ANASTOMOTIC LEAKAGE AFTER ANTERIOR RESECTION OF THE RECTUM

Objective: To evaluate the results of elective colorectal operations, to identify factors that influence the anastomotic leak rate, and to a ssess the value of a covering colostomy. Design: Retrospective study. Setting: University hospital. Subject: 134 consecutive patients underg oing elective resection for a neoplasm (125 carcinomas and 10 villous adenomas) of the colon and rectum. Interventions: 135 operations (one for a recurrence in the anastomotic line). Main outcome measures. Morb idity and mortality. Results: Two patients died (1%) and there were 29 complications (21%); 16 patients developed clinical anastomotic leaks (12%) and 3 patients symptomatic strictures (2%). One patient who dev eloped a leak died and 11 of the 16 (69%) were re-operated on. The res t settled spontaneously. The only significant risk factor associated w ith leakage was distance of the anastomosis from the anal verge (all 1 6 leaks were within 7 cm, compared with 44/118 without leaks, p < 0.00 1). Four patients who developed leaks were left with permanent colosto mies. Protecting stomas had no significant influence, despite the fact that only 1 of those that developed leaks had a colostomy compared wi th 9/109 that did not. Conclusion: Prospective controlled studies are needed to establish the true value of a covering colostomy after low c olorectal anastomosis.