The standard treatment for Hirschsprung’s disease has been colostomy, followed by one of several pullthrough procedures 1.
Colostomy represents a significant source of morbidity and even mortality; these complications related to colostomy have been well documented in many series 2.
Nowadays, one-stage definitive repair has become the standard treatment. One-stage procedures are performed by laparotomy, minimal invasive techniques, and transanal pullthrough 3,4. Despite this, the staged operation still has a place 5.
The endorectal pullthrough for Hirschsprung’s disease was described by Soave in 1964 6. Since then, different modifications have been made to the procedure.
Here, we describe our modification of the endorectal pullthrough to perform it through the colostomy site.
Patients and methods
This study included 13 patients, four girls and nine boys. Their age ranged from 9 months to 7 years.
They presented to us with a leveling colostomy with a diagnosis of Hirschsprung’s disease. The colostomy was carried out in the neonatal period because of neonatal intestinal obstruction in three patients (one female and two males), recurrent attacks of enterocolitis in two patients, and inability of the patient to withstand major surgery in three patients. Four patients presented to us without a clear history of the cause for the colostomy. One patient aged 7 years presented with intestinal obstruction with a history suggestive of Hirschsprung’s disease. Plain radiograph indicated sigmoid volvulus. A trial of reduction by enema and a rectal tube failed. Exploration indicated sigmoid volvulus with the typical gross picture of Hirschsprung’s disease. The volvulus was untwisted, a biopsy from the narrow segment was taken, and a leveling colostomy was constructed.
The duration from the time of the colostomy till the pullthrough ranged from 3 to 8 months.
The operation was performed as a usual transabdominal pullthrough operation. The abdomen was entered through a skin incision all around and besides the mucocutaneous junction of the colostomy. The colostomy was dissected from the abdominal wall till the peritoneum was entered.
The colon was repaired temporarily and proximal to the colostomy site, the colon was mobilized by incising the peritoneum on both sides until a suitable length of the colon was mobilized. This step requires a good retraction of the abdominal wall. Distal to the colostomy site, the colon was mobilized by double ligation of its mesocolon.
The operation was completed as a modified endorectal Soave pullthrough. A circumferential seromuscular incision of the rectum was carried out and the rectal mucosa was dissected till the anorectal junction was reached. The mobilized colon was pulled through the seromuscular cuff and end-to-end rectoanal anastomosis was performed.
An intra-abdominal tube drain was placed and the defect in the abdominal wall was repaired.
In one patient, the colon proximal to the colostomy site was still considerably dilated because of its prolapse and extramobilization of the colon up to the cecum was required; this resulted in a change of the site of the cecum to the subhepatic position. Appendicectomy was required in this patient (Figs 1–6).
Mobilization of the colon proximal to the colostomy site could be difficult but with traction of the colon downward and medially and retraction of the abdominal wall upwards, the colon could be mobilized.
In two cases, endorectal dissection could not be completed abdominally as the colostomies were constructed in the lumbar region. We decided on transanal dissection of the remaining part of the rectum.
This study was approved by the ethical committee of the institution.
This study included 13 patients. The operation was completed as described in all patients. The time from colostomy to pullthrough ranged from 3 to 7 months (median 4.7 months).
The operative time ranged from 95 to 140 min (median 113 min).
The patient was kept on intravenous fluids for 48 h, systemic antibiotics, and metronidazol for 5 days and the drain was extracted after 24 h. The wound dressing was changed after 72 h and stitches were removed after 8 days.
All the patients passed stool within 24–48 h. Stool output ranged from two to six stools per day.
Optimal wound healing occurred in all patients without wound complications.
Postoperative dilatation started 2 weeks postoperatively to prevent stricture at the anastomotic site. Follow-up ranged from 5 to 13 months (median 5.7 months) (Table 1).
Early postoperative complications
Postoperative perineal excoriation occurred in four patients and this was treated with a zinc-based cream. Postoperative high-grade fever occurred in two patients without a specific cause and responded well to cold fomentations and antipyretics.
A urinary tract infection occurred in one patient on the fourth postoperative day, as indicated by urine analysis and culture, and responded well to proper treatment.
A patient with a history of recurrent attacks of preoperative enterocolitis developed mild enterocolitis 2 weeks after the operation; it occurred once and responded to conservative treatment using antibiotics, intravenous fluids, and rectal irrigation. There were no early anastomosis-related complications.
Late postoperative complications
Adhesive intestinal obstruction occurred in one patient 3 months after the operation and was treated conservatively.
There was recurrence of symptoms in two patients because of stricture at the anastomotic site: one responded to repeated dilatation and the other required internal sphincterotomy (Table 2).
Although a primary pullthrough is the preferred approach for use in children with Hirschsprung’s disease, some children benefit from an initial leveling colostomy, which is indicated in infants who have severe enterocolitis or a markedly dilated proximal colon that might preclude the performance of a primary pullthrough 5. Colostomy decompresses the bowel, and this decompression elevates the nutritional status of patients, decreases the risk of anastomotic leak and pelvic abscess, and facilitates smooth healing 7. Results of studies comparing one-stage with two-stage pullthrough have reported higher complication rates in the two-stage operation, which may be attributable to the presence of a stoma 7,8.
Other indications for staged surgery using a colostomy include patients with long segment, total colonic aganglionosis, complicated Hirschsprung’s disease with perforation, toxic megacolon, questionable pathology, or unavailability of frozen section 8,9.
In our institution, the standard treatment for Hirschsprung’s disease is a one-stage pullthrough operation either through a transabdominal or a transanal approach. Colostomy is performed only in indicated cases.
Three of our patients underwent a colostomy in the neonatal period because of neonatal intestinal obstruction, two because of recurrent enterocolitis, three because of poor general condition, one because of sigmoid volvulus, and three without a definite history.
The duration from performing a colostomy till the definite transcolostomy-site pullthrough varied considerably according to the cause of the colostomy. The operation was performed when the patient was in a fair general condition and the colon was decompressed enough for a safe pullthrough. This period ranged from 3 to 8 months.
The idea to do the pullthrough transcolostomy site is that it is a less invasive technique than the usual approach of Soave pullthrough, which uses a hockey stick incision including the colostomy site. The transcolostomy-site incision is smaller, it is not muscle cutting, and leads to much lower morbidity. None of our patients had developed wound-related complications.
Lesser time was taken to reach the peritoneal cavity, but the total time of the entire procedure was comparable with the standard time of the pullthrough operation. Mobilization of the colon proximal to the colostomy site takes longer than usual with the hockey stick incision. This step must be carried out carefully and it is aided by downward traction on both the colon at the colostomy site and the transverse colon. This facilitates rapid mobilization of the colon, even the difficult splenic flexure. In two cases with prolapsed colostomy, the mobilization of the colon proximal to the colostomy site was easy, but we found that this prolapsed part of the colon was not decompressed enough and it was resected.
Transanal mucosectomy has been used for many years in endorectal pullthrough for Hirschsprung’s disease. It was used by Georgeson et al.10, who described the use of laparoscopy for the abdominal stage. Saltzman et al. 11 used it with laparotomy for the abdominal stage.
In two of our patients, endorectal dissection was difficult abdominally; therefore, we performed it transanally. A combination of transcolostomy-site and transabdominal pullthrough is a feasible option.
Great care must be exercised to ensure that the colon is not twisted as it is pulled through the muscular cuff. Twisted pullthrough could result in intestinal obstruction or may be a cause for redo surgery 9,12. This occurred inone of our patients; fortunately, we noted this intraoperative and untwisted. Thus, we recommend orientation of the colon after pullthrough and before the coloanal anastomosis to exclude twist.
Postoperative perineal excoriation is common and can be treated with a zinc-based cream.
Dilatation of the anastomosis is necessary for several months after the Soave operation to prevent stricture formation 13. We start postoperative dilatation as a routine in all patients 15 days postoperatively and for variable periods. Two patients in our series developed stricture as they were lost in the follow-up period. They presented with symptoms of recurrence. Physical examination indicated stricture, and immediate dilatation was started. One patient responded well to repeated dilatation and in the other, posterior myectomy was required. We recommend routine dilatation after pullthrough operations.
Postoperative enterocolitis is a serious complication and it is a common cause of Hirschsprung’s disease-related mortality. Enterocolitis occurs in 17–50% of infants with Hirschsprung’s disease and is most commonly caused by intestinal obstruction and residual aganglionic bowel 14. Early symptoms of enterocolitis in patients with Hirschsprung’s disease include abdominal distention; foul-smelling, watery diarrhea; lethargy; and poor feeding. Treatment with rectal irrigation several times per day and antibiotics is usually effective. Oral metronidazole can be used with rectal irrigation in patients with milder disease. More serious disease should be treated intravenously with broad-spectrum antibiotics and rectal irrigation 14.
One of our patients who had a preoperative history of enterocolitis developed a mild attack of postoperative enterocolitis and responded well to conservative treatment.
The overall incidence of complications in our series is comparable with other Soave pullthrough series. Larger series and long-term follow-up are required to document our procedure.
The reported mortality in the literature is variable. An unacceptably high mortality (6–35%) has been reported 15. Sharma et al. 9, reported 112 cases without deaths; they attributed this to a proper preoperative preparation and the use of sound surgical principles.
There were no postoperative deaths in our series possibly because of the same reasons.
Our technique has the following advantages:
- It is associated with less pain and a shorter hospitalization than the classic endorectal pullthrough.
- Wound complications are rare.
- The cosmetic result is better than that with the classic Soave operation.
- It has no specific technique-related complications.
- Long-term outcome and functional results are good.
- To our knowledge, this approach has not been described before.
Conflicts of interest
There are no conflicts of interest.
1. Kleinhaus S, Boley SJ, Sheran M, Sieber WK. Hirschsprung’s disease. A survey of the members of the surgical section of the American academy of pediatrics. J Pediatr Surg. 1979;14:588–597
2. Patwardhan N, Kiely EM, Drake DP, Spitz L, Pierro A. Colostomy for anorectal anomalies: high incidence of complications. J Pediatr Surg. 2001;36:795–798
3. Cilley RE, Statter MB, Hirschl RB, Coran AG. Definitive treatment of Hirschsprung’s disease in the newborn with a one- stage procedure. Surgery. 1994;115:551–556
4. De La Torre-Mondragón L, Ortega-Salgado JA. Transanal endorectal pull-through for Hirschsprung’s disease. J Pediatr Surg. 1998;33:1283–1286
5. Teltelbaum DH, Coran AGO’Neill JA Jr, Fonkasrud EW, Coran AG, Grosfeld JL Hirschsprung’s disease and related neuromuscular disorders of the intestine. 2006;26th ed. Philadelphia Mosby-Elservier:1514–1559
6. Soave F. Hirschsprung’s disease: a new surgical technique. Arch Dis Child. 1964;39:116–124
7. Ghoroubi J, Imanzadeh F, Sadeghian N, Sayyari4 AA, Peyvasteh M, Javaherizadeh H, Aghaie E. Comparison between one and multiple stages surgery in the treatment of Hirschsprung’s disease. Ann Pediatr Surg. 2009;5:172–176
8. Langer JC, Fitzgerald PG, Winthrop AL, Srinathan SK, Foglia RP, Skinner MA, et al. One-stage versus two-stage Soave pull-through for Hirschsprung’s disease in the first year of life. J Pediatr Surg. 1996;31:33–37
9. Sharma S, Gupta DK. Hirschsprung’s disease presenting beyond infancy: surgical options and postoperative outcome. Pediat Surg Int. 2012;28:5–8
10. Georgeson KE, Fuenfer MM, Hardin WD, Holcomb G. Primary laparoscopic pull-through for Hirschsprung’s disease in infants and children. J Pediatr Surg. 1995;30:1017–1022
11. Saltzman DA, Telander MJ, Brennom WS, Telander RL. Transanal mucosectomy: a modification of the Soave procedure for Hirschsprung’s disease. J Pediatr Surg. 1996;31:1272–1275
12. Langer JC, Durrant AC, De La Torre L, Teitelbaum DH, Minkes RK, Caty MG, et al. One-stage transanal Soave pullthrough for Hirschsprung disease: a multicenter experience with 141 children. Ann Surg. 2003;238:569–576
13. Saldi W, Rasheed K, Al Mohaidly M, Kfoury H, Tariq M, Al Rawaf A. Management of Hirschsprung’s disease: a comparison of Soave’s and Duhamel’s pull-through methods. Pediat Surg Int. 2004;20:590–593
14. Coran AG, Teitelbaum DH. Recent advances in the management of Hirschsprung’s disease. Am J Surg. 2000;180:382–387
15. Bandré E, Kaboré RAF, Ouedraogo I, Soré O, Tapsoba T, Bambara C, Wandaogo A. Hirschsprung’s disease: management problem in a developing country. Afr J Paediatr Surg. 2010;7:166–168