The Risk Factors of Anastomotic Leakage and Influence of Fecal Diversion after Resection of Rectal Cancer

Purpose. The most important surgical complication following rectal resection with anastomosis is symptomatic anastomotic leakage. This study investigated factors in anastomotic leakage and the effect of fecal diversion after resection of middle and low rectal cancers. Methods. Prospective data collection from patients with rectal cancer at 16 cm or less from anal verge was reviewed and risk factors of anastomosis investigated. The relationship between anastomotic leakage and clinicopathologic variables was determined using logistic regression analysis. Multivariate analysis with a logistic regression model was done to determine independent factors of anastomotic leakage. Results. From January 1993 to June 2003, 999 rectal cancer patients received elective radical resection and anastomosis. Fifty-three of these patients experienced anastomotic leakage. Univariable analysis revealed that age ＞70 years old (P=0.008), tumor location between 6-12 cm (P=0.026), and surgery with ultra-LAR (P=0.002) were significantly associated with increased anastomotic leakage. Multivariate analysis showed only older patients (P=0.009) and operation method (P=0.002) were independent factors for the development of anastomotic leakage; tumor of the middle rectum (6-12 cm) had borderline significance (P=0.078). Thirty percent (n=3/10) of patients with diverting stoma and 100% (n=43/43) of patients without diverting stoma needed reoperation to treat abdominal sepsis. Conclusion. Older rectal cancer patients, or those who have had anastomosis at the anorectal junction or dentate line, have increased risk of anastomotic leakage. A diverting stoma seems not to decrease incidence of anastomotic leakage, but may decrease the necessity of reoperation and provide a positive oncological impact if leakage occurs.