Pancreaticojejunostomy Versus Pancreaticogastrostomy In Postoperative Pancreatic Fistula After Pancreaticoduodenectomy:A Single Centre Study Using The ISGPS 2016 Definition

ObjectivesPancreaticoduodenectomy(PD)is the gold standard to treat periampullary carcinoma.The main problem of PD is the incidence of complications such as postoperative pancreatic fistula(POPF)remains high,which significantly affects the patients’ postoperative recovery.It has been widely debated that whether pancreaticogastrostomy(PG)is superior to pancreaticojejunostomy(PJ)in the incidence of POPF.Previous studies had obvious defects in the diagnosis of POPF.The International Study Group Of Pancreatic Surgery updated the new definition of POPF in 2016.Our study is the first to use the new definition to compare PJ and PG in postoperative pancreatic fistula after PD.Methods210 patients underwent PD in our centre between January 2012 and December 2012,of which 136 cases received PJ and 74 cases received PG The main outcome is the incedence of postoperative pancreatic fistula.The secondary outcomes included other complications and clinical outcome such as reoperation and morality.Also,we verified the pancreatic fistula risk score system(FRS)in predicting POPF.Risk factors of postoperative pancreatic fistula were examined using regression analysis.ResultsOur trial included 210 patients.The entire rate of POPF was 16.2%.There was no significant difference in the rate of POPF between PJ and PG(16.9%vs 14.9%,p=0.701).The delayed gastric emptying(DGE)was less severe in PG(p=0.012),while the overall incidence had no significant difference between PJ and PG(18.4%vs 14.9%,p=0.518).Also,PG had less bile leakage than PJ(18.9%vs 33.8%,p=0.022).However,the main clinical outcomes didn’t present any significant difference between PJ and PG,such as the mortality(0.7%vs 0),morbidity(45.6%vs 40.5%,p=0.481)and the rate of severe postoperative complications(11.0%vs 13.5%,p=0.595).The incidence of POPF was 0 in negligible risk patients,5%in low risk patients,12%in intermediate patients and 48%in high risk patients.Only small duct(<3mm)predicted the occurrence of pancreatic fistula using univariate logistic regression(odds ratio 4.31,p<0.001)and multivariable logistic regression(odds ratio 4.93,p<0.001).ConclusionPJ and PG are both safe and optional pancreatic anastomosis after PD.The rate and severity of POPF after PJ versus PG was not different.Although PG had lighter severity of DGE and less bile leakage,the clinical outcomes remained parallel between the two anastomosis methods.The Fistula Risk Sore is an effective tool in predicting risk of POPF.Small duct(<3mm)is the only independent risk factor of POPF.