Journal of Pediatric Surgery 52 (2017) 1776–1781

Contents lists available at ScienceDirect

Journal of Pediatric Surgery

journal homepage: www.elsevier.com/locate/jpedsurg

Pathologically assessed grade of Hirschsprung-associated enterocolitis in

resected colon in children with Hirschsprung's disease predicts
postoperative bowel function☆
Siyang Cheng a,b,1, Jun Wang a,b,1, Weihua Pan a,b, Wenbo Yan a, Jia Shi a, Wenbin Guan c,
Yang Wang b,d, Wei Cai a,b,d,⁎
a
Department of Pediatric Surgery, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, No. 1665, Kongjiang Road, Shanghai, 200092, China
b
Shanghai Key Laboratory of Pediatric Gastroenterology and Nutrition, No. 1665, Kongjiang Road, Shanghai, 200092, China
c
Department of Pathology, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, No. 1665, Kongjiang Road, Shanghai, 200092, China
d
Shanghai Institute for Pediatric Research, No. 1665, Kongjiang Road, Shanghai, 200092, China

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: The aim of this study was to investigate the relationship between the grade of enterocolitis on patholog-
Received 6 November 2016 ical assessment of resected colon and postoperative bowel function in children with Hirschsprung's disease (HD).
Received in revised form 27 February 2017 Methods: Children with HD who were seen at a large tertiary center from January 2012 to December 2013 were
Accepted 20 March 2017 enrolled into this study. Resected colon was assessed using the histopathologic grade scoring system for
Hirschsprung-associated enterocolitis (HAEC), and the relationship of these scores to postoperative bowel func-
Key words:
tion was assessed. Time of recovery to normal defecation was the primary outcome measure. The t-test, analysis
Hirschsprung's disease
Hirschsprung-associated enterocolitis
of variance, and Kaplan–Meier, univariate, and multiple regression analyses were performed.
Aganglionosis Results: Eighty children with HD (median age at repair 7.9 months; range 1.3 months to 9 years) were included in
Pull-through procedure the study. Nineteen children dropped out of the study and were considered as providing censored data, giving a
Pathology follow-up rate of 76.3%. A total of 21 children (34.4%) were admitted to hospital with at least one episode of en-
terocolitis. Multivariate Cox proportional hazards models showed that compared with patients with a normal
proximal colon, those with an inflamed proximal segment had a 1.5-fold higher risk of a poor recovery. Logistic
regression analyses suggested that postoperative HAEC admissions increased by 57% with each HAEC patholog-
ical grade of the transitional segment and by 50% with each grade of the overall segment. Compared with normal
bowel in the transitional segment, the detection of grade ≥3 HAEC in the transitional area increased the incidence
of postoperative HAEC by 4.75-fold.
Conclusions: Children whose resected proximal colon showed inflammation on pathological assessment were at
risk of poor recovery after surgery. A higher pathological HAEC score for the sum of the overall three segments
suggested an increased risk for the subsequent development of enterocolitis. Among three segments, the severity
of enterocolitis in the transitional segment was the most significant factor in predicting postoperative HAEC.
Type of study: Clinical study.
Level of evidence: Moderate.
© 2017 Elsevier Inc. All rights reserved.

The past decades have witnessed huge advances in the diagnosis and
treatment of Hirschsprung's disease (HD). With the development of the
one-stage Soave procedure, the transanal endorectal pull-through pro-
cedure (TERPT) has been increasingly popular. This procedure allows
Abbreviations: ANOVA, analysis of variance; CI, confidence interval; HAEC,
Hirschsprung-associated enterocolitis; HD, Hirschsprung's disease; HR, hazard ratio; OR,
odds ratio; SD, standard deviation; TCA, total colonic aganglionosis; TERPT, transanal
endorectal pull-through procedure; TOR, time of postoperative recovery.
☆ The manuscript was funded by the Shanghai science committee, No.14411950405.
⁎ Corresponding author at: Department of Pediatric Surgery, Xinhua Hospital, Shanghai
Jiao Tong University School of Medicine, No. 1665, Kongjiang Road, Shanghai, 200092,
China.
E-mail address: caiw1978@163.com (W. Cai).
1
Siyang Cheng and Jun Wang contributed equally to this work.
minimally invasiveness surgery, and reduces the operation time and du-
ration of hospital stay. Nevertheless, outcomes remain unsatisfactory
for some patients, postoperative complications remain challenging for
surgeons, and the factors influencing postoperative bowel function re-
main uncertain. While many studies have focused on the incidence of
postoperative complications, particularly Hirschsprung-associated en-
terocolitis (HAEC), less attention has been paid to the time of postoper-
ative recovery (TOR) after a definitive procedure, which is another
factor that directly reflects bowel function.
HAEC is a serious and potentially life-threatening complication for
HD children. Today, the typical clinical characteristics of HAEC are sel-
dom seen in clinics before surgical treatment owing to the development
of antibiotics and the increasingly valued use of enemas in hospitals.

http://dx.doi.org/10.1016/j.jpedsurg.2017.03.056
0022-3468/© 2017 Elsevier Inc. All rights reserved.
 S. Cheng et al. / Journal of Pediatric Surgery 52 (2017) 1776–1781
Preoperative HAEC is considered to be a high risk factor for postopera-
tive HAEC, which can be a stubborn problem in children undergoing a
definitive procedure, increasing the duration of hospitalization and
costs of management. The histopathologic grading system for HAEC,
first proposed by Elhalaby et al. [1], has been suggested as an effective
method of predicting postoperative HAEC. However, the results of stud-
ies of this subject have been conflicting.
To evaluate these issues, we studied the effect of histopathologic
changes in HAEC grade in colon resected during a definitive procedure
for HD by measuring the TOR to normal defecation and testing the hy-
potheses that: (1) children with a pathologically inflamed proximal
colon would tend to have a poor recovery after surgical treatment;
and (2) a higher HAEC histopathologic grade in colon resected during
pull-through procedure would be associated with an increased risk of
postoperative HAEC.
1. Material and methods
1.1. Clinical data
The population for this retrospective study included all children with
HD who underwent surgical treatment at the Department of Pediatric
Surgery of Xinhua Hospital from January 2012 to December 2013. Spe-
cifically, all children underwent a Soave procedure. Children who
were diagnosed as having total colonic aganglionosis (TCA) were ex-
cluded from this study. There were no participants with Down's syn-
drome or other congenital malformation or patients with a family
history of HD in this cohort.
All the definitive procedures in the study were elective. The length of
colon that should be excised was determined by morphologic changes
in the colon and rapid frozen-section diagnosis in a definitive procedure.
Our method of definitive procedure referred to the pull-through meth-
od of Prof. Jacob C. Langer [2], so there is hardly muscular cuff remaining.
Occasionally a 1 cm muscular cuff remained, and a posterior myotomy
of the sleeve would then be performed on this muscular cuff through
a “V” shape on the back wall.
Data were collected on patient demographics, birth weight, details
of the definitive procedure (i.e. age, weight, type of repair, date of oper-
ation, dates of admission and discharge, intraoperative complications),
preoperative clinical HAEC, and type of HD. Data on postoperative re-
covery and postoperative complications, including stricture, anastomot-
ic leaking, fecal pollution, and postoperative admissions for HAEC, were
also analyzed.
1.2. Postoperative care
Patients were required to attend the follow-up visits approximately
every 3 months after their definitive procedure and the details of
children's recovery were recorded. We telephoned all of the children's
families 2 years after the repair operation to confirm their recovery
and that they had not recidivated, and again measured the TOR. Impor-
tantly, routine postoperative home dilatations are recommended for all
children seen in our center. The final event was designated as the recov-
ery of normal defecation; in other words, children who defecated with-
out constipation, diarrhea, soiling, or social problems were regarded as
cured. Considering that the return to normal defecation is a functional
recovery after definitive operation, a simple constipation or a diarrhea
would be regarded as a functional disorder and was defined according
to the diagnostic criteria for children aged 6 months to 3 years old in
the Rome III Diagnostic Criteria [3]. Postoperative HAEC was recorded
according to the Delphi analysis on HAEC [4], after children suffered
some high relative symptoms (i.e. fever, abdominal distension, explo-
sive diarrhea, constipation, foul-smelling or bloody stool) and went to
the hospital.
1777
1.3. HAEC pathological grade assessment
All paraffin sections of colon excised from HD patients, including ste-
nosed, transitional, and dilated segments, were determined according
to the initial records in pathological department which had noted
down the rapid frozen-section diagnosis in the definitive procedure.
Then these paraffin sections would be respectively assessed according
to the HAEC grade system proposed by Elhalaby et al. [1]. All assess-
ments were conducted by author Guan Wenbin. Details of the scoring
system are shown in Fig. 1. Since no perforation occurred in this cohort,
we regarded ulceration reaching to the smooth muscle layer as
transmural necrosis (marked as grade V).
1.4. Statistical analysis
All statistical analyses were conducted using SPSS version 23.0 (IBM
Corp., New York, NY, USA). P ≤ 0.05 was considered statistically signifi-
cant. The pathologically assessed HAEC grade scores of resected colon
are described as means ± SDs. Group comparisons of HD patients' base-
line characteristics were conducted using t-tests or analysis of variance
(ANOVA) followed by the Bonferroni multiple-comparison test, as ap-
propriate. Each participant's time from initial hospitalization to TOR
was identified, either censored or to the end of the study, whichever
came first. Cox proportional hazards models were used to assess the as-
sociation between the pathologically assessed HAEC grade of resected
colon and postoperative recovery in HD patients by calculating
multivariable-adjusted hazard ratios (HRs) and 95% confidence inter-
vals (CIs). We considered two sequential models in the analysis:
model 1 was adjusted for children's age and sex; and model 2 was addi-
tionally adjusted for preoperative HAEC (yes vs. no), type of HD (long vs.
rectosigmoid), and initial colostomy (yes vs. no). In addition, we exam-
ined the association between the pathologically assessed HAEC grade of
resected colon and postoperative HAEC incidence in HD patients using
logistic regression models. Multivariable adjusted odds ratios (ORs)
and 95% CIs were calculated using the same sequential adjustment
strategy as above.
2. Results
2.1. Details of overall population and HAEC grade score assessment
Data from 80 children (65 boys and 15 girls) with HD who received a
pull-through procedure were analyzed. Details of the overall study pop-
ulation and the operation are shown in Table 1. The incidence of preop-
erative clinical HAEC was 7.5% (six patients). No intraoperative
complications were reported. At the end of the study, 19 patients
were considered as providing censored data, as they stopped attending
postoperative appointments and could not be contacted by telephone.
Recovery of defecation during the first 2 years after the definitive proce-
dure was confirmed in 52 children. The median for the length of follow-
up was 2 years (range 2 months to 2 years).
Paraffin sections of excised colon were assessed and given a grade
score according to the HAEC pathological grade system. Mean ± SD
HAEC scores of stenosed, transitional, and dilated segments from the
80 patients were 2.69 ± 1.73, 1.11 ± 1.48, and 0.46 ± 0.98, respectively.
The differences among groups were statistically significant in one-way
ANOVA grouped by different segments (F = 49.766, P b 0.0001). Subse-
quent comparisons between two segments using the Bonferroni meth-
od all showed statistical significance (all P b 0.05). The mean ± SD score
of the overall resected colon was 4.31 ± 2.9 (range 0–12).
Differences in baseline HAEC scores according to various characteris-
tics are shown in Table 2. As shown, the majority of baseline character-
istics did not significantly influence the HAEC scores of the three
segments. The exceptions were the HAEC score for the transitional seg-
ment in boys, which was statistically higher than that of girls; and the
HAEC score for the proximal segment, which was higher in children
1778 S. Cheng et al. / Journal of Pediatric Surgery 52 (2017) 1776–1781

Fig. 1. Fig. 1 images were all from paraffin sections of HD patients in this study. The pathological grade system of HAEC with increasing severity is as follows: (A) Grade 0, normal mucosa;
(B) Grade I, crypt dilation and retained mucin; (C) Grade II, cryptitis or ≤2 crypt abscesses per high-power field (HPF); (D) Grade III, multiple crypt abscesses per HPF; (E) Grade IV, mucosal
ulcerations or intraluminal fibrinopurulent debris. (F) Grade V, muscular ulcerations [5].

who underwent a two-stage procedure than in those who had a one- The results of multivariate Cox regression analysis are shown in
stage operation. Table 3. Specifically, age was counted as a continuous variable. The
final multivariate Cox model indicated that compared with a normal
2.2. Survival analysis of recovery
Table 2
Data from 80 patients, with 19 children dropping out (follow-up rate Difference in HAEC pathological grade assessment of resected colon by different baseline
characteristics in HD patients (n = 80).
76.3%), were used in the analysis of recovery. The rate of recovery in the
first 2 years was 84.8%, with a median TOR of 9.0 months. Characteristics n HAEC pathological grade score
Kaplan–Meier analysis showed that children with an inflamed prox- Distal Transitional Proximal Overall
imal resected colon were significantly more likely to have a slower re-
Average of overall 80 2.69 ± 1.73 1.11 ± 1.48 0.46 ± 0.98 4.31 ± 2.86
covery (Fig. 2), with a log-rank chi-square value of 12.92 (P b 0.0001). Age
Median TOR was 8.0 months in patients with a normal proximal colon b1 year old 58 2.66 ± 1.71 1.28 ± 1.57 0.53 ± 1.01 4.53 ± 2.92
and 18.0 months in those with an inflamed proximal colon. ≥1 year old 22 2.77 ± 1.73 0.68 ± 1.17 0.27 ± 0.88 3.73 ± 2.69
P value – 0.787 0.072 0.289 0.263
Gender
Table 1 Girl 15 2.47 ± 1.85 0.47 ± 1.13 0.49 ± 1.06 3.40 ± 2.59
Details of the study population and operation. Boy 65 2.74 ± 1.71 1.26 ± 1.52 0.48 ± 0.97 4.52 ± 2.90
P value – 0.586 0.030⁎ 0.786 0.172
Details of population Overall
HD type
N 80 Rectosigmoid 61 2.87 ± 1.68 1.05 ± 1.43 0.44 ± 1.06 4.39 ± 2.78
No. of males(%) 65(81%) Long 19 2.11 ± 1.79 1.32 ± 1.67 0.53 ± 0.70 4.05 ± 3.19
Median age at operation 7.9 m (1.3 m-9y) P value – 0.092 0.498 0.747 0.653
No. of recovery in the first 2 years 52 Preoperative HAEC
No. of loss in follow up 19 No 74 2.66 ± 1.76 1.07 ± 1.48 0.41 ± 0.91 4.18 ± 2.78
Yes 6 3.00 ± 1.27 1.67 ± 1.51 1.17 ± 1.60 6.00 ± 3.69
Details of operation Overall P value – 0.563 0.345 0.300 0.134
Laparoscopic ERPT, n (mean(± SD) OR time) 21 (208 ± 73 min) Initial colostomy
Open ERPT, n (mean(±SD) OR time) 59 (187 ± 77 min) No 55 2.78 ± 1.80 1.05 ± 1.46 0.25 ± 0.80 4.15 ± 2.77
No. of one-stage repair 55 Yes 25 2.48 ± 1.56 1.24 ± 1.56 0.92 ± 1.19 4.68 ± 3.10
Hospitalization after one-stage repair 9.3 ± 7.0d P value – 0.472 0.608 0.015⁎ 0.442
Hospitalization after two-stage repair 12.4 ± 5.5d
Abbreviations: HAEC, Hirschsprung-associated enterocolitis; HD, Hirschsprung's disease.
Abbreviations: OR, operation room. Data are means ± standard deviations. P values were obtained by using t test.
 S. Cheng et al. / Journal of Pediatric Surgery 52 (2017) 1776–1781 1779

Fig. 2. Kaplan–Meier analysis on postoperative recovery between the children whose proximal resected colon were normal and those whose proximal resected colon were inflamed.

proximal segment, an inflamed proximal dilated segment of resected
colon (score 1–5) was a statistically significant risk factor for postoper-
ative recovery (HR 0.38, 95% CI 0.17–0.87). Furthermore, making the
HAEC score of the proximal colon into a continuous variable, even a
score of only 1–2 was associated with a significantly lower likelihood
of postoperative recovery (HR 0.39, 95% CI 0.16–0.94).
2.3. Logistic regression analysis of postoperative HAEC incidence
For postoperative complications, the incidences were calculated ex-
cluding the data from the 19 patients who were lost to follow-up. In the
remaining 61 children, no stricture and anastomotic leaking occurred.
Only five children (8.2%) suffered fecal pollution.
For postoperative HAEC,21 children (34.4%) suffered at least one ep-
isode of HAEC during the first 2 years. As clearly shown in Table 4, after
controlling for other covariates, children with a higher HAEC patholog-
ical grade score in the transitional zone had an increased risk of postop-
erative HAEC (OR 1.57 per grade). The same result was seen for the total
score of the overall segments (OR 1.50 per grade). Moreover, making
the scores into a continuous variable, children with HAEC scores of
3–5 in the transitional zone had almost a six-fold increased risk of ad-
mission for postoperative HAEC compared with children whose transi-
tional segment colon was not inflamed.

Table 3
Multivariable-adjusted HR (95% CI) of the association between HAEC pathological grade assessment of resected colon and postoperative recovery in HD patients (n = 80)a.

HAEC pathological grade score Linear association

0 1–2 3–5 1–5

Distal
No. of recovery 11 13 28 41 –
No. of participants 15 19 46 65 –
Model Ib 1.00 0.87(0.37–2.04) 1.10(0.51–2.41) 1.01(0.48–2.11) 1.03(0.87–1.22)
Model IIc 1.00 1.00(0.41–2.40) 0.92(0.42–2.02) 0.94(0.44–2.01) 0.98(0.83–1.16)

Transitional
No. of recovery 33 8 11 19 –
No. of participants 44 21 15 36 –
Model Ib 1.00 0.59(0.26–1.33) 0.69(0.33–1.45) 0.64(0.35–1.20) 0.89(0.73–1.08)
Model IIc 1.00 0.57(0.24–1.34) 0.60(0.27–1.32) 0.58(0.30–1.14) 0.85(0.69–1.05)

Proximal
No. of recovery 44 7 1 8 –
No. of participants 59 17 4 21 –
Model Ib 1.00 0.29(0.13–0.65) ⁎ 0.32(0.04–2.33) 0.29(0.13–0.63) ⁎ 0.57(0.35–0.93) ⁎
Model IIc 1.00 0.39(0.16–0.94) ⁎ 0.33(0.04–2.50) 0.38(0.17–0.87) ⁎ 0.70(0.44–1.10)
Overall 0 1–5 6–15 1–15
No. of recovery 9 30 13 43 –
No. of participants 10 44 26 70 –
Model Ib 1.00 0.73(0.32–1.67) 0.50(0.19–1.30) 0.66(0.29–1.50) 0.93(0.84–1.04)
Model IIc 1.00 0.90(0.39–2.07) 0.70(0.26–1.88) 0.86(0.38–1.97) 0.92(0.82–1.03)

Abbreviations: CI, confidence interval; HR, hazard ratio; HAEC, Hirschsprung-associated enterocolitis; HD, Hirschsprung's disease.
a
HR (95% CI) was estimated by using Cox proportional hazard model.
b
Model 1 was adjusted for children's age and gender.
c
Model 2 (final model) was additionally adjusted for preoperative HAEC (yes vs. no), type of HD (long vs. rectosigmoid), and initial colostomy (yes vs. no).
1780 S. Cheng et al. / Journal of Pediatric Surgery 52 (2017) 1776–1781

Table 4
Multivariable-adjusted OR (95% CI) of the association between HAEC pathological grade assessment of resected colon and postoperative HAEC incidence in HD patients (n = 61)a.

HAEC pathological grade score Linear association

0 1–2 3–5 1–5

Distal
No. of HAEC 3 4 14 18 –
No. of participants 12 16 33 49 –
Model Ib 1.00 0.76(0.12–4.64) 1.53(0.32–7.37) 1.23(0.27–5.66) 1.23(0.87–1.72)
Model IIc 1.00 0.25(0.03–2.17) 2.03(0.33–12.35) 1.03(0.21–5.18) 1.46(0.95–2.22)

Transitional
No. of HAEC 8 5 8 13 –
No. of participants 37 11 13 24 –
Model Ib 1.00 2.41(0.54–10.81) 4.25(1.03–17.46) ⁎ 3.29(1.00–10.80) ⁎ 1.47(1.02–2.11) ⁎
Model IIc 1.00 2.42(0.45–13.18) 5.75(1.13–29.17) ⁎ 3.85(0.99–14.93) 1.57(1.04–2.37) ⁎

Proximal
No. of HAEC 11 9 1 10 –
No. of participants 46 13 2 15 –
Model Ib 1.00 5.67(1.42–22.71) 3.68(0.18–76.11) 5.33(1.45–19.67) ⁎ 2.22(1.07–4.61)
Model IIc 1.00 3.02(0.63–14.49) 4.20(0.16–109.09) 3.19(0.74–13.81) 1.84(0.89–3.79)
Overall 0 1–5 6–15 1–15
No. of HAEC 1 9 11 20 –
No. of participants 9 34 18 52 –
Model Ib 1.00 2.17(0.22–21.22) 8.42(0.80–88.74) 3.62(0.39–33.57) 1.41(1.09–1.82) ⁎
Model IIc 1.00 1.31(0.12–13.91) 5.62(0.50–63.42) 2.28(0.23–22.32) 1.50(1.12–2.00) ⁎

Abbreviations: CI, confidence interval; OR, odds ratio; HAEC, Hirschsprung-associated enterocolitis; HD, Hirschsprung's disease.
a
OR (95% CI) was estimated by using Logistic model.
b
Model 1 was adjusted for children's age and gender.
c
Model 2 (final model) was additionally adjusted for preoperative HAEC (yes vs. no), type of HD (long vs. rectosigmoid), and initial colostomy (yes vs. no).

3. Discussion concomitant variables, an inflamed proximal segment continued to be

Thanks to medical advances, including diagnostic modalities, surgi-
cal techniques, and neonatal care, the curative rate of HD patients dra-
matically improved during the 20th century. Although there are few
arguments among the medical community with respect to the diagnosis
and treatment of HD, some children suffer unsatisfactory outcomes fol-
lowing a pull-through procedure. Furthermore, few studies have fo-
cused on the TOR.
Among the complications of HD, HAEC is responsible for the most se-
rious morbidity and mortality [5]. Elhalaby and Teitelbaum [1] initially
reported histopathologic changes associated with HAEC and suggested
the value of these changes in predicting postoperative HAEC. However,
their study did not clearly indicate whether they had assessed the HAEC
pathological grade of each of three segments of the resected colon, and
they excluded children with preoperative HAEC. Later studies, such as
that by Haricharan et al. [6], performed HAEC pathological grade assess-
ment on just the proximal segment of the resected bowel. In addition,
the potential impact of the pathologically assessed grade of HAEC of
the resected bowel on postoperative recovery has not previously been
assessed. As a supplement to previous studies, we performed the cur-
rent study with a series of 80 HD patients from a large tertiary center.
As shown in Table 2, the pathologic characteristics of HAEC could be
seen in both the ganglionic and aganglionic segments of the colon, as
previously reported by Elhalaby [1]. Three results are worth noticing:
Firstly, a total of 21 (26.3%) and 36 (45%) paraffin sections from the
proximal dilated and transitional zones, respectively, were found to be
inflamed, suggesting that this phenomenon is not a small-probability
event in the ganglionic area. Secondly, HAEC grade scores declined
sharply from the distal to the proximal colon. Thirdly, inflammation
was present to the same degree those patients that have undergone co-
lostomy. These results are contrary to the common clinical belief that
stools blocking the proximal colon will continually damage the intesti-
nal wall and induce enterocolitis and so that an undergone colostomy
could relieve the inflammation of the distal bowel.
As shown in Fig. 2, children with an inflamed proximal resected
colon had a markedly slower recovery during the first 2 years following
surgery (P b 0.0001). As shown in Table 3, after adjustment for all
a risk factor for poor postoperative recovery, with an HR of 0.38. In
other words, the rate of recovery to normal defecation in children
with a normal proximal colon was almost 2.5 times that of children
with an inflamed proximal colon in the first 2 years following surgery.
This result is similar to that reported by Schulten et al. [7], who found
that a histologically regular proximal segment of the colon resected pre-
dicted satisfactory postoperative bowel function. An inflamed proximal
segment suggests a high possibility that the remaining normal gangli-
onic colon will also be inflamed, as the proximal segment is closest to
the normal colon anastomosed to the anus. As a result, the postopera-
tive recovery of these children might be unsatisfactory.
At present, the best length of additional ganglionated bowel to resect
is unclear. Some authors have suggested optimizing the function of the
proximal bowel and placing more importance on rapid frozen-section
diagnosis in a definitive procedure to assess the maturation of ganglion
cells of the reversed bowel [7–10], while other authors have suggested
lengths of 2–15 cm [11]. The results of the current study may provide
a new method with which to evaluate the length of resection.
With the advancement of studies on etiology of HAEC, the proposed
potential etiologies have been classified into three main aspects (intes-
tinal barrier dysfunction, abnormal innate immune response and/or, ab-
normal microbiota) [4]. Meanwhile, many risk factors for HAEC have
been identified. For example, some genetic factors were considered to
possibly make contributions to the etiology of HAEC, such as the pres-
ence of trisomy 21, cartilage-hair hypoplasia, and a family history of
HD [12–18]. Several studies have reported that HAEC is more common
in children with long-segment HD [19,20]. In addition, preoperative
HAEC is widely accepted as a risk factor for postoperative HAEC.
Thanks to these findings, the incidence of preoperative HAEC has
markedly decreased over the past half century, from about 44% to
about 5% [21]; the incidence of preoperative HAEC among the patients
in this series was 7.5%. However, postoperative HAEC remains challeng-
ing. The rate of postoperative HAEC in the current study was 34.4%,
which is one of the highest incidence rates reported in studies to date
[22].
As shown in Table 4, we found that a HAEC grade score of ≥3 in the
transitional area was associated with a high risk of postoperative HAEC,
 S. Cheng et al. / Journal of Pediatric Surgery 52 (2017) 1776–1781
similar to the findings reported by Elhalaby et al. [1]. However, no cor-
relation was observed between the HAEC grade score of the proximal
segment and postoperative HAEC, which has also been previously re-
ported by Schulten et al. [7] and Hanimann et al. [23]. The results of
studies on the association between the histology of the resected bowel
and postoperative HAEC have been contradictory. This could be because
previous studies have not assessed the overall three segments of the
resected colon or studied only the proximal segment. Our analysis of lin-
ear associations further highlighted that the transitional segment has
the most significant role among the three segments in predicting post-
operative HAEC (OR 1.57 per grade). The same result in our analysis of
the overall resected colon suggests that the pathological assessment of
HAEC in the resected colon should be a reliable method of predicting
postoperative HAEC (OR 1.5 per grade), as the sum of HAEC grade scores
from the three segments provides the best reflection of the overall in-
flammation of the total colon.
For HAEC, the most ideal treatment may be its prevention. The use of
enemas is one of the most effective preventive strategies, for its contri-
butions to limiting the colonic distention and fecal stasis. This treatment
has been proved to be effective in reducing the incidence of postopera-
tive HAEC [24] and has also been recommended when surgical treat-
ment is delayed [25]. Probiotics is thought to confer a prophylactic
benefit to HD patients on HAEC, but the multicenter RCT study conduct-
ed by El-Sawaf et al. [26] failed to demonstrate any positive results of it.
In this cohort, we performed the rectal washouts twice a day before
surgical management on not only the patients who had been diagnosed
as HD but also the children who were suspicious of HD. This probably
explains why most patients appear to have had signs of bowel inflam-
mation on the resected specimens but only six children was diagnosed
as the preoperative HAEC in hospital.
This was a retrospective study with a limited number of patients. The
absence of patients with trisomy 21 in the study, which made it impossi-
ble for us to investigate the association between trisomy 21 and postoper-
ative HAEC, may be another limitation. Additionally, we excluded 7 TCA
patients, which might cause some loss of information, given the signifi-
cant influence of ileocecal valve on postoperative recovery and the diffi-
culty of determining the transitional zone in TCA patients. Therefore, the
results in the current study should be interpreted with care.
However, the results of this study suggest that pediatric surgeons
should counsel the parents of children with HD about the potential
risk of worse outcomes related to an inflamed proximal segment of
the resected colon. We should also be alert to children with a high
HAEC pathology grade in the transitional segment or in the sum of the
overall three segments, because of their increased risk of subsequently
developing postoperative HAEC. Furthermore, we suggest that resected
bowel tissue from children with HD should be routinely examined for
HAEC pathology grade. Further prospective comparative studies are re-
quired with a longer-term follow-up to confirm these results.
1781
References
[1] Elhalaby EA, Teitelbaum DH, Coran AG, et al. Enterocolitis associated with
Hirschsprung's disease: a clinical histopathological correlative study. J Pediatr Surg
1995;30:1023–7.
[2] Dasgupta R, Langer JC. Transanal pull-through for Hirschsprung disease. Semin
Pediatr Surg 2005;14:64–71.
[3] Guidelines–Rome III diagnostic criteria for functional gastrointestinal disorders, J
Gastrointestin Liver Dis 2006;15:307–12.
[4] Pastor AC, Osman F, Teitelbaum DH, et al. Development of a standardized definition
for Hirschsprung's-associated enterocolitis: a Delphi analysis. J Pediatr Surg 2009;
44:251–6.
[5] Austin KM. The pathogenesis of Hirschsprung’ s disease-associated enterocolitis.
Semin Pediatr Surg 2012;21:319–27.
[6] Haricharan RN, Seo JM, Kelly DR, et al. Older age at diagnosis of Hirschsprung disease
decreases risk of postoperative enterocolitis, but resection of additional ganglionat-
ed bowel does not. J Pediatr Surg 2008;43:1115–23.
[7] Schulten D, Holschneider AM, Meier-Ruge W. Proximal segment histology of
resected bowel in Hirschsprung's disease predicts postoperative bowel function.
Eur J Pediatr Surg 2000;10:378–81.
[8] White FV, Langer JC. Circumferential distribution of ganglion cells in the transition
zone of children with Hirschsprung disease. Pediatr Dev Pathol 2000;3:216–22.
[9] Farrugia MK, Alexander N, Clarke S, et al. Does transitional zone pull-through in
Hirschsprung's disease imply a poor prognosis? J Pediatr Surg 2003;38:1766–9.
[10] Estevao-Costa J, Fragoso AC, Campos M, et al. An approach to minimize postopera-
tive enterocolitis in Hirschsprung's disease. J Pediatr Surg 2006;41:1704–7.
[11] Antao B, Roberts J. Laparoscopic-assisted transanal endorectal coloanal anastomosis
for Hirschsprung's disease. J Laparoendosc Adv Surg Tech A 2005;15:75–9.
[12] Kleinhaus S, Boley SJ, Sheran M, et al. Hirschsprung's disease – a survey of the mem-
bers of the surgical section of the American Academy of Pediatrics. J Pediatr Surg
1979;14:588–97.
[13] Teitelbaum DH, Qualman SJ, Caniano DA. Hirschsprung's disease. Identification of
risk factors for enterocolitis. Ann Surg 1988;207:240–4.
[14] Quinn FM, Surana R, Puri P. The influence of trisomy 21 on outcome in children with
Hirschsprung's disease. J Pediatr Surg 1994;29:781–3.
[15] Menezes M, Corbally M, Puri P. Long-term results of bowel function after treatment
for Hirschsprung's disease: a 29-year review. Pediatr Surg Int 2006;22:987–90.
[16] Morabito A, Lall A, Gull S, et al. The impact of Down's syndrome on the immediate
and longterm outcomes of children with Hirschsprung's disease. Pediatr Surg Int
2006;22:179–81.
[17] Passarge E. The genetics of Hirschsprung's disease. Evidence for heterogeneous eti-
ology and a study of sixty-three families. N Engl J Med 1967;276:138–43.
[18] Makitie O, Heikkinen M, Kaitila I, et al. Hirschsprung's disease in cartilage-hair hypo-
plasia has poor prognosis. J Pediatr Surg 2002;37:1585–8.
[19] Rescorla FJ, Morrison AM, Engles D, et al. Hirschsprung's disease. Evaluation of mor-
tality and long-term function in 260 cases. Arch Surg 1992;127:934–41 [941–942].
[20] Moore SW, Albertyn R, Cywes S. Clinical outcome and long-term quality of life after
surgical correction of Hirschsprung's disease. J Pediatr Surg 1996;31:1496–502.
[21] Teitelbaum DH, Cilley RE, Sherman NJ, et al. A decade of experience with the primary
pullthrough for hirschsprung disease in the newborn period: a multicenter analysis
of outcomes. Ann Surg 2000;232:372–80.
[22] Murphy F, Puri P. New insights into the pathogenesis of Hirschsprung's associated
enterocolitis. Pediatr Surg Int 2005;21:773–9.
[23] Hanimann B, Inderbitzin D, Briner J, et al. Clinical relevance of Hirschsprung-
associated neuronal intestinal dysplasia (HANID). Eur J Pediatr Surg 1992;2:147–9.
[24] Marty TL, Matlak ME, Hendrickson M, et al. Unexpected death from enterocolitis
after surgery for Hirschsprung's disease. Pediatrics 1995;96:118–21.
[25] Vieten D, Spicer R. Enterocolitis complicating Hirschsprung's disease. Semin Pediatr
Surg 2004;13:263–72.
[26] El-Sawaf M, Siddiqui S, Mahmoud M, et al. Probiotic prophylaxis after pullthrough
for Hirschsprung disease to reduce incidence of enterocolitis: a prospective, ran-
domized, double-blind, placebo-controlled, multicenter trial. J Pediatr Surg 2013;
48:111–7.