WHI: The Role of Hormonal Therapy in Disease Prevention
The Women's Health Initiative (WHI) clinical trials on hormone therapy (HT) in postmenopausal women have changed clinical thinking about estrogen therapy in substantial ways, and have clarified clinical guidelines on disease prevention in postmenopausal women. The WHI trials, along with the HERS Trial, a secondary prevention trial of HT and coronary artery disease, have been the only randomized, double-blind trials of estrogen therapy with sufficient statistical power to explore a broad range of important disease outcomes in postmenopausal women to date.
WHI HT findings have been widely discussed and at times controversial, perhaps in part because their findings have been inconsistent with established clinical lore and practice, particularly with respect to the impact of HT on atherosclerotic diseases. WHI findings also caused some reconsideration of basic research programs predicated on the notion that HT was cardio- and neuroprotective in animal models. However, WHI methods and findings have stood up to substantial scrutiny and have, in my view, become the standard for complex, long-term trials of hormonal intervention in older women.
WHI comprised 2 HT trials: estrogen plus progestin (Prempro) for women with an intact uterus (E+P); and estrogen alone (E-alone) (Premarin) for women with prior hysterectomy. The trials were not intended to explore treatment of menopausal symptoms. Rather, they addressed the value of long-term HT in the prevention of major chronic conditions of women after menopause. The primary outcomes were all-cause mortality, coronary mortality, and several major conditions: coronary disease, stroke, breast cancer, hip and other fractures, pulmonary embolism, and colorectal cancer. In addition there was a global index of these disease outcomes.
In execution and interpretation, the WHI HT trials were complex, the discussion of which is beyond the scope of this column. Basic WHI HT trial findings are summarized briefly in the Table . Major outcomes, primary and secondary, were defined before trial initiation. An ancillary study of dementia outcomes (WHIMS: the WHI Memory Study) began within 2 years of WHI study onset. In general, the findings were similar across the age spectrum under study (50-79 years of age at entry), although there was some variation.
As the Table shows, some conditions were increased by HT while others were decreased. Neither intervention had an impact on total mortality. Coronary disease, breast cancer, and pulmonary embolism rates were significantly increased by E+P, but this was not seen with E-alone. Both interventions caused an increase in stroke rates and a decrease in hip fracture rates. E+P caused a decrease in colon cancer rates, but this was not seen with E-alone. Dementia rates were increased with E+P; a similar trend was seen with E-alone, but it was not statistically significant. There was a significant overall increase in dementia outcome rates when the findings from both trials were combined.
In the case of differential effects between E+P and E-alone, it has been suggested that the progestin (medroxyprogesterone acetate) may be responsible for some of these differences. This may be correct, but the issue will require further study, particularly because the 2 trial populations were dissimilar in some respects. For example, the E-alone participants had a higher mean body mass index and more years of prior exposure to HT than did E+P participants, so those who are interpreting trial differences must be cautious. Many of the findings were consistent with prior observational epidemiologic studies, with the most important exception being the cardiovascular outcomes. Also of note, in the E+P trial after 1 year, there were very few differences between intervention and placebo groups in a variety of quality-of-life measures, suggesting that in this population there were no important effects in the variety of physical and behavioral domains studied.
The WHI has made critically important contributions to our knowledge of disease prevention in postmenopausal women and has challenged current clinical practice and prevention policies. Given that some types of HT will likely cause some excess morbidity, a particular problem emerges in the treatment of menopausal symptoms, because few other effective treatments exist. Also, many women seek advice on the value of HT at the time of menopause, and provision of counseling is now more complex.
Perhaps most important, WHI has hopefully changed the decision process related to HT use at the time of menopause, and refocused attention on a very important issue: the best regimen for disease prevention in the patient's next 30 years of life. The US Preventive Services Task Force has recommended against HT for chronic disease prevention in postmenopausal women, reinforcing the need for an overall prevention focus.
More detailed analyses of WHI HT findings will be forthcoming. Additionally, 2 other WHI trials are not yet complete: long-term interventions with calcium/vitamin D, and low-fat diet. The WHI will continue to help guide our clinical preventive strategies and policies for several years to come.
[Dr. Wallace is the site principal investigator for the Women's Health Initiative at the University of Iowa. The opinions expressed are solely those of the author and not the Women's Health Initiative investigators, the National Institutes of Health, the American College of Preventive Medicine, or any other organization or association.]