Abstract:Objective To investigate the predictive value of human neutrophil lipocalin（HNL），procalcitonin（PCT），and C - reactive protein（CRP）for peripancreatic infection in patients with severe acute pancreatitis（SAP）, and to provide evidence-based reference for early anti-infective treatment of SAP complicated with peripancreatic infection.Methods A total of 76 SAP patients admitted to Jilin Province FAW General Hospital from October 2020 to February 2024 were enrolled，and divided into a peripancreatic infection group（n=28）and a non-peripancreatic infection group（n=48）. Serum HNL was measured by enzyme-linked immunosorbent assay，PCT by chemiluminescence，and CRP by latex immunoturbidimetry，and the differences of the three indicators between the two groups were compared. Logistic regression models were used to explore the association between serum biomarkers（HNL，PCT，CRP）and peripancreatic infection. Receiver operating characteristic（ROC）curve was used to evaluate diagnostic efficacy. Results Median levels of HNL，PCT，and CRP in the peripancreatic infection group were significantly higher than those in the non -peripancreatic infection group（P<0.01）. Multivariate logistic regression analysis revealed that high HNL level（OR=1.031，95%CI：1.013-1.049）and high PCT level（OR=1.451，95%CI：1.040-2.022）were independent risk factors for peripancreatic infection in SAP（P<0.05）. The ROC results indicated that the area under the curve（AUC）values for predicting SAP with peripancreatic infection using HNL，PCT，and the combination of both indicators were 0.893，0.872，and 0.935，respectively. The sensitivity，specificity，positive predictive value，and negative predictive value of the combined prediction were 82.14%，95.83%，92.00%，and 90.20%，respectively. Conclusion The construction of a combined diagnostic model of HNL and PCT is helpful for the early prediction of SAP combined with peripancreatic infection，providing a key time window for clinical intervention.