[Factors associated with anastomotic leakage after anterior resection in rectal cancer].

OBJECTIVE: To analyze the factors associated with anastomotic leakage after anterior resection in rectal cancer with the technique of total mesorectal excision (TME).

METHODS: From January 2005 and December 2007, 738 consecutive patients with rectal cancer underwent anterior resection. The data of those patients was collected and reviewed retrospectively. The associations between anastomotic leakage and 9 patient-related variables as well as 7 surgical-related variables were examined.

RESULTS: Low rectal cancer (located 7 cm or less above the anal edge), non-specialized surgeon and transanal tube use were the risk factors associated with anastomotic leakage on univariate analysis. The anastomotic leakage rate of low-rectal cancer was significantly higher than that of high-rectal cancer (5.9% vs. 0.9%, P = 0.003). The anastomotic leakage rate of the cases operated by colorectal surgeon was significantly lower than that of the cases operated by non-specialized surgeon (3.9% vs. 11.3%, P = 0.031). There was a tendency for colorectal surgeons to operate on a greater proportion of low rectal cancer than non-specialized surgeons (72.1% vs. 52.8%, P = 0.003). The leakage rate of transanal tube group was unexpectedly higher than that in patients without transanal tube (14.5% vs. 3.6%, P < 0.001). On multivariate logistic regression analysis, diabetes mellitus (P = 0.027), distance less than 1 cm from tumor to distal resection margin (P = 0.009) and defunctioning stoma (P = 0.031) were also associated with anastomotic leakage rate besides low rectal cancer, non-specialized surgeon and transanal tube use. In a further analysis of 522 patients with low rectal cancer, the leakage rate of defunctioning stoma group was significantly lower than that of non-stoma group (2.9% vs. 8.5%, P = 0.007). By contract, the leakage rate of transanal tube group was still higher than that in patients without transanal tube (15.1% vs. 4.9%, P = 0.008) because of its poor protective effect as well as the selection bias.

CONCLUSIONS: Low-rectal cancer, non-specialized surgeons and diabetes mellitus are risk factors of anastomotic leakage after rectal surgery. A defunctioning stoma was effective in preventing leakage after low-rectal cancer surgery.