Pancreaticoduodenectomy (PD) is a complex and high-skill demanding procedure often associated with significant morbidity and mortality. However, the results have improved over the past two decades. However, there is a paucity of research concerning the learning curve for PD. Our aim was to report the outcomes of 100 consecutive PDs representing a single surgeon's learning curve and to depict the factors that influenced the learning process. We reviewed the first 121 PDs performed at our academic center (2013-2019) by a single surgeon; 110 were PDs (5 laparoscopic and 105 open) and 11 were total PDs (1 laparoscopic and 10 open). Subsequent statistics was performed on the first 100 PDs, with attention paid to the learning curve and survival rate at 5 years. The data were analyzed comparing the first 50 cases (Group 1) to the last 50 cases (Group 2). The most frequent histopathological tumor type was pancreatic ductal adenocarcinoma (50%). A total of 39% of patients had preoperative biliary drainage and 45% presented with positive biliary cultures. The preferred reconstruction technique included pancreaticogastrostomy (99%), in situ hepaticojejunostomy (70%), and precolic gastro-jejunal anastomosis (88%). Postoperative complications included biliary fistula (1%), pancreatic fistula (8%), pancreatic stump bleeding (4%), and delayed gastric emptying (13%). The mean operative time decreased after the first 50 cases (p < 0.001) and blood loss after 60 cases (p = 0.046). R1 resections lowered after 25 cases (p = 0.025). Vascular resections (17%) did not influence the rate of complications (p = 0.8). The survival rate at 5 years for pancreatic adenocarcinoma was 32.93%. Outcomes improve as surgeon experience increases, with proper training being the most important factor for minimizing the impact of the learning curve over the postoperative complications. Analyzing the learning curve from the perspective of a single surgeon is mandatory for accurate statistical results and interpretation.
Read full abstract