| Objective:To evaluate the postoperative safety of pancreaticojejunostomy(PJ)and pancreaticogastrostomy(PG)in pancreaticoduodenectomy.The main outcome indicators were postoperative pancreatic fistula,postoperative hemorrhage,intra-abdominal infection,biliary leakage,enteric fistula,postoperative mortality,reoperation rate and readmission rate.The secondary outcome indexes were postoperative pancreatitis,delayed gastric emptying,peptic ulcer and total complication score.Methods:We searched the databases of Pub Med,Web of Science,EMBASE,Cochrane,CNKI,Wanfang and VIP,all randomized controlled trials(RCTs)about the comparison of the complications after pancreaticojejunostomy and pancreaticogastrostomy in pancreaticoduodenectomy in Chinese or English were included.Literature screening and data extraction were carried out by two independent researchers,data analysis was carried out by Rev Man5.3 and Stata13.0 software,and the quality of the included randomized controlled trials was evaluated by Cochrane intervention system review manual.Result:A total of 12 RCT trials were included in 9 countries,of which 936 patients received PG and 881 received PJ.Main outcomes: the total incidence of postoperative pancreatic fistula in PG was significantly lower than that in PJ [RR=0.72,95%CI(0.58-0.88),P<0.05].In the study included in ISGPF2005 criteria,the total incidence of postoperative pancreatic fistula in PG may be lower than that in PJ [RR=0.70,95%CI(0.49-0.99),P =0.05].However,there was no significant difference between PG and PJ in clinically relevant postoperative pancreatic fistula(CR-POPF)[RR = 0.69,95% CI(0.42-1.12),P> 0.05] and grade C postoperative pancreatic fistula [RR = 0.93,95% CI(0.39-2.23),P> 0.05].The total incidence of postoperative hemorrhage in PG was significantly higher than that in PJ [RR=1.42,95%CI(1.08-1.87),P<0.05].But there was no significant difference in clinically relevant postoperative hemorrhage(CR-PPH)[RR = 1.12,95% CI(0.73-1.71),P> 0.05] and grade C postoperative hemorrhage [RR = 1.17,95% CI(0.62-2.23),P> 0.05].The incidence of postoperative bile leakage in PG group was significantly lower than that in PJ group [RR=0.55,95%CI(0.32-0.95),P<0.05].The other major outcomes as follow between PG and PJ had no significant difference: Intra-abdominal infection [RR=0.90,95%CI(0.66-1.22),P>0.05],enteric fistula [RR=0.86,95%CI(0.41-1.80),P=0.05],Postoperative hospital mortality [RR=0.95,95%CI(0.57-1.59),P>0.05],Mortality within 90 days after operation [RR=1.70,95%CI(0.80-3.58),P>0.05].There was no significant difference between PG and PJ in all secondary outcomes: Total incidence of delayed gastric emptying [RR=1.02,95%CI(0.85-1.23),P>0.05],postoperative pancreatitis [RR=0.96,95%CI(0.63-1.45),P>0.05],peptic ulcer [RR=0.72,95%CI(0.19-2.66),P>0.05],Total postoperative complication score ? III [RR=1.34,95%CI(0.80-2.24),P>0.05].There was no publication bias found in the items(POPF,PPH,DGE)that included by more than 10 studies.Subgroup analysis:(1)CR-POPF: There was no significant difference in PG and PJ between the "dunking PG & duct-mucous PJ” group [RR=0.72,95%CI(0.29-1.78),P>0.05] and the "anastomotic procedure uncertainty" group [RR=0.64,95%CI(0.32-1.30),P>0.05].(2)PPH: PJ in ISGPS group was better than that in PG [RR=1.70,95%CI(1.18-2.45),P<0.05],but there was no significant difference in non-ISGPS group [RR=1.12,95%CI(0.73-1.71),P>0.05].(3)DGE: There was no significant difference between PG and PJ neither in ISGPS group [RR=1.12,95%CI(0.85-1.48),P>0.05] nor non-ISGPS group [RR=0.78,95%CI(0.40-1.51),P>0.05].The sensitivity analysis of CR-POPF suggests that the conclusion may be unstable.Conclusion:Untill now,it is possibly superior to choose PG than PJ in pancreaticoduodenectomy. |
PDF Full Text Request