Association between the triglyceride glucose-Chinese visceral adiposity index and new-onset stroke risk: a national cohort study

Numerous studies have investigated the effect of an integrated index that combines the triglyceride‒glucose (TyG) index with various obesity indicators on stroke incidence. However, how to use the TyG index and the Chinese Visceral Adiposity Index (CVAI) for stroke prevention remains unclear. This study examined the associations between dynamic changes in the TyG-CVAI index and cumulative, baseline, and new-onset stroke risk. Data from 3,769 participants in the China Health and Retirement Longitudinal Study(CHARLS) were analyzed, concentrating on the baseline TyG-CVAI, TyG-CVAI in 2015, and the cumulative TyG-CVAI derived from these. The fluctuations of the TyG-CVAI index were grouped into three clusters using K-means clustering analysis. Logistic regression models were used to examine the relationship between the TyG-CVAI index and new-onset stroke risk. Restricted cubic splines (RCS) were employed to investigate potential nonlinear relationships while assessing the predictive capability by receiver operating characteristic curve. During the follow-up period, 181 participants experienced stroke events. The stroke incidence rates in Clusters 1, 2, and 3 were 2.42%, 8.72%, and 4.37%, respectively. After adjustment for confounding factors, Cluster 2 with high and increasing TyG-CVAI index (OR = 3.16, 95% CI 1.94–5.22), the Q3 group with high cumulative TyG-CVAI index (OR = 2.53, 95% CI 1.60–4.02), and the Q3 group with high baseline TyG-CVAI index (OR = 2.49, 95% CI 1.57–3.95),which were all correlated with an elevated risk of new-onset stroke. The RCS analysis disclosed a U-shaped relationship between cumulative and baseline TyG-CVAI index and stroke risk. The fluctuations in and baseline, and cumulative TyG–CVAI index are independently correlated with an increased risk of stroke. The TyG–CVAI index is anticipated to be a more efficient and significant indicator for evaluating early stroke.