OBJECTIVE The purpose of this study was to compare long-term functional results of two methods of reconstruction after anterior rectal resection for cancer: low colorectal anastomosis and colonic J-pouch-anal anastomosis. BACKGROUND After anterior resection for mid or low rectal cancer, the decision to perform low colorectal or coloanal anastomosis is made intraoperatively, depending on the distance of the tumor from the anal verge. Functional results of these operations are considered to be similar one to two years after surgery. No study to date has compared long-term functional results after rectal excision followed by either low colorectal anastomosis or colonic J-pouch-anal anastomosis. METHODS From 1987 to 1992, 173 patients underwent anterior resection for cancer located between 2 to 12 cm from the anal verge. All patients alive without recurrence were contacted by telephone interview for assessment of functional results. There were 47 patients with colonic J-pouch-anal anastomosis and 34 patients with low colorectal anastomosis. Minimum follow-up was three years for all patients (mean, 5 years). RESULTS The two groups were well matched for gender, age, histologic stage, and use of adjuvant therapies. Patients with colonic J-pouch-anal anastomosis displayed significantly better function in terms of frequency of defecation (1.57+/-1 vs. 2.79+/-1; P=0.001) and presence of irregular transit or stool "clustering" (30 vs. 71 percent; P=0.003). Patients who underwent colonic J-pouch-anal anastomosis were significantly less likely to require constipating agents (4 vs. 21 percent; P=0.03) or need to follow a restricted diet (14 vs. 41 percent; P=0.01). Results concerning the need to defecate again within one hour and disruption of social or professional life as a consequence of surgery showed a tendency in favor of colonic J-pouch-anal anastomosis. CONCLUSIONS Colonic J-pouch-anal anastomosis offers superior long-term function compared with low colorectal anastomosis after radical treatment of rectal cancer. Preservation of a short rectal segment followed by a straight colorectal anastomosis does not offer any clinical advantage over colonic J-pouch-anal anastomosis.