Triglyceride-glucose and body shape index as metabolic-morphometric markers for metabolic dysfunction-associated steatotic liver disease identification across glycemic spectrum

BACKGROUNDMetabolic dysfunction-associated steatotic liver disease (MASLD) demands straightforward and dependable non-invasive biomarkers for early risk identification.The triglyceride-glucose index combined with a body shape index (TyG-ABSI), which jointly reflects insulin resistance and central adiposity, may enhance metabolic risk stratification; however, whether its diagnostic utility differs across the glycemic spectrum has not been fully clarified. AIMTo evaluate the utility of TyG-ABSI for identifying MASLD across normal glucose regulation (NGR), prediabetes (Pre-DM), and diabetes mellitus (DM). METHODSA retrospective cross-sectional study was conducted, including 7413 adults who underwent health examinations between January 2020 and March 2024.TyG-ABSI was calculated for all participants, and the study population were stratified into three glycemic subgroups: NGR (n = 2933), Pre-DM (n = 3549), and DM (n = 931).Boruta feature selection, multivariable logistic regression, restricted cubic spline analysis, and receiver operating characteristic (ROC) analysis were applied to examine its relationship with and discrimination for MASLD. RESULTSTyG-ABSI was significantly associated with MASLD status in the overall cohort and all glycemic subgroups.After full adjustment for confounders, each one-unit increment in TyG-ABSI was associated with an elevated odds of MASLD [odds ratio (OR) 3.36, 95% confidence interval (CI): 2.92-3.86].This significant association persisted in NGR (OR 5 / 46 3.11, 95%CI: 2.43-3.99),Pre-DM (OR 3.65, 95%CI: 2.97-4.50),and DM (OR 2.02, 95%CI: 1.39-2.92).The overall area under the ROC curve of TyG-ABSI for identifying MASLD was 0.809 (95%CI: 0.799-0.819),with corresponding area under the ROC curves of 0.802, 0.776, and 0.728 in the NGR, Pre-DM, and DM subgroups, respectively.Notably, the optimal TyG-ABSI cutoff values increased progressively with worsening glycemic status (NGR: 6.499; Pre-DM: 6.888; DM: 7.241). CONCLUSIONTyG-ABSI demonstrated a consistent cross-sectional association with MASLD across glycemic states, with greater discriminative capacity in early metabolic dysfunction than in diabetes.Given its preliminary nature, prospective cohort validation incorporating longitudinal therapy records is required prior to clinical application.