Abstract
BACKGROUND: The purpose of this study was to evaluate the relationship between triglyceride glucose (TyG)-related indices, including the TyG index and its modified versions, and the risk of developing diabetes mellitus (DM) in patients with non-alcoholic fatty liver disease (NAFLD). METHODS: Data were sourced from the Dryad repository, comprising 2,374 individuals diagnosed with NAFLD. The TyG index was calculated using the formula: Ln [(fasting triglyceride level in mg/dL) × (fasting blood glucose in mg/dL)]/2. Modified indices were created by incorporating the TyG index with body mass index (BMI), waist circumference (WC), and waist-to-height ratio (WHtR). Logistic regression analysis and restricted cubic spline regression were utilized to investigate the association of the TyG index and modified indices with the incidence of diabetes. Predictive capabilities were evaluated through receiver operating characteristic (ROC) curve analyses. RESULTS: In multivariable logistic regression models that accounted for confounding factors, a linear correlation was identified between the TyG index and its related indices (e.g., TyG-BMI, TyG-WC, and TyG-WHtR) with the development of DM, when considered as continuous variables. Each standard deviation (SD) increase was associated with odds ratios (ORs) and 95% confidence intervals (CIs) of 1.45 (1.23–1.72, P < 0.001), 1.57 (1.33–1.85, P < 0.001), 1.81 (1.53–2.14, P < 0.001), and 1.75 (1.48–2.06, P < 0.001), respectively. The area under the curve (AUC) values for TyG, TyG-BMI, TyG-WC, and TyG-WHtR were 0.626, 0.647, 0.675, and 0.681, respectively, indicating that the TyG-related indices had higher AUC values than TyG alone. Stratified analyses revealed a positive correlation between TyG-related indices and DM among NAFLD patients with various characteristics. CONCLUSIONS: A linear and positive correlation was observed between TyG-related parameters and the risk of incident DM. Among NAFLD patients, the TyG-BMI, TyG-WC, and TyG-WHtR indices were found to be more effective clinical predictors than the TyG index alone for forecasting the onset of DM.