Potential Performance Comparison of TyG-BMI and Inflammatory-Lipid Indices for Predicting Incident MASLD in Patients with Chronic Hepatitis B

Abstract

Background: Noninvasive predictors of incident metabolic dysfunction-associated steatotic liver disease (MASLD) in patients with chronic hepatitis B (CHB) are limited. We evaluated the predictive value of the triglyceride-glucose body mass index (TyG-BMI) compared with that of inflammatory-lipid indices. Objectives: This study aimed to compare the predictive value of TyG-BMI with that of six metabolic and inflammatory-lipid indices for incident MASLD in patients with CHB and to assess whether predictive performance varied according to antiviral treatment, HBeAg status, and HBV DNA load. Methods: This retrospective cohort study included 131 patients with CHB and no baseline MASLD. Incident MASLD occurred in 41 patients (31.3%). Cox regression, time-dependent receiver operating characteristic (ROC) analysis, and decision curve analysis were used to evaluate predictive performance. Results: TyG-BMI was the only one of the seven indices to reach statistical significance in the fully adjusted Cox model (HR = 1.021, P = 0.027). The highest TyG-BMI quartile was associated with a 4.12-fold higher risk than the lowest quartile (P = 0.027). TyG-BMI achieved time-dependent area under the curve (AUC) values above 0.70 and demonstrated numerically higher predictive performance than the other indices. Numerical differences in predictive performance were observed across virological subgroups. Calibration and decision curves indicated good model performance. Conclusions: In this exploratory cohort, TyG-BMI showed a numerically stronger association with incident MASLD than individual metabolic and inflammatory-lipid composite indices. External prospective validation is needed before clinical application.