I am fairly confident that most women—certainly those post-menopausal or peri-menopausal—are aware of the extensive media coverage and dire warnings following the release of the results of the Women’s Health Initiative (WHI) study on hormone replacement therapy (HRT) in 2002.  At that time, it was stated that HRT is detrimental to a woman’s health, with risks outweighing the benefits.  It stated, pretty unequivocally, that HRT increased risk of breast cancer, cardiac events and stroke.

It would be overstating to say that all of the 2002 results were inaccurate, since, as we know, science is rarely definitive and more information is constantly emerging and being revised; however, women should know that many of the initial results have been found to have been distorted, misunderstood, over-generalized, and in some cases flat out wrong.  While the updated findings have been published in various scientific papers and journals, the mass media continues to refer almost exclusively to the 2002 WHI negative results.  Having received the attention they wanted, they have moved on.  But what if some of the findings turned out to be wrong…

A few recent examples highlight this continuing problem of misinformation:

-The New York Times recently published an article on what is currently known about what causes cancer and what prevents it.  The underlying thesis was that we know very little, but two things we do know:  first, that smoking causes cancer (okay, we can all agree on that); second, that HRT causes breast cancer.

-A November 29 article on sex and menopausal issues, also in The Times, stated that many women stopped taking hormone therapies because of their link to small increases in breast cancer, heart attacks and strokes.

-A November 12 article in HealthDay News posited that the “declining use of HRT” may be driving down rates of a condition called atypical ductal hyperplasia—a known risk factor for breast cancer.  Other articles have gone even further, to say that the incidence of breast cancer is down for the same reason.

Seemingly no one in the mainstream media is quoting the analyses and research that questions all of the 2002 results.  For example, in an article published in The Cancer Journal in 2009, Dr. Avrum Bluming, Clinical Professor of Medicine at the University of Southern California and Master of the American College of Physicians, along with Dr. Carol Travis, debunk many of the previous 2002 research findings.  Their work joins a growing list of reports on the WHI results and newly published research studies, chipping away at the theory that HRT is bad for all women, all of the time.

There have been studies on the relationship between hormones and breast cancer since 1942 and the results, predictably, have been mixed.  The myriad of results has the pendulum continuously swinging between HRT having protective or harmful effects.  The information we have to base medical decisions on is only as good as the research seeking answers.  The better quality of research, the stronger the evidence to support taking certain medical actions.

One area where there is a growing body of evidence contradicting the 2002 WHI results is on the impact of HRT on developing heart disease.  The conclusion of most cardiologists remains that there is no reason for women to take hormones primarily to help forestall or prevent cardiovascular disease.  There are risks for women who begin taking HRT in their 60s, especially during the first year, and especially if they have any pre-existing artery disease.    However, newer studies and analysis of the data are showing that certain groups of women, in fact large numbers, may experience cardiovascular benefits from HRT use.

In 2007, the WHI investigators in a secondary analysis of the WHI suggested that women who start short-term HRT closer to menopause may actually reduce their risk of coronary heart disease compared to the increased risk seen in women starting HRT farther from menopause.  While these WHI findings, like those in 2002, did not reach statistical significance, it did introduce the “timing hypothesis”, an increasingly popular suggestion that proximity to menopause may be the key factor in determining when HRT transitions from being helpful to being harmful, at least in the case of CHD.  Additional research on the timing hypothesis and HRT use and cardiovascular health continues:

-Animal studies suggest that continuous estrogen keeps blood vessels healthy, but that estrogen replacement after a hormone-free interval post-menopause cannot reverse vascular damage done (Cardiovascular Res 2002).

-Additionally, just two years after the WHI report, the Nurses Health Study, an ongoing observational study of 121,000 nurses dating back to 1976, reported a 40% decrease in heart attacks in hormone users, 90% of who started therapy within 4 years of menopause.  Looking at those nurses who started HRT 10 or more years after menopause (as the vast majority of women in the WHI study did), a 37% increase in relative risk for heart attack is observed, similar to the WHI findings.

One area where the results are less clear, and confusion remains, is on the risk of developing breast cancer after using HRT.  To give just a few examples:

-Dr. Bluming states in his analysis of the 2002 WHI results that many of the increases in risk that were reported either did not reach statistical significance, or did not report data sufficiently so as to calculate whether a rise was significant.

-In 2006, another update of a follow-up study from the WHI cohort reported no increased incidence of breast cancer in the women randomized to get the estrogen/progesterone combination.  While one, given the 2002 HRT findings, may have expected to also find a greater risk of early, non-invasive forms of breast cancer in women taking estrogen/progesterone, it did not.

-Additional estrogen-alone studies reported in 2006 in the Journal of the American Medical Association have found no increased risk of breast cancer, and, in some cases, possibly cause a slight decrease. This same year, the International Menopause Society went on the record, stating that newer studies sent a “very clear message that ET [estrogen therapy] for postmenopausal women does not increase the risk of breast cancer”, and, for some users might even prove protective.

-A 2008 report in the Journal of the National Cancer Institute reported that in trials where estrogen was given to high-risk women with the implicated BRCA1 gene mutation they did not find an increased risk of breast cancer.

-Decreases in breast cancer rates started in 1999 and a 2009 report of decreases in breast cancer deaths are trends that can be traced back to starting in 1990—long before the first publication of the WHI results and the sharp decreases in HRT use that followed.

-At a 2009 meeting of the American Association for Cancer Research, it was reported that between 2002 and 2003 there was a 7 percent drop in breast cancer incidence for American women.  A decline of 3 percent is attributable to change in hormone therapy, leaving other pathways and influences responsible for a 4 percent decrease.

- A 2008 report at the San Antonio Breast Cancer Symposium suggested that using hormones in excess of 5 years can double the risk for breast cancer.  Timing and duration of HRT are proving to be key factors in determining when the medicines are protective and when they may be harmful.  It remains advisable to take the lowest dose necessary for the shortest amount of time necessary.

Let’s turn to stroke and blood clots.  Both are related to vascular health, and hormonal impact on vessels and blood clot development has been studied, going back decades.

-The 2007 re-analysis of the WHI by its investigators reconfirmed a link between HRT and risk of stroke, unaffected by proximity to menopause.

-A 2007 review article by Drs. Howard Hodis and Wendy Mack found that stroke and venous thromboembolism (a form of blood clot) in women using HRT are rare, and even rarer when hormone therapy is initiated in women less than 60 years old and in close proximity to menopause.  Dr. Hodis in a 2008 piece in the Cleveland Clinic Journal of Medicine reports that the new data has changed recommendations, with the benefits of HRT exceeding risks when initiated in menopausal women younger than 60 years.

-When the estrogen-only arm of the WHI was ended in 2004 due to an increased stroke risk, women were already being inundated with the risks of HRT.  The potential for women over-reporting subtle neurological deficits, fitting into the WHI’s broad definition of “stroke”, may have artificially raised the numbers.   When critics of the WHI controlled for this detection bias they found that the reported increased risk of stroke for women on HRT vanished.

On a final point, Dr. Bluming in his analysis also questions the increase in dementia and onset of Alzheimer’s found in the 2002 WHI report, noting that when women who had mild cognitive impairment already at the start of the WHI study are excluded, the results implicating HRT in cognitive impairment become statistically insignificant.

As I said at the onset, there is always more research needed and no one would ever state unequivocally that these or any results are final.  A more precise knowledge of when and for whom HRT may be helpful will only come if we learn lessons from all the science available.  My advice to women is to do your research and speak to your doctor.  Much of what you currently think, and much of what he or she may think about HRT, may no longer be accurate.

This entry was posted on Tuesday, January 12th, 2010 at 9:02 am and is filed under Women's Health. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.