Objective: To compare perioperative and oncologic outcomes of pure (totally) laparoscopic pancreaticoduodenectomy (TLPD) or robot-assisted pancreaticoduodenectomy (RAPD) with those of conventional open pancreaticoduodenectomy (OPD).
Methods: A systematic literature search was performed using PubMed, EMBASE, and Cochrane library databases. Studies comparing TLPD with OPD and RAPD with OPD were included; only original studies reporting more than 10 cases for each technique were included. Studies were combined using a random-effects model to report heterogeneous data, or a fixed-effects model was applied.
Results: TLPD involved longer operative time (weighted mean difference [WMD]: 116.85 min; 95% confidence interval [CI] 54.53-179.17) and significantly shorter postoperative hospital stay (WMD: -3.68 days; 95% CI -4.65 to -2.71). Overall morbidity and postoperative pancreatic fistula were not significantly different between TLPD and OPD. RAPD was associated with a longer operative time, less intraoperative blood loss, and shorter hospital stay. Oncologic outcomes were not significantly different among the procedure types.
Conclusions: Compared to OPD, TLPD and RAPD were feasible and oncologically safe procedures. However, there are no prospective studies, and the majority of the studies on TLPD and RAPD have remained in the early training phase. In addition to randomized controlled trials or prospective studies, new data from the late training phase of learning experiences should also be analyzed.
Keywords: Laparoscopic pancreatectomy; Laparoscopic pancreaticoduodenectomy; Meta-analysis; Pancreas; Pancreaticoduodenectomy.