Was The Women's Health Initiative Good Or Bad For Women's Health?
Most women are familiar with the Women’s Health Initiative (WHI), the largest randomized controlled trial to date, sponsored by the National Institute of Health (NIH) to evaluate the role of hormone therapy in menopause to protect cardiovascular and bone health. Begun in 1991 as a proposed 15‐year study, women in menopause with a uterus were randomized to take orally either a placebo or PremPro®, a combination of Premarin®, a conjugated equine estrogen (CEE), and medroxyprogesterone, a synthetic version of progesterone. Women with a hysterectomy were given either CEE alone or placebo. In part, this $725 million study was intended to resolve the controversy over whether menopause should be embraced as a natural transition in life, a position taken by the feminist movement at the time, or, as proposed by such books as Feminine Forever (Pocket Books, NY, 1968), that menopause was a hormone deficiency totally preventable with hormone therapy.
In 2002, the entire study was abruptly stopped due to a statistical increase in breast cancer and stroke and no apparent benefit for reducing cardiovascular risk. This bold action by the NIH prompted the New York Times article entitled, Hormone Replacement Study: A Shock to the Medical System. As one physician later wrote, “I may have taken my last pill this morning.” By 2003, there was a precipitous reduction in hormone prescriptions, ushering in a decade of menopausal women without hormone support. But was the WHI study done correctly?
The purpose of hormone replacement in menopause is to compensate for the lack of estrogen, primarily estradiol, of which 95% is produced by the ovaries during the reproductive years. The role of a progestin is to duplicate progesterone, produced cyclically by the premenopausal ovaries, whose sole purpose is to prevent overstimulation of the uterine lining by estradiol.
Three aspects of the WHI deserve closer scrutiny: the age of the women in the study, the choice of the hormones used, and the significance of the statistical risk.
With an average age of 63 years, many of these women may already have had vascular damage from age‐related changes, thus increasing their risks for cardiovascular events. The average age of menopause in the United States is 51 years.
In the WHI, Premarin®, derived from the urine of pregnant mares (Pre‐mar‐in) contains ten different estrogens but almost no estradiol, the most powerful of the body’s estrogens. Taken as an estrogen pill, it is absorbed in the intestine, with a “first pass” through the liver, increasing production of many clotting factors. This explains the increased incidence of stroke in both arms of the study. Conversely, transdermal estrogen does not stimulate these clotting factors. Medroxyprogesterone was used as the progestin. This is a synthetic chemical version of progesterone. Women on PremPro® did show a statistical increase in breast cancer. Yet, those women with a hysterectomy, and therefore taking Premarin® only, showed a reduction in breast cancer mortality. Was it the progestin? Recent data suggests that medroxyprogesterone in breast tissue may alter hormone receptors adjacent to the estrogen receptors, thus increasing the risks of estrogen‐stimulated cancers.
Applying statistical risk to clinical risk also is a challenge. In the WHI, out of 10,000 person‐years, there were seven more cardiac events, eight more strokes, eight more pulmonary emboli, and eight more breast cancers. The WHI frightened women (a negative), but it laid the groundwork for improved menopausal care (a positive). Today, there is a growing sense that if women are started on hormone replacement in their late 40s or early 50s and are administered pure estradiol by a patch or cream and pure progesterone, that the original hypotheses of the WHI might have been better served.
By James Woods, M.D., and Elizabeth Warner, M.D.
Dr. Woods treats patients for menopause at the Hess/Woods Gynecology Practice.
Disclaimer: The information included on this site is for general educational purposes only. It is neither intended nor implied to be a substitute for or form of patient specific medical advice and cannot be used for clinical management of specific patients. Our responses to questions submitted are based solely on information provided by the submitting institution. No information has been obtained from any actual patient, and no physician‐patient relationship is intended or implied by our response. This site is for general information purposes only. Practitioners seeking guidance regarding the management of any actual patient should consult with another practitioner willing and able to provide patient specific advice. Our response should also not be relied upon for legal defense, and does not imply any agreement on our part to act in a legal defense capacity.
James Woods |