P1.244. Robotic-Assisted Versus Conventional Minimally Invasive Oesophagectomy for Oesophageal Cancer: A Comparative Meta-Analysis of Perioperative and Oncological Outcomes

Abstract

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – technique Background Robotic-assisted minimally invasive oesophagectomy (RAMIE) is increasingly adopted to improve surgical precision, mediastinal dissection, and surgeon ergonomics. However, its clinical and oncological advantages over conventional minimally invasive oesophagectomy (cMIE) remain uncertain. This study presents the largest contemporary comparative meta-analysis between RAMIE and cMIE focusing on peri-operative and oncological outcomes. Methods MEDLINE, Embase and Cochrane databases were searched for comparative studies published from 2019 onwards evaluating RAMIE versus cMIE for oesophageal malignancy. Randomised controlled trials and observational cohorts were included. Primary outcomes were R0 resection rate and lymph node yield. Secondary outcomes included operative duration, blood loss, hospital and intensive care unit length of stay, recurrent laryngeal nerve palsy, conversion to open surgery, major complications (Clavien–Dindo ≥III), anastomotic leak, pneumonia, re-operation, and 30- and 90-day mortality. Continuous variables were pooled using mean differences and dichotomous outcomes using risk ratios with 95% confidence intervals under random-effects models. Risk of bias was assessed using ROBINS-I and Cochrane RoB-2 tools. Overlapping national database cohorts were excluded from pooled analyses. The protocol was prospectively registered (PROSPERO CRD420251045505). Results Twenty-two studies involving over 20,000 patients were included, comprising two randomised controlled trials and multiple matched retrospective cohorts. RAMIE was associated with a longer operative duration (MD 25.8 minutes, 95% CI 0.5–51.0) and a modest increase in lymph node yield (MD 1.4 nodes, 95% CI 0.1–2.6). R0 resection rates were equivalent between approaches (RR 1.01, 95% CI 0.99–1.03). Recurrent laryngeal nerve palsy showed a non-significant trend toward higher incidence with RAMIE (RR 1.18, 95% CI 0.93–1.49). Major complications, anastomotic leak and pneumonia were comparable. Conversion to open surgery trended in favour of RAMIE (RR 0.63, 95% CI 0.36–1.10). 30-day mortality did not differ significantly, while 90-day mortality trended toward reduction with RAMIE (RR 0.73, 95% CI 0.50–1.05). Blood loss and length of stay were not significantly different. Conclusion RAMIE demonstrates comparable peri-operative safety and oncological outcomes to cMIE. However, longer operative duration during the initial learning curve remains a consideration. Interpretation is limited by heterogeneity and predominantly retrospective data. Further evaluation of long-term functional outcomes and resource utilisation is required to determine whether technical advantages translate into meaningful overall benefit.