Does Stopping Menopausal Hormone Therapy Have Cardiovascular Consequences?

Summary and Comment |
October 27, 2015

Does Stopping Menopausal Hormone Therapy Have Cardiovascular Consequences?

  1. Andrew M. Kaunitz, MD

Findings from nationwide Finnish analysis are not consistent with randomized trial data.

  1. Andrew M. Kaunitz, MD

In a recently published large observational study from Finland, investigators concluded that stopping menopausal hormone therapy (HT) raises risk for death from cardiovascular disease (CVD; including cardiac and cerebrovascular events). Using nationwide data from 1994–2009, they analyzed CVD mortality in 332,202 women who discontinued HT compared with expected CVD mortality in the background population.

During the first year after HT discontinuation, elevations in cardiac and stroke mortality were observed (standardized mortality ratios, 1.26 and 1.63, respectively), while in subsequent years, reductions in such mortality occurred compared with the general population (P<0.05 for all comparisons). The absolute increased risk for CVD death reported in the first year after discontinuing HT corresponded to 4 deaths/10,000 women-years of exposure; the absolute risk for stroke death was 5 additional events/10,000.


In contrast to these data, findings from the Women's Health Initiative large randomized trial of HT did not show changes in mortality among women who stopped HT (JAMA 2008; 299:1036, JAMA 2011; 305:1305, and JAMA 2013; 310:1353). Limitations of the observational Finnish data probably account for this discordance. For example, the investigators did not know why women discontinued HT, raising the possibility that those who had symptoms suggesting CVD (or who developed new risk factors) preferentially stopped HT, potentially introducing bias. Women and their clinicians should opt to continue, lower the dose, or discontinue HT through shared decision making that focuses on quality of life as well as individual concerns about risk for cancer, CVD, and osteoporosis. Dramatic as they are, the results of this report should not affect how we counsel women about use or discontinuation of HT.

Dr. Kaunitz thanks Dr. JoAnn E. Manson and Dr. Cynthia A. Stuenkel for their valuable contributions to the preparation of this summary.

Editor Disclosures at Time of Publication

  • Disclosures for Andrew M. Kaunitz, MD at time of publication Consultant / Advisory board Actavis plc; Bayer AG; Merck Royalties UpToDate Grant / Research support Therapeutics MD; Bayer; Agile; NIH Editorial boards Contraception; Menopause; Contraceptive Technology Update; OBG Management; Medscape OB/GYN & Women’s Health Leadership positions in professional societies North American Menopause Society (Treasurer)


Reader Comments (6)

Kathy Maupin MD Physician, Obstetrics/Gynecology, Private Practice

It is astounding that there is a disclaimer attached to this study, to doctors not to use this information in treating women. This is just one of the hundreds of studies that reveal life saving effects from HT, especially ET. We SHOULD used medical studies to decrease mortality in female patients! If we don't why are we studying the effects of hormone replacement at all? The national colleges are so concerned about protecting their doctors from law suits they have forgotten why we practice medicine! The evidence for R estrogen replacement is in refutable and soon doctors will be sued for NOT providing HT for their menopausal patients!

Tomi S Mikkola, MD Physician, Obstetrics/Gynecology, Helsinki University Hospital

The NEJM Journal Watch summary and commentary of our large nationwide study (Mikkola TS et al. J Clin Endocrinol Metab early release 2015, Sep 28) state very frankly that the results are in contrast to Women’s Health Initiative (WHI) data and thus, “Limitations of the observational Finnish data probably account for this discordance”. This indicates a preconception that new studies may not challenge the “WHI dogma”. This should not be the mind set in science.

We fully agree that a randomized placebo controlled trial is the “golden standard” in clinical studies, but the results can be interpreted only to similar populations as in the trial. In contrast to our study population, the women in WHI trial did not represent the typical clinical setting where recently menopausal symptomatic women need hormone therapy (HT). In WHI HT was initiated to elderly (in average 63 years of age) women with little or no symptoms, and they were obese (BMI 29) with various existing cardiovascular risk factors (e.g. 36% with hypertension) (WHI Investigators JAMA 2002;288:321-33). Furthermore, WHI-trial took on characteristics of an observational study as participants became aware of their treatment assignment from unblinding (Shapiro S. Climacteric 2003;6:302-10, Climacteric 2007:10suppl2:2-12). To partially overcome these fundamental problems with the trial, various sub-analyses have been conducted within the WHI-study showing e.g. that HT is beneficial for cardiovascular system if initiated to healthy women soon after menopause (Rossouw et al. JAMA 2007;297:1465-77). These results are again in line with various observational studies including our study population (Mikkola et al. Menopause 2015;22:976-83).

The current NEJM Journal Watch statement that “Findings from nationwide Finnish analysis are not consistent with randomized trial data” is not fully correct. In WHI post-trial follow-up, mortality was increased within the three years of cessation of the estrogen+progestin (E+P) arm of the trial in the women who were assigned to E+P relative to those who were assigned to placebo (HR=1.15; 95% CI, 0.95-1.39) (Heiss et al. JAMA 2008;299:1036-45). Importantly, mortality was significantly increased in post-trial follow-up in women who were originally assigned to E+P relative to those who were assigned to placebo and were at least 80% compliant with intervention (HR=1.53; 95% CI, 1.04-2.24) (Heiss et al. JAMA 2008;299:1036-45). In addition to the WHI studies, data are available from at least two other randomized controlled trials addressing the issue of HT withdrawal. In the Heart and Estrogen/progestin Replacement Study (HERS) II, the unblinded post-trial 2.7-year follow-up to the HERS trial, women originally assigned to E+P had a 3.3-fold higher rate of ventricular arrhythmia requiring resuscitation compared with those women assigned to placebo (HR=3.30; 95% CI, 1.08-10.1) (Grady et al. JAMA 2002;288:49-57). During the first 6 months of post-trial follow-up of the Women’s Estrogen for Stroke Trial (WEST), there were three fatal strokes and 18 nonfatal strokes among the women originally randomized to estradiol therapy relative to one fatal stroke and 8 nonfatal strokes among the women originally assigned to placebo (HR=2.3; 95% CI, 1.1-5.0; p=0.03) (Viscoli et al. N Engl J Med 2001;345:1243-9).

We detected that women who stopped HT relative to women who continued HT had a 2.3-fold (95% CI, 2.12-2.50) greater risk of cardiac death within the first post-HT year and a 1.3-fold (95% CI, 1.21-1.31) greater risk of cardiac death more than one year after stopping HT (Mikkola TS et al. J Clin Endocrinol Metab early release 2015, Sep 28). In addition, women who stopped HT relative to women who continued HT had a 2.5-fold (95% CI, 2.28-2.77) greater risk of dying from stroke within the first post-HT year and a 1.3-fold (95% CI, 1.19-1.31) greater risk of dying from stroke more than one year after stopping HT. We believe that these data substantially furthers our understanding of the post-trial data from WHI, HERS and WEST since the totality of data is more important than talking about strengths and weaknesses of individual studies.

It will be of interest to see if the unconditional “WHI dogma” will pass a test of time, when increasing number of new studies challenge it. Several observational studies, randomized controlled trials and meta-analyses indicate that when initiated close to menopause, HT reduces mortality. Also cumulative data support that HT withdrawal has potentially detrimental implications for women. In total, the data are highly informative for the counselling and use of HT in the prevention and treatment of our menopausal patients.

Henry Lindner, MD Physician, Endocrinology, Pennsylvania

The authors of the study failed to emphasize that these women were taking estradiol, not conjugated equine estrogens (i.e. Premarin) as in the WHI study. Had they done so Dr. Kaunitz might have drawn a different conclusion. Estradiol is human estrogen; it should be expected to have superior benefits and less risk than non-human estrogen molecules until proven otherwise. Many studies show that it has a more beneficial effect upon various cardiovascular risk factors than Premarin, especially when taken transdermally. Indeed, the professional associations like NAMS, Endocrine Society and ACOG have lost credibility in this field, so vital to women's health, because they continue to attribute all the problems causes by various hormone substitutes to women's natural hormones.

Vikki Stefans, MD Physician, Pediatric Subspecialty, Academic/Private Not-For-Profit

I would also add that I agree we do not have enough assessment of estradiol versus synthetic estrogens; it is clear that transdermal estrogen is safer than oral*, but unclear if estradiol being the usual transdermal form versus a variety of estrogenic compounds available orally, again in combination with a variety of progestogens. Estrogen receptors accommodate a wide range of molecules with a wide variety of stimulatory, blocking, and antagonistic effects of varying strengths. The different types of estrogen receptors in different types of cells makes it a highly complex subject. The transgender MTF literature is very interesting in this regard.

"VTE risk may be lessened by use of transdermal estrogen in MTF adults....The “first-pass” hypothesis of liver metabolism of estrogen proposes that there is a decrease of thromboembolic and other cardiovascular events with the use transdermal as opposed to oral estrogen therapy [[10]

Vikki Stefans, MD Physician, Pediatric Subspecialty, Academic/Private Not-For-Profit

I think Darlene Zimmerman is basically right. Also, ethinyl estradiol is probably pro-coagulant as well.
It may depend on exactly which hormonally active molecules are used, and host factors; I'm not sure why women with cardiac symptoms would be more likely to stop therapy, but women whose general well-being is more dependent on it could be more likely to continue it.

Zimmermann, Darlene MPH Other Healthcare Professional, Preventive Medicine, Home

WHI was an extremely flawed study:The women in the study were too long passed menopause so that disease processes were already in play. Keeping in mind that HRT is preventative, not therapeutic!!! Also the type of HRT used in the study, Prempro, continuous combined, has been shown to be
associated with Cancer by European researchers, particularly Swedish studies. Further, the culprit in HRT is the synthetic progestin, a molecule with different chemical bonds than the progesterone made by a woman's body. Progestin ,again in European studies, has been shown to be detrimental to women's bodies...not the Estrogen.When progesterone molecules with the same identical chemical bonds as progesterone made by the ovary is used ,the results are dramatically different.A simple google search can provide attribution to my declarations. Most importantly, why have bio-identical vs synthetic HRT not yet been quantified in double blind placebo studies? For NAMS to disparage bio-identicals and to state that they are harmful to women, is an opinion not supported by well designed research studies and undermines the credibility of the entire Society and its member gynecologists.

Your Comment

(will not be published)

Filtered HTML

  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Do you have any conflict of interest to disclose?
This question is for testing whether you are a human visitor and to prevent automated spam submissions.

Vertical Tabs

* Required

Reader comments are intended to encourage lively discussion of clinical topics with your peers in the medical community. We ask that you keep your remarks to a reasonable length, and we reserve the right to withhold publication of remarks that do not meet this standard.

PRIVACY: We will not use your email address, submitted for a comment, for any other purpose nor sell, rent, or share your e-mail address with any third parties. Please see our Privacy Policy.