The WHI results are at odds with results of large epidemiological studies that showed protection against cardiovascular disease. Although the latter data are, in part, confounded by a “healthy user bias,” much of the inconsistency may be explained by the fact that women in the latter studies initiated HRT at the menopausal transition, whereas the WHI trial was conducted in older women (mean age 63.3), who were, on average, approx 12 yr postmenopausal. In addition, older trials included women on either unopposed estrogen therapy (ERT) or cyclic HRT regimens.

Whatever other forces may have been at work, observational and experimental evidence supports the conclusion that estrogen’s atheropreventive effects predominate early, in the absence of vulnerable plaque to be ruptured or thrombotic episodes propagated by narrowed lumens and intravascular turbulence. On the contrary, age-related adverse effects of HRT may prevail once complex atheromas and luminal narrowing/irregularity are established. It is known that prevalence of subclinical “at-risk” atherosclerotic lesions increases in women during the first 5–10 yr after menopause. Furthermore, animal and clinical evidence supports the use of lower doses of estrogen than were employed in the WHI in older/longer postmenopausal women.

Therefore, we suggest that conclusions from the WHI should be strictly limited to the WHI Writing Group’s own published interpretation that initiation of daily continuous treatment with combined oral conjugated equine estrogens (0.625 mg) and medroxyprogesterone acetate (2.5 mg) or 0.625 mg conjugated equine estrogen, alone, in older postmenopausal women is inadvisable for prevention of heart disease. Other conclusions on the use of such regimens are moot, since they are not appropriate clinical treatments. The allowance of “age creep” to generalize these conclusions to subjects not studied in adequate power by the WHI is neither scientifically correct nor appropriate for the development of clinical practice guidelines.

Because of the limitations on the interpretation of the WHI, new RCTs are needed to resolve these questions. These RCTs should be designed to resolve whether estrogen treatment started during the menopausal transition is cardioprotective. Meanwhile, decisions of whether to initiate HRT for peri-menopausal women or to maintain it in women on long-term HRT started for estrogen-deficiency symptoms in the perimenopause should continue to be individualized based on consideration of all available data.