Preoperative enterocolitis is associated with poorer long-term bowel function after Soave-Boley endorectal pull-through for Hirschsprung's disease☆

Materials and methods 

Between 1983 and 1998, 63 children (56 boys and 7 girls) underwent Soave-Boley ERPT for HD. At the time of review in 2001, all children were at least 3 years of age, which allowed meaningful assessment of bowel function.

The study consisted of (1) a review of case notes and (2) a detailed questionnaire on aspects of the child's bowel function completed by parents.

The case note review obtained information regarding presentation, age at diagnosis, associated conditions and family history of HD, extent of aganglionosis, age at ERPT, primary or 2-stage ERPT, early and delayed postoperative complications, and length of follow-up.

The questionnaire used a semiquantitative scoring system for 6 aspects of bowel function at the time of the study (Table 1).

Table 1. Scoring system asking parents to grade six aspects of their child's bowel function

	Ability to hold back defecation
	Always	3
	Weekly problems	2
	Problems < 1/week	1
	No voluntary controls	0
	Feels/reports urge to defecate
	Always	3
	Most of the time	2
	Uncertain	1
	Absent	0
	Frequency of defecation
	Every other day to twice a day	3
	More often	2
	Less often	1
	Soiling
	Never	3
	Staining < 1/week, no underwear change required	2
	Frequent staining, underwear often changed	1
	Daily soiling, protective aids	0
	Accidents
	Never	3
	Fewer than 1/week	2
	Weekly accidents, often requires protective aids	1
	Daily, requires protective aids during day & night	0
	Social problems
	No social problems	3
	Sometimes (foul odours)	2
	Problems causing restriction in social life	1
	Severe social and/or psychological problems	0

Results 

The demographic data of the 63 children who underwent Soave-Boley ERPT for HD were the following: 56 boys and 7 girls, mean age at the time of the study was 8.8 years (range, 3.0 to 17.4 years), and mean birth weight was 3,331 g (range, 870 g to 4,330 g). Forty-five (71%) patients presented within 72 hours of birth. Seven (11%) presented with features of enterocolitis (fever, leucocytosis, bleeding per rectum, pneumoperitoneum or intramural gas on plain abdominal radiograph). Mean age at ERPT was 135.5 days, median 50 days (range, 12 to 1,416 days). Primary ERPT was performed in 52 patients (83%), and 2-stage ERPT was done in 11 patients (17%). Early ERPT (at <50 days of age) was performed in 29 patients and late ERPT (>50 days of age) in 34 patients. The extent of the aganglionic segment was rectosigmoid (38, 60%), long segment (21, 33%), and total colonic (4, 6%). There was a family history of HD in 2 cases. Major associated abnormalities were: Down's syndrome (n = 7), cardiac lesions (n = 3), congenital hypoventilation syndrome (n = 1), congenital amputation of forearm (n = 1), and horse-shoe kidney (n = 1). Postoperative complications are shown in Table 2.

Table 2. List of postoperative complications after Soave-Boley ERPT

	Early
	Anastomotic stricture	14
	Enterocolitis	6
	Wound problems	7
	Infection	3
	Dehiscence	4
	Mucosal prolapse	2
	Pelvic sepsis	2
	Colon retraction	2
	Intestinal obstruction	1
	Late
	Enterocolitis	8
	Adhesive intestinal obstruction	3
	Incisional hernia	1
	Short bowel syndrome	1
	Cerebral palsy	1

Table 3. List of secondary operative procedures undertaken in patients as a result of complications after Soave-Boley ERPT

	Rectal biopsies for constipation	8
	Wound resutured (dehiscence)	4
	Colostomy for constipation and soiling	3
	Laparotomy for adhesive obstruction	2
	Repeat pullthrough (anastomotic stricture, colon retraction)	2
	Antegrade continent enema (ACE) for constipation	2
	Anal mucosectomy (mucosal prolapse)	2
	Laparotomy and biopsies for constipation	1

Of the 63 patients operated on, 8 were excluded from review (3 deaths, 3 with permanent colostomy, and 2 with antegrade continent enema procedures). Questionnaires were sent to parents of 55 children and 50 (91%) were returned.

TFS was “good” in 25 (45%), “fair” in 18 (33%), and “poor” in 12 (22%). Fewer than half of the patients, therefore, had what parents and clinicians would perceive as “normal” bowel function.

TFS was analysed with regard to subgroups of patients. Figure 1 shows the mean functional scores comparing boys and girls, early or late ERPT, levels of aganglionic segment, primary or 2-stage ERPT, and presence or absence of anastomotic stricture, Down's syndrome, and preoperative enterocolitis.

View Large Image Download to PowerPoint

Fig. 1. Bar chart shows comparison of mean functional scores (x axis) between various subgroups within the study population (y axis). (A) male/female, (B) rectosigmoid/long segment/total colonic, (C) early (<50 days)/late (>50 days) ERPT, (D) postoperative anastomotic stricture/no stricture, (E) primary/2-stage ERPT, (F) Down's syndrome/others, (G) preoperative HAEC/others.

Mean TFS was worse for those patients who had preoperative enterocolitis (TFS, 8.0) than for those who did not (TFS, 12.7). This difference was statistically significant (P < .01). There was, however, no significant difference in means for age at presentation, age at diagnosis, length of aganglionic segment, or age at ERPT between the 2 groups.

Discussion 

Analysis of long-term anorectal function in children after surgical management of HD or anorectal malformation by pull-through procedure is difficult because of inability to define normal bowel function; lack of precise definitions and objective measurements to assess “constipation” and “faecal continence”; usage of different scoring and grading systems by different investigators; usage of different study methods (eg, questionnaire/clinical examination/psychological profile/manometric studies/colonic transit times) to assess bowel function; grouping together of patients with different surgical procedures (eg, Duhamel, Swenson, ERPT); the influence of varying social, cultural, dietary, and economic conditions on bowel function.

The scoring system used in this series was devised to avoid the need for direct patient review or repeat investigation and is a composite of previous methods based principally on that of Rintala and Lindahl.10

However, for HD, some conclusions can be made: (1) children with HD can suffer considerable long-term morbidity,4, 5 (2) HAEC is a major factor in morbidity and mortality,6, 7 (3) bowel function is related to length of aganglionic segment (patients with total colonic aganglionosis [TCA] do less well than those with rectosigmoid disease),2, 5 (4) children with Down's syndrome have a poorer outcome,11 (5) evidence regarding improvement of bowel function with time is conflicting because there is an absence of longitudinal studies and possible underreporting by parents because of adaptation to the difficulties associated with the condition.

In our series, absence of a pattern of increasing severity of bowel malfunction with increasing extent of aganglionosis is probably because of the small number of patients with TCA (4 patients; 6.3%). Similarly, although patients with Down's syndrome had a lower mean TFS, the relatively small number precluded statistical analysis.

The reasons for poorer outcome in patients with preoperative HAEC are not clear, although a high mortality rate for this group has been reported previously. A survey of 1,196 patients by the Surgical Section of the American Academy of Pediatrics (Kleinhaus et al12) showed an incidence of prediagnosis HAEC of 15% with a 30% mortality rate within this group. Sarioglu et al7 reported a 26% incidence of preoperative HAEC and 7.6% mortality rate.

Questions remain as to whether (1) variations in factors between HD patients (eg, properties of gut mucosa6, 13) predispose both to HAEC and to poorer long-term outcome and whether (2) preoperative HAEC compromises outcome from ERPT surgery.

In clinical practice, suspicion of HAEC (signs of abdominal distension or tenderness, bleeding per rectum or pneumoperitoneum) should lead to prompt resuscitative and supportive measures in an effort to limit the morbidity and mortality of this potentially fulminant condition.