Combined blood sugar and muscle index flags cardiovascular risk in Chinese adults

A new index built from four routine blood biomarkers was associated with cardiovascular disease in middle-aged and older Chinese adults, but its ability to actually discriminate between people who do and do not develop disease was weak, according to a study published in Cardiovascular Diabetology. The researchers combined the triglyceride-glucose (TyG) index, a marker of insulin resistance, with the sarcopenia index (SI), a proxy for skeletal muscle mass, into a single measure they call the TyG-Sarcopenia Index (TyG-SI), calculated as TyG divided by SI and multiplied by 1000.

The analysis drew on the China Health and Retirement Longitudinal Study (CHARLS), using its 2015 and 2020 waves. In total, 10,266 adults aged 45 years or older were included in the cross-sectional analysis — 4,766 of them men (46.4%), with a median age of 61 years. A separate prospective analysis followed 6,934 participants who were free of cardiovascular disease at baseline for approximately five years. Associations were estimated with logistic regression and Cox proportional hazards models adjusted for age, sex, education, marital status, residence, smoking, alcohol consumption, body mass index, hypertension, and cancer. Restricted cubic splines, ROC comparisons, exploratory subgroup analyses, and five-fold out-of-fold XGBoost Cox SHAP analyses supplemented the main models.

In the cross-sectional data, 2,092 of the 10,266 participants (20.4%) had prevalent cardiovascular disease. Prevalence rose steadily across TyG-SI quartiles, from 14.4% in the lowest quartile to 26.3% in the highest, a trend the authors reported as statistically significant (P for trend < 0.001). Compared with participants in the lowest quartile, those in the highest had an odds ratio of 1.36 (95% confidence interval 1.15–1.60) for prevalent cardiovascular disease after multivariable adjustment.

The prospective findings pointed in the same direction. Over roughly five years of follow-up, 804 of the 6,934 initially disease-free participants (11.6%) developed cardiovascular disease. Each standard deviation higher TyG-SI at baseline was associated with an adjusted hazard ratio of 1.14 (95% CI 1.06–1.22) for incident cardiovascular disease, suggesting a modest but detectable association between the combined metabolic and muscle measure and future disease risk.

However, the study’s discrimination analyses tempered any enthusiasm for immediate clinical use. Single-indicator area under the ROC curve values were 0.580 for prevalent cardiovascular disease, 0.579 for heart disease, and 0.577 for stroke — barely better than chance. When the incremental value of TyG-SI beyond conventional risk factors was assessed with out-of-fold XGBoost methods, the change in AUC was small: 0.00188 for prevalent cardiovascular disease and 0.00226 for incident cardiovascular disease, with the latter confidence interval including zero.

In the machine-learning analysis, TyG-SI accounted for 18.0% of total out-of-fold feature contribution in the XGBoost-SHAP model, ranking behind age, hypertension, and body mass index. This suggests the index carries information that overlaps with, rather than replaces, established risk factors — consistent with the authors’ framing of it as a candidate complementary marker rather than an established clinical tool.

The appeal of the index lies in its simplicity. Because it is derived from four routine biomarkers already commonly measured in clinical practice, it could in principle be computed without additional testing. The authors concluded that TyG-SI was associated with both prevalent and incident cardiovascular disease in Chinese middle-aged and older adults after multivariable adjustment, but that its single-indicator discriminative ability was weak and its incremental discrimination beyond conventional risk factors was small, supporting further evaluation rather than clinical deployment.

As with any observational analysis of an existing cohort, the findings describe associations within a specific population — Chinese adults aged 45 and older participating in CHARLS — and cannot by themselves establish causation or generalizability to other populations. The authors note that the results support further evaluation of TyG-SI as a complementary risk marker, not its adoption as a decision-making instrument.

Subject of Research: Association of a combined triglyceride-glucose and sarcopenia index with cardiovascular disease in Chinese middle-aged and older adults

Article Title: Beyond lipids: sarcopenia as a cardiovascular risk indicator in Chinese middle-aged and older adults

Article References: Feng, L., Yan, H., Yang, G., Xin, M., Duan, W., Pan, Y., Chen, D., Yang, S., Zeng, H., Wan, X., & He, X. (2026). Beyond lipids: sarcopenia as a cardiovascular risk indicator in Chinese middle-aged and older adults. Cardiovascular Diabetology. https://doi.org/10.1186/s12933-026-03421-8

Image Credits: AI Generated

DOI: 10.1186/s12933-026-03421-8

Keywords: TyG-SI, insulin resistance, sarcopenia, cardiovascular disease, CHARLS, risk marker, cohort study, biomarkers, middle-aged adults, older adults, China, Beyond