New Study on Estrogen Yields Confusion but Same Recommendations
This week a study published in the Journal of the American Medical Association revealed new results about estrogen use and generated major media hubbub. (See the links below for some of the media coverage, the article and an accompanying editorial.) To understand the kerfuffle it helps to review how our understanding of estrogen replacement therapy has evolved.
A generation ago, based largely on intuition and on epidemiologic studies, we were convinced that long term estrogen replacement therapy was beneficial for women’s health. The theory made a lot of sense: Women before menopause have fewer heart attacks then men, but their risk of heart attack increases after menopause. Well, that’s just when their ovaries stop making estrogen, so estrogen must be protective. Like most untested things that make perfect sense, this was false.
But before we get too far ahead, we have to understand progesterone. Estrogen taken for a prolonged duration stimulates the lining of the uterus and increases the likelihood of uterine cancer. Estrogen taken with progesterone doesn’t have that risk. So for women who still have a uterus hormone replacement therapy (HRT) has come to mean combination therapy with estrogen and progesterone.
So finally in 1991, a randomized trial was done to test what everyone already thought they knew – the Women’s Health Initiative (WHI). Women were randomized to combination HRT and placebo. The results were that women on HRT had more strokes and at least as many heart attacks as the placebo group. There was much wailing and gnashing of teeth, from some doctors because they refused to believe it, from others because we realized that we had been harming countless women for a generation based on shoddy science.
Since WHI our understanding of the benefits and risks of HRT has been the following.
- HRT increases the risk of stroke
- HRT does not prevent heart attacks
- HRT effectively treats osteoporosis and prevents fractures
- HRT increases the risk of breast cancer
- HRT effectively treats the symptoms of menopause
Since there are far safer treatments for osteoporosis (the family of medicines called bisphosphonates), the evidence-based recommendation has been that HRT is only indicated for women who suffer from intolerable menopausal symptoms and should only be used in the lowest dose and the shortest duration possible. The goal of HRT is symptom relief and not any long-term health benefit.
After the WHI findings HRT use declined dramatically, though some physicians and researchers would not give up their infatuation with estrogen. They argued that it was the progesterone that was causing all the risks in HRT, and that estrogen alone would be safer.
Now remember that estrogen without progesterone increases the risk of uterine cancer, so it can only be given safely to women who have had a hysterectomy. A branch of the WHI randomized over 10,000 women with prior hysterectomy to estrogen alone or to placebo. This trial was stopped early because of the increased frequency of stroke in the estrogen group. (See link below to my post about this finding back in 2006.) This should have been the final nail in the coffin of HRT.
This week’s study adds very little to our knowledge. It followed the women in the estrogen-only trial for 10 years after the completion of the trial, meaning for 10 years after they stopped taking estrogen or placebo. The increased risk of stroke went away after the women stopped estrogen, but so did the decreased risk of fractures. Some small health benefits were found in younger women (in their 50s) taking estrogen, but far too much has been made of that since to achieve this small benefit the women first had to be exposed to an increased risk of stroke.
This week’s revelation does not change the indication for HRT in any way. It should simply make women in their 50s less anxious about the long-term dangers of short-term HRT taken only for symptom relief.
New York Times article: ” target=”_blank”>Estrogen’s Effects Tied to Age
My last review of the evidence on estrogen in 2006: ” target=”_blank”>Short-term Use of Unopposed Estrogen