Risk factors for significant hepatic fibrosis in alcoholic liver disease and construction of a nomogram prediction model.

backgroundSignificant fibrosis represents a critical stage in the progression of Alcoholic Liver Disease (ALD) to cirrhosis. Early identification of significant hepatic fibrosis is crucial for treatment and prognosis. This study aims to investigate risk factors for significant hepatic fibrosis in ALD patients via clinical data analysis and establish a nomogram for risk prediction.

methodsA retrospective study was conducted on 455 ALD patients from three hospitals, divided into training (n = 249), internal validation (n = 108), and external validation (n = 98) cohorts. Based on liver stiffness measurements by shear wave elastography (≥ 10.2 kPa), patients were divided into a significant hepatic fibrosis group and a non-significant fibrosis group. General information, laboratory indicators, and ultrasound measurements were collected. Independent risk factors were identified through stepwise multivariate logistic regression analysis. A logistic regression prediction model was constructed using the training set, and a nomogram was developed. Internal validation and external validation were performed. The model’s discriminative ability, calibration, and clinical utility were evaluated using ROC curve, calibration curves, and decision curve respectively. The ability of the prediction model in this study, APRI, and FIB-4 to diagnose significant alcoholic liver fibrosis was compared using the area under the receiver operating characteristic curve (AUC) index.

findingsStepwise multivariate logistic regression identified splenic area (SA), albumin (ALB), portal vein diameter (PV), prothrombin time (PT), and gamma - glutamyl transferase (GGT) as independent risk factors for significant hepatic fibrosis in ALD. A prediction model incorporating these five indicators was developed, and a nomogram was constructed. ROC curve demonstrated an AUC of 0.9440 for the training set, 0.9468 for internal validation set and 0.884 for external validation set. Calibration curves and decision curve indicated good calibration and clinical utility in three sets.The AUC of the prediction model in this study was higher than APRI and FIB-4 in the training set, internal verification set and external verification set.

conclusionsSA, PV, ALB, PT, and GGT were identified as independent risk factors for significant hepatic fibrosis. Utilizing these risk factors, we successfully developed a prediction model for significant hepatic fibrosis in ALD patients. Internal validation and external validation demonstrated that this model can be effectively applied in clinical decision-making.