The incidence of anastomotic leaks in patients undergoing colorectal surgery

Background There is evolving interest in auditing and credentialling the performance of surgeons. The incidence of anastomotic leakage has been proposed as a measure of performance following colorectal surgery. The aim of this study was to evaluate the incidence and risk factors associated with anastomotic leakage in patients undergoing resections of the colon and rectum. Methods A prospective database was developed for all patients undergoing colorectal surgery. Anastomotic leakage was defined prior to the commencement of the study. A logistic regression analysis was performed to determine independent predictors of leakage. The variables analysed included age, sex, American Society of Anesthesiology (ASA) score, anatomical location, pathology, emergency surgery, type of anastomosis, a covering stoma and radiotherapy. Significance was defined as the probability of a type 1 error of < 5%. The results are presented as odds ratios (ORs) and 95% confidence intervals (95% CIs). Results There were 1598 patients who underwent 1639 anastomoses. Their mean age was 63 years, 34% of patients were ASA 3 or 4, and 16% of the operations were emergencies. Anastomotic leaks occurred in 2.4% (40/1639) of anastomoses. The leak rate for intraperitoneal anastomoses was 1.5% (19/1283) vs 6.6% for extraperitoneal anastomoses (21/316). Half of these leaks (20/40) were managed with re-operation or percutaneous drainage procedures. Ultra-low anterior resections were associated with the highest leak rate (8%, 18/225). A logistic regression analysis identified a covering stoma (P = 0.0001, OR 5.078, 95% CI 2.527-10.23) and diverticular disease (P = 0.037, OR 2.304, 95% CI 1.053-5.042) as independent predictors of a leak. Conclusions Within this surgical unit, the incidence of leaks from intraabdominal anastomoses was relatively low. However, leaks in patients undergoing extraperitoneal anastomoses continue to be a major cause of morbidity and mortality.