ObjectiveTo compare the short-and mid-term outcomes of patients with esophageal cancer after subtotal esophagectomy via thoracoscopy in lateral prone position, prone position, or left lateral position. MethodsThis randomized prospectively controlled study was conducted in 121 patients receiving subtotal esophagectomy via thoracoscopy between January 2010 and February 2013. The patients were randomly assigned into three groups to underwent esophagectomy in lateral prone position, prone position, or left lateral position, respectively. Forty-three patients (24 males, 19 females, 61.5±1.5 years) underwent surgery in lateral prone position, 39 patients (21 males, 18 females, 63.2±1.7 years) in prone position and other 39 patients (22 males, 17 females, 60.1±1.6 years) in left lateral position. Esophagogastric anastomosis was performed in the left neck. ResultsThe median operative time in the three groups was 232 (165-296) min, 230 (170-310) min, and 280 (190-380) min, respectively (P < 0.05). The median perioperative bleeding was 262 (185-330) ml, 275 (100-320) ml and 350 (120-560) ml, respectively (P > 0.05). The average number of harvested lymph nodes was 19.1 (9-26), 18.4 (11-23), 10.9 (6-21), respectively (P < 0.05). The postoperative medical complications occurred in 10, 9 and 11 patients in three groups, respectively, with no statistical difference. Twenty patients died in the lateral prone position group after a median follow-up period of 19.2 (6-31) months, 18 patients died in the prone position group after a median follow-up period of 20.7 (8-29) months, and 21 patients died in the left lateral position group after a median follow-up period of 18.5 (12-33) months. ConclusionThe results confirm the feasibility and safety of this minimally invasive esophagectomy via thoracoscopy in lateral prone position, prone position, or left lateral position for patients with esophageal carcinoma. A possible advantage of lateral prone technique is that in case of an emergency, precious time could be saved in changing the position of the patient.