Diagnostic Value of CHOL and BMI for Metabolic Dysfunction-Associated Fatty Liver Disease in Qinghai Province, China: A Cross-sectional Study
TL;DR
The nomogram classification model based on BMI and CHOL demonstrated favorable discrimination, calibration, and clinical utility and may serve as a simple screening tool for severe fatty liver, facilitate the early identification of high-risk individuals, and guide individualized interventions.
Abstract
Background: The severity of metabolic dysfunction-associated fatty liver disease (MAFLD) is closely associated with metabolic factors. Early identification of severe fatty liver is critical for timely clinical intervention; however, simple and intuitive tools for assessing the risk of severe fatty liver are lacking. Objectives: This study aimed to develop and validate a nomogram-based classification model for severe fatty liver disease using routine clinical indicators. Methods: A total of 159 patients with fatty liver were retrospectively enrolled, including 39 patients in the severe group and 119 in the mild-to-moderate group. Univariate analyses were performed to screen indicators associated with fatty liver severity, followed by binary logistic regression to identify independent correlating factors. A nomogram model was constructed based on these factors. Model discrimination was assessed using receiver operating characteristic (ROC) curve analysis, calibration accuracy using a calibration curve, and clinical utility using decision curve analysis (DCA). Results: Univariate analysis revealed significant differences in Body Mass Index (BMI), total cholesterol (CHOL), and the triglyceride-glucose (TyG) index between the 2 groups (P < 0.05). Multivariate logistic regression showed that BMI (odds ratio [OR] = 1.585; 95% CI, 1.342 - 1.871) and CHOL (OR = 1.647; 95% CI, 1.081 - 2.509) were independent predictors of severe fatty liver. The nomogram model based on these 2 factors achieved an area under the curve (AUC) of 0.888 in ROC analysis, with a sensitivity of 82% and a specificity of 80%. The calibration curve demonstrated high agreement between the predicted probability and the observed risk, and DCA indicated a high clinical net benefit within a threshold probability range of 5%-80%. Conclusions: The nomogram classification model based on BMI and CHOL demonstrated favorable discrimination, calibration, and clinical utility. It may serve as a simple screening tool for severe fatty liver, facilitate the early identification of high-risk individuals, and guide individualized interventions. However, owing to the small sample size, which included only 39 patients with severe fatty liver, and the cross-sectional design, the stability of the model requires further validation in large-sample, multicenter prospective cohorts.