Managing leaks following anterior resections: a new classification system.

Background/aims: Anastomotic leak rates following rectal resections and anastomosis stand at 2.9-15.3%, with an attendant mortality rate of 6- 39.3%. The aim of our study was to identify those patients who had developed an anastomotic leak, and evaluate the indications for re-exploration as well as study the impact of covering colostomies on the subsequent outcome.

Methodology: We analyzed 266 consecutive anterior resections for rectal cancer performed at a single institution between 1st September 2002 and 31st December 2006.

Results: Twenty-one anastomotic leaks were encountered in 266 resections. Covering colostomies were performed in 56% (151/266) of the patients. Out of the 21 patients who developed a leak, 9 had a covering colostomy (42.8%). In this group, 3 of patients (33%) could be managed without surgical re-exploration, while all the 12 patients without a covering colostomy had to undergo a reexploration. With our new classification system for anastomotic dehiscence, the clinical decisions appear to follow a predictable pattern. There were 4 deaths (1.5%). However, the mortality rate in the patients undergoing surgical re-exploration for complications was 16.6% (3/18).

Conclusions: This large study provides an insight into the potential advantages of covering colostomies wherever indicated following anterior resections. With our new and effective classification system for clinical leaks, the management road map can be simplified and standardized.