ESTROGEN AFTER ISCHEMIC STROKE: Effect of estrogen replacement on risk of recurrent stroke and death in the Women’s Estrogen for Stroke Trial (WEST)

Abstract

71 Introduction: Observational research has produced conflicting findings concerning the effect of estrogen replacement therapy (ERT) on reducing risk for vascular events or death in women. To test the effect of ERT in women with established cerebrovascular disease, we designed a randomized trial of estradiol-17β(1 mg/day) vs. placebo. Methods: Participants were identified from 20 hospitals in New England. Eligibility criteria included age over 44, at least 1 year since last menstrual period, and TIA or non-disabling stroke within 90 days of entry. Randomization was stratified by baseline risk group and hospital. Primary trial outcomes were non-fatal stroke and all-cause death. Results: From December 1993-May 1998, 652 women were randomized (332 estradiol, 320 placebo). Index event was TIA in 164 subjects and stroke in 488. Mean age of subjects was 71 years (range 46–91); 84% were white, 13% black, and 4% other. Mean follow-up was 2.7 years (range: 18 days-5.8 years). At 1 year, 76% of subjects assigned to estradiol were on study drug. In the estradiol group, adverse events were diagnosed during follow-up in 8 subjects (1 pulmonary embolus (PE), 1 deep venous thrombosis (DVT), 5 breast cancers, 1 endometrial cancer) compared with 7 events in placebo subjects (2 PE, 1 DVT, and 4 breast cancers). Non-fatal strokes were confirmed in 51 subjects in the estradiol group vs. 52 in placebo subjects (rates at 3 years[R]: 16.8% estradiol vs. 17.4% placebo; logrank p-value[p]=.83). Death occurred in 46 estradiol subjects vs. 38 placebo subjects (R=13.0% vs. 12.6%, p=.89). At 3 years, combined rate of non-fatal stroke or death was 27.6% in the estradiol group vs. 27.7% in placebo [p=.80]. Conclusion: During an average follow-up of 2.7 years, estradiol treatment did not protect against recurrent cerebral ischemia or reduce all-cause mortality in postmenopausal women with pre-existing cerebrovascular disease. No increased risk for adverse events associated with estrogen was observed. This trial adds to the growing body of evidence that fails to confirm a protective role for ERT in populations with known vascular disease.