International validation of the distal pancreatectomy fistula risk score: evaluation in minimally invasive and open surgery.

Background Postoperative pancreatic fistula (POPF) remains the most severe complication following distal pancreatectomy (DP). The preoperative distal fistula risk score (D-FRS) was introduced to predict the POPF risk. The aim of this study was to externally validate the D-FRS in an international expert center cohort. Methods This international, multicenter, retrospective cohort study included open and minimally invasive DP for benign and malignant lesions performed from 01/2014 until 12/2023 in 12 centres from 6 countries, that each performed more than 50 pancreatectomies annually. The D-FRS was calculated from pancreatic thickness and duct size. Predicted and actual POPF were compared using sensitivity, specificity and area under the curve (AUC). Results A total of 778 patients underwent DP of whom 284 (39%) underwent robotic, 278 (38%) open and 165 (23%) laparoscopic DP. The rate of POPF was 32%. The sensitivity, specificity and AUC of the D-FRS for the overall cohort was 32%, 63% and 48% (95% CI 44-51%), respectively. The AUC for open, laparoscopic and robotic DP was 54% (48-60%), 47% (39-55%) and 45% (39-50%), respectively. For neoadjuvant therapy naïve patients the AUC was 52.3%. On multivariate analysis POPF was associated with body mass index (odds ratio 1.04 (95% CI 1.01-1.07)), protective factors were neoadjuvant therapy (OR 0.54 (0.22-0.94)) and the robotic approach (OR 0.64 (0.42-0.97)). Conclusions The preoperative D-FRS showed insufficient discrimination to identify patients who develop POPF after DP irrespective of the surgical approach. Novel preoperative POPF risk scores are needed, considering the standard minimally invasive approach and the widespread use of neoadjuvant therapy.