Original Article

Multicenter Study of 342 Anorectal Malformation

Patients: Age, Gender, Krickenbeck Subtypes, and
Associated Anomalies
Tahmina Banu1,2 Anwarul Karim2,3 Maryam Ghavami Adel1,4 Kokila Lakhoo1,5
Tasmiah Tahera Aziz 2 Arni Das Nugayer Sharmeen2 Benjamin Yapo1,6
2

Kazi Md Noor-ul Ferdous2,7 Kh Ahasanul Kabir2,7 Mirza Kamrul Zahid1,2,8 Kathryn Ford1,5
Md Qumrul Ahsan2,9 Mastura Akter2 Md Afruzul Alam2 Mozammel Hoque2,9

1 Global Initiative for Children’s Surgery (GICS) Address for correspondence Tahmina Banu, MS, MD, FRCS,
2 Chittagong Research Institute for Children Surgery, Chittagong, Chittagong Research Institute for Children Surgery, 29 Panchlaish,
Bangladesh Chittagong 4203, Bangladesh (e-mail: proftahmina@gmail.com).
3 Department of Surgery, Li Ka Shing Faculty of Medicine, The
University of Hong Kong, Hong Kong
4 Department of Pediatric Surgery, Tehran University of Medical
Sciences, Tehran, Iran

Downloaded by: University of Iowa Libraries. Copyrighted material.

5 Oxford Children’s Hospital and University of Oxford, Oxford,
United Kingdom
6 National Department of Health, Mt Hagen Highlands Regional
Hospital, Mount Hagen, Papua New Guinea
7 Dhaka Shishu (Children) Hospital, Dhaka, Bangladesh
8 Shaheed Suhrawardy Medical College and Hospital, Dhaka,
Bangladesh
9 Department of Pediatric Surgery, Chattogram Ma O Shishu Hospital,
Bangladesh

Eur J Pediatr Surg

Abstract Introduction Published studies based on Krickenbeck classification of anorectal

malformations (ARMs) are still insufficient to assess the global as well as regional
relative incidence of different ARM subtypes, gender distribution, and associated
anomalies. The primary purpose of this study was to provide an estimate of those in
Global Initiative for Children's Surgery (GICS) research group.
Materials and Methods We collected ARM data prospectively for 1 year from four
institutes of different geographic locations. A total of 342 patients were included in this
study (195, 126, 11, and 10 from Bangladesh, Iran, Papua New Guinea, and Oxford,
United Kingdom, respectively).
Results Overall male to female ratio was 1:1. The most frequent ARM subtype was
Keywords perineal fistula (23.7% ¼ 81/342). About 48.5% (166/342) patients had at least one
► Global Initiative for associated anomaly. Cardiac and genitourinary systems were the most commonly
Children’s Surgery affected systems, 31.6% (108/342) and 18.4% (63/342), respectively. These organ-
► congenital anomalies systems were followed by anomalies of vertebral/spinal (9.9% ¼ 34/342), musculoskel-
► anorectal etal (4.4% ¼ 15/342), and gastrointestinal/abdominal (3.2% ¼ 11/342) systems. Rec-
malformations tovesical fistula had the highest percentage (96.4% ¼ 27/28) of associated anomalies.
► VACTERL About 18.1% (62/342) patients had multiple anomalies. ARMs (both isolated and with

received © Georg Thieme Verlag KG DOI https://doi.org/

March 21, 2019 Stuttgart · New York 10.1055/s-0039-1695789.
accepted after revision ISSN 0939-7248.
July 14, 2019
Global Initiative for Children’s Surgery ARM study Banu et al.

associated anomalies) occurred equally in males and females. Comparison between

patients from Bangladesh and Iran showed differences in relative incidence in ARM
subtypes. In addition, Iranian patient group had higher percentage of associated
anomalies compared with Bangladeshi (73 vs. 35.4%).
Conclusion Our study provides important insights about ARM subtypes, gender
distribution and associated anomalies based on Krickenbeck classification especially
from Bangladesh and Iran.

Introduction for vertebral/spinal anomalies. Ultrasonography (USG) were

Anorectal malformations (ARMs) are congenital defects occur-
ring with an estimated worldwide incidence of 2 to 6 per
10,000 live births and are the commonly encountered congeni-
tal anomalies of the digestive tract.1 The defects occur primarily
due to abnormal development of the hindgut, affecting both
males and females. ARMs can occur as isolated form or with
other anomalies including syndromic forms and chromosomal
done to assess the abdominal, genitourinary, and spinal anom-
alies. If the USG was found to be abnormal for the urinary
system, it was followed up with micturating cystourethrogram.
Patients were grouped into neonate (0–28 days), infant
(29 days to 1 year), toddler (>1 to 3 years), children (>3
years to 10 years), adolescent (>10 years to 19 years), and
adult (>19 years). Sex of the patients was determined primar-
ily on the basis of clinical aspects (i.e., without karyotyping).

Downloaded by: University of Iowa Libraries. Copyrighted material.

aberrations.2 Most ARMs are thought to be multifactorial with
contribution of both non-genetic1 and genetic factors.3,4
Since the new Krickenbeck classification system was
introduced in 2005,5 published studies based on this classi-
fication system are still insufficient to assess the global as
well as regional incidence of different ARM subtypes, gender
distribution, and associated anomalies. The purpose of this
study was to assess the epidemiological status of ARMs with
special emphasis on low- and middle-income countries.
The associated anomalies were categorized into different
organ systems: cardiovascular, genitourinary, vertebral/spi-
nal, musculoskeletal, gastrointestinal/abdominal, facial/head,
brain, and syndromic. If a patient had associated anomalies in
multiple systems, it was considered in each system separately.
In addition, patients were considered to have VACTERL, if there
were ARM and at least two of the following congenital
malformations: vertebral defects, cardiac defects, tracheoeso-
phageal fistula, renal anomalies, and limb anomalies.7

Materials and Methods
Study Design and Data Collection
After obtaining ethical approval from the respective author-
ities, we collected data prospectively for 1 year (2017–2018)
from Bangladesh (Chittagong Research Institute for Children
Surgery [CRICS]), Iran (Children’s Medical Center, Tehran),
Papua New Guinea (Mt Hagen Highlands Regional Hospital),
and Oxford (Oxford Children’s Hospital). Most of the patients
in this study were from Bangladesh and Iran. CRICS is one of
the three referral centers for pediatric surgical patients in
Chittagong with a population of nearly 32 million.6 Child-
ren’s Medical Center is one of the four pediatric referral
hospitals in Tehran with a population of 8.7 million.
Both newly admitted patients and patients who came for
follow-up at that time were included except in Oxford, where
only newly admitted patients were included. For patients who
came for follow-up, previous medical records were reviewed
for data collection. Patients were included for once. Age rep-
resents the age of the patient at the time of data collection.
These patients were not further followed up for data collection
in this study. Data were primarily collected on four aspects: age
at the time of data collection for this study, gender, Krickenbeck
subtype, and associated anomalies. No predefined specific
screening protocol for identification of associated anomalies
was followed. It was based on patient’s clinical presentation
and surgeon’s clinical suspicion. In general, cardiac imaging
was done with two-dimensional echocardiography; plain
abdominal and thoracolumbosacral radiographs were taken
Statistical Analyses
Data were analyzed with Microsoft Excel and GraphPad
Prism 7. Figures were generated with GraphPad Prism 7.
Frequencies and percentages were used to describe categor-
ical variables, and median and interquartile ranges (IQR)
were used to describe continuous variable (age). Chi-squared
test, Fisher’s exact test, and Mann–Whitney U test were
performed for comparisons where appropriate. If needed
Bonferroni correction for multiple tests was done. A p-value
less than 0.05 was considered to be statistically significant.
Results
A total of 342 patients were included in this study (195, 126,
11, and 10 from Bangladesh, Iran, Papua New Guinea, and
Oxford, respectively). Data were analyzed considering all the
patients together. In addition, individual country-based
analyses for Bangladesh and Iran were also done and results
were compared whenever appropriate.
Age and Gender Distribution
Median age and IQR were calculated for all the patients (i.e.,
newly admitted patients and patients coming for follow-up).
Overall median age was 4.4 days (IQR: 1.8–211.7 days) and
60.5% patients were neonates. Median age for Bangladeshi
and Iranian patients were 14.6 days (IQR: 1.8 days–2 years)
and 2.9 days (IQR: 1.8–21.9 days), respectively (Mann–Whit-
ney U-test, p < 0.0001). Patient percentage according to
different age groups is shown in ►Fig. 1. Overall male to

European Journal of Pediatric Surgery

 Global Initiative for Children’s Surgery ARM study Banu et al.

167) was the predominant subtype followed by perineal

fistula (22.2% ¼ 37/167) and rectovaginal fistula
(21.6% ¼ 36/167). There were 12 (3.5%) female and 7 (2%)
male patients with cloacal malformation.
Bangladeshi and Iranian patient groups had different relative
proportions of ARM subtypes (►Table 1). In Bangladesh, the top
three subtypes were ARMs without fistula (28.7%), vestibular
fistula (20%), and perineal fistula (18.5%). Proportions of ARMs
without fistula and vestibular fistula were significantly higher
in Bangladesh than Iran. On the other hand, perineal fistula
(31%), rectovaginal fistula (25.4%), and rectovesical fistula
(20.6%) were the top three subtypes in Iran. The proportions
of rectovaginal and rectovesical fistula were significantly higher
in Iran than Bangladesh (►Table 1).

Associated Congenital Anomalies

Overall, 51.5% (176/342) patients had isolated ARM, whereas
Fig. 1 Distribution of patients into different age groups.
the remaining 48.5% (166/342) patients had at least one
associated anomaly. Cardiac and genitourinary systems were

Downloaded by: University of Iowa Libraries. Copyrighted material.

female ratio was 1:1 (174 vs. 167 patients, respectively). In
Bangladesh, 50.8% (99/195) patients were male and 48.7%
(95/195) were female. One patient (0.5%) had disorders of sex
development (DSD). In Iran, male and female patients were
48.4% (61/126) and 51.6% (65/126), respectively (►Fig. 1).
ARM Subtypes
Overall, the most frequent ARM subtype was perineal fistula
(23.7% ¼ 81/342), followed by ARMs without fistula
(17.8% ¼ 61/342). In males, the largest groups of ARM sub-
types were rectourethral fistulas (25.9% ¼ 45/174) and peri-
neal fistula (25.3% ¼ 44/174), and ARMs without fistula
(24.7% ¼ 43/174). In females, vestibular fistula (30.5% ¼ 51/
the mostly affected systems (including both the only associated
anomaly and as part of multiple anomalies) with 31.6% (108/
342) and 18.4% (63/342) patients, respectively. These organ
systems were followed by vertebral/spinal (9.9% ¼ 34/342),
musculoskeletal (4.4% ¼ 15/342), and gastrointestinal/abdom-
inal (3.2% ¼ 11/342) systems. Associated multiple anomalies
( 2 systems involved in addition to the ARM) were present in
18.1% (62/342) patients. Total 10.8% (37/342) cases (18 males,
18 females, and 1 DSD) had VACTERL association.
Among the ARM subtypes, rectovesical fistula had the
highest percentage (96.4% ¼ 27/28) of associated anomalies
followed by rectovaginal fistula (80.6% ¼ 29/36) and cloacal
malformations (male and female ¼ 78.9% ¼ 15/19) (►Table 2).

Table 1 Comparison of ARM subtypes between Bangladesh and Iran (p-values are shown after Bonferroni adjustment)

Subtypes Bangladesh Iran p-Value

n % n %
Perineal fistula 36 18.5 39 31.0 0.24
ARMs without fistula 56 28.7 – – < 0.0001
Vestibular fistula 39 20.0 9 7.1 0.02
Rectobulbar urethral fistula 24 12.3 11 8.7 > 0.9999
Rectovaginal fistula 4 2.1 32 25.4 < 0.0001
Rectovesical fistula 2 1.0 26 20.6 < 0.0001
Cloacal malformation (female) 7 3.6 5 4.0 > 0.9999
Anal stenosis 7 3.6 – – 0.72
Male cloaca 7 3.6 – – 0.72
H-type fistula 5 2.6 – – > 0.9999
Rectoprostatic urethral fistula 1 0.5 3 2.4 > 0.9999
Pouch colon 2 1.0 1 0.8 > 0.9999
Anorectal duplication 2 1.0 – – > 0.9999
Rectobladder neck fistula 2 1.0 – – > 0.9999
Posterior cloaca 1 0.5 – – > 0.9999
Total 195 100 126 100

Abbreviation: ARM, anorectal malformation.

European Journal of Pediatric Surgery

Global Initiative for Children’s Surgery ARM study Banu et al.

Table 2 ARM subtypes and associated abnormalities (subtypes Discussion

with < 2% patients are not shown)
ARMs present with a wide spectrum of clinical manifesta-
Types of ARM Total Patients with associated tions and associated anomalies and therefore pose a signifi-
number anomalies cant challenge to the healthcare professionals. Most patients
n n % need early surgical interventions in their life. However, the
Perineal fistula 81 24 29.6 majority of the patients do not obtain complete anatomical
and functional restoration of their bowel that affect their
ARMs without 61 21 34.4
fistula quality of life.1 In this study, we investigated 342 patients
with ARMs to estimate relative incidence of different ARM
Vestibular fistula 52 21 40.4
subtypes according to the Krickenbeck classification, gender
Rectobulbar 41 21 51.2 distribution, and associated anomalies. To our knowledge,
urethral fistula
this is the first study of this cohort size from Bangladesh and
Rectovaginal 36 29 80.6 Iran based on this classification.
fistula
Rectovesical 28 27 96.4 Gender Distribution of ARMs
fistula The reported male to female ratio of ARMs varies among
Cloacal 12 10 83.3 different studies. Some studies including the first report of
malformation the European consortium on anorectal malformations (ARM-

Downloaded by: University of Iowa Libraries. Copyrighted material.

(female)
Male cloaca 7 5 71.4
Abbreviation: ARM, anorectal malformation.
When comparison was made between Bangladesh and Iran,
we found that ARM patients with associated anomalies were
35.4 and 73% in these countries, respectively (chi-squared test,
p ¼ < 0.0001). No statistically significant differences in gender
distribution were observed between Bangladesh and Iran,
both in isolated ARM patients and in patients with associated
congenital anomalies (►Fig. 2). In other words, males and
females were equally affected in both isolated ARMs and ARMs
with associated anomalies in both countries.
Fig. 2 Gender distribution of isolated ARMs and ARMs with additional
associated congenital anomalies in Bangladesh and Iran. ARM, anorectal
malformation.
NET) of 203 patients8 and results from United States Midwest
Pediatric Surgery Consortium of 506 patients9 showed no
overall gender bias. In line with these studies, we also found
an overall male to female ratio of 1:1 in our study. We did not
find statistically significant differences in gender distribution of
isolated ARMs and ARMs with additional congenital malfor-
mations, which remained true even in individual country-
based analyses of patients from Bangladesh and Iran. Unlike
our study, a slight male predominance of 1.5- to 1.8-fold was
also found in large-scale studies like the European Surveillance
of Congenital Anomalies (EUROCAT) of 1,414 patients8 and a
study of 1,992 patients from Japan.10
Relative Incidence of ARM Subtypes
Although the relative incidence of ARM subtypes varies widely
across different studies, perineal fistula was consistently
reported to be the most predominant subtype (35–55%) in
most of the major studies.8,9,11,12 In our combined dataset, the
proportion of perineal fistula was relatively lower (23.7%) than
those studies, although it was still the most predominant
subtype. This relatively lower incidence of perineal fistula in
our study may be a result of selection bias as we included
patients coming for follow-up as well, thus including more
patients with severe forms of ARMs (as these patients are more
likely to come for follow-up). Interestingly, statistically signifi-
cant differences in relative incidence of different ARM subtypes
were observed between Bangladesh and Iran. In Bangladesh,
the top three subtypes were ARMs without fistula, vestibular
fistula, and perineal fistula, respectively. In Iran, the top three
subtypes were perineal fistula, rectovaginal fistula, and recto-
vesical fistula, respectively. The proportions of ARMs without
fistula and vestibular fistula were significantly higher in
Bangladesh than in Iran. On the other hand, rectovaginal and
rectovesical fistula were significantly higher in Iran than in
Bangladesh. Although factors responsible for these observed
differences in ARM subtypes are not clearly understood, other
factors might play a role including genetic and environmental
factors, differences in institutional expertise, and differences in
the patient referral system based on severity. Surprisingly, 3.6%

European Journal of Pediatric Surgery

 Global Initiative for Children’s Surgery ARM study
(7/195) of Bangladeshi ARM patients had male cloacal malfor-
mations, which was significantly higher than previously
reported percentage of 0.2%.13 We assume that this was due
to increased referral to the participating institute in Bangladesh
since the publication of male cloacal malformation in 2013.13
ARMs and Associated Congenital Anomalies
Patients with ARMs commonly have additional anomalies in
one or more organ systems including genitourinary, cardiac,
and spinal systems.2 The associated anomaly can itself be
another rare disease, for example, Hirschsprung’s disease.14,15
It might also occur as a part of syndromes like Currarino
syndrome, Pallister-Hall syndrome, or with chromosomal ab-
normalities like trisomy 21 and trisomy 18.2 The incidence of
the associated anomalies ranges from 40 to 70%.1 In our study,
overall 48.5% patients had at least one additional congenital
anomaly. However, the rate differed significantly between
Bangladesh and Iran (35.4 and 73%, respectively). As mentioned
earlier, in addition to the environmental and genetic factors,
referral system and institutional differences in screening pro-
tocols for the search of additional anomalies might be impor-
tant factors for variations across different studies as well as in
our study. Therefore, to get a better estimate, comparison
among national databases would be more appropriate.
Cardiovascular, genitourinary, and vertebral/spinal anoma-
lies were the most commonly affected organ systems in patients
with ARMs in our study which are consistent with other
studies.8,9,16 We found VACTERL association in 10.8% ARM
patients that is similar to the result of EUROCAT study (12%).8
Conclusion
In this study, ARMs (both isolated and with associated anoma-
lies) occurred equally in males and females. Overall, perineal
fistula was the most frequent ARM subtype; however, the
relative frequency of different ARM subtypes differed between
Bangladesh and Iran. The frequency of associated anomalies
increased with the complexity of the ARM subtype.
Limitations
The detection rate of associated anomalies might have varied
from country to country due to variation in expertise and
facilities. In addition, as only one institute was involved from
each country, our data might not represent the respective
country. Nevertheless, our study provides important insights
about ARMs based on Krickenbeck classification.
Funding
None.
Conflict of Interest
None declared.
Banu et al.
Acknowledgments
The authors would like to thank the institutions that
provided the data.
References
1 Wijers CH, van Rooij IA, Marcelis CL, Brunner HG, de Blaauw I,
Roeleveld N. Genetic and nongenetic etiology of nonsyndromic
anorectal malformations: a systematic review. Birth Defects Res C
Embryo Today 2014;102(04):382–400
2 Levitt MA, Peña A. Anorectal malformations. Orphanet J Rare Dis
2007;2:33
3 Zhang T, Tang XB, Wang LL, et al. Mutations and down-regulation
of CDX1 in children with anorectal malformations. Int J Med Sci
2013;10(02):191–197
4 Hsu JSJ, So M, Tang CSM, et al. De novo mutations in Caudal Type
Homeo Box transcription Factor 2 (CDX2) in patients with persis-
tent cloaca. Hum Mol Genet 2018; 27(02):351–358
5 Holschneider A, Hutson J, Peña A, et al. Preliminary report on the
International Conference for the Development of Standards for
the Treatment of Anorectal Malformations. J Pediatr Surg 2005;40
Downloaded by: University of Iowa Libraries. Copyrighted material.
(10):1521–1526
6 Government of the People’s Republic of Bangladesh. Ministry
of Health and Family Welfare (MOHFW). Health bulletin 2017.
Available at: http://www.dghs.gov.bd/images/docs/Publicaations/
HealthBulletin2017Final13_01_2018.pdf. Accessed on 29.1.2019
7 Solomon BD. VACTERL/VATER Association. Orphanet J Rare Dis
2011;6:56–56
8 de Blaauw I, Wijers CH, Schmiedeke E, et al. First results of a
European multi-center registry of patients with anorectal mal-
formations. J Pediatr Surg 2013;48(12):2530–2535
9 Minneci PC, Kabre RS, Mak GZ, et al; Midwest Pediatric Surgery
Consortium. Screening practices and associated anomalies in
infants with anorectal malformations: results from the Midwest
Pediatric Surgery Consortium. J Pediatr Surg 2018;53(06):
1163–1167
10 Endo M, Hayashi A, Ishihara M, et al; Steering Committee of
Japanese Study Group of Anorectal Anomalies. Analysis of 1,992
patients with anorectal malformations over the past two decades
in Japan. J Pediatr Surg 1999;34(03):435–441
11 van Rooij IA, Wijers CH, Rieu PN, et al. Maternal and paternal risk
factors for anorectal malformations: a Dutch case-control study.
Birth Defects Res A Clin Mol Teratol 2010;88(03):152–158
12 Nah SA, Ong CC, Lakshmi NK, Yap TL, Jacobsen AS, Low Y.
Anomalies associated with anorectal malformations according
to the Krickenbeck anatomic classification. J Pediatr Surg 2012;47
(12):2273–2278
13 Banu T, Chowdhury TK, Hoque M, Rahman MA. Cloacal
malformation variants in male. Pediatr Surg Int 2013;29(07):
677–682
14 Hofmann AD, Puri P. Association of Hirschsprung’s disease and
anorectal malformation: a systematic review. Pediatr Surg Int
2013;29(09):913–917
15 Karim A, Akter M, Aziz TT, et al. Epidemiological characteristics
of Hirschsprung’s disease (HSCR): results of a case series of
fifty patients from Bangladesh. J Pediatr Surg 2018;53(10):
1955–1959
16 Theron AP, Brisighelli G, Theron AE, Leva E, Numanoglu A.
Comparison in the incidence of anorectal malformations between
a first- and third-world referral center. Pediatr Surg Int 2015;31
(08):759–764
European Journal of Pediatric Surgery