Laparoscopic versus open pancreatoduodenectomy: a meta-analysis of randomized controlled trials

Introduction: the evidence with regard to the benefit of laparoscopic surgery for pancreatoduodenectomy is conflicting. The aim of this meta-analysis was to compare the short-term outcomes in patients undergoing laparoscopic or open pancreatoduodenectomy via randomized controlled trial studies. Methods: PubMed, Embase and Cochrane Library databases were searched for studies addressing laparoscopic versus open pancreatoduodenectomy up to February 2019. Only randomized controlled trial studies were included. Results: three randomized controlled trial studies were identified, which included a total of 224 patients. Statistically significant differences were found with regard to estimated blood loss in favor of laparoscopic pancreatoduodenectomy (WMD, -150.9 ml; 95% CI, -167.61 to -134.18; p < 0.001) but with longer operative time (WMD, 97.66 min; 95% CI, 21.28 to 174.05; p = 0.01). No significant differences were found for severe postoperative complications (defined as Clavien-Dindo grade ≥ III complications), complication-related mortality within 90 days, blood transfusion requirements, length of hospital stay, postoperative pancreatic fistula, postpancreatectomy hemorrhage, bile leakage, delayed gastric emptying, surgical site infection, readmission rate, reoperation rate, harvested lymph nodes and R0 resection rate. Conclusions: the perioperative safety of laparoscopic pancreatoduodenectomy, which may have an advantage of lower estimated blood loss, is comparable to that of open pancreatoduodenectomy. Currently, a small volume of cases may be an important reason that affects the evaluation between laparoscopic and open pancreatoduodenectomy. Further evaluation of laparoscopic pancreatoduodenectomy will require large randomized control trials.