Associations Between High-Density Lipoprotein Particles and Ischemic Events by Vascular Domain, Sex, and Ethnicity: A Pooled Cohort Analysis

BACKGROUND: High-density lipoprotein (HDL) cholesterol concentration (HDL-C) is an established atheroprotective marker, in particular for coronary artery disease; however, HDL particle concentration (HDL-P) may better predict risk. The associations of HDL-C and HDL-P with ischemic stroke and myocardial infarction (MI) among women and Blacks have not been well studied. We hypothesized that HDL-P would consistently be associated with MI and stroke among women and Blacks compared with HDL-C. METHODS: We analyzed individual-level participant data in a pooled cohort of 4 large population studies without baseline atherosclerotic cardiovascular disease: DHS (Dallas Heart Study; n=2535), ARIC (Atherosclerosis Risk in Communities; n=1595), MESA (Multi-Ethnic Study of Atherosclerosis; n=6632), and PREVEND (Prevention of Renal and Vascular Endstage Disease; n=5022). HDL markers were analyzed in adjusted Cox proportional hazard models for MI and ischemic stroke. RESULTS: In the overall population (n=15 784), HDL-P was inversely associated with the combined outcome of MI and ischemic stroke, adjusted for cardiometabolic risk factors (hazard ratio [HR] for quartile 4 [Q4] versus quartile 1 [Q1], 0.64 [95% CI, 0.52–0.78]), as was HDL-C (HR for Q4 versus Q1, 0.76 [95% CI, 0.61–0.94]). Adjustment for HDL-C did not attenuate the inverse relationship between HDL-P and atherosclerotic cardiovascular disease, whereas adjustment for HDL-P attenuated all associations between HDL-C and events. HDL-P was inversely associated with the individual end points of MI and ischemic stroke in the overall population, including in women. HDL-P was inversely associated with MI among White participants but not among Black participants (HR for Q4 versus Q1 for Whites, 0.49 [95% CI, 0.35–0.69]; for Blacks, 1.22 [95% CI, 0.76–1.98]; P(interaction)=0.001). Similarly, HDL-C was inversely associated with MI among White participants (HR for Q4 versus Q1, 0.53 [95% CI, 0.36–0.78]) but had a weak direct association with MI among Black participants (HR for Q4 versus Q1, 1.75 [95% CI, 1.08–2.83]; P(interaction)<0.0001). CONCLUSIONS: Compared with HDL-C, HDL-P was consistently associated with MI and ischemic stroke in the overall population. Differential associations of both HDL-C and HDL-P for MI by Black ethnicity suggest that atherosclerotic cardiovascular disease risk may differ by vascular domain and ethnicity. Future studies should examine individual outcomes separately.