Volume 74, No.10: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
693
Original Article
SMJ
Ravit Ruangtrakool, M.D., Cholapa Pintawekiat, M.D.
Division of Pediatric Surgery, Department of Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, ailand.
Types and Levels of Colostomy in Children with
Anorectal Malformation
ABSTRACT
Objective: Divided colostomy for anorectal management is oen recommended due to reports of higher complications
associated with loop colostomy. is study was conducted to compare outcomes and complications in colostomies
in children with anorectal malformations according to type and level of colostomy.
Materials and Methods: A retrospective study was performed in children with anorectal malformations who
underwent a colostomy at Siriraj Hospital between December 2003 and June 2018.
Results: Out of 167 patients, 159 had a loop colostomy while 8 had a divided colostomy. Overall complication
rates were 33.3% for loop colostomy and 62.5% for divided colostomy (p = 0.100). Urinary tract infection was the
most frequently encountered complication in both loop and divided colostomies, at 23.7% and 50%, respectively
(p = 0.094). e prolapse rate in the loop colostomy group was 8.8 % and 0% in the divided colostomy group (p =
0.376). Overall complication rates with respect to location of stoma also did not dier (p = 0.706). Prolapse rates
were 15.8 % in transverse colostomy and 7.1 % in sigmoid colostomy (p = 0.231). Overall complications rates of
colostomy closure in loop and divided colostomy was 7.5% and 12.5%, respectively (p = 0.672). Non-inferiority was
demonstrated by the dierences in overall complications of loop and divided colostomy (p = 0.008).
Conclusion: ere was no dierence in incidence of complications between type or location of colostomy performed
in children with anorectal malformations. Loop colostomy was non-inferior to divided colostomy in respect to
overall complications.
Keywords: Anorectal malformation; loop colostomy; divided colostomy; colostomy prolapse; urinary tract infection;
complication (Siriraj Med J 2022; 74: 693-698)
Corresponding author: Ravit Ruangtrakool
E-mail: sisuped@mahidol.ac.th
Received 14 February 2022 Revised 6 July 2022 Accepted 31 July 2022
ORCID ID:http://orcid.org/0000-0001-8162-2941
http://dx.doi.org/10.33192/Smj.2022.81
All material is licensed under terms of
the Creative Commons Attribution 4.0
International (CC-BY-NC-ND 4.0)
license unless otherwise stated.
INTRODUCTION
Among all congenital anomalies, the gastrointestinal
anomaly was the second most common system involved
(33.67%).
1
Anorectal malformations are a common
gastrointestinal anomaly encountered by pediatric
surgeons worldwide. There is a wide spectrum of
malformations, ranging from simple cutaneous stula
to cloacal malformations. Colostomy, with subsequent
denite repair is the standard treatment in those with
non-low type anorectal malformation. Loop colostomy
was the only preferred option in Division of Pediatric
Surgery at Siriraj Hospital for more than ve decades,
until divided colostomy was rstly introduced by PeñaA,
who developed posterior sagittal anorectoplasty, the most
popular denite operation for anorectal malformation in
1982.
2-5
Divided colostomy is generally preferred over
loop colostomy due to the higher rate of complications
associated with the latter, which includes prolapse, risk
of incomplete diversion of feces that causes subsequent
distension of distal rectal pouch, and possible contamination
Volume 74, No.10: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
694
of feces into the urinary tract in those with a connecting
stula between the rectum and genitourinary system.
2-6
However, controversy still surrounds the higher risk of
complications associated with loop colostomies compared
to divided colostomies with previously published studies
showing debatable results.
7-9
us, this study was conducted
to describe and compare outcomes and complications of
colostomies in children with anorectal malformations
according to type and level of colostomy. Outcomes
related to colostomy closure with respect to type of
colostomy were also compared.
MATERIALS AND METHODS
Following approval by the Siriraj Institutional
Review Board (Si 175/2019) a retrospective study was
conducted in children with anorectal malformations
who underwent a colostomy at Siriraj Hospital between
December 2003 to June 2018. Children with cloacal
exstrophy and major chromosomal anomalies incompatible
with life and those with incomplete medical information
were excluded from the study. Patients’ demographics,
type of malformation, location and type of colostomy
was collected. First, a colostomy was performed and this
was followed by denitive repair. Following achieving
an adequate neo-anus size as dilated by the parents,
colostomy closure was performed. Loop colostomy was
the preferred option in our division at Siriraj Hospital.
Complications during colostomy were recorded, including
prolapse, retraction, parastomal hernia, urinary tract
infection, bleeding, and skin excoriation. Upon colostomy
closure, operative time and complications were noted.
Complications during colostomy closure included wound
infection, wound dehiscence, and anastomosis leakage.
e collected data was analyzed using SPSS soware
version 18 (SPSS Inc. Released 2009. PASW Statistics for
Windows, Version 18.0. Chicago: SPSS Inc.). Continuous
data was expressed as median and IQR and categorical
data expressed as numbers and percentages. A Chi square
test or Fisher’s exact test was used to compare outcomes
in type and location of colostomy. Non-inferiority test
for dierence in overall complications between loop and
divided colostomy was conducted. Non-inferiority was
demonstrated when lower bound of the 95% one-sided
CI for dierence in overall complications was lower than
pre-specied non-inferior margin of 10%. A p-value of
<0.05 indicated statistical signicance.
RESULTS
Of the 178 patients whose medical records were
reviewed, 11 were excluded, which meant 167 patients
were included in the study. Out of the 167 patients
included, 159 underwent loop colostomies while eight
had a divided colostomy for fecal diversion. One hundred
and four out of 159 participants were male, of which 98
had a loop colostomy. Sixty-three patients were female,
of which 61 underwent a loop colostomy. For colostomy
level, there were four locations in total; ascending colon,
transverse colon, descending colon, and sigmoid colon.
e majority of patients underwent a (134 out of 167)
sigmoid loop colostomy. ere was a wide distribution
of malformation types ranging from imperforate anus
without stula to complex defects without signicant
dierences between the loop and divided colostomy
group (Table 1).
e dierences in complications found in the loop
and divided colostomy groups were not significant
(Table 2). Overall complication rates were 33.3% in
the loop colostomy group and 62.5% in the divided
colostomy group (p = 0.100). Urinary tract infections
were the most frequently observed complications in both
the loop (23.3%) and divided (50%) colostomy group.
e prolapse rate was 8.8% in the loop colostomy group
and 0% in the divided colostomy group, while skin
excoriation was 6.3% in the loop colostomy group and
12.5% in the divided colostomy group. When comparing
complications according to colostomy location, there
were no dierence in overall or individual complications
(Table 3). Interestingly, no statistically signicant dierence
was noted in overall complications or prolapse rates
between the transverse and sigmoid colostomy.
e median operative time for colostomy closures
was 160 minutes for loop colostomy and 195 minutes for
divided colostomy. e dierence was not statistically
signicant (p = 0.128). e incidence of complications
such as wound infection, wound dehiscence, and leakage
following closures in loop colostomy and divided colostomy
were not statistically signicant (Table 4).
When statistics for non-inferiority were performed,
with pre-specied non inferior margin of 10% between loop
and divided colostomies, non-inferiority was demonstrated
as the dierence in overall complications (p = 0.008).
DISCUSSION
Colostomy with subsequent definite repair is
the standard treatment in people with non-low type
malformation. Divided colostomies have been proposed
over loop colostomies due to reports of an increase in
complications associated with loop colostomy.
2-6
In fact,
loop colostomy was condemned by Pena A,
2,4,5
a world
authority in anorectal malformation management, due to
an increased prolapse rate, risk of incomplete diversion of
feces causing subsequent distension of distal rectal pouch,
Ruangtrakool et al.
Volume 74, No.10: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
695
Original Article
SMJ
TABLE 1. Comparison of the patient characteristics between the loop colostomy and the divided colostomy.
TABLE 2. Complications from colostomy, comparing between loop colostomy and divided colostomy.
Variable
Loop Colostomy Divided Colostomy Total
(n = 159) (n = 8) (n = 167)
P-value
Gender, n (%) 0.711
Male 98 (61.6%) 6 (75.0%) 104 (62.3%)
Female 61 (38.4%) 2 (25.0%) 63 (37.7%)
Level of Colostomy, n (%) 0.137
Ascending 3 (1.9%) 0 (0%) 3 (1.8%)
Transverse 18 (11.3%) 1 (12.5%) 19 (11.4%)
Descending 2 (1.3%) 1 (12.5%) 3 (1.8%)
Sigmoid 134 (84.3%) 6 (75.0%) 140 (83.8%)
Type of Malformation, n (%) 0.819
Imperforate Anus Without Fistula 30 (18.9%) 3 (37.5%) 33 (19.8%)
Perineal Fistula 10 (6.3%) 0 (0%) 10 (6.0%)
Vestibular Fistula 14 (8.8%) 0 (0%) 14 (8.4%)
Rectovaginal Fistula 6 (3.8%) 0 (0%) 6 (3.6%)
Rectobulbar Urethral Fistula 26 (16.4%) 0 (0%) 26 (15.6%)
Rectoprostatic Urethral Fistula 19 (11.9%) 1 (12.5%) 20 (12.0%)
Rectobladder Neck Fistula 11 (6.9%) 1 (12.5%) 12 (7.2%)
Rectovesicle Fistula 9 (5.7%) 0 (0%) 9 (5.4%)
Persistent Cloaca < 3 Cm 13 (8.6%) 1 (12.5%) 14 (8.4%)
Persistent Cloaca > 3 Cm 7 (4.4%) 0 (0%) 7 (4.2%)
Rectal Atresia 3 (1.9%) 0 (0%) 3 (1.8%)
Complex Defect 3 (1.9%) 0 (0%) 3 (1.8%)
Loop Colostomy Divided Colostomy Total
(n = 159) (n = 8) (n = 167)
P-value
Overall Complication, n (%) 53 (33.3%) 5 (62.5%) 58 (34.7%) 0.100
Prolapse 14 (8.8%) 0 (0%) 14 (8.4%) 0.376
Retraction 2 (1.3%) 0 (0%) 2 (1.2%) 0.747
Parastomal Hernia 1 (0.6%) 0 (0%) 1 (0.6%) 0.820
Urinary Tract Infection 37 (23.3%) 4 (50.0%) 41 (24.6%) 0.094
Bleeding 5 (3.1%) 0 (0%) 5 (3.0%) 0.607
Skin Excoriation 10 (6.3%) 1 (12.5%) 11 (6.6%) 0.502
Volume 74, No.10: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
696
TABLE 3. Complications from colostomy, comparing among dierent sites.
TABLE 4. Outcomes at colostomy closure, comparing between loop colostomy and divided colostomy.
Ascending Transverse Descending Sigmoid
Variables colostomy colostomy colostomy colostomy P-value
(n = 3) (n = 19) (n = 3) (n = 140)
Overall Complication, n (%) 2 (66.7%) 6 (31.6%) 1 (33.3%) 50 (35.7%) 0.706
Prolapse 1 (33.3%) 3 (15.8%) 0 (0%) 10 (7.1%) 0.231
Retraction 0 (0%) 0 (0%) 0 (0%) 2 (1.4%) 0.948
Parastomal Hernia 0 (0%) 0 (0%) 0 (0%) 1 (0.7%) 0.981
Urinary Tract Infection 2 (66.7%) 5 (26.3%) 1 (33.3%) 34 (24.3%) 0.409
Bleeding 0 (0%) 0 (0%) 0 (0%) 5 (3.6%) 0.820
Skin Excoriation 1 (33.3%) 0 (0%) 0 (0%) 10 (7.1%) 0.168
Variable
Loop Colostomy Divided Colostomy Total
(n = 159) (n = 8) (n = 167)
P-value
Operative time (min) 0.128
Median (min. max) 160 (35, 457) 195 (120, 215) 160 (35, 157)
Complications, n (%) 12 (7.5%) 1 (12.5%) 13 (7.7%) 0.672
Wound Infection 8 (5.0%) 1 (12.5%) 9 (5.39%) 0.410
Wound Dehiscence 1 (0.6%) 0 (0%) 1 (0.6%) 0.814
Leakage 1 (0.6%) 0 (0%) 1 (0.6%) 0.814
Gut Obstruction 3 (1.9%) 0 (0%) 3 (1.8%) 0.682
Incisional Hernia 1 (0.6%) 0 (0%) 1 (0.6%) 0.814
TABLE 5. Overall complications in loop and divided colostomy.
Loop colostomy Divided colostomy Difference Non-inferiority test
(n=159) (n=8) (95% one-sided CI) (P-value)
Overall complications 53 (33.3%) 5 (62.5%) -28.5% (-1.7, ∞)* 0.008**
*Non-inferiority was demonstrated (lower bound of the 95% one-sided CI for dierence in overall complications between loop and divided
colostomy was lower than pre-specied non-inferiority margin of 10%)
**Non-inferiority was demonstrated and p-value of non-inferiority test was less than signicant level of 0.05
Ruangtrakool et al.
Volume 74, No.10: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
697
Original Article
SMJ
and possible contamination of feces into the urinary tract
in patients with a connecting stula between the rectum
and genitourinary system.
2-5
However, in our study,
there was no statistically signicant dierence between
the loop and divided colostomy group regarding overall
complications. e complication rate associated with loop
colostomies in this study was consistent with previously
published studies at about 20%-30%.
3,7,9
However, this
study elicited a higher rate of complications in divided
colostomies compared to other studies (62.5% vs. 8%-
30%).
3,7,9
e prolapse rate of loop colostomies was quite low
in our study (8.8%) compared to other published studies
which reported rates of up to 18%.
3,9
A low prolapse
rate in loop colostomies in our institution might be the
result of the stoma creation technique used at our center
where loop colostomy was performed at the descending-
sigmoid colonic junction and the proximal and distal
limb of colostomy site were sutured together prior to
exteriorization and xation at sheath and skin. e suturing
of the proximal and distal limb may have decreased the
mobility of colon.
Divided colostomies were preferred over loop
colostomies due to risk of fecal contamination into the
distal rectourinary stula in the latter.
2,4
Although urinary
tract infection was the most common complication,
there was no signicant dierence between the loop
and divided colostomy group in this study. is nding
was consistent with previously published studies.
3,7
A
loop colostomy conducted in the proper way was able
to complete fecal diversion and was not dierent from
divided colostomy.
Regarding location of colostomy, no statistically
signicant dierence in complication rates was elicited
in this study. However, our study had higher rates of
overall complications for both transverse and sigmoid
colostomies compared to results published by van den
Hondel et al
9
and Demirogullari et al.
10
is might be the
result of including urinary tract infection as a complication
in our study while other studies did not include it. Previous
literatures have revealed that transverse colostomies have
a higher prolapse rate than other colostomy locations.
9-11
Regarding transverse colostomies, our study had a lower
prolapse rate than others. As mentioned previously,
this might be due to our surgical technique of placing
sutures between the proximal and distal limb of colon
prior to exteriorization of stoma at sheath and skin. Also,
we had more experience performing a loop colostomy
regardless of location when compared to other studies.
Since there were no significant differences in
complications in loop and divided colostomies, we
attempted to determine whether loop colostomy was
non-inferior compared to divided colostomy in respect
to complication rates. e non-inferior margin was
pre-determined to be 10%. Interestingly, we found that
non-inferiority, which was shown as a p-value in the
non-inferiority test, to be less than the signicant level
of 0.05. is had not been shown in previously published
studies.
Since there was no dierence in complications
between loop and divided colostomies during the stoma
creation period, outcomes during and aer colostomy
closure were investigated to demonstrate the advantage
of one stoma over the other. e operative time for
colostomy closure seemed shorter for loop colostomy at 160
minutes compared to 195 minutes for divided colostomy.
However, there was no signicant dierence to suggest
easier closure in loop colostomy. Complications such as
wound infection, wound dehiscence and anastomotic
leakage were also not signicantly dierent.
e limitation of this study was its retrospective
design which means some information might be missing.
Moreover, the number of subjects was relatively small at
167 patients. Also, there were a smaller number of divided
colostomy patients compared to the loop colostomy
group as it is our division’s preference to perform the
latter. is made comparison between the two groups
dicult in the study. However, our results were similar
to previously published studies in which loop colostomy
had good results compared to divided colostomy. A
multicenter study may be performed in the future to
increase the number of patients and data of divided
colostomy cases.
CONCLUSION
Loop colostomy is non-inferior to divided colostomy
in terms of overall complications and is a feasible diversion
procedure for anorectal malformation. Proper technique
and experience with loop colostomy helps achieve complete
diversion of feces with outcomes similar to that of divided
colostomy.
ACKNOWLEDGEMENTS
We would like to thank Dr. Sasima Tongsai from
the Division of Clinical Epidemiology, Department of
Research and Development, Faculty of Medicine Siriraj
Hospital, Mahidol University for her continuous help
with data processing and statistical analysis.
Conicts of interest: e authors have no conicts of
interest to declare.
Volume 74, No.10: 2022 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
698
REFERENCES
1. Sahoo S, Ganguly R, Dash M, Pradhan A, Priya TG, Mohanty.
Study of Congenital Malformation in a Tertiary Care Teaching
Hospital.Siriraj Med J. 2021;73(9):609-13.
2. Pena A, Migotto-Krieger M, Levitt MA. Colostomy in anorectal
malformations: a procedure with serious but preventable
complications. J Pediatr Surg. 2006;41(4):748-56.
3. Oda O, Davies D, Colapinto K, Gerstle JT. Loop versus divided
colostomy for the management of anorectal malformations.
J Pediatr Surg. 2014;49(1):87-90.
4. Levitt MA, PeñaA.Imperforate anus and cloacal malformations.
In:Holcomb III GW, Murphy JA, ed.Ashcra’s Pediatric
Surgery, 5
th
edition, Philadelphia: Saunders Elsevier; 2010.p.468-
90.
5. WilkinsS,PeñaA. e role of colostomy in the management
of anorectal malformations.Pediatr Surg Int.1988;3:105-9.
6. Gardikis S, Antypas S, Mamoulakis C, Demetriades D, Dolatzas T,
Tsalkidis A, et al. Colostomy type in anorectal malformations:
10-years experience. Minerva Pediatr 2004; 56(4):425-9.
7. Liechty ST, Barnhart DC, Huber JT, Zobell S, Rollins MD.
e morbidity of a divided stoma compared to a loop colostomy
in patients with anorectal malformation. J Pediatr Surg. 2016;
51(1):107-10
8. Patwardhan N, Kiely EM, Drake DP, Spitz L, Pierro A. Colostomy
for anorectal anomalies: high incidence of complications. J
Pediatr Surg. 2001;36(5):795-8.
9. van den Hondel D, Sloots C, Meeussen C, Wijnen R. To split or
not to split: colostomy complications for anorectal malformations
or hirschsprung disease: a single center experience and a
systematic review of the literature. Eur J Pediatr Surg. 2014;24(1):
61-9.
10. Demirogullari B, Yilmaz Y, Yildiz GE, Ozen IO, Karabulut R,
Turkyilmaz Z, et al. Ostomy complications in patients with
anorectal malformations. Pediatr Surg Int. 2011;27(10):1075-8.
11. Almosallam OI, Aseeri A, Shanafey SA. Outcome of loop versus
divided colostomy in the management of anorectal malformations.
Ann Saudi Med. 2016;4:352-5.
Ruangtrakool et al.
