Establishment and Validation of a Dynamic Nomogram for Persistent Organ Failure in Acute Biliary Pancreatitis: A Retrospective Study.

Abstract

Purpose: The objective of this study was to create a predictive model for the onset of persistent organ failure (POF) in individuals suffering from acute biliary pancreatitis (ABP) by utilizing indicators observed within 24 hours of hospital admission. Early detection of high-risk POF patients is crucial for clinical decision-making.

Patients and methods: Clinical data and laboratory indicators within 24 hours of admission from ABP patients diagnosed at The First Affiliated Hospital of Wenzhou Medical University between January 1, 2016, and January 1, 2024 were collected and retrospectively analyzed. The Least Absolute Shrinkage and Selection Operator (LASSO) regression and multivariate logistic regression (stepwise regression) methods were employed to identify variables for constructing the prediction model. The prediction model's performance was evaluated using the area under the curve (AUC), calibration curve, and decision curve analysis (DCA). It was compared with other scoring systems such as SIRS, BISAP, APACHE II, CTSI, and MCTSI. Additionally, a web-based calculator was created to simplify the calculation process.

Results: Out of 324 ABP patients, 25 developed POF. Initial screening identified 18 variables; through LASSO regression and multivariable logistic regression analysis, five variables including BMI, Hb, ALB, Ca, and LIP were determined as independent predictors of POF. According to these factors to build prediction model, draw the nomogram. The AUC's receiver operating characteristic curve analysis demonstrated a significantly higher value in comparison to other scoring systems. Calibration curve and DCA show that the established model to predict the accuracy of POF is higher, clinical decision of net benefit is also higher. A network calculator utilizing this predictive model was developed.

Conclusion: A predictive model incorporating five risk indicators has been established exhibiting high discriminatory power and accuracy which aids in early identification of ABP patients at risk for developing POF. This holds significant value in guiding clinical decision-making.