External Validation Of Alternative Fistula Risk Core(a-FRS) For Predicting Pancreatic Fistula After Pancreatoduodenectomy

BackgroundThe incidence of pancreatic fistula after pancreaticoduodenectomy remained high.The fistula risk score(FRS)for predicting risk of clinically relevant POPF(CR-POPF)after pancreaticoduodenectomy has been validated widely.Alternative fistula risk(a-FRS)was developed to refine the FRS.ObjectiveThe aim of this study was to validate the utilities of the recently proposed a-FRS and updated a-FRS(ua-FRS),using data from a high-volume pancreatic center.MethodsThis retrospective analysis included data of 370 consecutive patients undergoing PD(317 open PD[OPD]and 53 laparoscopic PD[LPD])during March 2012-May 2018.The predictive abilities of FRS,a-FRS and ua-FRS were compared by Pearson correlation coefficients and receiver-operating characteristic(ROC)curve analysis.Risk factors for CR-POPF were evaluated by multivariate regression analysis.ResultOf the 370 patients,80(21.62%)developed CR-POPF:64(80%)in the OPD cohort and 16(20%)in the LPD cohort.The incidences of CR-POPF in patients classified as low risk,intermediate risk,and high risk by a-FRS were 5.88%,24.38%,and 57.69%,respectively(R~2=0.97).The incidences of CR-POPF in patients classified as negligible risk,low risk,intermediate risk,and high-risk by FRS were 0%,8.62%,21.51%,and 52.50%,respectively(R~2=0.92).Multivariate logistic regression analysis showed increasing BMI(p=0.003),soft gland texture(p<0.001)and thin pancreatic duct diameter(p<0.001)were independent risk factors for CR-POPF.The area under the ROC curve(AUC)was 0.74 for a-FRS vs.0.70 for FRS.The a-FRS performed better than FRS for prediction of CR-POPF in OPD patients(AUC:0.74 vs.0.70)and slightly lower than FRS in LPD patients(AUC:0.70 vs.0.72).However,the AUC of ua-FRS was 0.78 in LPD patients,which was better than FRS and a-FRS.ConclusionThe a-FRS appears to be an accurate and convenient tool for predicting occurrence of CR-POPF after PD.