Objective To investigate the application value of semi-end-to-end esophagojejunal anastomosis for the Roux-en-Y digestive tract reconstruction after laparoscopy-assisted total gastrectomy (LATG). Methods The retrospective cohort study was conducted. The clinical data of 205 gastric adenocarcinoma patients who underwent LATG at the Southwest Hospital of the Third Military Medical University from January 2012 to December 2015 were collected. Among 205 patients, 140 who underwent Roux-en-Y digestive tract reconstruction with end-to-side esophagojejunal anastomosis were allocated into the control group, and 65 who underwent Roux-en-Y digestive tract reconstruction with semi-end-to-end esophagojejunal anastomosis were allocated into the study group. All the patients underwent LATG according to Japanese gastric cancer treatment guidelines (ver.3). Observation indicators included: (1) surgical situations: operation completion, operation time, time of digestive tract reconstruction, volume of intraoperative blood loss and number of patients with intraoperative esophagojejunal anastomosis-site complications (anastomosis-site stenosis and bleeding). (2) Postoperative situations: time to initial anal exsufflation, time of postoperative drainage tube removal, number of patients with postoperative esophagojejunal anastomosis-site complications (anastomosis-site stenosis, bleeding and leakage), number of patients with postoperative non-esophagojejunal anastomosis-site complications (pulmonary infection, pleural effusion, wound infection, abdominal abscess, intra-abdominal bleeding, duodenal stump fistula, intestine obstruction and internal abdominal hernia) and duration of postoperative hospital stay. (3) Follow-up situations. Follow-up using outpatient examination or telephone interview was performed to detect the survival of patients and tumor recurrence or metastasis up to April 2016. Measurement data with normal distribution were represented as ±s and comparison between groups was analyzed using the t test. Measurement data with skewed distribution were represented as M (range) and comparison between groups was analyzed using the nonparametric test. Comparison of count data was analyzed using the chi-square test, and ranked data was analyzed using the nonparametric test. Results (1) Surgical situations: all the patients received successful LATG and Roux-en-Y digestive tract reconstruction. Operation time and time of digestive tract reconstruction were (254±57)minutes, (53±10)minutes in the control group and (233±55)minutes, (41±9)minutes in the study group, respectively, with statistically significant differences between the 2 groups (t=2.508, 8.191, P 0.05). Of 8 patients with anastomosis-site stenosis in the control group, 4 didn′t receive special treatment, 1 underwent end-to-side esophagojejunal anastomosis again after dismantling anastomosis-site and 3 underwent side-to-side anastomosis between jejunal stump and distal jejunum again due to higher anastomosis-site surface. One patient with intraoperative anastomosis-site bleeding in the study group underwent strengthening suture of anastomosis-site and then bleeding was stopped. (2) Postoperative situations: number of patients with anastomosis-site stenosis, bleeding and leakage (postoperative esophagojejunal anastomosis-site complications) was respectively 11, 0, 6 in the control group and 0, 0, 1 in the study group, with a statistically significant difference between the 2 groups (χ2=6.232, P 0.05). Of 11 patients with postoperative anastomosis-site stenosis in the control group, 5 didn′t received special treatment and 6 were improved through endoscopic balloon dilatation. Patients with postoperative anastomosis leakage were improved after adequate drainage, anti-infection and symptomatic treatments. Patients with pulmonary infection were improved after anti-infection treatment. Patients with pleural effusion, wound infection, abdominal abscess and duodenal stump fistula were improved after adequate drainage, anti-infection and symptomatic treatments. Bleeding of patients with intra-abdominal bleeding in the control group was controlled by reoperation, and hemostasis and symptomatic treatment were conducted for patients with intra-abdominal bleeding in the study group. Patients with intestine obstruction and internal abdominal hernia were improved after reoperation. (3) Follow-up situations: among 205 patients, 192 were followed up for 4-51 months with a median time of 28 months, including 130 in the control group and 62 in the study group. During the follow-up, death and tumor recurrence or metastasis were respectively detected in 19, 23 patients in the control group and 8, 10 patients in the study group. Conclusion Semi-end-to-end esophagojejunal anastomosis is safe and feasible for the Roux-en-Y digestive tract reconstruction after LATG, with advantages of shorter time of digestive tract construction and fewer postoperative esophagojejunal anastomosis-site complications. Key words: Gastric neoplasms; Gastrectomy; Semi-end-to-end anastomosis; Roux-en-Y reconstruction; Laparoscopy