Association of triglyceride-glucose index and its related parameters with atherosclerotic cardiovascular disease: evidence from a 15-year follow-up of Kailuan cohort

Abstract

Background: Triglyceride glucose (TyG) index and its related parameters have been introduced as cost-effective surrogate indicators of insulin resistance, while prospective evidence of their effects on atherosclerotic cardiovascular disease (ASCVD) remained scattered and inconsistent. We aimed to evaluate the association of TyG and its related parameters with new-onset ASCVD, and the predictive capacity were further compared.

Method: A total of 95,342 ASCVD-free participants were enrolled from the Kailuan study. TyG and its related parameters were defined by fasting blood glucose, triglyceride, body mass index (BMI), waist circumstance (WC) and waist-to-height ratio (WHtR). The primary outcome was incident ASCVD, comprising myocardial infarction (MI) and ischemic stroke (IS). Cox proportional hazard models and restricted cubic spline (RCS) analyses were adopted to investigate the association between each index and ASCVD. The C-index, integrated discrimination improvement (IDI), and net reclassification improvement (NRI) were used for comparison of their predictive value for ASCVD.

Results: During a median follow-up of 15.0 years, 8,031 new cases of ASCVD were identified. The incidence rate of ASCVD increased along with elevated levels of each index, and the relationships were found to be nonlinear in the RCS analyses. The hazard ratio (HR) and 95% confidence interval (95% CI) for ASCVD was 1.39 (1.35, 1.43), 1.46 (1.41, 1.50), 1.50 (1.46, 1.55), and 1.52 (1.48, 1.57) per 1 IQR increase of baseline TyG, TyG-BMI, TyG-WC, and TyG-WHtR, respectively, and the association were more pronounced for females and younger individuals aged < 60 years (P_{for interaction}<0.05). Using the updated mean or time-varying measurements instead of baseline indicators did not significantly alter the primary findings. Additionally, TyG-WC and TyG-WHtR showed better performance in predicting risk of ASCVD than TyG, with the IDI (95% CI) of 0.004 (0.001, 0.004) and 0.004 (0.001, 0.004) and the category-free NRI (95% CI) of 0.120 (0.025, 0.138) and 0.143 (0.032, 0.166), respectively. Similar findings were observed for MI and IS.

Conclusions: Both the TyG index and its related parameters were significantly and positively associated with ASCVD. TyG-WC and TyG-WHtR had better performance in predicting incident ASCVD than TyG, which might be more suitable indices for risk stratification and enhance the primary prevention of ASCVD.

Keywords: Atherosclerotic cardiovascular disease (ASCVD); Triglyceride glucose (TyG); Triglyceride glucose-body mass index (TyG-BMI); Triglyceride glucose-waist circumference (TyG-WC); Triglyceride glucose-waist-height ratio (TyG-WHtR).

Fig. 1

Forest plots of subgroup analyses for the association of TyG and its related parameters with atherosclerotic cardiovascular diseases

Fig. 2

Sensitivity analyses for association of TyG and its related parameters with atherosclerotic cardiovascular diseases. A Excluding ASCVD events that occurred during the initial two years of follow-up; B Using the updated mean instead of baseline measurements of TyG index and its related parameters; C Further adjusting for hypertension, diabetes and hyperlipidemia; D Adopting the Cox regression model with time-varying variables updated in each survey before the occurrence of ASCVD events; E Adopting the Fine & Gary sub-distribution hazard models to further consider the competing risk of non-ASCVD deaths