Fistula Risk Score for Auditing Pancreatoduodenectomy

Abstract

Objective: To develop a fistula risk score for auditing, to be able to compare postoperative pancreatic fistula (POPF) after pancreatoduodenectomy among hospitals. Background: For proper comparisons of outcomes in surgical audits, case-mix variation should be accounted for. Methods: This study included consecutive patients after pancreatoduodenectomy from the mandatory nationwide Dutch Pancreatic Cancer Audit. Derivation of the score was performed with the data from 2014 to 2018 and validation with 2019 to 2020 data. The primary endpoint of the study was POPF (grade B or C). Multivariable logistic regression analysis was performed for case-mix adjustment of known risk factors. Results: In the derivation cohort, 3271 patients were included, of whom 479 (14.6%) developed POPF. Male sex [odds ratio (OR)=1.34; 95% confidence interval (CI): 1.09–1.66], higher body mass index (OR=1.07; 95% CI: 1.05–1.10), a final diagnosis other than pancreatic ductal adenocarcinoma/pancreatitis (OR=2.41; 95% CI: 1.90–3.06), and a smaller duct diameter (OR=1.43/mm decrease; 95% CI: 1.32–1.55) were independently associated with POPF. Diabetes mellitus (OR=0.73; 95% CI: 0.55–0.98) was independently associated with a decreased risk of POPF. Model discrimination was good with a C-statistic of 0.73 in the derivation cohort and 0.75 in the validation cohort (n=913). Hospitals differed in particular in the proportion of pancreatic ductal adenocarcinoma/pancreatitis patients, ranging from 36.0% to 58.1%. The observed POPF risk per center ranged from 2.9% to 25.4%. The expected POPF rate based on the 5 risk factors ranged from 11.6% to 18.0% among hospitals. Conclusions: The auditing fistula risk score was successful in case-mix adjustment and enables fair comparisons of POPF rates among hospitals.