Abstract
BACKGROUND: Atherosclerotic cardiovascular disease (ASCVD) is a leading global cause of mortality, and reliable risk assessment tools are critical for prevention. While TyG-related parameters are used for ASCVD risk stratification, the association between the novel TyG-ABSI and ASCVD in the general population requires further characterization, and its potential improvement in discrimination over traditional TyG parameters has not been fully elucidated. METHODS: We analyzed data from the National Health and Nutrition Examination Survey (NHANES, 1999–2018), a nationally representative cross-sectional study including 22,466 participants. ASCVD was defined by self-reported physician diagnosis of coronary heart disease (CHD), angina pectoris (AP), myocardial infarction (MI), or stroke. TyG-ABSI and traditional TyG parameters were calculated using standardized laboratory and anthropometric data. Statistical analyses included weighted logistic regression (to assess associations), restricted cubic spline analysis (to explore dose–response relationships), subgroup analyses (to test effect modification), receiver operating characteristic (ROC) curves (to evaluate discrimination ability), and net reclassification improvement (NRI) and integrated discrimination improvement (IDI) (to quantify improvement in reclassification and risk separation). Sensitivity analyses were conducted to verify robustness. RESULTS: In fully adjusted models, TyG-ABSI was significantly associated with ASCVD (odds ratio [OR] per standard deviation [SD] increase: 1.15, 95% confidence interval [CI]: 1.09–1.22) and its subtypes (CHD: OR = 1.14, 95%CI: 1.05–1.24; AP: OR = 1.10, 95%CI: 1.01–1.21; MI: OR = 1.15, 95%CI: 1.06–1.25), with a significant linear dose–response relationship observed for ASCVD and all subtypes (all p-trend < 0.05). TyG-ABSI yielded the highest ROC area under the curve (AUC) across all outcomes (ASCVD: 0.69; CHD: 0.70; AP: 0.69; MI: 0.70; stroke: 0.65). Furthermore, TyG-ABSI demonstrated modest but significant improvement in discrimination over traditional TyG parameters for both ASCVD and its subtypes, as evidenced by significant NRI and IDI metrics. Specifically for ASCVD, it correctly reclassified an additional 7.88% (vs TyG-WHtR) to 16.36% (vs TyG-BMI) of patients based on NRI. The association between TyG-ABSI and ASCVD was robust across demographic, lifestyle, and clinical subgroups, as well as in sensitivity analyses. CONCLUSION: This study indicates a significant positive correlation between TyG-ABSI and ASCVD in the general population. Moreover, TyG-ABSI demonstrates modest improvement in discrimination compared to traditional TyG parameters, identifying additional high-risk individuals via reclassification. These findings suggest that TyG-ABSI holds promise as a candidate marker for optimizing ASCVD risk stratification in clinical practice, though future prospective validation is warranted. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12933-025-03069-w.