Thursday, March 23, 2006

Will Hormones HuRT You?

In 2002, a study was published which claimed that hormone replacement therapy (HRT) increased the risk of stroke for older women. And this was no piddling fly-by-night affair - this was a randomized, double blind, placebo-controlled Women's Health Initiative study of 20,000 post-menopausal subjects conducted by the NIH. Since the investigators had hoped to show that estrogen had a protective effect on women's health, this unexpected result generated a lot of publicity and caused many alarmed women and their physicians to back off hastily from widely used hormonal treatments of menopausal symptoms.

But not so fast. How relevant are the results of this study to women in the real world? In a recent lecture at the University of Washington, Seattle, Dr. Phyllis Wise clued me in to the considerable caveats.

First of all, the women in the experimental group did not receive the standard doses of hormones used in treating menopausal symptoms; instead they were given relatively high, continuous doses of CEE (conjugated equine estrogen) along with MPA (which mimics progesterone and androgen and could interfere with estrogen). This was so that the investigators would be more likely to observe an effect, but was not an accurate re-creation of usual medical treatments.

Second, because this was a double blind study (no one was supposed to be able to tell who received treatment and who received placebo), only women who exhibited no menopausal symptoms were selected for the study. In other words, the investigators deliberately gave hormones to a group of women who would NOT normally seek hormone replacement therapy. Therefore, there is every reason to believe that the experimental group was not representative of the population of women who undergo HRT, meaning that results from this study cannot safely be generalized to the patient population.

Most critically, the participants were women aged 50-79 (average 63) years at the beginning of the study, and many of them had never received hormone replacement therapy. Prior to the experimental treatment, these women had no estrogen circulating in their bodies for an average of TWELVE YEARS.

Why is this so important? Because, biologically speaking, as many lactose-intolerants and former high school athletes have realized, if you don't use it, you lose it. Estrogen, like other hormones, is a signal that has to be received in order to work. Only tissues that make a protein receptor for the estrogen molecule can be instructed by it. And guess what - if there hasn't been any estrogen around for years, the body stops wasting its resources on making estrogen receptors. After that, taking estrogen won't help a damned thing. It's like holding up a stop sign to a driver who has his eyes closed.

This idea is supported by a really cool series of studies by Dr. Wise's group in an animal model for stroke. Researchers removed the ovaries from rats to mimic menopause and later used a suture to block a cerebral artery, temporarily stopping bood flow to one side of the brain. If the rats were given estrogen treatments immediately after removal of their ovaries, they suffered much less tissue death upon blockage of blood flow. However, if estrogen treatment was delayed (for a time equivalent to the years human women spent between menopause and hormone treatment in the WIH study), there was no longer any health benefit. This was because the estrogen receptor (ER-alpha) that mediated the effect of estrogen on brain tissue damage, and would normally be produced at high levels in response to brain injury, was no longer present following prolonged estrogen deprivation.

So the authors of the famous WHI study can't conclude anything about the possible health risks or benefits of standard hormone replacement therapy for women who come to their doctors complaining of menopausal symptoms. No matter how large the study, if it can't accurately represent the relevant treatment method or patient population, then the results cannot reliably inform our medical decisions.


References

Writing Group for the Women’s Health Initiative Investigators, Risks and benefits of estrogen plus progestin in healthy postmenopausal women principal results from the Women’s Health Initiative randomized controlled trial, JAMA 288 (2002), pp. 321–333.

P.M. Wise (2006). Estrogen therapy: Does it help or hurt the adult and aging brain? Insights derived from animal models. Neuroscience 138(3):831-5.

Wednesday, March 22, 2006

Just the Fats, Ma'am

In a recent widely publicized Women's Health Initiative study, the NIH found no significant differences in the incidence of breast cancer, heart disease, or stroke for women following a low-fat diet. This pronouncement was met with surprise by many clinicians (and one suspects with glee by the National Restaurant Association). But should we really conclude that years of warnings about reducing fat intake were mistaken?

Too little?
The study was meant to compare one group of women who had reduced their total fat intake to 20% of dietary calories with a control group who got 35-37% of their calories from fat. Investigators expected that such a substantial difference in fat consumption would reveal any impact of fats on women's health. And it might have... except that the low-fat group never met the study's goal. The women managed to reduce their fat consumption to 24% of total calories in the first year, but by year six were up to 29%, a difference of only 8% of calories from the control group*. This might not have been enough to affect their chances of developing cancer, heart disease, or stroke, even if fat intake were an important factor. In fact, the subset of women who reduced their fat intake the most did seem to reduce their risk of breast cancer.

Too late?
The participating women were aged 50-79 years at the start of the study and followed for about 8 years. This means that they were already old enough to be at fairly high risk for various cancers as well as heart disease before the changes in diet. These illnesses develop over time, and it is still quite possible that a low-fat diet earlier in life could help to prevent them.

Good fat, bad fat
Most troubling was the failure of this study to distinguish between the various types of fat that make up total fat intake. Saturated fats (mainly from animal products like red meat and dairy), and especially trans fats (from hydrogenated oils), have long been linked to increased risk for heart disease. But unsaturated fats, from fish, nuts, and vegetable oils, are considered "good fats" that lower risk for heart disease. So instructing the participants specifically to reduce their trans and saturated fat intake (and maybe even increase unsaturated fat) could have produced very different results.

All in all, just the kind of situation that confuses everyone and makes me crazy. No wonder people lose faith in the ever-fluctuating recommendations of scientists and clinicians. Even studies like this - huge, well-funded, controlled studies that should be setting the standards - have built-in design flaws that finally allow us to conclude... absolutely nothing.


*This made me grateful that I study the effects of diet in fruit flies and not human beings. If I want to look at brain development on a protein-free diet, my experimental subjects will stick to sucrose, and dammit, they'll like it.

Monday, March 13, 2006

Dealing with India

On March 7, the New York Times ran an editorial that lamented President Bush's recent trip to India and Pakistan. The authors contend that, while a similar offer should not have been made to our ally Gen. Musharraf, who pardoned the Pakistani scientist that helped transfer nuclear technology to leading rogue states, "Granting India a loophole that damages a vital treaty and lets New Delhi accelerate production of nuclear bombs makes no sense either."

The treaty in question is, of course, the Nuclear Non-Proliferation Treaty, signed by the five nations possessing nuclear weapons in 1968 as well as the vast majority of the remaining sovereign states, including both Iran and North Korea (which has since withdrawn). It was not signed by India, Pakistan, or Israel, all three of which have since developed nuclear technology and are believed to possesss more than 60 missiles apiece. Although the US signed the treaty, we were engaged in secret weapons sharing agreements with non-nuclear members of NATO, and still provide nuclear bombs to Belgium, Germany, Italy, the Netherlands, and Turkey, thus violating Articles I and II. (This fine example of global leadership fits right in with our refusal to sign the Comprehensive Nuclear Test Ban, along with China, Israel, India, and Pakistan.)

So the US is hardly in a position to take the moral high ground when it comes to nuclear non-proliferation. Clearly we are willing to operate by a double standard in our own interests - or, let's say, in the interest of global stability - by providing trusted democratic allies with weapons of mass destruction.

And the new deal with India doesn't even go that far. The proposed agreement 1) does not introduce nuclear technology to a non-nuclear power, 2) provides nuclear materials for civilian energy use, not military purposes, and 3) encourages India to reclassify many existing nuclear plants as "civilian" and subject them to monitoring by the IAEA.

In fact, my main question is, why has it taken so long for the US to create official agreements like this with the world's largest democracy? Apparently we're still suffering the after-effects of the cold war strategy that involved support of military coups (especially in Central/South America) in opposition to those left-leaning legitimately elected leaders who might turn out to be (cue ominous music)... COMMIE SYMPATHIZERS!!! And India, which chose to pursue many socialist policies as a fledgling nation to keep itself from being economically overrun by more powerful nations, has been regarded with similar suspicion... in contrast to neighboring Pakistan, currently ruled by the military dictator who usurped their democratically elected leader.

And let's face it, having Musharraf on our side really hasn't accomplished very much in the fight against Al-Qaeda. Although, as our President said recently, finding Bin Laden isn't really that important anyway.

Because clearly it's much more critical to eliminate a sovereign government with no links to 9/11 or Al-Qaeda - a secular regime, however abhorrent, that had been providing a critical balance against fundamentalist Iran. And to maintain good relations with the Sauds and Emirati who supplied 17 of the 19 identified 9/11 hijackers.

***
Declaration of competing interests: I am currently dating an Indian national, discussions with whom provided the basis for this post.