Abstract: Objective To evaluate the safety of minimally invasive surgery and open surgery on the prognosis of pancreaticoduodenectomy. Methods Database including CNKI, WanFang and PubMed (from January 1990 to December 2014) were searched to collect articles on comparing the safety of minimally invasive pancreaticoduodenectomy with open pancreaticoduodenectomy. The Cochrane network RevMan 5.0 software was used for Meta - analysis. Results A total of 7 articles with 557 patients were included. The Meta-analysis showed significant differences in operative time, estimated blood loss and length of hospital stays between minimally invasive pancreaticoduodenectomy group and open pancreaticoduodenectomy group (P< 0.05). There was no significant difference in pancreatic fistula, delayed gastric empty and mortality between two groups (P> 0.05). Conclusion Minimally invasive pancreaticoduodenectomy is as safe and feasible as open pancreaticoduodenectomy, and it has the advantages of less bleeding, milder pain, faster recovery and shorter hospital stay.