OBJECTIVES: The outcome of colovesical fistula management may be unsatisfactory; complications are reported in up to 45% of patients. Published studies are retrospective and tend to lack standardized management strategies and long-term follow-up. This cohort study assesses a policy of resection of colovesical fistulae in continuity with any distal colorectal stricture, and includes 5-year follow-up. METHOD: All patients undergoing surgery in our institution for colovesical fistula between February 1991 and April 1995 were entered into the study. The fistulae were resected in continuity with any distal bowel stricture, according to a standard single-stage operative protocol. Postoperative mortality and morbidity were recorded, and prospective review was undertaken at April 2000. RESULTS: Nineteen consecutive patients entered the study. The source of the fistula was diverticular disease (n = 14), colorectal cancer (n = 3), trauma (n = 1) or Crohn's (n = 1) disease. Thirteen patients had a colorectal stricture. One patient died due to ischaemic colitis within 30 days of surgery. Eleven other patients died of unrelated causes before April 2000, in whom there was no evidence of fistula recurrence before death at a median of 37 months after operation (range 2-95 months). At 5-year follow-up there was no evidence of fistula recurrence in the seven remaining patients. CONCLUSIONS: A policy of resection of the fistula and associated colorectal stricture with primary bowel anastomosis and bladder drainage, resulted in no recurrences and low morbidity. However comorbidity is important in this patient population, most of whom will die from unrelated causes within a few years.