PURPOSE: Different studies have shown that low colorectal and coloanal anastomosis often yield poor functional results. The aim of the present study was to investigate whether a colonic reservoir is able to improve functional results. METHODS: Thirty-eight consecutive patients subjected to low anterior resection were randomized following rectal excision in two groups. One (n = 19) had a stapled straight coloanal anastomosis, and the other (n = 19) had a IO-cm stapled colonic pouch low rectal anastomosis. Median anastomotic distance above the anal verge was 3.38 +/- 0.56 cm and 2.14 +/- 0.36 cm in both groups, respectively. Continence alterations, urgency, tenesmus, defecatory frequency, anal resting and maximum voluntary squeezing pressures, and maximum tolerable volume were evaluated one year later. RESULTS: One patient died of pulmonary embolism, and seven presented with a recurrence and were excluded from the study. Stool frequency was greater than three movements per day in 33.3 percent of cases with a reservoir and in 73.3 percent of those with a straight coloanal anastomosis (P < 0.05). Maximum tolerable volume was significantly greater in patients with a reservoir (335 +/- 195) than in those without (148 +/- 38) (P < 0.05). There were no significant differences in other variables studied. CONCLUSIONS: This study shows that some aspects of defecatory function after rectal excision could improve with a colonic reservoir.