Abstract 2503: Plasma Estradiol and Testosterone Levels and Risk of Ischemic Stroke in Postmenopausal Women
Background: Large randomized trials have shown that use of exogenous estrogens with or without progestin increase the risk of ischemic stroke among women. Current data on the association between endogenous sex hormones and ischemic stroke are very limited. Data are also lacking on the association with endogenous androgens, which, through aromatization, are the primary source of endogenous estrogen in postmenopausal women. Thus, we evaluated plasma estradiol and testosterone levels and risk for ischemic stroke among women in the Nurses’ Health Study (NHS).
Methods: Among 32,826 participants in the NHS who provided blood samples in 1989-1990, 405 postmenopausal women (209 never/past users and 196 current users of hormone therapy) had a confirmed diagnosis of ischemic stroke through 2006. An equal number of controls who were free of stroke during follow-up were matched for age, use of postmenopausal hormones, and smoking status. Estradiol and testosterone levels were measured by mass spectroscopy. Conditional logistic regression was used to evaluate the association of log-transformed levels of estradiol and testosterone with stroke risk, within strata of hormone therapy use.
Results: Circulating estradiol was not associated with stroke risk among never/past users; the relative risk (RR) (95% confidence interval) for highest to lowest estradiol quartiles was 1.27 (0.74-2.19), when adjusted for matching factors (p trend=0.75). However, there was a nonsignificantly increased risk of stroke among current users with higher estradiol levels; conditional on matching factors, the RRs were 2.28 (1.20-4.33), 1.55 (0.81-2.96), and 1.90 (0.96-3.79), for quartiles 2-4, respectively (p for trend=0.56). No association was observed between total testosterone levels and stroke risk in either group; the RRs for highest to lowest quartiles were 0.86 (0.50-1.49) and 1.06 (0.60-1.87) in never/past and current users, respectively (p for trends ≥0.39). The associations with free sex hormone levels were similar to those observed with total circulating levels. Additional adjustment for body mass index, aspirin use, alcohol consumption, and history of diabetes, elevated cholesterol, hypertension, or coronary heart disease, as well as levels of c-reactive protein (CRP) and the total/HDL cholesterol ratio did not materially change the associations. No association was observed between total and free testosterone levels and stroke risk in either never/past or current users (p for trends ≥0.39). These associations did not vary across levels of aspirin use smoking status, or CRP (p for interactions >0.05).
Conclusions: Estradiol and testosterone levels were not associated with risk of ischemic stroke in this cohort of postmenopausal women.
- © 2012 by American Heart Association, Inc.