Anastomotic leaks in colorectal surgery

BACKGROUND: Anastomotic leaks are a serious complication of bowel surgery. This study aimed to evaluate the rate and severity, and identify risk factors for leaks in patients undergoing bowel anastomoses. METHODS: Prospective evaluation was performed on patients undergoing bowel surgery within a colorectal surgical unit. Anastomotic leak was defined and graded according to severity. A nurse independently collected the information. Stepwise logistic regression analysis was performed. RESULTS: Two thousand three hundred and sixty-three patients underwent 2994 anastomoses. Their median age was 64 years. Seven per cent were emergency operations. Anastomotic leak occurred in 82 patients (2.7%). Sixty-three per cent of leaks were managed with drainage or re-operation. Ultra-low anterior resection (ULAR) was associated with the highest subgroup leak rate (7.3%). In multivariable analysis, independent predictors for a leak included 'other' pathologies (iatrogenic injury, ischaemia, radiation enteritis) (P=0.016, odds ratio (OR): 6.3, 95% confidence interval (CI): 1.4-28.0), ULAR (P=0.001, OR: 8.5, 95% CI: 2.3-31.2) and the surgeon (A: P<0.001, OR: 3.4, 95% CI: 2.1-5.6). CONCLUSION: Majority of predictors for anastomotic leak were fairly intuitive. Nonetheless, it was relevant to note the importance of the individual surgeon as an independent predictor for leaks.