Prior research underscores the significant impact of remnant cholesterol (RC) on stroke occurrence due to its proatherogenic and proinflammatory traits. This study aims to explore diverse risks of new-onset stroke associated with RC, considering distinct inflammation levels in the middle-aged and senior population in China.

We analyzed 6509 participants from the China Health and Retirement Longitudinal Study (CHARLS) across four waves (2011–2018). We employed a multivariable Cox proportional hazards regression model, incorporated restricted cubic spline techniques, and conducted sensitivity analyses to evaluate the association among RC, high-sensitivity C-reactive protein (hsCRP), and the risk of new-onset stroke. Results: Over 7 years, 540 new-onset strokes occurred. Individuals in the highest quartile of RC levels exhibited a heightened risk of new-onset stroke, with a multivariable-adjusted hazard ratio (HR) peaking at 1.50 (95% confidence interval 1.12–2.00, P for trend = 0.021), showing a non-linear correlation (P nonlinearity = 0.049). High hsCRP alone had an adjusted HR of 1.10 (95% CI 0.87–1.39), compared to 1.40 (95% CI 1.00–1.96) for high RC alone. Additionally, concurrent high RC and hsCRP showed an adjusted HR of 1.43 (95% CI 1.05–1.96). Consistency persisted across various hsCRP thresholds, after adjusting for additional parameters, or excluding chronic diseases in the primary model, reinforcing result robustness.

Our findings reveal a substantial and non-linear association between higher baseline RC levels and an elevated risk of new-onset stroke. Moreover, elevated levels of both RC and hsCRP jointly pose the highest risk for new-onset stroke, surpassing the risk associated with each factor individually.