“Timing Hypothesis” for Hormone Therapy: Still Viable, or Time to Let Go?

Postmenopausal hormone therapy was associated with an unfavorable balance of benefits and harms in findings from the landmark Women's Health Initiative (WHI) randomized trials, published in 2002 for estrogen-progestin1 and in 2004 for estrogen alone.2 At enrollment, women in these trials ranged in age from 50 to 79 (average age, 63), and only about one third were 50 to 59. Thus, soon after the initial WHI findings were published, researchers and clinicians began to wonder whether the outcomes were applicable to women who start HT shortly after menopause. This issue seemed particularly relevant for coronary heart disease (CHD), because previous observational studies had suggested that HT use lowered risk for coronary disease; conversely, combination HT with estrogen-progestin led to excess risk for coronary events in the WHI trial.

In subsequent WHI analyses, data trends suggested that HT did not lead to excess coronary risk when it was started close to menopause.3 Although this finding did not reach statistical significance, the idea that HT was safe or even beneficial (from the coronary perspective) when started shortly after menopause gained some currency and became known as the “timing hypothesis.” This hypothesis was promoted by some WHI researchers,4 but others remained skeptical.5 Notably, WHI researchers who advanced the timing hypothesis did not advocate prescribing HT to prevent coronary disease; rather, they suggested that “health care providers need not be unduly concerned about coronary risk” when they prescribe short-term HT to recently menopausal women for relief from vasomotor symptoms.4

To address the timing hypothesis more conclusively, the WHI researchers now have published new data from both the WHI randomized trials (which involved >17,000 women) and a parallel WHI observational study of nearly 100,000 women. Information on date of menopause and hormone use before enrollment in the WHI was used to calculate a “gap time” between menopause and first use of HT. The analysis is quite complex: Essentially, the researchers noted trends toward excess risk for CHD, venous thromboembolism, and breast cancer among early users of estrogen-progestin, and trends toward elevated risk for stroke and venous thromboembolism among early users of estrogen alone. A caveat is that the authors don't indicate whether the excess CHD events actually occurred during the first several years of HT use (i.e., when women were still in their fifties) or later on. Nevertheless, the authors conclude that “... the unfavorable balance of benefits and risks observed in the [estrogen-progestin] trial as a whole also applies to recently menopausal women .... WHI data provide little support for the estrogen timing hypothesis concerning CHD risk ....”6

However, the timing hypothesis is not dead and buried. Primate studies and other lines of basic research suggest that estrogen slows initial development of atherosclerotic plaques. In addition, the specific hormones used in the WHI (conjugated equine estrogens and medroxyprogesterone) might have harmful effects that are not shared by estradiol and progesterone. Randomized trials — for example, the KEEPS and ELITE trials — are in progress to study the effects of HT on surrogate endpoints (e.g., carotid intima-media thickness) in recently postmenopausal women.

But, until new trials with clinical endpoints are completed, a prudent conclusion is that clinical research has not identified a window of time after menopause during which HT can confidently be considered to be free of risk.7 Women of any age who receive HT for symptom relief still should be informed of all potential risks, including CHD.