Background Laparoscopic distal gastrectomy for early gastric cancer has been widely accepted, but laparoscopic total gastrectomy has still not gained popularity because of technical difficulty and unsolved safety issue. We conducted a single-arm multicenter phase II clinical trial to evaluate the safety and the feasibility of laparoscopic total gastrectomy for clinical stage I proximal gastric cancer in terms of postoperative morbidity and mortality in Korea. The secondary endpoint of this trial was comparison of surgical outcomes among the groups that received different methods of esophagojejunostomy (EJ). Methods The 160 patients of the full analysis set group were divided into three groups according to the method of EJ, the extracorporeal circular stapling group (EC; n  = 45), the intracorporeal circular stapling group (IC; n  = 64), and the intracorporeal linear stapling group (IL; n  = 51). The clinicopathologic characteristics and the surgical outcomes were compared among these three groups. Results There were no significant differences in the early complication rates among the three groups (26.7% vs. 18.8% vs. 17.6%, EC vs. IC vs. IL; p  = 0.516). The length of mini-laparotomy incision was significantly longer in the EC group than in the IC or IL group. The anastomosis time was significantly shorter in the EC group than in the IL group. The time to first flatus was significantly shorter in the IL group than in the EC group. The long-term complication rate was not significantly different among the three groups (4.4% vs. 12.7% vs. 7.8%; EC vs. IC vs. IL; p  = 0.359), however, the long-term incidence of EJ stenosis in IC group (10.9%) was significantly higher than in EC (0%) and IL (2.0%) groups ( p  = 0.020). Conclusions The extracorporeal circular stapling and the intracorporeal linear stapling were safe and feasible in laparoscopic total gastrectomy, however, intracorporeal circular stapling increased EJ stenosis.