The clinicalvalue of pancreatic fistula risk scoring system in pancreaticoduodenectomy / 中华肝胆外科杂志

ABSTRACT

Objective To explore the value of pancreatic fistula risk score (FRS) in pancreaticoduodenectomy (PD).Methods A total of 491 patients undergoing complete pancreaticoduodenectomy in the Department of Biliary Pancreatic Surgery,Affiliated Provincial Hospital of Anhui Medical University from January 1st,2012 to January 1st,2017 were included in this retrospective study.Among them,288 males and 203 females.Univariate and multivariate logistic regression analysis were used to determine the independent risk factors for pancreatic fistula.Receiver operating characteristic (ROC) curve was used to prospectively analyze the value of the FRS in predicting the incidence of postoperative pancreatic fistula in 90 PD patients.Results A total of 90 patients (18.3％) of 491 patients with PD developed pancreatic fistula,including 56 cases (11.4％) of grade A pancreatic fistula,26 cases (5.3％) of grade B pancreatic fistula,and 8 cases (1.6％) of grade C pancreatic fistula.Univariate and multivariate logistic regression analysis found that gender male,main pancreatic duct diameter ≤ 3 mm were independent risk factor for pancreatic fistula after PD.FRS scores without pancreatic fistula risk group (0 points) clinical pancreatic fistula incidence rate 0,low pancreatic fistula risk group (1 ～ 2 points) clinical pancreatic fistula incidence rate 5.4％,moderate pancreatic risk group (3 ～ 6 points) The incidence of clinical pancreatic fistula was 12.1％,and the incidence of clinical pancreatic fistula was 60％ in the high pancreatic risk group (7 ～ 10points).The FRS system prospectively predicts the sensitivity of pancreatic fistula after surgery for PD was 80％,specificity was 70.5％,and the area under the curve was 0.852 (95％ CI0.740 ～ 0.963).Conclusions The FRS has a certain clinical value in predicting pancreatic fistula after PD,which can predict and interfere with pancreatic fistula risk in early postoperative surgery.