Rectal Irrigation’s for the Prevention of Postoperative

Enterocolitis in Hirschsprung's Disease

By Tern L. Marty, Takahiko Seo, John J. Sullivan, Michael E. Matiak, Richard E. Black, and Dale Gi. Johnson

Salt Lake City, Utah

The purpose of this clinical trial was to examine the role of rectal Irrigation’s in the prevention of postoperative Enterocolitis in children with Hirschsprung’s disease. Over the past 22 years 177 children had surgical treatment for Hirschsprung’s disease at a single pediatric hospital. Five chil-dren have died of other causes. of the remaIning 172 pa-Patient., follow-up clinical information was obtained from 135 (78% l In 1gW, all of the children undergoing surgical recon-struction for Hlrschsprung'e disease were placed on routine postoperative rectal Irrigation’s with normal saline. The par-parents were instructed in the irrigation technique before leav-ing the hospital. Irrigation’s were started 1 to 2 weeks postoperatively and were performed two times a day for 3 months. then once a day for an additional 3 months. There were no complications from the irrigations themaelvec. A previous report from the authors' institution has documented five sudden deaths over the past 22 years from fulminate postoperative enterocolitle. These five children all had an unremarkable postoperative recovery, then developed a rapidly progressive diarrheal illness. Ml of these deaths occurred before the Institution of routine postoperative irrigation’s. of the 135 patients this review, 40 had the postoperative rectal Irrigation’s. The remaining .5 children serve as historical controls. Data analysis showed that 34 of the 95 children in the nonirrigation cohort developed postoperative enterocolitle compared with 3 of 40 in the rectal Irrigation cohort: P ~ .001 using Reher's exact tact. In the authors' experience, routine postoperative rectal irrigation’s have signlficanity decreased the Incidence and severity of Enterocolitis In children after surgical correction of Hirschsprung’s disease.

All who treat this disease would surely agree that when the diagnosis is suspected it should be confirmed. and operation. either colostomy or definitive, should be performed before the baby is allowed home lest the baby succumb to a crisis of enterocolitis.

Enterocolitis is a well-known and some-times fatal complication of Hirschsprung's disease. The original description of this entity is credited to Bill and Chapman in 1962.2 In their initial series they report a mortality rate of 33% in infants with Enterocolitis occurring before any operation.2 In early as well as more recent literature it is well recognized that Enterocolitis can occur before or after definitive operative treatment. The incidence of Enterocolitis before the diagnosis of Hirschsprung's disease is established as between 15% and 50%with a mortality rate between 20% and 50%. After surgical reconstruction the incidence of Enterocolitis is between 2% and 33%, with a mortality rate varying between 0% and 30%.6.8.l In addition, there are numerous case reports in the literature of children succumbing to sudden death from Enterocolitis after an unremarkable surgical reconstruction and postoperative course.3.t~t3 The purpose of this clinical trial is to examine the role of rectal irrigation’s in the prevention of postoperative Enterocolitis.

MATERIALS AND METHODS

Over the past 22 years, 177 children have undergone surgical treatment for Hirschsprung's disease at a single pediatric tertiary referral center. Five children have died from causes unrelated to Hirschsprung’s disease. Of the remaining 172 patients. follow-up clinical information was obtained from 135 (78%). Mean length of follow-up is 7.9 years. with a range of 3 months to 21.5 years.

In 1989, all of the children undergoing surgical reconstruction for Hirschsprung's disease w6e placed on routine postoperative rectal irrigation’s with normal saline. The parents were instructed in the irrigation technique before leaving the hospital. Irrigation’s were started 1 to 2 weeks postoperatively and were performed two times a day for 3 months, then once a day for an additional 3 months. The volume of the rectal irrigation’s is at least 10 mil kg of body weight: some children require a larger volume for effective evacuation.

A separate report from our institution has documented five sudden deaths over the past 22 years from fulminate postoperative Enterocolitis. These five children all had an unremarkable postoperative recovery. then a rapidly progressive diarrhea illness developed. All of these deaths occurred before the institution of routine postoperative irrigation’s. Of the 135 patients in this review. 40 had the postoperative rectal irrigation’s. The remaining 95 children serve as historical controls.

RESULTS

Data analysis showed that 34 of the 95 children in the non-irrigation cohort had postoperative Enterocolitis, compared with 3 of 40 in the rectal irrigation cohort; < .001 using Fisher's exact test. These results are shown in Table 1.

 

'In these children, Enterocolitis developed during the rectal Irrigation’s.

There was no significant difference in the percentage of children in each group with total colonic aganglionosis. There have been no complications from the irrigation’s themselves. In addition, there have been no deaths from postoperative Enterocolitis since the introduction of the postoperative irrigation’s.

DISCUSSION

From the results stated above, it appears that the routine postoperative irrigation’s have decreased the incidence and severity of Enterocolitis. It is also interesting that two children had Enterocolitis before their prophylactic irrigation’s were instituted, and two other children had Enterocolitis after their irrigation’s had been discontinued. In all four of these cases, with prompt initiation of routine daily irrigation’s, no further Enterocolitis developed.

Of the three children in whom symptoms of Enterocolitis-colitis developed during the irrigation’s, two had residual bowel stenosis, and the third child developed diarrhea after antibiotics for otitis media. These three children will be discussed individually in further detail. The first patient has congenital anal stenosis, a small pelvis, and an anteriorly placed anorectal canal. This young child has suffered three bouts of postoperative Enterocolitis despite the daily irrigation’s. It is unclear whether partial obstruction caused by abnormal-mal pelvic anatomy is a contributing factor to her recurrent Enterocolitis.

The second child, in whom the prophylactic irrigation’s failed, had done well for approximately 8 months following modified Duhamel reconstruction and had been off of the irrigation’s completely for a number of months without any evidence of Enterocolitis. The irrigation’s were then restarted secondary to functional obstruction and inadequate evacuation. He had been on twice daily irrigation’s for 2.5 months when he developed Hirschsprung's Enterocolitis and recurrent fecal impaction. The patient required a week of hospitalization, intravenous fluids, nasogastric decompression, manual disimpaction, and rectal irrigation’s every 6 hours. Barium enema in this child showed a narrowed segment near the anastomosis, which may be a contributing factor to the recurrent impaction and subsequent Enterocolitis.

The third child, who suffered abdominal distention, explosive diarrhea, and vomiting despite twice daily irrigation’s, had received oral Bicillin (Wyeth Laboratories, Philadelphia, PA) for otitiS media shortly before her acute gastrointestinal illness. She responded to intravenous hydration, nasogastric de-compression, and frequent (every 4 to 6 hours) rectal irrigation’s. In each of these last two cases, empiric antibiotics against CIostridium difficile were started when the diagnosis of Enterocolitis was suspected. Stool cultures, including Rotazyme (Cambridge Bio-tech, Worchester, MA) and C difficile, were negative in both children, and the antibiotics were subsequently discontinued. The role of oral antibiotics (ic, Bicillin)~n precipitating Enterocolitis is debated in the literature; some series support an etiologic role ,3.IA whereas others do not believe there is an associa-tion.5~15

CONCLUSION

The importance of rectal decompression in the treatment of Enterocolitis has been well known for many years.~'6~'7 This is the first report that we are aware of that used routine rectal irrigation’s to pre-vent postoperative Enterocolitis. In our experience, routine postoperative rectal irrigation’s have significantly decreased the incidence and severity of Enterocolitis-colitis in children after surgical correction of Hirschsprung’s disease.

REFERENCE

1. Nixon HH: Hirschsprung’s disease. Arch Dis Child 39:109-115 1964
2. Bill AH Chapman ND: The Enterocolitis of Hirschsprung’s disease its natural history and treatment. Am J Surg 103:70-74 1962
3. Bagwell CE, Langham MR, Mahaffey SM, et al: Pseudomembrananous Colitis following resection for Hirschsprung’s disease J Pediatric surgery 27:1261-1264, 1992
4. Boley SJ, Dinari G Cohen MI, Hirschsprung’s Disease in the newborn, Clin Perinatol 5:45-60 1978
5. Brearly S Armstrong GR, Nairin R er al : Pseudomembranous Colitis: a Lethal Complication of Hirschsprung’s disease unrelated to antibitioic usage, J pediatric surg 22;257-259 1987
6. Kleinhaus S Boley SJ, Sieber WK, et al: Hirschsprung’s disease a survey of the members of the surgical section of the American academy of pediatrics j pediatric surg 14;588-597 1979
7. Frank JD, Nixon HH,: causes of death in Hirschsprung’s disease, analysis and conclusions for therapy, in bill AH ed progress in pediatric surgery Baltimore md urband schwarzenberg 1979, pp 199-205

From the University of Utah. Primary Children 's Medical Center. Salt Lake City. UT

Presented at the 27th Annual Meeting of the Pacific Association of Pediatric Surgeons. Kagoshima. Japan. May 23-27, 1994.

Address Reprint requests to Dale G. Johnson, MD. Professor of Surgery. Professor of Pediatrics, University of Ut4~ Surgeon in Chief. Primary Children s Medical Center, 100 North Medical Dr, Suite 2600.

Salt Lake City. UT84113-1100.

Copyright t 1995 ly WB. Saunders Company 0022.340119513005.0003503.00IO