Menopause and stroke and the effects of hormonal therapy

The incidence of stroke increases substantially after menopause, and in the United States it is the third leading cause of death. Data exist suggesting that women have worse outcomes for stroke than do men. Trials of aspirin use further suggest that there is a gender difference regarding stroke. While men may have a coronary benefit from aspirin, postmenopausal women do not; yet ischemic stroke may be decreased in women but not in men. Among the traditional risk factors for stroke (such as smoking, hypertension, diabetes, obesity), hormonal therapy (HT) has been suggested to be a risk as well, although the data are not consistent. The previous Position Statement of the IMS published in 2004 was relatively silent on the issue of stroke. The annual rate of stroke in women increases rapidly with aging in postmenopausal women. While the rate is approximately 0.6–0.8/1000/year at age 50–59, it is over 2/1000 after age 60. In white women in the USA, it is 4.2/1000 at 65–74 years of age, and 11.3/1000 between ages 75 and 84 years. Thus, in trials such as the Women's Health Initiative (WHI), most of the strokes occurred in older women. Both the conjugated equine estrogen/medroxyprogesterone acetate (CEE/MPA) and CEE-alone trials in the WHI reported an increased risk of stroke in the entire population using nominal statistics: 1.41 (95% confidence interval (CI) 1.07–1.85) and 1.39 (95% CI 1.10–1.77), respectively. The increased risk was related to ischemic stroke and not hemorrhagic stroke. The absolute risk for the entire population was 0.8/1000 and 1.2/1000 woman-years (<1/1000 signifies a 'rare' event using the CIOMS classification). However, the risk was not increased in the 50–59-year-old age group, although the numbers are small. Here, the background prevalence of stroke is much lower as noted above. The results of the observational trial of the WHI were not consistent with the randomized clinical trial data and were more in keeping with older observational data showing no increased risk of stroke. The authors reconcile these differences by suggesting differences in the timing of initiation of hormones, which was at an earlier age in the observational cohort. Several recent observational studies, which will be presented, show no increased risk of ischemic stroke in younger cohorts, but possibly an increase in the risk of transient ischemic attack. These recent studies suggested the risk to be less with lower doses of estradiol ≤1 mg and to be consistent with older studies showing no risk with doses < 0.625 mg CEE. In addition, the risk was possibly lower with non-oral therapy, and was reduced if started prior to menopause. The existence of hypertension was shown to substantially increase the risk. However, data on progestogen use versus unopposed estrogen have not been consistent. At the same time, a recent body of evidence from basic science studies has reaffirmed the neuronal and stroke protective effects of estrogen. Thus, the discrepancy between these data and clinical data showing no benefit or increased risk of stroke remains to be explained. Recent trials in older women with osteoporosis have suggested an increased risk of stroke with tibolone and of stroke mortality with raloxifene. In conclusion, the current data suggest no increased risk of stroke with hormone therapy in younger (50–59 years) normotensive postmenopausal women, particularly when lower doses are prescribed soon after menopause.