Triglyceride-Glucose-Frailty Index Shows Strong Link to Cardiovascular Risk

A composite measure that blends blood sugar, blood fat and frailty signals has emerged as a powerful predictor of future cardiovascular disease, according to a large prospective cohort study drawing on nearly a decade of health data from middle-aged and older adults in China. The research, published in Cardiovascular Diabetology, evaluated nine triglyceride-glucose-frailty composite indices and their ability to flag incident cardiovascular disease across the full spectrum of cardiovascular-kidney-metabolic syndrome stages, from stage 0 through stage 3.

The study drew on 6,083 participants from the China Health and Retirement Longitudinal Study, tracked between 2011 and 2020. Over a median follow-up of 9.0 years, 1,644 incident cardiovascular disease events were recorded, representing 27.03 percent of the cohort. The incidence of cardiovascular events rose steadily across cardiovascular-kidney-metabolic syndrome stages, climbing from 15.01 percent at stage 0 to 31.82 percent at stage 3, underscoring the progressive nature of metabolic and vascular risk accumulation.

Among the nine composite indices tested, one designated TFI-WC stood out as the strongest performer. Each one-standard-deviation increase in TFI-WC was associated with a hazard ratio of 1.258 for incident cardiovascular disease, with a 95 percent confidence interval of 1.204 to 1.316 and a p-value below 0.0001. The association followed a nonlinear pattern, and the remaining eight indices also reached statistical significance, all with p-values below 0.0001.

The relationships held consistently across all cardiovascular-kidney-metabolic syndrome stages, with no significant interaction detected between stage and index performance. This consistency suggests that the composite measure captures cardiovascular risk information that is relevant whether someone is at the earliest metabolic stage or already further along the disease continuum.

Beyond simple association, the researchers tested whether adding TFI-WC to a base prediction model genuinely improved risk discrimination. The time-dependent area under the receiver operating characteristic curve rose from 0.623 to 0.641 at three years, from 0.652 to 0.669 at five years, and from 0.639 to 0.665 at nine years. The index also produced the highest five-year integrated discrimination improvement, measured at 0.0075 with a 95 percent confidence interval of 0.0059 to 0.0091. The other eight indices similarly contributed incremental predictive value, all with p-values below 0.0020.

The team also explored machine-learning approaches. Logistic regression delivered the highest discrimination among the models tested, achieving an area under the curve of 0.639. SHAP analysis, a technique for interpreting model outputs, identified each composite index as a leading contributor within its respective model, reinforcing the central role these measures play in cardiovascular risk stratification.

Clustering analysis added another dimension to the findings. The researchers identified three distinct burden strata within the cohort, and participants in the highest stratum carried a 1.797-fold increased risk of cardiovascular disease compared with the lowest, with a 95 percent confidence interval of 1.530 to 2.111. This graded pattern aligns with the broader concept that metabolic, renal and cardiovascular risk factors accumulate along a continuum rather than appearing as isolated binary conditions.

Subject of Research: Composite indices combining triglyceride-glucose metrics with frailty measures for predicting incident cardiovascular disease across cardiovascular-kidney-metabolic syndrome stages.

Article Title: Nine triglyceride-glucose-frailty composite indices and incident cardiovascular disease across cardiovascular-kidney-metabolic syndrome stages 0–3: a prospective cohort study

Article References: Chen, Q., Cao, C., Li, J., Hu, H., & Xu, Y. (2026). Nine triglyceride-glucose-frailty composite indices and incident cardiovascular disease across cardiovascular-kidney-metabolic syndrome stages 0–3: a prospective cohort study. Cardiovascular Diabetology. https://doi.org/10.1186/s12933-026-03399-3

Image Credits: AI Generated

DOI: 10.1186/s12933-026-03399-3

Keywords: triglyceride-glucose index, cardiovascular disease, frailty, cardiovascular-kidney-metabolic syndrome, risk prediction, cohort study, CHARLS, biomarker, metabolic health, Cox regression, machine learning, risk stratification