Abstract
PURPOSE: Early detection of hyperlipidemic acute pancreatitis (HLAP) with exacerbation tendency is crucial for clinical decision-making and improving prognosis. The aim of this study was to establish a reliable model for the early prediction of HLAP severity. PATIENTS AND METHODS: A total of 225 patients with first-episode HLAP who were admitted to Fujian Medical University Union Hospital from June 2012 to June 2023 were included. Patients were divided into mild acute pancreatitis (MAP) or moderate-severe acute pancreatitis and severe acute pancreatitis (MSAP+SAP) groups. Independent predictors for progression to MSAP or SAP were identified through univariate analysis and least absolute shrinkage and selection operator regression. A nomogram was established through multivariate logistic regression analysis to predict this progression. The calibration, receiver operating characteristic(ROC), and clinical decision curves were employed to evaluate the model's consistency, differentiation, and clinical applicability. Clinical data of 93 patients with first-episode HLAP who were admitted to the First Affiliated Hospital of Fujian Medical University from October 2015 to October 2022 were collected for external validation. RESULTS: White blood cell count, lactate dehydrogenase, albumin, serum creatinine, serum calcium, D-Dimer were identified as independent predictors for progression to MSAP or SAP in patients with HLAP and used to establish a predictive nomogram. The internally verified Harrell consistency index (C-index) was 0.908 (95% CI 0.867-0.948) and the externally verified C-index was 0.950 (95% CI 0.910-0.990). The calibration, ROC, and clinical decision curves showed this nomogram's good predictive ability. CONCLUSION: We have established a nomogram that can help identify HLAP patients who are likely to develop MSAP or SAP at an early stage, with high discrimination and accuracy.