Colostomy

Published on December 2017 | Categories: Documents | Downloads: 48 | Comments: 0 | Views: 1080
of x
Download PDF   Embed   Report

Comments

Content

Int J Colorectal Dis (2009) 24:479–488
DOI 10.1007/s00384-009-0662-x

REVIEW

Loop ileostomy versus loop colostomy for fecal diversion
after colorectal or coloanal anastomosis: a meta-analysis
F. Rondelli & P. Reboldi & A. Rulli & F. Barberini &
A. Guerrisi & L. Izzo & A. Bolognese & P. Covarelli &
C. Boselli & C. Becattini & G. Noya

Accepted: 20 January 2009 / Published online: 12 February 2009
# Springer-Verlag 2009

Abstract
Background Sphincter-saving surgery for the treatment of
middle and low rectal cancer has spread considerably when
total mesorectal excision became standard treatment. In
order to reduce leakage-related complications, surgeons
often perform a derivative stoma, a loop ileostomy (LI), or
a loop colostomy (LC), but to date, there is no evidence on
which is the better technique to adopt.

Financial support: none
F. Rondelli : A. Rulli : F. Barberini : P. Covarelli :
C. Boselli : G. Noya
Department of General and Oncologic Surgery,
University of Perugia,
P.zza Università 1,
Perugia, Italy
P. Reboldi : C. Becattini
Department of Internal Medicine, University of Perugia,
P.zza Università 1,
Perugia, Italy
A. Guerrisi
Department of Radiological Sciences,
University of Rome “Sapienza”,
Viale Regina Elena 324,
Rome, Italy
L. Izzo : A. Bolognese
General Surgery Unit, University of Rome “Sapienza”,
Rome, Italy
L. Izzo (*)
Medical School, University of Rome “Sapienza”,
Via G. Tomasi di Lampedusa 9,
00144 Rome, Italy
e-mail: [email protected]

Methods We performed a systematic review and metaanalysis of all randomized controlled trials until 2007 and
observational studies comparing temporary LI and LC for
temporary decompression of colorectal and/or coloanal
anastomoses.
Clinically relevant events were grouped into four study
outcomes:





general outcome measures: dehydratation and wound
infection GOM
construction of the stoma outcome measures: parastomal hernia, stenosis, sepsis, prolapse, retraction,
necrosis, and hemorrhage
closure of the stoma outcome measures: anastomotic leak
or fistula, wound infection COM, occlusion and hernia
functioning of the stoma outcome measures: occlusion
and skin irritation.

Results Twelve comparative studies were included in this
analysis, five randomized controlled trials and seven
observational studies. Overall, the included studies reported
on 1,529 patients, 894 (58.5%) undergoing defunctioning
LI. LI reduced the risk of construction of the stoma
outcome measure (odds ratio, OR=0.47). Specifically,
patients undergoing LI had a lower risk of prolapse (OR=
0.21) and sepsis (OR=0.54). LI was associated with an
excess risk of occlusion after stoma closure (OR=2.13) and
dehydratation (OR=4.61). No other significant difference
was found for outcomes.
Conclusion Our overview shows that LI is associated with a
lower risk of construction of the stoma outcome measures.
Keywords Loop ileostomy . Loop colostomy .
Derivative enterostomy

480

Int J Colorectal Dis (2009) 24:479–488

Introduction

Materials and methods

The surgical approach to rectal cancer has changed after the
wide acceptance of total mesorectal excision (TME) and
the reduction of the distal resection margin to 1 cm from the
tumor edge. As a consequence, the number of sphinctersaving procedures has increased with a reduction of
abdominoperineal excisions [1].
The low and ultra-low colorectal or coloanal anastomoses
(less than 3 cm from the sphincter) which are currently
performed have a relevant risk of leakage. Devascularization
of the rectal stump, the wide space created by the TME, and
the use of neo-adjuvant treatment are currently accepted as
being responsible for this high rate of leakage [2–4].
Derivative enterostomies are often performed to temporarily to reduce the effects of anastomotic leakage and also
the rate of leakage-related reinterventions [2, 5].
In an observational study in 258 patients, the rate of
anastomotic leakage was 4.9% and 17% in patients
undergoing derivative enterostomies and in patients treated
without stoma, respectively [6]. Similar differences have
been reported by other authors (8.2% vs. 16%) who have
reported a lower rate of surgical reintervention when stoma
was performed [3].
Furthermore, a recent randomized controlled trial of 234
patients evidenced that a defunctioning stoma dramatically
decreased the rate of symptomatic anastomotic leakage
(10.3% vs. 28%) [7].
Presently, no consensus exists on the best surgical procedure for derivative enterostomy, despite many case
controls and cohort studies that have evaluated the clinical
course of patients undergoing LI or LC after colorectal
anastomoses.
Recently, two published meta-analyses [8, 9], both
including five randomized controlled trials (RCTs) comparing the efficacy and safety of LI and LC, failed to
demonstrate any significant advantage of one technique
over the other. In fact, they were only able to report a
smaller non-significant rate of prolapse associated with LI.
Both papers conclude that larger RCTs are needed in order
to establish the best surgical procedure for performing a
transitory enterostomy. An even more recent meta-analysis,
including three RCTs and four cohort studies, has suggested
that LI may be preferable to LC when used to defunction a
distal colorectal anastomosis [10]; wound infections following stoma reversal were reduced, as were overall stomarelated complications and incisional hernia following stoma
reversal for LI patients in high quality studies [10].
Given this reality, this systematic review and metaanalysis of all RCTs and observational studies was carried
out to compare temporary LI with LC.

Search strategies
A systematic review of the literature was performed using
the following databases: MEDLINE, the Cochrane database
of systematic reviews, and the Cochrane controlled trials
register. Also, a manual search was done on important
relevant literature. The following terms were used for the
search: ‘loop ileostomy, loop colostomy, colorectal anastomosis, coloanal anastomosis’. The following restrictions
were used: age >19 and study in humans.
Reference lists of available reviews and selected studies
were cross-searched for additional literature. The authors
also searched independently on the internet and in libraries
for published and unpublished abstracts. Moreover, experts
in the field of colorectal surgery were consulted.
Inclusion and exclusion criteria
All published and unpublished studies comparing LI and
LC were considered for analysis as well as cohort studies
including only patients undergoing a specific derivative
procedure; either LI or LC was excluded from the
analysis.
Study outcomes
The following outcomes were considered for the analysis:
wound infection (during open stoma), dehydratation,
necrosis, prolapse, retraction, parastomal hernia, stenosis,
sepsis, hemorrhage, occlusion, wound infection (post-stoma
closure), anastomotic leak or fistula, hernia (post-stoma
closure), skin irritation, and bowel occlusion.
To obtain results which could be compared from present
literature and to assess a higher number of events, the study
outcomes were grouped into four categories, as was done in
the Guenaga study [9].
A—GENERAL OUTCOME MEASURES: wound infection and dehydratation.
B—CONSTRUCTION OF THE STOMA OUTCOME
MEASURES: necrosis, prolapse, retraction, parastomal
hernia, stenosis, sepsis, and hemorrhage.
C—CLOSURE OF THE STOMA OUTCOME MEASURES: occlusion, wound infection, anastomotic leak or
fistula, and hernia.
D—FUNCTIONING OF THE STOMA OUTCOME
MEASURES: skin irritation and occlusion.
The rates of tardive bowel occlusion and late reintervention were not examined, since data pertaining to these
were not present in the studied literature.

Int J Colorectal Dis (2009) 24:479–488

481

Data collection
Two reviewers independently extracted the following from
each study: first author, year of publication, study population characteristics (urgent or elective surgery), study
design, inclusion and exclusion criteria, matching criteria,
number of patients operated on with each technique,
duration of follow-up, mean age, male-to-female ratio, and
stoma closure rates.
Only published data were considered for analysis.
Statistical analysis
The meta-analysis was performed according to the recommendations of the Cochrane Collaboration and the Quality
of Reporting of Meta-analyses guidelines [11, 12]. Statistical analysis of dichotomous outcomes was performed
using odds ratios (OR) as the summary statistic and was
reported with 95% confidence intervals (CI). Odds ratios
for the outcomes of interest were combined using Der
Simoman and Laird. Haldane correction was applied when
studies contained a zero in one cell for the number of events
of interest in either of the two groups.
Study quality was assessed using the Star Rating Scale
[13], modified according to Tilney criteria. Pre-planned
subgroup analyses included study design (RCT vs. nonRCT) and clinical setting (elective vs. urgent).
Heterogeneity was assessed by two methods: graphical
exploration with funnel plots to evaluate publication bias
[14, 15] and Cochran’s chi-square test with I-squared test
for heterogeneity to assess between-studies heterogeneity.
Statistically significant heterogeneity was considered to be
present when p<0.10 and I squared>50% [16].
Analysis was conducted using the statistical Review
Manager Version 4.2 (The Cochrane Collaboration, Software Update, Oxford) and stat aver 9.2 (Statacorp LP).

Results
Our search retrieved 93 papers (Fig. 1). Seventy-two papers
were excluded after reviewing the title as they reported on:
the complications of colorectal anastomosis, the clinical
outcome of patients treated with or without derivative
enterostomy after colorectal surgery, and different surgical
technique or outcomes. While the remaining 21 papers
were evaluated in their full text, of these, 16 were
excluded as they did not report on this meta-analysis’
prespecified outcomes. From an analysis of the selected
papers’ reference lists, an additional ten studies were
found to compare the clinical course of LI and LC after

Fig. 1 Papers retrieved during the search

colorectal surgery. Overall, 15 studies were deemed
eligible for this analysis, five RCTs, which had been
examined in two previous meta-analyses [8, 9], and ten
non-RCT reports. Three out of eight non-RCT papers were
found in literature but their full text evaluations were not
available [17–19].
Thus, in conclusion, 12 comparative studies were
included in the analysis [5, 20–30]. Five were randomized
controlled trials [20–22, 27, 28], three were prospective
non-randomized studies [5, 24, 29], and four were
retrospective reviews [23, 25, 26, 30].
Globally, the included studies reported on 1,529 patients:
894 LI (58.5%) and 635 LC (41.5%).
The patient and study characteristics are summarized in
Table 1.
The underlying diagnoses for colorectal surgery were
reported in 11 studies (1,487 patients) [5, 20–28, 30]:
colorectal cancer in 1,320 (89%), diverticular disease in 116
(7.8%), and other diagnoses in 51 patients (3.2%). Five
studies (994 patients) reported outcomes on defunctioning
stomas exclusively following resections for colorectal
cancer [5, 20, 22, 23, 25] including only patients
undergoing elective surgery. All the other studies included
both patients undergoing elective and urgent colorectal
surgery [21, 24, 26–30].
The outcomes of interest reported by each of the
included studies are summarized in Table 2.

482

Int J Colorectal Dis (2009) 24:479–488

Table 1 Study characteristics
Reference

Design Ileostomy Colostomy Diagnosis Matching

Indication Stoma not
for stoma reversed N
(%)

Age at surgery
(mean unless
stated)

Females
N (%)

Edwards
et al.
[20]
Gastinger
et al. [5]
Khoury et
al. [21]
Law et al.
[22]
Rullier et
al. [23]
Rutegard
and
Dahlgren
[24]
Tocchi et
al. [25]
Fasth et
al. [29]
Gooszen
et al.
[28]
Williams
et al.
[27]
Gohring
et al.
[26]
Sakai et
al. [30]

RCT

34

36

A

1, 2, 5, 7

n/c

I=2 (5.9)
C=5 (13.9)

I=63 (median)
C=68 (median)

21 (30.0) ********

PNR

407

229

A

1, 2, 4, 5

n/c

n/c

n/c

RCT

32

29

A, B

1, 2, 4, 5

42

38

A

Retro

107

60

A

PNR

32

29

A, B, C

1,
6,
1,
5,
1,
5,

I=65
C=65
I=65.2
C=67.8
I=63 (median)
C=64 (median)
I=67
C=72

25 (41.0) ********

RCT

a, b, c,
d, e
f

I=48 (10.5)
C=40 (14.9)
n/c

Retro

17

24

A

PNR

21

21

n/c

RCT

37

39

RCT

23

Retro

Retro

2,
7,
2,
7
2,
9

4, 5,
8
3, 4,
3, 4,

a, e
n/c

I=7 (16.7)
C=0 (0)
I=11 (10.3)
C=10 (16.7)
I=24 (75.0)
C=14 (48.3)

Study quality
(star rating)
(max 11)

****

31 (38.8) ********
46 (27.5) ******
27 (44.3) *****

1, 2, 3, 4,
5
1, 2, 8, 9

g

n/c

n/c

A, B, C

1, 2, 3, 4,
6

a, e, g

I=0 (0)
C=0 (0)
I=8 (21.6)
C=7 (17.9)

I=63.3
14 (34.1) *****
C=63.2
I=73 (median)
24 (57.1) ****
C=69 (median)
I=63.2 (median) 49 (64.5) ********
C=64.7 (median)

24

A, B, C

1, 2, 4, 6,
8

n/c

I=3 (13.0)
C=4 (16.7)

I=71 (median)
23 (48.9) ********
C=66.5 (median)

79

43

A, B, C

1, 2, 4, 6,
8

n/c

I=8 (10.2)
C=7 (16.3)

I=59 (median)
C=63 (median)

n/c

63

63

A, B, C

1, 2, 3, 4

a, b, e

I=15 (23.9)
C=23 (36.6)

I=64 (median)
C=64 (median)

64 (50.8) *****

*****

I ileostomy, C colostomy, Retro retrospective, PNR prospective non-randomized, RCT randomized controlled trial, n/c no comment; A colorectal
cancer, B diverticular disease, C other; 1 age, 2 gender, 3 body mass index, 4 diagnosis, 5 tumor level, 6 anastomosis level, 7 tumor stage,
8 anastomosis method, 9 mode of surgery; a poor bowel preparation, b colonic obstruction, c severe cardiovascular disease, d extensive local
malignancy, e technical anastomotic problem, f anastomosis below 5 cm, g “anastomosis at risk”

General outcome measures
Overall, eight studies reported on general outcome measures (42 total events): four wound infection (290 patients)
and five dehydratation (481 patients) (Fig. 2).
At the analysis of singular outcomes, no significant
difference was observed regarding the incidence of wound
infection between patients undergoing LI or LC (OR 0.82;
95% CI 0.39–1.74). None of the 221 patients undergoing
LC experienced dehydratation compared with 11/120 LI
patients (OR 4.61; 95% CI 1.15–18.53).
Construction of the stoma outcome measures
Nine studies reported on construction of the stoma outcome
measures (149 total events). Regarding the singular

outcome events, studies reported on: two necroses (237
patients), six prolapses (481 patients), five retractions (481
patients), eight parastomal hernias (673 patients), two
stenoses (237 patients), seven sepses (571 patients), and
three hemorrhages (354 patients) (Fig. 3).
An advantage was observed in terms of a lower
incidence of prolapse in LI patients (6/261 patients)
compared to LC patients (35/220 patients), resulting in an
OR 0.21 with 95% CI 0.09–0.51. Similarly, LI patients had
a lower incidence of sepsis (21/302) compared to LC (44/
269) patients (OR 0.54; 95% CI 0.30–0.99). A trend to a
lower incidence of stenosis was observed in LI patients (1/
139) compared to LC (four out of 98) (OR 0.24; 95% CI
0.04–1.55).
No difference was observed between patients undergoing
LI or LC regarding the incidence of necrosis (1/139 and 2/98;

Int J Colorectal Dis (2009) 24:479–488

483

Table 2 Outcomes of interest reported by each of the included studies
Reference

General
outcome measures

Construction
outcome measures

Closure
outcome measures

Edwards et al. [20]
Gastinger et al. [5]
Khoury et al. [21]
Law et al. [13]
Rullier et al. [23]
Rutegard and Dahlgren [24]
Tocchi et al. [25]
Fasth et al. [29]
Gooszen et al. [28]
Williams et al. [27]
Gohring et al. [26]
Sakai et al. [30]

a2

b2, b4, b6

c1, c2
c1, c3

a1
a2
a2
a1
a2
a1

b4,
b2,
b1,
b2,
b3,

b6,
b3,
b2,
b4,
b4,

b7
b4
b3, b4, b5, b6, b7
b6

b1, b2, b3, b4, b5, b6
b2, b6
a1, a2

c1, c2, c3
c1, c2, c3, c4
c1

b3, b4, b6, b7

c2,
c1,
c1,
c1,
c1,

c3
c2, c3
c2
c2, c3
c2

Functioning
outcome measures

d2
d1
d1, d2
d2
d1
d1, d2
d1
d1, d2
d1

A—GENERAL OUTCOME MEASURES: wound infection (a1) and dehydratation (a2); B—CONSTRUCTION OF THE STOMA OUTCOME
MEASURES: necrosis (b1), prolapse (b2), retraction (b3), parastomal hernia (b4), stenosis (b5), sepsis (b6), hemorrhage (b7); C—CLOSURE OF
THE STOMA OUTCOME MEASURES: occlusion (c1), wound infection (c2), anastomotic leak or fistula (c3), hernia (c4); D—FUNCTIONING
OF THE STOMA OUTCOME MEASURES: skin irritation (d1) and occlusion-ileus (d2)

OR 0.48; 95% CI 0.06–3.99), retraction (4/258 and 7/223; OR
0.53; 95% CI 0.08–3.63 with I squared 45.7%), parastomal
hernia (7/356 and 11/317; OR 0.59; 95% CI 0.23–1.55), and
hemorrhage (4/202 and 2/152; OR 1.33; 95% CI 0.23–7.53).
Closure of the stoma outcome measures
Overall, ten studies reported on closure of the stoma
outcome measures (124 total events), nine occlusion
Fig. 2 General outcome
measures

(1,237 patients), eight wound infection (620 patients), six
anastomotic leaks or fistulae (1,065 patients), and one
hernia (146 patients) (Fig. 4).
A benefit was observed in terms of a lower incidence of
occlusion in LC patients (10/491 patients) compared to LI
(34/746) patients, resulting in an OR 2.13 with 95% CI
1.01–4.53, with no evidence of heterogeneity.
No difference was observed between patients undergoing
LI or LC regarding the incidence of wound infection (18/

484

Int J Colorectal Dis (2009) 24:479–488

Fig. 3 Construction of the
stoma outcome measures

352 and 28/268; OR 0.51; 95% CI 0.20–1.28), anastomotic
leak or fistula (18/659 and 4/406; OR 1.87; 95% CI
0.47–7.47).
The incidence of hernia was reported only in one trial
with results in favor of LI (4/96 and 8/50; OR 0.26; 95% CI
0.07–0.91).
Functioning of the stoma outcome measures
Overall, nine studies reported on functioning of the stoma
outcome measures (113 total events): seven skin irritations
(559 patients) and six occlusions (491 patients) (Fig. 5).
LI patients reported a lower incidence of skin irritation
(31/296) compared to LC (51/263) for an OR 0.64 with
95% CI 0.39–1.08, with no evidence of heterogeneity. No
difference was observed regarding the incidence of occlusion between LI and LC (OR 1.36; 95% CI 0.64–2.91).

Overall outcome measures
Overall, 25 and 17 general outcome events were observed
in LI and LC patients, respectively. The cumulative analysis
of the construction of the stoma outcome measures did not
evidence a significant difference between LI and LC (OR
1.21; 95% CI 0.63–2.35).
Overall, 44 and 105 events related to the construction of
the stoma were observed in LI and LC patients, respectively. The cumulative analysis of construction of the stoma
outcome measures demonstrated a significant difference
between LI and LC (OR 0.47; 95% CI 0.32–0.69).
Overall, 74 and 50 events related to the closure of the
stoma were observed in LI and LC patients, respectively.
The cumulative analysis of closure of the stoma outcome
measures did not reveal a significant difference between LI
and LC (OR 1.02; 95% CI 0.56–1.86).

Int J Colorectal Dis (2009) 24:479–488

485

Fig. 4 Closure of the stoma
outcome measures

Overall, 51 and 62 events related to the functioning of
the stoma were observed in LI and LC patients, respectively. Finally, the cumulative analysis of functioning of the
stoma outcome measures did not show a significant difference between LI and LC (OR 0.81; 95% CI 0.53–1.25).
Subgroup analyses
Among the studies included in the analysis, five included
only patients undergoing elective surgery, while seven
included patients undergoing elective or urgent fecal
diversion for colorectal or coloanal anastomosis.
Setting Subgroup analysis of construction of the stoma
outcome measures showed a significant benefit from LI in
elective surgery patients (OR 0.31; 95% CI 0.17–0.57). A
trend in favor of LI was confirmed in emergency surgery
patients; however, it was not statistically significant (OR
0.60, 95% CI 0.33–1.09).
Design The analysis of construction of the stoma outcome
measures in RCTs demonstrated the superiority of LI

compared to LC (OR 0.33; 95% CI 0.19–0.60). Despite
the fact that a trend in favor of LI compared to LC was
confirmed in non-RCTs, this difference did not result as
being statistically significant (OR 0.62; 95% CI 0.37–1.04).
Pathology The analysis restricted to five trials exclusively
focusing on rectal cancer confirmed a significant advantage
for LI compared to LC in terms of a lower incidence of
prolapse (OR 0.26; 95% CI 0.09–0.79). A trend to a lower
incidence of sepsis in favor of LI compared to LC was still
reported (OR 0.41; 95% CI 0.12–1.47) but was not
statistically significant. A significant advantage was observed in terms of a lower incidence of occlusion after
closure of a LC compared to LI, resulting in an OR 3.03
with 95% CI 1.27–7.24.

Discussion
It has been demonstrated that the systematic use of a
temporary diverting stoma to decompress low and ultra-low

486

Int J Colorectal Dis (2009) 24:479–488

Fig. 5 Functioning outcome
measures

colorectal or coloanal anastomoses significantly reduces the
rate of anastomotic leakage [7] and that the development of
a leak is correlated with worse prognosis after a curative
resection for colorectal cancer [31–33]. Since a stoma can
be constructed involving either the ileum or the colon, the
surgeon often faces the decision on whether to perform LI
or LC. There is much literature on this specific subject, but
only a few papers have really compared ileal with colonic
diversion. For this reason, this meta-analysis was performed
on all published randomized controlled trials (RCTs) and
observational studies comparing loop ileostomy (LI) with
loop colostomy (LC) for temporary decompression of
colorectal and/or coloanal anastomosis. Whenever a paper
reported a LI versus LC comparison, we reviewed it
without exclusion criteria and of these only three papers
were excluded due to the fact that their full texts were not
available [17–19].
Heterogeneity was assessed by two methods: graphical
exploration with funnel plots to evaluate publication bias
and Cochran’s chi-square test and I-squared test for
heterogeneity to assess between-study heterogeneity. Statistically significant heterogeneity was considered to be
present when p<0.10 and I squared>50%.

This study design allowed to analyze a larger population
and a larger number of post-operative complications
compared to previously published meta-analyses on the
same topic, with no evidence of heterogeneity.
This study adopted the same end-point classification
used by Guenaga et al. [9], which reflects the time course of
clinical issues a surgeon may face when dealing with
temporary fecal diversion surgery. In addition, some events
were added that the past authors had not considered in their
analyses. This kind of classification is useful to evaluate the
efficacy and safety of different phases of this type of
surgery. Also, the uniformity in the end-point definitions
better facilitates both comparison among studies and
everyday clinical practice.
The first group of results analyzed the general outcome
measures, including wound infection and dehydratation.
Even if there was a significantly less dehydratation
reported in LC patients, the cumulative analysis of the
two parameters failed to show a statistically significant
difference.
In the second group analyzed, which included the stoma
construction outcome measures (parastomal hernia, stenosis, sepsis, prolapse, retraction, necrosis, and hemorrhage),

Int J Colorectal Dis (2009) 24:479–488

better results were observed in LI mainly due to sepsis and
prolapse, which showed a significantly better outcome even
when considered separately.
The third group of variables included the stoma closure
outcome measures: anastomotic leak or fistula, wound
infection, occlusion, and hernia. Neither of the two
procedures presented significant advantages when compared, even if a trend in favor of LC was registered
regarding the post-operative bowel occlusion.
The fourth examined group (functioning outcome measures) included ileus and skin irritation; neither the cumulative nor the single parameter analysis succeeded in
revealing a significant difference.
Findings here indicated an LI advantage over LC
regarding the reduction of prolapse and sepsis in the stoma
construction outcome and less dehydratation in LC in
general outcome measures. These were the only significant
variables that could be demonstrated.
The authors here believe that the clinical relevance of the
single complications is greatly different, since some
complications usually demand a reintervention, while
others are time-limited complications often managed with
non-invasive treatments. Specifically, regarding the abovementioned parameters, it is believed here that LI has a
minor impact on the quality of life, since prolapse and
sepsis are much worse complications than dehydratation.
Examining the subgroup of patients treated with elective
surgery, the advantages of LI construction appear to
increase significantly.
The reason of this result is not clear, but several relevant
differences between the elective and the emergency
procedures such as pathologies, bowel preparation, bowel
distention, mesenteric edema and thickness, and, last but
not least, different patient and surgeon statuses, should be
considered.
As for the trends, intended as better but non-significant
results, reduced incidences of occlusion after stoma closure
were observed in LC patients compared to LI. In fact, this
difference was mainly evident in the oncologic patient
subgroup. The reason for this could possibly be found in
the more extensive dissection and mobilization required in
oncologic procedures, which foster adhesion formation
[34]; the subsequent stoma closure in a patient with
abundant bowel adhesions could more likely lead to an
ileus after a LI shutdown because of the relatively greater
mobility of the ileal compared to that of the colonic tract.
The likelihood of serious complications is lower when
closing a LI compared to LC, as the latter is associated with
a higher risk of traction and vascular damage to the residual
vascular support of the suture [35].
The conclusion reached from this meta-analysis is that
the superiority of one treatment over another cannot be

487

definitively declared; however, the authors here endorse LI
over LC.
Ad hoc controlled clinical trials with definitive end
points are needed to once and for all clarify which is better:
LI or LC? To the purpose, the authors here estimate that a
randomized trial enrolling between 650 and 760 patients
per arm would be necessary to establish, with 90% power
(β) and a 1% type 1 error rate (α), the superiority of LI
over LC in terms of construction outcome measures.
Acknowledgments The authors thank Professor—blinded for the
review process—for his thoughtful advice on this study.

References
1. Heald RJ, Ryall RD (1986) Recurrence and survival after total
mesorectal excision for rectal cancer. Lancet 28:1479–1482
2. Marusch F, Koch A, Schmidt U et al (2002) Value of a protective
stoma in low anterior resections for rectal cancer. Dis Colon
Rectum 45:1164–1171
3. Peeters KC, Tollenaar RA, Marijnen CA et al (2005) Risk factors
for anastomotic failure after total mesorectal excision of rectal
cancer. Br J Surg 92:211–216
4. Giuliani D, Willemsen P, Van Elst F, Vanderveken M (2006) A
defunctioning stoma in the treatment of lower third rectal
carcinoma. Acta Chir Belg 106:40–43
5. Gastinger I, Marusch F, Steinert R et al (2005) Protective
defunctioning stoma in low anterior resection for rectal carcinoma.
Br J Surg 92:1137–1142
6. Dehni N, Schlegel RD, Cunningham C, Guiguet M, Tiret E, Parc
R (1998) Influence of a defunctioning stoma on leakage rates after
low colorectal anastomosis and colonic J pouch-anal anastomosis.
Br J Surg 85:1114–1117
7. Matthiessen P, Hallbook O, Rutegard J, Simert G, Sjodahl R
(2007) Defunctioning stoma reduces symptomatic anastomotic
leakage after low anterior resection of the rectum for cancer. A
randomized multicenter trial. Ann Surg 246:207–214
8. Lertsithichai P, Rattanapichart P (2004) Temporary ileostomy
versus temporary colostomy: a meta-analysis of complications.
Asian J Surg 27:202–210
9. Guenaga KF, Lustosa SA, Saad SS, Saconato H, Matos D (2007)
Ileostomy or colostomy for temporary decompression of colorectal
anastomosis. Cochrane Database Syst Rev 24:CD004647. Review
10. Tilney HS, Sains PS, Lovegrove RE, Reese GE, Heriot AG,
Tekkis PP (2007) Comparison of outcome following ileostomy
versus colostomy for defunctioning colorectal anastomoses. World
J Surg 3:1142–1151. Review
11. Norris SL, Atkins D (2005) Challenges in using nonrandomized
studies in systematic reviews of treatment interventions. Ann
Intern Med 142:1112–1119. Review
12. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD,
Rennie D et al (2000) Meta-analysis of observational studies in
epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA
283:2008–2012
13. Athanasiou T, Al-Ruzzeh S, Kumar P et al (2004) Off-pump
myocardial revascularization is associated with less incidence of
stroke in elderly patients. Ann Thorac Surg 77:745–753
14. Egger M, Davey Smith G, Schneider M et al (1997) Bias in metaanalysis detected by a simple, graphical test. BMJ 315:629–634

488
15. Egger M, Smith GD (1995) Misleading meta-analysis. BMJ
311:753–754
16. Higgins JP, Thompson SG (2002) Quantifying heterogeneity in a
meta-analysis. Stat Med 21:1539–1558
17. Popovic M, Petrovic M, Matic S, Milovanovic A (2001)
Protective colostomy or ileostomy. Acta Chir Iugosl 48:39–42
18. Lassalle FAB, Benati M, Quintana GO, Moscone CJ (1990) Loop
ileostomy as alternative to transverse colostomy to protect distal
anastomosis. Rev Argent Chirurg 58:160–164
19. Nordstrom G, Hulten L (1983) Loop ileostomy as an alternative to
transverse loop ileostomy. J Enterostomal Ther 10:92–94
20. Edwards DP, Leppington-Clarke A, Sexton R, Heald RJ, Moran
BJ (2001) Stoma-related complications are more frequent after
transverse colostomy than loop ileostomy: a prospective randomized clinical trial. Br J Surg 88:360–363
21. Khoury GA, Lewis MC, Meleagros L, Lewis AA (1987)
Colostomy or ileostomy after colorectal anastomosis?: a randomised trial. Ann R Coll Surg Engl 69:5–7
22. Law WL, Chu KW, Choi HK (2002) Randomized clinical trial
comparing loop ileostomy and loop transverse colostomy for faecal
diversion following total mesorectal excision. Br J Surg 89:704–708
23. Rullier E, Le Toux N, Laurent C, Garrelon JL, Parneix M, Saric J
(2001) Loop ileostomy versus loop colostomy for defunctioning
low anastomoses during rectal cancer surgery. World J Surg
25:274–277; discussion 277–278
24. Rutegard J, Dahlgren S (1987) Transverse colostomy or loop ileostomy
as diverting stoma in colorectal surgery. Acta Chir Scand 153:229–232
25. Tocchi A, Mazzoni G, Piccini M et al (2002) Use of ileostomy
and colostomy as temporal derivation in colorectal surgery. G Chir
23:48–52
26. Gohring U, Lehner B, Schlag P (1988) Ileostomy versus
colostomy as temporary deviation stoma in relation to stoma
closure. Chirurg 59:842–844

Int J Colorectal Dis (2009) 24:479–488
27. Williams NS, Nasmyth DG, Jones D, Smith AH (1986) Defunctioning stomas: a prospective controlled trial comparing
loop ileostomy with loop transverse colostomy. Br J Surg 73:566–
570
28. Gooszen AW, Geelkerken RH, Hermans J, Lagaay MB, Gooszen
HG (1998) Temporary decompression after colorectal surgery:
randomized comparison of loop ileostomy and loop colostomy. Br
J Surg 85:76–79
29. Fasth S, Hulten L, Palselius I (1980) Loop ileostomy—an
attractive alternative to a temporary transverse colostomy. Acta
Chir Scand 146:203–207
30. Sakai Y, Nelson H, Larson D, Maidl L, Young-Fadok T, Ilstrup D
(2001) Temporary transverse colostomy vs loop ileostomy in
diversion. Arch Surg 136:338–342
31. McArdle CS, McMillan DC, Hole DJ (2005) Impact of
anastomotic leakage on long-term survival of patients undergoing curative resection for colorectal cancer. Br J Surg 92:1150–
1154
32. Bell SW, Walker KJ, Richard MJ, Sinclair G, Dent OF, Chapuis
PH, Bokey EL (2003) Anastomotic leakage after curative anterior
resection results in a higher prevalence of local recurrence. Br J
Surg 90:1261–1266
33. Walker KG, Bell SW, Richard MJ, Mehanna D, Dent OF, Chapuis
PH, Bokey EL (2004) Anastomotic leakage is predictive of
diminished survival after potentially curative resection for
colorectal cancer. Ann Surg 240:255–259
34. Wong KS, Remzi FH, Gorgun E, Arrigain S, Church JM, Preen
M, Fazio WF (2005) Loop ileostomy closure after restorative
proctocolectomy: outcome in 1504 patients. Dis Colon Rectum
48:243–250
35. Barrier A, Martel P, Dugue L, Gallot D, Malafosse M (2001)
Direct and reservoir colonic-anal anastomoses. Short and long
term results. Ann Chir 126:18–25

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