Robotic Pancreatic Surgery - Learning Curve and Implementation

Minimally invasive resection techniques for the treatment of various pathologies of the pancreas are potentially advantageous for the treated patients in terms of restitution time and postoperative morbidity, but are a technical challenge for the responsible surgeon. The introduction of robotic assistance in visceral surgery offers a possibility for further distribution of minimally invasive procedures in pancreatic surgery. The aim of this study was to examine the possibilities for developing robotic pancreatic surgery in Germany. The data are based on the quality reports of the hospitals for the years 2015-2019 combined with a selective literature search. The number of quality reports available decreased from 1635 to 1594 between 2015 and 2019. A median of 96 clinics performed 11-20, 56 clinics 21-50 and 15 clinics more than 50 pancreaticoduodenectomies. For distal resections, there were 35 clinics with 11-20, 14 clinics with 21-50 and two clinics with more than 50 procedures. In relation to all clinics with at least five distal resections per year, minimally invasive procedures were performed at only 29 clinics; a ratio to laparoscopic left resections of over 50% was reported in only seven clinics. According to the literature, the learning curves for robotic pancreatic distal resection and pancreaticoduodenectomy diverge. While the learning curve for robotic distal resection is completed after around 20 procedures, the learning curve for robotic pancreaticoduodenectomy has several plateaus, which are reached after around 30, 100 and 250 procedures. Due to the decentralised structure of pancreatic surgery in Germany, a nationwide introduction of robotic pancreatic surgery is unlikely. The routine use of robotic pancreaticoduodenectomy will probably be restricted to high volume centres in the foreseeable future.