Effect of atorvastatin in elderly patients with a recent stroke or transient ischemic attack.

BACKGROUND: It is unclear whether patients age 65 years and over with a recent stroke or TIA benefit from statin treatment to a similar degree as younger patients. METHODS: The 4,731 patient cohort in the SPARCL study was divided into an elderly group (65 and over) and a younger group. The primary endpoint (fatal or nonfatal stroke) and secondary endpoints were analyzed, with calculation of the hazard ratio (HR) and p values from a Cox regression model. RESULTS: There were 2,249 patients in the elderly group and 2,482 in the younger group. The baseline LDL (133 mg/dL) and total cholesterol were comparable in the two groups. The elderly and younger groups had a 61.4 mg/dL and 58.7 mg/dL decrease in mean LDL during the trial. The primary endpoint was reduced by 26% in younger patients (HR 0.74, 0.57-0.96, p = 0.02) and by 10% in elderly subjects (HR 0.90, 0.73-1.11, p = 0.33). A test of heterogeneity for a treatment-age interaction was not significant (p = 0.52). The risk of stroke or TIA (HR 0.79, p = 0.01), major coronary events (HR 0.68, p = 0.035), any coronary heart disease event (HR 0.61, p = 0.0006), and revascularization procedures (HR 0.55, p = 0.0005) was reduced in the elderly group. CONCLUSIONS: There was no heterogeneity in the stroke reduction seen with atorvastatin in the elderly and younger groups. Cardiac events and revascularization procedures were also lower in both the elderly and younger subgroups treated with atorvastatin. These results support the use of atorvastatin in elderly patients with recent stroke or TIA.     «

BACKGROUND: It is unclear whether patients age 65 years and over with a recent stroke or TIA benefit from statin treatment to a similar degree as younger patients. METHODS: The 4,731 patient cohort in the SPARCL study was divided into an elderly group (65 and over) and a younger group. The primary endpoint (fatal or nonfatal stroke) and secondary endpoints were analyzed, with calculation of the hazard ratio (HR) and p values from a Cox regression model. RESULTS: There were 2,249 patients in the...     »