Abstract 4182: Endogenous Estrogen and Risk of Stroke in Elderly Men: The Honolulu Heart Program

Background : Little is known about the relationship between endogenous estrogen and the risk of stroke, particularly in elderly men. The purpose of this report is to determine if levels of total and bioavailable endogenous estrogen and testosterone can predict stroke in a population-based sample of elderly men. Methods : Endogenous estrogen and testosterone were measured in 2,197 men aged 71 to 93 years who participated in the Honolulu Heart Program from 1991 to 1993. All were free of prevalent stroke, coronary heart disease, and cancer. Participants were followed to the end of 1998 for thromboembolic and hemorrhagic events. Results : During the course of follow-up, 124 men developed a stroke (9.1/1000 person-years). After age-adjustment, men in the top quintile of bioavailable estrogen (>82.8 pmol/L [22.6 pg/mL]) experienced a 2-fold excess risk of stroke versus men whose estrogen levels were lower (15.5 vs 7.5/1000 person-years, p<0.001). Among the lower quintiles, there were little differences in the risk of stroke. Findings were also significant and comparable for total estrogen and for thromboembolic and hemorrhagic events. After additional adjustment for hypertension, diabetes, adiposity, cholesterol concentrations, atrial fibrillation, and other characteristics, the relationship between estrogen and stroke between the top versus lower quintiles persisted (relative hazards = 2.2; 95% confidence interval = 1.4 − 3.3, p<0.001). Testosterone was not related to the risk of stroke. Conclusions : Findings suggest that high levels of endogenous estrogen are associated with an elevated risk of stroke in elderly men.