One-stage Soave pull-through for Hirschsprung's disease: A comparison of the transanal and open approaches

Purpose: The authors reviewed their experience using the transanal Soave te chnique, to determine (1) if it offers any advantages over the standard ope n approach and (2) whether routine laparoscopic visualization is necessary. Methods: The case reports of 37 consecutive children less than 3 years old undergoing Soave pull-through were reviewed. Patients were excluded from an alysis if they had total colon disease or had a previous colostomy. The pat ients were divided into 3 groups: open Soave (OS, n = 13), transanal Soave with routine laparoscopic visualization (LVS, n = 9), and transanal Soave w ith selective laparoscopy or minilaparotomy (TAS, n = 151. Cost was calcula ted based on hospital stay, operating room time, and use of laparoscopic eq uipment. Results: In the TAS group, suspicion of a longer segment led to the selecti ve use of laparoscopy with or without biopsy in 2 children, and the use of a small umbilical incision for mobilization of the splenic flexure in 2. Th ere were no differences among groups with respect to age, weight, gender, t ransition zone, operating time, blood loss, intraoperative complications, e nterocolitis, or stricture or cuff narrowing. Hospital stay was significant ly longer in the OS group (median, 7 days; range, 3 to 47) than the LVS (me dian, 1; range 1 to 6) or TAS (median, 1, range, 1 to 3) groups. Cost tin t housands of dollars) was also higher in the OS group (median, 6.9; range, 3 .9-25.7) than the LVS (median, 3.9; range, 3.6 to 6.4) or TAS (median, 3.4; range, 2.2 to 9.4) groups. Repeat surgery was necessary for 4 OS patients: 2 adhesive small bowel obstructions (1 of whom died), 1 twisted pull-throu gh, and 1 recurrent aganglionosis. Three TAS patients required repeat surge ry: 1 twisted purl-through, 1 anastomotic leak, and 2 cuff narrowing. Conclusions: These data suggest that the transanal pull-through is associat ed with a significantly shorter hospital stay and lower cost than the open approach, without an increased risk of complications. Because there is no i ntraabdominal dissection, there probably is a lower incidence of adhesive b owel obstruction. Routine laparoscopic visualization or minilaparotomy is n ot necessary but should be used in children who are at higher risk for long segment disease. Copyright (C) 2000 by W.B. Saunders Company.