Optimal Timing of Serum Lipase for Early Prediction of Clinically Relevant Pancreatic Fistula After Pancreaticoduodenectomy.

Background: Pancreatic fistula remains the leading cause of morbidity after pancreaticoduodenectomy. However, its diagnosis is frequently delayed. We evaluate the predictive accuracy of serum lipase in detecting a clinically relevant pancreatic fistula (CR-POPF) and assess whether its combination with procalcitonin (PCT) improves risk stratification. Methods: A single-center study was conducted that included 115 patients who underwent pancreaticoduodenectomy between 2015 and 2019. Serum lipase levels were measured preoperatively and on postoperative day (POD) 1, 2, and 3. Elevated lipase was defined as >3× the institutional upper limit of normal (>177 U/L). Results: Serum lipase levels were higher in patients who developed CR-POPF at all postoperative days (p < 0.01). The area under the curve (AUC) increased from POD 1 (0.76) to POD 2 (0.82) and remained stable at POD 3 (0.82), while specificity improved from 52.7% to 89.2%, with consistently high negative predictive values (92.5-93.3%). Elevated lipase on POD 3 predicts CR-POPF (OR 6.89, 95% CI 1.61-29.53; p = 0.009). The combination of elevated lipase, regardless of the preoperative value, and PCT ≥ 1.5 ng/mL on POD 3 achieved full specificity. Conclusions: Early postoperative serum lipase levels are useful to exclude CR-POPF after pancreaticoduodenectomy. POD 3 provided a proper diagnostic value. The combination of serum lipase and PCT on POD 3 may further improve accuracy, enabling tailored postoperative surveillance.