Dynamic reversibility of biological aging and risk of cardiovascular disease and stroke: a longitudinal cohort study.

BACKGROUND: Aging status, indexed by biological age acceleration (BAacc), is closely associated with cardiovascular disease (CVD). However, transitioning from reactive healthcare to predictive, preventive, and personalised medicine (PPPM/3PM) requires elucidating how longitudinal, dynamic transitions in aging status impact incident CVD risk. METHODS: We analyzed 4,756 participants from the China Health and Retirement Longitudinal Study. Biological age was calculated using the Klemera-Doubal method (KDM). KDM-BAacc was measured at Wave 1 (baseline) and Wave 3 (follow-up) to derive multidimensional dynamic metrics: change groups, K-means clusters, aging rates, cumulative burden, and extreme phenotypes. Wave 2 was excluded from baseline calculations due to the absence of blood biomarker collection. Incident CVD (stroke and heart disease) was ascertained over a median 9.13-year follow-up. Cox models were adjusted for sociodemographic, lifestyle, and clinical risk factors. RESULTS: Compared to sustained acceleration, reversing aging status ("accelerated to non-accelerated") significantly lowered risks of composite CVD (HR: 0.72, 95% CI: 0.59-0.89) and stroke (HR: 0.46, 95% CI: 0.33-0.64). Consistent protective effects were observed in "high-to-low" cluster trajectories. Conversely, the "Rapid aging" phenotype (sustained highest quartile) nearly doubled stroke risk (HR: 1.97) versus the "Anti-aging" group. Faster aging rates (≥ 1) were associated with a 32% increased stroke risk, demonstrating a linear dose-response relationship. Notably, associations with heart disease were attenuated after adjusting for comorbidities. CONCLUSIONS: Biological aging is dynamically reversible, and reversing KDM-BAacc significantly mitigates CVD and stroke risks. For routine medical application, we recommend integrating longitudinal KDM-BAacc profiling into annual primary care check-ups. Crucially, these findings strongly support the paradigm shift from reactive medical services to PPPM/3PM. By identifying reversible suboptimal health prior to disease onset, this predictive tool enables targeted prevention and personalised interventions based on individual aging trajectories, decisively surpassing conventional generalized approaches. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at https://doi.org/10.1007/s13167-026-00460-9.