[Anastomotic leak following colorectal surgery: incidence, risk factors and treatment].

The aim of this study was to assess the incidence and identify the risk factors associated with colorectal anastomotic leakage. A further objective was to investigate the therapeutic choices. We reviewed the clinical files of 124 patients who underwent mechanical end-to-end anastomosis after colorectal resection during the period 2000-2007. The mean age was 66.9 years and the male:female ratio was 1.2:1. Indications for surgery were malignant neoplasms in 109 cases (87.9%) and benign disease in 15 cases (12.1%). The anastomosis was always performed in an elective setting in the framework of a procedure of one or more stages (87.9% and 7.3% respectively). A pelvic drain was positioned in a routine manner and no protective ileostomies were constructed. Among the various risk factors, co-morbidities, ASA risk and low anastomotic level were the most important we examined. Student's t-test, the Chi-square test and Fisher's test were used for comparative univariate analysis, with significant results for p < or = 0.05. The incidence of anastomotic leak was 10.5% (13/124), but only 4.8% (6/124) required a second operation or failed to heal with simple conservative therapy. The pelvic drain was always effective in allowing the early diagnosis of leakages and limiting the spread of peritoneal inflammation. Mortality was 1.6% (2/124) overall, but, if leakages alone were considered, it increased to 15.4%. Among the risk factors examined, only low anastomotic level was confirmed by statistical analysis. In conclusion, the study confirms low anastomotic level as a negative prognostic factor for the healing of colorectal anastomosis. The pelvic drain proved to be effective in allowing early diagnosis and conservative management of leakages.