Background & Objectives: The role of lipid and lipoprotein biomarkers in risk prediction of ischemic heart disease is well established, but their value as independent predictors for ischemic stroke is less certain. Certain LDL subclasses based on particle sizes are associated with atherosclerotic risk. However, there is little information available on the relation of LDL particle number (LDL-P, nmol/L), LDL particle size (nm) with stroke. We evaluated associations between LDL-P and LDL particle size with the risk of ischemic stroke (IS) and transient ischemic attack (TIA).
Methods: We studied US adults aged of 45-84 years not on lipid-lowering medications enrolled in the Multi-Ethnic Study of Atherosclerosis between 2000 and 2002 with follow-up through December 2011. We evaluated associations of LDL-P (total LDL-P, large LDL-P and small LDL-P) and LDL particle size measured by NMR with, (1) composite of incident IS and TIA, and (2) IS alone. Standardized hazard ratios (HRs) were adjusted for age, sex, race (model 1), plus diabetes, hypertension medication, systolic and diastolic blood pressure, family history, smoking, HDL (model 2), and plus LDL concentration (model 3).
Results: We identified 129 subjects with incident IS, 61 with TIA and 5,328 without any type of stroke during follow-up; mean ages were 67.4, 66.7 and 61.2 years, respectively (p<0.0001). White ethnicity was 37.2%, 42.6% and 37.6%), respectively. For the composite outcome of IS and TIA, while small LDL-P (HR 1.20, 95%CI 1.04 to 1.39) was predictive in model 1, this was attenuated after additional adjustment in models 2 and 3. For the outcome of IS alone, total LDL-P (HR 1.25, 95% CI 1.06 to 1.48) and small LDL-P (HR 1.34, 95% CI 1.13 to 1.60) were associated with incident IS in model 1, but there was no significant association of large LDL-P or LDL size with IS. Only small LDL-P remained predictive of IS in model 2 (HR 1.32, 95% CI 1.05 to 1.65) and in model 3 (HR 1.31, 95% CI 1.03 to 1.67).
Conclusions: Increased small LDL-P, not particle size, is most closely associated with incident IS and could be useful in risk stratification beyond standard risk factors and lipid measures.