BACKGROUND: Colonic J pouch reconstruction has been found to be associated with a lower incidence of anastomotic leakage than straight anastomosis. However, studies on this topic are underpowered and retrospective. This randomized trial evaluated whether the incidence of anastomotic leakage was reduced after colonic J pouch reconstruction compared with straight colorectal anastomosis following anterior resection for rectal cancer.METHODS: This multicentre RCT included patients with rectal carcinoma who underwent low anterior resection followed by colorectal anastomosis. Patients were assigned randomly to receive a colonic J pouch or straight colorectal anastomosis. The main outcome measure was the occurrence of major anastomotic leakage. The incidence of global (major plus minor) anastomotic leakage and general complications were secondary outcomes. Risk factors for anastomotic leakage were identified by regression analysis.RESULTS: Of 457 patients enrolled, 379 were evaluable (colonic J pouch arm 190, straight colorectal arm 189). The incidence of major and global anastomotic leakage, and general complications was 14·2, 19·5 and 34·2 per cent respectively in the colonic J pouch group, and 12·2, 19·0 and 27·0 per cent in the straight colorectal anastomosis group. No statistically significant differences were observed between the two arms. In multivariable logistic regression analysis, male sex (odds ratio 1·79, 95 per cent c.i. 1·02 to 3·15; P = 0·042) and high ASA fitness grade (odds ratio 2·06, 1·15 to 3·71; P = 0·015) were independently associated with the occurrence of anastomotic leakage.CONCLUSION: Colonic J pouch reconstruction does not reduce the incidence of anastomotic leakage and postoperative complications compared with conventional straight colorectal anastomosis. Registration number NCT01110798 (http://www.clinicaltrials.gov).