Postmenopausal Hormone Therapy and Risk of Cardiovascular Disease by Age and Years Since Menopause

Observational studies have demonstrated a lower risk of coronary heart disease (CHD) events in postmenopausal women taking hormone therapy, but, in contrast, clinical trials have shown no benefit and in some instances suggest an increased risk of CHD. This discrepancy may in part result from the timing of initial hormone therapy in relation to age and the interval since the onset of menopause. This report is a secondary analysis of Women’s Health Initiative studies—all randomized controlled trials—in which 10,739 postmenopausal women with a history of hysterectomy were assigned to receive 0.625 mg daily of conjugated equine estrogens (CEE) or placebo, and another 16,608 not having hysterectomy received either CEE plus 2.5 mg of medroxyprogesterone acetate (MPA) or placebo. Hormone therapy did not lower the overall risk of CHD. After adjusting for risk factors, the hazard ratio (HR) for CHD was lower in women taking CEE alone than in those given both CEE and MPA. The overall risk of stroke was increased, with no apparent difference between individual trials. Although events became more numerous with advancing age, there was no significant added effect of hormone therapy by age for any outcome in the combined trials. The HR for CHD was 0.76 in women within 10 years of menopause, 1.10 for those 10 to 20 years after menopause, and 1.28 for those more than 20 years after menopause. Findings were similar for women with and those without previous cardiovascular disease. The effect of hormone therapy on the risk of stroke was not influenced by years since menopause. Estimated absolute excess risk figures associated with hormone therapy became progressively larger with increasing age starting at age 60 years. Age and years since menopause correlated with one another at a high level. There were no significant trends for hormone therapy by years since last treatment, and no significant interactions were noted between either previous hormone use or oophorectomy status on the one hand and, on the other, in-trial hormonal effects by age or years since menopause. These findings offer some reassurance that hormone therapy still is a reasonable option for the management of menopausal symptoms over the short term, but they do not imply that prolonged hormone use is safe. The investigators agree that hormone therapy is appropriate for the short-term relief of moderate or severe vasomotor symptoms—but not over the long-term to prevent cardiovascular disease.