Harvard Women's Health Watch

Hormone therapy: The next chapter

Is it time to rewrite the book on postmenopausal hormone use?

What a difference a decade makes! When the new millennium dawned, hormone therapy (HT) was a mainstay of postmenopausal health care. Besides relieving hot flashes and other troublesome symptoms, it was thought to offer protection against a host of degenerative disorders. In 2001, perimenopausal women were commonly urged to consider HT, not just for symptom relief but also for benefits that included protection against osteoporosis and possibly heart disease, colon cancer, and Alzheimer's disease. Its risks were thought to be limited — mainly increased susceptibility to breast cancer and gallbladder disease. Today, however, HT is linked not only to these conditions but also to an increased risk for stroke, blood clots, and Alzheimer's disease. Amid the growing evidence of harmful effects, many women are loath to consider HT even for short-term symptom relief.

Despite this history of dashed expectations, it's too early to close the book on HT. In particular, there's still little evidence that short-term use for symptom relief does any harm. And some experts believe that it may help protect younger women against heart disease.

Chapter 1: High hopes for hormones

The high expectations for HT were based on more than 50 years of experience and observational studies in which researchers compared the medical records of women who used hormones with those who did not. The most convincing data came from the Nurses' Health Study (NHS), which began in 1976 and included 122,000 women who were ages 30 to 55 at the start of the study. Every two years, the nurses completed detailed questionnaires about their lifestyles and health. By the mid-1990s, reports from the study suggested an increased incidence of breast cancer and gallbladder disease among women who had taken HT for several years. Long-term users also seemed to get some health benefits, particularly a reduced risk of cardiovascular disease. But these trials weren't conclusive, partly because women who chose to take hormones might have been healthier to begin with.

To eliminate this "healthy woman" bias, the National Institutes of Health and drug companies conducted clinical trials in which women with similar health profiles were assigned at random to HT or a placebo. The first such study — the Postmenopausal Estrogen/Progestin Intervention (PEPI) Trial, which involved 875 women, ages 45 to 64 — seemed to confirm the NHS findings. In 1995, the researchers reported that HT raised HDL (good) cholesterol and lowered both LDL (bad) cholesterol and circulating levels of the clotting factor fibrinogen, which, at higher levels, contributes to cardiovascular disease. It also increased bone density. But PEPI didn't determine whether these improvements paid off in fewer heart attacks, strokes, blood clots, or bone fractures.

Chapter 2: Shadow of a doubt

Then, in 1998, there came a sharp surprise — the results of the second large controlled study, called the Heart and Estrogen/Progestin Replacement Study (HERS). HT didn't prevent heart attacks in 2,763 postmenopausal women with heart disease. In fact, participants taking hormones had 50% more heart attacks during the first year. But many experts speculated that the hormone's benefits simply came too late for women who already had angina or heart disease. They thought the physiological changes produced by HT in the PEPI study would show up as better health outcomes in larger, longer trials.

Risks and benefits of estrogen alone in WHI studies

Condition

Increase or decrease in relative risk

Increase or decrease in cases per year*

Ages of women in study**

Gallbladder disease

67%

31 more

50–79

Dementia

49%

12 more

65–79

Blood clots in the legs

47%

4 more

50–79

Stroke

39%

12 more

50–79

Blood clots in the lungs

34%

3 more

50–79

Kidney stones

19%

5 more

50–79

Colon cancer

8%

1 more

50–79

Heart disease

9%

6 fewer

50–79

Diabetes

12%

8 fewer

50–79

Breast cancer

23%

7 fewer

50–79

Osteoporotic fractures

39%

56 fewer

50–79

*Per 10,000 postmenopausal women
**70% of participants were ages 60–79.

Chapter 3: The tide turns

Those trials came under the umbrella of the Women's Health Initiative (WHI), which was launched in 1991 and involved more than 161,000 women ages 50 to 79. The WHI investigated strategies for preventing chronic disease in postmenopausal women and was designed to deliver the definitive word on HT. In more than 27,000 women, the WHI's hormone trials evaluated the two most common HT regimens — estrogen plus a progestin (synthetic progesterone) in women with an intact uterus, and estrogen alone in women who'd undergone a hysterectomy. The study was planned to last until 2005, but by 2002 it was apparent that — despite reductions in colorectal cancer, hip fractures, and spinal fractures — the women taking the estrogen/progestin combination had more, not fewer, health problems, including significant increases in heart disease, stroke, blood clots in the veins, and blood clots in the lungs. As expected, the women on combined therapy also had an increased risk for breast cancer — and the cancers detected in HT users tended to be larger and more advanced than those diagnosed in women taking a placebo.

In 2004 the other arm of the study was also halted, because the women taking estrogen alone had a higher rate of stroke than the women on placebo. Research on estrogen's cognitive impact appeared in 2003 as part of the WHI Memory Study (WHIMS), which involved WHI participants ages 65 to 79. The WHIMS data indicated that women taking estrogen plus progestin had few improvements in overall well-being or cognitive function and a doubled risk of dementia. In another study, the Women's Health Initiative Study of Cognitive Aging, researchers investigated the effects of both forms of HT on women ages 65 to 84 who didn't have dementia. This study ended early along with the other arms of the WHI, but it found evidence linking long-term HT taken late in life to a decline in the ability to understand and use language.

Risks and benefits of estrogen + progestin in WHI studies

Condition

Increase or decrease in relative risk

Increase or decrease in cases per year*

Ages of women in study**

Blood clots in the lungs

113%

8 more

50–79

Blood clots in the legs

107%

10 more

50–79

Dementia

105%

23 more

65–79

Gallbladder disease

59%

20 more

50–79

Stroke

41%

8 more

50–79

Heart disease+

29%

6 more

50–79

Breast cancer

26%

8 more

50–79

Kidney stones

24%

5 more

50–79

Diabetes

21%

15 fewer

50–79

Osteoporotic fractures

24%

19 fewer

50–79

Colon cancer

37%

6 fewer

50–79

*Per 10,000 postmenopausal women
**67% of participants were ages 60–79
+No increased risk in women ages 50–59

Chapter 4: Where things stand now

In 2009, the Cochrane Collaboration, an international biomedical think tank that reviews the results of randomized controlled trials, took stock of all the trials that assessed the long-term effects of HT versus a placebo in healthy women. Cochrane researchers analyzed 19 randomized clinical trials involving a total of 41,904 women ages 26 to 91. Study participants taking HT were assigned to different types of estrogen and natural or synthetic progesterone (progestogens), and they took the hormones in different ways, including pills, patches, and nasal sprays.

The Cochrane report supported the findings of the WHI. The analysts concluded that combination HT — that is, estrogen plus a progestogen — increased the risk of heart attack, stroke, blood clots in the veins, breast cancer, gallbladder disease, and dementia, while estrogen alone increased the risk of blood clots, stroke, and gallbladder disease. On the other hand, both estrogen alone and combined HT reduced the risk of fractures, and combined HT reduced the risk of colorectal cancer.

Chapter 5: Still being written

The WHI didn't answer all the questions about hormone therapy. For one thing, only one type of estrogen (Premarin) and only one estrogen/progestin combination (Prempro) were tested in the WHI. Also, the participants, at an average age of 63, were substantially older than women who take estrogen for symptom relief during perimenopause or early postmenopause. Moreover, the adverse effects of HT were more likely to occur in the older women, which suggests that HT might be less harmful — and possibly offer some benefits — for younger women. Two randomized controlled trials are exploring those possibilities.

The Kronos Early Estrogen Prevention Study (KEEPS) is following 720 women ages 42 to 58 — a decade younger, on average, than WHI participants. They are also healthier, with lower body mass indexes, lower cholesterol and blood pressure levels, and less smoking. As a result, they have one-fourth the average baseline risk for heart disease. At the study outset, the women were all within three years of their last period. Half are taking estrogen by pill or patch, along with progesterone, and the other half are taking a placebo. All are regularly monitored for atherosclerotic plaque and other signs of coronary artery disease. The study is scheduled to end in early 2012.

The Early Versus Late Intervention Trial with Estradiol (ELITE) is comparing estrogen's effects in two groups of women — those within six years of menopause and those at least 10 years past menopause. The 643 women enrolled in the study have been randomly assigned to take 17-beta estradiol (Estrace) or a placebo daily for an average of five years. (Women with a uterus will also use vaginal progesterone or a placebo the last 10 days of each month.) All are being monitored with carotid artery ultrasound and cardiac CT for the development of atherosclerosis. ELITE is scheduled to end in 2013.

What does this mean for me?

The hormone saga has different messages for different women:

  • If you are a decade or more past menopause, HT probably isn't for you. It would likely increase your risk for conditions that are already more prevalent in older women — stroke, heart attack, and blood clots — and, if you're taking a progestin, for breast cancer as well. Moreover, these risks increase in the first couple of years of HT, while the benefits (reduced risk of fractures and colon cancer) don't kick in until you've been taking hormones for four or five years — and by that time, the hormone-related risk for dementia begins to mount. So you (and your clinician) may want to explore other options for hot-flash relief and bone building. And don't overlook colon cancer screening: colonoscopy reduces colon cancer risk by 77%, making it twice as effective as HT for that purpose.

  • If your hot flashes have subsided, but you're troubled by vaginal dryness, vaginal estrogen is a reasonable option. There's little evidence that estrogen-containing vaginal creams, inserts, or rings have any serious health risk, and they could improve your quality of life.

  • If you're perimenopausal or recently postmenopausal and find that symptoms (hot flashes, night sweats, or vaginal dryness) are disrupting your life, you may want to discuss HT with your clinician. Approach it as you might pain relievers or cough suppressants: take the smallest dose that does the job, for the shortest possible time. (It may take several weeks of trial and error to find the proper dose.) You may also want to ask your clinician about using transdermal (patch) HT rather than a pill. A patch may have fewer health risks than oral HT: observational studies suggest that, compared with oral estrogen, the transdermal form is associated with a lower risk for stroke, blood clots, and gallbladder disease.

  • And stay tuned for the results of the KEEPS and ELITE trials.