Abstract
OBJECTIVE: Current literature lacks evidence on the association between the triglyceride glucose-weight-adjusted waist index (TyG-WWI) and incident stroke. Therefore, this study aims to elucidate the relationship between TyG-WWI and stroke risk in a nationally representative cohort of middle-aged and elderly Chinese adults by: (1) Quantitatively assessing its prospective association with incident stroke, and (2) Investigating potential nonlinear relationships and inflection points. METHODS: This national longitudinal study examined data from five waves of the CHARLS (2011-2020). TyG-WWI served as the key exposure variable, while stroke incidence, as determined by physician diagnosis, was the principal outcome of interest. Associations were evaluated using Cox proportional hazards models alongside nonlinear analyses, with careful adjustment for a range of covariates. To further ensure the reliability of our findings, we performed a series of sensitivity and subgroup analyses. RESULTS: In this study, we investigated the association between TyG-WWI and stroke risk. The results demonstrated a significant positive correlation between TyG-WWI quartiles and stroke risk, with HRs of 1.35 (95% CI 1.08-1.69, p = 0.0094), 1.38 (95% CI 1.09-1.73, p = 0.0069), and 1.42 (95% CI 1.09-1.85, p = 0.0086) for Q2, Q3, and Q4, respectively, compared to Q1 in Model III (P for trend = 0.0158). A two-piecewise linear regression model revealed an inflection point of TyG-WWI at 43.32, with a significant association between TyG-WWI and stroke risk only above this point (HR = 1.03, 95% CI 1.01-1.05, p = 0.0008). Stratified analysis indicated a more robust association in non-hypertensive individuals (HR = 1.94, 95% CI 1.29-2.91, p = 0.0014 in Q4) and a significant interaction for hypertension status (P for interaction = 0.0493). CONCLUSION: In conclusion, our study reveals a significant association between TyG-WWI and stroke risk in the middle-aged and elderly population in China, with stroke risk increasing progressively across higher TyG-WWI quartiles. Notably, this association was significantly more pronounced in non-hypertensive individuals. Furthermore, an inflection point indicates a stronger association beyond this threshold.