortality for pancreaticoduodenectomy. Ann Surg 2003; 237(4): 509–14. 173. Lieberman MD, Kilburn H, Lindsey M, Brennan MF. Relation of perioperative deaths to hospital volume among patients undergoing pancreatic resection for malignancy. Ann Surg 1995; 222(5): 638–45. 174. Simunovic M, To T, Theriault M, Langer B. Relation between hospital surgical volume and outcome for pancreatic resection for neoplasm in a publicly funded health care system. CMAJ 1999; 160(5): 643–8. 175. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative mortality for major cancer surgery. JAMA 1998; 280(20): 1747–51. 176. Sosa JA, Bowman HM, Gordon TA, et al. Importance of hospital volume in the overall management of pancreatic cancer. Ann Surg 1998; 228(3): 429–38. 177. Welsch T, Büchler MW, Schmidt J. Surgery for pancreatic cancer. Z Gastroenterol 2008; 46(12): 1393–403. 178. Fong Y, Gonen M, Rubin D, et al. Long-term survival is superior after resection for cancer in high-volume centers. Ann Surg 2005; 242(4): 540–4; discussion 544–7. 179. van Heek NT, Kuhlmann KFD, Scholten RJ, et al. Hospital volume and mortality after pancreatic resection: a systematic review and an evaluation of intervention in the Netherlands. Ann Surg 2005; 242(6): 781–8, discussion 788–90. 180. Büchler MW, Wagner M, Schmied BM, et al. Changes in morbidity after pancreatic resection: toward the end of completion pancreatectomy. Arch Surg 2003; 138(12): 1310–4; discussion 1315. 181. Winter JM, Cameron JL, Campbell KA, et al. 1423 pancreaticoduodenectomies for pancreatic cancer: A single-institution experience. J Gastrointest Surg 2006; 10(9): 1199–210; discussion 1210–1. 182. Berberat PO, Friess H, Kleeff J, et al. Prevention and treatment of complications in pancreatic cancer surgery. Dig Surg 1999; 16(4): 327–36. 183. Dimick JB, Pronovost PJ, Cowan JA, et al. Variation in postoperative complication rates after high-risk surgery in the United States. Surgery 2003; 134(4): 534–40; discussion 540–1. 184. Cunningham JD, Weyant MT, Levitt M, et al. Complications requiring reoperation following pancreatectomy. Int J Pancreatol 1998; 24(1): 23–9. 185. Farley DR, Schwall G, Trede M. Completion pancreatectomy for surgical complications after pancreaticoduodenectomy. Br J Surg 1996; 83(2): 176–9. 186. Halloran CM, Ghaneh P, Bosonnet L, et al. Complications of pancreatic cancer resection. Dig Surg 2002; 19(2): 138–46. 187. Welsch T, Frommhold K, Hinz U, et al. Persisting elevation of C-reactive protein after pancreatic resections can indicate developing inflammatory complications. Surgery 2008; 143(1): 20–8. 188. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997; 226(3): 248–57; discussion 257–60. 189. van Berge Henegouwen MI, van Gulik TM, DeWit LT, et al. Delayed gastric emptying after standard pancreaticoduodenectomy versus pyloruspreserving pancreaticoduodenectomy: an analysis of 200 consecutive patients. J Am Coll Surg 1997; 185(4): 373–9. 190. Tran KTC, Smeenk HG, van Eijck CHJ, et al. Pylorus preserving pancreaticoduodenectomy versus standard Whipple procedure: a prospective, randomized, multicenter analysis of 170 patients with pancreatic and periampullary tumors. Ann Surg 2004; 240(5): 738–45. 191. Hartel M, Wente MN, Hinz U, et al. Effect of antecolic reconstruction on delayed gastric emptying after the pylorus-preserving Whipple procedure. Arch Surg 2005; 140(11): 1094–9. 192. Yeo CJ, Barry MK, Sauter PK, et al. Erythromycin accelerates gastric emptying after pancreaticoduodenectomy. A prospective, randomized, placebo-controlled trial. Ann Surg 1993; 218(3): 229–37; discussion 237–8. 193. Böttger TC, Junginger T. Factors influencing morbidity and mortality after pancreaticoduodenectomy: critical analysis of 221 resections. World J Surg 1999; 23(2): 164–71; discussion 171–2. 194. Cullen JJ, Sarr MG, Ilstrup DM. Pancreatic anastomotic leak after pancreaticoduodenectomy: incidence, significance, and management. Am J Surg 1994; 168(4): 295–8. 195. Bassi C, Dervenis C, Butturini G, et al. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005; 138(1): 8–13. 196. Bassi C, Falconi M, Salvia R, et al. Management of complications after pancreaticoduodenectomy in a high volume centre: results on 150 consecutive patients. Dig Surg 2001; 18(6): 453–7; discussion 458. 197. Sohn TA, Yeo CJ, Cameron JL, et al. Pancreaticoduodenectomy: role of interventional radiologists in managing patients and complications. J Gastrointest Surg 2003; 7(2): 209–19. 198. Tien Y, Lee P, Yang C, et al. Risk factors of massive bleeding related to pancreatic leak after pancreaticoduodenectomy. J Am Coll Surg 2005; 201(4): 554–9. 199. Trede M, Schwall G. The complications of pancreatectomy. Ann Surg 1988; 207(1): 39–47. 200. Wente MN, Veit JA, Bassi C, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery 2007; 142(1): 20–5. 201. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy for cancer of the head of the pancreas. 201 patients. Ann Surg 1995; 221(6): 721–31; discussion 731–3. 202. Shimizu Y, Yasui K, Fuwa N, et al. Late complication in patients undergoing pancreatic resection with intraoperative radiation therapy: gastrointestinal bleeding with occlusion of the portal system. J Gastroenterol Hepatol 2005; 20(8): 1235–40. 203. de Castro SMM, Busch ORC, Gouma DJ. Management of bleeding and leakage after pancreatic surgery. Best Pract Res Clin Gastroenterol 2004; 18(5): 847–64. 204. Brennan MF, Pisters PW, Posner M, et al. A prospective randomized trial of total parenteral nutrition after major pancreatic resection for malignancy. Ann Surg 1994; 220(4): 436–41; discussion 441–4.

398

ADENOCARCINOMA OF THE PANCREAS
205. van Berge Henegouwen MI, Akkermans LM, van Gulik TM, et al. Prospective, randomized trial on the effect of cyclic versus continuous enteral nutrition on postoperative gastric function after pyloruspreserving pancreatoduodenectomy. Ann Surg 1997; 226(6): 677–85; discussion 685–7. 206. Berberat PO, Ingold H, Gulbinas A, et al. Fast track-different implications in pancreatic surgery. J Gastrointest Surg 2007; 11(7): 880–7. 207. Büchler MW, Friess H. Evidence forward, drainage on retreat: still we ignore and drain!? Ann Surg 2006; 244(1): 8–9. 208. Kawai M, Tani M, Terasawa H, et al. Early removal of prophylactic drains reduces the risk of intra-abdominal infections in patients with pancreatic head resection: prospective study for 104 consecutive patients. Ann Surg 2006; 244(1): 1–7. 209. Connor S, Alexakis N, Garden OJ, et al. Meta-analysis of the value of somatostatin and its analogues in reducing complications associated with pancreatic surgery. Br J Surg 2005; 92(9): 1059–67. 210. Oettle H, Post S, Neuhaus P, et al. Adjuvant chemotherapy with gemcitabine vs observation in patients undergoing curative-intent resection of pancreatic cancer: a randomized controlled trial. JAMA 2007; 297(3): 267–77. 211. Neoptolemos JP, Stocken DD, Friess H, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med 2004; 350(12): 1200–10. 212. Stocken DD, Büchler MW, Dervenis C, et al. Meta-analysis of randomised adjuvant therapy trials for pancreatic cancer. Br J Cancer 2005; 92(8): 1372–81. 213. Sener SF, Fremgen A, Menck HR, Winchester DP. Pancreatic cancer: a report of treatment and survival trends for 100,313 patients diagnosed from 1985–1995, using the National Cancer Database. J Am Coll Surg 1999; 189(1): 1–7. 214. Knaebel HP, Märten A, Schmidt J, et al. Phase III trial of postoperative cisplatin, interferon alpha-2b, and 5-FU combined with external radiation treatment versus 5-FU alone for patients with resected pancreatic adenocarcinoma—CapRI: study protocol [ISRCTN62866759]. BMC Cancer 2005; 5: 37. 215. Picozzi VJ, Kozarek RA, Traverso LW. Interferon-based adjuvant chemoradiation therapy after pancreaticoduodenectomy for pancreatic adenocarcinoma. Am J Surg 2003; 185(5): 476–80. 216. Crippa S, Domínguez I, Rodríguez JR, et al. Quality of life in pancreatic cancer: analysis by stage and treatment. J Gastrointest Surg 2008; 12(5): 783–93; discussion 793–4. 217. Makary MA, Winter JM, Cameron JL, et al. Pancreaticoduodenectomy in the very elderly. J Gastrointest Surg 2006; 10(3): 347–56. 218. Bilimoria KY, Bentrem DJ, Ko CY, et al. National failure to operate on early stage pancreatic cancer. Ann Surg 2007; 246(2): 173–80. 219. Lüttges J, Vogel I, Menke M, et al. The retroperitoneal resection margin and vessel involvement are important factors determining survival after pancreaticoduodenectomy for ductal adenocarcinoma of the head of the pancreas. Virchows Arch 1998; 433(3): 237–42. 220. Delcore R, Rodriguez FJ, Forster J, et al. Significance of lymph node metastases in patients with pancreatic cancer undergoing curative resection. Am J Surg 1996; 172(5): 463–8; discussion 468–9. 221. Millikan KW, Deziel DJ, Silverstein JC, et al. Prognostic factors associated with resectable adenocarcinoma of the head of the pancreas. Am Surg 1999; 65(7): 618–23; discussion 623–4. 222. Raut CP, Tseng JF, Sun CC, et al. Impact of resection status on pattern of failure and survival after pancreaticoduodenectomy for pancreatic adenocarcinoma. Ann Surg 2007; 246(1): 52–60. 223. Benassai G, Mastrorilli M, Quarto G, et al. Survival after pancreaticoduodenectomy for ductal adenocarcinoma of the head of the pancreas. Chir Ital 2000; 52(3): 263–70. 224. Schnelldorfer T, Ware AL, Sarr MG, et al. Long-term survival after pancreatoduodenectomy for pancreatic adenocarcinoma: is cure possible? Ann Surg 2008; 247(3): 456–62. 225. Carpelan-Holmström M, Nordling S, Pukkala E, et al. Does anyone survive pancreatic ductal adenocarcinoma? A nationwide study re-evaluating the data of the Finnish Cancer Registry. Gut 2005; 54(3): 385–7. 226. Evans DB, Abbruzzese JL, Willett CG. Cancer of the pancreas. In: DeVita VT, Hellman S, Rosenberg SA, eds. Cancer—Principles and Practice of Oncology, 6th edn. Philadelphia: Lippincott Williams and Wilkins, 2001: 1126–61. Cameron JL, Riall TS, Coleman J, Belcher KA. One thousand consecutive pancreaticoduodenectomies. Ann Surg 2006; 244(1): 10–5. Lim JE, Chien MW, Earle CC. Prognostic factors following curative resection for pancreatic adenocarcinoma: a population-based, linked database analysis of 396 patients. Ann Surg 2003; 237(1): 74–85. Meyer W, Jurowich C, Reichel M, et al. Pathomorphological and histological prognostic factors in curatively resected ductal adenocarcinoma of the pancreas. Surg Today 2000; 30(7): 582–7. Bouvet M, Gamagami RA, Gilpin EA, et al. Factors influencing survival after resection for periampullary neoplasms. Am J Surg 2000; 180(1): 13–7. Cameron JL, Crist DW, Sitzmann JV, et al. Factors influencing survival after pancreaticoduodenectomy for pancreatic cancer. Am J Surg 1991; 161(1): 120–4; discussion 124–5. Allema JH, Reinders ME, van Gulik TM, et al. Prognostic factors for survival after pancreaticoduodenectomy for patients with carcinoma of the pancreatic head region. Cancer 1995; 75(8): 2069–76. Welsch T, Kleeff J, Friess H. Molecular pathogenesis of pancreatic cancer: advances and challenges. Curr Mol Med 2007; 7(5): 504–21. Wagner M, Redaelli C, Lietz M, et al. Curative resection is the single most important factor determining outcome in patients with pancreatic adenocarcinoma. Br J Surg 2004; 91(5): 586–94. Pawlik TM, Gleisner AL, Cameron JL, et al. Prognostic relevance of lymph node ratio following pancreaticoduodenectomy for pancreatic cancer. Surgery 2007; 141(5): 610–8. Kuhlmann KFD, de Castro SMM, Wesseling JG, et al. Surgical treatment of pancreatic adenocarcinoma; actual survival and prognostic factors in 343 patients. Eur J Cancer 2004; 40(4): 549–58. Shimada K, Sakamoto Y, Sano T, Kosuge T. Prognostic factors after distal pancreatectomy with extended lymphadenectomy for invasive pancreatic adenocarcinoma of the body and tail. Surgery 2006; 139(3): 288–95. Takahashi T, Niino N, Ishikura H, et al. Predictive factors for long-term survival in patients with pancreatic carcinoma. Hepatogastroenterology 1997; 44(17): 1463–8. Ozaki H, Hiraoka T, Mizumoto R, et al. The prognostic significance of lymph node metastasis and intrapancreatic perineural invasion in pancreatic cancer after curative resection. Surg Today 1999; 29(1): 16–22. Cleary SP, Gryfe R, Guindi M, et al. Prognostic factors in resected pancreatic adenocarcinoma: analysis of actual 5-year survivors. J Am Coll Surg 2004; 198(5): 722–31. Han S, Jang J, Kim S, et al. Analysis of long-term survivors after surgical resection for pancreatic cancer. Pancreas 2006; 32(3): 271–5. Connor S, Bosonnet L, Ghaneh P, et al. Survival of patients with periampullary carcinoma is predicted by lymph node 8a but not by lymph node 16b1 status. Br J Surg 2004; 91(12): 1592–9. Tseng JF, Raut CP, Lee JE, et al. Pancreaticoduodenectomy with vascular resection: margin status and survival duration. J Gastrointest Surg 2004; 8(8): 935–49; discussion 949–50. Geer RJ, Brennan MF. Prognostic indicators for survival after resection of pancreatic adenocarcinoma. Am J Surg 1993; 165(1): 68–72; discussion 72–3. Neoptolemos JP, Stocken DD, Dunn JA, et al. Influence of resection margins on survival for patients with pancreatic cancer treated by adjuvant chemoradiation and/or chemotherapy in the ESPAC-1 randomized controlled trial. Ann Surg 2001; 234(6): 758–68. Verbeke CS, Leitch D, Menon KV, et al. Redefining the R1 resection in pancreatic cancer. Br J Surg 2006; 93(10): 1232–7. Esposito I, Kleeff J, Bergmann F, et al. Most pancreatic cancer resections are R1 resections. Ann Surg Oncol 2008; 15(6): 1651–60. Bueno de Mesquita HB, Maisonneuve P, et al. Intake of foods and nutrients and cancer of the exocrine pancreas: a population-based casecontrol study in the Netherlands. Int J Cancer 1991; 48(4): 540–9. Lyon JL, Slattery ML, Mahoney AW, Robison LM. Dietary intake as a risk factor for cancer of the exocrine pancreas. Cancer Epidemiol Biomarkers Prev 1993; 2(6): 513–8. Larsson SC, Håkansson N, Näslund I, et al. Fruit and vegetable consumption in relation to pancreatic cancer risk: a prospective study. Cancer Epidemiol Biomarkers Prev 2006; 15(2): 301–5.

227. 228.

229.

230. 231.

232.

233. 234.

235.

236.

237.

238.

239.

240.

241. 242.

243.

244.

245.

246. 247. 248.

249.

250.

399

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
251. Lin Y, Tamakoshi A, Hayakawa T, et al. Nutritional factors and risk of pancreatic cancer: a population-based case-control study based on direct interview in Japan. J Gastroenterol 2005; 40(3): 297–301. 252. Anderson KE, Sinha R, Kulldorff M, et al. Meat intake and cooking techniques: associations with pancreatic cancer. Mutat Res 2002; 506–507: 225–31. 253. Anderson KE, Kadlubar FF, Kulldorff M, et al. Dietary intake of heterocyclic amines and benzo(a)pyrene: associations with pancreatic cancer. Cancer Epidemiol Biomarkers Prev 2005; 14(9): 2261–5. 254. Ghadirian P, Baillargeon J, Simard A, Perret C. Food habits and pancreatic cancer: a case-control study of the Francophone community in Montreal, Canada. Cancer Epidemiol Biomarkers Prev 1995; 4(8): 895–9. 255. Michaud DS, Liu S, Giovannucci E, et al. Dietary sugar, glycemic load, and pancreatic cancer risk in a prospective study. J Natl Cancer Inst 2002; 94(17): 1293–300. 256. Silvera SAN, Rohan TE, Jain M, et al. Glycemic index, glycemic load, and pancreatic cancer risk (Canada). Cancer Causes Control 2005; 16(4): 431–6. 257. Harnack LJ, Anderson KE, Zheng W, et al. Smoking, alcohol, coffee, and tea intake and incidence of cancer of the exocrine pancreas: the Iowa Women’s Health Study. Cancer Epidemiol Biomarkers Prev 1997; 6(12): 1081–6. 258. La Vecchia C, Negri E, Franceschi S, D’, et al. Tea consumption and cancer risk. Nutr Cancer 1992; 17(1): 27–31. 259. Tavani A, La Vecchia C. Coffee and cancer: a review of epidemiological studies, 1990–1999. Eur J Cancer Prev 2000; 9(4): 241–56. 260. Silverman DT. Risk factors for pancreatic cancer: a case-control study based on direct interviews. Teratog Carcinog Mutagen 2001; 21(1): 7–25. 261. Wente MN, Bassi C, Dervenis C, et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS). Surgery 2007; 142(5): 761–8. 262. Takada T, Amano H, Yasuda H, et al. Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A phase III multicenter prospective randomized controlled trial in patients with resected pancreaticobiliary carcinoma. Cancer 2002; 95(8): 1685–95. 263. Bakkevold KE, Arnesjø B, Dahl O, Kambestad B. Adjuvant combination chemotherapy (AMF) following radical resection of carcinoma of the pancreas and papilla of Vater--results of a controlled, prospective, randomised multicentre study. Eur J Cancer 1993; 29A(5): 698–703. 264. Klinkenbijl JH, Jeekel J, Sahmoud T, et al. Adjuvant radiotherapy and 5-fluorouracil after curative resection of cancer of the pancreas and periampullary region: phase III trial of the EORTC gastrointestinal tract cancer cooperative group. Ann Surg 1999; 230(6): 776–82; discussion 782–4. 265. Regine WF, Winter KA, Abrams RA, et al. Fluorouracil vs gemcitabine chemotherapy before and after fluorouracil-based chemoradiation following resection of pancreatic adenocarcinoma: a randomized controlled trial. JAMA 2008; 299(9): 1019–26. 266. Further evidence of effective adjuvant combined radiation and chemotherapy following curative resection of pancreatic cancer. Gastrointestinal Tumor Study Group. Cancer 1987; 59(12): 2006–10. 267. Martignoni, ME, Friess H, Sell F, et al. Enteral nutrition prolongs delayed gastric emptying in patients after Whipple resection. Am J Surg 2000; 180(1) (July): 18–23. 268. Habbe N, Langer P, Bartsch DK. Hereditäres Pankreaskarzinom. Chirurg 2008; 79: 1029–37. 269. van der Gaag NA, Rauws EA, van Eijck CH, et al. Preoperative biliary drainage for cancer of the head of the pancreas. N Engl J Med 2010; 362: 129–37. 270. Neoptolemos JP, Dunn JA, Stocken DD, et al. Adjuvant chemotherapy with fluorouracil plus folinic acid vs. gemcitabine following pancreatic cancer resection: a randomized controlled trial. JAMA 2010; 304: 1073–81.

400

43

Palliation of pancreas cancer Michael G. House and Keith D. Lillemoe
considerable morbidity to the actual laparotomy and exploration (12,18,21). For patients with unequivocal evidence of unresectable or metastatic disease during preoperative evaluation, endoluminal methods for biliary and gastroduodenal stenting should be attempted when clinical findings of obstruction are present or imminent. Operative bypass procedures should be reserved for treatment failures of nonoperative methods (i.e., endoscopic or percutaneous) in patients who otherwise have reasonable life expectancy.

The surgical management of pancreatic adenocarcinoma is focused largely on complete tumor extirpation in patients with resectable tumors. Unfortunately, the majority of patients with pancreatic adenocarcinoma have advanced stage disease at the time of diagnosis and are not eligible for resection. Detectable metastases and/or extensive locoregional disease is frequently recognized during preoperative staging, thus consideration for a potentially curative resection is appropriate in less than onefourth of all patients with pancreatic adenocarcinoma (1–4). Even though high-resolution cross-sectional imaging and other dedicated staging modalities (e.g., pancreatic endoscopic ultrasound) have obviated the need for routine operative exploration to assess resectability in most patients, occult metastases or celiac/mesenteric vascular invasion precluding complete resection will be discovered at the time of exploration in approximately 20% of patients with localized, apparently resectable tumors on preoperative imaging (5,6). Symptom palliation becomes the goal of therapy for the majority of patients with pancreatic adenocarcinoma whose disease is not amenable to a potentially curative resection. Depending on individual performance status and medical co-morbidities, the life expectancy for all patients with unresectable cancer is typically less than 1 year. Patients with nonmetastatic, locally advanced cancer experience a median survival on the order of 9 to 12 months, whereas metastatic pancreatic adenocarcinoma is typically associated with a median survival of less than 6 months (1,7–10). Adequate palliation of biliary and duodenal obstruction, and most importantly control of cancer-related pain, has been shown to improve quality of life (10–18). Therefore, every attempt, whether nonoperative or operative, should be made to palliate obstruction and relieve pain in virtually all patients with unresectable pancreatic cancer who have a reasonable life expectancy. Contrarily, the benefits of palliative treatments must be weighed against the potential morbidity associated with them. Therefore, it is difficult to advocate prophylactic palliative procedures for asymptomatic patients, many of whom are at uncertain risk for developing symptoms prior to death. Operative treatment has served as the traditional modality for palliating the symptoms associated with locally advanced pancreatic cancer. However, nonoperative therapies offered by endoscopists and interventional radiologists have proved to be reliable and durable in select patients with biliary and/or duodenal obstruction (19,20). A decision to pursue operative versus nonoperative palliation typically arises in two clinical scenarios. For patients undergoing open exploration for equivocal radiographic signs of unresectability, operative palliation is almost always indicated for those found to have nonmetastatic (or low-volume metastatic), locally unresectable disease intraoperatively. In experienced hands, operative biliary and gastric bypass procedures should not add

indications for palliation of pancreatic cancer
The majority of pancreatic adenocarcinomas arise in the head of the pancreas and possess a desmoplastic biology. Not surprisingly, 80% of patients with pancreatic adenocarcinoma will seek medical attention for symptoms related to jaundice secondary to mechanical obstruction of the intrapancreatic portion of the distal common bile duct (16,22). Obstructive jaundice is the most common presenting symptom for patients with periampullary cancer, and if left untreated, it can be accompanied by refractory pruritus, anorexia, malabsorptive diarrhea, and liver failure. Although nausea and vomiting are common symptoms among patients with pancreatic cancer, possibly as a result of disease infiltration of autonomic nerve plexuses that causes gastric noncompliance and poor emptying, only a minority of patients will develop mechanical obstruction of the duodenum, either at the time of diagnosis (i.e., less than 5%) or during disease progression (i.e., 10–30%) (21–23). The development of gastric outlet obstruction only adds to the progressive malnutrition potentiated by the jaundiced state. Each of these conditions, particularly when combined, can lead to rapid generalized wasting and diminished quality of life. For these reasons, decompression of biliary obstruction and relief of duodenal obstruction lead to a dramatic improvement in the overall medical condition that contributes to a prolongation of comfortable survival. Diagnostic laparoscopy, either routinely or selectively, has become an integral part of the staging of many patients with pancreatic cancer. In most situations when unresectable disease is found at laparoscopy, as either liver metastasis or peritoneal implants, life expectancy is quite short and operative palliation is not generally indicated. In a series of 155 patients from Memorial Sloan-Kettering Cancer Center who were found to have unresectable pancreatic adenocarcinoma at the time of staging laparoscopy, only 2% required an open procedure to palliate biliary or gastric obstruction during their remaining lifetime (24). Jaundiced patients without gastric outlet obstruction, who are found to have metastatic disease at the time of staging laparoscopy, can be palliated successfully with biliary stenting alone in most circumstances. Laparoscopic biliary bypass is an option and surgical series,

401

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
involving limited numbers of patients, have shown satisfactory short- and long-term results from this approach (25–29). Patients with gastric outlet obstruction, who are determined to have unresectable disease at the time of staging laparoscopy, should be considered for laparoscopic gastrojejunostomy, especially when there are anatomic constraints that will limit the success of endoluminal gastroduodenal stenting. If unresectable disease is discovered at the time of laparotomy, both a biliary–enteric bypass and a gastrojejunostomy should be performed regardless of existing symptoms. At least three meta-analyses of surgical series have suggested that 15% to 25% of patients found to be unresectable at the time of laparotomy and not provided with a gastrojejunostomy will eventually develop symptomatic duodenal obstruction (22,23). Two prospective randomized trials have also provided Level Ib evidence that supports performing a biliary–enteric bypass and a gastrojejunostomy for patients who are determined to have unresectable disease at the time of laparotomy (12,18). are found to be unresectable at the time of laparotomy, should be provided with appropriate operative palliation. This point is obvious for patients who require transection of the bile duct as part of the operation to determine resectability. However, for patients who have existing metallic biliary stents and do not require division of the bile duct to assess resectability, a decision to perform a biliary bypass must factor the patient’s expected lifespan, the reported durability of metallic stents, and potential operative complications associated with biliary– enteric bypass. Unfortunately, there are no randomized data comparing operative versus nonoperative biliary decompression procedures using metallic biliary stents exclusively. After the abdomen is entered and assessed for metastatic disease, the duodenum is extensively mobilized out of the retroperitoneum to determine involvement of the superior mesenteric artery and to exclude the rare presence of aortic or caval invasion. Accurate assessment of tumor resectability in patients with equivocal radiographic findings usually necessitates a cholecystectomy and transection of the common bile or hepatic duct to facilitate identification and dissection of the portal vein. With extensive Kocherization of the third portion of the duodenum, the superior mesenteric vein (SMV) can be identified anteriorly and dissected along its surface under the neck of the pancreas to its connection with the portal vein. If extensive tumor encasement of the SMV or portal vein is discovered and the chance for a margin-negative resection is unlikely even with a major vein resection and reconstruction, a palliative double (biliary and gastric) bypass procedure should be considered at this point in the operation. If tissue confirmation of pancreatic adenocarcinoma was not obtained preoperatively, a transduodenal core needle biopsy of the pancreatic head should be obtained. Even though an operative biliary bypass can be accomplished with cholecystojejunostomy or choledochoduodenostomy, these two options are associated with overall inferior short- and long-term results and generally should be avoided (8,37–39). Our preferred approach uses hepatico- or choledochojejunostomy for internal drainage. Biliary bypass can be accomplished with either a simple jejunal loop or a Roux-en-Y limb (Fig. 43.1). While a loop anastomosis requires slightly less operative time, the use of a defunctionalized Roux-en-Y jejunal limb is associated with less anastomotic tension and facilitates the management of potential biliary leaks. The incidence of postoperative cholangitis also seems to be reduced with Roux limb drainage.

nonoperative techniques for biliary decompression
For jaundiced patients with unequivocally unresectable pancreatic cancer on preoperative evaluation, nonsurgical palliation is generally indicated except for the most terminally ill patients. Since its clinical inception in 1980, the use of endoscopically placed biliary endoprostheses has continued to evolve and now serves as the predominant modality for palliating obstructive jaundice in patients who are not candidates for curative resection. Endoscopic attempts at biliary drainage fail in less than 10% of patients, usually as the result of tumor infiltration into the duodenal wall that prevents access to the ampulla (30,31). A Cochrane review of endoscopic stents for the relief of distal biliary obstruction has provided Level Ia evidence that metal biliary stents, compared to plastic stents, have a lower risk of recurrent obstruction with no increased risk of complications (32). Obviously, the cost-effectiveness of metal stents over plastic stents is most apparent for patients with longer survival (33–36). In the uncommon event that endoscopic management is unsuccessful, percutaneous transhepatic access should be gained to allow external biliary drainage. In most cases, the initial external drainage procedure can be converted later to internal biliary drainage with a stainless steel alloy biliary stent (e.g., Wallstent; Boston Scientific, Natick, MA, USA) (19). Most symptomatic patients with preoperatively confirmed unresectable disease can be palliated adequately with nonoperative techniques, thus there is little role for surgical palliation for a large subgroup of patients with pancreatic cancer. However, there remains an important role for operative palliation in patients undergoing attempted resection.

nonoperative techniques for gastric decompression
Duodenal or gastric outlet obstruction has traditionally been managed by surgery, but there has been increased experience with endoluminal approaches to relieve gastric and duodenal obstruction over the past several years (40–42). In the past, endoscopic options included tube gastrostomy with jejunal extension for nutritional access; however, the development of self-expanding enteral stents has provided a reliable tool for palliating duodenal obstruction in patients who do not require surgical exploration to determine resectability (Fig. 43.2). Despite early success with enteral stents in small series,

operative techniques for biliary decompression
Despite advances in diagnostic radiography, open surgical exploration continues to serve as the standard for determining local tumor resectability. Thus, operative palliation of existing or potential biliary obstruction remains a major issue in the management of unresectable pancreatic cancer. Patients, who

402

PALLIATION OF PANCREAS CANCER
complications can arise and include mucosal ulceration, duodenal perforation, stent migration, and tumor ingrowth leading to recurrent obstruction (43,44). Patients with reasonable life expectancy, who fail endoscopic attempts at palliation or develop complications related to endoluminal stenting, may require an operative gastrojejunostomy.

operative techniques for gastric decompression
Historically, most surgeons advocated an antecolic gastrojejunostomy due to concerns of placing the anastomosis in proximity to the tumor bed; however, there is now strong evidence that a retrocolic, isoperistaltic gastrojejunostomy is associated with a lower incidence of postoperative delayed gastric emptying and even late-occurring gastric outlet obstruction (45). The anastomosis should be fashioned with either a hand-sewn or a stapled technique at the most dependent aspect of the greater curvature of the stomach with a loop of jejunum approximately 20 to 30 cm from the ligament of Treitz. The posterior gastrojejunostomy should be delivered below the transverse mesocolon and tacked in place. Vagotomy is generally avoided during palliative gastrojejunostomy to prevent delayed gastric emptying.

palliation of pain
The last step of surgical palliation for unresectable pancreatic cancer includes chemical splanchnicectomy which can be readily accomplished with the injection of 20 ml of 50% alcohol on each side of the aorta at the level of the celiac axis at the time of laparotomy (Fig. 43.3). Chemical splanchnicectomy can be performed similarly at the time of staging laparoscopy or other laparoscopic palliative procedures (46). Celiac plexus blocks under endoscopic ultrasound or computed tomography guidance have also been described, and thoracic splanchnicectomy although used infrequently can provide adequate pain relief for patients with unresectable pancreatic cancer (47,48). Long-term, pain related to pancreatic cancer is perhaps the most debilitating symptom associated with this disease and can lead to the deterioration of quality of life rather quickly. While only 30% to 40% of patients with pancreatic cancer report moderate to severe pain at the time of diagnosis, over 80% of patients with advanced cancer experience severe pain prior to death (49–52). A single institution prospective randomized controlled trial (Level Ib) has demonstrated that chemical splanchnicectomy can achieve acute pain relief in over 80% of patients and can prevent the subsequent onset of

Figure 43.1 Illustration of one operative technique for a palliative double bypass procedure for unresectable pancreatic cancer. Here, the hepaticojejunostomy (HJ) is shown as an end-to-side anastomosis with a retrocolic Roux-en-Y jejunal limb. The gastrojejunostomy (GJ) is depicted as a retrocolic side-to-side anastomosis between the most dependent aspect of the stomach and an isoperistaltic loop of proximal jejunum just beyond the ligament of Treitz.

(A)

(B)

Figure 43.2 Coronal section of a representative CT scan showing a metallic endostent providing adequate relief of duodenal obstruction from a pancreatic head cancer (A). Plain film radiography demonstrating the long-term patency of palliative metallic biliary and duodenal stents which can be deployed serially as combined or separate endoscopic procedures (B).

403

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
pain for up to 6 months postoperatively (53). Furthermore, patients with severe preoperative pain who undergo a palliative chemical splanchnicectomy experience a significant improvement in overall survival (Fig. 43.4). Comparisons Between Operative Versus Nonoperative Techniques Several prospective randomized studies have compared operative versus nonoperative procedures to palliate patients with malignant obstruction of the distal common bile duct (Table 43.1) (54–56). The findings of these individual studies, which have been corroborated by a larger meta-analysis (Level Ia), show no difference in patient survival based on treatment approach (20,30). Compared to nonoperative techniques which carry a lower short-term morbidity, mortality, hospital stay, and cost; the major advantage of operative biliary bypass is the lower incidence of late complications, namely recurrent jaundice and cholangitis (20,30). The relative benefits of operative palliation for biliary obstruction are most apparent for patients with low operative risk and reasonable life expectancy (e.g., greater than 6 months). It is often difficult to estimate the lifespan of an individual patient with a determined burden of tumor, and recent developments in palliative chemotherapy may further affect the natural history of advanced stage pancreatic cancer. Similarly, meta-analyses of relatively small-scale randomized prospective and comparative studies of endoscopic

Figure 43.3 A chemical splanchnicectomy can be accomplished by injecting 20 ml of 50% alcohol into the celiac ganglia on each side of the aorta (A) at the level of the celiac axis (CA). The use of a 22 gauge or smaller caliber spinal needle ensures containment of the injection wheal within the retroperitoneum.

Table 43.1 Randomized Trials of Operative Versus Nonoperative Palliation of Malignant Biliary Obstruction
Study Shepherd et al. Surgery Stent Andersen et al. Surgery Stent Smith et al. Surgery Stent Overall Surgery Stent
a

Year 1988

No. of patients 25 23

Treatment failure (%)a 8 9 4 4 7 5 7 5

Major complications (%) 56 30 20 36 29 11 32 18

Need for reintervention (%) 8 43 8 52 2 36 4 40

1989 25 25 2004 101 100 151 148

Inadequate biliary decompression.

Table 43.2 Randomized Trials of Operative Versus Nonoperative Palliation of Malignant Gastroduodenal Obstruction
Study Mehta et al. Laparoscopic GJ Stent Fiori et al. Open GJ Stent Overall Surgery Stent
a

Year 2006

No. of patients 14 13

Treatment success (%)a 93 85 89 100 91 91

Major complications (%) 57 0 11 0 39 0

2004 9 9 23 22

Oral intake by 2 weeks post-procedure.

404

PALLIATION OF PANCREAS CANCER
100
Alcohol pain (n = 20)

80 60

Saline pain (n = 14)

7. 8.

40 20 0 0 3 6 9 12

p = 0.0001

9.

10.
15 18 21 24 27 30 33 36

11.

Months of survival

Figure 43.4 In a prospective, randomized, double-blind study by Lillemoe et al. (53), chemical splanchnicectomy (alcohol) in patients with unresectable pancreatic cancer and preoperative pain resulted in a significant reduction in pain at 2-, 4-, and 6-month follow-up and a significant improvement in overall survival compared to placebo (saline) injection.

12.

13.

stenting versus surgical gastrojejunostomy have shown that while endoscopic stenting for the palliation of malignant gastroduodenal obstruction is associated with higher early clinical success (i.e., shorter time to oral intake and shorter length of hospital stay), operative gastric bypass procedures are preferable in patients with a prolonged prognosis who are likely to benefit from the reliable durability of surgical palliation that is less likely to require reintervention (Table 43.2) (41,57–59).

14.

15.

16.

17.

summary
Based on the existing clinical evidence, we advocate operative biliary decompression, gastric bypass, and chemical splanchnicectomy for all patients who undergo laparotomy without an indwelling metallic biliary stent and are found to have locally unresectable pancreatic carcinoma in the periampullary region. Even though this has not been studied directly, the durability of operative palliation should influence patients’ quality of life positively by decreasing the need for reinterventions and future hospitalizations. Symptomatic patients with preoperatively confirmed locally unresectable cancer or metastatic disease can be palliated reliably with nonoperative techniques. Such patients should be offered surgical palliation only when nonoperative procedures are unsuccessful and they have a reasonable life expectancy. To take advantage of the longterm benefits of surgical palliation, operative bypass procedures must be performed with acceptable morbidity.
18.

19. 20.

21. 22. 23. 24.

references
1. Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet 2004 Mar 27; 363(9414): 1049–57. 2. Niederhuber JE, Brennan MF, Menck HR. The National Cancer Data Base report on pancreatic cancer. Cancer 1995 Nov 1; 76(9): 1671–7. 3. Schneider G, Siveke JT, Eckel F, Schmid RM. Pancreatic cancer: basic and clinical aspects. Gastroenterology 2005 May; 128(6): 1606–25. 4. Wray CJ, Ahmad SA, Matthews JB, Lowy AM. Surgery for pancreatic cancer: recent controversies and current practice. Gastroenterology 2005 May; 128(6): 1626–41. 5. House MG, Yeo CJ, Cameron JL, et al. Predicting resectability of periampullary cancer with three-dimensional computed tomography. J Gastrointest Surg 2004 Mar–Apr; 8(3): 280–8. 6. Soriano A, Castells A, Ayuso C, et al. Preoperative staging and tumor resectability assessment of pancreatic cancer: prospective study comparing

25. 26.

27.

28.

29.

endoscopic ultrasonography, helical computed tomography, magnetic resonance imaging, and angiography. Am J Gastroenterol 2004 Mar; 99(3): 492–501. de Rooij PD, Rogatko A, Brennan MF. Evaluation of palliative surgical procedures in unresectable pancreatic cancer. Br J Surg 1991 Sep; 78(9): 1053–8. Di Fronzo LA, Cymerman J, Egrari S, O’Connell TX. Unresectable pancreatic carcinoma: correlating length of survival with choice of palliative bypass. Am Surg 1999 Oct; 65(10): 955–8. Gouma DJ, Nieveen van Dijkum EJ, van Geenen RC, van Gulik TM, Obertop H. Are there indications for palliative resection in pancreatic cancer? World J Surg 1999 Sep; 23(9): 954–9. Moore MJ. Pancreatic cancer: what the oncologist can offer for palliation. Can J Gastroenterol 2002 Feb; 16(2): 121–4. Cubiella J, Castells A, Fondevila C, et al. Prognostic factors in nonresectable pancreatic adenocarcinoma: a rationale to design therapeutic trials. Am J Gastroenterol 1999 May; 94(5): 1271–8. Lillemoe KD, Cameron JL, Hardacre JM, et al. Is prophylactic gastrojejunostomy indicated for unresectable periampullary cancer? A prospective randomized trial. Ann Surg 1999 Sep; 230(3): 322–8; discussion 8–30. Lillemoe KD, Cameron JL, Yeo CJ, et al. Pancreaticoduodenectomy. Does it have a role in the palliation of pancreatic cancer? Ann Surg 1996 Jun; 223(6): 718–25; discussion 25–8. Nieveen van Dijkum EJ, Kuhlmann KF, Terwee CB, et al. Quality of life after curative or palliative surgical treatment of pancreatic and periampullary carcinoma. Br J Surg 2005 Apr; 92(4): 471–7. Ridwelski K, Meyer F, Ebert M, Malfertheiner P, Lippert H. Prognostic parameters determining survival in pancreatic carcinoma and, in particular, after palliative treatment. Dig Dis 2001; 19(1): 85–92. van den Bosch RP, van der Schelling GP, Klinkenbijl JH, et al. Guidelines for the application of surgery and endoprostheses in the palliation of obstructive jaundice in advanced cancer of the pancreas. Ann Surg 1994 Jan; 219(1): 18–24. van der Schelling GP, van den Bosch RP, Klinkenbij JH, Mulder PG, Jeekel J. Is there a place for gastroenterostomy in patients with advanced cancer of the head of the pancreas? World J Surg 1993 Jan–Feb; 17(1): 128–32; discussion 32–3. Van Heek NT, De Castro SM, van Eijck CH, et al. The need for a prophylactic gastrojejunostomy for unresectable periampullary cancer: a prospective randomized multicenter trial with special focus on assessment of quality of life. Ann Surg 2003 Dec; 238(6): 894–902; discussion5. Costamagna G, Pandolfi M. Endoscopic stenting for biliary and pancreatic malignancies. J Clin Gastroenterol 2004 Jan; 38(1): 59–67. Taylor MC, McLeod RS, Langer B. Biliary stenting versus bypass surgery for the palliation of malignant distal bile duct obstruction: a metaanalysis. Liver Transpl 2000 May; 6(3): 302–8. Sarr MG, Cameron JL. Surgical management of unresectable carcinoma of the pancreas. Surgery 1982 Feb; 91(2): 123–33. Singh SM, Longmire WP Jr., Reber HA. Surgical palliation for pancreatic cancer. The UCLA experience. Ann Surg 1990 Aug; 212(2): 132–9. Watanapa P, Williamson RC. Surgical palliation for pancreatic cancer: developments during the past two decades. Br J Surg 1992 Jan; 79(1): 8–20. Espat NJ, Brennan MF, Conlon KC. Patients with laparoscopically staged unresectable pancreatic adenocarcinoma do not require subsequent surgical biliary or gastric bypass. J Am Coll Surg 1999 Jun; 188(6): 649–55; discussion 55–7. Gentileschi P, Kini S, Gagner M. Palliative laparoscopic hepatico- and gastrojejunostomy for advanced pancreatic cancer. JSLS 2002 Oct–Dec; 6(4): 331–8. Hamade AM, Al-Bahrani AZ, Owera AM, et al. Therapeutic, prophylactic, and preresection applications of laparoscopic gastric and biliary bypass for patients with periampullary malignancy. Surg Endosc 2005 Oct; 19(10): 1333–40. Nieveen van Dijkum EJ, Romijn MG, Terwee CB, et al. Laparoscopic staging and subsequent palliation in patients with peripancreatic carcinoma. Ann Surg 2003 Jan; 237(1): 66–73. Rhodes M, Nathanson L, Fielding G. Laparoscopic biliary and gastric bypass: a useful adjunct in the treatment of carcinoma of the pancreas. Gut 1995 May; 36(5): 778–80. Rothlin MA, Schob O, Weber M. Laparoscopic gastro- and hepaticojejunostomy for palliation of pancreatic cancer: a case controlled study. Surg Endosc 1999 Nov; 13(11): 1065–9.

%

405

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
30. Moss AC, Morris E, Leyden J, MacMathuna P. Malignant distal biliary obstruction: a systematic review and meta-analysis of endoscopic and surgical bypass results. Cancer Treat Rev 2007 Apr; 33(2): 213–21. 31. Naggar E, Krag E, Matzen P. Endoscopically inserted biliary endoprosthesis in malignant obstructive jaundice. A survey of the literature. Liver 1990 Dec; 10(6): 321–4. 32. Moss AC, Morris E, MacMathuna P. Palliative biliary stents for obstructing pancreatic carcinoma. Cochrane Database Syst Rev 2006(2): CD004200. 33. Arguedas MR, Heudebert GH, Stinnett AA, Wilcox CM. Biliary stents in malignant obstructive jaundice due to pancreatic carcinoma: a cost-effectiveness analysis. Am J Gastroenterol 2002 Apr; 97(4): 898–904. 34. Kaassis M, Boyer J, Dumas R, et al. Plastic or metal stents for malignant stricture of the common bile duct? Results of a randomized prospective study. Gastrointest Endosc 2003 Feb; 57(2): 178–82. 35. Prat F, Chapat O, Ducot B , et al. A randomized trial of endoscopic drainage methods for inoperable malignant strictures of the common bile duct. Gastrointest Endosc 1998 Jan; 47(1): 1–7. 36. Yeoh KG, Zimmerman MJ, Cunningham JT, Cotton PB. Comparative costs of metal versus plastic biliary stent strategies for malignant obstructive jaundice by decision analysis. Gastrointest Endosc 1999 Apr; 49(4 Pt 1): 466–71. 37. DiFronzo LA, Egrari S, O’Connell TX. Choledochoduodenostomy for palliation in unresectable pancreatic cancer. Arch Surg 1998 Aug; 133(8): 820–5. 38. Sarfeh IJ, Rypins EB, Jakowatz JG, Juler GL. A prospective, randomized clinical investigation of cholecystoenterostomy and choledochoenterostomy. Am J Surg 1988 Mar; 155(3): 411–4. 39. Urbach DR, Bell CM, Swanstrom LL, Hansen PD. Cohort study of surgical bypass to the gallbladder or bile duct for the palliation of jaundice due to pancreatic cancer. Ann Surg 2003 Jan; 237(1): 86–93. 40. Holt AP, Patel M, Ahmed MM. Palliation of patients with malignant gastroduodenal obstruction with self-expanding metallic stents: the treatment of choice? Gastrointest Endosc 2004 Dec; 60(6): 1010–7. 41. Jeurnink SM, van Eijck CH, Steyerberg EW, Kuipers EJ, Siersema PD. Stent versus gastrojejunostomy for the palliation of gastric outlet obstruction: a systematic review. BMC Gastroenterol 2007; 7: 18. 42. Kozarek RA, Ball TJ, Patterson DJ. Metallic self-expanding stent application in the upper gastrointestinal tract: caveats and concerns. Gastrointest Endosc 1992 Jan–Feb; 38(1): 1–6. 43. Espinel J, Vivas S, Munoz F, Jorquera F, Olcoz JL. Palliative treatment of malignant obstruction of gastric outlet using an endoscopically placed enteral Wallstent. Dig Dis Sci 2001 Nov; 46(11): 2322–4. 44. Lopera JE, Brazzini A, Gonzales A, Castaneda-Zuniga WR. Gastroduodenal stent placement: current status. Radiographics 2004 Nov–Dec; 24(6): 1561–73. 45. Sohn TA, Lillemoe KD, Cameron JL, et al. Surgical palliation of unresectable periampullary adenocarcinoma in the 1990s. J Am Coll Surg 1999 Jun; 188(6): 658–66; discussion 66–9. 46. Strong VE, Dalal KM, Malhotra VT, et al. Initial report of laparoscopic celiac plexus block for pain relief in patients with unresectable pancreatic cancer. J Am Coll Surg 2006 Jul; 203(1): 129–31. 47. Gress F, Schmitt C, Sherman S, Ikenberry S, Lehman G. A prospective randomized comparison of endoscopic ultrasound- and computed tomography-guided celiac plexus block for managing chronic pancreatitis pain. Am J Gastroenterol 1999 Apr; 94(4): 900–5. 48. Pietrabissa A, Vistoli F, Carobbi A, et al. Thoracoscopic splanchnicectomy for pain relief in unresectable pancreatic cancer. Arch Surg 2000 Mar; 135(3): 332–5. 49. Kalser MH, Barkin J, MacIntyre JM. Pancreatic cancer. Assessment of prognosis by clinical presentation. Cancer 1985 Jul 15; 56(2): 397–402. 50. Lichtenstein DR, Carr-Locke DL. Endoscopic palliation for unresectable pancreatic carcinoma. Surg Clin N Am 1995 Oct; 75(5): 969–88. 51. House MG, Choti MA. Palliative therapy for pancreatic/biliary cancer. Surg Oncol Clin N Am 2004 Jul; 13(3): 491–503, ix. 52. Kelsen DP, Portenoy RK, Thaler HT, et al. Pain and depression in patients with newly diagnosed pancreas cancer. J Clin Oncol 1995 Mar; 13(3): 748–55. 53. Lillemoe KD, Cameron JL, Kaufman HS, et al. Chemical splanchnicectomy in patients with unresectable pancreatic cancer. A prospective randomized trial. Ann Surg 1993 May; 217(5): 447–55; discussion 56–7. 54. Andersen JR, Sorensen SM, Kruse A, Rokkjaer M, Matzen P. Randomised trial of endoscopic endoprosthesis versus operative bypass in malignant obstructive jaundice. Gut 1989 Aug; 30(8): 1132–5. 55. Shepherd HA, Royle G, Ross AP,et al. Endoscopic biliary endoprosthesis in the palliation of malignant obstruction of the distal common bile duct: a randomized trial. Br J Surg 1988 Dec; 75(12): 1166–8. 56. Smith AC, Dowsett JF, Russell RC, Hatfield AR, Cotton PB. Randomised trial of endoscopic stenting versus surgical bypass in malignant low bileduct obstruction. Lancet 1994 Dec 17; 344(8938): 1655–60. 57. Fiori E, Lamazza A, Volpino P, Burza A, Paparelli C, et al. Palliative management of malignant antro-pyloric strictures. Gastroenterostomy vs. endoscopic stenting. A randomized prospective trial. Anticancer Res 2004 Jan–Feb; 24(1): 269–71. 58. Mehta S, Hindmarsh A, Cheong E, et al. Prospective randomized trial of laparoscopic gastrojejunostomy versus duodenal stenting for malignant gastric outflow obstruction. Surg Endosc 2006 Feb; 20(2): 239–42. 59. Hosono S, Ohtani H, Arimoto Y, Kanamiya Y. Endoscopic stenting versus surgical gastroenterostomy for palliation of malignant gastroduodenal obstruction: a meta-analysis. J Gastroenterol 2007 Apr; 42(4): 283–90.

406

44 Cystic tumors of the pancreas
introduction

Peter J. Allen and Murray F. Brennan
Scenario 2. Otherwise healthy 63-year-old female with episode of severe bronchitis who underwent CT imaging of the chest. No significant pulmonary abnormality noted on CT; however, cystic lesion noted in the tail of the pancreas. Dedicated pancreatic imaging revealed 3.2 cm cyst in the tail of the pancreas and mild diffuse main pancreatic ductal dilatation (Fig. 44.2). No further diagnostic testing was performed. Scenario 3. Otherwise healthy 60-year-old female with nonspecific abdominal pains who underwent CT imaging of the abdomen. A 1.5-cm cyst noted in the body of the pancreas (Fig. 44.3). Review of imaging reveals no solid component to the cyst and no main pancreatic ductal dilatation. An upper endoscopy with endoscopic ultrasound was performed which confirmed the CT-imaging findings. Fine needle aspiration was performed with cytology revealing non-diagnostic material. Cyst fluid CEA was 2853 ng/ml.

A cystic tumor of the pancreas is a radiographic finding that has a broad histologic differential. This differential includes non-neoplastic pseudocysts, benign neoplastic cysts (serous cysts), pre-malignant cysts (mucinous cysts), and cystic lesions with invasive carcinoma (Table 44.1) (1). The current ability to determine the histologic diagnosis of these lesions without resection is improving but limited (2,3). Serum testing, radiographic imaging, endoscopic ultrasound, cyst fluid cytology, and cyst fluid marker analysis (CEA) have a combined overall accuracy of approximately 70% to 85% (4). This diagnostic limitation can make treatment decisions difficult: resection may provide the only form of cure in those patients with highrisk mucinous cysts or very early invasive disease; however, resection has a major morbidity rate of approximately 20% and mortality rate of 2% to 10%. Resection of benign cysts will expose the asymptomatic patient to all the risks of resection without identified benefit. As cross-sectional imaging improves this diagnostic and treatment dilemma will become more common. The increased use of high-quality cross-sectional imaging has resulted in an increasing number of patients identified with small (<3 cm) asymptomatic cysts, and the ability to determine histology in these patients is even more difficult (5–7). The natural history of these small, incidentally discovered lesions is unknown. Even in a patient with a small and asymptomatic pre-malignant cyst (mucinous) the future risk of progression to malignancy has not been determined with respect to both frequency and duration. The decision to resect a small, asymptomatic mucinous lesion, particularly in someone who is elderly or with significant comorbidities, must take into account the fact that the natural history of that lesion is unknown.

pathologic sub-types and clinical behavior
Although the differential diagnosis of a cystic lesion of the pancreas is broad, over 85% of resected lesions will represent serous or mucinous cysts (8). This section will focus on these more common lesions. Pancreatic Pseudocyst The most common cystic lesion of the pancreas is the non-neoplastic inflammatory pseudocyst that develops as a complication of pancreatitis (4,9). A pancreatic pseudocyst is a fluid collection that arises in or adjacent to the pancreas but lacks an epithelial lining. Pseudocysts have been reported to develop in 15% to 50% of patients who experience acute pancreatitis (9,10). Because of this, some studies have reported that pseudocysts represent 75% to 85% of all cystic lesions of the pancreas (9,10). Pseudocysts are a well-characterized complication of pancreatitis and are not the focus of this chapter. Pseudocysts are managed expectantly or with percutaneous, endoscopic, or operative drainage. Serous Cystadenoma Serous cystadenoma of the pancreas was first characterized in detail by Compagno and Oertel in 1978 (11,12). In this report they described the gross (microcystic) and microscopic (glycogen rich) characteristics of SCA and differentiated the appearance and behavior of these lesions from mucinous cysts of the pancreas. Serous cystadenomas may range in diameter from 1 cm to over 20 cm and are grossly characterized by thick fibrous walls and septa, innumerable small cysts (<1 cm in diameter) containing thin clear fluid, and often a calcified central scar which may or may not contain focal hemorrhage (Fig. 44.4). Microscopically these lesions are characterized by

clinical scenarios—presentation and diagnostic evaluation
The clinical scenarios below are from three patients with cystic lesions of the pancreas treated at our institution over the past 5 years. The presentation and diagnostic evaluation are presented. Treatment decisions and rationale are provided in the last section of this chapter. Scenario 1. Otherwise healthy 66-year-old female who underwent CT imaging of the abdomen for left lower quadrant abdominal pain. Symptoms resolved without treatment and imaging revealed no evidence of diverticulitis or other left abdominal/pelvic pathology. Imaging did reveal 4.5 cm cystic lesion in pancreatic tail (Fig. 44.1). Review of CT revealed imaging characteristic of serous cystadenoma (microcystic, central calcification). No further diagnostic testing was performed.

407

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
a bland cuboidal epithelial cell lining which is typically devoid of nuclear polymorphism or mitotic activity (Fig. 44.4). Glycogen is abundant in the cytoplasm and can occasionally be detected within the amorphous cystic material. In general SCA are considered to be benign as there are fewer than 10 cases of metastatic serous cystadenocarcinoma within the world’s literature (13). Many of the case reports that describe “malignant” SCAs describe local invasion rather than metastatic spread, and therefore the true incidence of malignancy within serous cystadenomas is certainly less than 1%. Within our institutional database there are currently over 120 patients who have undergone resection for serous cystadenoma, and not a single case of metastatic spread has been documented. The more common clinical problem with serous cystadenoma is local growth and the subsequent development of symptoms such as pain or jaundice. The presence of symptoms (pain) appears to be related to the size of the lesion as studies describing patients with larger tumors tend to have a greater percentage of patients with symptoms. Compagno and Oertel’s study in 1978 of 34 cases of serous cystadenoma reported an average tumor diameter of 10.8 cm, and 71% were symptomatic (11). In a previous report from our institution the average tumor diameter of patients with resected serous lesions was 4.9 cm, and 35% were symptomatic (14). Because the exact size at which a serous lesion will become symptomatic is unknown, and because the growth rate of serous cystadenomas has not been defined, the appropriate management of the young patient with an asymptomatic serous lesion is yet to be determined. Tseng et al. reported a growth rate of 0.6 cm/year for patients with serous cystadenoma (15). In Tseng’s report, serial radiography was obtained

Table 44.1 Kloppel’s Classification of Cystic Neoplasms of the Pancreas
Kloppel’s classification Epithelial tumors Serous cystadenoma Mucinous cystadenoma Cystadenocarcinoma Intraductal papillary mucinous tumor Pseudopapillary and solid tumor Teratoma Acinous cystadenocarcinoma Adenosquamous carcinoma Mucinous cystic adenocarcinoma Cystic islet cell tumor Vascular tumor Lymphangioma Hemangiopericytoma Leiomyosarcoma Lymphoma Single cyst Polycystic disease Exclusively pancreatic With hepatorenal disease Von Hippel–Lindau disease

Non-epithelial tumors

Pseudotumors

Figure 44.2 Cystic lesion in the tail of the pancreas (solid arrow) and dilated main pancreatic duct (broken arrow) in an otherwise healthy 63-year-old female.

Figure 44.1 Cystic lesion in the tail of the pancreas (arrow) in otherwise healthy 66-year-old female.

Figure 44.3 Cystic lesion in the body of the pancreas (arrow) in an otherwise healthy 60-year-old female.

408

CYSTIC TUMORS OF THE PANCREAS
for a group of 24 patients who had a median radiographic follow-up of 23 months. There was a significant difference in growth rates between patients with tumors <4 cm at presentation (0.48 cm/year) and patients with lesions ≥4 cm (1.98 cm/year). Because of the observed increased rate of growth in larger lesions, this report recommended resection for asymptomatic patients with serous cystadenomas >4 cm. We have previously reported a similar overall growth rate of approximately 0.5 cm/year, but have not found any association between the size of the lesion and the rate of growth. We feel that asymptomatic patients can be safely followed with the possible exception of those patients who have large lesions that are marginally resectable. Intraductal Papillary Mucinous Neoplasm (IPMN) Intraductal papillary mucinous neoplasms (IPMN) are mucinous cystic tumors of the pancreas which were first classified into a unified diagnosis by the World Health Organization in 1996 (16). Prior to this these neoplasms were described under a variety of names including mucinous ductal ectasia, papillary carcinoma, and villous adenoma. Because of the lack of a unifying diagnosis, older reports evaluating mucinous cysts of the pancreas may actually represent a combination of both IPMN and mucinous cystic neoplasm (MCN) which is a distinct histopathologic entity. Grossly, IPMNs are characterized by ductal dilatation and mucin production (Fig. 44.5). IPMN is considered a “whole gland” process; however, radiographically apparent disease may be evident in the main pancreatic duct alone, the branch ducts alone, or both. Microscopically, IPMN lesions are characterized by papillary projections of columnarlined epithelium with varying degrees of dysplasia (Fig. 44.5). Mucin is typically abundant both within the cytoplasm of the lining epithelial cells as well as within the acellular fluid matrix. Current nomenclature (WHO) divides these lesions into the categories of adenoma, borderline (lowgrade dysplasia), high-grade dysplasia (carcinoma in situ), and carcinoma.

(A)

(B) Figure 44.4 Gross and microscopic characteristics of serous cystadenoma. Arrow notes central scar.

BD

PD

(A)

(B) Figure 44.5 Gross and microscopic characteristics of main duct IPMN. Abbreviations: PD, pancreatic duct; BD, bile duct.

409

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Invasive malignancy is well documented for this group of lesions. Several large series of resected IPMN have been reported in the literature (17,18). In a series from Johns Hopkins of 136 patients who underwent resection for IPMN there was invasive carcinoma identified in 38% of patients and an additional 55% of patients had in situ carcinoma (18). Similar results have been reported from our institution (17). D’Angelica et al. reported on 62 patients with resected IPMN of the pancreas, and in this series the prevalence of invasive carcinoma was 48% (n = 30), and the prevalence of in situ carcinoma was 27% (n = 17). The presence of malignancy has been found to be associated with the presence of main duct disease (vs. branch duct), as well as with the radiographic characteristics of a solid component and cyst size (14,18–20). Recent reports from the Massachusetts General Hospital, as well as from our own institution, have failed to identify invasive malignancy in small (<3 cm) branch duct IPMNs of the pancreas (14,20). The frequency and length of time it takes for IPMN to progress to malignancy are unknown. Increased age has been reported to be associated with malignancy in IPMN in several studies (18,19). Because of this association, a report from Johns Hopkins concluded that the lag time from adenoma to carcinoma in IPMN was approximately 5 years (18). In a recent study from our institution we also found that patients with IPMN adenoma were significantly younger than those with carcinoma (65 vs. 74 years, p = 0.02). We feel that this finding likely represents the course of progression from benign to malignant; however, the timing of this progression and if it occurs in even the majority of lesions remain unknown. Within our institutional cyst registry we have identified approximately 20 patients with small IPMNs of the pancreas who have been followed radiographically over a period of 5 to 120 months. None of these patients have been noted to have significant growth of the lesions or other evidence of the development of malignancy. Initial reports with limited follow-up suggested a significantly improved survival for patients with malignant IPMN as compared to patients with conventional pancreatic adenocarcinoma (19,21). These differences have become smaller as series with longer follow-up are reported; however, there does appear to be a biologic spectrum in the aggressiveness of invasive IPMN (17,22). The rate of nodal positivity is typically lower (33% to 54%) than what is seen after resection for conventional pancreatic adenocarcinoma, and some histopathologic sub-types, such as colloid papillary mucinous carcinoma, appear to have a more favorable long-term outcome (17). After resection for non-invasive IPMN distant recurrence should not occur; however, these patients have been found to be at risk for recurrence within the pancreatic remnant. A study from the Mayo clinic reported an 8% gland recurrence rate after partial pancreatectomy for non-invasive IPMN with a median follow-up of 37 months (22). These results are similar to those observed at our institution and highlight the need to follow patients for a long term after resection of non-invasive IPMN for the development of disease within the pancreatic remnant (23). Mucinous Cystic Neoplasm Current histopathologic data support the distinction between IPMN and MCN of the pancreas (12,24,25). MCNs are much less common than IPMN, and are defined as tumors that lack communication with the pancreatic ductal system, contain a mucin-producing columnar epithelium, and are supported by ovarian-like stroma. The ovarian stroma is a unique and defining feature of MCNs of the pancreas, is the presumed reason that MCNs are almost exclusively identified in women, and is a characteristic that pancreatic MCNs share with mucinous cysts of the ovary and liver (24,26). These lesions are most commonly located in the body and tail and can range in size from 2 to 25 cm (12,24). Grossly these tumors are round with a smooth surface and fibrous pseudocapsule (Fig. 44.6). MCNs may also progress to a malignant process and the reported malignancy rate in most large series has ranged between 10% and 50% (12,24,27). This rate may be underestimated as both benign and malignant epithelium may coexist within the same cyst and thus extensive histologic sampling and assessment are necessary. Factors found associated with the presence of malignancy in MCN of the pancreas have included the presence of septations, mural nodularity, and cyst

PD

Cyst

(A)

(B) Figure 44.6 Gross and microscopic characteristics of MCN. Abbreviation: PS, pancreatic duct.

410

CYSTIC TUMORS OF THE PANCREAS
size (14,24). Because these tumors are uncommon the natural history of mucinous cystadenocarcinoma has not been well defined. Patients with extension of malignancy beyond the tumor capsule have been shown to be at risk for recurrence and death from disease (24). The limitations to cyst fluid cytology are the result of the typically small volume and low cellular content of the aspirates, and the contamination of the samples with mucin and mucin-producing cells from the stomach or duodenum through which the needle is passed. The typical radiographic appearance of a serous cystadenoma is of a spherical lesion, with multiple small cysts, and central calcification. Because of the fibrous nature of these lesions a solid component is often described, and in the setting of other findings that are characteristic for SCA should not be viewed as concerning for malignancy. Like all cystic lesions, some SCA will present with atypical radiographic findings. Oligocystic SCA is a recently identified variant of SCA with a radiographic appearance that is indistinguishable from MCN or branch-duct IPMN (Fig. 44.5) (36). The radiographic appearance of IPMN is dependent on whether the lesion is predominantly involving the main duct, branch ducts, or both. Main duct IPMN will characteristically present with diffuse ductal dilatation. Any solid component or focal mass within these lesions should be viewed as concerning for malignancy. Branch duct IPMN may be unilocular or multilocular and there is by definition no dilation of the main pancreatic duct. In the absence of septations, solid component, or mural nodularity these lesions may be indistinguishable from MCN, retention cysts, small cystic endocrine tumors, or even pseudocysts. Any macrocystic lesion in the tail of the pancreas in a female patient should be suspected as an MCN. These lesions are typically several centimeters in diameter, solitary, and there should not be dilation of the main pancreatic duct. Peripheral calcifications, described as eggshell calcifications may be present. Any mural nodularity or solid component should be viewed as concerning for malignancy.

diagnostic evaluation
High-quality cross-sectional imaging is essential for the evaluation of patients with cystic lesions of the pancreas. Multidetector CT (MdCT) allows thin section scanning of the pancreas and has become the most common method for assessing pancreatic cysts (28). MdCT has the ability to provide excellent visualization of septa, mural nodules, and calcifications. MdCT also allows excellent visualization and characterization of the pancreatic parenchyma. Evaluation of the parenchyma adjacent to the cyst is critical as we have recently reported on several patients with pancreatic adenocarcinoma who presented with isolated small retention cysts adjacent to a radiographically occult malignancy (14). MRCP also provides excellent characterization of cyst morphology (4). MRCP may also allow for the ability to diagnose branch duct IPMN through identification of communication between the cyst and the pancreatic duct (29). Endoscopic evaluation with endoscopic ultrasound (EUS) has played an increasingly important role in the evaluation of pancreatic cysts. In general, endoscopy with or without endoscopic retrograde cholangiopancreatography (ERCP) has a limited role in the evaluation of pancreatic cysts; however, these tests may have indications in the evaluation of suspected IPMN. Endoscopic ultrasound (EUS) with or without cyst aspiration is highly operator dependent, but the information gained from EUS by an experienced gastroenterologist can be very valuable. EUS can provide detailed images of the cyst wall as well as internal cyst architecture and can be used to perform fine needle aspiration biopsy. The fluid obtained by EUS FNA can be used both for cytologic analysis as well as for various tumor marker analyses. The diagnostic utility of cyst fluid analysis has been studied extensively (4,30–32). A variety of tumor markers including CA19-9, CEA, CA15-3, M1 mucin, and amylase have been evaluated. The most consistent results have been reported for cyst fluid CEA levels. In a prospective study by Brugge et al. of 112 patients with cystic lesions, an elevated cyst fluid CEA level (>192 ng/ml) was the best predictor of a mucinous lesion and accurately identified these lesions in 79% of cases (3). Elevated CEA levels and the presence of extracellular mucin have been shown to have a positive predictive value for mucinous lesions as high as 85% (4,33,34). The degree to which the cyst fluid CEA level is elevated has not been found to be predictive of malignancy within mucinous cysts. Serous cystadenomas and retention cysts have been shown to have almost uniformly undetectable cyst fluid CEA levels (4,34,35). The ability of cyst fluid cytology to differentiate between serous versus mucinous cysts as well to identify malignancy within the mucinous sub-group is limited. Most studies have shown accuracy rates of cyst fluid cytology in the range of 50% and thus cytology is probably inferior to cyst fluid CEA alone in discriminating between serous and mucinous cysts (3).

treatment recommendations
Because of the frequent inability to determine histology without resection, and because of the unknown natural history of some cystic sub-types, many authors have recommended routine resection of all pancreatic cysts (2,37,38). These authors argue that because the preoperative distinction between benign and malignant is unreliable, and because the potential adverse consequences of not resecting a pre-malignant or malignant cyst are significant, all medically fit patients should undergo resection. Although this approach provides a guarantee to patients that no pre-malignant or malignant lesions will be observed, it exposes patients with benign lesions to the risks of pancreatectomy. Several recent reports, including a study from our own institution, have recommended a more selective approach to resection (39–41). Proponents of this approach argue that with current imaging techniques, and with an improved understanding of the various histologic entities, a group of patients can be identified who have an extremely low risk of malignancy. Most reports evaluating this approach have recommended non-operative management (radiographic follow-up) for selected patients with small, incidentally discovered cysts of the pancreas that do not have a solid component or other concerning clinical or radiographic features of malignancy

411

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
such as main pancreatic ductal dilatation (14,39,40). This approach avoids the risks of operation in patients with benign lesions, but with current limitations in non-resectional diagnosis cannot guarantee that a malignancy is not mistakenly being observed. In some instances the histopathology of a given cyst can be determined with a high level of certainty without resection. In these instances treatment recommendations can be made based on the known natural history of the specific histologic entity. In other circumstances, typically patients with small cysts, the exact histopathology of the lesion cannot be determined without resection. In these instances treatment recommendations must be based on the radiographic characteristics and the inferred histopathology once the diagnostic work-up is complete. Serous Cystadenoma When diagnostic evaluation identifies a patient with a serous cystadenoma, resection should be reserved for the symptomatic patient or in a healthy patient in whom significant growth has been observed. In the asymptomatic patient the risk of mortality from resection exceeds the risk of malignancy. As noted above, data from our institution as well as others confirm the non-metastatic nature of serous cystadenomas. However, these lesions can become symptomatic and resection remains indicated in the presence of symptoms. Intraductal Papillary Mucinous Neoplasm (IPMN) and Mucinous Cystic Neoplasm (MCN) Resection has been previously recommended for all patients with IPMN of the pancreas. These recommendations should in general be considered because of the previously reported high rate of malignancy within these lesions as well as the ability of non-invasive IPMN to progress to invasive malignancy. When cross-sectional imaging and endoscopic studies are characteristic of main duct IPMN, and/or when there are any concerning radiographic features such as a solid component, septations, or size >3 cm our standard approach is to perform resection (14). The most difficult clinical scenario is the management of the patient who presents with a small branch duct IPMN, particularly when it arises in the head of the pancreas of a 50-year-old patient. A recent review of our institutional experience with these lesions identified the size of the lesion to be associated with the presence of malignancy as well as with the decision to recommend operative or non-operative management (14). We have not identified invasive malignancy in any mucinous lesion less than 3 cm in diameter in the absence of solid component, symptoms, or main ductal dilatation. Multiple other studies have also not shown invasive disease in small (<3 cm) mucinous cysts of the pancreas and a recent consensus statement supported a non-operative approach in patients with small mucinous cysts of this nature (42). Our typical follow-up schedule for patients undergoing non-operative management consists of highquality cross-sectional imaging every 6 months for 2 years and then annually thereafter. Resection is typically performed when there is any significant growth in the lesion, or the development of a solid component, or other concerning features of malignancy.

clinical scenarios: treatment
Scenario 1. The radiographic features of this lesion are characteristic of serous cystadenoma and therefore no further diagnostic evaluation was performed. Radiographic follow-up was recommended because the patient was asymptomatic and the lesion was not marginally resectable. The patient has now been imaged (MRCP) annually for 4 years and there has been no growth of the lesion and the patient remains asymptomatic. Scenario 2. The radiographic features of this lesion are characteristic of combined branch and main duct IPMN, because of this no further testing was performed. Because of the main duct dilation operative resection was recommended. The risks and benefits of several resectional procedures (distal pancreatectomy, total pancreatectomy) were discussed in detail with the patient. Because the majority of the cystic disease was in the pancreatic tail and because the pancreatic head was relatively spared a laparoscopic spleen-sparing distal pancreatectomy was performed. The patient had low-grade dysplasia within the resected specimen. Follow-up at 2 years reveals no evidence of progressive disease within the head of the gland. Scenario 3. EUS was performed because of the radiographic ambiguity of the lesion and the young age of the patient. The CEA within the aspirated fluid was characteristic of a mucinous lesion and this patient almost certainly has a small branch duct IPMN. Radiographic follow-up has been recommended and the patient has had no significant change within this lesion after 2 years.

summary
In summary, many institutions are now reporting a selective approach to resection in patients with cystic lesions of the pancreas. Routine resection of all pancreatic cysts is currently impractical, and given the large numbers of patients being identified with <2 cm lesions this approach would result in a mortality rate that is much higher than the rate of malignancy. Most studies that have advocated a selective approach have reported the radiographic characteristics of main duct dilatation, a solid component, cyst size, and symptoms to be associated with treatment recommendations. In the absence of these findings we feel that radiographic follow-up is warranted. In the young patient with a small mucinous tumor the additional factors are the likelihood of progression to malignancy and the patient anxiety about radiographic follow-up. No data are available for the former. Efforts should be made to improve the ability to distinguish histopathologic sub-types without resection. The current challenges are to improve the sensitivity and specificity for the identification of mucinous sub-type, to better characterize the progression of IPMN and mucinous cystic tumors, and to develop better methods for identifying the presence of in situ or invasive disease in these patients. Continued improvements in cross-sectional imaging and endoscopic techniques, and further investigation into markers in the serum and cyst fluid, should allow better stratification of mucinous sub-types.

412

CYSTIC TUMORS OF THE PANCREAS

references
1. Kloppel G, Kosmahl M. Cystic lesions and neoplasms of the pancreas. The features are becoming clearer. Pancreatology 2001; 1: 648–55. 2. Ooi LL, Ho GH, Chew SP, et al. Cystic tumours of the pancreas: a diagnostic dilemma. Aust N Z J Surg 1998; 68: 844–6. 3. Brugge WR, Lewandrowski K, Lee-Lewandrowski E, et al. Diagnosis of pancreatic cystic neoplasms: a report of the cooperative pancreatic cyst study. Gastroenterology 2004; 126: 1330–6. 4. Brugge WR, Lauwers GY, Sahani D, et al. Cystic neoplasms of the pancreas. N Engl J Med 2004; 351: 1218–26. 5. Gorin AD, Sackier JM. Incidental detection of cystic neoplasms of the pancreas. Md Med J 1997; 46: 79–82. 6. Fernandez-Del CC, Targarona J, Thayer SP, et al. Incidental pancreatic cysts: clinicopathologic characteristics and comparison with symptomatic patients. Arch Surg 2003; 138: 427–3. 7. Laffan TA, Horton KM, Klein AP, et al. Prevalence of unsuspected pancreatic cysts on MDCT. AJR 2008; 191: 802–7. 8. Le Borgne J., De Calan L., Partensky C. Cystadenomas and cystadenocarcinomas of the pancreas: a multiinstitutional retrospective study of 398 cases. French Surgical Association. Ann Surg 1999; 230: 152–61. 9. Grace PA, Williamson RC. Modern management of pancreatic pseudocysts. Br J Surg 1993; 80: 573–81. 10. O’Malley VP, Cannon JP, Postier RG. Pancreatic pseudocysts: cause, therapy, and results. Am J Surg 1985; 150: 680–2. 11. Compagno J, Oertel JE. Microcystic adenomas of the pancreas (glycogenrich cystadenomas): a clinicopathologic study of 34 cases. Am J Clin Pathol 1978; 69: 289–98. 12. Compagno J, Oertel JE. Mucinous cystic neoplasms of the pancreas with overt and latent malignancy (cystadenocarcinoma and cystadenoma). A clinicopathologic study of 41 cases. Am J Clin Pathol 1978; 69: 573–80. 13. Matsumoto T, Hirano S, Yada K, et al. Malignant serous cystic neoplasm of the pancreas: report of a case and review of the literature. J Clin Gastroenterol 2005; 39: 253–6. 14. Allen PJ, D’Angelica M, Gonen M, et al. A selective approach to the resection of cystic lesions of the pancreas: results from 539 consecutive patients. Ann Surg 2006; 244: 572–82. 15. Tseng JF, Warshaw AL, Sahani DV, et al. Serous cystadenoma of the pancreas: tumor growth rates and recommendations for treatment. Ann Surg 2005; 242: 413–9. 16. Kloppel G, Solcia E, Longnecker DS. World Health Organization International Classification of Tumors. Berlin: Springer, 1996. 17. D’Angelica M, Brennan MF, Suriawinata AA, et al. Intraductal papillary mucinous neoplasms of the pancreas: an analysis of clinicopathologic features and outcome. Ann Surg 2004; 239: 400–8. 18. Sohn TA, Yeo CJ, Cameron JL, et al. Intraductal papillary mucinous neoplasms of the pancreas: an updated experience. Ann Surg 2004; 239: 788–97. 19. Salvia R, Fernandez-Del CC, Bassi C, et al. Main-duct intraductal papillary mucinous neoplasms of the pancreas: clinical predictors of malignancy and long-term survival following resection. Ann Surg 2004; 239: 678–85. 20. Sahani DV, Saokar A, Hahn PF, et al. Pancreatic cysts 3 cm or smaller: how aggressive should treatment be? Radiology 2006; 238: 912–19. 21. Sohn TA, Yeo CJ, Cameron JL, et al. Intraductal papillary mucinous neoplasms of the pancreas: an increasingly recognized clinicopathologic entity. Ann Surg 2001; 234: 313–21. 22. Chari ST, Yadav D, Smyrk TC, et al. Study of recurrence after surgical resection of intraductal papillary mucinous neoplasm of the pancreas. Gastroenterology 2002; 123: 1500–7.

23. White R, D’Angelica M, Tang L, et al. The fate of the remnant pancreas following resection of non-invasive intraductal papillary mucinous neoplasm. J Am Coll Surg 2007; 204(5): 987–93. 24. Zamboni G, Scarpa A, Bogina G, et al. Mucinous cystic tumors of the pancreas: clinicopathological features, prognosis, and relationship to other mucinous cystic tumors. Am J Surg Pathol 1999; 23: 410–22. 25. Yamada M, Kozuka S, Yamao K, et al. Mucin-producing tumor of the pancreas. Cancer 1991; 68: 159–68. 26. Erdogan D, Lamers WH, Offerhaus GJ, et al. Cystadenomas with ovarian stroma in liver and pancreas: an evolving concept. Dig Surg 2006; 23: 186–91. 27. Warshaw AL, Compton CC, Lewandrowski K, et al. Cystic tumors of the pancreas. New clinical, radiologic, and pathologic observations in 67 patients. Ann Surg 1990; 212: 432–43. 28. Sahani DV, Kadavigere R, Saokar A, et al. Cystic pancreatic lesions: a simple imaging-based classification system for guiding management. Radiographics 2005; 25: 1471–84. 29. Koito K, Namieno T, Ichimura T, et al. Mucin-producing pancreatic tumors: comparison of MR cholangiopancreatography with endoscopic retrograde cholangiopancreatography. Radiology 1998; 208: 231–7. 30. Sand JA, Hyoty MK, Mattila J, et al. Clinical assessment compared with cyst fluid analysis in the differential diagnosis of cystic lesions in the pancreas. Surgery 1996; 119: 275–80. 31. Hammel PR, Forgue-Lafitte ME, Levy P, et al. Detection of gastric mucins (M1 antigens) in cyst fluid for the diagnosis of cystic lesions of the pancreas. Int J Cancer 1997; 74: 286–90. 32. Hammel P, Levy P, Voitot H, et al. Preoperative cyst fluid analysis is useful for the differential diagnosis of cystic lesions of the pancreas. Gastroenterology 1995; 108: 1230–5. 33. Walsh RM, Henderson JM, Vogt DP, et al. Prospective preoperative determination of mucinous pancreatic cystic neoplasms. Surgery 2002; 132: 628–33. 34. Lewandrowski KB, Southern JF, Pins MR, et al. Cyst fluid analysis in the differential diagnosis of pancreatic cysts. A comparison of pseudocysts, serous cystadenomas, mucinous cystic neoplasms, and mucinous cystadenocarcinoma. Ann Surg 1993; 217: 41–7. 35. van der Waaij LA, van Dullemen HM, Porte RJ. Cyst fluid analysis in the differential diagnosis of pancreatic cystic lesions: a pooled analysis. Gastrointest Endosc 2005; 62: 383–9. 36. Goh BK, Tan YM, Yap WM, et al. Pancreatic serous oligocystic adenomas: clinicopathologic features and a comparison with serous microcystic adenomas and mucinous cystic neoplasms. World J Surg 2006; 30: 1553–9. 37. Horvath KD, Chabot JA. An aggressive resectional approach to cystic neoplasms of the pancreas. Am J Surg 1999; 178: 269–74. 38. Siech M, Tripp K, Schmidt-Rohlfing B, et al. Cystic tumours of the pancreas: diagnostic accuracy, pathologic observations and surgical consequences. Langenbecks Arch Surg 1998; 383: 56–61. 39. Spinelli KS, Fromwiller TE, Daniel RA, et al. Cystic pancreatic neoplasms: observe or operate. Ann Surg 2004; 239: 651–7. 40. Walsh RM, Vogt DP, Henderson JM, et al. Natural history of indeterminate pancreatic cysts. Surgery 2005; 138: 665–70. 41. Allen PJ, Jaques DP, D’Angelica M, et al. Cystic lesions of the pancreas: selection criteria for operative and nonoperative management in 209 patients. J Gastrointest Surg 2003; 7: 970–7. 42. Tanaka M, Chari S, Adsay V, et al. International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas. Pancreatology 2006; 6: 17–32.

413

45

Neuroendocrine pancreatic tumors Steven N. Hochwald and Kevin Conlon
hormone produced may be too small to cause symptoms. Third, the tumor may secrete a precursor hormone that is functionally inert or the hormonal product of the tumor may not yet be identified. Pancreatic endocrine tumors may occur at any age, although they are rare in children. The age range in one series of 125 patients was 3 months to 80 years, with a mean of 51 years (10). There have been no described significant differences in incidence by sex. No differences in histologic pattern have been found in nonfunctioning as compared to functioning tumors. During embryogenesis pancreatic islets are known to form mostly through cellular buds originating from intralobular ductules. Although this process normally ends before birth, it may persist or reappear in many proliferative diseases of the endocrine pancreas, including pancreatic involvement of type I MEN syndrome or solitary endocrine tumors arising in the adult pancreas (2). Since islet cells often have hormone co-expression during early fetal development, it is thought that the origin of pancreatic endocrine tumors is from multipotent cells in ductular epithelium, which can differentiate toward the various cell lines found in these tumors (11,12). Clinically, functional and non-functional tumors present in diverse manners with varied treatment dilemmas. Presentation in functional tumors is usually due to symptoms from the hypersecretion of a particular hormone, while in nonfunctional tumors it is usually due to an effect of the tumor mass. We will separately discuss the treatment challenges of these two tumor types, but attempts will be made to identify where therapeutic algorithms may overlap for these tumors.

introduction
Pancreatic endocrine tumors are benign or malignant epithelial tumors that show evidence of endocrine cell differentiation. Pancreatic endocrine tumors are uncommon, representing <5% of pancreatic tumors in surgical series (1,2). Clinically silent endocrine tumors have been detected in 0.3% to 1.6% of unselected autopsies in which only a few sections of the pancreas were examined, and in up to 10% of autopsies the whole pancreas was systematically investigated both grossly and microscopically. Most tumors from these series are small (less than 1 cm), in elderly patients (mean age of 70 years), and benign (clinically silent microadenomas). Pancreatic endocrine tumors can be broadly classified as functional or non-functional. Despite changing trends, the majority of clinically relevant pancreatic endocrine tumors are functional (3). The proportion of non-functioning tumors, in series of islet cell neoplasms, has varied over time, ranging between 15% and 53% of cases (4–7). While the definition of non-functional has been inconsistent in many reports, increased use of more sophisticated imaging modalities has allowed clinically silent intra-abdominal masses to be identified incidentally and many series report an increased incidence of non-functioning neoplasms (8,9). Overall, the reported 35% to 50% incidence of non-functioning endocrine tumors suggest that non-functional tumors are at least as common as insulinomas and more common than all of the remaining pancreatic endocrine tumor types (2). Functional endocrine tumors of the pancreas are peptidesecreting neoplasms leading to clinical presentation with a defined syndrome related to the effects of an abnormally elevated plasma peptide level. These peptides may or may not occur naturally in the pancreas and a given tumor may secrete multiple peptides. It is on the basis of the primary functional peptide hormone secreted that each tumor is named; e.g., gastrinoma, insulinoma. Non-functioning islet cell tumors are pancreatic neoplasms with endocrine differentiation in the absence of a clinical syndrome of hormone hyperfunction. Despite the presence of hormones in tumor cells at immunohistochemistry, many of these tumors lack evidence of increased serum hormonal levels. Tumors releasing increased amounts of hormone in the blood stream without evidence of a hyperfunctional syndrome are also often reported as non-functioning tumors. Several explanations can be given for why these non-functioning tumors are hormonally silent. One reason is that the principal hormone secreted by the tumor may cause no specific clinical signs, although it is released in excess. In some situations, the tumor makes a functionally inert hormone, which is recognized by an antibody directed against a functional hormone. As the specificity of antibodies improves, such false positives should disappear. A second possibility is that the amount of

functional islet cell tumors
Functional tumors vary with regard to size, location, age distribution, sex distribution, propensity for malignancy, and metastatic potential in accordance with the individual tumor type (Table 45.1) (13). As with most endocrine tumors, the clinical morbidity and mortality associated with the tumor are due to hormone hypersecretion. Most tumors are slow growing and well differentiated.

insulinoma
Insulinoma is a neoplasm that arises from the pancreatic insulin-producing beta-cells. Unlike other gastrointestinal endocrine tumors, which are malignant in more than 60% of cases, 90% of insulinomas are benign, solitary growths that occur almost exclusively within the pancreatic parenchyma (Table 45.1). They occur throughout the head, body, and tail of the pancreas with equal frequency (14). Three percent are in the uncinate and 2% to 3% are ectopic. Ectopic insulinomas are usually found in the duodenal mucosa, the hilum of the spleen, or in the gastrocolic ligament (13).

414

NEUROENDOCRINE PANCREATIC TUMORS
The well-known symptoms associated with hypoglycemia and inappropriate hyperinsulinism occur in a fasting state. The symptoms of headache, blurred vision, incoherence, convulsions, and coma are due to the deleterious effect of hypoglycemia on cerebral function. The symptoms of sweating, weakness, hunger, palpitation, and trembling are homeostatic responses to hypoglycemia, involving secretion of catecholamines (15). The development of these symptoms and the presence of fasting hypoglycemia (glucose <40 mg/dl) and hyperinsulinism (>5 U/ml) during a supervised 72-hour inhospital fast have been the gold standard in establishing the diagnosis (16). In fact, a 72-hour fast is rarely needed, since one-third of patients develop symptoms within only 12 hours, at least 80% within 24 hours, 90% in 48 hours, and 100% in 72 hours (17). Elevated plasma levels of C peptide and proinsulin are confirmatory. Once the biochemical diagnosis of insulinoma has been achieved, diagnostic testing should be performed in an effort at tumor localization. The preoperative localization of insulinomas has received a tremendous amount of attention in recent years. However, as of yet, no single technique has been accepted that is accurate, independent of operator expertise, safe, non-invasive, and inexpensive. A multitude of localization modalities has been devised with a wide disparity in reported success rates and expenses (Table 45.2) (17). Imaging for suspected islet cell tumors is important in the preoperative period for planning therapy. Metastases must be identified preoperatively so that the operative approach can be determined or unnecessary surgery can be abandoned. Noninvasive imaging studies most frequently utilized include CT and MRI and are considered standard radiological modalities for imaging of suspected insulinomas. Because results with dynamic CT for localization of insulinomas have been poor (Table 45.2), spiral CT has replaced dynamic CT for pancreatic imaging in most centers (Fig. 45.1). In one report, nine of 11 tumors could be located using two-phase spiral CT (18). However, the sensitivity of spiral CT in tumor localization remains to be determined in larger numbers of patients (17). Pancreatic endocrine tumors typically have a low signal intensity on T1-weighted MR images. They demonstrate high signal intensity on T2-weighted images. MRI with gadolinium contrast is more sensitive for the detection of vascular tumors than is CT with standard intravenous iodinated contrast agents and may therefore permit detection of insulinomas that cannot be identified on CT (19,20). MRI is likely to become more important for localization of insulinomas, but its role is not yet established. Other modalities that have been utilized in the localization of insulinomas include somatostatin receptor scintigraphy and endoscopic ultrasound. The sensitivity of somatostatin receptor scintigraphy for the detection of islet cell tumors should be independent of tumor size and depends only on tumor expression and cellular and total number of somatostatin receptors. Unfortunately, tumors with low somatostatin receptor density may not be imaged. Only 60% to 70% of insulinomas have been found to express this receptor (21,22) and, therefore, the sensitivity of somatostatin receptor scintigraphy is approximately 50% (23). Endoscopic ultrasound, in experienced hands, may be quite sensitive in localization of insulinomas. In one review, seven of

Table 45.1 Comparison of Functional Endocrine Tumors
Tumor type Gastrinoma Insulinoma Glucagonoma Vi Poma Somatostatinoma
Source: Adapted from Ref. (13).

Pancreatic associated primary (%) 30 95–99 100 100 68

Malignancy rate (%) 60 5–16 82 50 >90

Metastatic rate (%) 50–80 31 >50 50 75

Multicentricity (%) 20–40 10 2–4 20 10

Size Medium Small Large Small Large

MEN-1 (%) 18–41 4–10 Rarely 4 Unknown

Table 45.2 Sensitivity (%) of Localization Modalities for Insulinoma
Center Ann Arbor (25) NIH (103) Sweden (104) France (105) Italy (106) Mayo Clinic (107) (1982–87) Mayo Clinic (1980–95)
Source: Adapted from Ref. (17).

Transabdominal ultrasonography 26 11 40 15 59 64

CT 26 17 43 50 60 36 26

Angiography 44 35 54 44 75 53 47

Portal venous sampling 94 77 63 89

MR imaging 0 25

Intraoperative ultrasonography 92

Palpation 64

100 90 16 95

82 90 90

415

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
10 patients had insulinomas ranging in size from 1.5 to 2.2 cm. Two of three missed tumors were in the head of the pancreas (24). Despite this, endoscopic ultrasound has several disadvantages. It has difficulty visualizing tumors in the tail of the pancreas. In addition, it is invasive, requiring monitored sedation, and is highly operator dependent. Invasive localization tests include arteriography and portal venous sampling. Routine arteriography is no longer recommended in the localization of insulinomas, it has been replaced by other modalities due to its invasive and user-dependent nature. Portal venous sampling has been considered by some to be the single best test for localizing insulinomas (25). The technique involves catheter placement through the liver percutaneously and positioned in the portal system. More than 20 blood samples are taken from different veins that drain the pancreas and are tested for insulin to localize the insulinoma. The false positive rate has been shown to be low and sensitivity ranges from 63% to 94% (Table 45.2). However, this technique is user dependent and does not localize the tumor precisely, rather indicating a region of the pancreas that may harbor the tumor. In addition, complications such as hemobilia and hepatic bleeding may occur as a result of this procedure. An invasive technique utilizing selective arteriographic injection of calcium while measuring hepatic venous insulin levels has yielded excellent tumor localization rates (26). This study is becoming the invasive localizing study of choice, since it may be less user dependent and can be performed more easily than selective portal venous sampling. Due to the limitations of both invasive and non-invasive imaging in the work-up of suspected insulinomas, some centers recommend non-invasive imaging followed by surgical exploration with the use of intraoperative ultrasound. In selected patients, the Mayo Clinic reports a cure rate of 97.7% using this regimen for benign insulinomas (17). The use of intraoperative ultrasonography has enhanced the surgeon’s ability to localize insulinomas. Additional information provided by intraoperative ultrasonography includes defining the relationship of the tumor to the pancreatic and bile ducts and adjacent blood vessels (17). Our current recommendations in patients who meet biochemical criteria for insulinoma are to perform preoperative non-invasive radiologic imaging consisting of spiral CT scanning or quality MR imaging. If this is negative and the patient has no family history or signs of MEN syndrome, the patient should go to operative exploration. At surgery, intraoperative ultrasound should be utilized, if necessary to localize the tumor. In general, insulinomas are usually reddish purple or white and easily identified at operation. To confirm findings or to help localize deep parenchymal lesions, intraoperative ultrasonography can be utilized. In this way, the relationship of the pancreatic duct and other vascular structures to the tumor can be demonstrated. Even if one tumor is found, the entire pancreas should be explored. Enucleation is carefully done to avoid injury to the pancreatic duct. The tumor is enucleated by dissecting immediately adjacent to the tumor, bluntly separating the tumor from normal pancreas using fine instruments and a small sucker. The area is left open and is often drained. The use of a drain is not mandatory if there is minimal disruption of pancreatic parenchyma. Distal pancreatectomy may be necessary in larger, deeper tumors, or if the tumor involves the pancreatic duct. For larger lesions of the pancreatic head, subtotal pancreatectomy or a pancreaticoduodenectomy may be required. When an insulinoma has not been identified at the first operation and reoperation is contemplated, referral to a center with considerable experience is mandatory. Since insulinomas are equally distributed throughout the pancreas, success is proportional to the percentage of pancreas removed. Therefore, a blind distal resection would be effective in only 50% of cases and is not recommended. We and others would recommend that the abdomen be closed and the diagnosis be reconfirmed. Extensive preoperative localization should be performed prior to a re-exploration (17). In one report, with the introduction of intraoperative ultrasound, 15 of 16 reoperated patients have been cured of their disease (17). Although infrequent, malignant insulinomas may be found at exploration. The most common sites for metastasis are the liver and adjacent lymph nodes. In the event that metastases are found, there is a relevant role for debulking the tumor, because a survival advantage has been demonstrated for removing as much tumor as possible (27). In addition, debulking may assist in temporary alleviation of hypoglycemia. Attempts at resection or debulking of hepatic metastases may also be palliative or, rarely, curative (13).

gastrinoma
Gastrinomas are the second most common functioning islet cells tumors of the pancreas, occurring one-half as often as insulinomas (14). These tumors are generally small and occur more frequently (3:2) in males than in females (28,29). Gastrinomas may occur from childhood into old age, but the majority of cases occur between the fourth and the sixth decades of life (13). The Zollinger–Ellison syndrome (ZES), originally described in 1955, includes non-insulin-producing tumors of the

Figure 45.1 Spiral CT of insulinoma in body of pancreas. Arrow indicates tumor which is enhancing with contrast material. The patient underwent an enucleation of this tumor.

416

NEUROENDOCRINE PANCREATIC TUMORS
pancreas, acid hypersecretion and fulminant peptic ulcer disease (30). The more proper designation for this syndrome today is gastrinoma, as one or more of the initially described components of ZES may not be present. Historically, this disease was recognized following a protracted course of ulcer disease with delays in diagnosis ranging from 3 to 9 years (14). At present, patients with gastrinoma resemble the typical peptic ulcer patient. The most common presenting symptom of gastrinoma is epigastric pain and most patients will have a solitary ulcer. These ulcers are often <1 cm in diameter and 75% occur in the first portion of the duodenum. Less commonly, patients with gastrinomas may have recurrent, multiple, and atypically located ulcers, for example, in the distal duodenum (14%) or jejunum (11%) (31). Perforated ulcer remains a common complication with 7% of patients with gastrinomas presenting with perforation of the jejunum (32). Interestingly, as many as 20% of patients have no evidence of ulcer disease and present with the secretory effects of the tumor (33). Diarrhea occurs in 40% of patients with gastrinoma and is caused by gastric acid hypersecretion that increases intestinal transit time, leading to malabsorption. Control of stomach acid output by either total gastrectomy or medications has been shown to control the diarrhea in nearly all patients (34). A significant proportion of patients with gastrinoma will experience esophageal abnormalities including dysphagia and esophagitis. Indeed, ulceration, stricture formation, and perforation have been reported in this disease (35). Medical management of esophageal disease requires strict control of acid secretion by the stomach. Measurement of the fasting serum concentration of gastrin is the best single screening test for gastrinoma, as more than 99% of patients with gastrinomas will have abnormally elevated levels (>100 pg/ml). Ideally, for gastrin levels to be most accurate, all antisecretory medications should be stopped for several days prior to testing. The second critical exam is the determination of basal acid output (BAO). This is defined as a BAO greater than 15 mEq/h in patients without previous surgery to reduce gastric acid secretion, or greater than 5 mEq/h in patients with prior acid-reducing operations (31). The measurement of gastric acid output helps to exclude other causes for hypergastrinemia such as gastric outlet obstruction, antral G-cell hyperplasia, postvagotomy state, and retained antrum. In patients with achlorhydria, such as those with pernicious anemia and atrophic gastritis, failure of acid-induced feedback inhibition results in elevated serum gastrin levels. Therefore, measurement of serum gastrin levels alone in these patients will be inaccurate 50% of the time in the diagnosis of gastrinoma (34). If there is any diagnostic uncertainty or if the serum gastrin level is only moderately elevated, a secretin stimulation test is indicated. This test involves an intravenous bolus of 2 U/kg of secretin and then serum levels of gastrin are determined at 0, 2, 5, 10, and 20 minutes. Patients with gastrinomas have gastrin level elevations of 200 pg/ml or more above the fasting value (14). A positive secretin test is very useful in the differential diagnosis of gastrinoma from antral G-cell hyperplasia. The latter is also characterized by gastrin hypersecretion and hyperacidity. However, gastrin secretion in antral G-cell hyperplasia does not rise after administration of a secretin bolus (13). The secretin test has also been used to follow patients following surgical resection of gastrinoma to evaluate for the presence of recurrent or persistent disease. Patients with persistent or recurrent gastrinoma will have an abnormal secretin test before they develop an elevated basal gastrin level or imageable disease (36). With the advent of potent gastric antisecretory medications, acid hypersecretion can be effectively controlled in all patients with gastrinoma. Therefore, for control of gastric acid output, total gastrectomy is no longer indicated in the management of these patients (31). Proton pump inhibitors are the medical treatment of choice when H2-receptor antagonists have failed because of escape or unwanted side effects. Patients with gastrinoma require greater doses of medication than patients with typical peptic ulcer disease. In one study, the mean total dose of omeprazole to control gastric acid output in 63 patients with gastrinoma was 80 mg per day (37). Despite this, there is concern whether prolonged high dose omeprazole is safe in humans. Experiments in rodents have shown that prolonged omeprazole use is associated with the development of gastric carcinoid tumors (38). In fact, the development of diffuse malignant gastric carcinoids has been observed in a few patients with gastrinoma and MEN-1 maintained on omeprazole for prolonged time periods (37). Precise localization of all gastrinomas is critical for defining an appropriate therapeutic strategy. Gastrinomas are mainly located in the gastrinoma triangle (Fig. 45.2) (39). Primary gastrinomas can also be observed, but less frequently, in the distal duodenum or jejunum and in other parts of the pancreatic gland. Ectopic gastrinomas are rare but hard to localize preoperatively (ovaries, gallbladder). In sporadic gastrinoma, the primary tumor is often single or associated with peripancreatic metastatic lymph nodes (33,40). It is thought that primary gastrinomas can be located in lymph nodes because resection of lymph nodes has rarely been associated with

90%

10%

Figure 45.2 Anatomic triangle in which gastrinomas are most often found. Source: From Ref. (13).

417

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
long-term cure (33,40,41). Data indicate that about 30% of gastrinomas are pancreatic and the others are extrapancreatic, mainly duodenal (33,41,42). In MEN-1, the endocrine tumors are often multiple and can be located both in the pancreatic gland and in the duodenum (37). The ability of imaging modalities to localize primary gastrinoma and gastrinoma metastatic to the liver is summarized in Tables 45.3 and 45.4. At the time of diagnosis, approximately 25% to 40% of patients will have liver metastases, therefore, imaging studies must carefully assess the liver. Although noninvasive modalities for tumor localization have high specificity, their sensitivity is often low. In addition, the sensitivities of ultrasonography and CT scanning are much lower for duodenal gastrinomas than for pancreatic gastrinomas and depend on tumor size (43). Approximately 30% of gastrinomas between 1 and 3 cm are seen on CT, while nearly all larger tumors are imaged. CT detects 80% of pancreatic gastrinomas but only 35% of extrapancreatic tumors. Modern MRI technology has demonstrated improved ability to detect liver metastases with 83% sensitivity and 88% specificity (44). Despite the widespread use of computed tomography and magnetic resonance imaging, 50% of primary tumors will not be identified on conventional preoperative imaging studies (37). Somatostatin receptor scintigraphy (SRS) has been suggested to be the non-invasive imaging study of choice to localize primary and metastatic gastrinomas. Studies have demonstrated that gastrinomas have high densities of somatostatin receptors and that these can be used to image these tumors using radiolabeled somatostatin analogs (Table 45.3) (22,23). In a study of 35 patients, SRS detected 67% of all gastrinomas found at exploration, including 52% of primary gastrinomas found and 80% of lymph nodes containing metastatic gastrinoma. Therefore, SRS missed approximately one-third of all extrahepatic gastrinomas that were found at exploration. Of note, SRS detected only 30% of duodenal gastrinomas but detected 90% of pancreatic gastrinomas. Similar to several other studies, the investigators found that SRS was significantly more sensitive than conventional imaging studies and, on a lesion-by-lesion basis, was even more sensitive than all conventional imaging studies combined. The addition of all conventional studies to SRS detected only three (4%) additional lesions found at exploration in three patients (45). The potential benefit of endoscopic ultrasound for preoperative imaging is that it can visualize small tumors and may help distinguish the primary tumor from lymph node and liver metastases. In a study of 22 patients, EUS had a sensitivity of 50%, 75%, and 63% for duodenal, pancreatic, and lymph node gastrinoma, respectively (46). Currently, studies suggest that EUS has a sensitivity of 50% to 75% and a specificity of 95% in the localization of gastrinomas (24,46). Techniques such as portal venous sampling of gastrin and angiography with intra-arterial injection of secretin with venous sampling of gastrin have been shown to have good sensitivity for gastrinoma detection (Table 45.4) However, since 80% of gastrinomas are located in a relatively small anatomical area, the gastrinoma triangle, performing a study that indicates a region that may harbor a tumor does not add much information. Secretin angiography may be indicated for selection of patients that may benefit from an aggressive approach (e.g., pancreaticoduodenectomy) for a locally advanced or locally recurrent gastrinoma. A secretin angiogram may be indicated to determine whether all tumor is localized within the planned resection (37). In our opinion, preoperative evaluation in patients with gastrinoma should include 1. Endoscopy to evaluate for the presence of duodenal gastrinoma, 2. Spiral CT scan to evaluate the pancreas, lymph nodes, and liver (consider MR imaging if CT is not adequate or further evaluation of liver is needed), and 3. SRS for global evaluation of tumor extent. Despite these tests, a significant percentage of patients will not have their tumors detected before surgery. Therefore, a thorough surgical exploration is indicated. All patients with sporadic gastrinoma should undergo localization studies and be considered for exploratory laparotomy for potential cure. Since gastrinomas are relatively indolent, it has not been shown that resection of the primary tumor extends survival. However, evidence suggests that resection of primary gastrinoma decreases the incidence of liver metastases. Patients with liver metastases from gastrinoma will eventually die of disease. Therefore, it is reasonable to assume that resection of the primary should prolong survival (37). When exploring a patient for gastrinoma, a meticulous intraoperative approach is necessary. Gastrinomas that were previously missed have often subsequently been found to be in the duodenal wall. Gastrinomas are located in more proximal portions of the duodenum and the tumor density decreases more distally (Fig. 45.3). Simple palpation through the bowel wall without opening the duodenum will miss small duodenal tumors. Endoscopic transillumination and extensive duodenotomy have been found to improve localization of duodenal

Table 45.3 Sensitivity of Non-invasive Imaging Studies for Localization of Primary and Metastatic Gastrinoma
Imaging study Ultrasound CT MR SRS
Source: Adapted from Ref. (57).

Extrahepatic primary (%) 9 31 30 58

Liver (%) 48 42 71 92

Table 45.4 Sensitivity of Invasive Imaging Studies for Localization of Gastrinoma
Imaging modality Endoscopic ultrasound Angiogram Portal venous sampling Secretion angiogram
Source: Adapted from Ref. (37).

Primary (%) 50–75 28 73 78

Liver (%) NA 62 NA 41

418

NEUROENDOCRINE PANCREATIC TUMORS
primaries (16). Care to identify the ampulla and the pancreatic duct must be used when attempting to remove medial wall gastrinomas. Gastrinomas within the pancreatic head should be enucleated and those in the body or tail of the pancreas should be resected by either subtotal or distal pancreatectomy. Whether tumors are actually in lymph nodes near the pancreatic head or in the pancreas itself is determined by frozen section. If tumor is found only in lymph nodes, other lymph nodes should be identified and removed. A search must then be made for a duodenal wall primary gastrinoma. Duodenal tumors should be resected with a narrow margin of duodenum around the tumor. Distant metastases to the liver should be resected if all tumor can be completely and safely removed. With increased awareness of duodenal tumors, the operative detection rate is greater than 90% and the immediate cure rate is between 60% and 90% (Table 45.5) (28,48,49). The longterm cure rate is about one-half the immediate cure rate. Patients with resectable disease have excellent survival (5 year: 70%, 10 year: 50%) but those with unresectable multiple metastases have a 5-year survival rate of only 20% to 38% (33,50). Removal of all tumor or surgical debulking prolongs life expectancy in selected patients with metastatic disease (16). Prospective studies indicate that aggressive resection of metastatic disease in patients who have resectable disease by radiologic criteria had a 5-year survival of 79% compared with 28% in patients with inoperable metastatic disease (50,51). Patients who have solitary, localized metastatic disease appear to benefit most from this aggressive approach. The management of gastrinoma in patients with MEN-1 is controversial (52). Neuroendocrine tumors of the pancreas and duodenum are frequently multiple in this syndrome, making it difficult to determine which tumor is responsible for the clinical features. The role of surgery in these patients is not clear as few are cured and the islet cell tumors are thought to be less malignant than those seen in non-familial forms of the disease. A study has shown that, even when procedures to explore the duodenum and remove duodenal tumors were used, complete remission was uncommon because 86% of tumors had metastasized to lymph nodes and 43% of patients had multiple tumors (53). Nevertheless, some authors advocate an aggressive approach to these patients including 1. Performance of a distal pancreatectomy in every patient to remove tumors in the neck, body, or tail, 2. Duodenotomy in every patient to remove small duodenal wall tumors, and 3. Peripancreatic lymph node dissection in any MEN-1 patient with a duodenal neuroendocrine tumor or a pancreatic tumor 2 cm in diameter or larger (54). Using this approach in 38 patients resulted in 5-, 10-, and 15-year survival rates of 98%, 98%, and 96%, respectively. There was no operative mortality and no patient subsequently died of MEN-1 related disease. Two-thirds of the patients with gastrinomas remained eugastrinemic (54). A practical approach may be to perform imaging studies consisting of a CT scan and SRS in patients with MEN-1 and suspected gastrinoma. Those patients with tumors that are large (>2 cm) have a significant probability of metastases to the liver and the tumors are resected according to their malignant potential. Other patients with MEN-1 and gastrinoma may be treated with medication to control gastric secretion.

glucagonoma
Glucagonomas are usually large tumors (>5 cm) and occur most often in the body and tail of the pancreas and are rarely extrapancreatic (Table 45.1) (13,16). The incidence of glucagonoma is considerably less than insulinoma and gastrinoma and is estimated to be one in 20 million to one in 30 million. The exact incidence is unknown because glucagonomas may

Table 45.5 Results of Surgery for Localized Sporadic and MEN-1 Gastrinoma
Sporadic or MEN S S S S MEN MEN S/MENa No. with tumor found (%) 56 (77) 10 (91) 5 (100) 9 (41) 17 (90) 10 (100) 17 (100) No. diseasefree (%) 37 (50) 9 (82) 5 (100) 2 (9) 1 (5) 0 (0) 5 (30)b

Series Norton et al. (33) Howard et al. (48) Thompson et al. (49) McArthur et al. (108) Melvin et al. (109) MacFarlane et al. (53) Jaskowial et al. (110)
Figure 45.3 Location of 24 duodenal gastrinomas in patients with Zollinger– Ellison syndrome. Seventeen tumors were in the first portion, five were in the second, and two were in the third. Source: From Ref. (47).

N 73 11 5 22 19 10 17

Source: Adapted from Ref. (37). a Reoperation for localized recurrent gastrinoma. b Each of the five patients who were disease-free had sporadic gastrinoma. Abbreviations: MEN, Multiple endocrine neoplasia type 1; S, sporadic.

419

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
be underdiagnosed since they remain asymptomatic until they grow large, and may be misdiagnosed since the symptoms of excess glucagon can be attributed to other causes. The mean age at diagnosis is approximately 55 years. As glucagonoma may be associated with MEN-1 syndrome, patients and their families should be screened for other endocrinopathies. Patients usually present with symptoms of weight loss, glucose intolerance, and migratory necrolytic dermatitis. This rash begins as erythematous macules and papules on the face, abdomen, groin, and extremities. Other findings which are less common include stomatitis, glossitis, and diarrhea. Glucagonoma is diagnosed by characteristic findings on biopsy of the rash. However, serum levels of glucagon exceeding 1000 pg/ml are diagnostic (13,16). Although several symptomatic patients with glucagonomas have been found with glucagon concentrations below 1000 pg/ml, most glucagon concentrations of this magnitude have been in asymptomatic patients with small tumors. Glucagon concentrations can be elevated in other disease processes such as hepatic or renal insufficiency and after excessive exercise or severe stress. Elevation can also occur with diabetic ketoacidosis, septicemia, and the use of oral contraceptives. Glucagonomas are less difficult to localize preoperatively than other endocrine tumors of the pancreas because of their size at presentation. The improved technology and widespread use of CT scanning have decreased the importance of angiography, and CT scanning is the technique of choice both to localize the primary tumor and to demonstrate metastatic disease. The management of patients with a glucagonoma is directed toward control of the tumor and its hormonally related symptoms. When the tumor is still localized to the pancreas, surgical resection is the optimal treatment because it can completely reverse all clinical manifestations of the syndrome and result in a lasting cure. Only occasionally are these tumors amenable to enucleation. The high rate of malignancy together with a greater than 50% rate of metastases demands an aggressive resection strategy for glucagonomas (13). As much tumor as possible should be removed, including nodal metastases and lesions in the liver that can be safely wedged out. Formal hepatic resection is indicated if all gross tumor can be excised. Tumor debulking can provide dramatic and rapid improvement in many glucagonoma symptoms. A marked and prolonged decline in serum glucagon is possible with palliative resection, so the hormonal manifestations of this disorder disappear or are relieved for many years. Repeat debulking of recurrent or metastatic disease may also prolong survival (55). Since these tumors are slow growing, the results of resection plus chemotherapy have resulted in 5-year survival rates of 50% (56). Streptozotocin as a single agent has produced a biochemical response (reduction in serum hormone levels by at least 50%) in 64% of patients treated and a tumor response (regression of tumor size by at least 50%) in 50% of patients (57). However, most investigators have found a lower response rate than this. Symptoms of glucagonoma may be successfully controlled with the use of somatostatin analog (58).

vipoma
VIPoma syndrome characterized by watery diarrhea, hypokalemia, and achlorhydria (WDHA) was first described in 1958 by Vemer and Morrison (59). Vasoactive intestinal peptide was first isolated from bovine intestine in 1970 and soon after it was shown that extracts of peptides from tumor and plasma of patients with WDHA produced similar symptoms in dogs (13,60). The first report of a surgical cure was in 1978 when excision of a VIPoma in a patient with WDHA syndrome completely relieved the symptoms as plasma VIP levels dropped to normal (13). Since that time, 201 cases have been reported in the literature (61). VIPomas are predominantly in the pancreas (90%) and extrapancreatic tumors are very rare, except in children (14). Pancreatic VIPomas are usually solitary, small in diameter, and in the body or tail of the pancreas 75% of the time (Table 45.1) (61). Approximately 50% of pancreatic VIPomas are malignant and one-half of these are metastatic to the liver or regional lymph nodes at diagnosis. Less than 5% of patients with VIPoma of the pancreas have MEN-1 (14). The tumors have a bimodal distribution which is related to histologic type. In a study of 62 patients, 52 had pancreatic VIPomas and 10 had extrapancreatic ganglioneuromas. The extrapancreatic sites (adrenal, mediastinal, retroperitoneal) occur predominantly in the pediatric population and demonstrate a less aggressive course, with only a 10% rate of metastasis (61). Diagnosis includes an evaluation of the diarrhea for a secretory nature. This can be confirmed with a trial of fasting for 48 to 72 hours, which will have no major effect on the diarrhea due to VIPomas. The fecal content of potassium and sodium is determined. The sum of twice the sodium plus the potassium should equal isotonicity in a secretory diarrhea. All infectious causes of diarrhea must be excluded. A fasting plasma VIP level of more than 200 pg/ml is required to establish the diagnosis (14,62). Spiral CT scanning should be used to localize VIPomas. With small lesions, other modalities such as SRS or angiography may occasionally be necessary to help identify tumor location (63). Surgical excision remains the only effective method of cure. Preoperative restoration of extracellular volume status and correction of electrolyte abnormalities must be accomplished. The fluid and electrolyte imbalance should be reversed slowly because of its chronic nature. Octreotide acetate is useful in promptly inhibiting VIP secretion from the tumor and stopping the diarrhea (55) A careful surgical exploration including evaluation of the retroperitoneum, both adrenal glands, and the pancreas should be performed. Surgical therapy consists of enucleation or pancreaticoduodenectomy for pancreatic head tumors. Body and tail tumors are best managed by distal pancreatectomy. Complete resection of these tumors delivers full symptomatic relief. If all apparent tumor cannot be resected or if there is metastatic disease, surgical debulking remains a useful option. VIP levels are an excellent tumor marker for recurrence. If there is recurrence, repeat debulking may be a viable and therapeutic option (64). Survival of patients with malignancy and/or metastatic disease is disappointing. The average survival is approximately 1

420

NEUROENDOCRINE PANCREATIC TUMORS
year. There are few well documented chemotherapeutic agents for the treatment of the VIPoma syndrome. The combination of streptozotocin and 5-fluorouracil was effective in 65% of patients studied in the Eastern Cooperative Oncology Group (65). DTIC, human leukocyte interferon, and adriamycin in combination with streptozotocin have been used in small numbers of patients with variable success (65,66). Octreotide provides symptomatic relief for most patients with metastatic disease (67). However, due to the frequent malignant behavior of this tumor, aggressive treatment is warranted and outcome is frequently poor. In one review, survival rates for these patients were 48% at 1 year and 13% at 5 years (68). The experience with chemotherapy in this tumor is limited but streptozotocin and 5-FU has caused tumor regression and symptomatic remission in some patients and is therefore a reasonable consideration in the symptomatic patient with recurrent and/or metastatic disease (69).

somatostatinoma
Somatostatinomas are among the rarest of the functional endocrine tumors. There have been 41 males and 42 females reported with somatostatinomas. These patients have an average age of 54 years with a range of 24 to 84 years (64). The most common location for somatostatinomas was within the pancreas (68%); 19% occur in the duodenum and 3% each in the ampulla of Vater and the small bowel (Table 45.1) (68). Most commonly, pancreatic tumors are found in the head and body of the gland (13). The classic somatostatinoma syndrome has been characterized by the triad of diabetes, diarrhea/steatorrhea and gallstones. Diarrhea/steatorrhea occurs in approximately one-third of patients and results from an increase in stool osmolarity secondary to malabsorption of fats, sugars, and amino acids. Steatorrhea results from decreased pancreatic exocrine secretion and the associated impairment of fat absorption. Excessive somatostatin inhibits the release of cholecystokinin, which decreases gallbladder contraction and leads to malabsorption and cholelithiasis. Weight loss is one of the most common findings and is seen in about 40% of patients and may be attributable to malabsorption. Diabetes occurs in 25% of patients and may occur secondary to greater suppression of insulin secretion over glucagon release by somatostatin as well as indirectly by somatostatin suppression of gastric inhibitory peptide (13,14,68). Somatostatinomas are often large tumors at the time of presentation and imaging studies are associated with a high degree of accuracy. CT scanning correctly localized the somatostatinoma in 34 of 37 patients with pancreatic tumors. Angiography has also been used to localize difficult neoplasms (68). The diagnosis can be established by measuring an elevated fasting somatostatin level (normal, <100 pg/ml). When evaluated preoperatively, 34 of 35 patients had elevated levels of circulating somatostatin. Plasma somatostatin measurements may also be useful in evaluating the success of treatment and to follow patients for possible recurrence. Surgical resection is the preferred treatment for patients with somatostatinomas. Unfortunately, successful surgical management is difficult because of the high rate of metastases present at exploration (Table 45.1). Most tumors are large and enucleation is usually not possible. Pancreatic resection including a pancreaticoduodenectomy or distal pancreatectomy is often necessary. Debulking a large tumor or hepatic metastases may effectively palliate symptoms, often for prolonged periods of time. As with other islet cell cancer, somatostatinomas may have an indolent tumor biology and long-term survival can occur.

unusual functional islet cell tumors
Other functional islet cell tumors are exceedingly rare and include growth hormone-releasing factor secreting tumors, adrenocorticotropic hormone secreting tumors, parathyroid hormone like secreting tumors, and neurotensinomas. The treatment of choice for these rare tumors is uncertain but appears to be primarily surgical resection. Palliative resection (debulking) of metastatic disease is also frequently indicated since it may provide symptom benefit. Octreotide acetate may prove to be useful in the symptomatic treatment of many of these tumors (13).

non-functional islet cell tumors
Non-functioning tumors are slow growing and occur most commonly in the head of the pancreas (4,70). In surgical series, 38% to 80% of lesions are found in the pancreatic head. They tend to be relatively large, symptomatic, and often present with metastatic disease. In a similar fashion to adenocarcinoma of the pancreas, the clinical presentation of these tumors is related to either local invasion or metastatic spread. Jaundice, abdominal pain, weight loss, or the appearance of an abdominal mass are the predominant signs and symptoms (Table 45.6). Many authors report the presence, in a small number of patients, of multiple non-functioning tumors scattered throughout the pancreas (Table 45.7). Lesions in the pancreatic head may induce back pain but much less commonly than with ductal adenocarcinoma, which displays a characteristic ability to infiltrate retroperitoneal nerves. Often non-functional islet cell neoplasms in the body and tail of the pancreas do not present with many symptoms but may have a palpable mass on examination. In addition to these findings, patients may complain of nausea, vomiting, diarrhea, or lethargy (71). In contrast to adenocarcinoma of the pancreas where systemic effects are seen quite early in the presentation of the disease, patients with non-functioning islet cell tumors can present with advanced metastatic disease and relatively few systemic symptoms. An incidental presentation is not uncommon. Kent and colleagues from the Mayo Clinic noted that four (16%) of 25 non-functional tumors presented as an incidental finding (4). Rarely, these tumors may present with massive hemorrhage as a result of either penetration into the gastrointestinal tract or erosion of vessels in the retroperitoneum (72). It is not clear whether the interval from onset of symptoms to diagnosis differs in patients with non-functioning neoplasms and those with functional neoplasms. It has been suggested that as many small functional islet cell neoplasms cause symptoms related to hormone excess without the development of biliary

421

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 45.6 Signs and Symptoms of Non-functional Tumors at Presentation
Author Cheslyn-Curtis et al. (80) Broughan et al. (73) Kent et al. (4) Phan et al. (10) N 20 21 25 58 Study (yr) 1982–91 1948–84 1960–78 1949–96 Jaundice (%) 40 24 28 35 Abdominal pain (%) 35 48 36 56 Weight loss (%) 30 24 – 46 Mass (%) 40 24 8 –

Table 45.7 Location of Non-functional Islet Cell Tumors in the Pancreas
Author Kent et al. (4) Dial et al. (111) Broughan et al. (73) Eckhauser et al. (112) Yeo et al. (113) Cheslyn-Curtis et al. (80) Evans et al. (84) Madura et al. (81) Lo et al. (7) N 25 11 19 10 13 20 73 14 34 Study years 1960–78 1963–83 1948–84 1973–85 1985–90 1982–91 1953–92 1972–96 1985–96 Head No. (%) 14 (56) 7 (64) 9 (47) 8 (80) 5 (38) 14 (70) 43 (59) 11 (79) 16 (47) Body and tail No. (%) 5 (20) 4 (36) 7 (37) 2 (20) 8 (62) 5 (25) 30 (41) 1 (14) 16 (47) Multiple No. (%) 6 (24) – 3 (16) – – 1 (5) – 1 (7) 2 (6)

or gastrointestinal obstruction due to the primary tumor, that they present earlier (14). The duration of symptoms related to the tumor may be at least as long or longer in patients with functioning tumors as in non-functioning tumors. In a review of 64 patients with islet cell carcinomas (34 non-functional, 30 functional) treated at the Mayo Clinic, the duration of symptoms prior to diagnosis for all patients ranged from less than 1 to 120 months (median 10 months). For functional neoplasms the median duration of symptoms was 11 months versus 8 months for non-functional neoplasms (p > 0.1) (7). Others have also found that the median duration of symptoms was significantly longer for functional tumors (insulinomas: 22 months; gastrinomas: 36 months) as compared to nonfunctional neoplasms (6 months) (73). It appears that most patients with islet cell tumors undergoing surgical treatment have a lengthy duration of symptoms. A high index of suspicion is necessary in attempting to make the diagnosis of an islet cell tumor of the pancreas.

diagnosis
Since the clinical presentation of non-functioning islet cell tumors is similar to other pancreatic neoplasms, the major challenge is to distinguish this tumor from other forms of pancreatic neoplasia, especially from the much more common ductal adenocarcinoma. Non-functioning islet cell tumors tend to be larger than ordinary pancreatic adenocarcinomas at the time of diagnosis. Viable portions of islet cell tumors are typically well vascularized and often hypervascular relative to the unaffected portion of the pancreas. As a result this neoplastic tissue usually undergoes an appreciable degree of enhancement on images made with appropriate techniques of intravenous contrast enhancement (74). Other morphologic features that distinguish islet cell tumors include the lack of

Figure 45.4 Computed tomography of patient with large non-functioning tumor of tail of pancreas. Note area of cystic degeneration and hypodense areas of necrosis (arrows). This patient underwent a potentially curative resection of the tumor.

vascular encasement and lack of obstruction of the pancreatic duct as compared to adenocarcinoma of the pancreas. For non-functioning islet cell neoplasms, CT is the radiologic imaging modality of choice. Like other large, hypervascular neoplasms, islet cell carcinomas may have internal necrotic areas. These appear on CT as hypodense zones of low attenuation, sometimes with features of cystic degeneration or necrosis (Fig 45.4). Dystrophic calcification may develop within a tumor and is observed on CT in 20% of non-functioning islet cell tumors (75). The characteristic features on CT of nonfunctioning islet cell carcinomas, including large tumor mass, hyperenhancement, cystic degeneration, and calcification,

422

NEUROENDOCRINE PANCREATIC TUMORS
Table 45.8 CT Characteristics of Pancreatic Neoplasms
Tumor type Adenocarcinoma Acinar cell Solid and papillary Islet cell Serous cystadenoma Sex (M:F) 1.3:1.0 1.0:1.0 1.0:9.0 1.0:1.0 1.0:2.0 Vascularity Hypovascular Hypovascular Hypovascular Hypervascular Hypovascular Degree of necrosis Little Much Much Variable None Size Small Large Large Mod-large Large Enhancing capsule No Variable Yes No Yes Calcification No No Yes—peripheral Yes (25%)—internal Yes—central

help distinguish them from ductal adenocarcinoma. Similar features are present and identified by CT in the metastatic lesions from these tumors. Metastases are found most often in the liver and in regional lymph nodes. Whereas metastases from ductal adenocarcinoma tend to be small, those from islet cell carcinoma are often large (74). Pancreatic tumors other than ductal adenocarcinomas can be difficult to distinguish from non-functioning islet cell tumors. In a study of 45 patients with 50 non-functioning islet cell tumors, 36 of the tumors had heterogeneous areas and over half (N = 27) had cystic degeneration visualized on CT or MRI. Only 14 of the 50 non-functioning tumors were solid homogeneous masses (76). Therefore, an islet cell carcinoma could resemble atypical forms of serous or mucinous cystic neoplasms. More likely, a partially necrotic tumor, such as solid and papillary epithelial neoplasm, might resemble an islet cell neoplasm with similar structure. Unlike solid and papillary epithelial neoplasms, however, non-functioning islet cell tumors are often associated with large metastases and they do not typically have visible capsules. Moreover, they tend to occur in older age groups and they do not have the predilection for female subjects (Table 45.8). While MEN-1 syndrome has been associated with functioning endocrine tumors of the pancreas, small clinically silent endocrine tumors are often numerous in the pancreas of patients with this syndrome. Non-functioning microadenomatoses of MEN-1 patients have been found mainly to be composed of glucagon and pancreatic polypeptide producing cells. In addition to the microadenomatosis, larger, discrete adenomas, mainly composed of pancreatic polypeptide producing cells, have been found (77). Although increased blood levels of glucagon and pancreatic polypeptide have been frequently detected in such patients, as a rule related clinical syndromes have not been observed. Most clinically non-functioning pancreatic tumors in patients with MEN-1 syndrome have benign histologic patterns and behavior (2). Metastatic tumors to the pancreas such as those from renal cell carcinoma are particularly likely to have appearances indistinguishable from those of islet cell carcinoma. In these cases, clinical correlation should predict the nature of the lesion. In addition to characteristic clinical presentations and radiologic findings, serum markers have been utilized in the diagnosis of non-functional endocrine tumors. To be able to detect a tumor by its secretion(s) implies that the tumor is no longer biochemically silent. Nonetheless, such neoplasms are not associated with any distinct clinical symptoms.

Pancreatic polypeptideoma (PPoma) is one such tumor and among the most common of the non-functional islet cell tumors. Human pancreatic polypeptide is frequently associated with other hormones in pancreatic endocrine tumors and pancreatic polypeptide cells are most often found in glucagonomas and in non-secreting tumors (11). In a series of eight patients with isolated pancreatic polypeptide producing islet cell tumors, clinical features included abdominal pain (N = 4), weight loss (N = 4), diarrhea (N = 2), gastrointestinal bleeding (N = 2), and jaundice in one patient. Serum basal pancreatic polypeptide was elevated in most patients with a marked response to secretin. Six of eight patients underwent tumor resection with two patients being not surgical candidates due to hepatic metastases (77). After curative resection, elevated serum pancreatic polypeptide levels fell to normal. Contrary to most reported nonfunctioning tumors, PPomas seem to have a benign course even when of large size and producing local symptoms, as found in six of eight cases reported above and in 10 of 10 cases reported elsewhere (11). Other markers which have been evaluated in non-functioning islet cell tumors include neuron-specific enolase (78). Its role in monitoring patient course or response to therapy is not known at present. The presence of hormones in tumor cells at immunochemical staining provides useful information for the diagnosis of non-functioning islet cell tumors. Multiple hormones can be produced by these neoplasms. In a series of 61 non-functioning tumors, pancreatic polypeptide immunoreactivity was detected in 35% of cases, with a mean of 33% of all tumor cells; glucagon was found in 30% of cases and 30% of cells; somatostatin in 15% of cases and 20% of cells; serotonin in 20% of cases and 36% of cells; calcitonin in 20% of cases and 10% of cells; neurotensin in 8% of cases and 8% of cells; and insulin in 15% of cases and 2% of cells. No gastrin or VIP immunoreactivity was detected (79). It seems clear that tumors producing pancreatic polypeptide, glucagon, somatostatin, calcitonin, and serotonin are more likely to be without an associated syndrome than are those producing clinically powerful hormones such as insulin, gastrin, or VIP.

localization and extent of disease
There are sparse data on the accuracy of non-invasive and invasive radiologic imaging in tumor localization and predicting resectability of non-functioning islet cell tumors. With improvements in CT scanning, angiography is no longer necessary in their diagnostic work-up. In a series of 20 patients studied between 1982 and 1991, dynamic CT localized the tumor in 17 of 20 patient (85%). No pancreatic lesion was seen

423

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
in three patients who had obstructive jaundice. In two, the lesions were seen with angiography and the third was found at operation (80). In another series, dynamic CT demonstrated a mass in nine of 13 patients (69%) while three of the patients had no mass seen but dilated common and/or pancreatic ducts were present (81). In a study which evaluated both functioning and non-functioning islet cell tumors from 1949 to 1996, dynamic CT successfully localized the tumor 76% of the time. Angiography localized the tumor in 58% of patients and CT plus angiography did no better at tumor localization (79%) as compared to CT alone (76%) (10). Usually when CT shows a pancreatic mass presumed to be a non-functional islet cell neoplasm, arteriography is not warranted unless there is a question of invasion of a major vascular structure. Even then, non-invasive MR angiography has been shown to be accurate in predicting tumor vascular involvement (82). Due to the relatively good prognosis with islet cell neoplasms, many surgeons would attempt en bloc resection with vascular reconstruction if there was vascular involvement at the time of surgery (83). At present, the yield of new generation helical CT scanning in localizing non-functioning islet cell tumors is not known. Although a study suggests that MR imaging is superior to CT in identification of pancreatic endocrine tumors, this study can be faulted since dynamic CT was utilized for comparison (19). To our knowledge, no studies that compared state-ofthe-art helical CT with MR imaging have been reported in the literature. The role of laparoscopy in the management of patients with non-functioning islet cell tumors is unknown. With the goal of ruling out metastatic disease, laparoscopy would spare patients from undergoing an unnecessary laparotomy. In multiple series, evidence of metastatic disease has been found at the time of surgery in more than 50% of patients and resectability rates are low (8,73,84). Unfortunately, most of these series have spanned long time frames and it is unknown whether metastatic disease was visible on quality preoperative imaging studies. A number of studies have demonstrated improved accuracy of laparoscopy in predicting extent of disease for pancreatic adenocarcinoma (85). No study has specifically evaluated the use of laparoscopy in patients with islet cell tumors of the pancreas. Extrapolating from data obtained with pancreatic adenocarcinoma, it would appear that laparoscopy could spare asymptomatic patients with metastatic disease from undergoing unnecessary laparotomy (Fig. 45.5). High-affinity somatostatin receptors have been identified in the pancreatic islet cells (86). With this in mind, somatostatin receptor scintigraphy (SRS) has been developed for in vivo imaging of somatostatin receptor-positive tumors (Fig. 45.6A–C). Studies with small numbers of patients have shown accurate localization of non-functioning islet cell tumors with SRS (22,87). In a collective review of results from 15 centers in Europe with SRS, 82% of non-functioning tumors (N = 60) were visualized at scintigraphy (23). Investigations in patients with non-functioning pancreatic tumors, which have compared SRS with CT scanning and MRI, have shown a similar sensitivity (60%) for both SRS and conventional imaging in detecting tumors. Somatostatin receptor

Figure 45.5 Photograph of liver metastases of a non-functioning islet cell tumor (arrows) detected at laparoscopy. The primary tumor was in the head of the pancreas. The preoperative CT scan did not reveal metastatic disease.

scintigraphy was superior in detecting intra-abdominal and bony metastases. Conversely, tumor masses shown by conventional scanning techniques were missed by SRS in several patients. Since large tumor masses were missed by SRS in some cases, it is felt that the low density of somatostatin receptors on some tumors may be the major factor causing false negative results (88). Since somatostatin receptors appear to be present in similar concentrations in non-functioning and functioning tumors, SRS should be of equal accuracy in detecting both types of tumors. The advantage of radionuclide scintigraphy is that adenocarcinoma of the pancreas does not take up the tracer and therefore false positive rates are low (23). The disadvantage of this methodology is that the absence of tracer uptake by a tumor does not rule out an islet cell tumor. With rapid improvements in conventional imaging modalities (e.g., CT and MRI) in the visualization of the pancreas, liver and other intra-abdominal organs, the benefit of SRS in islet cell tumor identification may decrease. The capacity to take up and decarboxylate amine precursors such as 5-hydroxytryptophan (5-HTP) and l-dihydroxyphenylalanine (l-DOPA) and store their amine precursors is characteristic of neuroendocrine cells and tumors. Utilizing this concept, positron emission tomography (PET) has been evaluated in the diagnosis of pancreatic endocrine tumors. In a study utilizing l-DOPA, the ability to detect the pancreatic tumor and possible metastases was evaluated in 22 patients with islet cell tumors. In the six patients with non-functioning islet cell tumors, there was relatively lower uptake of l-DOPA and in two cases the primary tumor as well as the metastases were not identified with PET, despite the fact that they were large and clearly visible with CT (89). At present, with few available data, it appears that for localization of non-functioning endocrine tumors and their metastases, there is no general advantage of PET compared with CT. A larger patient population is needed to determine the usefulness of l-DOPA and 5-HTP as tracers for visualization of the various subgroups of pancreatic endocrine tumors.

424

NEUROENDOCRINE PANCREATIC TUMORS

(A)

(B)

(C) Figure 45.6 (A) CT scan of patient with cirrhosis and large non-functioning tumor of tail of pancreas (arrow). (B) Another image of the liver of the same patient shows a cystic lesion in the liver (arrow), which was suspicious for metastatic disease. (C) Octreotide scan of the same patient. Top panel demonstrates two intense areas of uptake in the left abdomen corresponding with the tumor and the spleen (arrowheads). In the bottom panel, the large cyst in the liver is visualized and there is no uptake within this cyst in the liver (arrowhead). At operation, these findings were confirmed, no liver metastases were identified and the patient underwent a distal pancreatectomy with splenectomy.

treatment: curative surgery for the primary tumor
Operative resection is the only potentially curative form of therapy for patients with pancreatic islet cell tumors. Operation can be performed to resect local disease and to prevent the development of distant metastases or local recurrence. Patients with non-functional islet cell tumors may have similar clinical symptoms to ductal adenocarcinoma of the pancreas. However, in contrast to ductal adenocarcinoma, survival rates following surgical resection are quite good, and aggressive surgical therapy is indicated for non-functional neoplasms. Even though these tumors tend to be slow growing, they often present when the tumor mass is quite large. In some studies mean tumor size is 8 to 10 cm at presentation (73,80). Despite this, with surgical resection, actuarial 5-year survival rates range from 44% to 63% with a median survival of 30 months to 4.8 years (Table 45.9). Surgical therapy should not be denied to patients on the basis of tumor size alone. While ductal adenocarcinoma tends to directly encompass and invade adjacent structures early in its disease course, nonfunctioning islet cell tumors more often displace structures

without actual invasion. The large tumor bulk may cause near obstruction of blood vessels via a compressive effect which can be relieved by tumor resection. For instance, tumors in the tail of the pancreas can compress the splenic vein leading to splenomegaly and even varices (Fig. 45.7A,B). Nevertheless, these tumors can be frequently resected with negative margins. Similarly, tumors in the head of the pancreas can compress the superior mesenteric vein. Again, in the absence of known metastatic disease, every attempt should be made at tumor resection. Tumors in the head of the pancreas should usually be treated by pancreaticoduodenectomy, while tumors in the body and tail should be treated with distal pancreatectomy. Extension of tumor from the pancreas into surrounding organs such as the stomach and colon should be resected en bloc. If the tumor is small (<2 cm) and located in the tail of the pancreas consideration should be given to distal pancreatectomy with splenic preservation. In patients with significant co-morbid conditions or for small lesions in the head of the pancreas, enucleation can be entertained. With endocrine tumors, some authors have advocated that surgical removal of involved lymph nodes should be performed

425

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 45.9 Rates of Curative Resection and Survival in Non-functional Islet Cell Tumors
Author Kent et al. (4) Broughan et al. (73) Thompson et al. (8) Venkatesh et al. (102) Cheslyn-Curtis et al. (80) Evans et al. (84) Legaspi and Brennan (71) Lo et al. (7) Phan et al. (10)
a b

N 25 21 27 43 20 73 33 34 58

Study years 1960–78 1948–84 1965–84 1950–87 1982–91 1953–92 1983–88 1985–96 1949–96

Tumor sizea (cm) >5d 10 – – 8 – – 6.2 5.1

Curative resection (%) 9 (36) 12 (57) 10 (37) 22 (51) 10 (50) 19 (26) 12 (36) 12 (35) 46 (79) 44% 63% 58% 40 30 4.8 76% – 52%

Survival allb – – – – 42 6.8 100% – –

Curativec (actuarial 5-yr) (actuarial 5-yr) (actuarial 3-yr) months (mean) months (median) yr (median) (actuarial 3-yr) (actuarial 5-yr)

Mean or median tumor size. Survival in entire study group: curative and non-curative resections. c Survival in curatively resected patients. d 18 of 25 tumors >5 cm in size.

(A)

(B)

Figure 45.7 (A) Large tumor in the tail of the pancreas with splenic vein obstruction and varices (arrows). This tumor was resected with negative margins. (B) Intraoperative picture of varices encountered at the time of resection (arrows).

as it may improve survival. Patients with malignant endocrine tumors whose disease is confined to regional lymph nodes have a greater chance of benefit by removal of tumor than patients who have distant metastases. Therefore, regional nodes which are involved with tumor should be resected as completely as possible in an attempt to eliminate all disease and decrease the probability of distant metastases (90). With non-functioning islet cell tumors, in the absence of metastases, it is controversial whether nodal disease outside the field of dissection should be resected. However, utilizing the data from neuroendocrine tumors of the small bowel in which extensive lymphadenectomy is recommended, it can be concluded that for non-functional islet cell neoplasms reasonable attempts should be made to remove all nodal disease.

treatment: curative surgery for metastatic disease
There are few data available on the role for attempted curative resections of metastatic non-functional islet cell tumors. Since these tumors may follow a relatively indolent clinical course, a case to resect anatomically localized and surgically resectable metastases within the liver can be made. Of course, for this to

be in the best interest of the patient, the primary disease and metastases must be completely resected. No data exist to validate this approach. It should only be considered if the metastatic lesion can be resected with acceptable morbidity and mortality. Patients with unresected hepatic metastases from gastrointestinal neuroendocrine tumors have been reported to have 5-year survival rates of between 13% and 43% (64). In an effort to improve survival and even cure patients who have failed other forms of therapy, liver transplantation has been applied in these patients. Reports of liver transplantation for metastatic neuroendocrine tumors have been confined to small numbers of patients and short follow-up, typically less than 3 years from the time of transplantation (91). In the best results published to date, 12 patients (five pancreatic islet cell tumors) received transplants at a single center. Long-term survival was achieved in most patients with a median survival of 55 months (92). However, in a multicenter report from France of 16 cases of liver transplantation for metastatic pancreatic islet cell tumors, the 4-year survival was 8%. The authors concluded that liver transplantation was not indicated for metastatic pancreatic islet cell tumors (93). Despite some

426

NEUROENDOCRINE PANCREATIC TUMORS
Table 45.10 Surgical Palliation and Survival in Non-functional Islet Cell Tumors
Author Kent et al. (4) Prinz et al. (96) Dial et al. (111) Eckhauser et al. (112) Evans et al. (84) Cheslyn-Curtis et al. (80) N 25 8 11 11 73 20 Palliative procedure no. (%) 3 (12) 4 (50) 2 (18) 2 (18) 12 (16) 7 (35) Type Biliary bypass (N = 2) Unknown (N = 1) Double bypass (N = 3)a Biliary bypass (N = 1) Biliary bypass (N = 2) Double bypass (N = 2) Distal pancreatectomy (N = 9) Pancreaticoduodenectomy (N = 3) Distal pancreatectomy (N = 2) Biliary bypass (N = 3) Pancreaticoduodenectomy (N = 2) Survival 7 yr (mean) 4.5 yr (mean) 1.5 yr (AWD) 8 yr (AWD) 13 mo (AWD) 26 mo (DOD) 4.5 yr (median) 16 mo (median)

a Biliary and gastric bypass. Abbreviations: AWD, alive with disease; DOD, dead of disease.

encouraging results from individual groups, until larger numbers of patients and longer follow-up are accumulated, or until the supply of donor livers increases, liver transplantation for metastatic pancreatic islet cell carcinomas should be applied with great caution.

treatment: palliative surgery
Palliative resections of the primary tumor in the presence of metastatic disease have been performed in non-functioning islet cell tumors. In the series from M.D. Anderson, nine distal pancreatectomies and three pancreaticoduodenectomies were performed in the presence of liver metastases (84). One perioperative death occurred in a patient who underwent distal pancreatectomy. Five of the 12 patients died of disease at a median of 3.7 years. Uncontrolled local tumor recurrence contributed to the cause of death in only one of the five patients. Overall, the median survival was 4.5 years (Table 45.10). These results can be compared to 22 patients who had metastatic disease at diagnosis and did not undergo resection of the primary tumor. Sixteen of the 22 died of disease at a median of 3.2 years. The cause of death in one patient was gastrointestinal hemorrhage caused by progression of the primary tumor in the pancreatic head. The other 15 patients died of liver metastases. The authors determined that the trend toward improved median survival in patients undergoing resection of their primary tumor in the presence of metastases could probably be explained by the smaller tumor volume in this group at the time of surgery. In a second study, resection for palliation of local tumor symptoms was performed in two patients with non-functioning tumors. Complete relief of symptoms was achieved in both patients. One patient died 76 months after palliative resection and the other patient was still alive with disease at 10 months (94). In the asymptomatic patient who has metastases from non-functioning islet cell tumor, due to prolonged survival regardless of primary tumor resection, it is difficult to justify primary tumor resection. In our opinion, in the absence of compelling data, resection of the primary tumor in the

presence of liver metastases should be considered only in the presence of debilitating symptoms (e.g., pain) or in the presence of complications (e.g., bleeding). Resection of metastases for palliation of symptoms has been reported more often in functioning than in non-functional islet cell carcinomas. It may play a more important role in alleviating the effects of excess production of endocrine hormones in functioning tumors. For non-functioning tumors it is rarely indicated, especially when other less invasive modalities can be utilized for treatment of symptoms due to metastatic tumor growth. Hepatic artery embolization has been performed to palliate symptoms from metastatic disease. In a study of 22 patients with metastatic functional or non-functioning neoplasms, sequential hepatic artery embolization was performed. Patients underwent a median of four embolizations. Partial remission was achieved in 12 patients. Hormonal syndromes were frequently relieved. Moderately toxic reactions were incurred after each embolization, but they were brief. Median survival was 33.7 months (95). Sequential hepatic artery occlusion with microembolic material may provide prolonged palliation for selected symptomatic patients with islet cell carcinoma metastatic to the liver. Palliative biliary and/or enteric bypasses for patients presenting with unresectable disease and biliary or gastric outlet obstruction have been performed (Table 45.10). Prinz et al. (96) reported four of eight patients with unresectable tumors who underwent either a biliary bypass and gastrojejunostomy or a biliary bypass alone. The mean survival in these four patients was 4.5 years. For patients with obstructive jaundice secondary to a locally unresectable non-functioning islet cell neoplasm, operative biliary bypass should be strongly considered. The superiority of operative biliary bypass for long-term management of malignant biliary obstruction justifies this approach in patients who have islet cell tumors and potential for prolonged survival. Duodenal bypass via a gastrojejunostomy for lesions in the head of the pancreas should also be considered in these patients.

427

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

treatment: chemotherapy
Recommended chemotherapy for advanced islet cell tumors is based on the results of a multicenter randomized trial. In this trial, 105 patients were randomized to receive one of three regimens: streptozotocin plus fluorouracil, streptozotocin plus doxorubicin, or chlorozotocin alone. Patients with either nonfunctioning or functioning islet cell tumors were included in this study. Streptozotocin plus doxorubicin was superior to streptozotocin plus fluorouracil in terms of the rate of tumor regression, measured objectively (69% vs. 45%, p = 0.05, respectively), and the median length of time to tumor progression (20 vs. 6.9 months, p = 0.001, respectively). Streptozotocin plus doxorubicin also had a significant advantage in terms of survival (median 2.2 vs. 1.4 years, p = 0.004) (97). This regimen is now considered the standard treatment for advanced islet cell tumors, but should be considered in the context of each individual patient, as some may have prolonged survival untreated, and response to treatment is greater than the effect on overall survival.

treatment: radiation therapy
The reported experience with radiation therapy in patients with islet cell carcinoma has been scarce. Torrisi et al. (98) treated three patients with locally advanced islet cell carcinoma. Objective responses were seen in two patients, whereas the third patient had a prolonged stabilization of the tumor. In addition, case reports of locally advanced islet cell tumors that have had complete responses to radiation therapy exist (99). It would appear that selected patients may be palliated by radiotherapy. Radiation has, however, been reserved in the main for the treatment of pain from localized bone metastases.

treatment: octreotide
Octreotide is a synthetic octapeptide with structure and activities similar to those of the native hormone somatostatin, but with significantly longer half-life and duration of action that the native substance. Octreotide has been reported to be effective in controlling the hormone-induced symptoms of patients with functional islet cell tumors (100). Although rare reports of tumor regression to octreotide have been published, the drug has been primarily utilized to ameliorate the symptoms from hormonal excess in patients with functional islet cell tumors. In a study which included 13 patients with advanced, incurable non-functional islet cell tumors, octreotide was administered via subcutaneous injection. No patient experienced a major objective response. Of the 21 patients with functional tumors, 15 demonstrated either a symptomatic improvement or an objective decrease in hormone level. Octreotide is useful in controlling symptoms due to hormonal excess in functional islet cell tumors. There is no evidence that octreotide would benefit patients with non-functional islet cell tumors.

distant metastases and who did not undergo resection. These patients had a 38% 5-year survival with a median survival of 3.3 years. In addition, among patients with localized disease at presentation, the survival was significantly better in patients who were able to undergo tumor resection as compared to those who did not. Therefore, important predictors of survival are the ability to undergo curative resection and the degree of local spread. It is unclear whether non-functional tumors have a worse prognosis than functional islet cell tumors. Several studies have shown that the survival of patients with non-functional tumors is poorer than for those with functional tumors. In a report from the Cleveland Clinic, actuarial 10-year survival was 55% for non-functioning tumors while it was 92% for insulinomas and 68% for gastrinomas (73). In a study from Johns Hopkins, median survival was 121 months in functional tumors while it was 96 months in non-functional neoplasms (p < 0.025) (10). In an early report from the Mayo Clinic, survival was statistically better at 3 years in those patients with gastrinomas compared with patients with non-functioning tumors, 91% versus 58% (8). However, in a later report from the same institution, there was no survival difference between functional and non-functional neoplasms (7). Other studies have found no difference in survival between functioning and non-functioning tumors (101,102). Size of the primary tumor rather than functional status may be a more important prognostic variable. In the study by Phan et al. (10) the median tumor size in the non-functional tumors was 4.0 cm as compared to 1.9 cm in the functional neoplasms. In addition, the malignancy rates were correspondingly lower in the functional tumors (47%) as compared to the non-functional ones (60%). It may be that non-functional tumors are detected at larger tumor burdens because of the absence of an endocrine syndrome. This may account for poorer survival with non-functioning neoplasms seen in some studies. Efforts have been made to develop more sophisticated markers to determine prognosis in patients with non-functioning islet cell neoplasms. In a preliminary study of 61 non-functioning tumors, vascular or perineural microinvasion and Ki67 proliferative index were the most sensitive and specific variables that were predictive of malignancy. These variables were utilized in tumors lacking evidence of malignancy at the time of surgery, to separate cases with increased risk of malignancy from cases with limited risk, and were found to be somewhat predictive of survival (79). Rigorous studies in larger numbers of patients with islet cell neoplasms, which examine different immunohistochemical markers, may help identify prognostic variables in these tumors.

conclusions
Clinically, there are similarities and differences between the various functional endocrine tumors of the pancreas. These tumors vary with regard to size, location, age distribution, sex distribution, propensity for malignancy, and metastatic potential in accordance with the individual tumor type. The clinical morbidity and mortality associated with the tumor are due to peptide hypersecretion and not to tumor mass. In addition,

prognosis
Evans et al. (84) found that patients who underwent curative resection of their primary tumor in the absence of metastases had a 72% 5-year survival with a median survival of 6.8 years. This was in sharp contrast to patients who presented with

428

NEUROENDOCRINE PANCREATIC TUMORS
there is no correlation between tumor size and the severity of functional manifestations. Non-functioning islet cell tumors are pancreatic tumors with endocrine differentiation in the absence of a clinical syndrome of hormone hyperfunction. These tumors are hormonally silent because either the principal hormone secreted by the tumor may cause no specific clinical signs; although it is released in excess, the amount of hormone produced may be too small to cause symptoms, or the tumor may secrete a precursor hormone that is functionally inert, or the hormonal product of the tumor may not yet be identified. Non-functioning islet cell tumors are slow growing and occur most commonly in the head of the pancreas. Imaging modalities such as CT scanning or MR imaging are accurate in the localization of these tumors within the pancreas. While false negative rates for tumor detection are high, octreotide scintigraphy may provide additional useful information in the evaluation for suspected metastases. Operative resection is the only potentially curative form of therapy for patients with functioning and non-functioning islet cell neoplasms. In contrast to ductal adenocarcinoma of the pancreas, survival rates following surgical resection are quite good. Since metastatic functional and non-functional islet cell carcinomas may follow a relatively indolent course, anatomically localized and surgical resectable metastases within the liver should be considered for resection. The primary site of disease must be controlled if this is to be considered. The survival of patients with non-functional tumors may be similar to those with functional neoplasms. Patients with nonfunctional tumors often present with larger tumor size, which may account for the decreased survival seen with these neoplasms in some series. Prognostic variables in islet cell tumors are limited and further work evaluating immunohistochemical markers in this disease is needed.

references
1. Kimura W, Kuroda A, Morioka Y. Clinical pathology of endocrine tumors of the pancreas. Dig Dis Sci 1991; 36: 933–42. 2. Solcia E, Sessa F, Rindi G, Bonato M, Capella C. Pancreatic endocrine tumors: general concepts; nonfunctioning tumors and tumors with uncommon function. In: Dayal Y, ed. Endocrine Pathology of the Gut and Pancreas. Boca Raton: CRC Press, 1991: 105–31. 3. Solcia E, Capella C, Kloppel G. Tumors of the endocrine pancreas. In: Anonymous Atlas of Tumor Pathology. Washington: AFIP, 1997: 145–209. 4. Kent RB, Van Heerden JA, Weiland LH. Nonfunctioning islet cell tumors. Ann Surg 1981; 193: 185–90. 5. Kloppel G, Heitz PU. Pancreatic endocrine tumors. Pathol Res Pract 1988; 183: 155–68. 6. Howard JN, Moss NH, Rhoads JE. Collective review: hyperinsulinism and islet cell tumors of the pancreas. Int Abstr Surg 1950; 90: 417–55. 7. Lo CY, Van Heerden JA, Thompson GB, et al. Islet cell carcinoma of the pancreas. World J Surg 1996; 20: 878–84 8. Thompson GB, Van Heerden JA, Grant CS, Carney A, Ilstrup DM. Islet cell carcinomas of the pancreas: a twenty-year experience. Surgery 1988; 104: 1011–7 9. Grant CS. Surgical management of malignant islet cell tumors. World J Surg 1993; 17: 498–503 10. Phan GQ, Yeo CJ, Hruban RH, et al. Surgical experience with pancreatic and peripancreatic neuroendocrine tumors: review of 125 patients. J Gastrointest Surg 1998; 2: 473–82 11. Heitz PU, Kasper M, Polak JM, Kloppel G. Pancreatic endocrine tumors: immunocytochemical analysis of 125 tumors. Hum Pathol 1982; 13: 263–71 12. Alpert S, Hanahan D, Teitelman G. Hybrid insulin genes reveal a development lineage for pancreatic endocrine cells and imply a relationship with neurons. Cell 1988; 53: 295 13. Mozzell E, Stenzel P, Woltering EA, Rosch J, O’Dorisio TM. Functional endocrine tumors of the pancreas: clinical presentation, diagnosis, and treatment. Curr Probl Surg 1990; 27: 303–86 14. Bieligk S, Jaffe BM. Islet cell tumors of the pancreas. Surg Clin N Am 1995; 75: 1025–40 15. Friesen SR. Tumors of the endocrine pancreas. N Engl J Med 1982; 306: 580–90 16. Norton JA. Neuroendocrine tumors of the pancreas and duodenum. Curr Probl Surg 1994; 31: 77–164 17. Grant CS. Insulinoma. Surg Oncol Clin N Am 1998; 7: 819–44 18. Van Hoe L, Gryspeerdt S, Marchal G, Baert AL, Mertens L. Helical CT for the preoperative localization of islet cell tumors of the pancreas. AJR 1995; 165: 1437–9 19. Smelka RC, Cumming MJ, Shoenut JP, et al. Islet cell tumors: comparison of dynamic contrast-enhanced CT and MR imaging with dynamic gadolinium enhancement and fat suppression. Radiology 1993; 186: 799–802 20. Muller MF, Meyenberger C, Bertschinger P, Schaer R, Marincek B. Pancreatic tumors: evaluation with endoscopic US, CT and MR imaging. Radiology 1994; 190: 745–51 21. Van Byck CHJ, Bruining HA, Reubi JC, et al. Use of isotope-labeled somatostatin analogs for visualization of islet cell tumors. World J Surg 1993; 17: 444–7 22. Krenning EP, Kwekkeboom DJ, Reubi J, et al. 111In-octreotide scintigraphy in oncology. Metabolism 1992; 41: 83–6 23. Krenning EP, Kwekkeboom DJ, Oei HY, et al. Somatostatin-receptor scintigraphy in gastroenteropancreatic tumors. Ann NY Acad Sci 1994; 733: 416–24 24. Thompson NW, Czako PF, Fritts LL, et al. Role of endoscopic ultrasound in the localization of insulinomas and gastrinomas. Surgery 1994; 116: 1131–8 25. Pasieka JL, MeLeod MK, Thompson NW, et al. Surgical approach to insulinomas: assessing the need for preoperative localization. Arch Surg 1992; 127: 442–7 26. Doppman JL, Miller DL, Chang R, et al. Intraarterial calcium stimulation test for insulinomas. World J Surg 1993; 17: 439–43 27. Danforth DN, Gordon P, Brennan MF, et al. Metastatic insulin secreting carcinoma of the pancreas: clinical course and role of surgery. Surgery 1984; 96: 1027–37

key points
● ●







Clinically silent endocrine tumors have been found in up to 10% of detailed autopsy examinations. Functional islet cell tumors include Insulinoma Gastrinoma Glucagonoma VIPoma Somatostatinoma. Localization of pancreatic endocrine tumors: Helical CT: sensitivity 85% MRI: sensitivity not known Angiography: sensitivity 58% Somatostatin receptor scintigraphy: 82% PET scan: sensitivity not known. Surgical excision, usually with curative intent, is the major objective of treatment. Surgical debulking of primary and metastatic functional islet cell tumors often achieves good palliation of symptoms.

429

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
28. Delcore R Jr, Cheung LY, Friesen SR. Characteristics of duodenal wall gastrinomas. Am J Surg 1990; 160: 621–4 29. Ellison EH, Wilson SD. The Zollinger–Ellison syndrome: reappraisal and evaluation of 260 registered cases. Ann Surg 1964; 160: 512–30 30. Zollinger RM, Ellison EH. Primary peptic ulceration of the jejunum associated with islet cell tumors of the pancreas. Ann Surg 1955; 142: 709–23 31. Meko JB, Norton JA. Management of patients with Zollinger–Ellison syndrome. Ann Rev Med 1995; 46: 395–411 32. Waxsman I, Gardner JD, Jensen RT, Maton PN. Peptic ulcer perforation as the presentation of Zollinger–Ellison syndrome. Dig Dis Sci 1991; 36: 19–24 33. Norton JA, Doppman JL, Jensen RT. Curative resection in Zollinger– Ellison syndrome: results of a 10 year prospective study. Ann Surg 1992; 215: 8–18 34. Wolfe MM, Jensen RT. Zollinger–Ellison syndrome, current concepts in diagnosis and management. New Engl J Med 1987; 317: 1200–9 35. Bondeson AG, Bondeson L, Thompson NW. Stricture and perforation of the esophagus: overlooked threats in the Zollinger-Ellison syndrome. World J Surg 1990; 14: 361–4 36. Pisegna J, Norton JA, Slimak GG, et al. Effects of curative gastrinoma resection on gastric secretory function and antisecretory drug requirements in the Zollinger–Ellison syndrome. Gastroenterology 1992; 102: 767–78 37. Norton JA. Gastrinoma: advances in localization and treatment. Surg Oncol Clin N Am 1998; 7: 845–61 38. Larsson H, Caisson E, Matsson H. Plasma gastrin and gastric enterochromaffin cell (activation and proliferation-studies with omeprazole and ranitidine in intact and adrenalectomized rats. Gastroenterology 1986; 90: 391–9 39. Stabile BE, Morrow DJ, Passaro E Jr. The gastrinoma triangle: operative implications. Am J Surg 1984; 147: 25–31 40. Farley DR, Van Heereden JA, Grant CS, Miller JL, Ilstrup DM. The Zollinger–Ellison syndrome. A collective surgical experience. Ann Surg 1992; 215: 561–9 41. Fraker DL, Norton JA, Alexander R, Venzon DJ, Jensen RT. Surgery in Zollinger–Ellison syndrome alters the natural history of gastrinoma. Ann Surg 1994; 220: 320–30 42. Ellison EC. Forty year appraisal of gastrinoma. Back to the future. Ann Surg 1995; 222: 511–24 43. Miller DL, Norton JA, Jensen RT. Prospective assessment of abdominal ultrasound in patients with Zollinger–Ellison syndrome. Radiology 1991; 178: 763–7 44. Pisegna JR, Doppman JL, Norton JA, Metz DC, Jensen RT. Prospective comparative study of ability of MR imaging and other imaging modalities to localize tumors in patients with Zollinger–Ellison syndrome. Dig Dis Sci 1993; 38: 1318–28 45. Alexander HR, Fraker DL, Norton JA, et al. Prospective study of somatostatin receptor scintigraphy and its effect on operative outcome in patients with Zollinger–Ellison syndrome. Ann Surg 1998; 228: 228–38 46. Ruszniewski P, Amouyal P, Amouyal G, et al. Localization of gastrinomas by endoscopic ultrasonography in patients with Zollinger–Ellison syndrome. Surgery 1995; 117: 629–35 47. Thorn AK, Norton JA, Axiotis CA, Jensen RT. Location, incidence and malignant potential of duodenal gastrinomas. Surgery 1991; 110: 1086–93 48. Howard TJ, Zinner MJ, Stabile BE, et al. Gastrinoma excision for cure. Ann Surg 1990; 211: 9–14 49. Thompson NW, Vinik Al, Eckhauser FE. Microgastrinomas of the duodenum: a cause for failed operations of the Zollinger–Ellison syndrome. Ann Surg 1989; 209: 396–404 50. Carty SE, Jensen RT, Norton JA. Prospective study of aggressive resection of metastatic pancreatic endocrine tumors. Surgery 1992; 112: 1024–32 51. Norton JA, Sugarbaker PH, Doppman JL, et al. Aggressive resection of metastatic disease in selected patients with malignant gastrinoma. Ann Surg 1986; 203: 352–9 52. Veldhius JD, Norton JA, Wells SAJ, et al. Therapeutic controversy; surgical versus medical management of multiple endocrine neoplasia (MEN) type 1. J Clin Endocrinol 1997; 82: 357–64 53. MacFarlane MP, Fraker DL, Alexander HR, et al. Prospective study of surgical resection of duodenal and pancreatic gastrinomas in multiple endocrine neoplasia type 1. Surgery 1995; 118: 973–80 54. Thompson NW. Management of pancreatic endocrine tumors in patients with multiple endocrine neoplasia type 1. Surg Oncol Clin N Am 1998; 7: 881–91 55. Modlin IM, Lewis JJ, Ahiman H, Bilchik AJ, Kumar RR. Management of unresectable malignant endocrine tumors of the pancreas. Surg Gynecol Obstet 1993; 176: 507–18 56. Rothmund M, Stinner B, Arnold R. Endocrine pancreatic tumors. Eur J Surg Oncol J 1991; 17: 191–9 57. Broder LE, Carter SK. Pancreatic islet cell carcinoma: results of therapy with streptozotocin in 52 patients. Ann Int Med 1973; 79: 101–7 58. Boden G, Ryan IG, Eisenschmid BL, et al. Treatment of inoperable glucagonoma with the long acting somatostatin analogue SMS 201995. N Engl J Med 1986; 314: 1686–9 59. Verner JV, Morrison AB. Islet cell tumor and a syndrome of refractory watery diarrhea and hypokalemia. Am J Med 1958; 25: 374–80 60. Said SI, Mutt V. Isolation from porcine-intestinal wall of a vasoactive octacosapeptide related to secretin and glucagon. Eur J Biochem 1972; 28: 199–204 61. Friesen SR. Update on the diagnosis and treatment of rare neuroendocrine tumors. Surg Clin N Am 1987; 67: 379–93 62. Mekhjian HS, O’Dorisio TM. VIPoma syndrome. Semin Oncol 1987; 14: 282–91 63. Inamoto K, Yoshino F, Nakao N, et al. Angiographic diagnosis of a pancreatic islet tumor in a patient with the WDHA syndrome. Gastrointest Radiol 1980; 5: 259–61 64. Delcore R, Friesen SR. Gastrointestinal neuroendocrine tumors. J Am Coll Surg 1994; 178: 188–211 65. Moertel CG, Hanley JA, Johnson LA. Streptozocin alone compared with streptozocin plus fluorouracil in the treatment of advanced islet-cell carcinoma. N Engl J Med 1980; 303: 1189–94 66. O’Dorisio TM, Mekhjian HS, Gaginella TS. Medical therapy of VIPomas. Endocrinol Metab Clin N Am 1989; 18: 545–56 67. Maton PN. Use of octreotide for control of symptoms in patients with islet cell tumors. World J Surg 1993; 17: 504–10 68. Harris GJ, Tio F, Cruz AB. Somatostatinoma of the duodenum: case report and review of the literature. J Surg Oncol 1987; 36: 8–16 69. Konomi K, Chijiiwa K, Katsuta T, et al. Pancreatic somatostatinoma: case report and review of the literature. J Surg Oncol 1990; 43: 259–69 70. Thompson NW, Eckhauser FE. Malignant islet cell tumors of the pancreas. World J Surg 1984; 8: 940–51 71. Legaspi A, Brennan ME. Management of islet cell carcinoma. Surgery 1988; 104: 1018–23 72. Debas HT, Mulvihill SJ. Neuroendocrine gut neoplasms. Arch Surg 1994; 129: 965–72 73. Broughan TA, Leslie JD, Soto JM, Hermann RE. Pancreatic islet cell tumors. Surgery 1986; 99: 671–8 74. Stephens DH. CT of pancreatic neoplasms. Curr Probl Daign Radiol 1997; 26: 53–108 75. Eelkema EA, Stephens DH, Ward EM, Sheedy PF. CT features of nonfunctioning cell carcinoma. AJR 1984; 143: 943–8 76. Buetow PC, Parrino TV, Buck JL, et al. Islet cell tumors of the pancreas: pathologic-imaging correlation among size, necrosis and cysts, calcification, malignant behavior, and function. AJR 1995; 165: 1175–9 77. Strodel WE, Vinik Al, Lloyd RV, et al. Pancreatic polypeptide-producting tumors. Arch Surg 1984; 119: 508–14 78. Prinz RA, Marangos PJ. Serum neuron-specific enolase: a serum marker for nonfunctioning pancreatic islet cell carcinoma. Am J Surg 1983; 145: 77–81 79. La Rosa S, Sessa F, Capella C, et al. Prognostic criteria in nonfunctioning pancreatic endocrine tumors. Virchows Arch 1996; 429: 323–33 80. Cheslyn-Curtis S, Sitaram V, Williamson RCN. Management of nonfunctioning neuroendocrine tumors of the pancreas. Br J Surg 1993; 80: 625–7 81. Madura JA, Cummings OW, Wiebke EA, et al. Nonfunctioning islet cell tumors of the pancreas: a difficult diagnosis but one worth the effort. Am Surg 1997; 63: 573–8

430

NEUROENDOCRINE PANCREATIC TUMORS
82. McFarland EG, Kaufman JA, Saini S, et al. Preoperative staging of cancer of the pancreas: value of MR angiography versus conventional angiography in detecting portal venous invasion. AJR 1996; 166: 37–43 83. Harrison LE, Klimstra D, Brennan MF. Portal vein resection for adenocarcinoma of the pancreas: a contraindication for resection. Ann Surg 1996; 224: 342–9 84. Evans DB, Skibber JM, Lee JE, et al. Nonfunctioning islet cell carcinoma of the pancreas. Surgery 1993; 114: 1175–82 85. Conlon KC, Dougherty E, Klimstra DS, et al. The value of minimal access surgery in the staging of patients with potentially resectable peripancreatic malignancy. Ann Surg 1996; 223: 134–40 86. Patel YC, Amherdt M, Orci L. Somatostatin-receptor scintigraphy in gastroenteropancreatic tumors. Science 1982; 217: 1155–6 87. Van Eijck CHJ, Lamberts SWJ, Lemaire LCJM, et al. The use of somatostatin receptor scintigraphy in the differential diagnosis of pancreatic duct cancers and islet cell tumors. Ann Surg 1996; 224: 119–24 88. Scherubl H, Bader M, Fett U, et al. Somatostatin-receptor imaging of neuroendocrine gastroenteropancreatic tumors. Gastroenterology 1993; 105: 1705–9 89. Ahlstrom H, Eriksson B, Bergstrom M, Biurling P, Langstrom B, Oberg K. Pancreatic neuroendocrine tumors: diagnosis with PET. Radiology 1995; 195: 333–7 90. Zogakis TG, Norton JA. Palliative operations for patients with unresectable endocrine neoplasia. Surg Clin N Am 1995; 75: 525–38 91. Makowka L, Tzakis AG, Mazzaferro V, et al. Transplantation of the liver for metastatic endocrine tumors of the intestine and pancreas. Surg Gynecol Obstet 1989; 168: 107–11 92. Lang H, Oldhafer KJ, Weimann A, et al. Liver transplantation for metastatic neuroendocrine tumors. Ann Surg 1997; 225: 347–54 93. Le Treut YP, Delpero JR. Dousset B, et al. Results of liver transplantation in the treatment of metastatic neuroendocrine tumors. Ann Surg 1997; 225: 355–64 94. McEntee GP, Nagorney DM, Kvols LK, Moertel CG, Grant CS. Cytoreductive hepatic surgery for neuroendocrine tumors. Surgery 1990; 108: 1091–6 95. Ajani JA, Carrasco CH, Chamsangavej C, et al. Islet cell tumors metastatic to the liver: effective palliation by sequential hepatic artery embolization. Ann Int Med 1988; 108: 340–4 96. Prinz RA, Badrinath K, Chejfec G, Freeark RJ, Greenlee HB. ‘Nonfunctioning’ islet cell carcinoma of the pancreas. Am Surg 1983; 49: 345–9 97. Moertel CG, Lefkopoulo M, Lipsitz S, Hahn RG, Klaassen D. Streptozocin-doxorubicin, streptozocin-fluorouracil, or chlorozotocin in the treatment of advanced islet-cell carcinoma. N Engl J Med 1992; 326: 519–23 Torrisi JR, Treat J, Zeman R, Dritschilo A. Radiotherapy in the management of pancreatic islet cell tumors. Cancer 1987; 60: 1226–31 Tennvall J, Ljungberg O, Ahren B, Gustavsson A, Nillson LO. Radiotherapy for unresectable endocrine pancreatic carcinoma. Eur J Surg Oncol 1992; 18: 73–6 Saltz L, Trochanowski B, Buckley M, et al. Octreotide as an antineoplastic agent in the treatment of functional and nonfunctional neuroendocrine tumors. Cancer 1993; 72: 244–8 White TJ, Edney JA, Thompson JS, Karrer FW, Moor BJ. Is there a prognostic difference between functional and nonfunctional islet cell tumors. Am J Surg 1994; 168: 627–30 Venkatesh S, Ordonez NG, Ajani J, et al. Islet cell carcinoma of the pancreas. Cancer 1990; 65: 354–7 Doherty GM, Doppman JL, Shawker TH, et al. Results of a prospective strategy to diagnose, localize, and resect insulinomas. Surgery 1991; 110: 989–97 Grama D, Eriksson B, Martensson H, et al. Clinical characteristics, treatment and survival in patients with pancreatic tumors causing hormonal syndromes. World J Surg 1992; 16: 632–9 Menegaux F, Schmitt G, Mercadier M, Chigot JP. Pancreatic insulinomas. Am J Surg 1992; 165: 243–8 Angelini L, Bezzi M, Tucci GJ, et al. The ultrasonic detection of insulinomas during surgical exploration of the pancreas. World J Surg 1987; 11: 642–7 Grant CS, Van Heerden JA, Charboneau JW, et al. Insulinoma: the value of intraoperative ultrasonography. Arch Surg 1988; 123: 843–8 McArthur KE, Richardson CT, Barnett CC, et al. Laparotomy and proximal gastric vagotomy in Zollinger–Ellison syndrome: results of a 16-year prospective study. Am J Gastroenterol 1996; 91: 1104–11 Melvin WS, Johnson JA, Sparks J, et al. Long-term prognosis of Zollinger– Ellison syndrome in multiple endocrine neoplasia. Surgery 1993; 114: 1183–8 Jaskowiak NT, Fraker DL, Alexander HR, et al. Is reoperation for gastrinoma excision indicated in Zollinger–Ellison syndrome. Surgery 1996; 120: 1055–63 Dial PF, Braasch JW, Rossi RL, Lee AK, Jin G. Management of nonfunctioning islet cell tumors of the pancreas. Surg Clin N Am 1985; 65: 291–9 Eckhauser FE, Cheung PS, Vinik Al, et al. Nonfunctioning malignant neuroendocrine tumors of the pancreas. Surgery 1986; 100: 978–87 Yeo CJ, Wang BH, Anthone GJ, Cameron JL. Surgical experience with pancreatic islet cell tumors. Arch Surg 1993; 128: 1143–8

98. 99.

100.

101.

102. 103.

104.

105. 106. 107. 108.

109.

110.

111. 112. 113.

431

46 Rare tumors of the pancreas
introduction

Jooyeun Chung, Lisa J. Harris, Hamid Abdollahi, and Charles J. Yeo
were more aggressive, again raising the role of sex hormones in the pathogenesis of SPPN (8).

Although ductal adenocarcinoma is the most common malignant neoplasm of the pancreas, there are multiple unusual tumors of the pancreas that surgeons must keep in mind. A thorough history and physical exam and a high-resolution dedicated pancreatic CT scan are crucial in differentiating these rare tumors from the more commonly seen “tumors.” At times, MRI/MRCP, endoscopic retrograde cholangiopancreatography, and endoscopic ultrasound can aid in making the diagnosis. Here, we discuss nine rare “tumors” of the pancreas and their distinguishing features (Table 46.1).

acinar cell carcinoma (acc)
ACCs account for approximately 1% of all primary pancreatic neoplasms. ACC tends to occur in the fifth to seventh decade of life with 2:1 ratio of males to females. The presenting symptoms are nonspecific, including abdominal pain, bloating, postprandial vomiting, or a change in bowel habits (9–11). The neoplastic cells have a unique ability to produce pancreatic enzymes such as trypsin, chymotrypsin, lipase, and amylase. Lipase hypersecretion syndrome, though well described in the literature, is seen in less than 15% of patients with ACC and is characterized by subcutaneous fat necrosis, polyarthralgia, and eosinophilia. Although ACC was previously thought to be found predominantly in the head of the pancreas, a recent review by Wisnoski et al. suggests that ACC is more often found in the body or the tail of the pancreas (12). ACC is usually an exophytic, well demarcated, and large hypodense mass on CT. The larger tumors can be heterogenous with internal calcifications or intratumoral hemorrhage (13,14). On gross examination, these tumors are soft, fleshy, and well-circumscribed masses. Histologically, ACCs have four patterns of growth: acinar, solid, trabecular, and glandular with the acinar pattern being the most common. They are often highly cellular with minimal stroma. In addition, they have a unique immunohistochemical staining pattern for trypsin, lipase, amylase, phospholipase, and chymotrypsin. In one series of 28 patients, ACC was positive for trypsin in 100% of the cases. They typically fail to stain for synaptophysin, chromogranin, and other islet cell hormones, thus helping to differentiate them from pancreatic neuroendocrine tumors (15). ACC had been previously thought to be a cancer of poor prognosis, comparable or worse than pancreatic ductal adenocarcinoma. However, review of the current literature for survival analysis suggests otherwise. The stage-specific survival is significantly better for ACC than adenocarcinoma, and resection helps to greatly improve this survival. On multivariate analysis, age less than 65, well-differentiated tumor, and R0 resection were the independent prognostic factors (17). Current survival data are summarized in Table 46.2.

solid pseudopapillary neoplasm (sppn)
SPPN of the pancreas accounts for 1% to 3% of all pancreatic neoplasms. It has a low malignant potential and is predominantly seen in young females in their 20s and 30s. SPPN is often an incidental finding when patients undergo imaging for other reasons or the patient may present with nonspecific gastrointestinal symptoms (1). SPPN vary widely in size. The larger tumors may undergo cystic or hemorrhagic degeneration and occasionally have calcifications (Fig. 46.1). They are typically well circumscribed and surrounded by a pseudocapsule. Histologically, the tumors show solid regions composed of nests and sheets of uniform epithelioid cells alternating with cystic spaces and pseudopapillae (2). Much has been hypothesized about the pathogenesis of SPPN, given its predilection for young females. One theory involves the genital ridge-related cells that could have been incorporated into the pancreas during early embryogenesis secondary to the close proximity of the genital ridge and the pancreatic anlage (3). There has been a case report of SPPN in a 2-year-old female which would support the embryology theory (4). The role of sex hormones also has been investigated. To date, the expression of progesterone receptors has been well documented but the status of the estrogen receptors has remained questionable. Geers et al. reported on the differential expression of ER versus ER but this observation warrants further investigation (5). On a molecular level, greater than 90% of SPPN have mutations in the B-catenin gene, resulting in the disruption of E-cadherin, a key regulator of cell–cell junctions (6). Although SPPN is typically larger than pancreatic adenocarcinoma at the time of diagnosis, it is usually resectable and R0 resection is generally curative. The overall mortality from the neoplasm is 2% and the recurrence rate is 10% to 15%. The cases which recur or present with metastatic disease have nuclear pleomorphism, a high mitotic count with diffuse infiltrate growth pattern, and vascular invasion (7). A study by Marchado et al., which had the highest number of male patients (7 out of 34), reported that the male patients with SPPN were older, had higher rate of vascular invasion, and

adenosquamous carcinoma (asc)
ASC accounts less than 1% to 4% of pancreatic tumors. There are many case reports but only a few series discussing ASC exist in current literature. According to Kardon et al., which is the largest report on ASC to date, the mean age of the patients is 65 years, the tumors vary widely in size (2–12 cm), and ASC is predominantly located in the head of the pancreas. The clinical symptoms are similar to ductal adenocarcinoma, including weight loss, obstructive jaundice, and abdominal pain (18,19).

432

RARE TUMORS OF THE PANCREAS
Table 46.1 Rare Tumors of the Pancreas
Percentage of all pancreas tumors 1–3%

Type of tumor Solid pseudopapillary neoplasm

Demographics Young females

Imaging Large, well circumscribed, cystic or hemorrhagic degeneration Large, hypodense, exophytic, well demarcated Wide range of size Well circumscribed lesion or infiltrating lesion with poorly defined borders Often hypervascular spherical lesions Cysts, adenomas, and hypervascular neuroendocrine tumors Common local infiltration and metastases Similar to ductal adenocarcinoma

Treatment Resection

Prognosis Low malignant potential, cured with R0 resection

Acinar cell carcinoma

<1%

5th to 7th decade of life 2:1 M:F 6th decade of life 5th to 6th decade of life M > F

Resection

Better than adenocarcinoma Worse than adenocarcinoma Better than adenocarcinoma

Adenosquamous carcinoma Pancreatic lymphoma

<1–4% <0.5%

Resection Chemotherapy ± radiotherapy. Occasionally resection

Metastases to the pancreas Von Hippel–Lindau syndrome

<1% Rare

Variable M=F

Different for various Variable primaries Resection if Often determined by neuroendocrine other VHL lesions tumor Resection, if appropriate ?chemoradiation Resection Similar to adenocarcinoma Better than adenocarcinoma if positive for microsatellite instability Good

Giant cell tumor

Rare

6th decade of life M=F 6th to 7th decade of life M > F

Medullary carcinoma

?<5%

Autoimmune pancreatitis

Uncommon

6th to 7th decade of life M > F

Focal enlargement of pancreas

Steroid therapy

Note: Autoimmune pancreatitis (also termed lymphoplasmacytic sclerosing pancreatitis) is not a neoplasm, but may be mistaken for a neoplastic process.

Figure 46.1 CT scan showing a large heterogenous mass in the tail of the pancreas (arrow) with central calcification. The surgical specimen was consistent with a solid pseudopapillary neoplasm.

By definition, ASC must have both histologic components of squamous and adenocarcinoma (20). Some authors in the past had arbitrarily set 30% as the minimum amount of squamous carcinoma component required to diagnose ASC in a specimen.

However, Kardon et al. showed a wide range in the proportion of squamous carcinoma in individual specimens. Thus, it was their opinion that an adenocarcinoma showing any degree of malignant squamous cell differentiation should be considered as ASC. It has been hypothesized that ASC is the result of malignant squamous metaplasia (dedifferentiation) within an adenocarcinoma. To further support this theory, Kardon et al. studied K-ras oncogene mutation, which has been strongly linked to ductal adenocarcinoma (>95% have K-ras mutation). In their series of ASC, they found that greater than 50% of their specimen had K-ras mutation, even in the tumors that were almost exclusively squamous carcinoma. However, the presence of K-ras mutation did not have any prognostic implications. Recent molecular studies of ASC have shed more light on the pathogenesis of this aggressive tumor, indicating that K-ras mutations may be crucial for the development of ASC. ASC is a more aggressive tumor than ductal adenocarcinoma. Kardon et al. reported average survival of 11 months for ASC patients who underwent R0 resection versus a median survival of 18 to 20 months for resectable adenocarcinoma. Patients who underwent palliative procedures had short median survival of 3.8 months. Smoot et al. found similar survival data in their study; patients who underwent R0 resection had median survival of 14.4 months versus 4.8 months

433

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 46.2 Summary of Survival Data on Acinar Cell Carcinoma in the Current Literature
Author Wisnoski (12) Seth (11) Kitagami (16) Holen (10) Year 2008 2008 2007 2002 Number of patients 672 14 115 39 Mean age (year) 56 57 59.6 60 Median survival (month) 47 33 19 Median survival after resection (month) 123 41

without resection (21). Obviously, a complete surgical resection is recommended for ASC whenever possible. While the standard of care currently recommends adjuvant chemo- or chemoradiation therapy for resected ASC patients, no trials have proven a survival benefit in this patient cohort.

pancreatic lymphoma
Primary pancreatic lymphoma (PPL) is a rare disease, representing less than 0.5% of all pancreatic tumors and less than 2% of extra nodal non-Hodgkin’s lymphomas (NHL) (22,23). PPL occurs more frequently in men than women and usually presents in the 5th to 6th decade (median age 57.1 years) (24). The clinical presentation of PPL is often nonspecific. The most common presenting symptom is abdominal pain (83%), followed by abdominal mass (58%), weight loss (50%), jaundice (37%), acute pancreatitis (12%), small bowel obstruction (12%), and diarrhea (12%) (25). The classic B-symptoms of nodal NHL are seen in less than 2% of PPL patients (27). Laboratory studies are often non-diagnostic. Two different morphologic patterns of involvement are seen on CT: a well-circumscribed mass or an infiltrating lesion with poorly defined borders (27). Certain CT findings such as a bulky head of pancreas tumor without significant dilation of the bile duct or main pancreatic duct and enlarged lymph nodes below the level of the renal vein suggest PPL over adenocarcinoma (23,27). However, cytohistological examination of tissue, obtained by image-guided fine needle aspiration (FNA) or endoscopically guided FNA, is required to confirm the diagnosis of PPL (24,28). Surgery is usually reserved for rare cases where less invasive modalities fail to provide a tissue diagnosis. The treatment of PPL has varied, due to the lack of prospective randomized studies to delineate management. Chemotherapy has been the mainstay of treatment with CHOP (Cyclophosphamide, Adriamycin, Vincristine, Prednisone), CVP (Cyclophosphamide, Vincristine, Prednisolone), and MACOP-B (Methotrexate, Leucovorin, Doxorubicin, Cyclophosphamide, Vincristine, Bleomycin) being used most commonly (26,29). Radiotherapy has been used variably, alone and as consolidation after chemotherapy (26). Recently rituximab has been used in combination with CHOP resulting in improved response rates (30). Surgical resection for stage I and II disease has been shown to increase cure rates and may be used as part of multimodality therapy for resectable lesions (31). In general, the outcomes for pancreatic lymphoma are better than for ductal adenocarcinoma.

Figure 46.2 CT scan showing a large heterogenous mass involving the body and the tail of the pancreas (arrow) in a 70-year-old male who had previously undergone right upper lobectomy for lung adenocarcinoma. This mass was found during a follow-up PET CT scan. The pathology specimen confirmed the diagnosis of a metastasis from the lung primary.

metastatic disease to the pancreas
The pancreas is the site of metastases from a wide variety of primary neoplasms. The five most common cancers with

isolated metastases to the pancreas are renal, lung, breast, colon, and melanoma (32). Metastases to the pancreas are usually identified during initial metastatic work-up of a primary tumor, during routine follow-up after resection of the primary cancer, or via the presence of symptoms secondary to the pancreatic lesion (32). On CT imaging, hypervascularity of the lesion, absence of enlarged lymph nodes, and multicentric lesions suggest metastatic disease to the pancreas over a primary neoplasm (Fig. 46.2) (33). Treatment options (both surgical and nonsurgical) depend on the type of primary tumor, location of the metastasis, the extent of disease, and the presence or absence of symptoms related to the metastatic tumor (32). Renal cell carcinoma (RCC) is the most common cancer associated with isolated metastasis to the pancreas (34). Pancreatic metastasis from RCC is usually metachronous, occurring an average of 9.2 years after initial resection (32,35). Most patients are asymptomatic, with lesions detected through routine post-RCC surveillance (36,37). Surgical resection yields a 5-year survival between 53% and 75% compared to 5% to 30% for those with unresectable lesions (32). Metastectomy in patients with RCC does seem to confer a survival benefit and should be performed in eligible patients.

434

RARE TUMORS OF THE PANCREAS
Table 46.3 Summary of the Most Common Isolated Metastasis to the Pancreas
Primary tumor Renal cell cancer Lung Breast Colon Melanoma Clinical setting Isolated pancreatic metastasis Diffuse metastatic disease Diffuse metastatic disease Direct extension from right or transverse colon primary Multiple intra-abdominal metastasis Recommended treatment Metastectomy Metastectomy not recommended Metastectomy in conjunction with multimodality therapy En bloc resection (typically pancreaticoduodenectomy) Metastectomy if complete resection of all intra-abdominal sites is possible Survival benefit Improved survival No survival benefit May improve survival Improved survival Improved survival

Isolated pancreatic metastases from lung cancer are rare. These patients have a poor prognosis and most do not benefit from metastectomy (32,38). Pancreatic metastasis from breast cancer usually occurs in the setting of diffusely metastatic disease; while controversial, surgical resection, in conjunction with multimodality therapy, may improve survival (32). Pancreatic metastasis from colon cancer often occurs secondary to local invasion; en bloc resection has been found to improve prognosis and is advocated in carefully selected cases (32). Pancreatic metastasis from malignant melanoma is generally found in conjunction with metastasis to other intraabdominal organs (39). Though the data are limited with regard to pancreatic resection for melanoma, improved survival has been documented when complete resection of all intra-abdominal sites of metastatic melanoma can be achieved (39,40). The different types of primary tumors metastatic to the pancreas and their recommended treatment are summarized in Table 46.3.

for pancreatic cysts or serous cystadenomas (43) but surgical resection is recommended for selected neuroendocrine tumors due to their malignant potential. The criteria for resection of VHL neuroendocrine tumors are (1) no metastatic disease and size greater than 3 cm in the body or tail, or greater than 2 cm in the head or (2) patient undergoing laparotomy for other lesions (47). Patients who do not meet criteria for resection should be successfully followed with yearly CT scan to assess for tumor enlargement (42). The most common sites of metastatic neuroendocrine tumor in VHL are the liver and peripancreatic lymph nodes; treatment consists of chemotherapy (and duodenal or biliary stenting as needed for palliation of space occupying lesions) (42).

giant cell tumors
Giant cell tumors of the pancreas are very rare. They appear as two distinct histopathologies: pleomorphic giant cell carcinoma and osteoclast-like giant cell tumor of the pancreas (OGTP) (49). The pleomorphic giant cell variant is highly anaplastic with pleomorphic mononucleated and multinucleated giant cells (50). OGTP is rarer and has a more favorable prognosis than the undifferentiated pleomorphic giant cell carcinoma (51). OGTP is characterized by osteoclast-like giant cells and mononuclear stromal cells identical to the those found in giant cell tumors of the bone (49). The histogenesis of this tumor continues to remain controversial, with both epithelial and mesenchymal origins being suggested (52). OGTP accounts for much less than 1% of all malignant neoplasms of the pancreas (53). The presenting symptoms include abdominal pain, weight loss, jaundice, and occasionally a palpable mass. The average age of onset is 60 years, although there is a wide range of 30 to 80 years of age, and it affects males and females equally. The majority of tumors are found in the head of the pancreas. Local invasion is common at presentation. Approximately 50% of patients have metastasis at the time of diagnosis (49). The overall prognosis appears to be marginally better than pancreatic adenocarcinoma, with median survival less than 1 year (54). These cancers may arise in association with an adenocarcinoma of the pancreas or a mucinous cystic neoplasm (55). Although there is no definitive treatment algorithm because of the rarity of the neoplasm, surgical resection should be pursued if appropriate (51). The role for radiation and chemotherapy is unclear. Such treatment is often used, extrapolating from data generated in the treatment of giant cell tumors of the bone (49).

von hippel-lindau syndrome
A germline mutation of a tumor suppressor gene on the short arm of chromosome 3 (3p25–26) causes a multisystem cancer syndrome, called the Von Hippel–Lindau (VHL) syndrome (41,42). VHL has an autosomal dominant inheritance pattern. The penetrance of the disease is greater than 90% by 65 years of age (43). It is found in all ethnic groups and affects both genders equally. The prevalence of VHL disease ranges from 1:30,000 to 1:50,000 individuals (44). Affected individuals may develop CNS and retinal hemangioblastomas, renal cysts, renal carcinoma, pheochromocytoma, pancreatic cysts, pancreatic neuroendocrine tumors, as well as epididymal and broad ligament cystadenomas (42). In patients with a family history of VHL, the diagnosis can be made with the finding of a single cerebellar or retinal hemangioblastoma, RCC, or pheochromocytoma. For patients without a family history, the diagnosis requires two or more hemangioblastomas or one hemangioblastoma and another visceral lesion (45). Pancreatic involvement in VHL can present in the form of pancreatic cysts (30–70% of affected individuals), serous cystadenomas (10%), or neuroendocrine tumors (11–17%) (46–48). Pancreatic neuroendocrine tumor typically appears on CT scan as a well circumscribed, enhancing mass in early phase images and on EUS as a homogenous, hypoechoic mass (46). Somatostatin receptor scintigraphy (SRS) can be used as an adjunct in diagnosis. Surgical intervention is not required

435

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

autoimmune pancreatitis (aip)
AIP, also known as lymphoplasmacytic sclerosing pancreatitis (LPSP), is characterized by a chronic inflammation of the pancreas with prominent lymphocytic infiltration leading to fibrosis (61). It was initially described by Sarles et al. in 1961 as a case of sclerosing pancreatitis with hypergammaglobulinemia (62,63). However, it was not until 1995 that the term “autoimmune pancreatitis” was introduced by Yoshida et al. (64). In the past 10 years, many clinicopathological aspects of AIP have been clarified and the disease has become a discrete entity. While not a neoplasm, AIP can mimic neoplasia. There are four histologic characteristics of AIP: dense infiltration of the pancreas with lymphoplasmacytic cells, interstitial fibrosis, “collar like” periductal inflammation around medium-sized interlobular ducts, and obliterative periphlebitis (30). The infiltrating cells comprise CD4+ and CD8+ lymphocytes and IgG4+ plasma cells (61). On gross examination, the fibrous process in the pancreas can mimic ductal pancreatic adenocarcinoma (65). The disease is frequently centered at the head of the gland, although the extent of disease can vary with some reports suggesting a diffuse involvement of the entire pancreas. The gland is gray to white in color and firm, often described as “sausage like.” There is also an associated loss of lobular architecture, and the inflammation can spread to involve the common bile duct and gallbladder (66). The two most common presenting symptoms of AIP are jaundice (67%) and abdominal pain (35%). Patients may also present with anorexia, weight loss, diabetes, and pancreatic insufficiency (30,61,65). In about 20% of patients, AIP is associated with another autoimmune disease, including Sjögren’s syndrome, primary sclerosing cholangitis, ulcerative colitis, and Crohn’s disease (66). AIP usually presents in the 6th and 7th decade of life and has a male predominance (65,66). Serologically, patients have been found to have elevations in several autoantibodies, including carbonic anhydrase, anti-lactoferrin, anti-nuclear factor, and various gamma globulins, especially IgG4 (64,67). On CT imaging, patients will frequently have focal enlargement of the entire pancreas, most prominently in the head (Fig. 46.3) (64,68). Other findings include a capsulelike rim on MRI T2-weighted images and diffuse, irregular stenosis of the pancreatic duct as seen by MRCP or ERCP (68). Treatment for confirmed AIP has evolved to medical management, with many patients having a dramatic response to steroid therapy within 4 weeks. There are no strict rules for dosage and duration, although most publications report doses of prednisone starting at 30 to 40 mg/day for at least 1 to 2 weeks, with a gradual taper. There is a reported relapse rate of between 10% and 20%, which usually responds to additional steroids. There has been a small group of patients which have required maintenance therapy (61,64). Differentiating AIP from adenocarcinoma is crucial due to the differences in treatment of these two entities. There have been several sets of diagnostic criteria proposed for AIP including the Japanese, the Korean, and the Mayo Clinic criteria (Table 46.4) (69,70). Despite these criteria, there remain a small group of patients who may come to surgical resection to rule out the presence of neoplasia in an abnormal gland (71). Additionally, there have been reports of

Figure 46.3 Typical CT scan demonstrating the “sausage-like” appearance of the body and tail of the pancreas in autoimmune pancreatitis (arrow). The gallbladder is prominent and the intrapancreatic portion of the distal common bile duct is dilated.

medullary carcinoma
Initially described by Goggins et al. in 1998 as pancreatic adenocarcinomas with DNA replication errors (RER+), medullary carcinomas of the pancreas are poorly differentiated carcinomas that have been frequently grouped with conventional ductal carcinoma of the pancreas. However, they are histologically distinct, described by syncytial growth pattern, expanding tumor borders, and extensive necrosis (56). Furthermore, they have unique genetic features and tumor pathogenesis. Although not clearly known, up to 5% of pancreatic cancers may be medullary variants. Clinically, patients with these tumors present in the 6th and 7th decade of life with a slight male to female predominance and most patients have a family history of cancer in a first degree relative. On a molecular level, medullary carcinomas of the pancreas were initially reported to have a wild-type K-ras gene. However, later studies have not confirmed this as a universal finding. In addition, medullary cancers harbor microsatellite instability (MSI) in about 22% cases as compared to a lack of MSI in ductal adenocarcinomas (56). This suggested a distinct genetic pathway for medullary tumor development that is different from pancreatic adenocarcinoma (57). MSI is caused by mutations or methylation in such DNA repair genes as MLH1 and MSH2 (and others), which can be inherited or acquired (58). Patients with MSI-positive medullary tumors are observed to have better prognosis than those with ductal adenocarcinoma. Bunzo et al. examined the predictive value of MSI on prognosis of pancreatic cancer and found that MSI-positive tumors had significant longer survival times than their MSInegative counterpart (59,60). Such a beneficial association of MSI positivity with prognosis has been observed in other types of cancers including colorectal, gastric, and ampullary cancer. The treatment for localized disease is surgical resection. The optimal post-resection chemotherapy remains unknown, although this tumor’s unique molecular profile suggests a role for specifically targeted therapy.

436

RARE TUMORS OF THE PANCREAS
Table 46.4 Japanese, Korean, and Mayo Criteria for Diagnosis of Autoimmune Pancreatitis
Diagnostic criteria Histology Japanese criteria Marked interlobular fibrosis and prominent infiltration of lymphocytes and plasma cells in the periductal area, occasionally with lymphoid follicles in the pancreas Korean criteria Fibrosis and lymphoplasmocytic infiltration Mayo criteria At least one of the following: (1) Periductal lymphoplasmacytic infiltrate with obliterative phlebitis and storiform fibrosis (2) Lymphoplasmacytic infiltrate with storiform fibrosis showing abundant (>10 cell/HPF) IgG4-positive cells Elevated serum IgG4 levels

Serology

Imaging

High serum gamma-globulins, IgG or IgG4, or the presence of autoantibodies such as antinuclear antibodies and rheumatoid factor Diffuse or segmental narrowing of the main pancreatic duct with diffuse or localized enlargement of the pancreas by imaging studies such as abdominal ultrasound, CT, and MRI

At least one of the following: (1) Elevated levels of IgG and/IgG4 (2)Detected autoantibodies (1) CT: diffuse enlargement (swelling) of the pancreas (2) ERCP: diffuse or segmental narrowing of the pancreatic duct

Involvement of other organs

Not included

Association with other postulated autoimmune diseases

Response to steroid therapy

Not included

Dramatic resolution of narrowing of the pancreatic duct

Common: diffusely enlarged gland with delayed (rim) enhancement; diffusely irregular, attenuated main pancreatic duct. Others: focal pancreatic mass/ enlargement; focal pancreatic duct stricture, pancreatic duct atrophy, pancreatic calcifications; or pancreatitis Hilar/intrahepatic biliary strictures, persistent distal biliary stricture, parotid/lacrimal gland involvement, mediastinal lymphadenopathy, retroperitoneal fibrosis Resolution/marked improvement of pancreatic/extrapancreatic manifestation with steroids

Source: Modified from Refs. (69,70,71).

synchronous AIP and ductal adenocarcinoma, treated via pancreatic resection (72).

references
1. Campbell F, Azadeh B. Cystic Neoplasms of the exocrine pancreas. Histopathology 2008; 52: 539–51. 2. Adams AL, Seigal GP, Jhala NC. Solid pseudopapillary tumor of the pancreas. Adv Anat Pathol 2008; 15: 39–45. 3. Liu X, Rauch TM, Siegal GP, et al. Solid-pseudopapillary neoplasm of the pancreas. Three cases with a literature review. Appl Immunohistochem Mol Morpho 2006; 14: 445–53. 4. Mao C, Guvendi M, Domenico DR, et al. Papillary cystic and solid tumors of the pancreas: a pancreatic embryonic tumor? Studies of three cases and cumulative review of the world’s literature. Surgery 1995; 118: 821–8. 5. Geers C, Moulin P, Gigot JF, et al. Solid and pseudopapillary tumor of the pancreas—review and new insights into pathogenesis. Am J Surg Pathol 2006; 30: 1243–9. 6. Shi C, Daniels JA, Hruban RH. Molecular characterization of pancreatic neoplasms. Adv Anat Pathol 2008; 15: 185–95. 7. Tipton SG, Smyrk TC, Sarr MG, et al. Malignant potential of solid pseudopapillary neoplasm of the pancreas. Br J Surg 2006; 93: 733–7. 8. Machado MCC, Machado MAC, Bacchella T, et al. Solid pseudopapillary neoplasm of the pancreas: distinct patterns of onset, diagnosis, and prognosis for male versus female patients. Surgery 2008; 143: 29–34. 9. Riechelmann RP, Hoff PM, Moron RA, et al. Acinar cell carcinoma of the pancreas. Int J Gastrointest Cancer 2003; 23: 67–72.

10. Holen KD, Klimsra DS, Hummer A, et al. Clinical characteristics and outcomes from an institutional series of acinar cell carcinoma of the pancreas and related tumors. J Clin Oncol 2002; 20: 4673–8. 11. Seth AK, Argani P, Campbell KA, et al. Acinar cell carcinoma of the pancreas: an institutional series of resected patients and review of the current literature. J Gastrointest Surg 2008; 12: 1061–7. 12. Wisnoski NC, Townsend CM Jr, Nealon WH, et al. 672 Patients with acinar cell carcinoma of the pancreas: a population-based comparison to pancreatic adenocarcinoma. Surgery 2008; 144: 141–8. 13. Tatli S, Mortele KJ, Levy AD, et al. CT and MRI features of pure acinar cell carcinoma of the pancreas in adults. AJR 2005; 184: 511–9. 14. Chiou YY, Chiang JH, Hwang JI, et al. Acinar cell carcinoma of the pancreas clinical and computed tomography manifestations. J Comput Assist Tomogr 2004; 28: 180–6. 15. Mortenson MM, Katz MHG, Tamm EP, et al. Current diagnosis and management of unusual pancreatic tumors. Am J Surg 2008; 196: 100–13. 16. Kitagami H, Kondo S, Hirano S, et al. Acinar cell carcinoma of the pancreas: clinical analysis of 115 patients from Pancreatic Cancer Registry of Japan Pancreas Society. Pancreas 2007; 35: 42–6. 17. Schmidt CM, Matos JM, Bentrem DJ, et al. Acinar cell of the pancreas in the United State: prognostic factors and comparison to ductal adenocarcinoma. J Gastrointest Surg 2008; 12: 2078–86. 18. Kardon DE, Thompson LDR, Przygodki RM, et al. Adenosquamous carcinoma of the pancreas: a clinicopathologic series of 25 cases. Mod Pathol 2001; 14: 443–51. 19. Hsu JT, Yeh CN, Chen YR, et al. Adenosquamous carcinoma of the pancreas. Digestion 2005; 72: 104–8.

437

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
20. Rahemtullah A, Misdraji J, Pitman MB. Adenosquamous carcinoma of the pancreas: cytologic features in 14 cases. Cancer 2003; 99: 372–8. 21. Smoot RL, Zhang L, Sebo TJ, et al. Adenosquamous carcinoma of the pancreas: a single-institution experience comparing resection and palliative care. J Am Coll Surg 2008; 207: 368–70. 22. Grimison PS, Chin MT, Harrison ML, et al. Primary pancreatic lymphoma—pancreatic tumours that are potentially curable without resection, a retrospective review of four cases. BMC Cancer 2006; 6: 117. 23. Arcari A, Anselmi E, Bernuzzi P, et al. Primary pancreatic lymphoma. A report of five cases. Haematologica 2005; 90: e23–6. 24. Lin H, Li SD, Hu XG, et al. Primary pancreatic lymphoma: report of six cases. World J Gastroenterol 2006; 12: 5064–7. 25. Saif MW. Primary pancreatic lymphomas. J Pancreas 2006; 7: 262–3. 26. Basu A, Patil N, Mohindra P, et al. Isolated non-Hodgkin’s lymphoma of the pancreas: case report and review of literature. J Cancer Res Therap 2007; 3: 236–9. 27. Merkle EM, Bender GN, Brambs HJ. Imaging findings in pancreatic lymphoma: differential aspects. AJ R 2000; 174: 671–5. 28. Ji Y, Kuang TT, Tan YS, et al. Pancreatic primary lymphoma: a case report and review of the literature. Hepatobiliary Pancreatic Dis Int 2005; 4: 622–6. 29. Battula N, Srinivasan P, Prachalias A, et al. Primary pancreatic lymphoma: diagnostic and therapeutic dilemma. Pancreas 2006; 33: 192–4. 30. Mortenson MM, Katz MH, Tamm EP, et al. Current diagnosis and management of unusual pancreatic tumors. Am J Surg 2008; 196: 100–13 31. Koniaris LG, Lillemoe KD, Yeo CJ, et al. Is there a role for surgical resection in the treatment of early-stage pancreatic lymphoma? J Am Coll Surg 2000; 190: 319–30 32. Showalter SL, Hager E, Yeo CJ. Metastatic disease to the pancreas and spleen. Semin Oncol 2008; 35: 160–71. 33. Wente MN, Kleeff J, Esposito I, et al. Renal cancer cell metastasis into the pancreas: a single-center experience and overview of the literature. Pancreas 2005; 30: 218–22. 34. Volmar KE, Jones CK, Xie HB. Metastases in the pancreas from nonhematologic neoplasms: report of 20 cases evaluated by fine-needle aspiration. Diagn Cytopathol 2004; 31: 216–20. 35. Sellner F, Tykalsky N, De Santis M, et al. Solitary and multiple isolated metastases of clear cell renal carcinoma to the pancreas: an indication for pancreatic surgery. Ann Surg Oncol 2006; 13: 75–85. 36. Law CH, Wei AC, Hanna SS, et al. Pancreatic resection for metastatic renal cell carcinoma: presentation, treatment, and outcome. Ann Surg Oncol 2003; 10: 922–6. 37. Zerbi A, Ortolano E, Balzano G, et al. Pancreatic metastasis from renal cell carcinoma: which patients benefit from surgical resection? Ann Surg Oncol 2008; 15: 1161–8. 38. Hiotis SP, Klimstra DS, Conlon KC. Results after pancreatic resection for metastatic lesions. Ann Surg Oncol 2002; 9: 675–9. 39. Gutman H, Hess K, Kokotsakis J. Surgery for abdominal metastases of cutaneous melanoma. World J Surg 2001; 25: 750–8. 40. Wood T, DiFranzo L, Rose DM, et al. Does complete resection of melanoma metastatic to solid intra-abdominal organs improve survival? Ann Surg Oncol 2001; 8: 658–62. 41. Kaelin WG. Von Hippel-Lindau disease. Annu Rev Pathol 2007; 2: 145–73. 42. Lonser RR, Glenn GM, Walther M, et al. von Hippel-Lindau disease. Lancet 2003; 361: 2059–67. 43. Shehata BM, Stockwell CA, Castellano-Sanchez AA, et al. Von HippelLindau (VHL) disease: an update on the clinico-pathologic and genetic aspects. Adv Anat Pathol 2008; 15: 165–71. 44. Elli L, Buscarini E, Portugalli V, et al. Pancreatic involvement in von Hippel-Lindau disease: report of two cases and review of the literature. Am J Gastroenterol 2006; 101: 2655–8. 45. Mukhopadhyay B, Sahdev A, Monson JP, et al. Pancreatic lesions in von Hippel-Lindau disease. Clin Endocrinol (Oxf) 2002; 57: 603–8. 46. Corcos O, Couvelard A, Giraud S, et al. Endocrine pancreatic tumors in von Hippel-Lindau disease: clinical, histological,and genetic features. Pancreas 2008; 37: 85–93. 47. Blansfield JA, Choyke L, Morita SY, et al. Clinical, genetic and radiographic analysis of 108 patients with von Hippel-Lindau disease (VHL) manifested by pancreatic neuroendocrine neoplasms(PNETs). Surgery 2007; 142: 814–8. 48. Delman KA, Shapiro SE, Jonasch EW, et al. Abdominal visceral lesions in von Hippel-Lindau disease: incidence and clinical behavior of pancreatic and adrenal lesions at a single center. World J Surg 2006; 30: 665–9. 49. Nai GA, Amico E, Gimenez VR, et al. Osteoclast-like Giant Cell tumors of the pancreas associated with mucous-secreting adenocarcinoma: case report and discussion of the histogenesis. Pancreatology 2005; 5: 279–84. 50. Layfield LJ, Bentz J. Giant-cell containing neoplasm of the pancreas: an aspiration cytology study. Diagn Cytopathol 2007; 36: 238–44. 51. Joo E, Heo T, Park C, et al. A case of osteoclast-like giant cell tumor of the pancreas with ductal adenocarcinoma: Histopatholgical, immunohistochemical, ultrastructural, and molecular biologic basis. J Korean Med Sci 2005; 20: 516–520. 52. Loya AC, Ratnakar KS, Shastry RA. Combined osteoclast giant cell and pleomorphic giant cell tumors of the pancreas: A rarity. An immunohistochemical analysis and review of literature. J Pancreas 2004; 5: 220–4. 53. Sauto-Vial N, Rahili A, Karimdjee-Soihili B, et al. Hepatobiliary and pancreatic: osteoclast-like giant cell tumor of the pancreas. J Gastroenterol Hepatol 2006; 21: 1072. 54. Bauditz J, Rudolph B, Wermke W. Osteoclast-like giant cell tumors of the pancreas and liver. World J Gastroenterol 2006; 12: 7878–83. 55. Hruban RH, Fukushima N. Pancreatic adenocarcinoma: update on the surgical pathology of carcinomas of ductal origin and PanINs. Mod Pathol 2007; 20: 61–70. 56. Wilentz RE, Goggins M, Redston M, et al. Genetic, immunohistochemical, and clinical features of medullary carcinoma of the pancreas. Am J Pathol 2000; 156: 1641–51. 57. Goggins M, Offerhaus GJ, Hilgers W, et al. Pancreatic adenocarcinomas with DNA replication errors (RER+) are associated with wild type K-ras and characteristic histopathology: poor differentiation, a syncytial growth pattern, and pushing borders suggest RER+. Am J Pathol 1998; 152: 1501–7. 58. Jaffee EM, Hruban RH, Canto M, et al. Focus on pancreas cancer. Cancer Cell 2002; 2: 25–8. 59. Nakata B, Wang YQ, Yashiro M, et al. Prognostic value of microsatellite instability in resectable pancreatic cancer. Clin Cancer Res 2002; 8: 2536–40. 60. Yamamoto H, Itoh F, Nakamura H, et al. Genetic and clinical features of human pancreatic ductal adenocarcinoma with widespread microsatellite instability. Cancer Res 2001; 61: 3139–44. 61. Finkelberg DA, Sahani D, Deshpande V, et al. Autoimmune pancreatitis. N Engl J Med 2006; 355: 2670–6. 62. Sarles H, Sarles JC, Muratore R, et al. Chronic inflammatory sclerosis of the pancreas – an autonomous pancreatic disease? Am J Dig Dis 1961; 6: 688–98. 63. Klöppel G, Sipos B, Zamboni G, et al. Autoimmune pancreatitis: histo- and immunopatholgical features. J Gastroenterol 2007; 42: 28–31. 64. Yoshida K, Takeuchi T, Watakanabe S, et al. Chronic pancreatitis caused by autoimmune abnormality: proposal of the concept of autoimmune pancreatitis. Dig Dis Sci 1995; 40: 1561–8. 65. Krasinskas AM, Raina A, Khalid A, et al. Autoimmune pancreatitis. Gastroenterol Clin North Am 2007; 36: 239–57. 66. Rao AS, Palazzo F, Chung J, et al. Diagnostic and treatment modalities for autoimmune pancreatitis. Curr Treat Opt Gastroenterol 2006; 9: 377–84. 67. Okazaki K, Uchida K, Fukui T. Recent advances in autoimmune pancreatitis: concept, diagnosis, and pathogenesis. J Gastroenterol 2008; 43: 409–18. 68. Kawamoto S, Siegelman SS, Hruban et al. Lymphoplasmacytic sclerosing pancreatitis (autoimmune pancreatitis): evaluation with multidetector CT. Radiographics 2008; 28: 157–70. 69. Otsuki M, Chung JB, Okazaki K, et al. Asian diagnostic criteria for autoimmune pancreatitis: concensus of the Japan-Korea symposium on autoimmune pancreatitis. J Gastroenterol 2008, 43: 403–8. 70. Kamiwasa T. Diagnostic criteria for autoimmune pancreatitis. J Clin Gastroenterol 2008; 42: 404–7. 71. Kamisawa T, Imai M, Chen PY, et al. Strategy for differentiating autoimmune pancreatitis from pancreatic cancer. Pancreas 2008; 37: 62–7. 72. Witkiewicz AK, Kennedy EP, Kennyon L, et al. Synchronous autoimmune pancreatitis and infiltrating pancreatic ductal adenocarcinoma. Mod Pathol 2008, 39: 1548–51.

438

47

Acute pancreatitis C. Ross Carter, A. Peter Wysocki, and Colin J. McKay
gallstone or stricture (8) or an alteration in biochemical homeostasis at a cellular level, triggering a rise in intracellular calcium inducing activation of pro-enzymes within the acinar cell, leading to activated trypsin release into the cytosolic compartment (9). While the cellular initiation mechanisms are an area of significant research interest, the management of the acute attack is in general not affected by the etiology, and the importance of identifying an etiological factor lies in the potential for preventing a further episode.

introduction
In the last 20 years, there are few disease processes where the understanding and the approach to management have changed more than that of acute pancreatitis. Acute pancreatitis presents a spectrum of disordered physiology, ranging from a mild and rapidly resolving attack (80%) requiring little more than analgesia and a short period of intravenous fluid resuscitation, to an overwhelming system illness characterized by multi-organ dysfunction refractory to aggressive resuscitation. Various clinical and biochemical scoring systems (1–4) have been used in an attempt to differentiate between mild or severe acute pancreatitis, but out-with the study environment, management is reactive aimed at normalizing altered physiology or the management of local complications. The majority of patients with severe early organ dysfunction will have pancreatic necrosis on CT scan. There is an association between the development of necrosis and the severity of organ dysfunction (5), although patients with edematous pancreatitis may manifest clinical features of a severe attack. Clinical practice has changed rapidly over the last decades, and the previous focus on parenchymal necrosis and particularly the role of prevention/treatment of infection are now considered in a wider concept tending toward organ support and less aggressive intervention. Most patients who develop organ failure have evidence of this at the time of admission or very shortly thereafter (6), and worsening or persistent organ failure is associated with an adverse outcome, and/or the development of late complications. There is no evidence that aggressive early surgical intervention to address the necrosis has a beneficial effect on outcome and indeed is potentially harmful. The majority of patients with acute pancreatitis will have a mild clinical course and other than maintenance of fluid volume and analgesia, subsequent treatment is aimed at prevention of a further attack. Early definitive treatment of cholelithiasis is recommended by cholecystectomy or endoscopic sphincterotomy as recurrent mild attacks are not uncommon (7). The main focus of this chapter is on the 15% to 20% of patients who present with severe disease and the options and rationale for clinical management.

early management of the acute admission
Following a severe acute attack, there is a dynamic process of evolution, in some cases over several months, with changing local and systemic clinical function and morphology, and while the mode of death in almost all cases is multi-organ failure, the pathophysiology and therefore the need for intervention changes with time. The majority of patients, whether mild or severe, present with acute-onset abdominal pain and vomiting. Most patients will settle within 24 to 48 hours. A proportion of patients develop a cytokine-driven systemic inflammatory response (10), which may progress to establish multi-organ dysfunction. The systemic nature of the disease process has been recognized for over 30 years, the “predictive” biochemical scoring systems, dating from the 1970s, reflecting altered organ homeostasis established early organ dysfunction. The magnitude of the physiological disturbance, however, does not affect the principles of management which remains reactive, dealing with maintenance of organ function. In severe acute pancreatitis (SAP), local and systemic inflammatory process, coupled with increases in capillary permeability and third space losses, leads to relative hypovolemia, reduced tissue perfusion, and oxygenation. Initial management therefore involves aggressive fluid resuscitation, monitoring the response through urine output, intra-arterial, intravenous monitoring, and pulse oximetry, but there are no established endpoints of resuscitation to confirm that tissue perfusion and oxygen delivery have been adequately restored. Respiratory and renal supports are often required. Inotropic cardio-vascular support should be delayed until after an adequate circulating volume has been achieved. The pattern and severity of organ dysfunction vary considerably from patient to patient and management can therefore only be optimized on an individual patient basis. In patients often with minimal prior co-morbidity, the development of rapidly progressive or refractory organ failure following initial presentation led to a desire, and often belief, that specific interventions may improve outcome, over- and above-optimized organ support. Unfortunately there are no

etiology
Pancreatic inflammation was first reported in association with alcohol excess in 1878 by Freidrich, and 20 years later, Opie proposed the bile reflux theory potentially underlying the most common cause of gallstones. A detailed description of the pathophysiological mechanisms that are thought to be involved in the initiation of an attack is beyond the scope of this chapter. An attack may be initiated by obstruction of the duct either through passage of a

439

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
measures that have been shown to reduce overall mortality, and local protocols are often biased by personal preference rather than an evidence base. pancreatitis, and indeed this may be harmful. At present we cannot advocate monitoring intra-abdominal pressure out with a clinical trial. Nutrition Median inpatient stay for a patient with MODS-associated SAP is in excess of 2 months. SIRS-associated catabolism, coupled with gastric stasis is common. There is universal consensus that maintenance of nutritional integrity is essential, the main consideration relating to optimum mode of delivery. Gut rest was considered mandatory throughout the 1980s but a total of seven studies, all delivering enteral feed distal to the ligament of Treitz, have now shown a benefit for enteral nutrition, the consistent finding being a reduction in TPN-associated side effects and reduced cost, rather than any reduction in pancreas-specific morbidity (23–26). Three subsequent studies (27–30) have shown proximal feeding into the stomach to be a practical alternative to jejunal feeding with no apparent effect on the pancreatitis. Tolerance of feed therefore governs delivery and our own approach is to allow the patient to eat if tolerated and to use nasogastric, naso-jejunal, or total parenteral nutrition support as required to maintain an adequate intake. A secondary consideration is the potential to alter the disease process through the use of immuno-modulating feeds. Small studies from the United Kingdom (31,32) provided some support for this contention, showing a reduction in the inflammatory response and organ failure in those receiving enteral support, but unfortunately small numbers limited the validity of the conclusions. In the critical care environment (burns, trauma, post-surgical) there have been several trials comparing “immuno-nutrition” with standard enteral feed, with promising results. However, the only study in acute pancreatitis (PROPATRIA trial (33)—The Dutch Acute Pancreatitis Study group), showed probiotics to be associated with increased mortality, and enhanced feeding should at present remain within a study context. Specific Pharmacological Intervention Over the last three decades there have been a number of studies evaluating the concept of supporting the endogenous anti-protease defense mechanisms, the inhibition of pancreatic enzymes, or the inflammatory response. Double-blind randomized trials of iv aprotinin (34) (Trasylol) and gabexate mesilate (35), including a meta-analysis (36), showed no advantage over placebo. Intra-peritoneal aprotinin and the use of both low- and high-dose fresh frozen plasma proved unhelpful. The five randomized trials of octreotide have shown no benefit (37,38), and despite initial promising results (39) with the platelet-activating factor antagonist, Lexipafant, a multi-center randomized controlled trial (recruiting >1500 patients), again showed no advantage over placebo. The potential for other agents which modify the inflammatory response or to influence outcome in acute pancreatitis has only been assessed in experimental models, and there is currently no evidence supporting the use of any specific agent in SAP.

role for specific interventions
ERCP A small group of patients with cholangitis present with associated hyperamylasemia, organ dysfunction being driven by biliary sepsis and the raised amylase is coincidental. This is less common in the Western World than in the Far East. Pyrexia, being a common feature of a delayed Systemic inflammatory response syndrome (SIRS) response in SAP, is very unusual within 48 hours, and when associated with rigors and jaundice, an erroneous diagnosis of pancreatitis should be considered. There have now been three published randomized trials (11–13) and four smaller studies. The most recent meta-analysis (14) has shown early endoscopic retrograde cholangiopancreatography (ERCP) in patients without acute cholangitis, which did not lead to a significant reduction in the risk of overall complications and mortality (Level 1a). Antibiotics Until the early peak in multiple organ dysfunction syndrome (MODS)–associated mortality was recognized, late sepsis due to secondary infection of pancreatic necrosis was thought to be the major cause of death. Infection occurs in 15% to 20% of patients with a minimum 30% pancreatic necrosis. Secondary infection manifests as escalating sepsis or a deterioration in organ failure scores. The early promise of the antibiotic trials of the 1990s (15,16) led to the widespread adoption of the use of prophylactic antibiotics, but with the addition of larger, albeit still underpowered, trials (17,18); the most recent meta-analysis concluded that prophylactic antibiotics do not reduce infected pancreatic necrosis and mortality in patients with SAP (19). In patients with necrosis, SIRS-associated pyrexia is common and may last for weeks in the absence of infection. Persistence of antibiotics beyond a prophylactic time frame carries the danger of the emergence of subsequent resistant or nosocomial infections. At present the decision to use prophylactic antibiotics is based on personal preference rather than evidence base and further well-designed trials are required. Surgical Intervention A small proportion of patients are initially diagnosed at laparotomy and other than washout and closure no pancreatic procedure or drainage is required. Historical attempts at early resection/debridement were associated with prohibitive mortality and the only randomized trial (20) was stopped due to unacceptable mortality in the surgery group. Very rarely, hemorrhage or intestinal ischemia may demand intervention in the first week. Recent interest has focused on the role of intra-abdominal hypertension (IAH) as a contributing factor to organ dysfunction. There are data to suggest that raised intra-abdominal pressure may be associated with disease severity, organ failure, and mortality in SAP (21,22). There are, however, no data to suggest outcome improves following surgical decompression for raised IAP in acute

440

ACUTE PANCREATITIS

summary of early management
The mainstay of early management is the early recognition and the proactive management of compromised organ function. With the exception of the cholangitic patient, there is no role for early surgical, endoscopic, or pharmacological intervention. Nutrition should be by the enteral route where possible. The use of prophylactic antibiotics remains controversial, but when prescribed should be for a time-limited course to prevent the selection of resistant species.

radiological assessment
Trans-abdominal ultrasound is often performed within 24 hours but in the absence of jaundice, this has little effect on clinical management. Bowel gas and a restless patient often compromise the examination and exclusion of cholelithiasis may require a repeated examination in the recovery period. Early post-admission CT may be appropriate where the diagnosis is in doubt or a complication suspected; however, as the evolution of necrosis is not complete until at least 72 hours, in some patients axial scanning is best delayed until this time. Subsequent management and need for further radiological assessment are determined by the clinical condition and the trend of biochemical and organ failure scores.

the pus surrounding the necrosis often resulted in at least a temporary improvement in organ failure challenged this dogma; however, further sepsis was common. The importance of maintaining a sustainable drainage system was recognized in the era of major open debridement leading to the techniques of open packing (42) and closed lavage (43). The solid component within these collections tends to block the lumen of small diameter drains but provided drainage is maintained, organ failure will resolve despite residual necrosis. The mortality that followed an open necrosectomy was common in the initial (72 hours) post-operative period due to overwhelming organ failure. This post-interventional escalation in organ failure is significantly less following all minimally invasive approaches, but often at the cost of multiple interventions and prolonged inpatient stay. A balanced approach is therefore able to utilize a number of techniques dependent on the clinical condition of the patient, the presence of sepsis, the degree of organ failure, the position of the collection, and the maturity of the collection.

surgical intervention in acute pancreatitis
Open Surgery Although minimally invasive approaches have revolutionized management in many centers, on a worldwide basis open debridement remains a keystone of management. Laparotomy/Debridement (Table 47.1) The technique of pancreatic debridement involves a wide exposure of the abdomen, usually through a bilateral subcostal/rooftop or midline incision. The lesser sac is entered via the gastrocolic omentum, or occasionally the transverse mesocolon. Pus is aspirated from the abscess cavity, leaving the solid component behind which is then removed by “blunt finger” dissection. Tissue which will not come away by finger teasing should be left in situ to demarcate and subsequent removal at a further procedure. The procedure may also include a cholecystectomy, operative cholangiogram, and a feeding jejunostomy. Prevention of recurrent sepsis led to several approaches to the management of the debridement cavity. With Drainage/“Closed Packing” Simple drainage, often with multiple retro-peritoneal tube drains, was the conventional approach, with second look laparotomies for recurrent sepsis. This technique has been modified using multiple soft Penrose drains containing cotton gauze to pack the cavity following completion of the necrosectomy, which are subsequently removed at intervals allowing the cavity to collapse around the drains. Laparostomy with Open Packing The lesser sac is packed with lubricated cotton gauze, and the abdomen left open, allowing planned re-explorations every few days until granulation tissue forms. Enteric fistula and secondary hemorrhage are not uncommon, and the technique is rarely performed as a first option. Surgical

management of necrosis
Until recently, sterile necrosis was considered to be a driver of organ dysfunction, leading to a necrosectomy for patients failing to progress after a few weeks. It was also considered essential to identify the development of infection as early as possible, leading to the popularization of protocol-driven radiologically guided fine needle bacterial aspiration (40), and “pre-emptive” necrosectomy following a positive result. Current opinion is that outcome can be improved by the avoidance of early major intervention, especially in patients with organ failure, utilizing minimally invasive approaches to sepsis control where possible. CT-guided fine needle aspiration no longer plays any role in our practice. For many years most specialist centers have addressed all collections utilizing a single surgical technique. This dogmatic procedure-based algorithm failed to address the changing requirements and risk profile with maturation of the collection. The indications for intervention vary with the dynamic evolution of pancreatic/peri-pancreatic collections (41). In the first 4 weeks, the “solid necrotic phase,” demarcation of devitalized tissue is incomplete and an attempt to remove the devitalized tissue often incomplete and associated with bleeding. By about 8 to 10 weeks demarcation results in formation of a walled off fluid collection containing a variable amount of solid necrosis. In the early weeks, achieving control of sepsisdriven organ failure is the primary consideration, whereas following maturation when organ failure is rare and morbidity low, indications include failure to thrive, SIRS, nutritional failure, gastric outlet obstruction, or pain. The presence of proven infection within necrosis (bacteria or gas on CT) was previously seen as a mandatory indication for urgent debridement, as it was believed until recently that recovery would only occur once the necrosis was completely removed. The observation that drainage of

441

442
Mean age in yrs (range) Mortality 6% – 53 (30–81) 64 (56%) 46 (100%) 14/–/– 23 – 13% – 9/–/– 31 41 – Number with infected necrosis (%) Mean AII/RS/ CT-SI Timing of intervention from onset (mean days) Mean postoperative length of stay (days) ITU preoperatively Morbidity Reoperations 17% ITU postoperatively 64 46 – – – 45% (median 6 days) – 29 57 (28–87) 29 (100%) 255 (75%) 16/–/– 12 (1–31) 55a 13/4/– 22 85 24% – 44% 26% – 53 (16–85) 39% 90% preoperative organ failure – 79% (mean 2.2) – 58b (19–98) 140 (100%) 11/5/– 20 64 27% mean 3 days 78% 51% (median 1) Mean 27 days

Table 47.1 Major Series of Laparotomy and Necrosectomy

Unit, year

Method

Number of patients in series

Warshaw, 1998 (44)

Bradley, 1999 (45)

Buchler, 2000 (46)

Götzinger, 2002 (47)

340

Beger, 2005 (43)

Necrosectomy with closed packing Necrosectomy with scheduled re-explorations Necrosectomy with closed lavage Necrosectomy with scheduled re-explorations or resection Necrosectomy with closed lavage

140

a

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

b

Survivors. Median. Abbreviations: AII, APACHE II score; RS, Ranson score; CT-SI, Balthazar CT severity index; ITU, intensive therapy unit.

ACUTE PANCREATITIS
packing and planned re-operation are, however, sometimes required to control blood loss from the retroperitoneum following the development of an intraoperative coagulopathy, a lavage system being created, following correction of the coagulopathy, at the time of interval pack removal. With Closed Lavage The Ulm group popularized the use of post-operative closed lavage and this remains the most popular method for postoperative sepsis control following open debridement. Several (4–6) large diameter tube drains are inserted in the lesser sac and throughout the abdomen and the abdomen closed. Continuous lavage is then commenced, the aim being the continuous removal of devitalized necrotic material and bacteria. The lavage is continued, for around 3 to 4 weeks on average, until the return fluid is clear, and the patient has no residual signs of systemic sepsis. Minimally Invasive Approaches A number of minimally invasive techniques have been developed over the last 15 years and these are often viewed as complimentary rather than exclusive. These will be initially described and then the potential advantages/disadvantages discussed. Percutaneous Catheter Drainage (Table 47.2) Interventional radiological drainage of abscesses has been commonplace for many years. The major difficulty in the acute pancreatitis patient is the tendency for these to block, leading to recurrent sepsis. Utilizing simple drainage as the primary modality of treatment is possible, but is extremely labor intensive and delayed surgical intervention is commonly required. Endoscopic Drainage (Table 47.3) Transmural drainage of lesser sac collections was initially performed for established pseudocyst. Baron first described extending the role into the management of pancreatic abscess, where the collections contained some solid component. Endoscopic ultrasound–guided drainage can increase the technical success rate and reduce the risk of bleeding. Small diameter stent drainage could lead to incomplete drainage, and the use of tract dilatation, multiple catheters, and intracystic lavage became commonplace. Seifert described the performance of a necrosectomy through the endoscopic cystgastrostomy and this is becoming increasingly popular. Potential advantages include that it can be performed without a general anesthetic, potentially performed as a day patient in suitable patients and the lack of an external drain, but inadequate drainage and hemorrhage are potential hazards. Percutaneous Necrosectomy (Table 47.4) This combines minimally invasive drainage with closed postoperative lavage. A CT-guided radiological drain in inserted ideally in the left flank to promote gravitational drainage. The drainage tract is dilated often releasing pus under pressure and the cavity explored using a rigid urological endoscope. Loose necrotic material can be removed in a piecemeal fashion as it is this that tends to block the drains. A wide bore 32FG drain (with a parallel catheter for post-operative lavage) is left within the cavity. Continuous lavage is maintained and further procedures performed when incomplete drainage is suspected. This is probably the easiest way of maintaining adequate drainage and sepsis control but results in a prolonged hospital stay and commonly late pancreatic fistula. In the 1980s Fagniez described an open minimally invasive approach utilizing a left flank incision and blind retroperitoneal debridement (Table 47.5). In 2001, Horvath modified this technique using video-assistance to allow direct visualization of the debridement through a 8 to 10 cm left flank incision. This minimally invasive technique is currently being assessed in the first randomized comparison of a minimally invasive approach against open surgery (PANTER trial, Dutch Acute Pancreatitis Study Group).

laparoscopy
Early attempts at laparoscopy attempted to mimic the open debridement, but proved technically challenging and has been superseded by the other surgical approaches. Laparoscopic cystgastrostomy has been reported for the management of late walled off necrosis. Initial descriptions involved intra-luminal laparoscopy but the position of the collection relative to the stomach is critical. Modification by performing a longitudinal anterior gastrostomy and then addressing the posterior cystgastrostomy greatly simplified the procedure and allows a onestep approach to organized necrosis. The procedure requires a well-formed cavity and complete separation of the necrotic tissue, so is less appropriate for collections in an early stage of the disease.

summary
The choice of primary procedure is determined by the relative importance of sepsis control and the stage of evolution of the necrosis-associated peri-pancreatic collection. Optimal sepsis control is obtained by percutaneous necrosectomy but completion of the process takes some time and requires multiple procedures. At the other end of the spectrum, where sepsis is not an issue but intervention demanded through pressure effects or failure to thrive, a single procedure laparoscopic cystgastrostomy expedites completion of treatment, particularly where a predominant solid component makes endoscopic clearance likely to require multiple procedures. Endoscopic cystgastrostomy has a role between these extremes. In the septic patient, initial endoscopic drainage is easy but maintenance of sepsis control often demands repeated intervention and adjuvant radiological drainage. A particular problem is that the first indication of a blocked internal cystgastrostomy drainage system is a clinical escalation of sepsis whereas a blocked external drain is easily recognized and addressed. Similarly large or para-colic collections are unsuitable. However, in the fluid-predominant

443

444
Number with infected necrosis managed percutaneously Technique 25–152 3.3 Number of deaths in series 34 –/–/8 9 (1–48) 16 Mean timing Mean AII/ of intervention from RS/CT-SI for patients onset days in series (range) Successful percutaneous management alone of those with infected necrosis 17 Average duration of drainage (days or range) Acute operative management Average catheter of those with infected exchanges per necrosis patient (range) 26 19 –a 19 –a –/5/E 23 20 Average of 3 4 transperitoneal catheters per patient (10–28 Fr) 8–16 Fr 5 43 14–56 –a 0.4b 6 6 15b 12 (2–31) 10 (1–58) 15 3 23 –/4/8 –a 2 Average of 1.4 drains per patient (10–12 Fr) –a 2 –a –a 0.1 –a 12 3 80 18/–/6c >11c 42 37 (1–260)c 2 (1–9)c 20
b

Table 47.2 Percutaneous Drainage of Infected necrosis

Author, year

Number of patients in series

Mean age in years (range)

Freeney, 1998 (48)

34

56 (31–71)

32

53

Gouzi, 1999 (49) Baril, 2000 (50)

25

40 (17–68)

Oláh, 2006 (51) Lee, 2007 (52)

15

59 (36–78)

31

49

Bruennler, 2008 (53)

80

57 (17–79)c

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Szentkereszty, 2008 (54)

61

44 (25–87)d

23/4/–d

>4

7d

1 14 Fr catheter dilated to 20 Fr combined with irrigation 27 Median of (18/60 mantwo aged without catheters surgery) per patient (8–24 Fr) with active necrosectomy in 18 patients b –a (10/61)d

–a

–a

15d

a

b

Data not stated. Limited/unclear data published. Median. d For entire reported series of patients with various diagnoses/treatments. Abbreviations: AII, APACHE II score; RS, Ranson score; CT-SI, Balthazar CT severity index; PA, pancreatic abscess; PC, post-operative collection.

c

Table 47.3 Endoscopic Necrosectomy
Mean AII/RS/ CT-SI for patients in series Timing of intervention from onset (mean days) Technique 7 Fr, 10 Fr stents One or more 7-10 Fr stents 7 Fr nasopancreatic catheter Transpapillary stenting 10 Fr stent 7 Fr nasocyst catheter 7 Transpapillary stenting 10 Fr stent 7 Fr nasocyst catheter 9 6 (2–12) 4 10 Fr stent 10 Fr strent
b

Author, year 55 (37–55) 13–28 8 3 62 (18–68) 9 12–/–/9 42 1 (1 SN, 1 RC) –a

Number of patients in series 0 0

Median age in years (range)

Number with infected necrosis

Patients in series cured by endoscopic treatment alone Number of deaths in series

Number of endoscopic reinterventions (average) –a 0.3

Post-procedure length of stay (days) 24–96 17

Seifert, 2000 (55)

3

Park, 2002 (56)

9

ACUTE PANCREATITIS

Seewald, 2005 (57)

5

63 (38–84)

5

–a 4

–a

0

28

–a

Seewald, 2005 (57)

8

56 (38–80)

0 (8 PA)

–a

–a

0

7

–a

Charnley, 2006 (58) 49 5 –/4/8

13

53 (30–64)

11 (2 SN)

8/–/6

?24

15% 0

3 –
a

–a –a

Lee, 2007 (52)

5

a

b

Data not stated. Limited/unclear data published. Abbreviations: AII, APACHE II score; RS, Ranson score; CT-SI, Balthazar CT severity index; OPN, organized pancreatic necrosis; PA, pancreatic abscess; SN, sterile necrosis; RC, residual collection after endoscopic treatment.

445

446
Number with infected necrosis –a 24 23 28 –a 6 1 0 21(19%) 1
d

Table 47.4 Percutaneous Necrosectomy
Patients in series cured by retroperitoneal necrosectomy Without laparotomy Median ITU Number post-operative preoperaof deaths length of stay tively in series (days) (number) 2 0 9 64 84c >50 –a 42 3 0 16 –a 1 38% 9 6 28 1 1 43 2 –a

Author, yr

Number of patients in series

Median age in years (range)

Mean AII/RS/ CT-SI for patients in series

Timing of intervention from onset (mean days)

Early morbidity

Median number of reoperations per patient (range) 2.5 (1–4) 2.3 (1–4) –a 3 (2–8) 8 3(1–6) 4 0

ITU post-operatively (number)

10 –/10/E 9/–/9 –a –a –a 30 (11–91)

45 (32–74) 10

6

46b (23–78) 6

47

56 (18–85) 38 (9 SN)

Mean of 0 days Mean of 11 days
d

9 1 107

58 (41–80) 7 (2 PA)

1

51

Carter, 2000 (33) Risse, 2004 (59) Connor, 2005 (60) Mui, 2005 (61) Shelat, 2007 (62) Carter2007

110

54(18–89)

55%

a

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

Data not stated. Mean. c Total hospital stay. d Limited/unclear data published. Abbreviations: AII, APACHE II score; RS, Ranson score; CT-SI, Balthazar CT severity index; SN, Sterile necrosis; PA, Pancreatic abscess.

b

Table 47.5 Retroperitoneal Laparostomy
Number with infected necrosis 18 (17 SN) –/4a/E 40 13 70 14a Mean AII/RS/ CT-SI Timing of intervention from onset (mean days) Patients in series cured by lumbotomy without laparotomy Number of deaths in series Mean length of stay (days) Technique Morbidity Mean number of reoperations per patient

Author, year

Number of patients in series

Median age in years (range)

Fagniez, 1989 (63)

40

46 (26–80)

25

2.6

Gambiez, 1998 (64) 6 –/–/7 41 (27–77) 4 0 –

20

51 (–)

13 (7 SN)

–/3/E



13 (7 with IPN)

2

62

6

4

ACUTE PANCREATITIS

Horvath, 2001 (65)

6

36 (16–48)

3

2 percutaneous drainages

18 14 (1 SN) 9/–/8 41 (15–164) 11

53 (29–63)

18

–/–/8

48 (0–181)



2 1

100 (43–240) 100 (45–240)

– 6

2 (1–11) 1

Besselink, 2006 (66) van Santvoort, 2007 (67)

15

52 (34–66)

10–15 cm left upper quadrant incision anterior to 12th rib including mobilization of descending colon 6 cm incision centered on 12th rib debridement, 23 cm mediastinoscope Percutaneous drainage, debridement through 4–5 cm flank incision, retroperitoneoscopy via ports placed through lumbotomy incision, CO2 insufflation Various left-sided approaches with videoscopic assistance 5 cm subcostal incision along percutaneous drain, initial blind debridement followed by videoscopic debridement

a For entire reported series of patients with various diagnoses/treatments. – data not stated, ? limited/unclear data published Abbreviations: AII, APACHE II score; RS, Ranson score; CT-SI, Balthazar CT severity index; ITU, intensive therapy unit; ASIS, anterior superior iliac spine; IPN, infected pancreatic necrosis; SN, sterile necrosis.

447

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C)

(D)

Figure 47.1 (A) Early phase CT (48hrs) with head necrosis and peripancreatic edema. (B) CT at 6 weeks confirming extensive loss of parenchyma with early demarcation and no evidence of infection. (C) CT at 7 weeks with extensive gas indicating infection—clinical sepsis addressed by percutaneous necrosectomy x3. (D) CT at 12 weeks prior to discharge.

There are no comparative data between techniques, and our multimodal approach has evolved largely through experience, they should be seen as complimentary, and often combined in a single patient. There is undoubtedly a move away from open surgery, particularly in the septic patient. The complexity and need for a spectrum of interventional options demand that these patients are cared for within an organized multidisciplinary regional team network.

references
1. Ranson JH, Rifkind KM, Roses DF, et al. Objective early identification of severe acute pancreatitis. Am J Gastroenterol 1974; 61(6): 443–51. 2. Blamey SL, Imrie CW, O’Neill J. Prognostic factors in acute pancreatitis. Gut 1984; 25(12), 1340–6. 3. Larvin M, McMahon MJ. APACHE-II score for assessment and monitoring of acute pancreatitis. Lancet 1989; 2(8656): 201–5. 4. Marshall JC, Cook DJ, Christou NV, et al. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995; 23(10): 1638–52. 5. Balthazar EJ. Acute pancreatitis: assessment of severity with clinical and CT evaluation. Radiology 2002; 223(3): 603–13. 6. Buter A, Imrie CW, Carter CR, Evans S, McKay CJ. Dynamic nature of early organ dysfunction determines outcome in acute pancreatitis. Br J Surg 2002; 89(3): 298–302.

Figure 47.2

well-organized collection without organ failure, endoscopic cystgastrostomy, with dilatation and limited necrosectomy is potentially the procedure of choice.

448

ACUTE PANCREATITIS
7. Johnson CD, et al. UK guidelines for the management of acute pancreatitis. Gut 2005 May; 54 (Suppl 3): iii1–9. 8. Acosta JM, Ledesma CL. Gallstone migration as a cause of acute pancreatitis. N Engl J Med 1974; 290(9): 484–7. 9. Raraty MGT, Petersen OH, Sutton R, Neoptolemos JP. Intracellular free ionized calcium in the pathogenesis of acute pancreatitis. Baillieres Best Pract Clin Gastroenterol 1999; 13(2): 241–51. 10. Imrie CW, McKay CJ. The scientific basis of medical therapy of acute pancreatitis: could it work, and is there a role for lexipafant? Gastroenterol Clin North Am 1999; 28(3): 591–9. 11. Neoptolemos JP, London NJ, James D, et al. Controlled trial of urgent endoscopic retrograde cholangiopancreatography and endoscopic sphincterotomy versus conservative treatment for acute pancreatitis due to gallstones. Lancet 1988; 2(8618): 979–83. 12. Fan S-T, Lai ECS, Mok FPT, et al. Early treatment of acute biliary pancreatitis by endoscopic papillotomy. New Engl J Med 1993; 328(4): 228–32. 13. Folsch UR, Nitsche R, Ludtke R, Hilgers RA, Creutzfeldt W. Early ERCP and papillotomy compared with conservative treatment for acute biliary pancreatitis. New Engl J Med 1997; 336(4): 237–42. 14. Petrov MS, van Santvoort HC, Besselink MG, et al. Early endoscopic retrograde cholangiopancreatography versus conservative management in acute biliary pancreatitis without cholangitis: a meta-analysis of randomized trials. Ann Surg 2008; 247(2): 250–7. 15. Pederzoli P, Bassi C, Vesentini S, Campedelli A. A randomized multicenter clinical trial of antibiotic prophylaxis of septic complications in acute necrotizing pancreatitis with imipenem. Surg Gynecol Obstet 1993; 176(5): 480–3. 16. Sainio V, Kemppainen E, Puolakkainen P, et al. Early antibiotic treatment in acute necrotising pancreatitis. Lancet 1995; 346: 663–7. 17. Dellinger EP, Tellado, J.M., Soto N.E., et al Early antibiotic treatment for severe acute necrotising pancreatitis: randomised double blind placebo controlled study. Ann Surg 2007; 245(5), 684–5. 18. Isenmann R, Runzi M, Kron M, Kahl S, Kraus D, Jung N, et al. Prophylactic antibiotic treatment in patients with predicted severe acute pancreatitis: a placebo-controlled, double-blind trial. Gastroenterology 2004; 126(4): 997–1004. 19. Bai Y, Gao J, Zou DW, Li ZS. Prophylactic antibiotics cannot reduce infected pancreatic necrosis and mortality in acute necrotizing pancreatitis: evidence from a meta-analysis of randomized controlled trials. Am J Gastroenterol 2008; 103(1): 104–10. 20. Meir J., Luque-de Leon E, Castillo A, Robledo F, Blanco R. Early versus late necrosectomy in severe necrotising pancreatitis. Am J Surg 1997; 173: 71–5. 21. Al Bahrani AZ, Abid GH, Holt A, McCloy RF, Benson J, Eddleston J et al. Clinical relevance of intra-abdominal hypertension in patients with severe acute pancreatitis. Pancreas 2008; 36(1): 39–43. 22. Zhang WF, Ni YL, Cai L, et al. Intra-abdominal pressure monitoring in predicting outcome of patients with severe acute pancreatitis. Hepatobiliary Pancreat Dis Int 2007; 6(4): 420–3. 23. Kalfarentzos F, Kehagias J, Mead N, Kokkinis K, Gogos, CA. Enteral nutrition is superior to parenteral nutrition in severe acute pancreatitis: results of a randomized prospective trial. Br J Surg 1997; 84(12): 1665–9. 24. McClave SA, Greene LM, Snider HL, et al. Comparison of the safety of early enteral vs parenteral nutrition in mild acute pancreatitis. JPEN: J Parenter Enteral Nutr 1997; 21(1): 14–20. 25. Nakad A, Piessevaux H, Marot J-C, et al. Is early enteral nutrition in acute pancreatitis dangerous? About 20 patients fed by an endoscopically placed nasogastrojejunal tube. Pancreas 1998; 17(2): 187–93. 26. Abou-Assi S, Craig K, O’Keefe SJ. Hypocaloric jejunal feeding is better than total parenteral nutrition in acute pancreatitis: results of a randomized comparative study. Am J Gastroenterol 2002; 97: 2255–62. 27. Eatock FC, Chong P, Menezes N, et al. A randomized study of early nasogastric versus nasojejunal feeding in severe acute pancreatitis. Am J Gastroenterol 2005; 100(2): 432–9. 28. Eckerwall GE, Axelsson JB, Andersson RG. Early nasogastric feeding in predicted severe acute pancreatitis: a clinical, randomized study. Ann Surg 2006; 244(6): 959–65. 29. Kumar A, Singh N, Prakash S, Saraya A, Joshi YK. Early enteral nutrition in severe acute pancreatitis: a prospective randomized controlled trial comparing nasojejunal and nasogastric routes. J Clin Gastroenterol 2006; 40(5): 431–4. Windsor ACJ, Li A, Barnes E, et al. Feeding the gut in acute pancreatitis: a randomised trial of enteral vs parenteral nutrition. Br J Surg 1996; 83(Suppl. 1): 31. Windsor ACJ, Kanwar S, Li AGK, et al. Compared with parenteral nutrition, enteral feeding attenuates the acute phase response and improves disease severity in acute pancreatitis. Gut 1998; 42(3): 431–5. Pearce CB, Sadek SA, Walters AM, et al. A double-blind, randomised, controlled trial to study the effects of an enteral feed supplemented with glutamine, arginine, and omega-3 fatty acid in predicted acute severe pancreatitis. JOP 2006; 7(4): 361–71. Besselink MG, van Santvoort HC, Buskens E, et al. Probiotic prophylaxis in predicted severe acute pancreatitis: a randomised, double-blind, placebo-controlled trial. Lancet 2008; 371(9613): 651–9. Imrie CW, McKay AJ, Neill JO. Short duration megadosage IV Trasylol in primary acute pancreatitis – a double-blind trial. Gut 1980; 21(5): 431–5. Pederzoli P, Cavallini G, Falconi M, Bassi C. Gabexate mesilate vs aprotinin in human acute pancreatitis (GA.ME.P.A.): a prospective, randomized, double-blind multicenter study. Int J Pancreatol 1993; 14(2): 117–24. Andriulli A, Leandro G, Clemente R, et al. Meta-analysis of somatostatin, octreotide and gabexate mesilate in the therapy of acute pancreatitis. Aliment Pharmacol Ther 1998; 12(3): 237–45. McKay C, Baxter J, Imrie C. A randomized, controlled trial of octreotide in the management of patients with acute pancreatitis. Int J Pancreatol 1997; 21(1): 13–9. Uhl W, Buchler MW, Malfertheiner P, et al. A randomised double blind multicentre trial of octreotide in moderate to severe acute pancreatitis. Gut 1999; 45: 97–104. McKay CJ, Curran F, Sharples C, Baxter JN, Imrie CW. Prospective placebo-controlled randomized trial of lexipafant in predicted severe acute pancreatitis. British J Surg 1997; 84(9): 1239–43. Rau B, Pralle U, Mayer JM, Beger HG. Role of ultrasonographically guided fine-needle aspiration cytology in the diagnosis of infected pancreatic necrosis. British J Surg 1998; 85(2): 179–84. Carter R. Percutaneous management of necrotizing pancreatitis. HPB (Oxford) 2007; 9(3): 235–9. Bradley EL III. A fifteen year experience with open drainage for infected pancreatic necrosis. Surg Gynaecol Obstet 1993; 177: 215–22. Rau B, Bothe A, Beger HG. Surgical treatment of necrotizing pancreatitis by necrosectomy and closed lavage: changing patient characteristics and outcome in a 19-year, single-center series. Surgery 2005; 138(1): 28–39. Fernandez-Del CC, Rattner DW, Makary MA, et al. Debridement and closed packing for the treatment of necrotizing pancreatitis. Ann Surg 1998; 228(5): 676–84. Bradley EL. Operative vs. Nonoperative therapy in necrotizing pancreatitis. Digestion 1999; 60(Suppl 1): 19–21. Buchler MW, Gloor B, Muller CA, et al. Acute necrotizing pancreatitis: treatment strategy according to the status of infection. Ann Surg 2000; 232(5): 619–26. Gotzinger P, Sautner T, Kriwanek S, et al. Surgical treatment for severe acute pancreatitis: extent and surgical control of necrosis determine outcome. World J Surg 2002; 26(4): 474–8. Freeny PC, Hauptmann E, Althaus SJ, Traverso LW, Sinanan. Percutaneous CT-guided catheter drainage of infected acute necrotizing pancreatitis: Techniques and results. AJR 1998; 170(4): 969–75. Gouzi JL, Bloom E, Julio C, et al. Percutaneous drainage of infected pancreatic necrosis: an alternative to surgery. Chirurgie 1999; 124(1): 31–7. Baril NB, Ralls PW, Wren SM, et al. Does an infected peripancreatic fluid collection or abscess mandate operation? Ann Surg 2000; 231(3): 361–7. Olah A, Belagyi T, Bartek P, Poharnok L, Romics L Jr. Alternative treatment modalities of infected pancreatic necrosis. Hepatogastroenterology 2006; 53(70): 603–7. Lee JK, Kwak KK, Park JK, et al. The efficacy of nonsurgical treatment of infected pancreatic necrosis. Pancreas 2007; 34(4): 399–404. Bruennler T, Langgartner J, Lang S, et al. Outcome of patients with acute, necrotizing pancreatitis requiring drainage-does drainage size matter? World J Gastroenterol 2008; 14(5): 725–30.

30.

31.

32.

33.

34. 35.

36.

37.

38.

39.

40.

41. 42. 43.

44.

45. 46.

47.

48.

49. 50. 51.

52. 53.

449

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
54. Szentkereszty Z, Kotan R, Posan J, Arkossy P, Sapy P. Therapeutic tactics in the treatment of acute necrotizing pancreatitis. Hepatogastroenterology 2008; 55(81): 266–9. 55. Seifert H, Faust D, Schmitt T, et al. Transmural drainage of cystic peripancreatic lesions with a new large-channel echo endoscope. Endoscopy 2001 Dec; 33(12): 1022–6. 56. Park JJ, Kim SS, Koo YS, et al. Definitive treatment of pancreatic abscess by endoscopic transmural drainage. Gastrointest Endosc 2002; 55(2): 256–62. 57. Seewald S, Groth S, Omar S, et al. Aggressive endoscopic therapy for pancreatic necrosis and pancreatic abscess: a new safe and effective treatment algorithm. Gastrointest Endosc 2005; 62(1): 92–100. 58. Charnley RM, Lochan R, Gray H,et al. Endoscopic necrosectomy as primary therapy in the management of infected pancreatic necrosis. Endoscopy 2006; 38(9): 925–8. 59. Risse O, Auguste T, Delannoy P, et al. Percutaneous video-assisted necrosectomy for infected pancreatic necrosis. Gastroenterol Clin Biol 2004; 28(10 Pt 1): 868–71. 60. Connor S, Alexakis N, Raraty MG, et al. Early and late complications after pancreatic necrosectomy. Surgery 2005; 137(5): 499–505. 61. Mui LM, Wong SK, Ng EK, Chan AC, Chung SC. Combined sinus tract endoscopy and endoscopic retrograde cholangiopancreatography in management of pancreatic necrosis and abscess. Surg Endosc 2005; 19(3): 393–7. 62. Shelat VG, Diddapur RK. Minimally invasive retroperitoneal pancreatic necrosectomy in necrotising pancreatitis. Singapore Med J 2007; 48(8): e220–3. 63. Fagniez PL, Rotman N, Kracht M Direct retroperitoneal approach to necrosis in severe acute pancreatitis. Br J Surg 1989; 76: 264–7. 64. Gambiez LP, Denimal FA, Porte HL, et al. Retroperitoneal approach and endoscopic management of peripancreatic necrosis collections. Arch Surg 1998; 133(1): 66–72. 65. Horvath KD, Kao LS, Ali A, et al. Laparoscopic assisted percutaneous drainage of infected pancreatic necrosis. Surg Endosc 2001; 15(7): 677–82. 66. Besselink MG, de Bruijn MT, Rutten JP, et al. Surgical intervention in patients with necrotizing pancreatitis. Br J Surg 2006; 93(5): 593–9. 67. van Santvoort HC, Besselink MG, Horvath KD, et al. Videoscopic assisted retroperitoneal debridement in infected necrotizing pancreatitis. HPB (Oxford) 2007; 9(2): 156–9.

450

48 Chronic pancreatitis
introduction

Jakob R. Izbicki, OliverMann, Asad Kutup, and Kai A. Bachmann
ongoing inflammatory disease characterized by irreversible structural changes associated with abdominal pain and permanent loss of function. In the Marseille–Rome classification of 1988, obstructive CP, chronic inflammatory pancreatitis (with loss of exocrine parenchyma and replacement by fibrosis), and the chronic calcifying pancreatitis were described. Recently a new classification of CP has been suggested: probable CP is characterized by a typical history and one or more of the following criteria: recurrent or persistent pseudocysts, ductal alterations, endocrine insufficiency (abnormal glucose tolerance test), or pathological secretin test. Definite CP is characterized by typical history and at least one of the following criteria: typical histology of an adequate surgical specimen, moderate or marked ductal alterations, pancreatic calcification, marked exocrine insufficiency defined as steatorrhea, normalized or markedly reduced enzyme substitution (14).

Chronic pancreatitis (CP) is a common disorder of the gastrointestinal tract with enormous social and personal impact. The prevalence of CP is 10 to 30 per 100,000 population and it affects about eight new patients per 100,000 population per year in the United States (1–3). Autopsy series, however, suggest a higher prevalence of 0.04% to 5%. CP is an inflammatory disease characterized by the progressive conversion of pancreatic parenchyma to fibrous tissue. The most frequent causes are excessive alcohol consumption, cholelithiasis, autoimmune or individual genetic predisposition, and anatomic variants such as pancreas divisum. In up to 20% of the patients the reasons or predisposing factors are not identifiable. The peak of presentation of the disease occurs in patients between 35 and 55 years of age (4). Abdominal pain is the main symptom of CP leading to inability to work, early retirement, and addiction to analgesics. Severe pain attacks are the leading causes for hospitalization. The natural course is characterized by a consecutive loss of pancreatic parenchyma by fibrosis leading to exocrine insufficiency with maldigestion and finally in advanced stages endocrine insufficiency. The clinical course and histological morphological changes that characterize the disease are extremely variable. Overall the life expectancy is shortened by 10 to 20 years. The mortality is increased 3.6-fold, as compared with a population without CP (2). The annual treatment costs are approximately $17,000 per patient (2). Due to improvements in the treatment 20% to 50% of the patients live more than 20 years with chronic inflammation of the pancreas (5,6). Besides pain, exocrine and endocrine malfunction, mechanical complications occur such as stricture of the bile duct, duodenal stenosis (7), or the formation of pancreatic pseudocysts. The process of continuing organ destruction cannot be interrupted by abstinence from alcohol consumption, which seems to be the causative agent in most cases. Despite thousands of reports that have been published in the last decades dealing with this disease, pathogenesis and pathophysiology of CP are poorly understood and the clinical course is unpredictable. Therefore adequate treatment of CP and its complications remain a major challenge (8,9).

natural course of cp
In the past the main rationale for conservative approaches derives from the assumption, that most patients with longstanding CP will become pain free due to a progressive “burning out” of the organ (15). Recently it was shown that the natural course of CP is characterized by progressive loss of pancreatic function by fibrosis of the parenchyma with consecutive endocrine and exocrine insufficiency supplementary to pain (12,16,17). After an initial period without noticeable pain, the disease progresses into the next stage characterized by pain and exocrine, later endocrine insufficiency. In the third stage the burn-out pancreas with global insufficiency is found, and pain might subside. The pancreatic parenchyma is irreversibly converted to fibrous tissue with diabetes and steatorrhea (18). At time of onset of CP 8% of the patient had a at least moderate endocrine insufficiency, whereas in long-term follow-up approximately 80% had endocrine insufficiency (19,20). Studies revealed that it takes 10 to 20 years of a progressive inflammatory process to lead to exocrine insufficiency by destroying the pancreatic parenchyma (21). Ten years after onset of CP, 50% to 93% of the patients with CP were still suffering from abdominal pain (17,20–22). At least 50% to 68% of the patients with CP need surgery for management of complications or for intractable pain (5,23). Reduction of alcohol intake did not influence the course of pain in chronic alcoholic pancreatitis, but continued alcohol abuse was associated with significantly lower survival rates (24,25). Patients that quit drinking showed improvement in exocrine function (20,24). Endocrine insufficiency did not alter the course of pain. For the individual patient the course of the disease is unpredictable (20,25).

definition of cp
The classification of CP as a separate disease was described in 1946 by Comfort et al. (10). Before, the term CP has been used for a variety of pancreatic diseases without a generally accepted definition (11,12). Since then, different classifications of CP have been presented. According to the Marseille Classification, CP is characterized by histological changes persisting after the etiologic agent has been removed (13). The Cambridge Classification (1983) defined CP as an

451

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

etiology pathomorphological findings in cp
Long-term alcohol intake is associated with an increased risk of developing CP. High caloric intake of protein and fat, smoking, and lack of vitamins and trace elements have been described as additional, predisposing risk factors (12). Ammann and colleagues suggested that acute pancreatitis and chronic alcoholic pancreatitis are different stages of the same disease (17,24). CP represents the remaining damage after episodes of severe acute pancreatitis (26,27). Alcohol consumption is the leading cause of CP in western countries (70–90%) (2,4,11,14). The acinar cells are directly damaged by alcohol. A change in microcirculatory perfusion and alterations of the epithelial permeability lead to an imbalance in the pancreatic juice or decreased fluids or bicarbonate secretion. Parenchymal necrosis with hemorrhage of the pancreas may induce perilobular fibrosis that leads to intralobular fibrosis, ductal obstruction, and periductal inflammation. Altered amounts of lithostatin in the pancreatic juice can lead to formation of the protein plugs and stones in ducts and ductuli (11,27,28). Pathomorphological findings in CP such as inflammatory infiltration of the pancreatic tissue, fibrosis, atrophy of the acinar cells, calcifications, pancreatic duct stricture, and pseudocysts can be found isolated, segmental, or diffuse throughout the whole organ (11,12,27,28). Histomorphologically different forms of CP can be distinguished. The most common form, the calcifying CP is characterized by recurrent bouts of clinically acute pancreatitis with abdominal pain and development of intraductal calculi, protein plugs, and parenchyma calcifications (2,12). Radiographically (ERCP) these impose as chain of lakes. These alterations of various degrees in different stages of the disease lead to pancreatic duct stenosis and consecutively to prestenotic dilatations. Additionally epithelial alterations, inflammatory periductal infiltrations, parenchymal atrophy, necrosis, and fibrosis can be found (12,18,26,27,29,30). Obstructive CP is often painless and caused by blockage of the main pancreatic duct due to tumor or an inflammatory process (post-acute pancreatitis) that leads to an atrophy of the pancreatic tissue and a prestenotic dilatation. No alteration of the ductal epithelium is found (12,18,26,30). Pancreatic duct stones are uncommon. Periductal fibrosis and inflammatory infiltration are mainly found around the larger ducts and in the pancreatic head. Diffuse fibrotic changes occur throughout the organ without lobular topography. Pancreatic main duct stenosis may be caused by papillary stenosis (tumor) or inflammation, duodenal diverticula, pancreatic tumors, congenital or acquired duct abnormalities (pancreas divisum), or rarely by traumatic pancreatic duct injuries. Small duct pancreatitis is an extremely rare form of CP that is defined as a throughout fibrous, inflammatory tissue with a main duct diameter of ≤ 3 mm, (31) . The autoimmune pancreatitis is characterized by the absence of typical risk factors for developing CP or hereditary factors. In the past this subtype was named primary inflammatory sclerosis of the pancreas, non-alcoholic duct destructive pancreatitis, or lymphoplasmacytic sclerosing pancreatitis (17,18,32). The term autoimmune pancreatitis was introduced by Yoshida

and colleagues in 1995 (33). Autoimmune pancreatitis can present with focal event or with multiple lesions. Pseudocysts and calculi are rarely found. Four histological features are characteristic for autoimmune pancreatitis. Lymphoplasmacytic infiltration, consisting of lymphocytes and plasma cells (often with high level of IgG4), macrophages, neutrophils, and eosinophils, result in an intestinal fibrosis (34). Additionally periductal inflammation and periphlebitis can lead to luminal strictures or obliterative venulitis, respectively. Obstructive jaundice is caused by an affection of the common bile duct (CBD) which may extend to the gallbladder and biliary tree. An increased level of IgG4 is a sensitive marker (35). Autoimmune pancreatitis is associated with other autoimmune disorders such as ulcerative colitis, Cohn’s disease, primary sclerosing cholangitis, Sjörgren’s syndrome, lymphocytic thyroiditis, and primary biliary cirrhosis (36). Hereditary CP is a rare form with an incidence of approximate 3.5 to 10 per 100,000 inhabitants (3,37). The morphologic findings in HCP are irregular sclerosis with focal, segmental, or diffuse destruction of the parenchyma. Different mutations have been detected to be associated with hereditary CP, most common R122H, an N291 mutation of the PRSS1 gene, and mutations of the CFTR and SPINK1 gene (36,38–40). The risk of developing pancreatic cancer is increased in HCP with PRSS1 mutation as compared with normal population and chronic alcoholic pancreatitis (41,42). Rare reasons for CP besides pancreatic duct obstruction due to tumors, strictures, diverticula, and pathoanatomical variations like pancreas divisum or annular pancreas are trauma and genetic mutation (14). In up to 20% of the patients the reason for CP remains unclear.

pathogenesis of pain in cp
Pain is the cardinal symptom in patients with CP. Together with the often ongoing consumption of alcohol it is most difficultly to treat. The permanent pain reduces the quality of life, leads to addiction of analgesics, and unemployment or early retirement. CP cannot be cured; therefore the aim of treatment is directed against symptoms (e.g. pain) and complications. Pain in CP is still only fragmentarily understood and a multifactorial nature is assumed, including inflammation, duct obstruction, high pancreatic tissue pressure, fibrotic encasement of sensory nerves, and a neuropathy characterized by both increased numbers and sizes of intrapancreatic sensory nerves and by inflammatory injury to the nerve sheaths allowing exposure of the neural elements to toxic substances. The pain is often localized in the upper part of the abdomen and is frequently nocturnal; sometimes it radiates to the back. Development of pain in the course of the disease is seen in 85% of the patients (43). It is described to be deep, penetrating, and debilitating and may increase after eating (44). In the initial stage of the disease the pain is intermittent and recurrent; later it is persistent. The painless pancreatitis is found rarely in alcoholinduced pancreatitis (<10%), while pain-free periods are seen in late-onset idiopathic pancreatitis. Pain pattern and histopathologic/radiologic findings have to be correlated in consideration of therapy especially surgery. Histological picture

452

CHRONIC PANCREATITIS
and diameter of the main pancreatic duct in CT scan, MRI/MRCP, and ERCP are necessary for optimal planning of the operation. Small duct disease requires other procedures compared to obstruction of the main pancreatic duct and inflammatory mass in the pancreatic head. The assessment of pain is very difficult. Most trials in CP use classifications for description of pre- and postoperative pain or outcome, such as excellent (no pain), good (better), fair (nil), poor (worse) (45), therefore no comparison between different trials is possible. Pain relief is more common in patients that quit drinking. The underlying mechanism for pain in CP is poorly understood. Different concepts have been hypothesized, but none of them can completely explain the pain in this disease. Present hypotheses include increased pressure on the ductal system and parenchyma by obstruction, neuritis, ischemia of the pancreatic tissue and intra- and extrapancreatic causes such as pseudocysts and CBDs or duodenal stenosis. The impact of the mentioned factors for the pathogenesis of the pain remains unclear and may vary between the patients. A higher intraductal pressure was measured in patients with CP compared to controls (46). The reason for increased pressure can be postinflammatory scarring of the pancreatic (main and side) ducts, pancreatic duct stones or strictures, or hemosuccus pancreaticus that leads to obstruction. Other reasons are pancreatic abscess, ascites, bile duct stenosis, or duodenal stenosis. The patients that were found to have a reduced intraductal pressure had a better pain relief compared to patients with higher intraductal pressure in the follow-up (46,47). Additionally it had been reported that phenotypic modification of primary sensory neurons may play a role in production of persisting pain (48). Focal release and uptake of mediators in the peptidergic nerves were changed by initial pancreatic inflammation. Previous trials revealed that number and diameter of the pancreatic nerves, as well as activity are significantly increased in CP (49,50). A correlation between pain and expression of growthassociated protein (43) and level of methionine–enkephalin was detected (50). It is hypothesized that the increases in pressure facilitate the intox of pain mediators into the nerves and result in a neuritis and therefore causing the pain in CP. Another hypothesis is that pancreatic ischemia is responsible for the pain. The ischemia activates the xanthine oxygenase, leading to the production of toxic oxygen metabolites. An increased level of cytochrome P450 in CP was found in several trials (51,52), but treatment with an inhibitor of the xanthine oxygenase did not reduce the pain. Causal for the development of CBD stenosis is the close anatomical relationship of the distal common to the head of the pancreas; hereby the risk of CBD stricture is increased in patients with enlarged pancreatic head. In patients with CP bile duct stricture is found in 5% to 9% and in up to 35% after surgical procedures for CP (55–58). Patients with CBD stricture can present asymptomatic with elevated liver enzymes, alkaline phosphatase, or bilirubin or being septic with cholangitis. In patients with CBD strictures secondary to CP interventional, i.e., surgical therapy is indicated. Ruling out a local malignancy is of greatest importance in patients with duodenal or CBD obstruction. Pancreatic ascites is found in approximately 4% of patients with CP and in 6% to 14% of those with pancreatic pseudocysts and is defined as massive accumulation of pancreatic fluid in the peritoneal cavity. The level of amylase in the ascitic fluid is typically above 1000 IU/L and the ascitic fluid to serum amylase ratio is approximately 6.0 (59,60). In those patients an endoscopic retrograde pancreatography (ERCP) should be performed to localize the site of leakage and to perform endoscopic stenting of the leak (61). Additional treatment with somatostatin or octreotide together with diuretics and repeated paracentesis may be beneficial for some patients (62,63). In patients with persistent or recurrent accumulation of ascites and/or sudden deterioration of clinical status surgery is indicated (64). Pancreaticopleural fistulas, result from a disruption of the pancreatic duct or leakage from a pseudocyst, are rare, but associated with a significant morbidity and mortality (65,66). Three main types of thoracic manifestations include mediastinal pseudocysts, pancreaticopleural fistulas, and pancreaticobronchial fistulas (67,68). Once a pancreaticopleural fistula is suspected, the concentration of amylase in the pleural effusion should be measured. Conservative treatment has an efficacy of 30% to 60%, a recurrence rate of 15%, and a mortality rate of 12% (69). If conservative therapy fails, endoscopic shincterotomy or stenting and surgical procedures should be considered aiming to reduce the hypertension intraductal or within pseudocyst, because the hypertension inhibits the spontaneous closure of fistula. Extrahepatic portal hypertension is a less common complication of CP; it may be confined to either the superior mesenteric or the splenic venous branch or may involve the whole splenomesentericoportal axis (70). It is defined as extrahepatic hypertension of the portal venous system in the absence of liver cirrhosis. The pathogenesis of extrahepatic portal hypertension in CP may include several factors. The inflammatory process is capable of causing initial damage to vascular walls and generating venous spasm, venous stasis, and thrombosis (71). A fibrosis of the pancreas can lead to progressive constriction of the splenomesentericoportal axis. Other reasons are considerable pancreatic head enlargement or compression by pancreatic pseudocysts or inflammatory swelling of the gland (72,73). At present an extrahepatic portal hypertension per se seems not to be an indication for surgical intervention in CP, because there was no evidence of hemorrhage (74) even though a potential risk of esophageal or gastric varices exists (75). Additionally it has to be mentioned that those patients have a

complications of cp
In the course of CP, several complications with less or more life-threatening potential may occur. In 12% of the patients that underwent surgery for CP, duodenal obstruction was detected; additionally it was found to be associated with CBD stenosis (53). Duodenal obstruction can also occur secondarily to pancreatic pseudocysts (54). The patients typically suffer from nausea, vomiting, upper abdominal pain, and weight loss. If duodenal obstruction does not resolve within 1 to 2 weeks of conservative therapy, an irreversible duodenal obstruction should be considered and therefore interventional/ surgical treatment is indicated.

453

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
considerable increased surgical risk. If the varices start to bleed, therapeutic options include interventional measures such as sclerotherapy, variceal ligation, and interventional (TIPSS) or surgical portosystemic shunting procedures (72,76). In patients with thrombosis of the portal vein with cavernous transformation, a trans-section of the pancreatic parenchyma above the portal vein as required for the Beger procedure and pancreatoduodenectomy should be avoided by all means as this carries unpredictable risks. Different endoscopic procedures have been used in treatment of CP including sphincterotomy, endoscopic stone extraction (in some trial combined with ESWL), and stenting of the pancreatic duct. Endoscopic pancreatic sphincterotomy in CP is technically challenging. Indications are sphincter oddi dysfunction or papillary stenosis/stricture. Another indication is to gain better access to the pancreatic duct for dilatation, transpapillary drainage, and stenting. The overall response rate was found to be 55% to 95%, in a follow-up an improvement of pain was found to be 60% after 14 months (78). Using the endoscopic stenting with changes on regularly base, a complete pain relief was found in 45% to 95% of the patients. Early complications (pancreatitis, cholangitis) occurred in 10% to 15% and late complications (strictures, ductal changes) in 10% to 30% of the patients (79). In endoscopic stenting the rate of complications was 32%, the initial pain relieve was 89% (80). In an other small trial stenting was performed in 25 patients with low morbidity (8–10%) and good results in 80% after 13 months (81). In another trial pain control was achieved after stenting in 70% in 12-month follow-up and 62% of the patients in 27-month follow-up; the morbidity was 25% overall (82). Therefore endoscopic stenting plays a role in patients who are unfit for surgery, but it is not recommended as definitive therapy, particularly with regard to the necessity of repeated endoscopic interventions due to infection, stent displacement, or stent occlusion (83,84). A randomized controlled trial (n = 39) recently found that patients who underwent pancreatojejunostomy (Partington– Rochelle) had a better quality of life and better pain relief (pain score 53 ± 21 vs. 25 ± 15) as compared to endoscopic drainage procedures in patients without pathology of the pancreatic head (85). Exocrine insufficiency, mortality, and rate of complications showed no significant differences. Since the inclusion criteria in this trial were obstruction of the pancreatic duct but without inflammatory mass (enlargement <4 cm), these results should be even more true in patients suffering from inflammatory tumor of the pancreatic head with potential organ complications such as stenosis of the CBD and duodenal outlet obstruction (85). In a previous trial including 72 patients comparing endoscopy and surgery a significant advantage of the surgical group was found in a 5-year follow-up with a rate of complete pain relief of 33.8% compared to 15% after endoscopy. The rate of new-onset diabetes mellitus did not show significant differences while patients in the surgical group had a higher increase of the body weight (47.2% vs. 28.6%) (86). It is mandatory that a good selection of patients in that endoscopic treatment for pain relief can be considered, but endoscopy has an important role in management of complications of CP.

conservative and interventional treatment of cp
The treatment of CP and its complications remain a major challenge (11). The most distressing symptom is intractable pain with consecutive abuse of analgesics. Medical therapies such as abstinence from alcohol, dietary alterations, analgesics (such as non-steroidal anti-inflammatory drugs, paracetamol, prednisolone, dextropropoxiphene, and tricylic antidepressiva and later on opioids), oral enzyme supplements, and somatostatin analoga offer improvement of the symptoms for some patients. However, abstaining from excessive alcohol consumption does not interrupt the progression of organ destruction, exocrine and endocrine insufficiency, and the presence of pain (8,17). Therefore different interventional techniques have been presented for management of CP and especially the pain associated with the disease. Endoscopic treatment in patients with CP was established in the last decades. The aim of this intervention is to alleviate outflow obstructions of the pancreatic duct and the CBD (16). Endoscopy has its established role in the management of pancreatic complications, especially drainage of pancreatic pseudocysts by cystic-enteric drainage. Additionally percutaneous catheter drainage is available as a temporizing measure in poor surgical patients with complicated or infected pancreatic pseudocysts. For pain control, endosonography guided or percutaneous celiac nerve block with alcohol or steroids and thoracoscopic splanchnicectomy have been described. Pain relief and rate of responders range from 20% to 87%, but the data on the results are rare. No prospective randomized trial is available at present. However, it was pointed out, that this is a symptomatic therapy (16). These procedures can be repeated as needed, but they are associated with severe complications and symptoms and usually recurrence of pain after a few months. The rate of CP-associated complications such as pseudocysts or progression of endocrine and exocrine insufficiency is not improved. Up to 60% of the patients with CP have pancreatic duct stones, which cause an obstruction and consecutively an increase of the intraductal pressure. This can lead to hypertension and ischemia, causing the pain attacks. Many patients are asymptomatic though they have proven duct stones. Extracorporal shockwave lithotripsy (ESWL) can be used in painful, chronic, calcified pancreatitis. The median delay to pain relief was 1.1 years, but 38% had pain relapse after 2 years. With a combination of ESWL and endoscopy (sphincterotomy and fragment extraction after ESWL), the rate increased to 45%. The mortality was 0, and the morbidity in the endoscopy group was 3% (77).

indications for surgical intervention
The primary therapy for CP is conservative, symptom-related treatment. Surgery should be considered in patients with failure of conservative and endoscopic interventions. Established indication for surgery are intractable pain, complications related to adjacent organs, suspicion of neoplasm, non-resolving stenosis of

454

CHRONIC PANCREATITIS
duodenum or CBD, intractable pain, pseudoaneurysm, or vascular erosion that cannot be controlled by radiological intervention, large pancreatic pseudocysts that cannot be endoscopically controlled, especially in conjunction with ductal pathology, and neither conservatively nor interventionally tractable internal pancreatic fistula (8,9,87–89). The indication for surgical interventions in CP has seen its ups and downs over the last decades. Due to optimized surgical procedures, improvements in intensive care and in selection of the patients, the perioperative risk was reduced and the outcome has improved. In preoperative diagnostic it is a major challenge to differentiate a malignant tumor from an inflammatory mass in the pancreatic head (2). For sufficient histopathological examination it is necessary to provide an adequate specimen to exclude malignancy; only resections or limited resections and extended drainage procedures can provide this (8). In approximately 10% of the patients with pancreatic carcinoma even in experienced centers the initial diagnose of malignancy is based on the histological specimen at the time of operation. An optimal surgical intervention should manage the problems and complications of the CP (Fig. 48.1). Additionally it should guarantee a low relapse rate, preserve a maximum of endocrine and exocrine function, and most importantly, restore quality of life (8). A new drainage procedure was described by Puestow and Gillesby in 1956 (98). Decompression of the main pancreatic duct was performed by a longitudinal latero-lateral pancreaticojejunostomy after resection of the pancreatic tail and splenectomy. A modification of the Puestow–Gillesby procedure, performing a spleen-preserving longitudinal pancreaticojejunostomy without pancreatic tail resection, was introduced by Partington and Rochelle (99). The procedure described by DuVal (96) and Zollinger (97) proved to be effective only if there was a single dominant obstruction between pancreatic tail and the ampulla of Vater. A single dominant stricture is found rarely, especially in chronic alcohol-induced pancreatitis, which is common among the majority of patients in the western hemisphere. It has to be mentioned that recurrent episodes of severe pain were frequently observed even after sufficient drainage of the duct system. For many years the longitudinal pancreaticojejunostomy introduced by Partington–Rochelle was favored in surgical treatment of CP. Even in the presence of multiple strictures (“chain of lakes”) the main pancreatic duct could be effectively drained. No resection is included in this procedure, therefore it was associated with lower perioperative morbidity and mortality compared to resection procedures. The advantages of simple drainage procedures are the maximal preservation of pancreatic tissue. Performing drainage procedure, ruling out a malignancy, is not possible, because no adequate specimen is available for pathological examination. Pain relief was found in 80% to 90% of the patients with non alcohol-induced CP and only 50% to 60% of the patients with alcohol-induced pancreatitis because the inflammatory mass in the pancreatic head including its strictures as well as the local intraductal hypertension of the ducts of second and third order are left behind (8,91). In the long-term follow-up the failure rate of drainage procedure was found to be up to 45%. The reasons were inadequate duct decompression, biliary stenosis, and most importantly inflammatory mass of the pancreatic head. Studies have shown that drainage procedure can prevent or delay the loss of pancreatic function. The rationale for the wide-spread application of drainage procedures in CP was the considerable morbidity and mortality

rationale for drainage procedures
Up to 60% of the patients with CP present with ductal ectasia that may arouse suspicion of intraductal hypertension (8,88,90,91). Therefore decompression of the pancreatic head by drainage has become a major procedure. At the turn of the 19th century the operative removal of pancreatic stones was described (92,93). The rationale for this operative intervention was an alleviation of pain and prevention of pancreatic atrophy (93). Coffey (94) and Link (95) were the first to describe the drainage of the pancreas with bypass by opening the pancreatic main duct. The groups of DuVal (96) and Zollinger (97) independently performed the decompression of the main pancreatic duct by resection of the pancreatic tail and retrograde drainage of the pancreatic duct via a termino-terminal or termino-lateral pancreaticojejunostomy.

Drainage
Cysto(gastro-)jejunostomy Pancreaticojejunostomy (Partington-Rochelle) Drainage and resection of pancreatic tail (Puestow-Mercadier) Left-resection of the pancreas (DuVal) Extended drainage (limited excision of pancreatic head (Frey) Duodenum preserving resection of pancreatic head (Beger) Pylorus-preserving partial duodenopancreatectomy Partial duodenopancreatectomy (Whipple) Pancreatectomy

Resection
Figure 48.1 Surgical armamentarium for the treatment of chronic pancreatitis.

455

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
rates of resectional procedures, i.e., partial pancreatoduodenectomy in the beginnings of pancreatic surgery. Nowadays the only suitable indication for a simple drainage procedure (Partington–Rochelle) with longitudinal pancreaticojejunostomy is in patients with a dilated ductal diameter (>7 mm) or “chain of lakes,” without an inflammatory mass in the pancreatic head and a normal ductal system (8,31).

duodenum-preserving resection of the pancreatic head
In order to combine the advantages of drainage procedures and pancreatoduodenectomy the duodenum-preserving resection was developed. The technique was first described by Beger in 1980 (107). The aim of this procedure was the prevention of sacrificing undiseased organ and achieving optimal control of the symptoms of the pancreatitis, especially pain. Due to minimizing the loss of normal pancreatic tissue this procedure is aiming for a better pancreatic function (108). This Beger procedure consists of a subtotal resection of the pancreatic head following transection of the pancreas above the portal vein. In CP with an inflammatory tumor of the pancreatic head, the transection is the most challenging part of the operation due to displacement or compression of the mesenterico-portal vein axis (Fig. 48.2). The body of the pancreas is drained by an end-to-end or end-to-side pancreatojejunostomy using a Roux-en-Y loop. The resection cavity is drained by the same jejunal loop drains by a side to side anastomosis to the rim of the resection cavity of the pancreatic head. An other advantage of this procedure is that the gastroduodenal passage and CBD continuity may be preserved (102,109). Extensive resection of the pancreatic head with decompression of the bile duct and the duodenum will allow adequate management even in cases of distal CBD stenosis or segmental duodenal obstruction The identification of the intrapancreatic course of the distal bile duct can be facilitated by insertion of a metal probe into the CBD through a proximal choledochotomy (110). The Beger procedure provides long-term pain relief in 75% to 95% of the patients (102,103,106,111–115). The mortality rate ranges from 0% to 3% in experienced centers (111,112,116). The morbidity rate was found to be 15% to 32% (75,108,112). In a randomized trial Berger’s procedure was superior in terms of pain relief as compared with pylorus-preserving partial pancreatoduodenectomy (75). No significant differences in pain relief were found in randomized trails comparing duodenumpreserving resection and partial pancreatoduodenectomy or

rationale for resectional procedures
Approximately 90% to 95% of the patients suffering from CP irrespective of the width of the pancreatic duct present with a pathology in the pancreatic head such as inflammatory mass, neuronal or proximal duct alterations (75,100,101). Thus, the head of the pancreas has been referred to as the pacemaker of the disease and its complications (75,102). The inflammatory enlargement of the pancreatic head is found in 30% to 50% of patients with CP, causes an obstruction of the pancreatic duct and sometimes even obstruction of the duodenum or CBD or segmental portal hypertension (103). Besides the mechanical disturbances the pain is thought to be caused by alteration of the parenchyma and nerve fibers in quality and quantity. An inflammatory mass in the pancreatic head is considered as a contraindication for a simple drainage procedure. The therapeutic principle of the resection of the pancreatic head is to eliminate the obstructive mass to drain the entire pancreatic duct and pancreatic ducts of second or third order. Because pain is related to hypertension of the duct and parenchyma the inflammatory mass is resected for achieving pain relieve. Initial the original Whipple–Kausch Procedure has been performed, which has been replaced by the pylorus-preserving pancreatoduodenectomy (PPPD) introduced by Longmire/ Traverso. The two procedures are basically the same except the preservation of antrum and pylorus resulting in preservation of the normal gastric function and prevention of biliary reflux gastritis. The pancreatoduodenectomy is the standard procedure in cancer of the pancreatic head; therefore pancreatic centers have high experience in this procedure. Comparing classic Whipple procedure with PPPD in a 5-year follow-up, a significantly higher rate of pain and nausea and lower quality of life were found (104). PPPD is an effective procedure in treatment of CP with improvement of the quality of life and short- and long-term pain relief in up to 90% of the patients (105). But the major disadvantage of the procedure is the sacrifice of the surrounding non-diseased organs with loss of the natural bowl continuity. Additionally a significant reduction of pancreatic exocrine and endocrine function was found. Long-term pain relief is reported in 66% to 89% after resection of the pancreatic head (105). After partial pancreatoduodenectomy the perioperative morbidity was found to be 20% to 53% (75,106). Nowadays the procedure can be performed with low mortality and adequate morbidity. Resections of the distal part of the pancreas are often associated with endocrine insufficiency. They offer only short pain relief. Therefore this procedure is in obsolete in treatment of pain. The only suitable indications are localized severe complications of the pancreatic tail such as pseudoaneurysm.

Figure 48.2 Beger procedure.

456

CHRONIC PANCREATITIS
pylorus-preserving partial pancreatoduodenectomy after longterm follow up (117). A modification of the Beger procedure was presented by Frey in 1985. The Frey procedure (115,116,118) combines a longitudinal pancreaticojejunostomy according to Partington and Rochelle (99) with a local excision of the pancreatic head. It is marked by leaving a rim of pancreatic tissue across the portal vein and superior mesenteric vein (Fig. 48.3). Therefore this operation is easier to learn and perform; additionally only one pancreatojejunostomy is necessary. The drainage of the resection cavity of pancreatic head, body, and tail is performed with a longitudinal pancreatojejunostomy using a Roux-en-Y loop. The Frey procedure can be performed without mortality (<1%) and low morbidity (9–39%) (7,106,112). In these patients, 56% were free of pain, and 32% had substantial pain relief. The main complications are hemorrhage, pancreatic fistula (0–5%), and intra-abdominal abscess. After 5 years 78% suffered from exocrine and 60% from endocrine; 44% were professionally rehabilitated. By preserving the duodenum, the physiological gastroduodenal passage and the continuity of the CBD are sustained. Additionally exocrine and endocrine pancreatic functions are preserved and the procedure is able to control complications of adjacent organs such as CBD stenosis, duodenal stenosis, and internal pancreatic fistulas comparable to the Beger procedure (118,119). The Hamburg procedure is a modification of the Frey procedure and was proposed by Izbicki and coworkers. The extent of pancreatic head excision can be modified up to a subtotal excision including the uncinate process combined with a longitudinal V-shaped excision of the ventral aspect of the pancreas into the pancreatic duct (Fig. 48.4). If ductal irregularities are present in the pancreatic body and tail, the operation can be extended as a drainage operation much in the way of a Partington–Rochelle procedure into the pancreatic tail. Therefore the major advantage of this procedure is that the extent of the resection can be customized to the individual morphology of the pancreas (113). In a recently published trial, the mortality and morbidity of the Hamburg procedure were 0% and 19.6%, respectively; 89% of the patients were free of pain in the follow-up; the increase of the body weight was significantly better as compared with classic PPPD (31). The Berne procedure is combining the Frey and Beger procedure (120). A duodenum-preserving resection of the pancreatic head is performed according to the Beger technique. Compared to the Frey procedure the extent of the excision of the pancreatic is much larger and is therefore definitely decompressing the CBD and preventing a potential recurrence. According to the Frey procedure the hazardous dissection is avoided by leaving a small shape of the pancreatic tissue on the anterior wall of the portal vein (120,121). Performing the Berne procedure the mortality was found to be 0% and the morbidity 20% (108).

comparison of different surgical approaches
Recently Müller published the long-term follow-up of an RCT, comparing Beger and PPPD, that was originally presented by Büchler 1995 (n = 40). No significant differences were found in terms of morbidity, mortality, and survival. After 24 months a significant better gain of body weight was reported after Beger procedure. Favorable results concerning loss of appetite could be detected after 14 years. Interestingly the rate of painfree patients was significantly higher after Beger (75%) compared to PPPD (40%) in short-term follow-up, while no significant differences could be detected concerning the pain score after 14 years. No significant differences were found comparing new-onset diabetes mellitus, enzyme substitution, and global health status (75,122). Additionally Farkas presented the results of a 12-month follow-up comparing Beger and PPPD in 2006 (n = 40). Significant advantages for the Beger procedure were noticed in terms of operating time (142.5 ± 4.9 vs. 278.5 ± 6.9 min), hospital stay (8.5 ± 0.9 vs. 13.8 ± 3.9 days), morbidity (0% vs. 30%), and increase of body weight (7.8 ± 0.9 vs. 3.2 ± 0.3 kg), while rates of pain-free patients (86% vs. 83%), mortality, and diabetes missed significance (123). The results of a prospective randomized trial comparing Frey versus PPPD were presented by Izbicki 1998 including a

Figure 48.3 Frey procedure.

Figure 48.4 Hamburg procedure.

457

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 48.1 Outcome after Frey procedure and Pylorus preserving pancreatoduodenectomy
Pyloruspreserving pancreatoduodenectomy 0% 53% 10 12.5 0 50 18.1 25 25 0 0 18.75 15% 0% 65% 96% 0% 39%

Table 48.2 Outcome after Frey procedure and Beger procedure
Beger procedure Perioperative mortality Perioperative morbidity 30-mo follow-up VAS Frequency of pain attacks Pain medications Inability to work Pain score 8-yr follow-up VAS Frequency of pain attacks Pain medications Inability to work Pain score Late mortality Late mortality (chronic pancreatitis associated) Endocrine insufficiency Exocrine insufficiency Reoperation Professional rehabilitation
Source: From Refs. (12,125).

Frey procedure 0% 22% 16 0 0 0 4 20 25 0 0 11.25 24% 6% 60% 78% 12% 38%

Frey procedure Perioperative mortality Perioperative morbidity 30-mo follow-up VAS Frequency of pain attacks Pain medications Inability to work Pain score 7-yr follow-up VAS Frequency of Pain attacks Pain medications Inability to work Pain score Late mortality Late mortality(chronic pancreatitis associated) Endocrine insufficiency Exocrine insufficiency Reoperation Professional rehabilitation
Source: From Ref. (106,124).

0% 32% 12 0 0 0 3 20 25 0 0 11.25 24% 3% 56% 88% 8% 59%

3% 19% 12 12.5 0 0 6.1 20 25 0 0 17.75 20% 3% 61% 86% 8% 42%

24-month follow-up (n = 61). In terms of morbidity rate the Frey procedure (19%) was superior to PPPD (53%). Additionally the Quality of Life was significantly higher (85.7 vs. 75.1) and the pain score better (6.1 vs. 18.1) after Frey procedure. In the long-term follow-up recently presented by Strate the favorable results still exists, but statistical significance is lost. Additionally no difference was found comparing late mortality, rate of endocrine and exocrine insufficiency, and need of reoperations (106,124) (Table 48.1). The only randomized trial comparing Frey and Beger procedure has been published by Izbicki 1995 (n = 42) (112). The perioperative mortality was 0 in both groups, while the morbidity rate was significantly lower after Frey (9%) compared to 15% after Beger procedure. In the 8-year follow-up published by Strate 2005 the rate of endocrine and exocrine insufficiency and rate of reoperation were comparable in both groups. No significant differences concerning Quality of Life and pain score were found at short-term and long-term follow-up (125) (Table 48.2). Recently a trial comparing Beger versus Berne modification was published (n = 65). The quality of life measured with the EORTC QLQ-PAN 26 was better, while no difference was found using the EORTC QLQ C30. The duration of hospital stay and the mean operating time were significantly shorter in the Berne group. No significant differences were found comparing the rates of reoperation, complications, and need of blood products. It had to be mentioned that the analysis was performed on an intention to treat basis, with 14 of the

65 patients were treated with other operation compared to randomization due to intraoperative findings (108). Short-term results favor the organ sparing operation, and additionally it is the easier operation to perform (easier to learn anyway) (106). In patients with small duct disease (diameter <3 mm), extended resection procedures are suggested (31). Additionally pain, nausea, and fatigue were significantly reduced. In terms of pain relief and quality of life and exocrine and endocrine function no significant differences were detected (7,125).

small duct disease
Recently a new entity of the sclerosing ductal pancreatitis characterized by a maximal Wirsungian duct diameter of less than 3 mm was described (126). In literature there is a controversy on the definition of a dilated pancreatic duct (90,127). The normal diameter of the main duct is 3 to 5 mm in relation to age (90,91). Independently of the discussion of the definition of a dilation of the main pancreatic ducts there is controversy concerning surgical technique. The majority of members of the American Pancreas Club considered a duct size of a minimum of 8 mm sufficient to justify a pancreaticojejunostomy. Others regarded a duct size of 5 mm as the limit to perform a pancreatojejunostomy rather than a pancreaticojejunostomy (90). An another extended drainage procedure has been proposed for treatment of the small duct disease: a longitudinal V-shaped excision of the ventral aspect of the pancreas combined with a longitudinal pancreatojejunostomy (Fig. 48.5).

458

CHRONIC PANCREATITIS

references
1. Lohr JM. What are the useful biological and functional markers of earlystage chronic pancreatitis? J Gastroenterol 2007 January; 42(Suppl 17): 66–71. 2. Mayerle J, Lerch MM. Is it necessary to distinguish between alcoholic and nonalcoholic chronic pancreatitis? J Gastroenterol 2007 January; 42(Suppl 17): 127–30. 3. Andersen BN, Pedersen NT, Scheel J, Worning H. Incidence of alcoholic chronic pancreatitis in Copenhagen. Scand J Gastroenterol 1982 March; 17(2): 247–52. 4. Ammann RW, Mullhaupt B. Do the diagnostic criteria differ between alcoholic and nonalcoholic chronic pancreatitis? J Gastroenterol 2007 January; 42(Suppl 17): 118–26. 5. Ammann RW, Akovbiantz A, Largiader F, Schueler G. Course and outcome of chronic pancreatitis. Gastroenterology 1984; 86: 820–8. 6. Thorsgaard PN, Nyboe AB, Pedersen G, Worning H. Chronic pancreatitis in Copenhagen. A retrospective study of 64 consecutive patients. Scand J Gastroenterol 1982 October; 17(7): 925–31. 7. Pessaux P, Kianmanesh R, Regimbeau JM, et al. Frey procedure in the treatment of chronic pancreatitis: short-term results. Pancreas 2006 November; 33(4): 354–8. 8. Knoefel WT, Eisenberger CF, Strate T, Izbicki JR. Optimizing surgical therapy for chronic pancreatitis. Pancreatology 2002; 2: 379–85. 9. Warshaw AL. Pain in chronic pancreatitis – patients, patience, and the impatient surgeon. Gastroenterology 1984; 86: 987–9. 10. Kloppel G. Toward a new classification of chronic pancreatitis. J Gastroenterol 2007 January; 42(Suppl 17): 55–7. 11. Ammann RW. A clinically based classification system for alcoholic chronic pancreatitis: summary of an international workshop on chronic pancreatitis. Pancreas 1997; 14: 215–21. 12. Strate T, Knoefel WT, Yekebas E, Izbicki JR. Chronic pancreatitis: etiology, pathogenesis, diagnosis, and treatment. Int J Colorectal Dis 2003 March; 18(2): 97–106. 13. Banks PA. Classification and diagnosis of chronic pancreatitis. J Gastroenterol 2007 January; 42(Suppl 17): 148–51. 14. Ammann RW. Diagnosis and management of chronic pancreatitis: current knowledge. Swiss Med Wkly 2006 March 18; 136(11–12): 166–74. 15. Ammann RW, Akovbiantz A, Largiader F. Pain relief in chronic pancreatitis with and without surgery. Gastroenterology 1984; 87: 746–77. 16. Adler DG, Lichtenstein D, Baron TH, et al. The role of endoscopy in patients with chronic pancreatitis. Gastrointest Endosc 2006; 63(7): 933–7. 17. Strate T, Yekebas E, Knoefel WT, Bloechle C, Izbicki JR. Pathogenesis and the natural course of chronic pancreatitis. Eur J Gastroenterol Hepatol 2002; 14: 929–34. 18. Chari ST. Chronic pancreatitis: classification, relationship to acute pancreatitis, and early diagnosis. J Gastroenterol 2007 January; 42(Suppl 17): 58–9. 19. Malka D, Hammel P, Sauvanet A, et al. Risk factors for diabetes mellitus in chronic pancreatitis. Gastroenterology 2000; 119: 1324–32. 20. Lankisch PG, Happe-Loehr A, Otto J, Creutzfeldt W. Natural course in chronic pancreatitis. Pain, exocrine and endocrine pancreatic insufficiency and prognosis of the disease. Digest 1993; 54: 148–55. 21. Layer P, Yamamoto H, Kalthoff L, Clain JE, Bakken LJ, DiMagno EP. The different course of early- and late-onset idiopathic and alcoholic chronic pancreatitis. Gastroenterology 1994; 107: 1481–7. 22. Ammann RW, Buehler H, Muench R, Freiburghaus AW, Siegenthaler W. Differences in the natural history of idiopathic (nonalcoholic) and alcoholic chronic pancreatitis. A comparative long-term study of 287 patients. Pancreas 1987; 2(4): 368–77. 23. Levy P, Milan C, Pignon JP, Baetz A, Bernades P. Mortality factors associated with chronic pancreatitis. Unidimensional and multidimensional analysis of a medical-surgical series of 240 patients. Gastroenterology 1989; 96: 1165–72. 24. Ammann RW, Akovbiantz A, Largiader F, Schueler G. Course and outcome of chronic pancreatitis. Longitudinal study of a mixed medicalsurgical series of 245 patients. Gastroenterology 1984; 86: 820–88. 25. Lankisch PG. Natural course of chronic pancreatitis. Pancreatology 2001; 1: 3–14.

Figure 48.5 V-Shape.

If the pancreatitis is accompanied by an enlarged pancreatic head a pancreatic head resection should be performed.

salvage procedures
Due to improvement of surgical technique and selection of patients, pancreatic surgery for CP yields excellent results. In some cases recurrence develops; most frequently in the remnant of the pancreatic head indicating either insufficient surgical resection of the head of the pancreas or aggressive disease. In these patients “re-do” pancreas head resections are indicated. The procedures that should be considered are the partial pancreatoduodenectomy (Whipple procedure, PPPD) and in selected patients (i.e. re-recurrence) even total splenopancreatoduodenectomy. This procedure is indicated in patients that underwent partial pancreatoduodenectomy, and additional interventional nerve blocks or surgical denervation fail to achieve definitive pain relief. In patients that previously underwent duodenum-preserving resection of the pancreatic head or partial pancreatoduodenectomy with recurrence of the CP in the body or tail a V-shape drainage procedure is indicated.

conclusion
The aim of treatment of CP is mainly pain relief and improvement in the quality of life, which still poses a major challenge today. Duodenum-preserving pancreatic head resection is the ideal procedure for treatment of CP. If ductal pathology is present in the pancreatic body or tail the procedure can be combined with longitudinal duct drainage in various degrees which allows a tailored concept. Duodenum-preserving resections of the pancreas combine high safety with high efficacy and offer the best short-term outcome, while the long-term results of PPPD are comparable. In small duct pancreatitis (duct diameter <3 mm), a V-shape excision is the therapy of choice. Pancreatic surgery bears many pitfalls and potential complications and is technically demanding. It should be left to experts in high-volume hospitals in order to minimize mortality and morbidity.

459

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
26. Kloppel G, Maillet B. Pathology of acute and chronic pancreatitis. Pancreas 1993 November; 8(6): 659–70. 27. Kloppel G, Maillet B. The morphological basis for the evolution of acute pancreatitis into chronic pancreatitis. Virchows Arch A Pathol Anat Histopathol 1992; 420(1): 1–4. 28. Sarles H. Epidemiology and physiopathology of chronic pancreatitis and the role of the pancreatic stone protein. Clin Gastroenterol 1984; 13: 895–912. 29. Kloppel G. Pathology of chronic pancreatitis and pancreatic pain. Acta Chir Scand 1990; 156(4): 261–5. 30. Lehnert P. Ätiologie und Pathogenese der chronischen Pankreatitis. Internist 1979; 20: 321–30. 31. Yekebas EF, Bogoevski D, Honarpisheh H, et al. Long-term follow-up in small duct chronic pancreatitis: a plea for extended drainage by “V-shaped excision” of the anterior aspect of the pancreas. Ann Surg 2006; 244(6): 940–6. 32. Ectors N, Maillet B, Aerts R, et al. Non alcoholic duct destructive chronic pancreatitis. Gut 1997; 41: 263–8. 33. Yoshida K, Toki F, Takeuchi T, Watanabe S, Shiratori K, Hayashi N. Chronic pancreatitis caused by an autoimmune abnormality. Proposal of the concept of autoimmune pancreatitis. Dig Dis Sci 1995 July; 40(7): 1561–8. 34. Toomey DP, Swan N, Torreggiani W, Conlon KC. Autoimmune pancreatitis. Br J Surg 2007; 94(9): 1067–74. 35. Choi EK, Kim MH, Lee TY, et al. The sensitivity and specificity of serum immunoglobulin G and immunoglobulin G4 levels in the diagnosis of autoimmune chronic pancreatitis: Korean experience. Pancreas 2007; 35(2): 156–61. 36. Kloppel G. Chronic pancreatitis, pseudotumors and other tumor-like lesions. Mod Pathol 2007 February; 20(Suppl 1): S113–31. 37. Barkin JS, Fayne SD. Chronic pancreatitis: update 1986. Mt Sinai J Med 1986; 53(5): 404–8. 38. Whitcomb DC, Gorry MC, Preston RA, et al. Hereditary pancreatitis is caused by a mutation in the cationic trypsinogen gene. Nat Genet 1996; 14(2): 141–5. 39. Gorry MC, Gabbaizedeh D, Furey W, et al. Mutations in the cationic trypsinogen gene are associated with recurrent acute and chronic pancreatitis. Gastroenterology 1997; 113(4): 1063–8. 40. Teich N, Mossner J, Keim V. Mutations of the cationic trypsinogen in hereditary pancreatitis. Hum Mutat 1998; 12(1): 39–43. 41. Lowenfels AB, Maisonneuve P, DiMagno EP,et al. Hereditary pancreatitis and the risk of pancreatic cancer. International Hereditary Pancreatitis Study Group. J Natl Cancer Inst 1997; 89(6): 442–6. 42. Malka D, Hammel P, Maire F, et al. Risk of pancreatic adenocarcinoma in chronic pancreatitis. Gut 2002; 51(6): 849–52. 43. Bradley EL. Long-term results of pancreatojejunostomy in patients with chronic pancreatitis. Am J Surg 1987; 153: 207–13. 44. Glasbrenner B, Adler G. Evaluating pain and the quality of life in chronic pancreatitis. Int J Pancreatol 1997; 22(3): 163–70. 45. Mannell A, Adson MA, McIlrath DC, Ilstrup DM. Surgical management of chronic pancreatitis: long-term results in 141 patients. Br J Surg 1988; 75(5): 467–72. 46. Ebbehoj N, Svendsen LB, Madsen P. Pancreatic tissue pressure in chronic obstructive pancreatitis. Scand J Gastroenterol 1984; 19: 1066–8. 47. Ebbehoj N, Borly L, Bulow J, Rasmussen SG, Madsen P. Evaluation of pancreatic tissue fluid pressure and pain in chronic pancreatitis. A longitudinal study. Scand J Gastroenterol 1990 May; 25(5): 462–6. 48. Woolf CJ. Phenotypic modification of primary sensory neurons: the role of nerve growth factor in the production of persistent pain. Philos Trans R Soc Lond B Biol Sci 1996 March 29; 351(1338): 441–8. 49. Bockman DE, Buechler M, Malfertheiner P, Beger HG. Analysis of nerves in chronic pancreatitis. Gastroenterology 1988; 94: 1459–69. 50. Di SP, Fink T, Weihe E, et al. Immune cell infiltration and growth-associated protein 43 expression correlate with pain in chronic pancreatitis. Gastroenterology 1997; 112(5): 1648–55. 51. Foster JR, Idle JR, Hardwick JP, et al. Induction of drug-metabolizing enzymes in human pancreatic cancer and chronic pancreatitis. J Pathol 1993; 169(4): 457–63. 52. Acheson DW, Hunt LP, Rose P, Houston JB, Braganza JM. Factors contributing to the accelerated clearance of theophylline and antipyrine in adults with exocrine pancreatic disease. Clin Sci (Lond) 1989; 76(4): 377–85. Prinz RA, Aranha GV, Greenlee HB. Combined pancreatic duct and upper gastrointestinal and biliary tract drainage in chronic pancreatitis. Arch Surg 1985; 120(3): 361–6. Warshaw AL. Conservation of pancreatic tissue by combined gastric, biliary, and pancreatic duct drainage for pain from chronic pancreatitis. Am J Surg 1985; 149(4): 563–9. Bradley EL. Parapancreatic biliary and intestinal obstruction in chronic obstructive pancreatitis. Is prophylactic bypass necessary? Am J Surg 1986; 151(2): 256–8. Huizinga WK, Thomson SR, Spitaels JM, Simjee AE. Chronic pancreatitis with biliary obstruction. Ann R Coll Surg Engl 1992; 74(2): 119–23. Vijungco JD, Prinz RA. Management of biliary and duodenal complications of chronic pancreatitis. World J Surg 2003; 27(11): 1258–70. Wislooff F, Jakobsen J, Osnes M. Stenosis of the common bile duct in chronic pancreatitis. Br J Surg 1982; 69(1): 52–4. Uchiyama T, Yamamoto T, Mizuta E, Suzuki T. Pancreatic ascites-–a collected review of 37 cases in Japan. Hepatogastroenterology 1989; 36(4): 244–8. Runyon BA. Amylase levels in ascitic fluid. J Clin Gastroenterol 1987; 9(2): 172–4. Bracher GA, Manocha AP, DeBanto JR, et al. Endoscopic pancreatic duct stenting to treat pancreatic ascites. Gastrointest Endosc 1999 June; 49(6): 710–5. Oktedalen O, Nygaard K, Osnes M. Somatostatin in the treatment of pancreatic ascites. Gastroenterology 1990; 99(5): 1520–1. Uhl W, Buchler MW, Malfertheiner P, et al. A randomised, double blind, multicenter trial of octreotide in moderate to severe acute pancreatitis. Gut 1999; 45: 97–104. Gomez-Cerezo J, Barbado CA, Suarez I, et al. Pancreatic ascites: study of therapeutic options by analysis of case reports and case series between the years 1975 and 2000. Am J Gastroenterol 2003; 98(3): 568–77. Saps M. Re: Gomez-Cerezo et al. Pancreatic ascites: study of therapeutic options by analysis of case reports and case series between the years 1975 and 2000. Am J Gastroenterol 2003; 98(10): 2332–3. Chebli JM, Gaburri PD, de Souza AF, et al. Internal pancreatic fistulas: proposal of a management algorithm based on a case series analysis. J Clin Gastroenterol 2004; 38(9): 795–800. Stenger AM, Knoefel WT, Dahmen U, Blochle C, Izbicki JR. Pancreaticobronchial fistula with communication to a pseudoaneurysm of the arteria lienalis as a rare complication in chronic pancreatitis. Z Gastroenterol 1998; 36(12): 1047–51. Izbicki JR, Wilker DK, Waldner H, Rueff FL, Schweiberer L. Thoracic manifestations of internal pancreatic fistulas: report of five cases. Am J Gastroenterol 1989; 84(3): 265–71. Kaman L, Behera A, Singh R, Katariya RN. Internal pancreatic fistulas with pancreatic ascites and pancreatic pleural effusions: recognition and management. ANZ J Surg 2001; 71(4): 221–5. Izbicki JR, Yekebas EF, Strate T, et al. Extrahepatic portal hypertension in chronic pancreatitis: an old problem revisited. Ann Surg 2002; 236: 82–9. Little AG, Moossa AR. Gastrointestinal hemorrhage from left-sided portal hypertension. An unappreciated complication of pancreatitis. Am J Surg 1981; 141(1): 153–8. Bernades P, Baetz A, Levy P, et al. Splenic and portal venous obstruction in chronic pancreatitis. A prospective longitudinal study of a medical-surgical series of 266 patients. Dig Dis Sci 1992; 37: 340–36. Malka D, Hammel P, Levy P, et al. Splenic complications in chronic pancreatitis: prevalence and risk factors in a medical-surgical series of 500 patients. Br J Surg 1998; 85(12): 1645–9. Bloechle C, Busch C, Tesch C, et al. Prospective randomized study of drainage and resection on non-occlusive segmental portal hypertension in chronic pancreatitis. Br J Surg 1997; 84(4): 477–82. Buchler MW, Friess H, Muller MW, Wheatley AM, Beger HG. Randomized trial of duodenum-preserving pancreatic head resection versus pylorus-preserving Whipple in chronic pancreatitis. Am J Surg 1995; 169(1): 65–9.

53.

54.

55.

56.

57. 58. 59.

60. 61.

62. 63.

64.

65.

66.

67.

68.

69.

70. 71.

72.

73.

74.

75.

460

CHRONIC PANCREATITIS
76. Warshaw AL, Jin GL, Ottinger LW. Recognition and clinical implications of mesenteric and portal vein obstruction in chronic pancreatitis. Arch Surg 1987; 122: 410–45. 77. Dumonceau JM, Costamagna G, Tringali A, et al. Treatment for painful calcified chronic pancreatitis: extracorporeal shock wave lithotripsy versus endoscopic therapy. A randomised controlled trial. Gut 2007 Apr; 56(4): 545–52. 78. Martin RF, Hanson BL, Bosco JJ, et al. Combined modality treatment of symptomatic pancreatic ductal lithiasis. Arch Surg 1995; 130(4): 375–9. 79. Buscaglia JM, Kalloo AN. Pancreatic sphincterotomy: technique, indications, and complications. World J Gastroenterol 2007; 13(30): 4064–71. 80. Weber A, Schneider J, Neu B, et al. Endoscopic stent therapy for patients with chronic pancreatitis: results from a prospective follow-up study. Pancreas 2007; 34(3): 287–94. 81. Heyries L, Sahel J. Endoscopic treatment of chronic pancreatitis. World J Gastroenterol 2007; 13(46): 6127–33. 82. Eleftherladis N, Dinu F, Delhaye M, et al. Long-term outcome after pancreatic stenting in severe chronic pancreatitis. Endoscopy 2005; 37(3): 223–30. 83. Born P, Rosch T, Bruhl K, et al. Long-term results of endoscopic treatment of biliary duct obstruction due to pancreatic disease. Hepatogastroenterology 1998; 45(21): 833–9. 84. Kahl S, Zimmermann S, Genz I, et al. Risk factors for failure of endoscopic stenting of biliary strictures in chronic pancreatitis: a prospective follow-up study. Am J Gastroenterol 2003; 98(11): 2448–53. 85. Cahen DL, Gouma DJ, Nio Y, et al. Endoscopic versus surgical drainage of the pancreatic duct in chronic pancreatitis. N Engl J Med 2007; 356(7): 676–84. 86. Dite P, Ruzicka M, Zboril V, Novotny I. A prospective, randomized trial comparing endoscopic and surgical therapy for chronic pancreatitis. Endosc 2003; 35(7): 553–8. 87. Lankisch PG, Andren-Sandberg A. Standards for the diagnosis of chronic pancreatitis and for the evaluation of treatment. Int J Pancreatol 1993; 14: 205–12. 88. Frey CF, Suzuki M, Isaji S. Treatment of chronic pancreatitis complicated by obstruction of the common bile duct or duodenum. World J Surg 1990; 14: 59–69. 89. Saeger HD, Schwall G, Trede M. Standard Whipple in chronic pancreatitis. In: Beger HG, Buechler M, Malfertheimer P, eds. Standards in Pancreatic Surgery, 1st edn. Berlin: Springer, 1993, 385–91. 90. Frey CF. Why and when to drain the pancreatic ductal system. In: Beger HG, Buechler MW, Ditschuneit H, Malfertheiner P, eds. Chronic Pancreatitis. Berlin: Springer, 1990, 415–25. 91. Markowitz JS, Rattner DW, Warshaw AL. Failure of symptomatic relief after pancreaticojejunal decompression for chronic pancreatitis. Strategies for salvage. Arch Surg 1994; 129: 374–9. 92. Gould AP. Pancreatic calculi: transactions of the Clinical Society of London. Lancet 1898; 2: 1532. 93. Moynihan SB. Pancreatic calculus. Lancet 1902; 4: 335. 94. Coffey R. Pancreaticojejunostomy and pancreatectomy. Ann Surg 1909; 50: 1238–64. 95. Link G. Treatment of chronic pancreatitis by pancreatectomy. Ann Surg 1911; 53: 768–82. 96. DuVal MK. Caudal pancreatico-jejunostomy for chronic relapsing pancreatitis. Ann Surg 1954; 140: 775–85. 97. Zollinger RM, Keith LM, Ellison EH. Pancreatitis. N Engl J Med 1954; 251: 497–502. 98. Puestow CB, Gillesby WJ. Petrograde surgical drainage of pancreas for chronic pancreatitis. Arch Surg 1958; 76: 898–906. 99. Partington PF, Rochelle REL. Modified Puestow procedure for retrograde drainage of the pancreatic duct. Ann Surg 1960; 152: 1037–43. 100. Izbicki JR, Bloechle C, Knoefel WT, Rogiers X, Kuechler T. Surgical treatment of chronic pancreatitis and quality of life after operation. Surg Clin North Am 1999; 79: 913–44. 101. Traverso LW, Tompkins RK, Urrea PT, Longmire WP, Jr. Surgical treatment of chronic pancreatitis. Twenty-two years’ experience. Ann Surg 1979 September; 190(3): 312–19. 102. Beger HG, Buechler M. Duodenum preserving resection of the head of the pancreas in chronic pancreatitis with inflammatory mass in the head. World J Surg 1990; 14: 83–7. 103. Buechler M, Friess H, Isenmann R, Bittner R, Beger HG. Duodenumpreserving resection of the head of the pancreas: the Ulm experience. In: Beger HG, Buechler M, Malfertheimer P, eds. Standards in Pancreatic Surgery, 1st edn. Berlin: Springer, 1993, 436–49. 104. Mobius C, Max D, Uhlmann D, et al. Five-year follow-up of a prospective non-randomised study comparing duodenum-preserving pancreatic head resection with classic Whipple procedure in the treatment of chronic pancreatitis. Langenbecks Arch Surg 2007; 392(3): 359–64. 105. Schnelldorfer T, Lewin DN, Adams DB. Reoperative surgery for chronic pancreatitis: is it safe? World J Surg 2006; 30(7): 1321–8. 106. Izbicki JR, Bloechle C, Broering DC, et al. Extended drainage versus resection in surgery for chronic pancreatitis – Prospective randomized trial comparing the longitudinal pancreaticojejunostomy combined with local pancreatic head excision with the pylorus preserving pancreatoduodenectomy. Ann Surg 1998; 228: 771–9. 107. Beger HG, Witte C, Krautzberger W, Bittner R. Experiences with duodenum-sparing pancreas head resection in chronic pancreatitis. Chirurg 1980; 51: 303–7. 108. Koninger J, Seiler CM, Sauerland S, et al. Duodenum-preserving pancreatic head resection—a randomized controlled trial comparing the original Beger procedure with the Berne modification (ISRCTN No. 50638764). Surgery 2008; 143(4): 490–8. 109. Izbicki JR, Bloechle C, Knoefel WT, et al. Complications of adjacent organs in chronic pancreatitis managed by duodenum-preserving resection of the head of the pancreas. Br J Surg 1994; 81(9): 1351–5. 110. Wilker DK, Izbicki JR, Knoefel WT, Geissler K, Schweiberer L. Duodenum-preserving resection of the head of the pancreas in treatment of chronic pancreatitis. Am J Gastroenterol 1990; 85(8): 1000–4. 111. Buechler MW, Friess H, Bittner R, et al. Duodenum-preserving pancreatic head resection: long-term results. J Gastrointest Surg 1997; 1: 13–9. 112. Izbicki JR, Bloechle C, Knoefel WT, et al. Duodenum preserving resections of the head of the pancreas in chronic pancreatitis – a prospective randomized trial. Ann Surg 1995; 221: 350–8. 113. Izbicki JR, Bloechle C, Broering DC, Kuechler T, Broelsch CE. Longitudinal V-shaped excision of the ventral pancreas for small duct disease in severe chronic pancreatitis. Ann Surg 1998; 227: 213–19. 114. Izbicki JR, Bloechle C, Knoefel WT, et al. Drainage versus Resektion in der chirurgischen Therapie der chronischen Kopfpankreatitis: eine randomisierte Studie. Chirurg 1997; 68: 369–77. 115. Frey CF, Amikura K. Local resection of the head of the pancreas combined with longitudinal pancreaticojejunostomy in the management of patients with chronic pancreatitis. Ann Surg 1994; 220: 492–507. 116. Frey CF, Smith GJ. Description and rationale of a new operation for chronic pancreatitis. Pancreas 1987; 2: 701–7. 117. Muller MW, Friess H, Martin DJ. Long-term follow-up of a randomized clinical trial comparing Beger with pylorus-preserving Whipple procedure for chronic pancreatitis. Br J Surg 2008 Mar; 95(3): 350–6 118. Beger HG, Krautzberger W, Bittner R, Büchler M, Limmer J. Duodenumpreserving resection of the head of the pancreas in patients with severe chronic pancreatitis. Surgery 1985; 97: 467–73. 119. Beger HG, Buechler M, Bittner R, Oettinger W, Roscher R. Duodenumpreserving resection of the head of the pancreas in severe chronic pancreatitis. Ann Surg 1989; 209: 273–8. 120. Gloor B, Friess H, Uhl W, Buchler MW. A modified technique of the Beger and Frey procedure in patients with chronic pancreatitis. Dig Surg 2001; 18: 21–5. 121. Koninger J, Friess H, Muller M, et al. Duodenum-preserving pancreas head resection-an operative technique for retaining the organ in the treatment of chronic pancreatitis. Chirurg 2004; 75(8): 781–8. 122. Muller MW, Friess H, Martin DJ, et al. Long-term follow-up of a randomized clinical trial comparing Beger with pylorus-preserving Whipple procedure for chronic pancreatitis. Br J Surg 2008; 95(3): 350–6. 123. Farkas G, Leindler L, Daroczi M, Farkas G, Jr. Prospective randomised comparison of organ-preserving pancreatic head resection with pyloruspreserving pancreaticoduodenectomy. Langenbecks Arch Surg 2006; 391(4): 338–42.

461

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
124. Strate T, Bachmann K, Busch P, et al. Resection vs drainage in treatment of chronic pancreatitis: long-term results of a randomized trial. Gastroenterology 2008; 134: 1406–11. 125. Strate T, Taherpour Z, Bloechle C, et al. Long-term follow-up of a randomized trial comparing the Beger and Frey procedures for patients suffering from chronic pancreatitis. Ann Surg 2005; 241(4): 591–8. 126. Izbicki JR, Bloechle C, Broering DC, Kuechler T, Broelsch CE. Longitudinal V-shaped excision of the ventral pancreas for small duct disease in severe chronic pancreatitis: prospective evaluation of a new surgical procedure. Ann Surg 1998 February; 227(2): 213–19. 127. Delcore R, Rodriguez FJ, Thomas JH, Forster J, Hermreck AS. The role of pancreatojejunostomy in patients without dilated pancreatic ducts. Am J Surg 1994; 168: 598–601.

462

49 Pancreatic injury
history

Demetrios Demetriades, Beat Schnüriger, and Galinos Barmparas
diagnosis
Clinical Presentation The diagnosis of penetrating pancreatic injury is usually made intraoperatively and does not pose any significant diagnostic problems. However, the diagnosis in blunt trauma is often challenging. A missed or delayed diagnosis of pancreatic injury increases morbidity and mortality (15). Because of its retroperitoneal location and infrequent occurrence, timely diagnosis of blunt pancreatic injury requires a high index of suspicion and is a challenging task even to the most experienced surgeon. Clinical signs are often vague and nonspecific. Even significant pancreatic injuries may present initially with minimal epigastric pain, with signs of peritonitis developing many hours or even days after the injury. The cornerstone of early diagnosis is a combination of serial physical examinations, laboratory tests, and imaging studies. Laboratory Investigations Unfortunately, no laboratory test is either sensitive or specific enough in evaluating suspected pancreatic injuries. Serum amylase levels have long been used as a useful marker and may assist in the diagnosis (18,19). Takishima et al. found a timedependent increase of serum amylase in patients suffering blunt pancreatic trauma (20). Elevated serum amylase was present in all cases when the samples were collected more than 3 hours after the injury. Therefore, serum amylase on admission may be particularly unreliable and should be followed serially. Additionally, no relation between the grade of pancreatic injury and the level of hyperamylasemia was found (20). The sensitivity and specificity of serum amylase in detecting pancreatic trauma range from 48% to 89% and 64% to 81%, respectively (Table 49.2). When used as a screening tool after blunt abdominal trauma, a normal serum amylase has a negative predictive value of 93% to 98% (20–24). Various other injuries, such as brain injuries, salivary gland, duodenal, and small bowel trauma may be associated with increased serum amylase levels (25–28). The determination of amylase isoenzymes does not improve the sensitivity of specificity. Radiologic Investigations Plain abdominal films and ultrasonography are of limited value in the diagnostic work-up of patients with suspected pancreatic injury. Contrast-enhanced helical CT is considered as the imaging study of choice for these patients (Fig. 49.1). Its accuracy increases in parallel with the interval between the CT study and the injury. In suspicious injuries a repeat CT scan at least 6 to 8 hours after the initial investigation is recommended. The timing of the intravenous contrast bolus, as well as the experience of the radiologist involved affects the diagnostic precision (29,30). The overall sensitivity and specificity of CT for identification of pancreatic injuries of all grades is reported

The first reported case of pancreatic injury was discovered at an autopsy in 1827, in a woman who was hit and killed by a stagecoach (1). The first documented posttraumatic pancreatic fistula was published in 1905 (2). In 1904, Garre operated successfully on a patient with a transected pancreas (3). Advances over the next few decades led to significant improvements in the diagnosis and management of pancreatic injuries.

epidemiology
Pancreatic trauma remains fairly uncommon. The overall incidence in blunt trauma is reported to be 0.2% and in penetrating trauma about 1% (4–9). However, it is likely that some injuries, especially after blunt trauma may remain undiagnosed. Penetrating trauma accounts for the majority of injuries (70–80%), with gunshot wounds being the most common mechanism (72%) (8). The location of the injury is evenly distributed among the head/neck and body of the pancreas (about 40% each), with the tail being less frequently injured (about 20%) (8,10). Because of the retroperitoneal location of the pancreas, significant force is mandated to lead to its injury. This fact, in combination with the proximity of the pancreas to vital structures, makes isolated pancreatic injuries rare. Overall, about 60% of patients with blunt trauma and about 90% with penetrating trauma have associated intra-abdominal injuries (7,11–15). Pancreatic trauma should be considered as a marker of other intra-abdominal injuries. The most commonly associated injuries in blunt trauma are of the spleen (34%), liver (26%), and duodenum (6%). In penetrating trauma, first is the stomach (53%), followed by the liver (51%) (7). Associated intra-abdominal vascular injuries are of major concern since they are the most common cause of early mortality. More than 75% of penetrating injuries to the head of the pancreas are associated with a major vascular injury (8).

injury grading
There are numerous classification systems for pancreatic injuries. The most widely accepted grading system is the one proposed by the Organ Injury Scaling Committee of the American Association for the Surgery of Trauma (OIS-AAST) in 1990 (Table 49.1) (16). This classification scheme takes into account the type of injury (hematoma or laceration), the presence or absence of structural duct involvement, and the location of pancreatic injury (proximal or distal to superior mesenteric vein). OIS-AAST grades I and II are considered as low-grade and grades III–V as high-grade pancreatic injuries (17). The classification may be based on computed tomography (CT) and intraoperative or autopsy findings. It is useful in the evaluation and management of patients with pancreatic injuries, and it is an excellent research tool for the comparison of the safety and efficacy of the various therapeutic approaches (15).

463

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 49.1 American Association of the Surgery of Trauma Organ Injury Scale for the Pancreas
Grade I Type of injury Hematoma Laceration Hematoma Laceration III IV V Laceration Laceration Laceration Description of injury Minor contusion without duct injury Superficial laceration without duct injury Major contusion without duct injury or tissue loss Major laceration without duct injury or tissue loss Distal transection or parenchymal injury with duct injury Proximal transection or parenchymal injury with duct injury Massive destruction of pancreatic head

II

Table 49.2 Sensitivity and Specificity of the Diagnostic Adjunctive and Level of Evidence
Sensitivity (%) Amylase levels CT MRCP ERCP 48–89 80 >95 >95 Specificity (%) 64–81 80 >95 >95 Level of evidence (references) III (20–24) III (10,31–33) III–IV (37–39) III–IV (40,41)

Figure 49.2 MRCP with detection of a pancreatic duct disruption (white arrow).

technique for the evaluation of pancreatic injuries and ductal status (Fig. 49.2). MRCP is diagnostic in 95% to 99% of cases of pancreatic ductal injuries, and complete visualization of normal-sized pancreatic ducts occurs in 97% of the patients (37–39) (Table 49.2). Secretin administration may improve ductal visualization (39). However, MRCP is applicable only in those patients who are hemodynamically stable and have minimal other injuries and does not permit therapeutic intervention, such as stent placement. Endoscopic retrograde pancreatography (ERCP) is the gold standard in the evaluation of the pancreatic duct and may also have therapeutic application by placement of a stent over a ductal injury (40–43). The procedure should be performed as early as possible, preferably within 12 to 24 hours of injury, to prevent abdominal septic complications (44). ERCP is often not available in an emergency setting and requires hemodynamic stability. The procedure may cause pancreatitis and the long-term results of stenting are still not well known (45,46). Intraoperative Exposure and Evaluation Many patients with pancreatic injury present with peritonitis or hemodynamic instability, requiring immediate abdominal exploration, before any preoperative diagnostic evaluation. A midline incision is the preferred approach in trauma because it provides an optimal exposure for the evaluation of other associated injuries. The pancreas is explored only after control of any bleeding and contamination due to a hollow viscus injury. A pancreatic injury should be suspected by the presence of lesser sac fluid collection, retroperitoneal bile staining, retroperitoneal hematoma, and fat necrosis of the omentum and retroperitoneum (47). Most of the pancreas can be visualized by opening the lesser sac. This maneuver can be easily and rapidly done by dividing

Figure 49.1 Grade IV pancreatic injury (white circle) and concomitant grade IV liver injury after blunt abdominal trauma.

to be around 80%, and the positive predictive value ranges from 80% to 100% (10,31–33) (Table 49.2). The study tends to underestimate the severity of pancreatic injury (34). With the introduction of multidetector CT and improved techniques, the diagnostic accuracy is likely to improve (35,36). The sensitivity of detecting ductal injury with multidetector CT has been measured in a single study at 91% (32). However, to further assess the integrity of the pancreatic duct, additional investigations may be needed. Magnetic resonance imaging (MRI) with cholangiopancreatography (MRCP) is a noninvasive, alternative imaging

464

PANCREATIC INJURY

(A)

(B)

Figure 49.3 (A,B) Opening of the lesser sac allows exposure and evalutaion of the anterior surface of the body and tail of the pancreas. Kocher maneuver allows evaluation of the pancreatic head. Source : From Ref. 83.

the gastrocolic ligament between the stomach and the transverse colon. This maneuver exposes the anterior, inferior, and superior surfaces of the body and tail of the pancreas (Fig. 49.3A). Any attachments between the pancreas and the posterior wall of the stomach are divided. Exploration of the posterior portion of the pancreas can be performed by an incision in the peritoneum on the inferior border of the pancreas in the region of suspected injury. The avascular nature of the retropancreatic plane allows the index finger to be carefully slipped behind the pancreas where a laceration can often be felt. Improved access to the posterior portion of the body and tail can be achieved by dissecting the splenocolic, splenophrenic, and splenorenal ligament. The plane anterior to the kidneys usually can be developed with careful blunt dissection (48). To assess the head and uncinate process of the pancreas and the integrity of the duodenum and the bile duct, an extended Kocher maneuver should be performed. This includes the mobilization of the second and third portion of the duodenum, pancreatic head, and distal common bile duct from their retroperitoneal position. This procedure allows inspection and bimanual palpation of the anterior and posterior surfaces of the head and uncinate process (Fig. 49.3B). Intraoperative Evaluation of the Integrity of the Pancreatic Duct The major determinant of morbidity and mortality related to pancreatic trauma is the integrity of the main pancreatic duct (49–51). Most trauma patients have normal-sized pancreatic ducts, which can be difficult to visualize. The use of magnifying glasses and administration of secretin may facilitate visualization of the duct. Several radiological and endoscopic methods of intraoperative pancreatography have been described (52–56), but are rarely used in trauma. If there is a low probability of ductal injury or the patient is hemodynamically unstable, simple drainage of the peripancreatic space is the most appropriate

Figure 49.4 Duodenotomy and catheterization of the ampulla of Vater for intraoperative pancreatography. Source : From Ref. 83.

approach. Postoperative evaluation by means of CT or MRCP may be considered in patients with persistent elevation of the serum amylase or pancreatic leaks. Intraoperative pancreatography may be considered in selected stable patients with suspicious pancreatic head ductal injury. It can be performed by injecting contrast medium into the gallbladder after clamping the proximal common bile duct. The administration of morphine to cause contraction of the sphincter of Oddi may aid in visualizing the pancreatic duct. In about 10% of subjects, the common bile duct and pancreatic duct drain separately, and the pancreatic duct is not visible with this technique. However, the images obtained may be useful in assessing the intrapancreatic portion of the common bile duct and the integrity of the ampulla of Vater (57). If the duodenum is already open, the ampulla of Vater may be cannulated directly (Fig. 49.4). Identification of the ampulla of Vater can be difficult, especially in the presence of edema or hematoma, and magnifying glasses are strongly recommended in order to identify this structure. The major duct is cannulated

465

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
using a pediatric feeding tube, and several milliliters of contrast medium are instilled. A static radiograph or fluoroscopy is taken to identify contrast extravasation. Amputation of the tail to gain access to the pancreatic duct for pancreaticography is mentioned only to condemn. Intraoperative ERCP has been used to assess the ductal system. This is a time-consuming approach and is rarely used, especially in severe trauma (56). In summary, visualization of the pancreatic duct using complex techniques should be considered only if the patient is stable and the surgeon is prepared to act on the findings and perform major pancreatic resection if needed.

management
The rarity of these injuries precludes the development of evidence-based guidelines. Expert opinions and small series reports (levels III and IV evidence) constitute the major source of the knowledge gained throughout the years. Nonoperative Management (NOM) Selective NOM might have a place in the management of carefully selected patients with blunt abdominal trauma. Only hemodynamically stable patients without evidence of peritonitis are potential candidates for NOM. Experience of NOM in pediatric population showed that this approach is safe and effective. Keller et al. (58) in a National Pediatric Trauma Registry study reviewed 154 pancreatic injuries in children, 80% with low-grade and 20% with high-grade injuries. NOM was successful in about half of high-grade injuries and about 80% of low-grade injuries. The frequency of NOM increased significantly over the last year of the study. However, pediatric pancreatic trauma is different from adult trauma because of the much lower risk of pancreatic duct injury in children (less than 1% in children, about 15% in adults). Subsequent experience in adult blunt trauma patients confirmed the safety of this approach (58–62). In a recent study there were no failures of NOM in grade I and 17.3% failure in grade II pancreatic injuries (61). In summary, NOM is safe for low grade injuries and may be acceptable in selected high-grade injuries. Early evaluation by means of ERCP or MRCP helps to visualize the integrity of the pancreatic duct. In addition ERCP might be used for stent placement, which can be an excellent adjunctive tool to NOM (63). The downside to NOM is the development of a pseudocyst or fistulae and occasionally severe pancreatitis (64). Most of these complications are treatable by percutanous CT-guided drainage. There is no evidence that the use of somatostatin analogs, such as octreotide, increases the success rate of NOM, although they might have a role in the treatment of posttraumatic pancreatic fistulae (65). Operative Management The surgical treatment plan of pancreatic injury should be determined by the overall condition of the patient, the grade and location of pancreatic injury, the concomitant injuries, and the experience of the surgeon. The strengths of management recommendations for pancreatic trauma are all of level C or D.

Figure 49.5 Anastomosis of the distal pancreatic stump to a Roux-en-Y jejunal loop. Source : From Ref. 83.

Low-grade injuries discovered intraoperatively are best managed with sparse debridement of nonviable tissue, hemostasis, and wide external drainage with closed suction drains. Although some authors advocate repair of the pancreatic capsule (66), this may lead to further parenchymal damage and pseudocyst formation and should only be done to control bleeding. Application of topical hemostatics on the pancreatic laceration may facilitate hemostasis and reduce the risk of postoperative leaks. When dealing with high-grade injuries (AAST-OIS grades III, IV, and V), the choice of procedure depends on the general condition of the patient and the location of duct injury, i.e., in the head, neck, or tail of the pancreas. Distal ductal injuries (AAST-OIS grade III) are best treated by distal pancreatectomy (67,68) often en bloc with the spleen. Spleen-preserving distal pancreatectomy should be considered in hemodynamically stable patients especially in children (69–72). Splenic preservation is technically more challenging and may result in increased blood loss. After completion of the distal pancreatectomy, the main pancreatic duct is identified if possible and is suture-ligated with nonabsorbable suture. With the main duct ligated, the proximal parenchyma should be closed with a mattress sutures or a TA stapling device. Extensive pancreatic resection to the right side of the superior mesenteric vessels may lead to diabetes or exocrine insufficiency. In these cases, the distal pancreas may be preserved and anastomosed to a Roux-en-Y jejunal loop (Fig. 49.5). Closed suction drains should be placed around the remaining pancreas and left upper quadrant if the spleen has been removed.

466

PANCREATIC INJURY
Proximal ductal injuries and injuries to the head of the pancreas (AAST-OIS grade IV) are more challenging. In the presence of hemodynamic instability or major associated injuries, or if the surgeon has no experience with complex pancreatic surgery, no attempts should be made to evaluate the integrity of the duct or perform major resections. In these cases, the safest option is hemostasis and liberal external drainage (50,68). Damage control with packing and temporary abdominal closure may be necessary. In destructive injuries to the head of the pancreas or the duodenum, a pancreaticoduodenectomy may be necessary. It should only be performed as a primary procedure in hemodynamically stable patients by an experienced surgeon. In severely compromised patients the surgeon should opt for damage control and a two-stage procedure (73–75). Primarily, damage control surgery should be performed to control the hemorrhage and any intestinal spillage. Major associated vascular injuries may be managed by ligation or temporary arterial shunts reinforced by abdominal packing. Complex gastrointestinal injuries may be stapled off and the common bile duct exteriorly drained. The abdominal wall is temporarily closed and the patient is transferred to the intensive care unit for stabilization. The definitive Whipple’s procedure should be deferred for 24 to 48 hours after restoration of the hemodynamic status, coagulability, and normalization of body temperature. The reconstruction, including pancreatico-jejunostomy or pancreatico-gastrostomy (76,77), choledochojejunostomy, and gastro-enterostomy, is similar to that in elective cases. Insertion of a jejunal feeding tube beyond the ligament of Treitz is strongly recommended to allow feeding in prolonged complicated cases. Overall, pancreaticoduodenectomy is rarely indicated in trauma. In a review at our center over 7.5 years, only 16 of 214 (7.4%) patients with pancreatic injuries required pancreaticoduodenectomy (73).

Figure 49.6 Very large posttraumatic peripancreatic pseudocyst.

outcomes
Mortality Many patients with pancreatic injury die from associated exsanguinating injuries. The overall pancreas-related mortality is lower than 1% and is usually the result of sepsis and organ failure. In pancreatico-duodenal resection the overall mortality is about 30% to 40% (7,73). Complications Pancreas-related local complications occur in about 25% of patients. The complication rate depends on the severity of pancreatic injury and associated injuries. The most common complications include pancreatic fistulae, peripancreatic fluid collection, local sepsis, pseudocysts, and pancreatitis. Most of these complications can successfully be managed non-operatively, either expectantly or with percutaneous drainage or endoscopic stenting of the duct. Pancreatic Leaks and Fistulae Early pancreatic leaks are the most common complications after surgery for pancreatic trauma and occur in 10% to 35% of patients (7,78). The amylase levels of this fluid should be determined to confirm the pancreatic origin. The prognosis is excellent and the vast majority close spontaneously within days or

weeks (79). Administration of a somatostatin analog may accelerate the resolution of the pancreatic leaks (65). The role of prophylactic use of somatostatin following pancreatic trauma is controversial and there is no evidence to support its routine use. Routine parenteral nutrition is not necessary in this group of patients. Many of them tolerate oral or intestinal tube feeding well, without any significant increase of the drain output. Parenteral nutrition should only be considered in cases where enteral feeding is associated with an increase of the output. No invasive diagnostic procedures, such as ERCP, should be considered during the early postoperative period, because of the high rate of spontaneous healing. Specific investigations should be considered only after persistently high output which shows no evidence of improvement. An MRCP might delineate the anatomy of the pancreatic duct and select those patients who might benefit from ERCP duct stenting. The long-term experience with stenting of posttraumatic leaks is still limited but encouraging. Previous case series found a high incidence of long-term ductal stricture and therefore concluded that the role of pancreatic duct stenting remains uncertain (46). Other investigators however, have reported successful stent placement, even in complicated cases (80–82). Peripancreatic Fluid Collection and Pseudocysts These complications are fairly common and may occur with either operative or non-operative management of pancreatic injuries. They are usually diagnosed during CT scan evaluation of the abdomen (Fig. 49.6). The nature of the fluid collection varies and it can be residual bleeding, serosanguinous fluid, pancreatic exudate, or true pancreatic juice. The natural history of truly pancreatic collections depends on the integrity of the main pancreatic duct. If the main duct is intact, the majority of these collections resolve spontaneously without any intervention. If the main duct is injured, the collection may persist and evolve to a pancreatic pseudocyst. Asymptomatic patients with small peripancreatic collections should be managed expectantly. Symptomatic patients, or those with large collections, may require CT-guided percutaneous drainage. If external drainage fails, as shown by persistently high drain output without any signs of improvement, an

467

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
ERCP with duct stenting should be considered. In rare cases operative resection or internal drainage of a pseudocyst may be necessary. Pancreatic Insufficiency Endocrine and exocrine insufficiency may occur after major pancreatic resections and the patients should be monitored on a regular basis, at least during the first few months after the resection. Late Strictures Late stricture of the choledochojejunostomy following pancreaticoduodenectomy for trauma is common because of the usually small size of the common bile duct. This stricture may manifest many months after the operation. It is strongly recommended that the liver function is monitored regularly, especially during the early months after the injury, in order to identify and treat the problem before liver damage occurs.
5. Tyburski JG, Dente CJ, Wilson RF et al. Infectious complications following duodenal and/or pancreatic trauma. Am Surg 2001 Mar; 67(3): 227–30; discussion 30–1. 6. Feliciano DV, Burch JM, Sput-Parinely V. Abdominal gunshut wounds. An urban trauma center’s experience with 300 consecutive patients. Ann Surg 1988; 208(3): 362–70. 7. Vasquez JC, Coimbra R, Hoyt DB, Fortlage D. Management of penetrating pancreatic trauma: an 11-year experience of a level-1 trauma center. Injury 2001 Dec; 32(10): 753–9. 8. Asensio JA, Demetriades D, Hanpeter DE, Gambaro E, Chahwan S. Management of pancreatic injuries. Curr Probl Surg 1999 May; 36(5): 325–419. 9. Akhrass R, Yaffe MB, Brandt CP, et al. Pancreatic trauma: a ten-year multi-institutional experience. Am Surg 1997 Jul; 63(7): 598–604. 10. Leppaniemi A, Haapiainen R, Kiviluoto T, Lempinen M. Pancreatic trauma: acute and late manifestations. Br J Surg 1988 Feb; 75(2): 165–7. 11. Stone HH, Fabian TC, Satiani B, Turkleson ML. Experiences in the management of pancreatic trauma. J Trauma 1981 Apr; 21(4): 257–62. 12. Graham J, Mattox K, Jordan G. Traumatic injuries of the pancreas. Am J Surg 1978; 136: 744. 13. Sukul K, Lont HE, Johannes EJ. Management of pancreatic injuries. Hepatogastroenterology 1992 Oct; 39(5): 447–50. 14. Sorensen V, Obeid F, Horst H. Penetrating pancreatic injuries. Am Surg 1986; 52: 354. 15. Bradley EL 3rd, Young PR Jr, Chang MC, et al. Diagnosis and initial management of blunt pancreatic trauma: guidelines from a multiinstitutional review. Ann Surg 1998 Jun; 227(6): 861–9. 16. Moore EE, Cogbill TH, Malangoni MA, et al. Organ injury scaling, II: Pancreas, duodenum, small bowel, colon, and rectum. J Trauma 1990 Nov; 30(11): 1427–9. 17. Kao LS, Bulger EM, Parks DL, Byrd GF, Jurkovich GJ. Predictors of morbidity after traumatic pancreatic injury. J Trauma 2003 Nov; 55(5): 898–905. 18. Naffziger HC, McCorkle HJ. The recognition and management of acute trauma to the pancreas: with particular reference to the use of the serum amylase test. Ann Surg 1943 Oct; 118(4): 594–602. 19. Elman R. The variations of blood amylase during acute transient disease of the pancreas. Ann Surg 1937 Mar; 105(3): 379–84. 20. Takishima T, Sugimoto K, Hirata M, et al. Serum amylase level on admission in the diagnosis of blunt injury to the pancreas: its significance and limitations. Ann Surg 1997 Jul; 226(1): 70–6. 21. White PH, Benfield JR. Amylase in the management of pancreatic trauma. Arch Surg 1972 Aug; 105(2): 158–63. 22. Olsen WR. The serum amylase in blunt abdominal trauma. J Trauma 1973 Mar; 13(3): 200–4. 23. Moretz JA 3rd, Campbell DP, Parker DE, Williams GR. Significance of serum amylase level in evaluating pancreatic trauma. Am J Surg 1975 Dec; 130(6): 739–41. 24. Bouwman DL, Weaver DW, Walt AJ. Serum amylase and its isoenzymes: a clarification of their implications in trauma. J Trauma 1984 Jul; 24(7): 573–8. 25. Liu KJ, Atten MJ, Lichtor T, et al. Serum amylase and lipase elevation is associated with intracranial events. Am Surg 2001 Mar; 67(3): 215–9; discussion 9–20. 26. Takahashi M, Maemura K, Sawada Y, Yoshioka T, Sugimoto T. Hyperamylasemia in critically injured patients. J Trauma 1980 Nov; 20(11): 951–5. 27. Coleman EJ, Dietz PA. Small bowel injuries following blunt abdominal trauma. Early recognition and management. N Y State J Med 1990 Sep; 90(9): 446–9. 28. Lawrence DM. Gastrointestinal trauma. Crit Care Nurs Clin North Am 1993 Mar; 5(1): 127–40. 29. Venkatesh SK, Wan JM. CT of blunt pancreatic trauma: a pictorial essay. Eur J Radiol 2008 Aug; 67(2): 311–20. 30. Linsenmaier U, Wirth S, Reiser M, Korner M. Diagnosis and classification of pancreatic and duodenal injuries in emergency radiology. Radiographics 2008 Oct; 28(6): 1591–602. 31. Cirillo RL Jr, Koniaris LG. Detecting blunt pancreatic injuries. J Gastrointest Surg 2002 Jul–Aug; 6(4): 587–98. 32. Teh SH, Sheppard BC, Mullins RJ, Schreiber MA, Mayberry JC. Diagnosis and management of blunt pancreatic ductal injury in the era of highresolution computed axial tomography. Am J Surg 2007 May; 193(5): 641–3; discussion 3.

summary
Pancreatic injuries remain fairly uncommon with penetrating trauma accounting for most cases. The diagnosis of isolated pancreatic injury due to blunt trauma may be missed on the initial evaluation and this delay increases morbidity and mortality. The diagnosis in these cases can be made timely by a high index of suspicion, serial clinical examinations, serial amylase levels, and repeat CT scan evaluation. All penetrating injuries to the pancreas require surgical interventions but a significant number of isolated injuries due to blunt trauma can safely be managed non-operatively, provided that the main pancreatic duct is intact and the patient is hemodynamically stable and has no signs of peritonitis. The type of operative management of the pancreatic injury depends on the severity and site of the injury, associated injuries, hemodynamic condition of the patient, and the experience of the surgeon. The majority of injuries can be managed by hemostasis and closed suction drainage. Distal injuries are best managed by distal pancreatectomy, with or without splenic preservation. Proximal injuries may be managed with hemostasis and drainage or extended distal pancreatectomy. In many pancreatic head injuries, if the duodenum is not severely damaged, hemostasis and external drainage may be sufficient. Pancreaticoduodenectomy is very rarely indicated and should be reserved only for destructive injuries to the head of pancreas or the duodenum. Damage control procedures should be considered in hemodynamically unstable, coagulopathic patients. Pancreas-related complications are common after severe trauma but they can usually be managed successfully with non-operative methods.

references
1. Travers B. Rupture of the pancreas. Lancet. 1827; 12: 384. 2. Korte W. The Pancreas: Its Surgery and Pathology. Philadelphia: Saunders, 1907. 3. Garre C. Totaler Querriss des Pankreas durch Naht geheilt. Beit Clin Chir 1904; 46: 233. 4. Wind P, Tiret E, Cunningham C, et al. Contribution of endoscopic retrograde pancreatography in management of complications following distal pancreatic trauma. Am Surg 1999 Aug; 65(8): 777–83.

468

PANCREATIC INJURY
33. Ilahi O, Bochicchio GV, Scalea TM. Efficacy of computed tomography in the diagnosis of pancreatic injury in adult blunt trauma patients: a singleinstitutional study. Am Surg 2002 Aug; 68(8): 704–7; discussion 7–8. 34. Akhrass R, Kim K, Brandt C. Computed tomography: an unreliable indicator of pancreatic trauma. Am Surg 1996 Aug; 62(8): 647–51. 35. Shanmuganathan K. Multi-detector row CT imaging of blunt abdominal trauma. Semin Ultrasound CT MR 2004 Apr; 25(2): 180–204. 36. Mullinix AJ, Foley WD. Multidetector computed tomography and blunt thoracoabdominal trauma. J Comput Assist Tomogr 2004 Jul–Aug; 28(Suppl 1): S20–7. 37. Fulcher AS, Turner MA, Yelon JA, et al. Magnetic resonance cholangiopancreatography (MRCP) in the assessment of pancreatic duct trauma and its sequelae: preliminary findings. J Trauma 2000 Jun; 48(6): 1001–7. 38. Nirula R, Velmahos GC, Demetriades D. Magnetic resonance cholangiopancreatography in pancreatic trauma: a new diagnostic modality? J Trauma 1999 Sep; 47(3): 585–7. 39. Ragozzino A, Manfredi R, Scaglione M, et al. The use of MRCP in the detection of pancreatic injuries after blunt trauma. Emerg Radiol 2003 Apr; 10(1): 14–8. 40. Hata M, Murao Y, Konobu T, Okuchi K, Nakajima Y. Laparoscopic treatment for peripheral pancreatic duct injury after blunt abdominal trauma: report of a case. Surg Today 2002; 32(7): 659–62. 41. Wolf A, Bernhardt J, Patrzyk M, Heidecke CD. The value of endoscopic diagnosis and the treatment of pancreas injuries following blunt abdominal trauma. Surg Endosc 2005 May; 19(5): 665–9. 42. Varadarajulu S, Noone TC, Tutuian R, Hawes RH, Cotton PB. Predictors of outcome in pancreatic duct disruption managed by endoscopic transpapillary stent placement. Gastrointest Endosc 2005 Apr; 61(4): 568–75. 43. Telford JJ, Farrell JJ, Saltzman JR, et al. Pancreatic stent placement for duct disruption. Gastrointest Endosc 2002 Jul; 56(1): 18–24. 44. Degiannis E, Glapa M, Loukogeorgakis SP, Smith MD. Management of pancreatic trauma. Injury 2008 Jan; 39(1): 21–9. 45. Harrell DJ, Vitale GC, Larson GM. Selective role for endoscopic retrograde cholangiopancreatography in abdominal trauma. Surg Endosc 1998 May; 12(5): 400–4. 46. Lin BC, Liu NJ, Fang JF, Kao YC. Long-term results of endoscopic stent in the management of blunt major pancreatic duct injury. Surg Endosc 2006 Oct; 20(10): 1551–5. 47. Olah A, Issekutz A, Haulik L, Makay R. Pancreatic transection from blunt abdominal trauma: early versus delayed diagnosis and surgical management. Dig Surg 2003; 20(5): 408–14. 48. Aird I, Helman P. Bilateral anterior transabdominal adrenalectomy. Br Med J 1955 Sep 17; 2(4941): 708–9. 49. Cogbill TH, Moore EE, Kashuk JL. Changing trends in the management of pancreatic trauma. Arch Surg 1982 May; 117(5): 722–8. 50. Degiannis E, Levy RD, Velmahos GC, et al. Gunshot injuries of the head of the pancreas: conservative approach. World J Surg 1996 Jan; 20(1): 68–71; discussion 2. 51. Lewis G, Krige JE, Bornman PC, Terblanche J. Traumatic pancreatic pseudocysts. Br J Surg 1993 Jan; 80(1): 89–93. 52. Barkin JS, Ferstenberg RM, Panullo W, Manten HD, Davis RC Jr. Endoscopic retrograde cholangiopancreatography in pancreatic trauma. Gastrointest Endosc 1988 Mar–Apr; 34(2): 102–5. 53. Beckingham IJ, Krige JE, Bornman PC, Terblanche J. Long term outcome of endoscopic drainage of pancreatic pseudocysts. Am J Gastroenterol 1999 Jan; 94(1): 71–4. 54. Berni GA, Bandyk DF, Oreskovich MR, Carrico CJ. Role of intraoperative pancreatography in patients with injury to the pancreas. Am J Surg 1982 May; 143(5): 602–5. 55. Hayward SR, Lucas CE, Sugawa C, Ledgerwood AM. Emergent endoscopic retrograde cholangiopancreatography. A highly specific test for acute pancreatic trauma. Arch Surg 1989 Jun; 124(6): 745–6. 56. Wisner DH, Wold RL, Frey CF. Diagnosis and treatment of pancreatic injuries. An analysis of management principles. Arch Surg 1990 Sep; 125(9): 1109–13. 57. Bornman PC, Krige JE. Management strategies in pancreatic trauma. Johannesburg: University of the Witwatersrand Press, 2006. 58. Keller MS, Stafford PW, Vane DW. Conservative management of pancreatic trauma in children. J Trauma 1997 Jun; 42(6): 1097–100. 59. Holmes JHt, Wiebe DJ, Tataria M, et al. The failure of nonoperative management in pediatric solid organ injury: a multi-institutional experience. J Trauma 2005 Dec; 59(6): 1309–13. 60. Stringer MD. Pancreatic trauma in children. Br J Surg 2005 Apr; 92(4): 467–70. 61. Duchesne JC, Schmieg R, Islam S, Olivier J, McSwain N. Selective nonoperative management of low-grade blunt pancreatic injury: are we there yet? J Trauma 2008 Jul; 65(1): 49–53. 62. Kouchi K, Tanabe M, Yoshida H, et al. Nonoperative management of blunt pancreatic injury in childhood. J Pediatr Surg 1999 Nov; 34(11): 1736–9. 63. Lin BC, Fang JF, Wong YC, Liu NJ. Blunt pancreatic trauma and pseudocyst: management of major pancreatic duct injury. Injury 2007 May; 38(5): 588–93. 64. Canty TG Sr, Weinman D. Management of major pancreatic duct injuries in children. J Trauma 2001 Jun; 50(6): 1001–7. 65. Phelan HA, Minei JP. Pancreatic trauma: diagnostic and therapeutic strategies. Curr Treat Options Gastroenterol 2005 Oct; 8(5): 355–63. 66. Glancy KE. Review of pancreatic trauma. West J Med 1989 Jul; 151(1): 45–51. 67. Patton JH Jr, Fabian TC. Complex pancreatic injuries. Surg Clin North Am 1996 Aug; 76(4): 783–95. 68. Patton JH Jr, Lyden SP, Croce MA, et al. Pancreatic trauma: a simplified management guideline. J Trauma 1997 Aug; 43(2): 234–9; discussion 9–41. 69. Cogbill TH, Moore EE, Morris JA Jr, et al. Distal pancreatectomy for trauma: a multicenter experience. J Trauma 1991 Dec; 31(12): 1600–6. 70. Pachter HL, Hofstetter SR, Liang HG, Hoballah J. Traumatic injuries to the pancreas: the role of distal pancreatectomy with splenic preservation. J Trauma 1989 Oct; 29(10): 1352–5. 71. Dawson DL, Scott-Conner CE. Distal pancreatectomy with splenic preservation: the anatomic basis for a meticulous operation. J Trauma 1986 Dec; 26(12): 1142–5. 72. Schein M, Freinkel W, D’Egidio A. Splenic conservation in distal pancreatic injury: stay away from the hilum! J Trauma 1991 Mar; 31(3): 431. 73. Asensio JA, Petrone P, Roldan G, Kuncir E, Demetriades D. Pancreaticoduodenectomy: a rare procedure for the management of complex pancreaticoduodenal injuries. J Am Coll Surg 2003 Dec; 197(6): 937–42. 74. Koniaris LG. Role of pancreatectomy after severe pancreaticoduodenal trauma. J Am Coll Surg 2004 Apr; 198(4): 677–8; author reply 8–9. 75. Krige JE, Beningfield SJ, Nicol AJ, Navsaria P. The management of complex pancreatic injuries. S Afr J Surg 2005 Aug; 43(3): 92–102. 76. McKay A, Mackenzie S, Sutherland FR, et al. Meta-analysis of pancreaticojejunostomy versus pancreaticogastrostomy reconstruction after pancreaticoduodenectomy. Br J Surg 2006 Aug; 93(8): 929–36. 77. Nakao A, Fujii T, Sugimoto H, et al. Is pancreaticogastrostomy safer than pancreaticojejunostomy? J Hepatobiliary Pancreat Surg 2006; 13(3): 202–6. 78. Bassi C, Dervenis C, Bultirini G. Postoperativ pancreatic fistula – an international study group (IS6PF) definition. Surgery 2005; 138: 8. 79. Subramanian A, Dente CJ, Feliciano DV. The management of pancreatic trauma in the modern era. Surg Clin North Am 2007 Dec; 87(6): 1515–32, x. 80. Hsieh CH, Liu NJ, Chen RJ, Fang JF, Lin BC. Endoscopically placed pancreatic stent in a patient with concomitant two locations of main pancreatic duct disruption following pancreatic trauma. Hepatogastroenterology 2003 Jan–Feb; 50(49): 269–71. 81. Kim HS, Lee DK, Kim IW, et al. The role of endoscopic retrograde pancreatography in the treatment of traumatic pancreatic duct injury. Gastrointest Endosc 2001 Jul; 54(1): 49–55. 82. Bagci S, Tuzun A, Erdil A, et al. Endoscopic treatment of pancreatic duct disruption due to blunt abdominal trauma: a case report. Mil Med 2007 May; 172(5): 548–50. 83. Blumgart MD, ed. Surgery of the Liver, Biliary Tract and Pancreas. London: Elsevier, 2006.

469

50 Pancreas transplantation
Khalid Khwaja
Over the past four decades, the field of pancreas transplantation has seen much evolution, both through refinement in surgical techniques and improvement in immunosuppressive strategies. Pancreas transplantation is primarily performed in uremic, type I diabetics in conjunction with a kidney transplant, and when successful, results in freedom from exogenous insulin therapy, amelioration, or even reversal, of diabetesassociated complications and improved quality of life. Pancreas transplants, contrary to other organ transplants, are not considered essential to patient survival. For each individual recipient, the potential benefit of a pancreas transplant must be carefully weighed against the risk of a major surgical operation and lifelong immunosuppression. The first successful pancreas transplant was performed at the University of Minnesota in 1966 (1). Early outcomes were poor, largely due to technical and infectious complications and the lack of effective and safe immunosuppressive regimens. Over the next decade, the efforts of investigators at several centers in Europe and North America resulted in refinement of surgical techniques and improvement in results. With the introduction of cyclosporine into clinical practice, the field of pancreas transplantation blossomed in the 1980s and outcomes became comparable to those of other transplanted organs. To date, over 20,000 pancreas transplants have been performed worldwide, as reported by the International Pancreas Transplant Registry and about 75% of these have been in the United States (2). In 2006 alone, 1386 pancreas transplants were performed in the United States, with approximately 4000 people waiting for pancreas transplants at the end of that year (Fig. 50.1) (3). centers, involves performing a living-donor kidney transplant at the same time as a deceased donor pancreas transplant (7,8). SPK using a kidney and a partial pancreas graft and a kidney from the same living donor has also been reported (9). Pancreas After Kidney Transplant (PAK) Here, the pancreas transplant is performed after the kidney transplant in a separate operation. The benefits of preemptive (before the initiation of dialysis) kidney transplantation have been well established (10,11). When a uremic diabetic presents for transplant consideration, they are candidates for either a SPK or a living donor kidney transplant followed by a PAK. The approach varies depending on the region of the country and the availability of a living kidney donor. Type I diabetics on dialysis have a waitlist mortality of almost 10% per year, which is higher than that for people with other causes of ESRD (12). If a living donor is available, than kidney transplant should be performed as soon as possible, as outcomes after transplant worsen in direct proportion to time spent on dialysis (11). Some regions of the United States allocate waitlist priority to diabetics listed for SPK (over kidney alone candidates) and in these areas, SPK may be the better option. Pancreas Transplant Alone (PTA) The fewest pancreas transplants are performed in this category. A select group of non-uremic, type I diabetics who have failed insulin therapy can be considered for solitary pancreas transplantation. These patients usually have severe and life-threatening metabolic complications, such as hypoglycemia unawareness, justifying the risk of surgery and immunosuppression.

categories of pancreas transplant
Pancreas transplantation is divided into three recipient categories as follows: Simultaneous Kidney and Pancreas Transplant (SPK) Pancreas transplants are most often performed in conjunction with a simultaneous kidney transplant (Fig. 50.2). About 10% of Type I diabetics will develop end stage renal disease (ESRD) during their lifetime (4,5) and over 40% of ESRD cases in the United States are due to diabetes (6). If a type I diabetic is a candidate for a kidney transplant, it is logical to consider them for a simultaneous pancreas transplant. Thus, two goals are accomplished (freedom from dialysis and normalization of glucose metabolism without exogenous insulin therapy) with one surgical procedure and one round of induction immunosuppresssion. As the two organs are from the same deceased donor, the kidney acts as a surrogate marker for pancreas rejection, allowing for better immune monitoring. A variant of SPK is a simultaneous pancreas and living-donor kidney transplant (SPLK). This technique, popularized by a few

indications for pancreas transplant
The goals of pancreas transplantation are to improve quality of life, normalize glucose metabolism, arrest or reverse end organ damage from diabetes, and protect the transplanted kidney from the effects of hyperglycemia. It is mainly indicated for uremic type I diabetics. However, successful outcomes have been reported with pancreas transplants in type II diabetics (13), but this group accounts for less than 10% of SPK transplants. The American Diabetes Association (ADA) criteria for solitary pancreas transplant (PTA) in type I diabetics are (1) a history of frequent, acute, and severe metabolic complications such as hypoglycemia, hyperglycemia, ketoacidosis; (2) incapacitating clinical and emotional problems with exogenous insulin therapy; and (3) consistent failure of insulin-based management to prevent acute complications (14). Pancreas transplant is a reasonable option for patients with traumatic, inflammatory, or surgical loss of pancreatic function and has the added advantage in these cases of restoring exocrine function (15). Transplantation is contraindicated in the setting of pancreatic malignancy.

470

PANCREAS TRANSPLANTATION
SPK 3000 Number of new registrations PAK PTA All pancreas

2000

1000

Patients with symptomatic coronary artery disease or stenoses greater than 75% should undergo pretransplant revascularization (19). All candidates should have optimal management of hypertension, hyperlipidemias, and be counseled on weight loss and smoking cessation if indicated. A BMI > 30 kg/m2 is an independent risk factor for technical graft failure (20).

the pancreas donor
0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 50.1 New registrations on pancreas waiting list by transplant type, 1995–2004. Data from 2007 OPTN/SRTR Annual Report. Accessed from www.ustransplant.org December 2008. Abbreviations: SPK, simultaneous kidney and pancreas; PAK, pancreas after kidney; PTA, pancreas transplant alone.

SPK 1600 Number of transplants

PAK

PTA

All pancreas

1200 800

400

0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 50.2 Pancreas transplants in the United States by transplant type, 1997–2006. Data from 2007 OPTN/SRTR Annual Report. Accessed from www.ustransplant.org December 2008.

the pancreas recipient
Pretransplant evaluation of the potential pancreas transplant recipient is geared toward excluding occult infection, malignancy, and coronary artery disease. After a thorough history and physical exam, screening for CMV, EBV, hepatitis B and C, HIV, syphilis, and tuberculosis is performed and vaccinations are updated. Pancreas transplantation should not be performed in the setting of active hepatitis B or C or if there is underlying cirrhosis. HIV-positive patients who have no history of opportunistic infections, preserved CD4 counts and low viral loads can be considered for transplantation (16). Minimum screening for malignancy should include a CXR, PAP smear for all women, mammogram for women over 40 years, PSA for men over 50 years, and a colonoscopy for those over 50 years. Active malignancy is an absolute contraindication to transplant, and a disease-free period of 2 to 5 years, depending upon the type and stage of the cancer, is recommended prior to transplant (17). Diabetics frequently have asymptomatic coronary artery disease, and should undergo baseline exercise or pharmacologic stress testing. However, the sensitivities of these tests vary and, in one study, 20% of patients with a negative dobutamine stress echo had a cardiac event posttransplant (18). Some centers routinely perform coronary angiography in all candidates, although this may be detrimental to those not yet on dialysis.

The majority of transplanted pancreata are from deceased donors who meet criteria for brain death. Selective use of pancreata from donors after cardiac death (DCD) is increasing, with no compromise in outcomes (21). Living donor pancreas transplants are not widely performed. The distal half of the donor pancreas, based on the splenic vessels, is removed for implantation, as a segmental graft (22). However, given the donor risk, the relatively short waiting times for solitary pancreas transplants and the fact that these transplants are not necessarily “life-saving,” use of live donors is hard to justify (23). All potential donors undergo a thorough medical evaluation and are screened for transmissible diseases such as HIV and hepatitis B and C. The presence of active or recent malignancy is generally a contraindication to pancreas donation. Pancreata from donors with a BMI greater than 30 kg/m2, age greater than 45 years and a cerebrovascular cause of death have higher technical failure rates after transplant (20), and should be used very selectively. Hyperglycemia, per se, does not contraindicate donation, as it is often related to donor stress or use of vasoactive drugs or steroids. Organs from pediatric donors (3–11 years) have been used for SPK transplants with good long-term results (24). A donor risk index has been developed and will allow more informed selection of donors and matching with potential recipients (25).

surgical techniques
Procurement The various methods for recovery of the pancreas from the deceased donor have been well described (26–28). Often, multiple recovery teams are present, and careful coordination between them is of paramount importance. Exposure is through a long midline or cruciate incision. The infrarenal and supraceliac aorta are controlled and looped and the pancreas is inspected through the lesser sac. The gland is assessed for quality, in particular, presence of edema, injury, fibrosis, and fat content. Pancreas dissection can be performed “in the warm” (prior to crossclamp) or “in the cold’” (after crossclamp); other surgeons prefer some form of en bloc recovery technique, with separation of the pancreas from the liver ex situ (29,30). The supraceliac aorta is clamped and the infrarenal aorta flushed with preservation solution. Various solutions are available and in the United States, typically University of Wisconsin (UW) or Histidine–Tryptophan–Ketoglutarate (HTK) solutions are used; however, recent studies suggest that the rate of posttransplant pancreatitis and graft failure is higher with HTK use (31,32). After flush, the pancreas is mobilized, using the spleen as a handle. The duodenum is maintained intact with the gland and the mesenteric root stapled and divided

471

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
away from the pancreatic head. The superior mesenteric artery (SMA) and splenic artery (SA) are kept with the pancreas, with the rest of the celiac axis reserved for the liver graft. The portal vein is divided at the midhilar level, with the distal half providing venous outflow for the pancreas. A donor iliac artery is retrieved, with its bifurcation, for subsequent vascular reconstruction. Living-donor pancreas procurement can be performed through an open or laparoscopic approach (22,33). The body and tail of the gland are mobilized while preserving the gastroepiploic arcade to allow preservation of the spleen. The splenic vessels are ligated at the splenic hilum. The gland is divided at the left border of the superior mesenteric vein (SMV), and the pancreatic duct identified. The splenic artery is divided just distal to its celiac origin and the splenic vein at its confluence with the SMV, the gland flushed and submitted to the recipient team for implantation.

Figure 50.3 A pancreas allograft procured from a deceased donor, with the duodenum and spleen intact.

backtable preparation
A carefully performed backtable preparation is probably the single most important technical aspect of pancreas transplantation. All work is carried out in a basin of preservative solution, cooled to about 4°C. The graft is received with the spleen and duodenum intact (Figs. 50.3 and 50.4). The spleen is removed by ligating the splenic vessels close to the distal end of the pancreas. Loose, fatty tissue on the upper and lower borders of the pancreas is carefully tied, the inferior mesenteric vein is ligated and the staple lines on the mesenteric stump and duodenal ends oversewn. On the posterior surface of the gland, the open ends of the PV, SMA, and SA are identified and any surrounding lymphatic and ganglionic tissue removed. The pancreatic head and duodenum will derive arterial supply from the SMA via the inferior pancreaticoduodenal arcade and the body and tail of the gland will be supplied by the SA. This dual supply is united by means of a “Y-graft” fashioned from the donor iliac vessels (Fig. 50.5), enabling a single arterial anastomosis during the recipient operation.

Figure 50.4 A procured pancreas allograft. The portal vein (PV), superior mesenteric artery (SMA), and splenic artery (SA) are preserved with the graft. The duodenal ends and the root of the small bowel mesentery are transected with a stapling device.

recipient operations
The recipient operation varies with respect to graft placement (intra- vs. extraperitoneal), venous drainage (systemic vs. portal), and exocrine drainage (bladder vs. enteric). Most grafts are placed intraperitoneally, although a few centers prefer extraperitoneal placement, similar to the approach used for kidney transplantation (34,35). Intraperitoneal placement allows placement of the kidney through the same incision in SPK transplants and may be associated with less wound problems (36). Systemic Venous Drainage Most pancreas transplants are drained systemically, usually into the iliac vein, on the right side (2). A lower midline incision is made and the cecum and right colon mobilized medially to expose the iliac vessels and ureter. The iliac vessels are completely mobilized, and all hypogastric venous branches ligated to allow full mobilization of the iliac vein and to reduce tension on the subsequent venous anastomosis. The graft is then brought into the operative field, with the head and duodenum directed caudally. The portal vein is anastomosed

Figure 50.5 Pancreas allograft after backtable preparation. The spleen is removed and the distal splenic vessels ligated. The inferior mesenteric vein is ligated. The duodenal staple lines are oversewn. A donor iliac artery “Y graft” is anastomosed to the SMA and SA.

472

PANCREAS TRANSPLANTATION

Figure 50.6 Transplanted[u1] pancreas allograft, just after reperfusion.

Figure 50.8 Pancreas transplant with portal venous drainage and enteric anastomosis. The graft PV is anastomosed, end-to-side, to the recipient superior mesenteric vein. The graft duodenum points cephalad.

Figure 50.7 A SPK transplant. The graft PV is anastomosed to the recipient right common iliac vein and the Y-graft to the recipient right common iliac artery. A side-to-side enteric anastomosis is depicted. The kidney is transplanted to the left external iliac vessels and a standard ureteroneocystostomy is constructed.

in an end-to-side fashion to the iliac vein and the long limb of the “Y-graft” to the iliac artery and the graft reperfused (Fig. 50.6). The graft duodenum is then drained into the bladder or bowel. Alternatively, the graft can be placed with the head cephalad, with venous drainage directly into the cava (37); this, however, precludes the option of bladder drainage. In an SPK transplant, the kidney is placed on the opposite side, either before or after the pancreas (Fig. 50.7).

Portal Venous Drainage About 20% to 30% of all pancreas transplants are currently drained into the portal venous system (2). Its proponents claim that this approach is more physiologic and associated with lower rates of rejection (38). Indeed, systemic transplants are associated with hyperinsulinemia, but the significance of this is unclear (39). Disadvantages of portal drainage are a graft that is less amenable to percutaneous biopsy and the inability to perform bladder exocrine drainage. However, registry data show no difference in graft survival between the two techniques (2). The author uses both methods, preferring portal drainage if a difficult pelvic dissection is anticipated or there is a prior kidney transplant on the right side. The approach is through an upper midline incision, and the SMV is exposed at the root of the small bowel mesentery and dissected as far proximally as possible. The right common iliac artery is exposed through a “window” in the mesentery. The graft is placed with the duodenum directed superiorly, and the graft portal vein anastomosed, end-to-side, to the SMV. The “Y-graft,” which is kept long, is anastomosed to the iliac artery through the mesenteric window. The graft is then reperfused and the enteric anastomosis completed (Fig. 50.8). Bladder Drainage Management of exocrine drainage has been the Achilles heel of pancreas transplantation. Over the years, numerous

473

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Surgical Complications Early surgical complications are particularly relevant after pancreas transplantation as they often lead to graft loss. Common, pancreas-specific complications are thrombosis, hemorrhage, infection/pancreatitis, and anastomotic leak. Graft thrombosis accounts for over 70% of technical failures (44) and can be arterial or venous. The incidence varies with transplant category and technique and ranges from 5% to 11% (2,44). Risk factors for thrombosis include older donor age, cerebrovascular cause of donor death, prolonged preservation time, retransplantation, segmental grafts, and non-standard vascular reconstruction (49). Rarely, with early surgical intervention, a graft may be salvaged after thrombosis (50). Due to improved antibiotic prophylaxis, better surgical techniques and more “targeted” immunosuppression, the incidence of perigraft infections is decreasing (2,44). Most infections will occur in the first few weeks posttransplant and are treated aggressively with antibiotics, antifungals, and surgical washout if necessary. Anastomotic leak rates are around 5% to 10% and tend to be higher with bladder-drained grafts (51,52). The most important risk factor for leaks is preservation time; when it exceeds 24 hours, leak rates as high as 25% have been reported (53). Bladder leaks can often be managed with drainage alone, whereas enteric leaks usually require aggressive surgical intervention.

Figure 50.9 Pancreas transplant with systemic venous drainage and exocrine drainage into the bladder. The graft duodenum points caudad.

techniques have been tried, including free drainage of the pancreatic duct into the peritoneum (40) and obliteration of the pancreatic duct with polymer injection (41,42). Currently, bladder or enteric drainage is the standard techniques. Bladder drainage of pancreatic exocrine secretions was first described in 1984 (43) and was the predominant technique in the 1990s (44). The graft duodenum is anastomosed directly to the dome of the bladder, using absorbable sutures (Fig. 50.9). A stapled technique, using an EEA device, has also been described (45). The main advantage of this technique is the early safety (small leaks, e.g., can be managed simply by catheter drainage of the bladder) and less technical failure when compared to enterically drained grafts (2). Urinary amylase output can also be monitored and this is a sensitive marker for pancreatic rejection (46). However, the technique is fraught with long-term complications, such as dehydration, metabolic acidosis, recurrent urinary tract infections, graft pancreatitis, and hematuria (47). About 20% to 30% of bladder drained grafts have to be converted to enteric drainage due to these complications (48). Enteric Drainage Most centers currently prefer enteric drainage, to avoid the metabolic complications associated with bladder drainage (2). The availability of better immunosuppressive agents and the lower rates of acute rejection have lessened the role of urinary amylase monitoring. The graft duodenum is anastomosed to a loop of proximal small bowel, either directly (Fig. 50.7) or using a Roux-en-Y technique. Technical complications are actually higher when a Roux limb is constructed (44). If the graft duodenum does not perfuse well or the ischemic time is long, it may be more prudent to use a Roux or perform bladder drainage.

outcomes
Historically, SPK transplant recipients have enjoyed much better pancreas graft survival than PAK and SPK recipients (54). Over the last decade, the 1-year survival for solitary grafts has been steadily improving. Current 1-year graft survivals for SPK, PAK, and PTA transplants are 86%, 79%, and 80%, respectively (Fig. 50.10) (3). At 10 years, 51% of SPK recipients still have functioning grafts, compared to only 28% of PAK recipients and 24% of PTA recipients (55). Chronic rejection and death with a functioning graft are responsible for most cases of late graft loss (56). Patient survival is similar for the three categories, and ranges from 95% to 97% at 1 year and 64% to 71% at 10 years (Fig. 50.11) (3). Recent studies have clearly demonstrated a survival advantage with SPK transplants (57–59). However, one analysis of registry data suggested that survival after pancreas transplant was worse with solitary pancreas transplants, when compared to waitlisted people who received conventional therapy for diabetes (58). Some patients were registered on several waitlists and counted more than once, and patients who dropped off the list for medical reasons were censored, biasing the outcomes of this study. When these differences were accounted for, there was no difference in survival at 4 years between waitlisted patients and those receiving PAK or PTA transplants (level of evidence: III) (59).

effects on secondary complications of diabetes
To date, there have been no prospective, randomized trials comparing the efficacy of pancreas transplant versus medical therapy. Moreover, most recipients already have advanced secondary complications at the time of transplant, making it

474

PANCREAS TRANSPLANTATION
Unadjusted graft survival (%)
100% 80% 60% 40% 20% 0% 1-Year 3-Year 5-Year 10-Year SPK PTA PAK

and a lower incidence of death from cardiovascular causes compared to waitlisted patients (73). Several studies have demonstrated improvement in quality of life after pancreas transplant, using various survey instruments (level of evidence IIa to III) (76).

immunosuppression
Up to 80% of pancreas recipients receive some form of induction therapy at the time of transplant, usually a T-cell depleting agent such as thymoglobulin (77). Standard maintenance regimen consists of a calcineurin inhibitor (tacrolimus or cyclosporine), an antimetabolite (mycophenolate mofetil or sirolimus), and steroids. Outcomes with tacrolimus-based therapy are better overall (78,79) but one must be wary of the potential of nephrotoxicity with this agent (80). Steroid avoidance and steroid withdrawal protocols are also increasingly used, with good short-term results (81). Acute rejection rates in the first year after pancreas transplant are currently less than 25% for SPK transplants and somewhat higher for solitary transplants (78). A rise in serum amylase or lipase, a fall in urinary amylase output (for bladderdrained grafts), and hyperglycemia are suggestive of acute rejection, and should prompt a biopsy. The histologic features for diagnosis and grading acute pancreas allograft rejection were recently standardized (82).

Figure 50.10 Unadjusted pancreas graft survival by transplant type. Data from 2007 OPTN/SRTR Annual Report. Accessed from www.ustransplant.org December 2008.

Unadjusted patient survival (%)

100% 80% 60% 40% 20% 0% 1-Year 3-Year

SPK

PTA

PAK

5-Year

10-Year

Figure 50.11 Unadjusted pancreas patient survival by transplant type. Data from 2007 OPTN/SRTR Annual Report. Accessed from www.ustransplant.org December 2008.

conclusions
Combined kidney and pancreas transplant is an effective option for uremic patients with type I diabetes. A select group of type I diabetics, who have normal renal function, may benefit from a solitary pancreas transplant. In the current era, graft survivals are comparable to those of other solid organ transplants. There is much enthusiasm over islet cell transplantation as a less invasive alternative, but, at present, the results are not as durable as those of whole organ transplant (83). No doubt, both fields will continue to evolve, with ongoing advances in techniques and immunosuppression.

difficult to demonstrate benefit after transplant. The Diabetes Control and Complication Trial (DCCT) clearly showed the benefits of normalizing blood glucose concentrations in diabetics, with an almost 50% reduction in retinopathy, neuropathy, and nephropathy (60). Several, small, non-randomized, but controlled studies provide evidence (level IIa) of the beneficial effects of transplant on secondary diabetic complications. Reversal of diabetic nephropathy was seen 10 years (but not at 5 years) after PTA, with decreased thickness of glomerular and tubular basement membranes and decreased mesangial fraction volume (61). Most studies investigating the effects of transplantation on diabetic retinopathy are limited by the presence of advanced disease at baseline. Proteinuria also decreases after successful PTA when compared to matched controls at 1-year posttransplant (62). Stabilization of retinopathy after pancreas transplantation has been well demonstrated (63–65) with less disease progression over time in comparison to controls (66). There is a sustained improvement in nerve conduction velocity and action potential amplitude after both SPK (67,68) and solitary pancreas transplantation (67). Recipients of SPK transplants have less progression of coronary atherosclerosis, (69) a better atherosclerotic risk profile (70), and significant improvement in cardiac geometry and function in comparison to kidney-alone recipients (71–73). There is improvement in systolic and diastolic blood pressure (74,75) and other cardiovascular parameters (75) after combined pancreas–kidney transplantation

references
1. Kelly WD, Lillehei RC, Merkel FK, et al. Allotransplantation of the pancreas and duodenum along with the kidney in diabetic nephropathy. Surgery 1966; 61: 827–37 2. Gruessner AC, Sutherland DE. Pancreas transplant outcomes for United States (US) and non-US cases as reported to the United Network for Organ Sharing (UNOS) and the International Pancreas Transplant Registry (IPTR) as of June 2004. Clin Transplant 2005; 19: 433–55 3. Leichtman AB, Cohen D, Keith D, et al. Kidney and pancreas transplantation in the United States, 1997–2006: the HRSA Breakthrough Collaboratives and the 58 DSA Challenge. Am J Transplant 2008; 8(4 Pt 2): 946–57. 4. Finne P, Reunanen A, Stenman S, et al. Incidence of end-stage renal disease in patients with type 1 diabetes. JAMA 2005; 294(14): 1782–7 5. Nishimura R, Dorman JS, Bosnyak Z, et al. Incidence of ESRD and survival after renal replacement therapy in patients with type 1 diabetes: a report from the Allegheny County Registry. Am J Kidney Dis. 200; 42(1): 117–24. 6. Foley RN, Collins AJ. End-stage renal disease in the United States: an update from the United States Renal Data System. J Am Soc Nephrol 2007; 18(10): 2644–8. 7. Farney AC, Cho E, Schweitzer EJ, et al. Simultaneous cadaver pancreas living-donor kidney transplantation: a new approach for the type 1 diabetic uremic patient. Ann Surg 2000; 232(5): 696–703.

475

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
8. Boggi U, Vistoli F, Del Chiaro M, et al. Simultaneous cadaver pancreasliving donor kidney transplantation. Transplant Proc 2004; 36(3): 577–9. 9. Gruessner RW, Kendall DM, Drangstveit MB, et al. Simultaneous pancreas-kidney transplantation from live donors. Ann Surg 1997; 226(4): 471–80 10. Becker BN, Rush SH, Dykstra DM, et al. Preemptive transplantation for patients with diabetes-related kidney disease. Arch Intern Med 2006; 166(1): 44–8. 11. Meier-Kriesche HU, Schold JD. The impact of pretransplant dialysis on outcomes in renal transplantation. Semin Dial 2005; 18(6): 499–504. 12. United States Renal Data System (USRDS): 2008 Annual Data Report. www.usrds.org Accessed December 2008. 13. Nath DS, Gruessner AC, Kandaswamy R, et al. Outcomes of pancreas transplants for patients with type 2 diabetes mellitus. Clin Transplant 2005; 19(6): 792–7. 14. Robertson P, Davis C, Larsen J, Stratta R, Sutherland DE; American Diabetes Association. Pancreas transplantation in type 1 diabetes. Diabetes Care 2004; 27 Suppl 1: S105 15. Gruessner RW, Manivel C, Dunn DL, Sutherland DE. Pancreaticoduodenal transplantation with enteric drainage following native total pancreatectomy for chronic pancreatitis: a case report. Pancreas 1991; 6(4): 479–88. 16. Toso C, Berney T, Oberholzer J, et al. Kidney-pancreas transplantation in a long-term non-progressor HIV-infected recipient. Am J Transplant 2003; 3(5): 631–3. 17. Buell JF, Beebe TM, Trofe J, et al. Donor transmitted malignancies. Ann Transplant 2004; 9(1): 53–6. 18. Herzog CA, Marwick TH, Pheley AM, et al. Dobutamine stress echocardiography for the detection of significant coronary artery disease in renal transplant candidates.Am J Kidney Dis 1999; 33(6): 1080–90. 19. Manske CL, Wang Y, Rector T, Wilson RF, White CW. Coronary revascularisation in insulin-dependent diabetic patients with chronic renal failure. Lancet 1992; 340(8826): 998–1002. 20. Humar A, Ramcharan T, Kandaswamy R, et al. Technical failures after pancreas transplants: why grafts fail and the risk factors–-a multivariate analysis. Transplantation 2004; 78(8): 1188–92. 21. Salvalaggio PR, Davies DB, Fernandez LA, Kaufman DB. Outcomes of pancreas transplantation in the United States using cardiac-death donors. Am J Transplant 2006; 6(5 Pt 1): 1059–65. 22. Sutherland DE, Gruessner R, Dunn D, et al. Pancreas transplants from living-related donors. Transplant Proc 1994; 26(2): 443–5. 23. Kumar AF, Gruessner RW, Seaquist ER. Risk of glucose intolerance and diabetes in hemipancreatectomized donors selected for normal preoperative glucose metabolism. Diabetes Care 2008; 31(8): 1639–43. 24. Fernandez LA, Turgeon NA, Odorico JS, et al. Superior long-term results of simultaneous pancreas-kidney transplantation from pediatric donors. Am J Transplant 2004; 4(12): 2093–101. 25. Axelrod DA, Sung RS, Meyer KH, et al. Systematic evaluation of pancreas allograft quality, outcomes and geographic variation in utilization. Am J Transplant 2010; 10(4): 837–45. 26. Starzl TE, Hakala TR, Shaw BW Jr., et al. A flexible procedure for multiple cadaveric organ procurement. Surg Gynecol Obstet 1984; 158(3): 223–30. 27. Delmonico FL, Jenkins RL, Auchincloss H Jr., et al. Simultaneous procurement of pancreas and liver from a single cadaveric donor. Transplant Proc 1989; 21(3): 3521. 28. C Marsh CL, Perkins JD, Sutherland DE, et al. Combined hepatic and pancreaticoduodenal procurement for transplantation. Surg Gynecol Obstet 1989; 168(3): 254–8. 29. Boggi U, Vistoli F, Del Chiaro M, et al. A simplified technique for the en bloc procurement of abdominal organs that is suitable for pancreas and small-bowel transplantation. Surgery 2004; 135(6): 629–41. 30. Moon JI, Nishida S, Butt F, et al. Multi-organ procurement and successful multi-center allocation using rapid en bloc technique from a controlled non-heart-beating donor. Transplantation 2004; 77(9): 1476–7. 31. Alonso D, Dunn TB, Rigley T, et al. Increased pancreatitis in allografts flushed with histidine-tryptophan-ketoglutarate solution: a cautionary tale. Am J Transplant 2008; 8(9): 1942–5. 32. Stewart ZA, Cameron AM, Singer AL, et al. Histidine-tryptophan-ketoglutarate (htk) is associated with reduced graft survival in pancreas transplantation. Am J Transplant. Early online view, 2008 Nov 4. 33. Tan M, Kandaswamy R, Sutherland DE, Gruessner RW. Laparoscopic donor distal pancreatectomy for living donor pancreas and pancreaskidney transplantation. Am J Transplant 2005; 5(8): 1966–70. 34. Adamec M, Janousek L, Saudek F, Tosenovský P. 100 pancreas transplantations with extraperitoneal graft placement. Ann Transplant 2001; 6(2): 41–2. 35. Boggi U, Vistoli F, Signori S, et al. Outcome of 118 pancreas transplants with retroperitoneal portal-enteric drainage. Transplant Proc 2005; 37(6): 2648–50. 36. Schweitzer EJ, Bartlett ST. Wound complications after pancreatic transplantation through a kidney transplant incision. Transplant Proc 1994; 26(2): 461. 37. Fridell JA, Shah A, Milgrom ML, et al. Ipsilateral placement of simultaneous pancreas and kidney allografts. Transplantation 2004; 78(7): 1074–6. 38. Philosophe B, Farney AC, Schweitzer EJ, et al. Superiority of portal venous drainage over systemic venous drainage in pancreas transplantation: a retrospective study. Ann Surg 2001; 234(5): 689–96. 39. Diem P, Abid M, Redmon JB, et al. Systemic venous drainage of pancreas allografts as independent cause of hyperinsulinemia in type I diabetic recipients. Diabetes 1990; 39(5): 534–40. 40. Sutherland DE, Baumgartner D, Najarian JS. Free intraperitoneal drainage of segmental pancreas grafts: clinical and experimental observations on technical aspects. Transplant Proc 1980; 12(4 Suppl 2): 26–32. 41. Dubernard JM, Traeger J, Neyra P, et al. A new method of preparation of segmental pancreatic grafts for transplantation: trials in dogs and in man. Surgery 1978; 84(5): 633–9. 42. Little JM, Lauer C, Hogg J. Pancreatic duct obstruction with an acrylate glue: a new method for producing pancreatic exocrine atrophy. Surgery 1977; 81(3): 243–9. 43. Sollinger HW, Cook K, Kamps D, et al. Clinical and experimental experience with pancreaticocystostomy for exocrine pancreatic drainage in pancreas transplantation. Transplant Proc 1984; 16(3): 749–51. 44. Gruessner AC, Sutherland DE. Pancreas transplant outcomes for United States (US) cases reported to the United Network for Organ Sharing (UNOS) and non-US cases reported to the International Pancreas Transplant Registry (IPTR) as of October, 2000. Clin Transpl 2000: 45–72. 45. Pescovitz MD, Dunn DL, Sutherland DE. Use of the circular stapler in construction of the duodenoneocystostomy for drainage into the bladder in transplants involving the whole pancreas. Surg Gynecol Obstet 1989; 169(2): 169–71. 46. Prieto M, Sutherland DE, Fernandez-Cruz L, et al. Experimental and clinical experience with urine amylase monitoring for early diagnosis of rejection in pancreas transplantation. Transplantation 1987; 43(1): 73–9. 47. Baktavatsalam R, Little DM, Connolly EM, et al. Complications relating to the urinary tract associated with bladder-drained pancreatic transplantation. Br J Urol 1998; 81(2): 219–23. 48. West M, Gruessner AC, Metrakos P, et al. Conversion from bladder to enteric drainage after pancreaticoduodenal transplantations. Surgery 1998; 124(5): 883–93. 49. Humar A, Kandaswamy R, Drangstveit MB, et al. Surgical risks and outcome of pancreas retransplants. Surgery 2000; 127(6): 634–40. 50. Boggi U, Vistoli F, Signori S, et al. Surveillance and rescue of pancreas grafts. Transplant Proc 2005; 37(6): 2644–7. 51. Gruessner RW, Sutherland DE, Troppmann C, et al. The surgical risk of pancreas transplantation in the cyclosporine era: an overview. J Am Coll Surg 1997; 185(2): 128–44. 52. Sollinger HW, Messing EM, Eckhoff DE, Pirsch JD, et al. Urological complications in 210 consecutive simultaneous pancreas-kidney transplants with bladder drainage. Ann Surg 1993 Oct; 218(4): 561–8 53. Humar A, Kandaswamy R, Drangstveit MB, et al. Prolonged preservation increases surgical complications after pancreas transplants. Surgery 2000; 127(5): 545–51. 54. Gruessner A, Sutherland DE. Pancreas transplant results in the United Network for Organ Sharing (UNOS) United States of America (USA) Registry compared with non-USA data in the International Registry. Clin Transpl 1994: 47–68. 55. Sutherland DE, Gruessner AC. Long-term results after pancreas transplantation. Transplant Proc 2007; 39(7): 2323–5.

476

PANCREAS TRANSPLANTATION
56. Humar A, Khwaja K, Ramcharan T, et al. Chronic rejection: the next major challenge for pancreas transplant recipients. Transplantation 2003; 76(6): 918–23. 57. Ojo AO, Meier-Kriesche HU, Hanson JA, et al. The impact of simultaneous pancreas-kidney transplantation on long-term patient survival. Transplantation 2001; 71(1): 82–90. 58. Venstrom JM, McBride MA, Rother KI, et al. Survival after pancreas transplantation in patients with diabetes and preserved kidney function. JAMA 2003; 290(21): 2817–23. 59. Gruessner RW, Sutherland DE, Gruessner AC. Mortality assessment for pancreas transplants. Am J Transplant 2004; 4(12): 2018–26. 60. The Diabetes Control and Complications Trial Research Group. the effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329: 977–86 61. Fioretto P, Steffes MW, Sutherland DER, et al. Reversal of lesions of diabetic nephropathy after pancreas transplantation. N Engl J Med 1998; 339; 69–75. 62. Coppelli A, Giannarelli R, Boggi U, et al. Disappearance of nephrotic syndrome in type 1 diabetic patients following pancreas transplant alone. Transplantation 2006; 81(7): 1067–8. 63. Konigsrainer A, Miller K, Steurer W, et al. Does pancreas transplantation influence the course of diabetic retinopathy? Diabetologia 1991; 34(S1): S86–8. 64. Koznarova R, Saudek F, Sosna T, et al. Beneficial effect of pancreas and kidney transplantation on advanced diabetic retinopathy. Cell Transplant 2000; 9: 903–8. 65. Giannarelli R, Copelli A, Sartini MS, et al. Early improvement of unstable diabetic retinopathy after solitary pancreas transplantation. Diabetes Care 2002; 25: 2358–9. 66. Ramsey RC, Goetz FC, Sutherland DE, et al. Progression of diabetic retinopathy after pancreas transplantation for insulin-dependent diabetes mellitus. N Engl J Med 1998; 318: 208–14. 67. Navarro X, Sutherland DE, Kennedy WR. Long-term effects of pancreatic transplantation on diabetic neuropathy. Ann Neurol 1997; 42: 727–36. 68. Allen RD, Al-Harbi IS, Morris JG, et al. Diabetic neuropathy after pancreas transplantation: determinants of recovery. Transplantation 1997; 63: 830–8. 69. Jukema J, Wouter MD, Smets YF, et al. Impact of simultaneous pancreas and kidney transplantation on progression of coronary atherosclerosis in patients with end-stage renal failure due to type 1 diabetes. Diabetes Care 2002; 25: 906–11. 70. Fiorina P, La Rocca E, Venturini M, et al. Effects of kidney-pancreas transplantation on atherosclerotic risk factors and endothelial function in patients with uremia and type 1 diabetes. Diabetes 2001; 50(3): 496–501. 71. Gaber AO, Wicks MN, Hathaway DK, Burlew BS. Sustained improvement in cardiac geometry and function following kidney-pancreas transplantation. Cell Transplant 2000; 9: 913–8. 72. Fiorina P, La Rocca E, Astorri E, et al. Reversal of left ventricular diastolic dysfunction after kidney-pancreas transplantation in type 1 diabetic uremic patients. Diabetes Care 2000; 23: 1804–10. 73. La Rocca E, Fiorina P, Di Carlo V, et al. Cardiovascular outcomes after kidney-pancreas and kidney-alone transplantation. Kid Int 2001; 60: 1964–71. 74. Elliot MD, Kapoor A, Parker MA, et al. Improvement in Hypertension in patients with diabetes mellitus after kidney/pancreas transplantation. Circulation 2001; 104: 563–9. 75. Coppelli A, Giannarelli R, Mariotti R, et al. Pancreas Transplant alone determines early improvement of cardiovascular risk factors and cardiac function in type 1 diabetic patients. Transplantation 2003; 76: 974–6. 76. Joseph JT, Baines LS, Morris MC, Jindal RM. Quality of life after kidney and pancreas transplantation: a review. Am J Kidney Dis 2003; 42(3): 431–45. 77. Meier-Kriesche HU, Li S, Gruessner RW, Fung JJ, et al. Immunosuppression: evolution in practice and trends, 1994–2004. Am J Transplant 2006; 6(5 Pt 2): 1111–31. 78. Sutherland DE, Gruessner RW, Dunn DL, et al. Lessons learned from more than 1,000 pancreas transplants at a single institution. Ann Surg 2001 Apr; 233(4): 463–501. 79. Saudek F, Malaise J, Boucek P, Adamec M; Euro-SPK Study Group. Efficacy and safety of tacrolimus compared with cyclosporin microemulsion in primary SPK transplantation: 3-year results of the Euro-SPK 001 trial. Nephrol Dial Transplant 2005; 20(Suppl 2): ii3–10, ii62. 80. Ojo AO. Renal disease in recipients of nonrenal solid organ transplantation. Semin Nephrol 2007; 27(4): 498–507 81. Cantarovich D, Vistoli F. Minimization protocols in pancreas transplantation. Transpl Int 2008 Aug 15. 82. Drachenberg CB, Odorico J, Demetris AJ, et al. Banff schema for grading pancreas allograft rejection: working proposal by a multi-disciplinary international consensus panel. Am J Transplant 2008; 8(6): 1237–49. 83. Shapiro AM, Ricordi C, Hering BJ, et al. International trial of the Edmonton protocol for islet transplantation. N Engl J Med 2006; 355(13): 1318–30.

477

51

Pediatric HPB disorders Maureen McEvoy and Michael P. La Quaglia
Clinical Features Signs and symptoms include jaundice, clay-colored stools, and hepatomegaly. In infancy, jaundice that persists beyond 2 weeks is no longer considered physiologic. The differential for neonatal cholestasis is presented in Table 51.1. The first step in diagnosis in an infant is to identify conjugated hyperbilirubinemia with prolonged jaundice, pale stools, or dark urine. A conjugated bilirubin greater than 20% of an elevated total serum bilirubin is diagnostic of homeostasis. Table 51.2 lists the diagnostic evaluation appropriate for the homeostatic neonate. Several authors consider liver biopsy to be the most reliable test for establishing the diagnosis (10,11). Liver biopsy can correctly predict extra hepatic biliary obstruction in more than 90% of cases (1,12). Ultrasound is safe and noninvasive and should be performed in all jaundiced infants as in initial evaluation. Treatment The surgical procedure of choice is the HPE, first described by Kasai in 1959 (13). The procedure has three components: (1) abdominal exploration and cholangiography to confirm the diagnosis, (2) dissection of the porta hepatis and transection of biliary tissue remnants at the portal plate, and (3) establishment of biliary drainage through the construction of a 40 to 50 cm Roux-en-Y jejunal conduit. Success in obtaining bile flow is better if HPE is done when the patient is between 60 and 90 days old. The sequential use of HPE followed by liver transplantation is the standard surgical paradigm followed around the world (14–16). Outcomes Left untreated, an infant with BA has a life expectancy of approximately 1 year. Following a Kasai procedure, 5-year survival rates with native liver have ranged between 48% and 60%. With the sequential treatment of a Kasai portoenterostomy and secondary liver transplant, if required, overall survival rate is approximately 90% (17). Outcomes of BA following Kasai HPE and liver transplant from various centers are summarized in Table 51.3. Survival with the native liver after the Kasai operation is approximately 30% at 10 years and 14% to 23% at 20 years (18–21). The largest North American long-term experience with HPE demonstrated survival with the native liver of 35% at 10 years and 21% at 20 years (22). A recent series of 755 BA patients listed for liver transplantation from North America reported a 3-year graft survival rate of 88% and a patient survival rate of 80% (23). Thus the current use of HPE followed by liver transplantation in children who subsequently develop cirrhosis provides excellent long-term survival for a disease that is fatal without surgery. Complications include cholangitis, cessation of bile flow, portal hypertension, intrahepatic cysts, hepatopulmonary syndrome, and others.

introduction
Advances in surgical technique have led various treatment options for patients with hepatobiliary and pancreatic disorders. There have also been advances in the understanding of pathogenesis and clinical behavior of specific diseases. This chapter addresses the common hepatobiliary and pancreatic disorders with a focus on surgical treatment modalities.

biliary atresia ⁽ba⁾

Classification BA is an obstructive condition of the bile ducts. It is of unknown etiology but results from a progressive obliterative process of variable extent. It has a worldwide incidence of 1 in 5000 to 18,000 live births, and it is more common in girls than in boys (1,2). Two clinical forms are described: acquired and embryonic (3). The acquired form accounts for 80% of affected infants. They are asymptomatic and anicteric at birth and develop jaundice in the first postnatal weeks. These otherwise normal infants are born with a patent biliary system which undergoes progressive inflammation and fibro-obliteration initiated by a perinatal insult. Infants with the embryonic form have no jaundice-free interval and suffer from one or more congenital anomalies, such as interruption of the suprarenal segment of the inferior vena cava with azygous continuation, preduodenal portal vein, midline symmetric liver, intestinal malrotation, situs anomalies, bronchial anomalies, and polysplenia or asplenia (2,4). Pathology BA can be classified by using macroscopic appearance and cholangiography findings according to three main categories. Types I, II, and III are defined as atresia at the site of the common bile duct, at the site of the hepatic duct, and up to the porta hepatis, respectively. Type III is most common. A patent duct that can be anastomosed to the intestine at the porta hepatis is present in 5% of cases. In more than 90% of cases, no normal ductal structures are seen at the porta hepatis (5). Early in the course the liver is enlarged. There are portal tract edema, bile duct proliferation, portal and periductal inflammation, and associated areas of hepatic cell injury. This process develops into end stage cirrhosis. The pathologic changes are generally considered to be panductal, affecting intrahepatic as well as extrahepatic structures (6–8). The degree of damage present in the intrahepatic biliary system is responsible for much of the morbidity after hepatic portoenterostomy (HPE). Paucity or absence of intralobular bile ducts along with architectural disturbances, even in jaundice-free infants after successful HPE, has been observed by some (9).

478

PEDIATRIC HPB DISORDERS
Table 51.1 Differential Diagnosis of Neonatal Cholestasis
1. Extrahepatic causes A. Biliary atresia B. Choledochal cyst C. Bile duct stenosis, strictures, or cholelithiasis D. Spontaneous perforation of the common bile duct E. Tumors or masses (extrinsic or intrinsic compression of bile ducts) 2. Intrahepatic causes A. Infectious: cytomegalovirus, rubella, herpes simplex, human herpesvirus 6, varicella zoster, adenovirus, enterovirus, parvovirus B19, hepatitis B virus, human immunodeficiency virus, toxoplasmosis, syphilis, tuberculosis, listeriosis, bacterial sepsis, and urinary tract infection B. Metabolic: alpha-1 antitrypsin deficiency, cystic fibrosis, galactosemia, hereditary tyrosinemia, hereditary fructose intolerance, glycogen storage disease type IV, Niemann-Pick type C, Gaucher’s disease, Wolman’s disease, cholesterol ester storage disease, panhypopituitarism, hypothyroidism, bile acid synthesis defects, peroxisomal disorders, arginase deficiency, and mitochondrial respiratory chain deficiencies C. Genetic: Alagille syndrome, Turner syndrome, trisomy 21, arthrogryposis-renal dysfunction-cholestasis syndrome, Aagenaes syndrome (cholestasis with lymphedema), progressive familial intrahepatic cholestasis (FIC1, BSEP, and multiple drug resistance 3 gene deficiencies), North American Indian childhood cirrhosis (cirrhin deficiency), congenital hepatic fibrosis/autosomal recessive polycystic kidney disease, Caroli’s disease, and neonatal Dubin–Johnson syndrome D. Toxic: total parenteral nutrition-associated, endotoxin from gram-negative infection, choral hydrate and other medications, and aluminum E. Cholangiopathies: nonsyndromic paucity of interlobular bile ducts and neonatal sclerosing cholangitis F. Miscellaneous: idiopathic neonatal hepatitis, congenital lupus, ischemia-reperfusion injury, histiocytosis X, erythrophagocytic lymphohistiocytosis, veno-occlusive disease, erythroblastosis fetalis (inspissated bile syndrome), and neonatal iron storage disease
Source: Reprinted from Ref. (105).

Table 51.2 Diagnostic Evaluation in the Cholestatic Neonate
Basic evaluation Serum aspartate aminotransferase and Serum alanine aminotransferase γ-glutamyl transferase Bilirubin—indirect and direct Prothrombin time/International normalized ratio Albumin level Complete blood count Initial diagnostic evaluation Blood and urine cultures Urine for reducing substances (if infant on a galactose-containing diet) Galactose-1-phosphate uridyl transferase Urine succinylacetone Thyroid-stimulating hormone/T4 Cortisol Hepatobiliary ultrasound Secondary diagnostic evaluation A1AT level and genotype Cystic fibrosis screen Liver biopsy Hepatobiliary scintigraphy MRCP or ERCP (depending on facility)
Source: Reprinted from Ref. (106).

The incidence is 1 in 13,000 to 15,000 in western countries, but rates as high as 1 in 1000 have been described in Japan (24). Choledochal cyst can be categorized according to their anatomic appearance into five types (25,26): I. Cystic or diffuse fusiform dilatation of the extrahepatic bile duct II. Diverticulum of the extrahepatic bile duct III. Choledochocele IV. Multiple cysts of the intra- or extrahepatic ducts (or both) V. Single or multiple intrahepatic cysts Type I accounts for at least 75% of all cases, and type IV accounts for most of the remainder. The other varieties are rare (27,28). Pathology Histologic sections of the wall of extrahepatic choledochal cysts have demonstrated a thick-walled structure of dense connective tissue interlaced with strands of smooth muscle. In most instances, some degree of inflammatory reaction is noted. The degree of histologic damage and the rate of epithelial metaplasia and dysplasia are related to the age of the patient (29). In a newborn, the histologic appearance of the liver is usually normal or having mild bile duct proliferation consistent with chronic biliary obstruction. Occasionally in older patients mild periportal fibrosis is noted. Choledochal cysts are congenital. There is predominance in females, suggesting a sex-linked defect, as well as a much higher incidence in Asian populations.

choledochal cyst
Classification Choledochal cysts are congenital anomalies of the biliary tract manifested by cystic dilatation of the extra hepatic biliary tree.

479

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
Table 51.3 Contemporary Outcome of Biliary Atresia Following Kasai Hepatoportoenterostomy (HPE) and Liver Transplantation
Country, year, and number of centers Japan, 1989–1999, 93 centers UK and Ireland, 1993–1995, 15 centers USA, 1997–2000, 9 centers France, 1997–2002, 22 centers England and Wales, 1999–2002, 3 centers
Source: Adapted from Ref. (107).

Number of patients 1381 93 104 271 148

Median age at HPE 61–70 days 54 days 61 days 57 days 54 days

Survival with native liver 5 years: 59.7% (actual) 5 years: 30.1% (actuarial) 2 years: 55.8% (actual) 4 years: 42.7% (actuarial) 4 years: 51% (actuarial)

Survival after liver transplantation _ 2.4 years: 89% (actual) 2 years: 88% (actual) 4 years: 88.8% (actuarial) 2 years: 89% (actuarial)

Overall survival of patients 5 years: 75.5% (actual) 5 years: 85% (actuarial) 2 years: 91.3% (actual) 4 years: 87.1% (actuarial) 4 years: 89% (actuarial)

The most plausible etiology is obstruction of the distal common bile duct. Clinical Features In the infantile form, patients present with obstructive jaundice, acholic stools, and hepatomegaly at 1 to 3 months of age (30). Patients do not tend to have abdominal pain of palpable mass. Infants do not ordinarily become jaundiced until 1 to 3 weeks after birth. In the adult forms, the clinical manifestations do not become evident until the patient is 2 years old; and most of these patients have fusiform deformities of the common duct without high grade or complete obstruction. The classic triad of abdominal pain, a palpable abdominal mass, and jaundice may be noted (31). Only partial obstruction occurs in the adult form, so the symptoms are intermittent. The pattern of pain has been described as similar to that of recurrent pancreatitis. Imaging Ultrasonography may be the only screening required in infants. If a choledochal cyst is suspected on US, 99Tcdi-isopropylphenylcarbamoyl-methylimidodiacetic acid (DISIDA) scintigraphy can confirm the diagnosis and provide information about drainage, obstruction, and hepatic function. Prenatal ultrasonography of fetal choledochal cyst has been reported by a number of investigators (32–34). A key question after prenatal diagnosis is the appropriate timing of surgical correction. Redkar suggests that asymptomatic patients are best operated around 3 months of age (35). Suita et al. noted that patients who undergo surgery within a month of life have a lower incidence of hepatic fibrosis than those operated on later (36). Treatment In 1970, Kasai et al. and Ishida et al. reported favorable results with cyst excision and Roux-en-Y jejunostomy (37,38). Outcome Radical cyst excision and hepaticojejunostomy yield consistently good results, even in small infants (27). In a large Japanese series of 200 children followed for a mean of

11 years, Roux-en-Y hepaticojejunostomy was performed in 188 patients. No operative mortality occurred, and 9% had complications including cholangitis, intrapancreatic terminal CBD calculi, pancreatitis, and bowel obstruction (39).

gallbladder disease
Cholelithiasis The prevalence of gallstones in children varies according to geography and age. The predominant factors in gallstone formation are biliary stasis, excess bilirubin pigment, and lithogenic bile. Hemolytic disorders, fasting and TPN, ileal resection/disease, cystic fibrosis, Down syndrome, childhood cancer, bone marrow transplantation, cardiac transplantation, spinal surgery, dystrophia myotonica, and chronic intestinal pseudo-obstruction have all been associated with increased incidence of cholelithiasis in children (40). Gallstones in infants occasionally resolve spontaneously. Early surgery can be deferred in the asymptomatic infant with gallbladder calculi. Management of asymptomatic cholelithiasis in older children is controversial. In children without hemolytic disorders, a conservative approach is recommended (41). Cholecystectomy is the standard treatment for symptomatic or complicated gallbladder stones. Hemolytic Cholelithiasis In the past, the usual cause of gallstones in children was hemolytic disease. Hereditary spherocytosis, sickle cell anemia, and thalassemia are the most common hemolytic disorders resulting in the development of gallstones. In patients with spherocytosis, ultrasound is indicated prior to splenectomy. If stones are present, cholecystectomy is performed. In sickle cell disease, only symptomatic patients require cholecystectomy, which should be preformed electively rather than emergently during a hemolytic crisis (42). Congenital Deformities Congenital deformities include a variety of abnormal configurations and locations of the gallbladder, such as gallbladder agenesis, duplication, bilobation, floating gallbladder, diverticula, and ectopia. They are usually of no clinical relevance; if

480

PEDIATRIC HPB DISORDERS
they are of clinical relevance, the symptoms are from gallbladder emptying, and cholecystectomy is recommended. Staging Most studies to date have used the clinical grouping defined by the Children’s Cancer Group and the Pediatric Oncology Group, which is presented in Table 51.4. Treatment Most studies support the effectiveness of systemic chemotherapy combined with complete surgical resection of the primary hepatic tumor (61,62). Survival depends on removal of the primary liver tumor in most cases. The first clinical decision is whether to initial neoadjuvant chemotherapy or proceed with resection. A completely resected tumor without the presence of metastatic disease is deemed stage I. If after resection pathology shows pure fetal histology, close observation ensues. For all other histologies and for stage II disease, four cycles of combination cisplatin, 5-fluorouracil, and incrusting are given. For a tumor deemed unrespectable at diagnosis (stage III) or a patient with metastatic disease (stage IV), current therapy consists of four cycles of chemotherapy with either resection or liver transplantation after cycle 4 followed by two more cycles. Extensive tumors usually shrink with chemotherapy, facilitating resection, whereas chemotherapy might be lessened or avoided in some patients by resection at diagnosis. About 46% of hepatic malignancies are resectable at diagnosis (59). Outcome Overall survival of 60% to 70% is achievable with non-stage IV hepatoblastoma except for with the very aggressive small cell variant. Approximately 50% of patients who present with pulmonary metastasis are curable. If gross disease remains in the primary site, survival falls to zero. Some patients with microscopic residual tumor are curable with continued chemotherapy and may benefit from external-beam radiotherapy to the primary hepatic site. In a multivariate analysis, factors that have been independent predictors of worsened prognosis include high TNM stage, unresectable tumor, bilobar involvement and multifocality, AFP less than 100 ng/ml or more than 105 ng/ml, distant metastases, embryonal versus fetal histology, and vascular invasion (63).

hepatoblastoma
Incidence Hepatoblastoma is the most common malignant hepatic tumor. Liver cancers constitute 0.5% to 2% of all pediatric solid tumors and about 5% of abdominal tumors in childhood (43). Hepatoblastomas are the most common primary hepatic tumors of childhood constituting 43% to 64% of all hepatic neoplasms in one large series (43–45). Approximately twothirds of all liver masses occurring in children are malignant. Eighty percent of 123 children in the United States registered with malignant liver tumors in 2000 had hepatoblastoma and they accounted for 91% of primary hepatic malignancies in children less than 5 years of age (46). There are approximately 50 to 70 new cases per year in the United States with a male to female ratio of 1.7:1 (47). The median age at diagnosis is about 18 months, and most cases occur before age 2½ to 3 years (48). Hepatoblastoma may occur in siblings (49–51). It is most strongly associated with familial polyposis (52,53), Gardner’s syndrome (54), and Beckwith–Wiedemann syndrome (55,56). Pathology The five histologic subtypes observed in hepatoblastoma are fetal, embryonal, mixed mesenchymal, macrotubular, and anaplastic or small cell. The importance of subtyping in hepatoblastoma is the association between prognostic risk and subtype (57,58). Patients with small cell undifferentiated tend to do worse. Figure 51.1 illustrates imaging and pathology of a child with Beckwith-Wiedemann syndrome with hepatoblastoma. Clinical Features The most common presenting sign of hepatoblastoma is an asymptomatic abdominal mass. A mild anemia with a markedly elevated platelet count is observed in most patients at diagnosis. The cause is probably secondary to abnormal cytokine release. In an abstract from the 1993 Annual Meeting of the American Society of Clinical Oncology, Van Tournet et al. stated that measurement of serum alpha-fetoprotein is well established as an initial tumor marker in the diagnosis of hepatoblastoma and a means of monitoring the therapeutic response. The normal level in most laboratories is less than 20 ng/ml whereas the AFP level at diagnosis in hepatoblastoma patients can range from normal to 7.7 × 106 ng/ml. It is estimated that the AFP is elevated in 84% to 91% of patients with hepatoblastoma (58). In comparison, the mean in pediatric patients with HCC was about 200,000 ng/ml (59). Imaging The first imaging study is usually an abdominal ultrasound. Computed tomography (CT) is useful to identify pulmonary metastases, identify diffuse hepatic involvement, and determine respectability. MRI is useful for evaluating hepatic lesions and their relationship to vascular structures (60). It can show the hepatic veins, the vena cava, and bile ducts.

Incidence HCC accounts for 23% of pediatric liver tumors (64). The incidence is bimodal with an early peak that occurs before 5 years and a second peak that occurs between 13 and 15 years. HCC is the most common hepatic malignancy of adolescence (65). There is a male predominance of 1.3 to 3.2:1. Hepatitis B and C correlate with the incidence of HCC. In Asia 85% of these patients (adults and children) are hepatitis B surface antigen positive, whereas this is found in only 10% to 25% of patients in the United States. The relative risk for the development of HCC is 250:1 for patients with chronic active hepatitis compared with patients without hepatitis surface antigen positivity (66). Other conditions associated with the development of HCC include cirrhosis, α1-antitrypsin deficiency, tyrosinemia, aflatoxin ingestion, hemochromatosis, hepatic venous obstruction, androgen and estrogen exposure, Alagille syndrome, and thorotrast administration (67).

hepatocellular carcinoma ⁽hcc⁾

481

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS

(A)

(B)

(C)

(D)

(E) Figure 51.1 Ten-month-old female with Beckwith–Wiedemann syndrome. (A,B) The hepatoblastoma is centered on the middle hepatic vein, with an extension into the right hepatic vein (arrow in A) and an encasement of the left hepatic vein (arrow in B). (C,D) Preoperative chemotherapy resulted in shrinkage of the tumor, with apparent involvement of a clear plane between the tumor and the right hepatic vein (arrow in C); panel D demonstrates the tumor encasing the left portal vein. An attempted extended left hepatic lobectomy was abandoned because of the presence of an occult tumor involvement of the right hepatic vein noted at surgery. (E) The patient underwent hepatic transplantation. Source: Reprinted from Ref. (104).

Pathology HCCs are highly invasive and often multicentric at diagnosis, with frequent hemorrhage and necrosis. Invasiveness, especially vascular invasion, is a hallmark of these tumors. Extrahepatic dissemination to portal lymph nodes, lungs, and bones is frequent at diagnosis and strongly affects survival.

Clinical Features Children and adolescents with HCC frequently present (68) with palpable abdominal masses (40%), but many are asymptomatic at diagnosis. Pain is common (38%) and may occur in the absence of an obvious mass. Constitutional disturbances such a as anorexia, malaise, nausea and vomiting, and significant

482

PEDIATRIC HPB DISORDERS
Table 51.4 Children’s Oncology Group Staging for Hepatoblastoma
Stage I Favorable histology Other histology Stage II Complete resection Purely fetal histology with a low mitotic index All other stage I tumors Gross total resection with microscopic residuals or total resection with preoperative or intraoperative rupture Unresectable tumors as determined by the attending surgeon, partially resected tumors with macroscopic residual, or any tumor with lymph node involvement Measurable metastatic disease to lungs or other organs

When encountered at laparotomy, they should usually be excised unless this would entail significant risk of morbidity. Annular Pancreas Annular pancreas is thought to be due to a faulty rotation of the ventral pancreatic bud in its course around the posterior aspect of the duodenal anlage during the sixth week of gestation. The duodenum is encircled by and obstructed with normal pancreatic tissue containing normal functioning acini, ducts, and islets of Langerhans (79,80). The theory is that half the ventral bud migrates anteriorly and half migrates posteriorly. Duodenal atresia and stenosis, intestinal malrotation, and trisomy 21 can often be found in combination with annular pancreas (81). The clinical symptoms relate to duodenal obstruction with bilious vomiting. Radiographic studies reveal the classic finding of the “double bubble” sign (82). Management consists of surgical bypass of the obstructing lesion with a duodenoduodenostomy. Resection or division of the annular pancreas should not be carried out. Pancreas Divisum Failure of the duct of Wirsung and the duct of Santorini to unite from the dorsal and ventral pancreas during development results in the anatomic variant known as pancreas divisum. In this disorder the duct of Wirsung is very small, and the duct of Santorini becomes the major ductal system and communicates with the duodenum through the minor papilla. If the orifice of the accessory papilla is stenotic, pancreatitis can occur. The goal of treatment is to establish adequate drainage of the duct of Santorini. Several reports show that adequate drainage can be obtained with accessory papilla sphincteroplasty (83). Acute Pancreatitis The causes of acute pancreatitis include trauma, biliary tract stone disease, choledochal cyst, ductal developmental anomalies, drugs, such as retrovirals, diuretics, anticonvulsants, as well as some chemotherapeutic agents, metabolic derangements, and infections. In children, trauma is the most common cause. Pancreas divisum is an anomaly present in 10% of the population, resulting from failure of the dorsal duct to fuse with the ventral duct. This relative obstruction may cause recurring episodes of pancreatitis (84). These patients should undergo a sphincteroplasty of the minor papilla. The pathogenesis entails the inappropriate activation of proenzymes, leading to autodigestion of the pancreas. Acute pancreatitis usually presents with the acute onset of midepigastric pain associated with back pain, severe vomiting, and low-grade fever (85,86). In severe cases of necrotizing or hemorrhagic pancreatitis, hemorrhage may dissect from the pancreas along tissue plans appearing as ecchymosis either in the flanks (Grey–Turner sign) or at the umbilicus (Cullen’s sign). Lipase levels have been proposed as a more specific test of pancreatic tissue damage, although intestinal perforation does cause an elevation of lipase throughout reabsorption via the peritoneum. Lipase is produced only in the pancreas and its measurement is particularly helpful for distinguishing pancreatic trauma from salivary trauma (87). CT scan offers better resolution than other modalities to determine size of pancreas, degree of edema, and the presence of fluid collections (88). It

Stage III

Stage IV
Source: Reprinted from Ref. (104).

weight loss occur with greater frequency. AFP is elevated in approximately 85% of patients, with most levels greater than 1000 ng/ml (69). Treatment Long-term survival is impossible without complete resection. However, because of the high incidence of multifocality within the liver, extrahepatic extension to regional lymph nodes, vascular invasion, and distant metastases, complete resection is often impossible. Unresectable HCCs can be palliated with embolization with or without added chemotherapeutic agents or radioisotopes (70). Percutaneous intralesional injection of ethanol also has been of palliative benefit when lesions are small (71). Radiofrequency ablation of these tumors, percutaneously or at laparoscopy/laparotomy, has been associated with tumor resolution and prolonged survival (72–75). Outcome The overall survival from HCC in childhood approaches zero and it remains a therapeutic problem. Occasionally, resection of localized lesions results in long-term survival. The trend is to separate HCC from hepatoblastoma in clinical studies because of its greatly differing biologic behavior.

pancreas
Congenital Anomalies Ectopic Pancreatic Rests The incidence of ectopic pancreatic rests in autopsy ranges from 1% to 2% (76) and is frequently encountered along foregut derivatives, such as the stomach and duodenum, as well as jejunum, ileum, and colon (77). They represent the most common anomaly of the gastric antrum and may cause a gastric outlet obstruction (78). Their origin is unknown, but one possible explanation suggests an aberrant epithelial–mesenchymal interaction, leading to trans-differentiation of heterotrophic embryonic epithelium into pancreatic epithelium. Ectopic rests are usually asymptomatic and found incidentally at laparotomy.

483

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
can also distinguish areas of necrosis. ERCP has been shown to have higher complication rates in children than in the adult population; however, it may be helpful with an impacted stone or in trauma patients with a pseudocyst (89). MRCP is a noninvasive, nontherapeutic technique to evaluate the biliary tree and pancreas. MRCP is currently the initial imaging study of choice in evaluation of pancreatic ductal anatomy in children with pancreatitis (90). Treatment requires aggressive fluid replacement and low threshold for transferring the patient to an intensive care unit (91,92). Surgical intervention in acute pancreatitis is not often necessary and is reserved for patients with severe necrotizing pancreatitis needing debridement or patients with pancreatic abscess (93,94). Pancreatic Abscess A pancreatic abscess may result from infection of necrotic pancreatic tissue or a peripancreatic fluid collection. Pancreatic abscess increases the mortality rate of pancreatitis threefold and is an absolute indication for surgical therapy (95,96). Diagnosis is made by Gram stain and culture of the suspected abscess by CT-guided needle aspiration. The indication is fever and leukocytosis persisting more than 7 to 10 days after onset of pancreatitis. Surgery consists of debridement of clearly necrotic tissue and placement of large sump suction drains. Pancreatic Pseudocyst Pancreatic pseudocysts result from damage of the pancreatic ductal system. The extravasated pancreatic enzymes and digested tissue are contained by the formation of a cavity composed of fibroblastic reaction and inflammation; however, there is no epithelial lining. Pseudocysts may be acute or chronic. Acute pseudocysts have an irregular wall on CT, and about 50% resolve without therapy. Chronic pseudocysts are spherical with a thick wall and rarely resolve on their own. In children, pseudocysts tend to resolve more frequently with medical therapy alone (97). Persistent pseudocysts require internal drainage, excision, or external drainage. Chronic Pancreatitis Chronic pancreatitis differs from acute pancreatitis in the irreversibility of the changes associated with the inflammation (98). Chronic pancreatitis is either calcifying or obstructive. The calcifying form is more common in children and is usually caused by hereditary pancreatitis, and is associated with intraductal pancreatic stones, pseudocysts and a more aggressive scar formation with significant damage. The obstructive type is associated with anatomic obstructions and is less severe. Chronic pancreatitis is uncommon in children, and the most common cause in North America is hereditary or familial pancreatitis (99). The inheritance is autosomal dominant with incomplete penetrance. The majority of patients express one of two mutations in the trypsinogen gene leading to alterations that prevent deactivation of trypsin within the pancreas causing autodigestion. The diagnosis of chronic pancreatitis depends on characteristic pain, diminished pancreatic function, and changes in radiographic appearances. Therapy is directed toward palliation of symptoms. Surgical or endoscopic therapy is indicated for bile or pancreatic duct obstruction or for pancreatic pseudocyst complications (100). Persistent Hyperinsulinemic Hypoglycemia of Infancy The defect in patients with persistent hyperinsulinemic hypoglycemia of infancy (PHHI) is related to four genes responsible for the ability of the beta-cell to regulate insulin secretion. Mutations prevent the normal feedback regulation of insulin production by serum glucose. Patients typically have hypoglycemia shortly after birth. It is critical to measure serum insulin levels and glucose levels simultaneously because the ratio is important. Initial treatment should consist of frequent feedings, with the addition of intravenous glucose as needed. Initial medical treatment should include antisecretory drugs such as diazoxide or a long-acting somatostatin analog. In patients with diffuse-type PHHI, adequate surgical treatment consists of a 90% to 95% pancreatectomy, leaving a residual remnant of the pancreas on the common bile duct (101). It is important to examine the tissue for adenoma. Approximately 75% of patients develop diabetes occurs after a 95% pancreatectomy (102). The long-term outcome for these patients depends on the age at onset, which relates to severity of disease. Most patients seem to outgrow the disease after several years, perhaps due to diminished activity of the beta-cell. Adenocarcinoma In general, pancreatic cancers in children are rare. Acinar cell adenocarcinoma has been seen in children and tends to be less aggressive with a better prognosis. Treatment is complete surgical resection. Another variant of adenocarcinoma seen in younger children has been termed pancreatoblastoma. This is the most common exocrine tumor of the pancreas in children. It is more often seen in boys and is thought to be of embryonic origin. These tumors are low malignancy and often arise in the head of the pancreas (103). Metastases are reported in one-third of cases, with the liver and lung being the most common sites. The prognosis is relatively good with a complete resection. As recurrence is common, close follow-up is required.

KEY POINTS

Biliary atresia is marked by obstruction of the bile ducts. In infancy, jaundice that persists beyond 2 weeks is physiologic. Signs and symptoms include jaundice, clay-colored stools, and hepatomegaly. Treatment is a hepatic portoenterostomy. Choledochal cysts are congenital anomalies of the biliary tract. There are five types. They present with obstructive jaundice. Treatment is cyst excision and Roux-en-Y jejunostomy. Cholelithiasis, hemolytic cholelithiasis, and congenital deformities are gallbladder diseases seen in pediatrics. Treatment varies for each and depends on symptoms.

484

PEDIATRIC HPB DISORDERS
Hepablastoma is the most common malignant hepatic tumor. There are five histologic subtypes. Patients with small cell undifferentiated have worse prognosis. It generally presents as an asymptomatic abdominal mass. Treatment is with systemic chemotherapy and complete surgical resection. Hepatocellular carcinoma is the most common hepatic malignancy of adolescence. It is highly invasive and generally presents with an asymptomatic abdominal mass. Treatment is complete resection; however, this is often difficult, and thus embolization may be helpful. Congenital anomalies consist of annular pancreas and pancreas divisum. Pancreatitis is rare, but both acute and chronic types are seen. Adenocarcinoma is rare in children.
20. Lykavieris P, Chardot C, Sokhn M, et al. Outcome in adulthood of biliary atresia: a study of 63 patients who survived for over 20 years with their native liver. Hepatology 2005; 41: 366–71. 21. Okazaki T, Kobayashi H, Yamataka A, et al. Long-term postsurgical outcome of biliary atresia. J Pediatr Surg 1999; 34: 312–5. 22. Altman RP, Lilly JR, Greenfeld J, et al. A multivariable risk factor analysis of the portoenterostomy (Kasai) procedure for biliary atresia: twenty-five years of experience from two centers. Ann Surg 1997; 226: 348–53; discussion 53–5. 23. Utterson EC, Shepherd RW, Sokol RJ, et al. Biliary atresia: clinical profiles, risk factors, and outcomes of 755 patients listed for liver transplantation. J Pediatr 2005; 147: 180–5. 24. McEvoy CF, Suchy FJ. Biliary tract disease in children. Pediatr Clin North Am 1996; 43: 75–98. 25. Todani T, Watanabe Y, Narusue M, et al. Congenital bile duct cysts: Classification, operative procedures, and review of thirty-seven cases including cancer arising from choledochal cyst. Am J Surg 1977; 134: 263–9. 26. Todani T, Watanabe Y, Toki A, et al. Co-existing biliary anomalies and anatomical variants in choledochal cyst. Br J Surg 1998; 85: 760–3. 27. Miyano T, Yamataka A, Kato Y, et al. Hepaticoenterostomy after excision of choledochal cyst in children: a 30-year experience with 180 cases. J Pediatr Surg 1996; 31: 1417–21. 28. Stringer MD, Dhawan A, Davenport M, et al. Choledochal cysts: lessons from a 20 year experience. Arch Dis Child 1995; 73: 528–31. 29. Tan CE, Howard ER, Driver M, et al. Non-communicating multiseptate gall bladder and choledochal cyst: a case report and review of publications. Gut 1993; 34: 853–6. 30. Davenport M, Betalli P, D’Antiga L, et al. The spectrum of surgical jaundice in infancy. J Pediatr Surg 2003; 38: 1471–9. 31. Alonso-Lej F, Rever WB Jr., Pessagno DJ. Congenital choledochal cyst, with a report of 2, and an analysis of 94, cases. Int Abstr Surg 1959; 108: 1–30. 32. Howell CG, Templeton JM, Weiner S, et al. Antenatal diagnosis and early surgery for choledochal cyst. J Pediatr Surg 1983; 18: 387–93. 33. Lugo-Vicente HL. Prenatally diagnosed choledochal cysts: observation or early surgery? J Pediatr Surg 1995; 30: 1288–90. 34. Mackenzie TC, Howell LJ, Flake AW, et al. The management of prenatally diagnosed choledochal cysts. J Pediatr Surg 2001; 36: 1241–3. 35. Redkar R, Davenport M, Howard ER. Antenatal diagnosis of congenital anomalies of the biliary tract. J Pediatr Surg 1998; 33: 700–4. 36. Suita S, Shono K, Kinugasa Y, et al. Influence of age on the presentation and outcome of choledochal cyst. J Pediatr Surg 1999; 34: 1765–8. 37. Ishida M, Tsuchida Y, Saito S, et al. Primary excision of choledochal cysts. Surgery 1970; 68: 884–8. 38. Kasai M, Asakura Y, Taira Y. Surgical treatment of choledochal cyst. Ann Surg 1970; 172: 844–51. 39. Yamataka A, Ohshiro K, Okada Y, et al. Complications after cyst excision with hepaticoenterostomy for choledochal cysts and their surgical management in children versus adults. J Pediatr Surg 1997; 32: 1097–102. 40. Stringer M. Gallbladder disease and cholelithiasis. In: Howard ER, Stringer MD, Columbani PM, eds. Surgery of the Liver, Bile Ducts and Pancreas in Children. 2nd edn. London, UK: Arnold; 2002: 189–208. 41. Bruch SW, Ein SH, Rocchi C, et al. The management of nonpigmented gallstones in children. J Pediatr Surg 2000; 35: 729–32. 42. Ware R, Filston HC, Schultz WH, et al. Elective cholecystectomy in children with sickle hemoglobinopathies. Successful outcome using a preoperative transfusion regimen. Ann Surg 1988; 208: 17–22. 43. Weinberg AG, Finegold MJ. Primary hepatic tumors of childhood. Hum Pathol 1983; 14: 512–37. 44. Stocker JT. Hepatoblastoma. Semin Diagn Pathol 1994; 11: 136–43. 45. Mann JR, Kasthuri N, Raafat F, et al. Malignant hepatic tumours in children: incidence, clinical features and aetiology. Paediatr Perinat Epidemiol 1990; 4: 276–89. 46. Darbari A, Sabin KM, Shapiro CN, et al. Epidemiology of primary hepatic malignancies in U.S. children. Hepatology 2003; 38: 560–6. 47. Lampkin BC, Wong KY, Kalinyak KA, et al. Solid malignancies in children and adolescents. Surg Clin North Am 1985; 65: 1351–86. 48. Exelby PR, Filler RM, Grosfeld JL. Liver tumors in children in the particular reference to hepatoblastoma and hepatocellular carcinoma: American Academy of Pediatrics Surgical Section Survey—1974. J Pediatr Surg 1975; 10: 329–37.

references
1. Mack CL, Sokol RJ. Unraveling the pathogenesis and etiology of biliary atresia. Pediatr Res 2005; 57: 87R–94R. 2. Carmi R, Magee CA, Neill CA, et al. Extrahepatic biliary atresia and associated anomalies: etiologic heterogeneity suggested by distinctive patterns of associations. Am J Med Genet 1993; 45: 683–93. 3. Perlmutter DH, Shepherd RW. Extrahepatic biliary atresia: a disease or a phenotype? Hepatology 2002; 35: 1297–304. 4. Silveira TR, Salzano FM, Howard ER, et al. Congenital structural abnormalities in biliary atresia: evidence for etiopathogenic heterogeneity and therapeutic implications. Acta Paediatr Scand 1991; 80: 1192–9. 5. O’Neill JA, Rowe MI, Grosfeld JL, et al. Pediatric Surgery. 5th edn. St. Louis, MO: Mosby, 1998. 6. Raweily EA, Gibson AA, Burt AD. Abnormalities of intrahepatic bile ducts in extrahepatic biliary atresia. Histopathology 1990; 17: 521–7. 7. Lilly JR, Altman RP. Hepatic portoenterostomy (the Kasai operation) for biliary atresia. Surgery 1975; 78: 76–86. 8. Ito T, Horisawa M, Ando H. Intrahepatic bile ducts in biliary atresia--a possible factor determining the prognosis. J Pediatr Surg 1983; 18: 124–30. 9. Sherlock S. The syndrome of disappearing intrahepatic bile ducts. Lancet 1987; 2: 493–6. 10. Balistreri WF. Neonatal cholestasis. J Pediatr 1985; 106: 171–84. 11. Brough AJ, Bernstein J. Conjugated hyperbilirubinemia in early infancy. A reassessment of liver biopsy. Hum Pathol 1974; 5: 507–16. 12. Moyer V, Freese DK, Whitington PF, et al. Guideline for the evaluation of cholestatic jaundice in infants: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr 2004; 39: 115–28. 13. Kasai M, Suzuki S. A new operation for “non-correctable” biliary atresia: hepatic portoenterostomy. Shujjutsu 1959; 13: 733–9. 14. Shteyer E, Ramm GA, Xu C, et al. Outcome after portoenterostomy in biliary atresia: pivotal role of degree of liver fibrosis and intensity of stellate cell activation. J Pediatr Gastroenterol Nutr 2006; 42: 93–9. 15. Azarow KS, Phillips MJ, Sandler AD, et al. Biliary atresia: should all patients undergo a portoenterostomy? J Pediatr Surg 1997; 32: 168–72; discussion 72–4. 16. Lang T, Kappler M, Dietz H, et al. Biliary atresia: which factors predict the success of a Kasai operation? An analysis of 36 patients. Eur J Med Res 2000; 5: 110–4. 17. Davenport M, De Ville de Goyet J, Stringer MD, et al. Seamless management of biliary atresia in England and Wales (1999–2002). Lancet 2004; 363: 1354–7. 18. Chardot C, Carton M, Spire-Bendelac N, et al. Prognosis of biliary atresia in the era of liver transplantation: French national study from 1986 to 1996. Hepatology 1999; 30: 606–11. 19. Laurent J, Gauthier F, Bernard O, et al. Long-term outcome after surgery for biliary atresia. Study of 40 patients surviving for more than 10 years. Gastroenterology 1990; 99: 1793–7.

485

SURGICAL MANAGEMENT OF HEPATOBILIARY AND PANCREATIC DISORDERS
49. Fraumeni JF Jr., Rosen PJ, Hull EW, et al. Hepatoblastoma in infant sisters. Cancer 1969; 24: 1086–90. 50. Napoli VM, Campbell WG Jr. Hepatoblastoma in infant sister and brother. Cancer 1977; 39: 2647–50. 51. Taat F, Bosman DK, Aronson DC. Hepatoblastoma in a girl with biliary atresia: coincidence or co-incidence. Pediatr Blood Cancer 2004; 43: 603–5. 52. Giardiello FM, Petersen GM, Brensinger JD, et al. Hepatoblastoma and APC gene mutation in familial adenomatous polyposis. Gut 1996; 39: 867–9. 53. Iwama T, Mishima Y. Mortality in young first-degree relatives of patients with familial adenomatous polyposis. Cancer 1994; 73: 2065–8. 54. Hartley AL, Birch JM, Kelsey AM, et al. Epidemiological and familial aspects of hepatoblastoma. Med Pediatr Oncol 1990; 18: 103–9. 55. Koishi S, Kubota M, Taniguchi Y, et al. Myelodysplasia in a child with Beckwith–Wiedemann syndrome previously treated for hepatoblastoma with multi-agent chemotherapy. J Pediatr Hematol Oncol 1996; 18: 419–20. 56. Tsai SY, Jeng YM, Hwu WL, et al. Hepatoblastoma in an infant with Beckwith–Wiedemann Syndrome. J Formos Med Assoc 1996; 95: 180–3. 57. Gonzalez-Crussi F, Upton MP, Maurer HS. Hepatoblastoma. Attempt at characterization of histologic subtypes. Am J Surg Pathol 1982; 6: 599–612. 58. Lack EE, Neave C, Vawter GF. Hepatoblastoma. A clinical and pathologic study of 54 cases. Am J Surg Pathol 1982; 6: 693–705. 59. Ortega JA, Krailo MD, Haas JE, et al. Effective treatment of unresectable or metastatic hepatoblastoma with cisplatin and continuous infusion doxorubicin chemotherapy: a report from the Childrens Cancer Study Group. J Clin Oncol 1991; 9: 2167–76. 60. Ohnuma N, Takahashi H, Tanabe M, et al. The role of magnetic resonance imaging for treatment in children with malignant solid tumor. Nippon Geka Gakkai Zasshi 1991; 92: 1143–6. 61. Gauthier F, Valayer J, Thai BL, et al. Hepatoblastoma and hepatocarcinoma in children: analysis of a series of 29 cases. J Pediatr Surg 1986; 21: 424–9. 62. von Schweinitz D, Hecker H, Harms D, et al. Complete resection before development of drug resistance is essential for survival from advanced hepatoblastoma–a report from the German Cooperative Pediatric Liver Tumor Study HB-89. J Pediatr Surg 1995; 30: 845–52. 63. von Schweinitz D, Hecker H, Schmidt-von-Arndt G, et al. Prognostic factors and staging systems in childhood hepatoblastoma. Int J Cancer 1997; 74: 593–9. 64. Finegold MJ. Tumors of the liver. Semin Liver Dis 1994; 14: 270–81. 65. Katzenstein HM, Krailo MD, Malogolowkin MH, et al. Hepatocellular carcinoma in children and adolescents: results from the Pediatric Oncology Group and the Children’s Cancer Group intergroup study. J Clin Oncol 2002; 20: 2789–97. 66. Bower M, Newlands ES, Habib N. Fibrolamellar hepatocellular carcinoma responsive to platinum-based combination chemotherapy. Clin Oncol (R Coll Radiol) 1996; 8: 331–3. 67. Wegmann W, Evison J, Schaub N, et al. [Liver cell carcinoma as a late complication of Alagille syndrome (arterio-hepatic dysplasia)]. Leber Magen Darm 1996; 26: 157–8, 61–3. 68. Ni YH, Chang MH, Hsu HY, et al. Hepatocellular carcinoma in childhood. Clinical manifestations and prognosis. Cancer 1991; 68: 1737–41. 69. Brower ST. Pancreatic cancer, hepatobiliary cancer, and neuroendocrine cancers of the GI tract. In: Padzur R, Hoskins WJ, Lawrence C, Wagman L, eds. Cancer Management: A Multidisciplinary Approach. 2nd edn. Huntington, NY: PRR, 1998: 113–48. 70. Maini CL, Scelsa MG, Fiumara C, et al. Superselective intra-arterial radiometabolic therapy with I-131 lipiodol in hepatocellular carcinoma. Clin Nucl Med 1996; 21: 221–6. 71. Ryu M, Shimamura Y, Kinoshita T, et al. Therapeutic results of resection, transcatheter arterial embolization and percutaneous transhepatic ethanol injection in 3225 patients with hepatocellular carcinoma: a retrospective multicenter study. Jpn J Clin Oncol 1997; 27: 251–7. 72. Inamori H, Ido K, Isoda N, et al. Laparoscopic radiofrequency ablation of hepatocellular carcinoma in the caudate lobe by using a new laparoscopic US probe with a forward-viewing convex-array transducer. Gastrointest Endosc 2004; 60: 628–31. 73. Koda M, Ueki M, Maeda Y, et al. Percutaneous sonographically guided radiofrequency ablation with artificial pleural effusion for hepatocellular carcinoma located under the diaphragm. AJR 2004; 183: 583–8. 74. Raut CP, Izzo F, Marra P, et al. Significant long-term survival after radiofrequency ablation of unresectable hepatocellular carcinoma in patients with cirrhosis. Ann Surg Oncol 2005; 12: 616–28. 75. Santambrogio R, Podda M, Zuin M, et al. Safety and efficacy of laparoscopic radiofrequency ablation of hepatocellular carcinoma in patients with liver cirrhosis. Surg Endosc 2003; 17: 1826–32. 76. Filippou DK, Vezakis A, Filippou G, et al. A rare case of ectopic pancreas in the ampulla of Vater presented with obstructive jaundice. Ann Ital Chir 2006; 77: 517–9. 77. Nakajima H, Kambayashi M, Okubo H, et al. Annular pancreas accompanied by an ectopic pancreas in the adult: a case report. Endoscopy 1995; 27: 713. 78. Ozcan C, Celik A, Guclu C, et al. A rare cause of gastric outlet obstruction in the newborn: Pyloric ectopic pancreas. J Pediatr Surg 2002; 37: 119–20. 79. Brambs HJ. Developmental anomalies and congenital diseases of the pancreas. Radiologe 1996; 36: 381–8. 80. Skandalakis JE, Gray SW, Ricketts R. The pancreas. In: Skandalakis JE, Gray SW, eds. Embrylology for Surgeons. 2nd edn. Baltimore: Lippincott, 1994: 381. 81. Sencan A, Mir E, Gunsar C, et al. Symptomatic annular pancreas in newborns. Med Sci Monit 2002; 8: CR434–7. 82. Berrocal T, Torres I, Gutierrez J, et al. Congenital anomalies of the upper gastrointestinal tract. Radiographics 1999; 19: 855–72. 83. Adzick NS, Shamberger RC, Winter HS, et al. Surgical treatment of pancreas divisum causing pancreatitis in children. J Pediatr Surg 1989; 24: 54–8; discussion 8. 84. Stimec B, Bulajic M, Korneti V, et al. Ductal morphometry of ventral pancreas in pancreas divisum. Comparison between clinical and anatomical results. Ital J Gastroenterol 1996; 28: 76–80. 85. Beger HG, Rau B, Mayer J, et al. Natural course of acute pancreatitis. World J Surg 1997; 21: 130–5. 86. Waldemar H, Buchler U, Buchler MW. Classification and severity staging of acute pancreatitis. Ann Ital Chir 1995; 66: 171–9. 87. Sternby B, O’Brien JF, Zinsmeister AR, et al. What is the best biochemical test to diagnose acute pancreatitis? A prospective clinical study. Mayo Clin Proc 1996; 71: 1138–44. 88. Fujiwara T, Takehara Y, Ichijo K, et al. Anterior extension of acute pancreatitis: CT findings. J Comput Assist Tomogr 1995; 19: 963–6. 89. Prasil P, Laberge JM, Barkun A, et al. Endoscopic retrograde cholangiopancreatography in children: a surgeon’s perspective. J Pediatr Surg 2001; 36: 733–5. 90. Neblett WW 3rd, O’Neill JA Jr. Surgical management of recurrent pancreatitis in children with pancreas divisum. Ann Surg 2000; 231: 899–908. 91. Ihse I, Andersson R, Andren-Sandberg A, et al. Conservative treatment in acute pancreatitis. Ann Ital Chir 1995; 66: 181–5. 92. Kaufmann P, Hofmann G, Smolle KH, et al. Intensive care management of acute pancreatitis: recognition of patients at high risk of developing severe or fatal complications. Wien Klin Wochenschr 1996; 108: 9–15. 93. Hwang TL, Chiu CT, Chen HM, et al. Surgical results for severe acute pancreatitis--comparison of the different surgical procedures. Hepatogastroenterology 1995; 42: 1026–9. 94. Beger HG, Rau B. Necrosectomy and postoperative local lavage in necrotizing pancreatitis. Ann Ital Chir 1995; 66: 209–15. 95. Wilson C. Management of the later complications of severe acute pancreatitis – pseudocyst, abscess and fistula. Eur J Gastroenterol Hepatol 1997; 9: 117–21. 96. Bittner R. Clinical significance and management of pancreatic abscess and infected necrosis complicating acute pancreatitis. Ann Ital Chir 1995; 66: 217–22. 97. Kisra M, Ettayebi F, Benhammou M. Pseudocysts of the pancreas in children in Morocco. J Pediatr Surg 1999; 34: 1327–9. 98. Shimizu M, Hirokawa M, Manabe T. Histological assessment of chronic pancreatitis at necropsy. J Clin Pathol 1996; 49: 913–5. 99. Sidhu SS, Tandon RK. The pathogenesis of chronic pancreatitis. Postgrad Med J 1995; 71: 67–70. 100. Dite P, Zboril V, Cikankova E. Endoscopic therapy of chronic pancreatitis. Hepatogastroenterology 1996; 43: 1633–7.

486

PEDIATRIC HPB DISORDERS
101. Soliman AT, Alsalmi I, Darwish A, et al. Growth and endocrine function after near total pancreatectomy for hyperinsulinaemic hypoglycaemia. Arch Dis Child 1996; 74: 379–85. 102. Taguchi T, Suita S, Ohkubo K, et al. Mutations in the sulfonylurea receptor gene in relation to the long-term outcome of persistent hyperinsulinemic hypoglycemia of infancy. J Pediatr Surg 2002; 37: 593–8. 103. Murakami T, Ueki K, Kawakami H, et al. Pancreatoblastoma: case report and review of treatment in the literature. Med Pediatr Oncol 1996; 27: 193–7. 104. Finegold MJ, et al. Liver tumors: pediatric population. Liver Transpl 2008; 14(11): 1545–56 105. Sokol RJ, et al. Screening and outcomes in biliary atresia: summary of a National Institutes of Health workshop. Hepatology 2007; 46(2): 566–81 106. Bassett MD, Murray KF, Biliary atresia: recent progress. J Clin Gastroenterol 2008; 42(6): 720–9. 107. Chardot C, Serinet MO. Prognosis of biliary atresia: what can be further improved? J Pediatr 2006; 148(4): 432–5

487

Index
AAST. See American Association for Surgery of Trauma Abdominal compartment syndrome, 274 Abdominal pain, 451 Ablative techniques, 57–58 ABO-incompatible liver, 291 ACC. See Acinar cell carcinoma Acinar cell carcinoma, 432 pancreatic lesions, 107 ACS. See Abdominal compartment syndrome Acute and chronic liver failure auxiliary liver transplantation auxiliary whole orthotopic liver transplantation, 294–295 immunosuppression, 295 outcome of, 295 donor selection ABO-incompatible liver, 291 donation after cardiac death, 290–291 donors with infection, 290 donors with malignancy, 290 elderly donors, 289–290 hepatitis B core antibody positive donors, 290 hepatitis C infection donors, 290 split-liver, 290 steatosis and abnormal liver function, 289 future perspectives hepatocyte transplantation, 297 immune tolerance, 297 liver support devices, 295, 297 orthotopic liver transplantation immunosuppression, 292–293 operative technique of, 292 outcomes in, 293–294 post-operative management, 292–293 recipient selection with acute liver failure, 288 end-stage liver disease, 288 with HIV, 288–289 with viral hepatitis, 288 retrieval of deceased donor liver graft donation after cardiac death retrieval, 292 standard retrieval, 291 splitting of deceased donor liver graft, 292 Acute cholangitis, gallstone disease, 374 Acute cholecystitis bile duct injuries, 360–362 gallstone disease, 373 Acute pancreatitis congenital anomalies, 483–484 early management of, 439–441 etiology of, 439 gallstone disease, 374–375 laparoscopy, 443 necrosis management, 441 radiological assessment, 441 role for specific interventions, 440 surgical intervention endoscopic drainage, 443, 445 laparotomy/debridement, 441–443 open surgery, 441 percutaneous catheter drainage, 443–444 percutaneous necrosectomy, 443, 446 Acute variceal hemorrhage, 282 ADA. See American Diabetes Association Adenocarcinoma, congenital anomalies, 484 Adenosquamous carcinoma, 432–434 Adjuvant chemotherapy, 391–393 hepatic metastasectomy, 58–59 Adjuvant intra-arterial chemotherapy, 123 Adjuvant regional chemotherapy, 141–142 Adjuvant systemic chemotherapy after liver resection, 141 colorectal liver metastases, 122 Adjuvant therapy gallbladder cancer, 331 laparoscopically discovered disease, 204–205 Adult hemangioma, 262 Adulthood, choledochal cyst clinical presentation, 354 complications, 355 diagnosis, 354–355 etiology and classification, 354 incidence and pathophysiology, 354 treatment, 355–358 Advanced cholangiocarcinoma combined liver and portal vein resection, 339–340 hepatopancreatoduodenectomy, 340 Aging, 46 AIP. See Autoimmune pancreatitis AJCC. See American Joint Committee on Cancer ALT. See Auxiliary liver transplantation Amebiasis, 253 Amebic liver abscess diagnosis of, 253–254 epidemiology of, 253 outcomes of, 255 pathogenesis of, 253 treatment of, 254–255 American Association for Surgery of Trauma, 271–272 American Diabetes Association, 470 American Joint Committee on Cancer, 192, 198 American Pancreas Club, 458 American Society of Anesthesiology, 47 American Society of Clinical Oncology, 137 Anatomic right trisectionectomy, 339 Anatomical hepatectomies, 5–6 Anatomical hepatic resection, 276–277 Anatomy of pancreas arterial anatomy anterior and posterior inferior pancreaticoduodenal arteries, 19 caudal and great pancreatic arteries, 20 dorsal pancreatic artery, 19–20 inferior pancreaticoduodenal artery, 19 posterior superior pancreaticoduodenal artery, 18–19 superior pancreaticoduodenal artery, 18 venous drainage of pancreas, 20–21 ductal anatomy of, 17 innervation of, 22 lymphatic drainage, 21–22 topography of, 17 Angioembolization, 275 Angiography, hydatid cyst, 313 Angiomyolipoma, 267 Annular pancreas, congenital anomalies, 483 Anti-angiogenic therapies, 219–220 ASA. See American Society of Anesthesiology ASC. See Adenosquamous carcinoma Aschoff–Rokitansky sinuses, 40 Ascites, 280, 284–285 ASCO. See American Society of Clinical Oncology Aspiration sclerotherapy, 301–302 laparoscopic surgery, 303 open surgery, 303 surgical treatment, 302–303 Associated intra-abdominal vascular injuries, 463 Autoimmune pancreatitis chronic pancreatitis, 452 rare tumors, 436–437 Auxiliary liver transplantation auxiliary whole orthotopic liver transplantation recipient selection criteria, 294 surgical technique, 294–295 immunosuppression, 295 outcome of, 295 Auxiliary whole orthotopic liver transplantation recipient selection criteria, 294 surgical technique, 294–295 AWOLT. See Auxiliary whole orthotopic liver transplantation BA. See Biliary atresia Backtable preparation, 472 BCS. See Budd–Chiari syndrome BD. See Biliary drainage

489

INDEX
Benign cystic diseases aspiration sclerotherapy, 301–303 polycystic liver disease, 303–304 rare lesions, 305 simple biliary hepatic cysts, 301 Benign disease, 24 Benign inflammatory pseudotumors, 370 Benign solid tumors adult hemangioma, 262 capillary hemangioma, 262 cavernous hemangioma imaging features of, 263 management of, 263 pathology of, 262 classification of, 262 congenital hepatic fibrosis, 267 focal nodular hyperplasia imaging features of, 264 management of, 264 pathology of, 263 telangiectatic, 263 hepatocellular adenoma imaging features of, 265 management of, 265 pathology of, 264–265 hepatocellular adenomatosis management of, 266–267 pathology of, 265–266 nodular regenerative hyperplasia, 266 pseudolipoma hereditary hemorrhagic telangiectasia, 268 heterotopic tissue, 268 inflammatory pseudotumor, 268 miscellaneous rare benign solid liver lesions, 268 peliosis hepatic, 267–268 Benzimidazole, 321 Bevacizumab, 140, 176, 220 Bile duct hamartomas, 305 ultrasound applications, 39–41 Bile duct injuries noniatrogenic bile duct strictures benign inflammatory pseudotumors, 370 biliary strictures secondary to pancreatitis, 369–370 calculous disease, 370 sclerosing cholangitis, 370 patient-related factors acute cholecystitis, 360–362 congenital abnormalities, 362–363 procedure-related factors critical view technique, 365 misidentification concepts, 363–365 technical problems, 365 surgeon/hospital-related factors laparoscopic equipment, 366 learning curve effect, 366 psychology of human error, 366 Bile ducts and liver, surgical anatomy anatomical hepatectomies, 5–6 anatomy of biliary exposure, 12–13 arterial blood supply of, 11–12 biliary anatomy, 10–11 biliary tract, 6 caudate lobe, surgical approach, 6 cystic duct, 9–10 early application of functional anatomy, 1 extrahepatic biliary anatomy, 8–9 falciform ligament, 14 gallbladder, 9–10, 14 hepatic veins, 13 intrahepatic biliary anatomy, 7–8 ligamentum venosum, 14 morphological anatomy, 1 portal system, 13–14 radiological anatomy of, 13 segmental anatomy of, 1–5 Bile fistula, 367 Bile leak, hepatectomy incidence, 65 prevention, 65 risk factors, 65–66 Bile pigment stone, 373 Biliary adenoma, 267 Biliary atresia classification, 478 clinical features, 478 outcomes, 478 pathology, 478 surgical treatment, 478 Biliary colic, 373 Biliary decompression nonoperative technique, 402 operative technique, 402 Biliary drainage, 333–334 Biliary dyspepsia, 375 Biliary hamartoma, 266–267 Biliary injuries avoidance of, 366–367 classification of, 360 iatrogenic, 360 incidence laparoscopic versus open cholecystectomy, 360 population-based studies, 360 management of intraoperative recognition, 367–368 operative technique, 368–369 preoperative preparation, 368 types of injuries, 368 outcome of treatment, 369 post-transplantation, 369 Biliary stricture, 356 after hepatectomy incidence of, 66 management of, 66 Bladder drainage, 473–474 Bland embolization, 216–218 Bleeding, hepatectomy incidence, 63 investigation and treatment, 65 presentation, 65 prevention, 63–65 Body habitus, 363 Borderline tumors, 385 Brachytherapy, malignant biliary obstruction, 349–350 Breast cancer, 54 noncolorectal, nonneuroendocrine metastases, 166 British Society of Gastroenterology, 346 BSG. See British Society of Gastroenterology Budd–Chiari syndrome, 280 CA19-9. See Carbohydrate antigen 19-9 Cambridge Classification of Chronic Pancreatitis, 451 Capecitabine, 173 Capillary hemangioma, 262 Carbamates, 321 Carbohydrate antigen 19-9, 382 Carcinoid symptom severity scale, 157 Carcinoid tumors, 154 Carcinoma fibrolamellar, 104 gallbladder, 104 hepatocellular, 104 Caroli’s disease, 357 Caspofungin, 259 Caudal pancreatic artery, 20 Caudate lobe, surgical approach, 6 Caval injury, 276 Cavernous hemangioma imaging features of, 263 management of, 263 pathology of, 262 CCA. See Cholangiocarcinoma CECT. See Contrast-enhanced computerized tomography Central pancreatectomy, 85 Cetuximab, 137, 176 CgA. See chromogranin A Charcot’s Triad of symptoms, 374 Chemical ablation, 195 Chemical splanchnicectomy, 403–404 Chemo-embolization, 216–218 Chemotherapeutic agents capecitabine, 173 irinotecan, 135 oxaliplatin, 135 Chemotherapeutic regimens, 173 Chemotherapy neuroendocrine tumors, 161 non-functional islet cell tumors, 428 Chemotherapy-associated hepatotoxicity diagnosis, 176–177 monoclonal antibodies, 176 nonalcoholic fatty liver disease sinusoidal obstruction syndrome, 174–176 steatohepatitis, 174 steatosis, 173–174 preoperative chemotherapy, 178

490

INDEX
prevention, 177 Chemotherapy-associated nonalcoholic fatty liver disease diagnosis of, 173 sinusoidal obstruction syndrome, 174–176 steatohepatitis, 174 steatosis, 173–174 CHF. See Congenital hepatic fibrosis Child–Turcotte–Pugh Score, 289 Chlorozotocin, 428 Cholangiocarcinoma liver and biliary tract lesions, 104 liver transplantation neoadjuvant chemoradiotherapy, 229–231 organ allocation, 231 Cholecystenteric fistula, 374 Choledochal cysts adulthood clinical presentation, 354 complications, 355 diagnosis, 354–355 etiology and classification, 354 incidence and pathophysiology, 354 treatment, 355–358 classification, 479 clinical features, 480 imaging, 480 outcomes, 480 pathology, 479–480 surgical treatment, 480 Choledochoduodenostomy, 91 Choledocholithiasis, 370 Choledochotomy, 90–91 Cholelithiasis, 480 Cholestatic neonate, diagnostic evaluation in, 479 Cholesterol stones, 373 chromogranin A, 154 Chronic pancreatitis classification of, 451 comparison of different surgical approaches, 457–458 complications of, 453–454 congenital anomalies, 484 conservative treatment, 454 definition of, 451 duodenum-preserving resection Beger procedure, 456–457 Berne procedure, 457 Hamburg procedure, 457 quality of life, 458 etiology pathomorphological findings, 452 indications for surgical intervention, 454–455 interventional treatment, 454 natural course of, 451 pathogenesis of pain, 452–453 rationale for drainage procedures, 455–456 rationale for resectional procedures, 456 salvage procedures, 459 small duct disease, 458–459 ultrasound findings in, 41 Ciliated foregut cysts, 305 Cirrhosis, 356 Cirrhotic liver, 266 cisPlatin, Interferon, Adriamycin, and 5-Fluorouracil, 219 Clinical risk score, 56 metastatic colorectal cancer, 120–121 Clonorchis sinensis, 242 CMC. See Conventional Milan Criteria Color Doppler, 36 Colorectal cancer, 47–48 Colorectal liver metastases, 47–48 clinical risk scores, 120–121 hepatic arterial infusion, 136–137 multimodal strategies chemotherapy and surgery, 149–150 computed tomography, 148–149 magnetic resonance imaging, 149 management strategies, 150–152 multidisciplinary team, 148 positron emission tomography, 149 preoperative staging, 148 resectability of, 149 resection margins, 149 strategies to improve respectability, 149 surgery, 149 tumor ablation, 150 natural history of, 118 patient evaluation patient selection, 118 preoperative imaging, 118–119 tumor resectability, 118 postoperative management adjuvant intra-arterial chemotherapy, 123 adjuvant systemic chemotherapy, 122 nonresectable metastatic disease, 124–126 outcomes of resection, 123–124 repeat liver resection, 129–130 resectability techniques, 126–129 prognostic factors, 119–120 resectable management preoperative management, 121 surgery approaches, 121–122 thermal ablation CLOCC study, 184 cryotherapy, 180–181 edge cryotherapy, 181 limitations of, 180 microwave coagulation, 183–184 percutaneous ethanol injection, 184 radiofrequency ablation, 181–183 Combined liver and portal vein resection, 339–340 Common bile duct (CBD) stenosis, 453 Common bile duct stones non-surgical management drug dissolution therapy, 377 endoscopic removal of bile duct stones/biliary stenting, 377 lithotripsy, 377 surgical management laparoscopic common bile duct stone removal, 378 open choledocholithotomy, 377 Computed tomography choledochal cyst, 355 colorectal liver metastases, 148–149 hydatid cyst, 311 liver and biliary tract lesions cholangiocarcinoma, 104 cross-sectional anatomy, 100 fibrolamellar carcinoma, 104 focal nodular hyperplasia, 102 gallbladder carcinoma, 104 hepatic hemangioma, 102 hepatocellular adenoma, 102, 104 hepatocellular carcinoma, 104 metastatic cancer to liver, 104 liver metastases, 109–110 localization of insulinomas, 415 neuroendocrine tumors, 155 pancreatic ductal adenocarcinoma, 382–383 pancreatic injuries, 463 pancreatic lesions acinar cell carcinoma, 107 cross-sectional anatomy, 100 metastatic cancer to pancreas, 107 pancreatic adenocarcinoma, 107 pancreatic neuroendocrine tumors, 106 solid pseudopapillary tumor, 106–107 scanning, radiological anatomy of liver, 13 Congenital abnormalities, bile duct injuries, 362–363 Congenital anomalies acute pancreatitis, 483–484 adenocarcinoma, 484 annular pancreas, 483 chronic pancreatitis, 484 ectopic pancreatic rests, 483 pancreas divisum, 483 pancreatic abscess, 484 pancreatic pseudocyst, 484 persistent hyperinsulinemic hypoglycemia of infancy, 484 Congenital deformities, 480–481 Congenital hepatic fibrosis, 267 CONKO-1, randomized controlled trail, 391–392 Contrast-enhanced computerized tomography, 192 Contrast-enhanced ultrasound, liver metastases, 109 Conventional Milan Criteria, 193 Covered stents, 349 CP. See Central pancreatectomy; Chronic pancreatitis CRLM. See Colorectal liver metastases Cross-sectional imaging. See CT imaging, MRI characteristics CRS. See Clinical risk score Cryoablation, neuroendocrine tumors, 160

491

INDEX
Cryotherapy, 126 thermal ablation, 180–181 Cryptogenic abscess, 256 CT. See Computed tomography Curative surgical resection, 394 Cyst aspiration, 301 Cyst fluid analysis, 411 Cystic dilatation. See Choledochal cyst Cystic fibrosis, 380–381 Cystic lesions, 36–37 Cystic metastases, 113 Cystic pancreatic neoplasms, 41 Cystic tumors clinical scenarios presentation and diagnostic evaluation, 407 treatment, 412 diagnostic evaluation, 411 pathologic sub-types and clinical behavior intraductal papillary mucinous neoplasm, 409–410 mucinous cystic neoplasm, 410–411 pancreatic pseudocyst, 407 serous cystadenoma, 407–409 treatment recommendations intraductal papillary mucinous neoplasm, 412 mucinous cystic neoplasm, 412 DCCT. See Diabetes Control and Complication Trial DCD. See Donation after cardiac death DDLT. See Deceased donor liver transplantation Deceased donor liver graft retrieval of donation after cardiac death retrieval, 292 standard retrieval, 291 splitting of, 292 Deceased donor liver transplantation, 208–213 Delayed gastric emptying, 81, 389 Detectable metastases, 401 Dexamethasone, 136 DGE. See Delayed gastric emptying Diabetes Control and Complication Trial, 475 Diabetic nephropathy, 475 Diagnostic laparoscopy palliation of pancreas cancer, 401–402 pancreatic ductal adenocarcinoma, 383, 385 Diagnostic radiography, 402 Diarrhea, 417, 421 Diet, 380 Diffuse liver disease liver, 36 pancreas, 41 Diffusion-weighted MR imaging, 111 Direct cholangioscopy, 346 endoscopic therapy, 347 methods of therapy, 347 percutaneous drainage of jaundice, 348 Distal cholangiocarcinoma, 336–337 Distal ductal pancreatic injuries, 466 Distal pancreatectomy, 73–74, 416 Distal splenorenal shunt, 283–284 Donation after cardiac death, 290–291 Donor, pancreas, 471 Doppler ultrasound, 329 Dorsal liver dissection, 338 Dorsal pancreatic artery, 19–20 DPPHR. See Duodenum-preserving pancreatic head resection Drug dissolution therapy common bile duct stones, 377 gallstone disease, 375 Drug-eluting microspheres, 218 DSRS. See Distal splenorenal shunt Ductal anatomy of pancreas, 17 Duodenal outlet obstruction, 402 Duodenal tumors, 419 Duodenum-preserving pancreatic head resection, 75–76. See also Chronic pancreatitis Dysplastic nodules, 266 Eastern Cooperative Oncology Group, 421 EBRT. See External beam radiotherapy ECD. See Extended/Expanded criteria donor Echinococcal cysts, 102 ECOG. See European Cooperative Oncology Group Ectopic insulinomas, 414 Ectopic pancreatic rest, 483 Edge cryotherapy, thermal ablation, 181 EHD. See Extrahepatic disease Elderly patients, liver surgery age-related liver changes, 46–47 colorectal liver metastases, 47–48 financial cost, 50 hepatocellular carcinoma, 48–50 surgical risk evaluation, 47 ELTR. See European Liver Transplant Registry Empyema, 373–374 Endogenic vesiculation, 308 Endoluminal ultrasound cystic tumors, 411 pancreatic ductal adenocarcinoma, 383 Endoscopic assessment, malignant biliary obstruction causes of, 343 covered vs. uncovered stents, 349 CT scanning, 343 direct cholangioscopy, 346 endoscopic therapy, 347 methods of therapy, 347 percutaneous drainage of jaundice, 348 endoscopic intervention, 344 endoscopic retrograde cholangio-pancreatography, 344–346 ERCP vs. PTC vs. surgery, 348 hilar strictures disease modifying treatment, 349 unilateral versus bilateral, 349 MR scanning, 343–344 plastic vs. metal, 348–349 radiological diagnostic imaging, 343 radiotherapy brachytherapy, 349–350 photodynamic therapy, 350 ultrasound, 343 Endoscopic drainage, 443, 445 Endoscopic pancreatic sphincterotomy, 454 Endoscopic retrograde cholangiopancreatography acute pancreatitis antibiotics, 440 nutrition, 440 specific pharmacological intervention, 440 surgical intervention, 440 cystic tumors, 411 gallbladder cancer, 329 gallstone disease, 377 malignant biliary obstruction, 344–346 pancreatic ductal adenocarcinoma, 383, 385 Endoscopic retrograde pancreatography, 464 Endoscopic therapy, 282–283 Endoscopic ultrasound, 415–416 malignant biliary obstruction, 344 Endoscopic variceal ligation, 282 End-stage liver disease, 288 Enteral feeding, 391 Enteric drainage, 474 Enucleation, 76–77, 416 EORTC. See European Organization for Research and Treatment of Cancer EPIC. See Erbitux Plus Irinotecan in Colorectal Cancer Erbitux Plus Irinotecan in Colorectal Cancer, 137 ERCP. See Endoscopic retrograde cholangio-pancreaticography ESLD. See end-stage liver disease ESPAC-1, randomized controlled trail, 391–393 ESWL. See Extracorporal shockwave lithotripsy European Cooperative Oncology Group, 137 European Liver Transplant Registry, 233 European Organization for Research and Treatment of Cancer, 141 EUS. See Endoluminal ultrasound EVL. See Endoscopic variceal ligation Exogenous vesiculation, 308, 310 Extended lymphadenectomy, 84–85, 387 Extended resections, 387 Extended/Expanded criteria donor ABO-incompatible liver, 291 donation after cardiac death, 290–291

492

INDEX
donors with infection, 290 donors with malignancy, 290 elderly donors, 289–290 hepatitis B core antibody positive donors, 290 hepatitis C infection donors, 290 split-liver, 290 steatosis and abnormal liver function, 289 Extensive locoregional disease, 401 External beam radiotherapy, 230 Extracorporal shockwave lithotripsy, 454 Extrahepatic biliary anatomy, 8–9 Extrahepatic cholangiocarcinoma operative procedures combined liver and portal vein resection, 339–340 distal cholangiocarcinoma, 336–337 hepatobiliary resection for hilar cholangiocarcinoma, 337–339 hepatopancreatoduodenectomy, 340 pancreatoduodenectomy, 336–337 preoperative management biliary drainage, 333–334 portal vein embolization, 334 staging of cholangiocarcinoma, 333 synbiotics treatments with bile replacement, 334–336 surgical anatomy of bile duct, 333 Extrahepatic disease, 181–183 Extrahepatic portal hypertension, 453 [18F] 2-fluoro-2-deoxyglucose, 111–112 Familial pancreatic cancer syndrome, 380 Familial predisposition, 380 FAST. See Focused Assessment for Sonographic examination of Trauma patient FDA. See Food and Drug Administration FDG. See [18F] 2-fluoro-2-deoxyglucose; Fluorine-18-labeled fluoro-deoxyglucose Fertile cyst, 315 Fibrolamellar carcinoma, 104 Fine needle aspiration, 383, 440 FISH. See Fluorescent in situ hybridization FLC. See Fibrolamellar carcinoma Floxuridine, 136–137 FLR. See Future liver remnant Fluorescent in situ hybridization, 229 Fluorine-18-labeled fluoro-deoxyglucose, 329 Fluorouracil, 135 5-Fluorouracil, 58–59 FNA. See Fine needle aspiration FNH. See Focal nodular hyperplasia FNH-like lesions, 264 Focal fatty variants, 267 Focal hepatic lesions ultrasound applications cystic lesions, 36–37 solid liver lesions, 37–39 Focal nodular hyperplasia imaging features of, 264 liver and biliary tract lesions, 102 management of, 264 pathology of, 263 telangiectatic, 263 Focused Assessment for Sonographic examination of Trauma patient, 271 FOLFIRI, 121 FOLFOX, 121 Food and Drug Administration, 176, 218 FPC. See Familial pancreatic cancer syndrome FUDR. See Floxuridine Functional endocrine tumors comparison of, 415 definition, 414 Functional islet cell tumors definition of, 414 gastrinomas management of, 419 precise localization of, 417 preoperative evaluation, 418 sensitivity of invasive and non-invasive imaging studies, 418 symptoms of, 417 glucagonomas, 419–420 insulinoma biochemical diagnosis of, 415 ectopic, 414 endoscopic ultrasound, 415–416 intraoperative ultrasonography, 416 localization modalities for, 415–416 symptoms, 415 somatostatinomas, 421 VIPomas, 420–421 Fungal liver abscess, 258–259 Future liver remnant, 67, 177 Gallbladder cancer adjuvant therapy, 331 clinical presentation and work-up, 329–330 history of, 329 laparoscopically discovered disease adjuvant therapy, 204–205 clinical presentation of, 199–200 epidemiology of, 197–198 palliative management, 205 pathology of, 198 patterns of spread, 198 radiologic workup, 200 staging systems, 198–199 surgical management, 201–204 liver and biliary tract lesions, 104 outcomes, 331 palliative care, 331–332 surgical management, 330–331 ultrasound applications, 39–41 Gallbladder disease cholelithiasis, 480 congenital deformities, 480–481 hemolytic cholelithiasis, 480 Gallstone acute cholangitis, 374 acute cholecystitis, 373 acute pancreatitis, 374–375 biliary colic, 373 biliary dyspepsia, 375 cholecystenteric fistula, 374 classification of, 373 empyema, 373–374 gallstone ileus, 374 ileus, 374 Mirrizi’s syndrome, 375 mucocele, 374 non-surgical management analgesia for biliary colic/ cholecystitis, 375 drug dissolution therapy, 375 percutaneous cholecystostomy, 375 obstructive jaundice, 374 surgical management, cholecystectomy laparoscopic cholecystectomy, 376 open cholecystectomy, 375–376 operative technique, 376–377 Gastric cancer, noncolorectal, nonneuroendocrine metastases, 167–168 Gastric decompression nonoperative technique, 402–403 operative technique, 403 Gastric outlet obstruction, 401–402 Gastrinomas management of, 419 precise localization of, 417 preoperative evaluation, 418 sensitivity of invasive and non-invasive imaging studies, 418 symptoms of, 417 Gastrointestinal cancers, 55 Gastrojejunostomy, 402 isoperistaltic, 403 palliative, 403 Gelfoam, 216 Genitourinary tumors, 55 Giant cell tumors, 435 Giant hemangiomas, 37 β-Glucuronidase, 242 Glucagonomas, 419–420 Grade A fistulas, 389 Grade B fistulas, 389 Grade C fistulas, 389–390 Great pancreatic artery, 20 Gynecological tumors, 166 HAE. See Hepatic artery embolization HAI. See Hepatic arterial infusion HALT. See Heterotopic auxiliary liver transplantation HAS. See Hepatic hemangiosarcoma HBV. See Hepatitis B Virus HCC. See Hepatocellular carcinoma Head and neck tumors, 168 HEHE. See Hepatic epithelioid hemangioendothelioma

493

INDEX
Hemangioma, 37 adult, 262 Hemolytic cholelithiasis, 480 Hemorrhage, 273–274 Hepatectomy, 5–6 bile leak consequences of, 66 incidence, 65 management of, 66 presentation, 66 prevention, 65 risk factors, 65–66 biliary stricture incidence of, 66 management of, 66 bleeding incidence, 63 investigation and treatment, 65 presentation, 65 prevention, 63–65 cardiac complications, 69 hepatic insufficiency incidence, 66 optimization of venous drainage, 67 portal vein embolization, 67 prevention, 67 treatment, 67–68 intra-abdominal infection abdominal drainage, 68 factors affecting, 68 pain relief, 68–69 renal failure consequences of, 69–70 etiology of, 69 renal impairment, 70 respiratory complications, 68–69 wound complications, 70 Hepatic abscess. See Liver abscess Hepatic arterial infusion, 136–137 Hepatic arterial infusional chemotherapy, 59 Hepatic artery embolization, 427 neuroendocrine tumors, 156–158 Hepatic cryoablation, 160 Hepatic epithelioid hemangioendothelioma, 233–235 Hepatic hemangioma, 102 Hepatic hemangiosarcoma, 237–238 Hepatic infantile hemangioendothelioma, 235–237 Hepatic metastasectomy adjuvant chemotherapy, 58–59 colorectal metastases, 53 neuroendocrine metastases, 53 noncolorectal metastases, 54–55 nonneuroendocrine metastases, 54–55 patient selection colorectal metastases, 55 neuroendocrine tumors, 56 noncolorectal tumors, 56 nonneuroendocrine tumors, 56 resection techniques ablative techniques, 57–58 morbidity and mortality, 57 segmental resections, 57 wedge resections, 56–57 Hepatic metastases, neuroendocrine tumors clinical features, 154 cryoablation, 160 diagnostic imaging, 155 hepatic artery embolization, 156–158 hepatic resection, 156 laboratory investigation, 154 liver transplantation, 160–161 medical treatment, 161 radiofrequency ablation, 158, 160 radionuclide therapy, 161 WHO classification of, 155 Hepatic portoenterostomy, 478 Hepatic radiofrequency ablation, 158, 160 Hepatic resection basic principles anesthetic techniques, 25 benign disease, 24 malignant disease, 24 patient selection, 24–25 preoperative imaging, 25 basic techniques exposure, 25 mobilization, 25–27 positioning, 25 skin incision, 25 left hemihepatectomy, 32 left lateral sectionectomy, 32–33 left trisectionectomy, 32 neuroendocrine tumors, 156 renal failure after consequences of, 69–70 etiology of, 69 right hemihepatectomy, 30–31 right trisectionectomy, 31 vascular isolation inflow control, 27–28 outflow control, 28–29 parenchymal transaction, 29 wedge vs. segmental resection central hepatectomy, 34 segment 4, 33 segmentectomy I, caudate resection, 33 segments 2/3, 33 segments 5 and 8, anterior sector, 33 segments 6 and 7, posterior sector, 33–34 Hepatic steatosis, 36 Hepaticojejunostomy, 337 Hepatitis B virus, 208 Hepatobiliary resection for hilar cholangiocarcinoma, 337–339 Hepatoblastoma clinical features, 481 imaging, 481 incidence, 481 outcomes, 481 pathology, 481 staging, 481 surgical treatment, 481 Hepatocellular adenoma imaging features of, 265 liver and biliary tract lesions, 102, 104 management of, 265 pathology of, 264–265 Hepatocellular adenomatosis management of, 266–267 pathology of, 265–266 Hepatocellular carcinoma advanced stage, 195 diagnosis, 192 early stages chemical ablation, 195 liver resection, 193–195 liver transplant, 195 thermal ablation, 195 intermediate stage, 195 liver and biliary tract lesions, 104 liver surgery, in elderly patients, 48–50 liver transplantation in Asia, 208 pretransplant neoadjuvant therapy, 211–212 recurrence treatment, 212–213 selection criteria and outcomes, 209–211 local regional therapies arterial embolization, 216 bland embolization, 216–218 chemo-embolization, 216–218 drug-eluting microspheres, 218 percutaneous chemical/thermal ablation, 218–219 radio-embolization, 218 pediatric hepato-pancreato-biliary disorders clinical features, 482–483 incidence, 481 outcomes, 483 pathology, 482 surgical treatment, 483 staging systems, 192 systemic therapies advanced cirrhosis treatment, 220 anti-angiogenic therapies, 219–220 cisPlatin, interferon, adriamycin, and 5-fluorouracil, 219 future developments, 220–221 historical background, 219 treatment options, 192–193 ultrasound applications, 38 variceal bleeding, 281 Hepatocyte transplantation, 297 Hepatopancreatoduodenectomy, 340 Hepatopulmonary syndrome, 285 Hepatotomy, 275–276 Hereditary chronic pancreatitis, 452 Hereditary hemorrhagic telangiectasia, 268 Hereditary pancreatitis, 380 Hereditary tumor syndromes, 380 Heterotopic auxiliary liver transplantation, 295

494

INDEX
Heterotopic tissue, 268 HHT. See Hereditary hemorrhagic telangiectasia 5-HIAA. See 5-Hydroxyindoleacetic acid High-affinity somatostatin receptors, 424 High-grade pancreatic injuries, 466 HIHE. See Hepatic infantile hemangioendothelioma Hilar cholangiocarcinoma anatomic right trisectionectomy, 339 intrahepatic cholangiojejunostomy, 339 left trisectionectomy, 338 left-sided hepatectomy, 337 liver transplantation neoadjuvant chemoradiotherapy, 229–231 organ allocation, 231 right hepatectomy, 338–339 right trisectionectomy, 338 Hilar strictures, malignant biliary obstruction disease modifying treatment, 349 unilateral vs. bilateral, 349 Histidine–Tryptophan–Ketoglutarate, 471 “Hockey stick” incision, 25 HPD. See Hepatopancreatoduodenectomy HPE. See Hepatic portoenterostomy HPS. See Hepatopulmonary syndrome HT. See Hepatocyte transplantation HTK. See Histidine–Tryptophan– Ketoglutarate 5-HTP. See 5-Hydroxytryptophan Human error, psychology of, 366 Hydatid cyst, 102 biological basis of surgery, 308 complications infection, 313 rupture, 313–314 conservative procedures, 316–317 diagnostic imaging angiography, 313 computed tomography, 311 magnetic resonance imaging, 311 radioisotope imaging, 313 ultrasonography, 311 intraoperative approach, 315–316 laparoscopy, 320 medical therapy, 321 pathological basis of surgery, 308 puncture aspiration injection reaspiration, 320–321 radical procedures, 317–320 serology of, 313 structure of, 308–310 topography, 314–315 treatment, 315 5-Hydroxyindoleacetic acid, 154 5-Hydroxytryptophan, 424 Hyperglycemia, 471 Hypersplenism, 280 Hypervascular metastases, 113 Hypoechoic halo, 38 Hypoechoic liver, 39 Iatrogenic biliary injuries, 360 ICC. See Intrahepatic cholangiocarcinoma IG-ICC. See Intraductal growth type of intrahepatic cholangiocarcinoma Immune tolerance, 297 Immunosuppression, 475 Indocyanine green (ICG) test, 67, 334 Infants, gallstones, 480 Inferior pancreaticoduodenal artery, 19 Inflammatory pseudotumor, 268 Infundibular technique, 364 Injury grading, 463 Institutional factors, 85 Insulinoma biochemical diagnosis of, 415 ectopic, 414 endoscopic ultrasound, 415–416 intraoperative ultrasonography, 416 laparoscopic enucleation, 93 localization modalities for, 415–416 symptoms, 415 Interferon, 161 International Pancreas Transplant Registry, 470 International Registry of Hepatic Metastases of Colorectal Cancer, 123 International Union against Cancer, 21 Intraabdominal abscesses, 390 Intra-abdominal infection, hepatectomy abdominal drainage, 68 factors affecting, 68 Intra-arterial chemotherapy, 125–126 Intra-arterial infusion therapies bland embolization, 216–218 chemo-embolization, 216–218 radio-embolization, 218 Intrabiliary metastases, 113 Intraductal growth type of intrahepatic cholangiocarcinoma, 223 Intraductal papillary mucinous neoplasm, 385, 409–410, 412 Intrahepatic biliary anatomy, 7–8 Intrahepatic cholangiocarcinoma, resection of adjuvant therapy, 226 classification and terminology, 223 epidemiology, 223 pre-operative diagnosis, 223–224 prognosis factors, 225–226 risk factors, 223 surgical strategy, 224–225 Intrahepatic cholangiojejunostomy, 339 Intrahepatic gallbladder, 363 Intrahepatic portal hypertension, 267, 280 Intraoperative cholangiography, 364–365 Intraoperative pancreatography pancreatic injuries, 465–466 Intraoperative ultrasonography, 416 Intraoperative ultrasound, 42–43 colorectal liver metastases, 122 Intravenous erythromycin, 389 IOC. See Intraoperative cholangiography IOUS, Intraoperative ultrasound IPMN. See Intraductal papillary mucinous neoplasm Irinotecan, 135 Isolated liver metastases, 109 Isolated pancreatic metastases, 435 Isoperistaltic gastrojejunostomy, 403 Japan Integrated Staging score, 192 Jaundice, 367 Juxtahepatic injury, 276 Laparoscopic cholecystectomy, 89 common bile duct stones, 378 gallstone disease, 376 Laparoscopic common bile duct exploration choledochoduodenostomy, 91 choledochotomy, 90–91 transcystic flushing, 89 transcystic stone extraction, 89–90 Laparoscopic cystgastrostomy, 443 Laparoscopic distal pancreatectomy, 93 Laparoscopic enucleation, 93 Laparoscopic liver resection, 93–95 Laparoscopic palliative bypass, 92 Laparoscopic pancreatectomy, 85, 92–93 Laparoscopic pancreaticoduodenectomy, 93 Laparoscopic port sites, 204 Laparoscopic staging, 91–92 Laparoscopically discovered gallbladder cancer adjuvant therapy, 204–205 clinical presentation of, 199–200 epidemiology of, 197–198 palliative management, 205 pathology of, 198 patterns of spread, 198 radiologic workup, 200 staging systems, 198–199 surgical management advanced tumors, 202 complications, 204 laparoscopic port sites, 204 liver resection, 203–204 lymph node dissection, 204 re-resection after laparoscopic cholecystectomy, 202–203 tumor invading into the subserosal layer, 201–202 tumors confined to the muscular propria, 201 Laparotomy/Debridement with closed lavage, 443 closed packing, 441 with drainage, 441 laparostomy with open packing, 441, 443 minimally invasive approaches, 443 Laser lithotripsy, 377 Late stricture, of choledochojejunostomy, 468 LDLT. See Living donor liver transplantation l-DOPA. See l-Dihydroxyphenylalanine Learning curve effect, 366 Left hemihepatectomy, 32

495

INDEX
Left lateral sectionectomy, 32–33 Left trisectionectomy hepatic resection, 32 hilar cholangiocarcinoma, 338 Left-sided hepatectomy, 337 Lexipafant, 440 LGSW. See Liver gunshot wounds Ligasure device, 94 Lipase, 483 Lipoma, 267 Lithotripsy, 377 Liver and bile ducts, surgical anatomy anatomical hepatectomies, 5–6 anatomy of biliary exposure, 12–13 arterial blood supply of, 11–12 biliary anatomy, 10–11 biliary tract, 6 caudate lobe, surgical approach, 6 cystic duct, 9–10 early application of functional anatomy, 1 extrahepatic biliary anatomy, 8–9 falciform ligament, 14 gallbladder, 9–10, 14 hepatic veins, 13 intrahepatic biliary anatomy, 7–8 ligamentum venosum, 14 morphological anatomy, 1 portal system, 13–14 radiological anatomy of, 13 segmental anatomy of, 1–5 and biliary tract lesions, CT and MRI imaging cholangiocarcinoma, 104 cross-sectional anatomy, 100 fibrolamellar carcinoma, 104 focal nodular hyperplasia, 102 gallbladder carcinoma, 104 hepatic hemangioma, 102 hepatocellular adenoma, 102, 104 hepatocellular carcinoma, 104 metastatic cancer to liver, 104 hydatid cyst biological basis of surgery, 308 complications, 313–314 conservative procedures, 316–317 diagnostic imaging, 311, 313 intraoperative approach, 315–316 laparoscopy, 320 medical therapy, 321 pathological basis of surgery, 308 puncture aspiration injection reaspiration, 320–321 radical procedures, 317–320 serology of, 313 structure of, 308–310 topography, 314–315 treatment, 315 ultrasound applications diffuse liver disease, 36 focal hepatic lesions, 36–39 Liver abscess, 356 amebic abscess diagnosis of, 253–254 epidemiology of, 253 outcomes of, 255 pathogenesis of, 253 treatment of, 254–255 fungal abscess, 258–259 pyogenic abscess diagnosis of, 256–257 epidemiology of, 255 microbiology of, 256 outcomes of, 258 pathogenesis of, 255–256 treatment of, 257–258 Liver biopsy, 478 Liver Cancer Study Group of Japan, 223 Liver failure, 48 Liver gunshot wounds, 274–275 Liver metastases computed tomography, 109–110 contrast-enhanced ultrasound, 109 detection, 114–115 imaging findings cystic metastases, 113 hypervascular metastases, 113 intrabiliary metastases, 113 pitfalls and limitations, 113 magnetic resonance imaging, 110–111 perfusion imaging, 112 positron emission tomography, 111–112 preoperative staging, 115–116 ultrasound techniques, 109 Liver resection laparoscopically discovered gallbladder cancer, 203–204 stages of hepatocellular carcinoma, 193–195 types of, 122 Liver support devices, 295, 297 Liver surgery, elderly patients age-related liver changes, 46–47 colorectal liver metastases, 47–48 financial cost, 50 hepatocellular carcinoma, 48–50 surgical risk evaluation, 47 Liver transplantation for acute and chronic liver failure auxiliary liver transplantation, 294–295 donor selection, 289–291 future perspectives, 295, 297 orthotopic liver transplantation, 292–294 recipient selection, 288–289 retrieval of deceased donor liver graft, 291–292 splitting of deceased donor liver graft, 292 in Asia, 208–209 hepatocellular carcinoma pretransplant neoadjuvant therapy, 211–212 recurrence treatment, 212–213 selection criteria and outcomes, 209–211 hilar cholangiocarcinoma neoadjuvant chemoradiotherapy, 229–231 organ allocation, 231 neuroendocrine tumors, 160–161 stages of hepatocellular carcinoma, 195 variceal bleeding, 284 Liver trauma anatomical hepatic resection, 276–277 classification of, 271 complications of non-operative management, 273–274 definitive surgical procedures, 275 early decision making during laparotomy, 275 hepatotomy, 275–276 liver gunshot wounds, 274–275 liver injuries, 274 manouevres, 276 non-anatomic liver resection, 276 non-operative management of liver injury, 272–273 perihepatic packing, 275 refractory bleeding, 275 selective vascular ligation, 275–276 LiverMetSurvey, 123 Living donor liver transplantation, 208–213 Living-donor pancreas procurement, 472 Local regional therapies arterial embolization, 216 bland embolization, 216–218 chemo-embolization, 216–218 drug-eluting microspheres, 218 percutaneous chemical/thermal ablation, 218–219 radio-embolization, 218 Low-grade pancreatic injuries, 466 LPSP. See Lymphoplasmacytic sclerosing pancreatitis LT. See Liver transplantation l–Dihydroxyphenylalanine, 424 Lymph node dissection, 337 laparoscopically discovered gallbladder cancer, 204 Lymphadenopathy, 343 Lymphatic drainage of pancreas, 21–22 Lymphoplasmacytic infiltration, 452 Lymphoplasmacytic sclerosing pancreatitis, 436 Macrocystic mucinous tumors, 41 Magnetic resonance cholangiography, 245 Magnetic resonance cholangiopancreaticography cystic tumors, 411 malignant biliary obstruction, 344 pancreatic ductal adenocarcinoma, 383 pancreatic injuries, 464 Magnetic resonance imaging colorectal liver metastases, 149 hydatid cyst, 311

496

INDEX
liver and biliary tract lesions cholangiocarcinoma, 104 cross-sectional anatomy, 100 fibrolamellar carcinoma, 104 focal nodular hyperplasia, 102 gallbladder carcinoma, 104 hepatic hemangioma, 102 hepatocellular adenoma, 102, 104 hepatocellular carcinoma, 104 metastatic cancer to liver, 104 liver metastases, 110–111 neuroendocrine tumors, 155 pancreatic lesions acinar cell carcinoma, 107 cross-sectional anatomy, 100 metastatic cancer to pancreas, 107 pancreatic adenocarcinoma, 107 pancreatic neuroendocrine tumors, 106 solid pseudopapillary tumor, 106–107 Malignant biliary obstruction causes of, 343 covered vs. uncovered stents, 349 CT scanning, 343 direct cholangioscopy, 346 endoscopic therapy, 347 methods of therapy, 347 percutaneous drainage of jaundice, 348 endoscopic intervention, 344 endoscopic retrograde cholangio-pancreatography, 344–346 ERCP vs. PTC vs. surgery, 348 hilar strictures disease modifying treatment, 349 unilateral vs. bilateral, 349 MR scanning, 343–344 operative vs. nonoperative palliation of, 404 plastic vs. metal, 348–349 radiological diagnostic imaging, 343 radiotherapy brachytherapy, 349–350 photodynamic therapy, 350 ultrasound, 343 Malignant disease, 24 Malignant gastroduodenal obstruction, 404 Malignant insulinomas, 416 Marseille–Rome classification of chronic pancreatitis, 451 Mass-forming type of intrahepatic cholangiocarcinoma, 223 Mayo Clinic protocol, 230 MCN. See Mucinous cystic neoplasm MCT. See Microwave coagulation MdCT. See Multidetector computed tomography Mebendazol, 321 Medullary carcinoma, 436 Melanoma, 54–55 noncolorectal, nonneuroendocrine metastases, 169 Memorial Sloan-Kettering Cancer Center, 201 MEN-1 syndrome, 423 “Mercedes” incision, 25 Mesenchymal hamartoma, 268, 305 Mesh hepatorrhaphy technique, 275 Metastasis resections, 78–79 Metastatic cancer to liver, 104 to pancreas, 107 Metastatic colorectal cancer clinical risk scores, 120–121 patient evaluation patient selection, 118 preoperative imaging, 118–119 tumor resectability, 118 postoperative management adjuvant intra-arterial chemotherapy, 123 adjuvant systemic chemotherapy, 122 nonresectable metastatic disease, 124–126 outcomes of resection, 123–124 repeat liver resection, 129–130 resectability techniques, 126–129 prognostic factors, 119–120 resectable management preoperative management, 121 surgery approaches, 121–122 unresectable liver disease coverting to resection, 138–140 regional chemotherapy, 136–137 systemic chemobiologic therapy, 137–138 systemic chemotherapy, 135–136 systemic therapy, 140–141 Metastatic disease, pancreas, 434–435 Metronidazole, 246 MF-ICC. See Mass-forming type of intrahepatic cholangiocarcinoma MIBG. See Radiolabeled metaiodobenzylguanidine Microsatellite instability, 436 Microwave ablation, 183–184 Microwave coagulation, 183–184 Mild postpancreatectomy hemorrhage, 391 Mirrizi’s syndrome, 375 Misidentification injuries, 367 Mixed stones, 373 Monoclonal antibodies, 176 MRCP. See Magnetic resonance cholangiopancreaticography MRI. See Magnetic resonance imaging MSI. See Microsatellite instability MSKCC. See Memorial Sloan-Kettering Cancer Center Mucinous cystic neoplasm, 410–412 Mucocele, 374 Multidetector computed tomography cystic tumors, 411 pancreatic ductal adenocarcinoma, 383 Multimodal strategies, colorectal liver metastases chemotherapy and surgery, 149–150 computed tomography, 148–149 magnetic resonance imaging, 149 management strategies, 150–152 multidisciplinary team, 148 positron emission tomography, 149 preoperative staging, 148 resectability of, 149 resection margins, 149 strategies to improve respectability, 149 surgery, 149 tumor ablation, 150 Multiple single antibiotics, 257 Multivesicular cyst, 315 Multivisceral resections, 78 NAFLD. See Nonalcoholic fatty liver disease National Cancer Database, 198 National Cancer Institute of Canada, 137 National Pediatric Trauma Registry, 466 Natural orifice transabdominal endoscopic surgery, 91 NCIC. See National Cancer Institute of Canada Necrosis, acute pancreatitis, 441 Neoadjuvant chemoradiotherapy liver transplantation, hilar cholangiocarcinoma, 229–231 Neoadjuvant chemotherapy, 121 Neoadjuvant systemic chemotherapy, 139 Neonatal cholestasis, 479 NETs. See Neuroendocrine tumors Neuroendocrine metastases, 53 Neuroendocrine tumor hepatic metastasis clinical features, 154 cryoablation, 160 diagnostic imaging, 155 hepatic artery embolization, 156–158 hepatic resection, 156 laboratory investigation, 154 liver transplantation, 160–161 medical treatment, 161 radiofrequency ablation, 158, 160 radionuclide therapy, 161 WHO Classification of, 155 Neuroendocrine tumors, 41, 56, 154 Nodular regenerative hyperplasia, 239, 266 NOMLI. See Non-operative management of liver injury Nonalcoholic fatty liver disease, 173 Non-anatomic liver resection, 276 Nonanatomical hepatectomies, 5 Noncolorectal metastases, 54–55 Noncolorectal, nonneuroendocrine metastases breast cancer, 166 gastric cancer, 167–168 gynecological tumors, 166 head and neck tumors, 168 melanoma, 169 pancreatic cancer, 167 predictive factors determining clinical outcome, 169–170 renal cell cancer, 166–167 respiratory tract, 168

497

INDEX
Noncolorectal, nonneuroendocrine metastases (Continued) sarcoma, 168–169 Noncolorectal tumors, 56 Non-functional islet cell tumors, 421–422 clinical presentation of, 422–423 curative resection and survival, 426 curative surgery treatment for metastatic disease, 426–427 for the primary tumor, 425–426 extent of disease and localization, 423–424 location of, 422 prognosis, 428 surgical palliation and survival, 427 treatment chemotherapy, 428 octreotide, 428 palliative surgery, 427 radiation therapy, 428 Non-functioning islet cell tumors, 414 Noniatrogenic bile duct strictures benign inflammatory pseudotumors, 370 biliary strictures secondary to pancreatitis, 369–370 calculous disease, 370 sclerosing cholangitis, 370 Nonneuroendocrine metastases, 54–55 Nonneuroendocrine tumors, 56 Non-operative management of liver injury, 272–273 Nonparasitic simple hepatic cysts, 102 Nonresectable metastatic disease intra-arterial chemotherapy, 125–126 management of, 124 systemic chemotherapy, 124–125 NOTES. See Natural orifice transabdominal endoscopic surgery NRH. See Nodular regenerative hyperplasia Nutrition, 440 Obesity, 363 Obstructive chronic pancreatitis, 452 Obstructive jaundice, 334, 374, 401 Octreotide, 84 non-functional islet cell tumors, 428 Octreotide acetate, 420 Ocular melanoma, 55 OGTP. See osteoclast-like giant cell tumor of pancreas OIS-AAST. See Organ Injury Scaling Committee of the American Association for the Surgery of Trauma Okuda Classification, 192 OLT. See Orthotopic liver transplant Open cholecystectomy common bile duct stones, 377 gallstone disease, 375–376 Organ Injury Scaling Committee of the American Association for the Surgery of Trauma, 463–464 Oriental cholangiohepatitis, 370 Orthotopic liver transplantation, 192–195 for acute and chronic liver failure immunosuppression, 292–293 operative technique of, 292 outcomes in, 293–294 post-operative management, 292–293 Osteoclast-like giant cell tumor of pancreas, 435 Oxaliplatin, 135 Oxaliplatin-associated neurotoxicity, 140 Pain, 367 palliation of, 403–405 Painless jaundice, 381 PAIR. See Puncture aspiration injection reaspiration PAK. See Pancreas after kidney transplant Palliation indications for, 401–402 nonoperative techniques biliary decompression, 402 gastric decompression, 402–403 operative techniques biliary decompression, 402 gastric decompression, 403 vs. nonoperative techniques, 404–405 of pain, 403–405 Palliative gastrojejunostomy, 403 Palliative surgery, non-functional islet cell tumors, 427 Palliative therapy, laparoscopically discovered disease, 205 Pancreas after kidney transplant, 470 Pancreas divisum, congenital anomalies, 483 Pancreas transplant alone, 470 Pancreas transplantation effects on secondary complications of diabetes, 474–475 immunosuppression, 475 indications, 470–471 outcomes, 474 pancreas donor, 471 pancreas recipient, 471 recipient categories pancreas after kidney transplant, 470 pancreas transplant alone, 470 simultaneous kidney and pancreas transplant, 470 recipient operations bladder drainage, 473–474 enteric drainage, 474 portal venous drainage, 473 surgical complications, 474 systemic venous drainage, 472–473 surgical techniques backtable preparation, 472 procurement, 471–472 Pancreatectomy bile leak, 82 death, 83 delayed gastric emptying, 81 factors affecting complication rates extended lymphadenectomy, 84–85 institutional factors, 85 laparoscopic pancreatectomy, 85 octreotide, 84 pancreatic duct management following resection, 83–84 patient factors, 85 peritoneal drainage, 84 pylorus-preserving pancreaticoduodenectomy, 84 resection of contiguous structures, 84–85 total and central pancreatectomy, 85 pancreatic anastomotic leak, 81 pancreatic fistula, 81 postpancreatectomy hemorrhage, 81–82 Pancreatic abscess, congenital anomalies, 484 Pancreatic adenocarcinoma, 107 Pancreatic anastomotic leak, 81 Pancreatic cancer common symptoms, 401 diet as risk factor, 380 dietary factors, 381 noncolorectal, nonneuroendocrine metastases, 167 randomized controlled trials, 382–383 Pancreatic duct, integrity of, 465–466 Pancreatic duct management following resection, 83–84 Pancreatic ductal adenocarcinoma adjuvant treatment, 391–394 clinical presentation, 381 diagnosis and staging, 381–385 epidemiology of, 380 etiology of, 380–381 management of, 385 mortality and morbidity, 387–399 delayed gastric emptying, 389 intraabdominal abscesses, 390 postoperative pancreatic fistula, 389–390 postpancreatectomy hemorrhage, 390–391 neoadjuvant treatment, 387 postoperative treatment, 391 preoperative biliary drainage, 385 prognosis, 394 risk factors, 380–381 surgical resection extended lymphadenectomy, 387 extended resections, 387 pancreatic–enteric anastomosis, 386–387 standard vs. pylorus-preserving pancreatoduodenectomy, 385–386 symptoms, 381 Pancreatic endocrine tumors classification, 414 functional islet cell tumors gastrinomas, 417–419 glucagonomas, 419–420 insulinoma, 414–416

498

INDEX
somatostatinomas, 421 VIPomas, 420–421 non-functioning islet cell tumors clinical presentation of, 422–423 curative resection and survival, 426 curative surgery treatment, 425–427 extent of disease and localization, 423–424 location of, 422 prognosis, 428 surgical palliation and survival, 427 treatment, 427–428 Pancreatic exocrine secretions, 474 Pancreatic fistula, 81 Pancreatic fluid, 389 Pancreatic inflammation, 439 Pancreatic injury diagnosis clinical presentation, 463 intraoperative exposure and evaluation, 464–465 intraoperative pancreatography, 465–466 laboratory investigations, 463 radiologic investigations, 463–464 epidemiology of, 463 injury grading, 463 nonoperative management, 466 operative management distal ductal injuries, 466 high-grade injuries, 466 low-grade injuries, 466 proximal ductal injuries, 467 outcomes complications, 467 late stricture, 468 mortality, 467 pancreatic insufficiency, 468 pancreatic leaks and fistulae, 467 peripancreatic fluid collection, 467–468 pseudocysts, 467–468 Pancreatic ischemia, 453 Pancreatic leaks, 467 Pancreatic lesions, CT and MRI imaging acinar cell carcinoma, 107 cross-sectional anatomy, 100 metastatic cancer to pancreas, 107 pancreatic adenocarcinoma, 107 pancreatic neuroendocrine tumors, 106 solid pseudopapillary tumor, 106–107 Pancreatic lymphoma, 434 Pancreatic main duct stenosis, 452 Pancreatic neoplasms, 41–42. See also Non-functioning islet cell tumors Pancreatic neuroendocrine tumors, 435 pancreatic lesions, 106 Pancreatic polypeptideoma, 423 Pancreatic pseudocyst, 407 congenital anomalies, 484 Pancreatic pseudocysts, 91 Pancreatic resections distal pancreatectomy, 73–74 duodenum-preserving pancreatic head resection, 75–76 enucleation, 76–77 metastasis resections, 78–79 multivisceral resections, 78 recurrence resections, 78 segmental resections, 76 total pancreatectomy, 74–75 vessel resections, 77–78 Whipple resection, 73 Pancreatic trauma, 463 Pancreaticopleural fistulas, 453 Pancreatic–enteric anastomosis, 386–387 Pancreatitis, choledochal cyst, 356 Pancreatoduodenectomy vs. pylorus-preserving pancreatoduodenectomy, 336–337 Panitumumab, 137 Parasympathetic innervation, 22 Partial pancreaticoduodenectomy, 73 Partial surgical shunts, 283 Patient-related factors, bile duct injury acute cholecystitis, 360–362 congenital abnormalities, 362–363 Paucity, 478 PBD. See Preoperative biliary drainage PCLD. See Polycystic liver disease PDAC. See Pancreatic ductal adenocarcinoma PDT. See Photodynamic therapy Pediatric hepato-pancreato-biliary disorders biliary atresia, 478 choledochal cysts, 479–480 congenital anomalies, 483–484 gallbladder disease, 480–481 hepatoblastoma, 481 hepatocellular carcinoma, 481–483 PEI. See Percutaneous ethanol injection Peliosis hepatic, 267–268 Percutaneous catheter drainage, 443–444 Percutaneous chemical/thermal ablation, 218–219 Percutaneous cholangio-drainage, 385 Percutaneous cholecystostomy, 375 Percutaneous drainage of jaundice, 348 Percutaneous ethanol injection, 184 Percutaneous necrosectomy, 443, 446 Percutaneous transhepatic biliary drainage, 247–248 Percutaneous transhepatic cholangiography, 348, 367 Perfusion imaging, 112 Perihepatic packing, 275 Peripancreatic fluid collection, 467–468 Peritoneal drainage, 84 Peroral cholangioscopy, 333 Persistent hyperinsulinemic hypoglycemia of infancy, 484 PET. See Positron emission tomography PET-CT. See Positron emission tomography with CT Pharmacologic therapy, 282–283 PHHI. See Persistent hyperinsulinemic hypoglycemia of infancy Photodynamic therapy, 350 Photofrin®, 350 PIAF. See cisPlatin, Interferon, Adriamycin, and 5-Fluorouracil Pleomorphic giant cell carcinoma, 435 PNET. See Pancreatic neuroendocrine tumors POC. See Peroral cholangioscopy Polycystic liver disease clinical presentation, 303–304 Gigot classification of, 304 novel treatments, 305 surgical treatment, 304–305 POPF. See Postoperative pancreatic fistula Portal hypertension causes of, 281 clinical manifestations ascites, 280 hepatocellular carcinoma, 281 hypersplenism, 280 pulmonary syndromes, 280–281 variceal bleeding, 280 etiology of, 280 evaluation, 281 and general surgeon, 285–286 pathophysiology of, 280 Portal vein embolization, 67, 334 tumor respectability, 126 Portal venous drainage, 473 Portopulmonary syndromes, 285 Positron emission tomography colorectal liver metastases multimodal approaches, 149 with computer tomography, 383 gallbladder cancer, 329 liver metastases, 111–112 neuroendocrine tumors, 155 Posterior inferior pancreaticoduodenal artery, 19 Posterior superior pancreaticoduodenal artery, 18–19 Posthepatectomy infections, 68 Postoperative bile leak consequences of, 66 management of, 66 presentation, 66 Postoperative pancreatic fistula definition, 390 grade A fistulas, 389 grade B fistulas, 389 grade C fistulas, 389–390 pancreatic fluid, 389 Postoperative treatment, pancreatic ductal adenocarcinoma, 391 Postpancreatectomy hemorrhage, 81–82 intraluminal and extraluminal, 390 mild, 391 severe, 391 PPH. See Portopulmonary syndromes; Postpancreatectomy hemorrhage PPL. See Primary pancreatic lymphoma PPoma. See Pancreatic polypeptideoma

499

INDEX
Ppower Doppler, 36 ppPD. See Pylorus-preserving pancreatoduodenectomy Preoperative biliary drainage, 385 Primary pancreatic lymphoma, 434 Primary prophylaxis, 281–282 Primary resection therapy, 194 Primary sclerosing cholangitis cholangiocarcinoma, 324–325 diagnosis of, 324 endoscopic treatment, 325 medical treatment, 325 natural history of, 324 surgical management, 325–327 Procedure-related factors, bile duct injury critical view technique, 365 misidentification concepts, 363–365 technical problems, 365 Procurement surgical technique, 471–472 Prophylactic octreotide, 84 Proton pump inhibitors, 417 Proximal ductal pancreatic injuries, 467 PSC. See Primary sclerosing cholangitis Pseudocysts, 467–468 Pseudolipoma hereditary hemorrhagic telangiectasia, 268 heterotopic tissue, 268 inflammatory pseudotumor, 268 miscellaneous rare benign solid liver lesions, 268 peliosis hepatic, 267–268 PTA. See Pancreas transplant alone PTC. See Percutaneous trans-hepatic cholangiography PTCD. See Percutaneous cholangio-drainage Pulmonary syndromes, 280–281 Puncture aspiration injection reaspiration, 320–321 PVE. See Portal vein embolization Pylorus-preserving pancreaticoduodenectomy, 84, 385–386 Pyogenic liver abscess diagnosis of, 256–257 epidemiology of, 255 microbiology of, 256 outcomes of, 258 pathogenesis of, 255–256 treatment of, 257–258 Pyrexia, 440 Radiation therapy, 428 Radiochemotherapy, 391 Radio-embolization, 218 Radiofrequency ablation, 126 thermal ablation, 181–183 Radioisotope imaging, hydatid cyst, 313 Radiolabeled metaiodobenzylguanidine, 155 Radionuclide therapy, neuroendocrine tumors, 161 Randomized controlled trail operative vs. nonoperative palliation malignant biliary obstruction, 404 malignant gastroduodenal obstruction, 404 pancreatic ductal adenocarcinoma CONKO-1, 391–392 ESPAC-1, 391–393 Rapamycin, 235 Rare benign cystic lesions, 305 Rare benign solid liver lesions, 268 Rare tumors acinar cell carcinoma, 432 adenosquamous carcinoma, 432–434 autoimmune pancreatitis, 436–437 giant cell tumors, 435 medullary carcinoma, 436 metastatic disease, pancreas, 434–435 pancreatic lymphoma, 434 renal cell carcinoma, 434 solid pseudopapillary neoplasm, 432 Von Hippel–Lindau syndrome, 435 Rare vascular liver tumors hepatic epithelioid hemangioendothelioma, 233–235 hepatic hemangiosarcoma, 237–238 hepatic infantile hemangioendothelioma, 235–237 nodular regenerative hyperplasia, 239 RCC. See Renal cell carcinoma Recurrence resections, 78 Recurrent metastatic disease, 129–130 Recurrent pyogenic cholangitis acute management, 246–247 clinical presentation, 244–245 complications, 250 definitive management, 247 definitive surgery, 248–250 hematological and biochemical investigations, 245–246 minimal access approach, 247–248 pathogenesis of, 242–243 pathology of, 243–244 Recurrent variceal bleeding, 282 Refractory bleeding, 275 Regenerative nodules, 266 Regional chemotherapy, 136–137 Renal cell cancer, 166–167, 434 Repeat liver resection, 129–130 Resectable liver disease, systemic therapy, 140–141 Resectable tumors, 385 Resection of contiguous structures, 84–85 Respiratory tract, 168 Retroperitoneal laparostomy, 447 RFA. See Radiofrequency ablation Right hemihepatectomy, 30–31 Right hepatectomy, 338 Right trisectionectomy hepatic resection, 31 hilar cholangiocarcinoma, 338–339 Routine arteriography, 416 RPC. See Recurrent pyogenic cholangitis Sarcoma, 54 noncolorectal, nonneuroendocrine metastases, 168–169 SCA. See Serous cystadenoma Sclerosing cholangitis, 370 Segmental resections, 57, 76 Selective surgical shunts, 283 Selective vascular ligation, 275–276 Self-expanding metal stents, 349 SEMS. See Self-expanding metal stents Sepsis, 367 Serous cystadenoma, 407–409, 412 Serum amylase, 463 Severe postpancreatectomy hemorrhage, 391 sFLR. See Standardized future liver remnant Short-term biliary stenting, 377 Simple biliary hepatic cysts, 301 Simple cystenterostomy, 357 Simultaneous kidney and pancreas transplant, 470 Sinusoidal obstruction syndrome, 174–176 SLT. See Split-liver transplantation SMA. See Superior mesenteric artery Small duct disease, 458–459 Small solitary hepatic metastases. See Hepatic metastasectomy SMV. See superior mesenteric vein Solid liver lesions, 37–39 Solid pseudopapillary neoplasm, 432 Solid pseudopapillary tumor, 106–107 Somatostatin analogs, 391 Somatostatin receptor scintigraphy, 424, 435 neuroendocrine tumors, 155 Somatostatinomas, 421 Sorafenib, 220 Special dedicated high-frequency transducers, 42 Spectral Doppler, 36 Spironolactone, 285 SPK. See Simultaneous kidney and pancreas transplant Split-liver transplantation, 290 SPPN. See Solid pseudopapillary neoplasm SPT. See Solid pseudopapillary tumor Spyglass™, 347 SRS, Somatostatin receptor scintigraphy SSTR-targeted therapy, 161 Staging laparoscopy, 91–92 Standard lymphadenectomy, 387 Standard vs. pylorus-preserving pancreatoduodenectomy, 385–386 Standardized future liver remnant, 177 Stapling devices, 94 Steatohepatitis, 174 Steatorrhea, 421 Steatosis, 173–174 Sterile cyst, 315 Streptozotocin, 420 Streptozotocin plus doxorubicin, 428 Streptozotocin plus fluorouracil, 428 Sunitinib, 220 Superior mesenteric artery, 381 Superior mesenteric vein, 381 Superior pancreaticoduodenal artery, 18 Surgeon/hospital-related factors laparoscopic equipment, 366

500

INDEX
Surgeon/hospital-related factors (Continued) learning curve effect, 366 psychology of human error, 366 Surgical resection, pancreatic ductal adenocarcinoma, 385–387 extended lymphadenectomy, 387 extended resections, 387 pancreatic–enteric anastomosis, 386–387 standard vs. pylorus-preserving pancreatoduodenectomy, 385–386 Surgical shunts, 283–284 Sympathetic innervation, 22 Symptom palliation, 401 Symptomatic hepatocellular carcinoma, 192 Synchronous liver metastases neoadjuvant chemotherapy, 130–131 surgery, 131 Systemic chemobiologic therapy, 137–138 Systemic chemotherapy nonresectable metastatic disease, 124–125 unresectable liver disease, 135–136 Systemic therapies advanced cirrhosis treatment, 220 anti-angiogenic therapies, 219–220 cisPlatin, interferon, adriamycin, and 5-fluorouracil, 219 future developments, 220–221 historical background, 219 unresectable liver disease, 140–141 Systemic venous drainage, 472–473 TACE. See Transarterial chemo-embolization; Trans-catheter arterial chemo-embolization TachoSil®, 65 Telangiectatic focal nodular hyperplasia, 263 Tenting injuries, 366–367 TFNH. See Telangiectatic focal nodular hyperplasia Therapeutic packing, 275 Thermal ablation CLOCC study, 184 cryotherapy, 180–181 edge cryotherapy, 181 limitations of, 180 microwave ablation, 183–184 percutaneous ethanol injection, 184 radiofrequency ablation, 181–183 stages of hepatocellular carcinoma, 195 Thermal injuries, 366–367 TIPS. See Transjugular intrahepatic portosystemic shunt “Top-Down” cholecystectomy, 365 Total cyst excision, 356 Total pancreatectomy, 74–75, 85 Total surgical shunts, 283 Total vascular exclusion, 128–129 TP. See Total pancreatectomy Trans-abdominal ultrasonography, 383 choledochal cyst, 355 Trans-abdominal ultrasound, 441 malignant biliary obstruction, 343 Transarterial chemo-embolization, 195 Trans-catheter arterial chemo-embolization, 216–217 Transcystic flushing, 89 Transcystic stone extraction, 89–90 Transjugular intrahepatic portosystemic shunt, 283 Tumor ablation multimodal approaches, 150 techniques, 126 Tumor respectability definition of, 118 portal vein embolization, 126 total vascular exclusion and cooling, 128–129 tumor ablation techniques, 126 two-stage hepatectomy, 127–128 TVE. See Total vascular exclusion Two-stage hepatectomy, 127–128 Two-stage liver resection, 127 Type B juxtahepatic injuries, 276 UCDA. See Ursodeoxycholic acid UCSF. See University of California at San Francisco UICC. See International Union against Cancer Ultrasonography hydatid cyst, 311 transabdominal, 382–383 Ultrasound gallbladder and bile ducts, 39–41 intraoperative, 42–43 liver diffuse liver disease, 36 focal hepatic lesions, 36–39 liver metastases, 109 malignant biliary obstruction, 343 pancreas diffuse pancreatic diseases, 41 pancreatic neoplasms, 41–42 radiological anatomy of liver, 13 University of California at San Francisco, 209–210 Univesicular cyst, 315 Unresectable liver disease coverting to resection, 138–140 regional chemotherapy, 136–137 systemic chemobiologic therapy, 137–138 systemic chemotherapy, 135–136 systemic therapy, 140–141 Unusual functional islet cell tumors, 421 Urine, 154 Ursodeoxycholic acid, 377 Vagotomy, 403 Variceal bleeding acute variceal hemorrhage, 282 clinical manifestations, 280 devascularization procedures, 284 endoscopic therapy, 282–283 liver transplantation, 284 pharmacologic therapy, 282–283 primary prophylaxis, 281–282 recurrent, 282 surgical shunts, 283–284 transjugular intrahepatic portosystemic shunt, 283 Variceal decompression, 283 Vascular endothelial growth factor, 137 Vascular injuries, 360 Vasoactive intestinal peptide, 420 VEGF. See Vascular endothelial growth factor Venous bleeding, 95 Venous drainage of pancreas, 20–21 Vessel resections, 77–78 VIPomas, 420–421 Von Hippel–Lindau (VHL) syndrome, 435 Watery diarrhea, hypokalemia, and achlorhydria, 420 WDHA. See Watery diarrhea, hypokalemia, and achlorhydria Wedge resections, 56–57 Whipple resection, 73 Y-graft, pancreas transplants, 473 ZES. See Zollinger–Ellison syndrome Zollinger–Ellison syndrome, 416–417

501

Surgical Management of Hepatobiliary and Pancreatic Disorders
Second Edition
Edited by Graeme J. Poston, Michael D’Angelica, and René Adam About the book
Hepato-Pancreato-Biliary (HPB) surgery is now firmly established within the repertoire of modern general surgery. This new edition has been completely rewritten by world-leading surgeons to reflect the considerable advances made in the surgical management of HPB disorders since the highly successful first edition. This new edition includes: • An in-depth coverage of benign and malignant disorders of the liver, pancreas, and bile ducts and gallbladder • A comprehensive section on anatomy, imaging, and surgical technique • Over 20 new chapters, including a complete account of pediatric HPB disorders • Almost 300 high-resolution images, many in full color Surgical Management of Hepatobiliary and Pancreatic Disorders, Second Edition, comprehensively covers the full spectrum of common HPB diseases and associated surgical techniques to assist not only the general surgeon in regular practice, but also surgical trainees and those in related specialties of oncology, radiology, gastroenterology, and anesthesia.

With a Foreword by Yuji Nimura, MD, President of the Aichi Cancer Center, Japan, and Past President of the IHPBA

This book demonstrates the wisdom of the new knowledge and technical skills of these diverse disciplines where cooperative efforts contribute toward the benefit of the patients with HPB disorders.
Also Available
Hepatocellular Carcinoma: A Practical Approach
Edited by Bandar Al Knawy, K. Rajendra Reddy and Luigi Bolondi ISBN: 9780415480802 e-ISBN: 9780203092880

About the Editors
Graeme j. Poston, MS, FRCS (Eng), FRCS (Ed), is Director of Surgery and Hepatobiliary Surgeon, University Hospital Aintree, Liverpool, UK. He is the President of the Association of Upper Gastrointestinal Surgeons of Great Britain and Ireland (AUGIS), PresidentElect of the European Society of Surgical Oncology (ESSO), Past President of the British Association of Surgical Oncology (BASO), and author of numerous publications and national/international guidelines relating to the practice of HPB surgery. Michael D’Angelica, MD, is an Associate Attending at Memorial Sloan-Kettering Cancer Center and an Associate Professor at Cornell University/Weill Medical Center. He is currently the Program Chairman of the American Hepato-Pancreato-Biliary Association and a writing member of the National Comprehensive Cancer Network (NCCN) practice guidelines for hepatobiliary malignancy. René Adam, MD, PHD, is Hepatobiliary Surgeon and Professor of Surgery, Hôpital Paul Brousse, Université Paris-Sud, Villejuif, France.

Improved Outcomes in Colon and Rectal Surgery
Edited by Charles B. Whitlow, David E. Beck, David A. Margolin, Terry C. Hicks and Alan E. Timmcke ISBN: 9781420071528 e-ISBN: 9781420071535

Textbook of Surgical Oncology
Edited by Graeme J. Poston, R. Daniel Beauchamp, and Theo J. M. Rogers ISBN: 9781841845074 e-ISBN: 9780203003220

Telephone House, 69-77 Paul Street, London EC2A 4LQ, UK 52 Vanderbilt Avenue, New York, NY 10017, USA

www.informahealthcare.com

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close