Dynamic graciloplasty versus implant of artificial sphincter for continent perineal colostomy after Miles' procedure: Technique and early results

Abdominoperineal resection (APR) is still the standard surgical treatment of anorectal cancers close to the dentate line. Unfortunately, a permanent iliac colostomy is a severe limitation of the quality of life. Attempts to construct a continent perineal colostomy after anorectal excision have been made over the last 15 years with uncertain benefits. We report on our early results of two different procedures consisting of a laparoscopic approach to abdominoperineal rectal excision, fashioning a perineal colostomy with dynamic graciloplasty or implant of an artificial sphincter. Between 2000 and 2004, a total of six patients underwent laparoscopic abdominoperineal resection or reversal of Miles' procedure and construction of perineal colostomy with dynamic graciloplasty (three cases) or implant of an artificial bowel sphincter (three cases). A diverting loop ileostomy was constructed in all patients to prevent contamination. Data concerning the perioperative management, postoperative morbidity and mortality and function after total anorectal reconstruction at the time of discharge, at postoperative month 1 and after ileostomy closure were collected and evaluated in a prospective non-randomised fashion. No early postoperative complications occurred in both groups. No late complication occurred in the dynamic graciloplasty group, whilst one patient of the artificial sphincter group had an ulceration of the tubing and the control pump through the suprapubic skin and the labium skin respectively on postoperative day 35. Another patient in this group, with an erosion of the transposed colon wall, died of myocardial infarction on postoperative day 75 after removal of the prosthesis. Postoperative stay after artificial sphincter implant and dynamic graciloplasty ranged from 12 to 27 days and 16 to 24 days, respectively. The loop ileostomy was closed at postoperative month 3 in all remaining patients except for one in the dynamic graciloplasty group, who died one day before hospitalization for ostomy closure because of an accidental, not disease/operation related reason. Follow-up of patients of the dynamic graciloplasty and artificial sphincter groups ranged from 3 to 24 months and 2.5 to 9.5 months, respectively. Patients in the dynamic graciloplasty group had no complications and follow-up showed satisfactory continence (SF36 form). All patients in the artificial sphincter group had late local complications with erosion of the prosthesis through the wall, its co