Background: Minimally invasive esophagectomy is a feasible technique shown to be safe and oncologically adequate for the treatment of esophageal cancer. This study aimed to describe one surgeon's learning curve for video-assisted thoracoscopic esophagectomy with the patient in lateral position. Methods: From May 2010 to June 2012, 89 thoracoscopic esophagectomies for esophageal cancer were performed by one surgeon. The patients were divided into three groups. Group A included the first 30 cases. Group B comprised cases 31 to 60, and group C included the final 29 cases. The demographic characteristics and the intra- and postoperative variables were collected retrospectively and analyzed. Results: One postoperative death occurred. Eight patients required conversion. No significant difference in background or clinicopathologic factors among the three groups was observed. Compared with group A, a significant decrease in intrathoracic operative time (107.7 ± 16.2 min; P = 0.0000), total operative time (326.3 ± 40.7 min; P = 0.0002), and blood loss (290.8 ± 114.3 ml; P = 0.0129) was observed in group B, whereas more retrieved nodes were harvested (20.1 ± 9.5; P = 0.0002). The last 29 patients (group C) involved significantly less intrathoracic operative time (82.8 ± 18.4 min; P = 0.0386), total operative time (294.7 ± 37.4 min; P = 0.0009), and blood loss (234.7 ± 87.8 ml; P = 0.0125) as well as a shorter postoperative hospital stay (12.4 ± 3.7 days; P = 0.0125) compared with group B. A significant decline in the overall morbidity from group A to group C ( P = 0.0005) also was observed. Conclusions: The results of this study suggest that at least 30 cases were needed to reach the plateau of thoracoscopic esophagectomy. After more than 60 cases of thoracoscopic esophagectomies had been managed, lower morbidity could be obtained. [ABSTRACT FROM AUTHOR]