I have outlined a comprehensive model of estrogen action on bone that explains the known physiologic effects of estrogen and that is consistent with the limited data of the available clinical studies. I cannot stress too highly that the model, however elegant and satisfying, is not the same as real postmenopausal women. The predictions of the model can be tested only in that world. Special problems make this testing unusually difficult, but the disease is too important in a population with the current and changing age distribution in the United States to excuse additional delay in undertaking the necessary studies. If the model is valid, it suggests that estrogen has a role in treatment of osteoporosis, but only if that treatment is confined strictly to palliation. The model also suggests that estrogen may play a much more important role in prevention. But to whom should such prevention be offered? Should it be postmenopausal women, when only about one fourth appear to be at risk? And who are the women in that risk group? Once again we must confront our ignorance. It is securely known that Black women in this country are relatively much less susceptible to osteoporotic fracture, but beyond that the best that one can offer is anecdotal information. Small, fine skinned, fair women, and those with less muscle mass and body hair, appear to be more at risk than larger, darker skinned, and more masculine women. But the overlap between the groups is unknown and possibly extensive. Determination of risk factors would appear to be of great importance if one is to justify long duration estrogen treatment. The morbidity and mortality of osteoporosis are more than sufficient to justify incurring the small risk of endometrial carcinoma that may accompany postmenopausal estrogen treatment, but is no justification at all in a patient who was never at risk for osteoporosis in the first place. Finally, what of dosage? It is not surprising that, with the paucity of qualitative data, there are at present no quantitative data. Nevertheless, the effects I have described in the context of the model all occur at physiologic levels of estrogens, and hence there can be no serious question of preventive doses exceeding levels of 1.25 mg of conjugated equine estrogens or their equivalent. Quite possibly, smaller doses, one half or one quarter as large, may be perfectly effective; but it would be rash to speculate beyond that. Presumably, unless the patient has had a hysterectomy, interrupted therapy to allow endometrial shedding would be the preferred regimen.
All Science Journal Classification (ASJC) codes
- Obstetrics and Gynecology