Anastomotic Leakage after Anterior Resection for Rectal Cancer with Mesorectal Excision: Incidence, Risk Factors, and Management

摘要

We investigated risk factors and prognostic implications of symptomatic anastomotic leakage after anterior resection for rectal cancer, and the influence of a diverting stoma. Our retrospective review of prospective collected data analyzed 475 patients who underwent anterior resection for rectal cancer. Uni- and multivariate analysis was made between anastomotic leakage and patient, tumor, and treatment variables, either for the overall group (n 5 475) and in the midlow rectal cancer subgroup (n 5 291). Overall rate of symptomatic leakage was 9 per cent (43 of 475) with no related postoperative mortality. At univariate analysis, significant factors for leak were a tumor less than 6 cm from the anal verge (13.7 vs 6.6%; P 5 0.011) and intraoperative transfusions (16.9 vs 4.3%; P 5 0.001). Similar results were observed in the midlow rectal cancer subgroup. At multivariate analysis, no parameter resulted in being an independent prognostic factor for risk of leakage. In patients with a leakage, a temporary enterostomy considerably reduced the need for reoperation (12.5 vs 77.8%; P < 0.0001) and the risk of a permanent stoma (18.7 vs 28.5%; P 5 0.49). The incidence of anastomotic failure increases for lower tumors, whereas it is not influenced by radiotherapy. Defunctioning enterostomy does not influence the leak rate, but it mitigates clinical consequences. Despite improvements in surgical techniques and perioperative management, symptomatic anastomotic leakage remains one of the most feared and dreadful complications after anterior resection for rectal cancer with considerable clinical implications. The average incidence is 10 per cent (range, 3 to 21%), whereas subclinical anastomotic failure may occur in up to 51 per cent of patients. Many factors relating to the patient and the type of anastomosis have been studied in connection with anastomotic leakage such as male sex, obesity, malnutrition, cardiovascular disease, perioperative blood transfusion, steroid use, and preoperative irradiation; anyway, the only single factor consistently shown to predict leakage in all published studies is a low rectal anastomosis. This study was designed to evaluate the incidence, risk factors, the treatment, and the prognosis of anastomotic leakage in a population of patients with rectal cancer submitted to potentially curative anterior resection with primary anastomosis at a single institution over a 20-year period and to assess the role of a temporary stoma in relation to anastomotic leakage.