Background : Although many formulas for predicting postoperative pancreatic fistula (POPF) have been introduced, POPF is generally predicted during pancreatic surgery due to pancreatic texture. This study aimed to suggest a fistula risk score (FRS) without pancreatic texture parameter that can be used before pancreatectomy using computed tomography (CT) body composition.

Methods : Data from 545 patients who underwent pancreatoduodenectomy for malignant disease between January 2008 and December 2019 were retrospectively reviewed. The odd ratio regarding clinically relevant (CR)-POPF according to CT body composition parameters was calculated with other previously known related factor for. Finally, using the selected parameters, preoperative fistula risk score (p-FRS) was suggested, and the results were compared with a previously reported updated alternative FRS (ua-FRS).

Results : Hounsfield unit (HU), pancreatic duct (PD) size, visceral fat area (VFA), and sarcopenia were significantly correlated with CR-POPF. (OR – HU: 1.05 [1.01-1.08], p=0.005; PD size: 0.76 [0.62-0.94], p=0.012; VFA: 1.01 [1.00-1.01], p<0.001; Sarcopenia: 1.78 [1.03-3.08], p=0.039) The p-FRS (p = exp(-4.052 + 0.045 [HU] - 0.272 [PD size] + 0.007 [VFA] + 0.577 [sarcopenia])/1 + exp(-4.052 + 0.045 [HU] - 0.272 [PD size] + 0.007 [VFA] + 0.577 [sarcopenia])) showed acceptable results in predicting CR-POPF (AUC = 0.712, p < 0.001). There was no statistically significant difference in DeLong’s test compared with the ua-FRS (Z=-0.352, p=0.725).

Conclusions : p-FRS could predict the risk of POPF before pancreatic surgery.