Objective: To perform a health maintenance organization-based case-contro l study to evaluate the association of total and high density lipoprotein (HDL) cholesterol with the risk of stroke subtypes and in patient subgroups. Methods: Cases had a confirmed incident ischemic stroke (n = 1,242) or hemorrhagic stroke (n = 313). Controls (n = 6,455) were identified in a companion myocardial infar ction study. Risk of stroke was modeled using logistic regression. Results: The highest total cholesterol quintile was associated with an increased risk of isch emic stroke compared to the lowest quintile (OR = 1.6, 95% CI 1.3 to 2.0) with the strongest subtype associations for atherosclerotic stroke (OR = 3.2) and la cunar stroke (OR = 2.4). The highest HDL cholesterol quintile was associated wit h a decreased risk of ischemic stroke compared to the lowest quintile (OR = 0.8, CI 0.6 to 1.0). Subgroup analyses suggested that the total cholesterol associat ion was more important for patients < 66 years of age and those with HDL < 50 mg /dL; the HDL association was more important for patients without diabetes or atr ial fibrillation. The second through fourth total cholesterol quintiles were ass ociated with a decreased isk of hemorrhagic stroke compared to the lowest quinti le (OR = 0.7, CI 0.5 to 1.0). Conclusions: Higher total and lower HDL cholestero l levels were associated with increased risk of ischemic stroke, especially cert ain stroke subtypes and patient subgroups. The lowest levels of total cholestero l were associated with an increased risk of all hemorrhagic strokes. Objective: To perform a health maintenance organization-based case-contro l study to evaluate the association of total and high density lipoprotein (HDL) cholesterol with the risk of stroke subtypes and in patient subgroups. Methods: Cases had a confirmed incident ischemic stroke ( Results (n = 6,455) were identified in a companion myocardial infarction study. Risk of stroke was modeled using logistic regression. Results: The highest total cholesterol quintile was associated with an increased risk of isch emic stroke compared to the lowest quintile (OR = 1.6, 95% CI 1.3 to 2.0) with the strongest subtype associations for atherosclerotic stroke (OR = 3.2) and la cunar stroke was associated wit ha decreased risk of ischemic stroke compared to the lowest quintile (OR = 0.8, CI 0.6 to 1.0). Subgroup analyzes suggest that the total cholesterol associat ion was more important for patients <66 ye ars of age and those with HDL <50 mg / dL; the HDL association was more important for patients without diabetes or atr ial fibrillation. The second through fourth total cholesterol quintiles were ass ociated with a decreased isk of hemorrhagic stroke compared to the lowest quinti (OR = 0.7, CI 0.5 to 1.0). Conclusions: Higher total and lower HDL cholestero l levels were associated with increased risk of ischemic stroke, especially cert ain stroke subtypes and patient subgroups. The lowest levels of total cholestero l were associated with an increased risk of all hemorrhagic strokes.