Analysis of the Impact of the Learning Curve on the Safety Outcome of the Totally Robotic-Assisted Biliopancreatic Diversion with Duodenal Switch: a Single-Institution Observational Study.

Introduction: Totally robotic-assisted biliopancreatic diversion with duodenal switch (BPD/DS) learning curve has been described to be longer at approximately 50 cases, at which point operative time and complications rate decrease and tend to stabilize. This study aimed to form an analysis of the impact of the learning curve on the safety outcomes of the totally robotic-assisted BPD/DS.

Methods: A retrospective review of patients who underwent primary totally robotic-assisted BPD/DS by one of our certified bariatric and metabolic surgeon member of our institution was performed. The patients were classified into two groups, the learning stage group (first 50 cases) and the mastery stage group. Differences in operative time in minutes and postoperative outcomes were analyzed.

Results: Two hundred seventy-six patients were included. The operative time and the postoperative length of stay were significantly higher in the learning stage group (173.8 ± 35.8 min vs. 139.2 ± 30.2 min, p= 0.0001; 3.4 ± 1.4 days vs. 2.6 ± 0.9 days, p= 0.0002). The overall leakage rate was significantly higher in the learning stage group (8% vs. 0.4%, p= 0.0001). The global rate of complications for the learning stage group was 14%, and for the mastery stage group was 6.6% (p= 0.08).

Conclusions: After the first 50 cases, the operative time, the length of stay, and the overall rate of complications decreased, being especially significant the decrease in the duodeno-ileal anastomosis leakage rate after reaching the learning curve.