The robotic-assisted pancreaticoduodenectomy with vascular reconstruction: our personal experience

Abstract

Introduction: Firstly described in 2003, the widespread of robotic pancreaticoduodenectomy (PD) is still slow. The complexity of the procedure, the higher cost of this technology and the lack of a specific training confined this procedure to high-selected centres. We investigated the perioperative outcomes of our first 50 robotic PD including our initial experience with resection and reconstruction of superior mesenteric/portal vein (PD-SMV/PV) in order to evaluate the safety and feasibility of this approach. Methods: We performed a retrospective analysis from January 2012 to April 2017 of our prospectively maintained database on robotic pancreatic surgery. Results: A total of 50 patients during the time period evaluated underwent to robotic PD. We performed 45 robotic PD and 5 robotic PD-SMV/PV. The overall operative time was 425 min (290–550), with a reduction over the course of learning curve, while the median blood loss was 170 ml (30–700). We observed a low conversion rate 5/50 (10%) and an acceptable overall postoperative morbidity rate of 12/50 (24%). The pancreatic fistula rate was 8/50 (16 %), but only 3 (6%) clinically significant. We achieved a R0 resection rate in 46/50 (92%) of patients with a mean lymph nodes retrieved of 20.5 (17–25). The reoperation rate for all series was low 6/50 (12%) as well as the 90-day mortality rate 3/50 (6%). Conclusions: Despite our initial learning curve, the robotic approach seems to be safe and effective giving some potential advantages in terms of estimated blood loss and conversion rate, preserving a good oncologic adequacy.