The Women's Health Initiative (WHI) trial recently was terminated three years before its planned completion date because the harmful effects of combination estrogen-progestin therapy outweighed the benefits. Among other findings, women taking estrogen-progestin therapy had an increased risk of stroke compared with women in the placebo group. The mechanism of stroke was hypothesized to be inflammatory or thrombotic. Wassertheil-Smoller and colleagues for the WHI investigators examined in more detail the subtypes, risk factors, and modifying factors to determine which additional factors might modulate the cerebrovascular effects of hormones.
The study population included WHI participants aged 50 to 79 years, with patients randomized to estrogen-progestin combination therapy or placebo. Outcomes included transient ischemic attacks and strokes, which were classified according to subtypes. Patients also were assessed for hypertension, physical activity levels, and baseline characteristics such as age, race, smoking status, and diabetes.
After an average of 5.6 years, 151 women (1.8 percent) in the combined therapy group had strokes compared with 107 women (1.3 percent) in the placebo group. Ischemic strokes accounted for 82.8 percent of the strokes in the combined therapy group and 75.7 percent in the placebo group, with an overall rate of 79.8 percent. Hemorrhagic strokes accounted for 11.9 percent of strokes in the hormone group and 18.6 percent in the placebo group, or 14.8 percent of strokes overall. Severity and subclassification of ischemic strokes were similar in both groups. For all stroke subtypes, the hazard ratio was 1.31, with ratios of 1.44 for ischemic stroke and 0.82 for hemorrhagic stroke. Subtypes and baseline characteristics were similar in both groups, with higher hazard ratios in the combined therapy group.
There was an adverse effect of hormones across all age groups. In general, stroke risk was greater in black women, current smokers, and women with hypertension, left ventricular hypertrophy, diabetes, higher Framingham risk scores, and higher white cell counts and hematocrit levels. Vitamin C supplementation and physical activity were associated with a decreased risk of stroke. Hormone therapy tended to increase systolic blood pressure, but this increase did not change the risk of stroke. Similarly, there were no significant interactions of baseline inflammatory markers and combined hormone therapy. Higher levels of C-reactive protein were associated with an increased risk of stroke in both groups.
WHI is a trial of healthy women, and only 5 percent of them have a history of cardiovascular disease. The current study, which extended beyond the original WHI findings by four months, found that the participants who took hormone therapy had a 31 percent increase in stroke risk compared with women who took placebo. The increased risk was significant only for ischemic stroke. No risk factor, including hypertension, appeared to mediate this increased risk—all women taking estrogen-progestin therapy were at increased risk, regardless of subtype.