Is the uneven reporting of HRT studies a “War on Women”??

Two patients asked me today about an article that came out this week in Annals of Internal Medicine, stating that the U.S. Preventive Services Task Force had recommended against the use of Hormone Replacement Therapy in menopausal women for anything other than short-term management of hot flashes or vaginal dryness. One of them in fact was quite upset—“This is my health we are talking about!”, she reminded me. What is frustrating about this is simply: How did these women hear about this paper—and yet did NOT hear about the randomized placebo-controlled trial of HRT that also came out this week showing tremendous benefit from HRT in preventing heart attacks and death (!) without any increased risk of cancer or blood clots?? Why is it so hard to get accurate information on HRT? I don’t know the answer to that… and truth be told, despite the inflammatory headline of this post, I’m not about to demand a Federal investigation into the reporting of HRT studies. I just want to get some useful info out into the world. Deconstructing reports like the U.S. Preventive Services Task Force paper is very time-consuming, because they lump a number of studies together without regard for some crucial distinctions between them. In fact, though most of the news releases completely missed this point, I am happy to say that at least one commentator got it right: Reproductive endocrinologist Dr. Wulf Utian (Case Western Reserve University) told Heartwire (10/23, Nainggolan):

This so-called ‘new’ review…hasn’t analyzed events by age. Where I think this document is weak is that it doesn’t look in context about the way preventive care should be considered or used for postmenopausal women. They should have broken the data down by decade, age 50 to 59, etc.’ Utian also pointed out that the updated guidelines fail to take into account data from the recent Danish Osteoporosis Prevention Study.

My thoughts exactly! Four gold stars for Dr. Utian… and four rotten tomatoes for the American Medical Association, who buried Dr Utian’s comments last among the six comments on the USPSTF paper. So what is Dr Utian talking about? Why would it be more informative to stratify the women “by decade, age 50-59, etc”? And what did the Danish study show?? I will absolutely answer those questions, but I want to try something even more ambitious: I want to try to give an overview of what ALL the HRT studies show, so that when the next inflammatory media report comes out, you, dear reader, are armed to analyze it yourself! So, let us travel back through time, to the Beginning: the year 2002, when the first results of the Women’s Health Initiative trial came out. (And everyone panicked.)

In The Beginning… there was the Women’s Health Initiative
To remind you: The Women’s Health Initiative was a huge prospective trial of HRT. It looked at a large group of women who used PremPro. PremPro is a combination of horse-estrogens (Premarin) and an altered, fake progesterone called Provera. The study compared these women to another large group of women who used ONLY Premarin (estrogens, but without the fake progesterone). Here are the results from the ESTROGEN-ONLY arm of the trial:

Results from the ESTROGEN-ONLY arm of the WHI. These women only took Premarin; they did not take Provera. (Click graph to enlarge.)

There are two things I want to point out on this chart:

  1. The risk of breast cancer– in all age groups– went DOWN. That’s right: contrary to what you hear in the news, taking estrogen alone without Provera actually DECREASED the risk of breast cancer. In the WHI trial, and in many other trials like it, the fake, non-bioidentical altered version of progesterone (called “progestins”) is what caused a small increase in breast cancer. Estrogen does not. Estrogen with bioidentical progesterone also does not increase breast cancer risk.
  2. In the women who were younger than age 60, the risk of heart attack and stroke went DOWN and their risk of “total death,” meaning death from any cause, went DOWN.

Isn’t that strange? You keep hearing in the news that HRT is bad, and that women shouldn’t use it… and yet here we see that women who started HRT before age 60 had a lower risk of death from any cause for the duration of the study. Surely that is just a fluke from this one study! Otherwise you would hear this trumpeted from every news station in the land, right?!? Actually, no, it’s not a fluke at all. In 2010 the Endocrine Society put out a position paper on HRT. The Endocrine Society is about as mainstream a group of doctors as you can find to comment on this topic. Here is what they said:

Menopausal Hormone Therapy was associated with a 40% reduction in mortality in women in trials in which participants had a mean age below 60 yr or were within 10 years of menopause onset.

Postmenopausal Hormone Therapy: An Endocrine Society Scientific Statement. JCEM July 1, 2010 vol. 95 no. 7 Supplement 1

That’s not just one trial—that’s their position paper from looking at ALL trials. I should get a T-shirt made: “Use HRT— Die 40% Less Often!” (Or something like that.) Is it really that simple? Almost… obviously there is some fine print, but I can help you through it. Let me give you Five Truths of HRT to guide you when you hear about these confusing studies:

Five Guiding Truths of Hormone Replacement Therapy

1. Earlier is better.
Starting HRT within ten years of menopause gives much greater benefit than starting later. As we saw in the WHI results above, in women who are younger than age 60, oral estrogens decrease the risk of both heart attack and stroke! In addition, starting HRT within eight years of menopause cuts your risk of Alzheimer’s disease in half. But didn't the Women's Health Initiative say that Estrogen causes Alzheimer's?? (Click to open.)

Despite the many positive studies on estrogen and memory (which can be viewed on my Hormone FAQ page, the Women’s Health Initiative reported in 2003 that HRT increased the risk of dementia in women 65 and older– as early as 12 months after starting therapy!–yet did not increase the risk of mild cognitive impairment. In other words, they seemed to say that women on HRT blew right past mild cognitive impairment and quickly became fully demented within a year of starting HRT!

(Shumaker SA et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women. The Women’s Health Initiative memory study: a randomized controlled trial. JAMA. 2003;289:2651–2662.)

How do we explain this? The answer is that the study group in the WHI were older women (average age 62) who had never been on HRT and thus had spent ten years since menopause building up plaques in their vessels. Then these women were placed on two drugs (Premarin and Provera) that are known to increase the risk of clots and strokes. These women did not suddenly develop Alzheimer’s: they developed mini-strokes in that first year on those drugs, and likely developed multiple-stroke dementia. That is why it appeared to be “sudden onset” Alzheimer’s: it wasn’t Alzheimer’s, it was small strokes. This is completely consistent with what we know about the WHI trial: within the first year of therapy (i.e. with PremPro, not bioidentical hormones), women over age 60 in the WHI trial who had not been on HRT previously did have an increased risk of strokes– because Premarin and Provera cause increased clotting and plaque rupture early on. This is why we instead use transdermal estrogen and bio-identical progesterone, both of which do NOT increase clotting or strokes. Further detail for those who like to understand the fine print:

In the Heart and Estrogen-Progestin Replacement Study (HERS), which also used PremPro and relatively older women (relative to menopause) just like the WHI: women who did NOT have a stroke or heart attack in that first year, got progressively LOWER risk of heart attacks/ strokes each subsequent year that they stayed on hormones. So, by year 4 and 5 on hormones, even those older women, who had never been on hormones until roughly age 62, and then began taking (relatively dangerous) Premarin and Provera, were at LOWER risk of stroke and heart attack than women who had never taken hormones. The reason is that Premarin seems to cause “reorganization” of plaques in your blood vessels (by increasing something called matrix metalloproteinase). Therefore in women who had plaque buildup for ten years since menopause (because they had no circulating estrogen/ weren’t on HRT), during that first year, this “reorganization” by Premarin increased their risk of plaque rupture and stroke/ heart attack. But if they made it past that first year, the plaque reorganization led to more stable and apparently smaller plaques that then were LESS likely to rupture every year on hormone therapy.

2. Using oral non-bioidentical estrogen (such as Premarin or ethinyl estradiol), at any age, will increase the risk of blood clots. When you swallow it, it gets metabolized in the liver, and increases the formation of clotting proteins. And using it together with fake, altered progestins increases the risk of clots even more. Using it in women with other risk factors for blood clots, such as obesity or smoking, raises the risk even further.

3. Transdermal estrogen (meaning through a cream or a patch) does not increase the risk of blood clots, in either older or younger women. Click here for further detail, if you like to understand the fine print...

Unlike horse estrogens (such as Premarin), bioidentical estrogen actually may not increase blood clots– even when taken by mouth. There are a few studies that suggest that, but the data is not fully resolved yet. To be safe, it is better to just use transdermal and not worry about it. If you want to read more about those studies, see my FAQ page under Hormone Replacement Therapy for Women.

4. Oral estrogen that is started more than ten years after menopause is more likely to cause a stroke or heart attack in that first year after starting HRT. The reason is, estrogen protects women from building up plaque in their arteries. After menopause, estrogen is not being produced– so unless she goes on HRT, she will start building up plaque. Therefore, if a woman has had ten years without any estrogen, she will have built up significant plaque in her arteries. If she then starts oral estrogen, the plaque that has formed starts to reorganize, and can become unstable in that first year — a piece can break off,  causing a heart attack or stroke. Click here for further detail, if you like to understand the fine print...

Oral estrogen increases something called matrix metalloproteinase, which seems to cause reorganization of plaque. Both the WHI and the HERS trial showed that if you make it past that first year without having a stroke or heart attack, then even if you stay on oral estrogen, your risk of heart attack or stroke decreases every year—and by year five, your risk is lower than someone who had never been on HRT. Side note #2: You can get away with using oral estrogens in women more than ten years after menopause without increasing the risk of stroke or heart attack, but only if the women are otherwise healthy: no hypertension, not obese, nonsmokers. BOTTOM LINE: This is the source of a lot of confusion as to whether HRT increases strokes or heart attacks. You can avoid this whole discussion by simply using transdermal estrogen.

5. Synthetic, altered progestins like Provera (medroxyprogesterone), norethindrone, and norethisterone, all increase the risk of breast cancer slightly. Estrogen with natural, bioidentical progesterone does not increase the risk of breast cancer. Estrogen given alone for HRT actually decreases the risk of breast cancer. Click here for further detail, if you like to understand the fine print...

Estrogen alone does not increase the risk of breast cancer, at least in trials where it is used for up to about ten years. But there is some indication that if you used estrogen alone –without progesterone or testosterone– for MORE than ten years, you might get a tiny increase in breast cancer risk. This is one of the reasons why we have our patients balance estrogen with natural progesterone, which seems to protect against breast tissue proliferation. For a review of many studies on this, see my Hormone FAQ page. The other reason we replace progesterone is: you need it to protect the uterus from proliferation caused by estrogen. Estrogen does not cause breast cancer, but it definitely CAN lead to endometrial cancer if it is not “opposed” by progesterone. We also replace testosterone to pre-menopausal levels. There is good reason to believe that that too helps reduce the risk of cancer, though no large studies have been done on that. For a review of some of the data we do have, go to the FAQ page under Testosterone for Women.

So, armed with those five Truths, you can now respond to most studies on HRT on your own! In fact, you can probably predict the results of the study before they come out:

  • Did breast cancer incidence increase? If so, they used a fake, altered progestin. They didn’t use bio-identical progesterone.
  • Was there an increase in blood clots? If so, they used oral estrogens in patients with some other risk factor for clots: either they were also taking altered progestins (which make the risk of clotting even worse), or the patients were overweight or smokers. If they had used transdermal, there would not have been an increase in clots.
  • Was there an increase in heart attacks or strokes? Then they did three things in combination that they should not have: They used oral estrogens rather than transdermal, and they used them in women who were already ten years post-menopausal when they started HRT, and those women were either hypertensive, obese, or smokers. If they had used oral estrogens in women who were within a few years of menopause and not hypertensive, the risk of heart attack or stroke would have gone DOWN, not up.
  • Did breast cancer incidence DECREASE? Then they most likely used estrogen alone, but may have used estrogen together with bioidentical progesterone.

So with that in mind, armed as you now are, let’s look at the Danish study that came out this past week that the news outlets completely ignored:

The Danish Osteoporosis Prevention Study is a large, prospective, placebo controlled trial on HRT (using bio-identical estradiol, no less!) The study included over 1000 women who were within a few years of menopause. They had mild high blood pressure (average BP 131/80) and, this being a European study, an astounding 43% were smokers. So, these women were not selected for being paradigms of good health. Half of the women received no hormones (the control group). The other 500 women all took bio-identical estradiol by mouth (rather than by transdermal patch or cream). If the women still had a uterus, they also gave them a non-bioidentical progestin (not real progesterone but an altered version), called norethisterone, to protect them from endometrial cancer—but, importantly, they only took norethisterone for ten days a month (the bare minimum they could get away with to protect the uterus), even though they took the estradiol every day. The women were treated for ten years, from about 1992 to 2002– and then, ironically, the study was halted because in 2002 the (inaccurately reported) results of the Women’s Health Initiative trial came out, and everyone began to panic about HRT. However, despite stopping the HRT, they followed the women for another six years to see what happened to them. So what happened?

Well, now you should be able to predict the answer! We see that they used oral estrogens in YOUNG women (meaning young relative to menopause—about age 50). So, sure enough, in accordance with Guiding Truth #1 above, the risk of heart attack and stroke decreased—in fact it decreased a LOT: As Dr. Raffaele pointed out in his analysis of the study last week,

These Danish women had over a 50 percent reduction in combined heart attacks, heart failure and death. Remarkably this reduction started to accrue very soon after initiation of therapy.

Okay, but what about the dreaded breast cancer?? These women were on HRT for ten years and were followed for 16 years total—surely they got breast cancer!! Actually, no. In keeping with Guiding Truth #5 above, the women who used estrogen-only had a decreased incidence of breast cancer. The women who took estradiol plus ten days per month of norethisterone had no change in their breast cancer risk—it didn’t go down, but it didn’t go up either (i.e. the norethisterone cancelled out the benefits of the estradiol in preventing breast cancer). Huh. So they didn’t get breast cancer (or any other cancer, by the way), and they had many fewer heart attacks… and everyone agrees that HRT protects them from osteoporosis… also we all know it stops hot flashes and makes women feel better… hmm but surely they died some horrible death, right?? Because HRT is terrible, I know that from the news! Well, remember that statement from the Endocrine Society back in 2010 that I mentioned? Here it is again:

Menopausal Hormone Therapy was associated with a 40% reduction in mortality in women in trials in which participants had a mean age below 60 yr or were within 10 years of menopause onset.

Postmenopausal Hormone Therapy: An Endocrine Society Scientific Statement. JCEM July 1, 2010 vol. 95 no. 7 Supplement 1

Those criteria fit the participants in this trial: They were all within ten years of menopause onset. So does that Endocrine Society statement hold true in this new trial as well? Did the women really have a 40% reduced risk of death from any cause?? Let’s see the results: over the ten years that they took HRT,

Deaths due to cardiovascular causes occurred in 18 women in the control group and five in the treated group. Deaths due to non-cardiovascular causes occurred in eight women in the control group and 10 in the treated group.

So, 26 women in the control group died, but only 15 of the women who received HRT died: Hormone Replacement Therapy reduced their risk of death from any cause by 42% for as long as they took it, exactly consistent with what the Endocrine Society saw in their 2010 review of HRT. After ten years, the women in this study stopped using HRT. But the authors followed the women for six more years. How did they do after stopping HRT? At the end of the 16 years,

In the control group there were 23 deaths due to cardiovascular causes and 17 due to non-cardiovascular causes. In the treatment group there were six deaths due to cardiovascular causes and 21 due to non-cardiovascular causes.

So: after 16 years, there had been 40 deaths in the women NOT on hormones (the control group), vs. 27 deaths in the women using HRT (ten years on HRT, 6 more not on HRT). In other words, their risk of death slowly headed back towards baseline after they stopped HRT, so that after six years not on HRT, their risk of death was lower by only 33%, not 42%. To put it another way: HRT = good; stopping HRT = bad. I hope you can now see why I am completely unmoved by the USPSTF (sounds like I’m trying to get a hair off my tongue) report. And the next time a paper on HRT makes the news headlines, you will be ready to comment on it!

Leave a Reply

Your email address will not be published. Required fields are marked *