Rethinking Postmenopausal Hormone Therapy
Caren G. Solomon,  NEJM 2003:348:579-580

In May 2002, the Women's Health Initiative (WHI) trial of daily combined therapy with estrogen and progestin was terminated early. The reason for stopping was an increased risk of breast cancer (and evidence of greater overall risk than benefit) in the hormone-therapy group. Far more surprising, however, was the associated increase in the risk of myocardial infarction. An expectation of coronary benefit had been a major reason for many women's decisions to use postmenopausal hormone therapy.

Earlier reports had failed to show improvement in cardiovascular outcomes in postmenopausal women with known cardiovascular disease who were treated with conjugated equine estrogen either alone or in combination with medroxyprogesterone. But it was still considered plausible that healthy women would benefit. Several observational studies involving women without coronary disease had shown roughly a halving of the risk of myocardial infarction among hormone users. These findings might have been explained at least in part by the tendency of healthier women to use this therapy. However, physiological data — such as improvement in lipid profiles and measures of endothelial function with estrogen therapy — suggested mechanisms for potential benefit.

This does not mean that postmenopausal hormone therapy should never be used. Postmenopausal symptoms — such as hot flashes and vaginal dryness or discomfort — remain a valid indication in the absence of contraindications such as a history of venous thromboembolism or coronary disease. For symptoms of genital atrophy alone, local estrogen or nonhormonal lubricants may be sufficient and should be considered. Although there are other possible treatments for vasomotor symptoms — for example, selective serotonin-reuptake inhibitors — hormone therapy is very effective and still reasonable as first-line treatment. Because vasomotor symptoms are generally transient, short-term use (for no more than two to three years) is all that is generally needed, and such use carries few risks. Using the minimal dose of estrogen that controls symptoms (e.g., 0.3 mg rather than 0.625 mg of conjugated estrogen) makes sense, although there are no long-term data indicating that a lower dose reduces risk.

What about women who are using postmenopausal estrogen–progestin therapy for reasons other than control of symptoms? On the basis of available data, these women should be advised to stop. Long-term use cannot routinely be encouraged for the protection of bone, given the availability of alternative therapies, and there are no data from large clinical trials to support the belief that long-term therapy will help women preserve cognitive function or maintain a youthful appearance.

That said, there is no urgency to stop hormone therapy abruptly. In women whose symptoms recur after stopping, therapy can be gradually tapered (by reducing the frequency of administration, the dose, or both) over a period of weeks to months. For a small number of women — those with persistent symptoms and reduced quality of life — continued treatment may be justified, as long as they understand the potential risks and the alternatives.

The findings of the WHI and other trials do not rule out the possibility that some postmenopausal women might derive cardiovascular benefit from hormone therapy. Nonetheless, our current ability to identify "good candidates" for hormone therapy is too rudimentary to support differential prescribing. Thus, the prudent approach is to avoid hormone therapy for the purpose of long-term prevention of disease.

The WHI findings have led some younger women who use oral contraceptives or who use hormone therapy after premature menopause to wonder whether they should stop. Studies of hormone therapy in women 50 years of age or older, however, cannot be generalized to these groups.

In response to the WHI, other hormonal regimens or preparations have been touted as alternatives to conjugated estrogen with medroxyprogesterone. Estrogen therapy alone (without a progestin) is not recommended unless a woman has had a hysterectomy, because it is associated with increased risks of endometrial hyperplasia and cancer. More information about the long-term effects of estrogen alone after hysterectomy should be forthcoming from an ongoing part of the WHI study. Different formulations (including "natural" estrogens or progesterone) or transdermal administration has also been suggested. However, their long-term effects have simply not been studied. On the basis of available data, the Food and Drug Administration recently recommended that labeling for all postmenopausal estrogen and estrogen–progestin products include a boxed warning emphasizing the associated risks of coronary disease, stroke, and breast cancer.

What should postmenopausal women do now? Women older than 65 years of age or younger women with other risk factors for osteoporosis should have their bone mineral density measured. Women should routinely be advised to consume adequate calcium and vitamin D and to engage in weight-bearing exercise. For women who have osteoporosis, the bisphosphonates alendronate and risedronate substantially reduce the risk of both hip and vertebral fractures. The selective estrogen-receptor modulator raloxifene also reduces the risk of vertebral fracture, although it has not been shown to reduce the risk of hip fracture. In contrast to estrogen, it appears to reduce the risk of invasive breast cancer but does not improve (and may cause) menopausal symptoms. Raloxifene also increases the risk of venous thromboembolism, although its effects on cardiovascular disease remain uncertain.

To reduce cardiovascular risk, coronary risk factors should be assessed, including reevaluation of the lipid profile, which may worsen after the cessation of hormone therapy. A healthful diet, exercise, and smoking cessation should be encouraged; medications including statins and antihypertensive agents should be used in appropriate patients. The combination of these approaches is much more likely than estrogen–progestin therapy to optimize health and longevity in postmenopausal women.