112 ROBOT-ASSISTED MINIMALLY INVASIVE ESOPHAGECTOMY CAN BE PERFORMED EVEN IN LEARNING CURVE PERIOD; PROPENSITY SCORE MATCH ANALYSIS

Da Vinci Surgical System Xi was introduced in our institution in 2017. Since then we started robot-assisted minimally invasive esophagectomy (RAMIE) as private practice. Japanese national insurance system started to reimburse RAMIE in 2018. Since then, the number of RAMIE is explosively increasing. Currently the indication of RAMIE in our institution purely depends on machine availability. However, the real clinical benefit of RAMIE over conventional minimally invasive esophagectomy (MIE) remains unknown. Methods Between January 2015 and January 2020, data of 156 consecutive patients with esophageal carcinoma undergoing esophagectomy was retrospectively reviewed. After exclusion of 12 salvage esophagectomy for patients after definitive chemoradiotherapy (>50Gy), 10 mediastinoscopic esophagectomy, 5 esophagectomy without upper mediastinal dissection, 2 two-stage esophagectomy and 1 total pharyngo-laryngo-esophagectomy, 126 patients (46 RAMIE and 80 MIE) were included in the analysis. The 1:1 propensity score match analysis was performed to compare the short-term outcomes between RAMIE and MIE. Results After matching, 45 RAMIE and 45 MIE were analyzed. RAMIE took significantly longer thoracic operation time (370 min vs. 288 min; P < 0.0001). Intraoperative blood loss (80 g vs. 50 g; P = 0.0759), harvested thoracic nodes (22 vs 25; P = 0.1188), harvested upper mediastinal nodes (12 vs 12; P = 0.4233) were similar. RAMIE showed less severe postoperative morbidity (Clavien-Dindo Grade III or higher) (9% vs. 22%; P = 0.0810) and lower incidence of recurrent laryngeal nerve palsy (Clavien-Dindo Grade II or higher) (7% vs. 18%; P = 0.1076). There was no postoperative death in both groups. Conclusion Although this result contains our early learning curve period of RAMIE, short term outcome of RAMIE is acceptable from a safety point of view.