Prediction Models of Severity in Acute Biliary Pancreatitis

• DOI: 10.3390/diagnostics15020126
• Publication Date: 2025-01-07T12:42:19Z
• Authors: Iulia Ratiu; Renata Bende; Camelia Nica; Oana Budii; Calin Burciu; Andreea Barbulescu; Tudor Moga; Bogdan Miutescu; Roxana Sirli; Mirela Danila; Alina Popescu; Felix Bende
• Keywords: acute biliary pancreatitis; Medicine (General); R5-920; comparison; severity prediction; multiparametric models; Article

Background: Acute pancreatitis is a common condition with variable prognosis. While the overall mortality rate of acute relatively low, ranging between 3 and 5% in most cases, severe forms can result significantly higher morbidity mortality. Therefore, early risk assessment crucial for optimizing management treatment. The aim present study wasto compare simple prognostic markers identify best predictors severity patients pancreatitis. Material Methods: A retrospective analysis was carried outon 108 admitted our center during one year biliary stratified based on revised Atlanta criteria. Results: subjects (mean age 60.1 ± 18.6, 65.7% females) diagnosed were included. Based criteria, 59.3% (64/108) classified as having mild pancreatitis, 35.2% (38/108) moderate-severe 5.5% (6/108) In univariate analysis, following parameterswere associatedwith at least form pancreatitis: Balthazar score, fasting blood glucose (mg/dL), modified CTSI CRP values 48 h, BISAP score admission, Ranson duration hospitalization (days), presence leukocytosis (×1000/µL) (all p < 0.05).BISAP admission (AUC-0.91), levels h (AUC-0.92), mCTSI (AUC-0.94), (AUC-0.93) had highest area under curve (AUC) predicting multivariate model including independent parameters predictive (p 0.0001), = 0.0082), 0.0091), respectively. showed slightly AUC compared to (AUC-0.96). Conclusions: use multiparametric prediction increase accuracy