Triglyceride-Glucose-Based Anthropometric Indices for Predicting Incident Cardiovascular Disease: Relative Fat Mass (RFM) as a Robust Indicator

Abstract

Background/Objectives: The triglyceride-glucose (TyG) index is a recognized marker for cardiovascular disease (CVD) risk linked to insulin resistance. Combining TyG with anthropometric indicators (AIs) may improve risk prediction, but the comparative value of different AIs, including novel ones like Relative Fat Mass (RFM), is unclear. This study aimed to identify which combination of TyG and AIs has the strongest association with incident CVD in a middle-aged and elderly Chinese cohort. Methods: In this prospective study, we evaluated the association between the cumulative average of TyG combined with eight AIs (TyG-AIs) and incident CVD, heart disease, and stroke. Using data from 5192 participants in the China Health and Retirement Longitudinal Study (CHARLS), we used multivariable logistic regression to compare the predictive value of these composite indices. Results: During follow-up, 1382 (26.6%) participants developed CVD. After full adjustment, the TyG index alone was only significantly associated with stroke. In contrast, most TyG-AIs showed stronger associations with all outcomes. Notably, the index combining TyG with Relative Fat Mass (TyG-RFM) exhibited the most robust associations with total CVD (OR = 2.236), heart disease (OR = 1.679), and stroke (OR = 3.288) when comparing the highest to lowest quartiles. Conclusions: Cumulative average TyG-AIs, particularly TyG-RFM, demonstrated more robust associations with incident CVD than the TyG index alone. The TyG-RFM index shows promise as a valuable tool to improve cardiovascular risk stratification, especially for identifying at-risk non-obese individuals.

Keywords: anthropometric indicator (AI); cardiovascular disease (CVD); triglyceride–glucose (TyG).

Figure 1

Flowchart of the study population inclusion.

Figure 2

Forest plot showing the associations of cumulative average TyG-AIs with total CVD, heart disease, and stroke, analyzed using multivariable-adjusted logistic regression (Model 3, adjusted for age, sex, smoking status, alcohol consumption status, marital status, educational attainment, diabetes, and hypertension).

Figure 3

Association between cumulative average TyG-AIs with total CVD in multivariable-adjusted RCS regression Model 3. Model 3 is adjusted for age, sex, smoking status, alcohol consumption status, marital status, educational attainment, diabetes, and hypertension. The solid red lines correspond to the central estimates of odds ratios (ORs), and the red-shaded regions indicate the 95% confidence intervals. The black dashed horizontal line at OR = 1 serves as the reference level.

Figure 4

Subgroup analyses using multivariable logistic regression (Model 3) examined the associations between cumulative average TyG-AIs and total CVD in obese and non-obese subgroups, with adjustments for age, sex, smoking status, alcohol consumption status, marital status, educational attainment, diabetes, and hypertension.

Figure 5

Subgroup analyses using multivariable logistic regression (Model 3) examined the associations between cumulative average TyG-AIs and total CVD in female and male subgroups, with adjustments for age, sex, smoking status, alcohol consumption status, marital status, educational attainment, diabetes, and hypertension.

Figure 6

Subgroup analyses using multivariable logistic regression (Model 3) examined the associations between cumulative average TyG-AIs and total CVD in age < 55 and age ≥ 55 subgroups, with adjustments for age, sex, smoking status, alcohol consumption status, marital status, educational attainment, diabetes, and hypertension.