Textbook Outcomes of Totally Robotic Versus Totally Laparoscopic Pancreaticoduodenectomy for Periampullary Neoplasm: A Propensity Score-Matched Cohort Study

摘要

Background/Objectives: Textbook outcome (TO) is a composite quality measure in surgery, but few studies have compared TO between robotic pancreaticoduodenectomy (RPD) and laparoscopic pancreaticoduodenectomy (LPD). This study aimed to evaluate and compare TO following RPD and LPD for periampullary neoplasms. Methods: We retrospectively analyzed 322 patients who underwent minimally invasive PD between 2010 and 2023 (RPD, n = 60; LPD, n = 262). LPD was first introduced in 2004, but only cases performed since 2010 were included, while RPD has been performed since 2019. Propensity score matching (1:2) yielded 48 RPD and 96 LPD patients. TO was defined as the absence of pancreatic fistula, bile leak, post-pancreatectomy hemorrhage, severe complications (Clavien-Dindo ≥ III), readmission, and in-hospital or 30-day mortality. Results: In the entire cohort, 240 of 322 patients (74.5%) achieved TO. After matching, TO rates were 64.6% in RPD and 76.9% in LPD (p = 0.656). Perioperative outcomes, including operative time, blood loss, transfusion, hospital stay, and major complications, were comparable, although RPD showed a higher incidence of hemorrhage (p = 0.032). Multivariate analysis identified body mass index < 25 kg/m2 as an independent predictor of achieving TO (OR 3.13, p = 0.008). Conclusions: RPD and LPD achieved comparable textbook outcomes in periampullary surgery. Both approaches are feasible when performed by experienced surgeons, but larger studies with long-term follow-up are needed to validate these findings.