Abstract
Esophageal resections are associated with significant postoperative morbidity and mortality. The standard surgical approach is minimally invasive, with a high proportion of robotic resections. The aim of this study was to evaluate the learning curve (LC) in the context of establishing a structured training program. As part of the retrospective analysis, all consecutive hybrid robot-assisted Ivor-Lewis esophageal resections performed by a defined team between May 2023 and October 2025 were evaluated. The LCs for the frequency of postoperative complications Clavien-Dindo (CD) ≥ III, in particular anastomotic leakage, and the textbook outcome rate were analyzed. The LCs for these target variables were evaluated using the cumulative sum (CUSUM) analysis. Between May 2023 and October 2025, a total of 71 patients underwent hybrid robot-assisted Ivor-Lewis esophagectomy. In-hospital mortality in the study population was 1.4% (1/71). Postoperatively, 29.6% (21/71) of patients experienced postoperative complications classified as grade IIIa or higher according to CD. A total of 9.9% of patients (n = 7) developed anastomotic leakage. After case no. 20 a significant reduction in anastomotic leakage rate was observed, with a leakage rate of 2% after completion of the LC. The rate of textbook outcomes in our cohort was 49% (35/71). The LC for robot-assisted esophageal resections is associated with the occurrence of postoperative complications. After 44 operations there is a reduction in complications. Total in-hospital mortality was low, demonstrating that robotic procedures can be learned safely at experienced centers while achieving textbook outcomes during the learning curve. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11701-026-03405-6.