Association between triglyceride-glucose index and total trabecular bone score in US adults

Abstract

This study examines the association between the triglyceride-glucose (TyG) index and trabecular bone score (TBS) among US adults. In this cross-sectional analysis of National Health and Nutrition Examination Survey 2005 to 2008, the descriptive cohort included 3880 adults with TBS and fasting laboratory data; 3571 with positive fasting-subsample weights contributed to survey-weighted regression. TyG was calculated as ln(triglycerides [mg/dL] × glucose [mg/dL]/2) and is dimensionless. Covariate missingness was handled by multiple imputation. The primary model excluded bone mineral density (BMD) to avoid overadjustment; BMD-adjusted, complete-case, and renal-function sensitivity models were fitted. In the primary model, higher TyG was associated with lower TBS (β per 1-unit increase = −0.0097; 95% confidence interval = −0.0182 to −0.0012; P = .026). Restricted cubic spline showed nonlinearity (P-overall < .001; P-nonlinear < .001). Notably, across TyG quartiles, spine BMD remained stable (P = .104), whereas TBS declined substantially (from 1.446 to 1.310, P < .001), suggesting TyG specifically associates with bone quality rather than quantity. An exploratory two-segment knot at TyG 8.58 requires validation and is not a clinical threshold. Higher TyG is modestly associated with lower TBS, independent of BMD. While causal inference is precluded by cross-sectional design, TyG – calculated from routine labs at no extra cost – may serve as a pragmatic signal to prompt TBS evaluation in metabolically unhealthy patients, particularly obese males, even when BMD is normal. Longitudinal validation is needed.