Postoperative Day 3 Serum Lipase Level as an Easy and Reliable Predictor of Clinically Relevant Pancreatic Fistula after Pancreaticoduodenectomy

Abstract

INTRODUCTION: Postoperative pancreatic fistula (POPF) represents the most common and feared complication after pancreatoduodenectomy (PD). POPF early detection could improve patient management and outcomes. Postoperative hyperlipasemia is used for the diagnosis of postoperative pancreatitis, but its role in predicting POPF needs to be proven. The aim of this study is to investigate the role of serum lipase (SL) in predicting POPF. METHODS: A prospective maintained database of patients who underwent PD between January 2012 and January 2020 was retrospectively analyzed. SL was measured on postoperative day (POD) 1, 2, and 3. The study groups were defined according to postoperative SL levels. Receiving operator characteristic (ROC) analysis was performed to assess the optimal cutoff of SL in predicting clinically relevant POPF (CR-POPF). The analysis was carried out using binary logistic regression analysis. RESULTS: One hundred fifteen patients fulfilled the inclusion criteria. The mean SL on POD 1, 2, and 3 were significantly higher in CR-POPF group (p < 0.01). The ROC analysis showed that, in the presence of SL ≥177 U/L on POD 1, 2, and 3, the negative predictive values for CR-POPF were 92.5%, 92.7% and 93.3%, respectively (Fig. 1). Notably, POD 3 SL ≥177 U/L was associated with 6-times higher odds of CR-POPF (odds ratio 6.89; p = 0.009).CONCLUSION: Postoperative SL on POD 1, 2, and 3 may be used to exclude the presence of CR-POPF. Notably, SL on POD 3 may represent an easy and reliable predictor of CR-POPF.