Abstract Background Anastomotic leakage is a major complication of rectal cancer surgery. The aim of this study was to investigate risk factors associated with symptomatic anastomotic leakage after total mesorectal excision (TME). Methods Between 1996 and 1999, patients with operable rectal cancer were randomized to receive short-term radiotherapy followed by TME or to undergo TME alone. Eligible Dutch patients who underwent an anterior resection (924 patients) were studied retrospectively. Results Symptomatic anastomotic leakage occurred in 107 patients (11·6 per cent). Pelvic drainage and the use of a defunctioning stoma were significantly associated with a lower anastomotic failure rate. A significant correlation between the absence of a stoma and anastomotic dehiscence was observed in both men and women, for both distal and proximal rectal tumours. In patients with anastomotic failure, the presence of pelvic drains and a covering stoma were both related to a lower requirement for surgical reintervention. Conclusion Placement of one or more pelvic drains after TME may limit the consequences of anastomotic failure. The clinical decision to construct a defunctioning stoma is supported by this study.