Related Experiment Video
Updated: Jun 28, 2026
Robotic-assisted Lateral Pancreaticojejunostomy for Chronic Pancreatitis
Published on: September 5, 2025
Impact of learning phase on complications and oncological quality in robotic left-sided pancreatectomy: A multicenter
Abdullah K Malik1, Bhargav Chikkala2, Ryan Heslin3
1Department of HPB and Transplant Surgery, Freeman Hospital, The Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK. Electronic address: https://twitter.com/G_MarquisGravel.
Background:
Robotic left-sided pancreatectomy is increasingly adopted worldwide, yet the impact of the learning curve on clinical and oncological outcomes remains unclear. The Brescia consensus defines the first 21 cases as the competency phase, but real-world validation is limited. We evaluated whether the surgical learning phase influenced major morbidity and oncological quality following robotic left-sided pancreatectomy.
Methods:
This multicenter retrospective cohort study included consecutive patients undergoing robotic left-sided pancreatectomy across 10 high-volume centers in the United Kingdom, Europe, United States, and Australia (2014-2025). The learning phase was categorized as competency (cases 1-21) or proficiency (>21). The primary endpoint was major morbidity (Clavien-Dindo ≥IIIa). Mixed-effects logistic regression, restricted cubic splines, and risk-adjusted cumulative sum analyses evaluated predictors of morbidity and center-level performance. In the pancreatic ductal adenocarcinoma subgroup, factors associated with R1 margins were assessed.
Results:
Among 521 patients, 200 (38.4%) were in the competency and 321 (61.6%) in the proficiency phase. Major morbidity was comparable between phases (14.5% vs 14.0%; P = .981). Independent predictors of major morbidity included male sex (odds ratio, 1.76; P = .037), preoperative pancreatitis (odds ratio, 2.23, P = .031), and multivisceral resection (odds ratio, 2.29; P = .034). In the pancreatic ductal adenocarcinoma subgroup (n = 110), R1 rates were higher in the proficiency phase but nonsignificant (39.7% vs 21.6%; P = .092). Increasing operative time (odds ratio, 1.39/h; P = .004) and node-positive disease (odds ratio, 2.62; P = .047) predicted R1 resection. Spline and risk-adjusted cumulative sum analyses showed no early excess harm.
Conclusion:
Robotic left-sided pancreatectomy can be safely adopted within structured programs without increased early morbidity. However, evolving case complexity during the proficiency phase necessitates ongoing oncological vigilance and robust institutional oversight.
