OBJECTIVETo evaluate the pancreatic fistula affected by different type of pancreaticojejunostomy after pancreaticoduodenectomy.

METHODSElectronic databases PubMed, EMBase, COCHRANE Library, Wanfang, and VIP etc were used to search for randomized controlled trials or non randomized prospective controlled trials reported before September 2013 on clinical effects of pancreaticojejunostomy after pancreaticoduodenectomy. The statistical analysis was done by Review Manager 5.0.

RESULTSA total of 8 trials were included in this meta-analysis. The effects of duct-to-mucosa pancreaticojejunostomy (dmPJ) and invaginating pancreaticojejunostomy (iPJ) on postoperative complication in five studies were compared, and no statistical significance were found in postoperative pancreatic fistula (POPF) (M-H:OR = 0.77, 95% CI:0.35-1.69, P = 0.52), reoperation (M-H:OR = 1.38, 95% CI:0.64-2.95, P = 0.41) and mortality (M-H:OR = 1.15, 95% CI:0.42-3.13, P = 0.79) between dmPJ and iPJ. The effects of binding pancreaticojejunostomy (bPJ) and conventional pancreaticojejunostomy (cPJ) (including duct-to-mucosa pancreaticojejunostomy and invaginating pancreaticojejunostomy) on postoperative complication were compared, and no statistical significance were found in postoperative pancreatic fistula (POPF) (M-H:OR = 0.57, 95% CI = 0.28-1.17, P = 0.13) , reoperation (M-H:OR = 1.18, 95% CI = 0.48-2.92, P = 0.72) and mortality (M-H:OR = 0.74, 95% CI = 0.27-1.99, P = 0.55) between bPJ and cPJ.

CONCLUSIONThere are no significant differences between dmPJ and iPJ in pancreatic fistula reoperation and mortality, and there are also no significant differences between bPJ and cPJ.