PURPOSE: Minimally invasive esophagectomy (MIE) has been increasingly used, but many reports have stated that recurrent laryngeal nerve (RLN) palsy after MIE is a major complication associated with postoperative pneumonia. Prevention of RLN palsy clearly has been a challenging task. The study aim was to determine if a three-dimensional (3-D) stereoscopic vision system can reduce the RLN palsy rate after MIE. METHODS: This was a retrospective study of MIE (McKeown esophagectomy) using a 3-D or 2-D stereoscopic vision system to treat 358 patients in the prone position between April 2010 and March 2019. The patients who underwent 3-D MIE (3-D group) or 2-D MIE (2-D group) were matched by using propensity score matching. After matching, the perioperative outcomes were compared between the groups. RESULTS: After propensity score matching, 154 patients were analyzed (77 patients, 3-D group; 77 patients, 2-D group). There were no significant differences in the patients' baseline characteristics in the matched cohort. There were no significant differences in the rates of pneumonia (Clavien-Dindo (C-D) grade ≥ II, 3-D vs. 2-D, 11 (14%) vs. 12 (16%)), anastomotic leakage (C-D grade ≥ II, 10 (13%) vs. 18 (23%)) and mortality. The rates of left RLN palsy (C-D grade ≥ IIIa, 1 (1.3%) vs. 7 (9.1%), P = 0.029), right RLN palsy (C-D grade ≥ I, 2 (3%) vs. 8 (10%), P = 0.049), comprehensive complication index (CCI®) (8.5 vs. 14.3, P = 0.011), and postoperative hospital stay period (median: 25 vs. 30 days, P = 0.034) were significantly lower in the 3-D group than in the 2-D group, respectively. CONCLUSIONS: In MIE, the 3-D viewing system was one of the factors that reduced postoperative morbidities such as the rates of each RLN palsy and CCI®, leading to shorter postoperative hospital stay.