Association between platelet-to-high-density lipoprotein cholesterol ratio and incident stroke among adults with hypertension: A longitudinal CHARLS cohort study

Objective Previous studies that have used data from the China Health and Retirement Longitudinal Study have linked the platelet-to-high-density lipoprotein cholesterol ratio to stroke in broader middle-aged and older populations. We evaluated whether this association persisted among adults with hypertension who were stroke-free at baseline and assessed platelet count and high-density lipoprotein cholesterol levels separately. Methods This retrospective secondary analysis used longitudinal data from the China Health and Retirement Longitudinal Study, with 2011 as the baseline year and 2013, 2015, 2018, and 2020 as the follow-up years. The primary cohort included 3963 adults aged ≥45 years who met the criterion for hypertension at baseline and were explicitly stroke-free at baseline and for whom complete and valid data were available for the platelet-to-high-density lipoprotein cholesterol ratio, follow-up, and primary-model covariates. Incident stroke was based on questionnaire-reported physician diagnosis without independent confirmation using medical records, neuroimaging, or stroke subtype adjudication. Cox models evaluated the platelet-to-high-density lipoprotein cholesterol ratio continuously and stratified by quartiles. Component, selection-bias, community-cluster robust, spline, and additional-adjustment analyses were performed. Results During 34,637.3 person-years of follow-up, 551 participants reported incident stroke. Each 1-SD increase in the platelet-to-high-density lipoprotein cholesterol ratio was associated with a higher stroke hazard in Model 3 (hazard ratio, 1.15; 95% confidence interval, 1.07–1.24; p < 0.001). A comparison of quartile 4 versus quartile 1 data yielded a hazard ratio of 1.33 (95% confidence interval, 1.05–1.69; p = 0.018 ; p for trend = 0.007). Platelet count was positively associated with stroke (hazard ratio per 1-SD increase, 1.16; 95% confidence interval, 1.07–1.25), while the inverse high-density lipoprotein cholesterol association attenuated after mutual adjustment. Community-cluster robust inference was similar. Restricted cubic spline analysis revealed an overall association between the platelet-to-high-density lipoprotein cholesterol ratio and hazard of first incident stroke ( p = 0.006), with no evidence of departure from linearity ( p for nonlinearity = 0.812). Conclusions A higher baseline platelet-to-high-density lipoprotein cholesterol ratio was associated with a higher hazard of first incident stroke among adults who met the criterion for hypertension at baseline and were stroke-free at baseline. The observational design and nonadjudicated outcome preclude causal or clinical-prediction claims; independent validation is required.