Nomogram combining inflammatory biomarkers for early prediction of organ failure in acute pancreatitis.

BACKGROUND

Organ failure (OF) is the main determinant of prognosis in acute pancreatitis. In recent years, promising biomarkers such as the C-reactive protein (CRP)/albumin ratio, the combined CRP and neutrophil-to-lymphocyte ratio score (C-NLR score), and waist circumference have shown potential for early risk prediction. This study aimed to develop a predictive nomogram integrating these markers for the early identification of OF.

METHODS

A retrospective analysis was conducted using a prospectively collected database of acute pancreatitis patients admitted to Hospital Clínico Universitario de Valladolid (Spain) between March 2014 and January 2023. Predictive variables for OF were identified through a combined approach of automated feature selection (Boruta algorithm) and multivariable logistic regression. The performance of the model was evaluated using the area under the ROC curve (AUC), calibration plots, and decision curve analysis.

RESULTS

A total of 830 patients were included, of whom 98 (11.8%) developed OF. Feature selection and multivariable analysis identified age, waist circumference, glucose, blood urea nitrogen, CRP/albumin ratio, and C-NLR score as independent predictors of OF, which were subsequently incorporated into the model. The nomogram showed good discrimination (AUC 0.81, 95% CI 0.76-0.85), accurate calibration with minimal underestimation at higher predicted probabilities, and favorable clinical utility.

CONCLUSIONS

The proposed nomogram, which integrates emerging biomarkers such as waist circumference, the CRP/albumin ratio, and the C-NLR score, could be used as complementary tool for the early prediction of OF in acute pancreatitis.