Objective: To compare the short-term outcomes of different techniques of colorectal anastomosis in extended left hemicolectomy. Methods: The clinical and follow-up data of 28 patients who underwent extended left colectomy from July 2000 to August 2013 were retrospectively analyzed. The patients were classified into conventional group (15 cases) and modified group (13 cases) according to the approaches used for colorectal anastomosis. Patients in conventional group underwent conventional colorectal anastomosis anterior to the small intestine, and those in modified group underwent colorectal anastomosis through the mesentery of the small intestine (8 cases) or posterior to the mesentery of the small intestine (5 cases). The intra- and postoperative variables between the two groups were compared. Results: There was no significant difference in operative time and intraoperative blood loss between the two groups (both P>0.05), but the average time to flatus and oral food intake, and length of postoperative hospital stay were shorter in modified group compared with conventional group (all P<0.05). The incidence of overall postoperative complications in modified group was significantly lower than that in conventional group (23.1% vs. 46.7%, P<0.05), where the main difference was due to the incidence of high-level small intestinal obstruction (26.7% vs. 0.0%, P<0.05). Conclusion: In extended left colectomy, using colorectal anastomosis through or posterior to the mesentery of the small intestine can decrease the tension at the anastomotic stoma, and avoid jejunal compression, and its efficacy is superior to that using anastomosis anterior to the small intestine.