Pablo Achurra, MD, Marcel Sanhueza, MD, Rolando Rebolledo, MD, Juan Francisco Guerra, MD, Fernando Pimentel, Jorge Martinez, MD, Julian Varas, Nicolas Jarufe. Experimental Surgery and Simulation Center, Department of Digestive Surgery, Pontificia Universidad Catolica de Chile
Introduction: Laparoscopic pancreaticoduodenectomy (LPD) is currently a feasible option in selected patients at high volume centers with available expertise. The long learning curve and associated high morbidity is mainly explained by the technical difficulty of the reconstruction, specifically the pancreaticojejunum anastomosis. Simulation training has demonstrated to reduce the learning curve of advanced laparoscopy.
Objective: To present a novel simulation training program for the acquisition of expertise in laparoscopic duct-to-mucosa pancreaticojejunostomy.
Methods: Experimental study of a training program based on sessions of ascending difficulty to perform a Blumgart laparoscopic duct-to-mucosa pancreaticojejunostomy in a novel ex-vivo based model. The program consists of 3 stages of ascending difficulty, wide (8mm), medium (5mm) and thin (less than 2mm) Wirsung duct diameter. Time, global rating scale (GRS), permeability and leakage of the anastomosis was recorded after each session. To carry on to the next stage the surgeon had to perform a permeable anastomosis with no leak, in less than 30min and with a GRS>20 points (Max: 25 points).
The trainees were assessed in a thin-duct model before and after the training program to assess the effects of training. After the simulation the surgeons performed a LPD in the operating room (OR).
Results: Two HPB surgeons with 10 years of experience in open and laparoscopic procedures, but minimal experience in LPD (less than 2 procedures) underwent the simulation training program to acquire expertise in laparoscopic pancreaticojejunostomy reconstruction.
In the initial assessment, procedural time was 52 and 55 min, GRS were 14 and 15 respectively and both surgeons had anastomosis leakage. The number of sessions to complete each stage were 3-4 in stage one, 4-4 in stage two and 3-2 in stage three for each surgeon respectively (10 sessions in total). In the final assessment, procedural time was 25 and 23 min, and GRS were 24 and 23. Both surgeons completed the anastomosis without leak and good permeability.
At the moment of submitting this article, only one of the surgeons had performed two LPD in the OR after completing the program (6mm and 2mm pancreatic duct). The time to perform the anastomosis was 50 and 40 min; GRS 22 and 23 respectively. The patients had no pancreatic fistula in the postoperative course.
Conclusion: Simulation may reduce the learning curve of complex minimally invasive HPB procedures like LPD duct to mucosa anastomosis. The skills acquired in the simulated ex-vivo model were transferred to the operating room.