Safety of Proctored Learning Curve in Upper Gastrointestinal Robotic Surgery

Background. Safe implementation of robotic upper gastrointestinal surgery requires an optimized learning process. We report the initial experience, outcomes and learning curve of a team introducing this technique in Argentina through the Upper GI International Robotic Association (UGIRA) structured training pathway. Methods. Consecutive patients who underwent robotic upper gastrointestinal surgery using the Da Vinci Xi® system were included from a prospective database. All procedures were performed by a single console surgeon, proctored by an experienced team from UMC Utrecht. Proctoring was performed on-site or remotely via telementoring system. The learning curve plateau was assessed using CUSUM analysis for operative time and lymph node yield. Results. Between 2022 and 2026, 66 robotic upper gastrointestinal procedures were performed by our team, including 26 esophagectomies (RAMIE), 8 total gastrectomies, and 32 benign procedures. No major complications, conversions, or mortality occurred in the benign group. In the RAMIE cohort, major complications (Clavien-Dindo ≥III) occurred in 34.6% of patients, with an anastomotic leak rate of 15.4% and a single case of 30-day mortality (3.8%). R0 resection was achieved in 96.2% of cases, with a median lymph node yield of 26 [IQR 22–32]. In the gastrectomy group, major complications occurred in 25.0% of patients, with no anastomotic leaks or mortality, and a median lymph node yield of 28 nodes [IQR 26–32]. CUSUM analysis for RAMIE identified an operative time plateau at case 7 and a lymph node yield trough at case 18. Conclusion. This study reports the initial implementation of robotic upper gastrointestinal surgery in Argentina following the UGIRA structured training pathway enhanced by telementoring. Despite limited case volume and follow-up, the approach appears feasible and safe, with acceptable short-term perioperative and oncologic outcomes.