Objective: To compare perineal healing with or without omentoplasty after abdominoperineal resection for carcinoma of the rectum.
Design: Prospective multicentre study.
Setting: 15 centres (three university, nine non-university teaching hospitals and three private clinics), France.
Subjects: 186 consecutive patients (between January 1983 and August 1990): 21 were withdrawn because of protocol violation leaving 165 for analysis.
Interventions: Abdominoperineal resection for adenocarcinoma of the distal third of the rectum followed by omentoplasty (n = 64) to the pelvic space or not (n = 101).
Main outcome measures: Number of healed perineums at one month, and the time interval to complete healing.
Results: 7 patients (4%) died, 4 of whom had had omentoplasty and 3 who had not (one perineal abscess). The number who developed immediate postoperative complications (11/64, 17% and 18/101, 18%) and median duration of hospital stay (21 days, range 8-191, and 22 days, range 8-132) were similar. The median time to complete healing (20 and 21 days), the rate of healed perineums at one month (42/62 and 67/99, both 68%) and the number of persisting sinuses at 12 months were also similar. The number of dehiscences of the perineum was significantly higher (p = 0.04) in the no omentoplasty group (16 compared with 3). There were 3 late deaths in the omentoplasty group and 7 in the no omentoplasty group, 1 and 5 with local recurrence, respectively. There were more recurrences in the no omentoplasty group but not significantly so.
Conclusions: Although this study was not randomised, the results suggest that omentoplasty to the pelvic space promotes perineal healing after abdominoperineal resection for carcinoma of the rectum by significantly reducing the need for secondary opening.