Hormone replacement therapy and stroke risk in older women

Studies of the effect of hormone replacement therapy (HRT) on the risk of stroke in postmenopausal women have yielded divergent results. Many of the studies had small numbers of estrogen-using women and relatively young women. All were confounded because women who use estrogen are healthier than the general population. Case-control studies, which included a larger number of cases, usually were limited to stroke survivors and had problems of recall bias regarding exposure. We describe the independent risk of fatal and nonfatal stroke in a socioeconomically homogeneous cohort of older, postmenopausal, community-dwelling women in Rancho Bernardo, California, who had a high rate of past or current HRT use. During 1984-1987, estrogen use and stroke risk factors were ascertained for 1031 women over age 60 without a history of stroke in the Rancho Bernardo cohort. Stroke deaths were monitored by obtaining 100% of death certificates, and nonfatal strokes were determined through questionnaires. At baseline, the average age was 73.1 years, 27% were current estrogen users, and 45% were past estrogen users. Over 8.8 follow-up years, 263 deaths occurred, and 37 of the death certificates listed stroke as an underlying or contributing cause of death. Only 7 of current HRT users had a stroke compared with 16 former users and 14 never users. The age-adjusted odds ratio (OR) was 0.92 (95% CI = 0.56-1.50) in HRT users versus never users. Results were not materially changed in multiply adjusted analyses. Similar analyses of nonfatal strokes and transient ischemic attacks (n = 20) showed an increased risk (OR = 3.02, 95% CI = 0.70-13.08). This study provides no evidence that HRT prevents stroke in older women, but the confidence intervals are wide, suggesting that more studies are needed.