CR03SMOKING IS A MAJOR RISK FACTOR FOR ANASTOMOTIC LEAK AFTER LOW ANTERIOR RESECTION

To identify risk factors for anastomotic leak after colonic and rectal anastomosis at our institution. Data was collected prospectively by Otago Clinical Audit between 1/1/2000 and 31/12/2004. All operations involving a colonic or rectal anastomosis were included. For all patients the association between clinical anastomotic leakage and seven demographic and clinical variables was assessed. For the subgroup of 118 patients who underwent low anterior resection, data on a further fifteen variables related to operative technique, lifestyle factors and pathology were obtained from chart review. The association between these and leakage was assessed. Continuous variables were analysed with t-tests. Categorical variables were analysed with univariate logistic regression, and Odds ratios (OR) with 95% confidence intervals (CI) were calculated. Over the five-year period, 878 patients had an operation with a large bowel anastomosis. The overall rate of anastomotic leakage was 4.8% (42 of 878). In this group, longer operating time (p < 0.01) and a low rectal anastomosis (OR 3.12; CI 1.44, 6.41) were significantly associated with anastomotic leak. The leak rate for low anterior resections was 12.7% (15 of 118). Involved lymph nodes (OR 6.7; CI 1.31, 64.7), metastatic disease (OR 4.32; CI 1.12, 16.7) and cigarette smoking (OR 17.6; CI 4.84, 63.8) were significantly associated with anastomotic leak in this subgroup. Current smokers and patients with advanced malignancy undergoing low anterior resection are at highest risk of anastomotic leak. These findings need to be confirmed prospectively and may have implications for clinical decision making. [PUBLICATION ABSTRACT]