Long-term outcomes of robotic-assisted versus thoraco-laparoscopic McKeown esophagectomy for esophageal cancer: a propensity score-matched study.

The long-term outcomes of robotic-assisted McKeown esophagectomy (RAME) compared to thoraco-laparoscopic McKeown esophagectomy (TLME) for the patients with esophageal squamous cell carcinoma (ESCC) remain unclear. The aim of this study was to compare the number of dissected lymph nodes and long-term survival between RAME and TLME using a propensity score-matched (PSM) analysis. A total of 721 patients undergoing minimally invasive McKeown esophagectomy at our department from February 2015 to October 2019 were analyzed, including 310 patients in RAME group and 411 in TLME group. The exact numbers of lymph nodes including those among thoracic and abdominal categories as well as those along the recurrent laryngeal nerve (RLN) were all recorded. PSM analysis was applied to generate matched pairs for further comparison. All patients with R0 resection were followed with a strict follow-up period which range from 1 to 56 months. The effect of lymphadenectomy was compared between all patients in unmatched and matched groups. Long-term outcomes consisting of overall survival (OS), disease-free survival (DFS) and recurrence rate (including regional recurrence rate, systemic recurrence rate and mediastinal lymph nodes recurrence rate) were compared in R0 resection patients. Finally, 292 patients were identified for each cohort after PSM. RAME was found to yield significantly more left RLN lymph nodes (mean: 2.27 ± 0.90 vs. 2.09 ± 0.79; P = 0.011) and more thoracic lymph nodes (mean: 12.60 ± 4.22 vs. 11.83 ± 3.12, P = 0.012) compared with TLME after PSM analysis. There was no significant difference in the OS and DFS between the RAME and TLME group. Besides, total recurrences were recognized in 33 (11.7%) patients in the RAME group and 36 (12.9%) in the TLME group (P = 0.676). The mediastinal lymph nodes recurrence rate in the RAME group was tended to be lower than that in the TLME group (2.5% vs. 5.4%, P = 0.079). Therefore, RAME might be an alternative approach for the treatment of ESCC with more lymph nodes dissected and similar long-term survival outcomes compared to TLME.