Therapy for fracture prevention.

NEARLY 60 YEARS AGO, FULLER ALBRIGHT ORIGInally recognized postmenopausal osteoporosis in older women who presented with height loss, kyphosis, and back pain as a result of vertebral fractures. He astutely surmised that the waning of estrogen levels after menopause resulted in loss of bone tissue and increased skeletal fragility and predicted that estrogen therapy would prevent fractures. The relationship between cessation of ovarian function and loss of bone mass is now well established. An interval of relatively rapid bone loss occurs in the years immediately following menopause, and bone loss continues and even accelerates again in the very elderly. Estrogen therapy preserves bone mineral density (BMD) in both younger and older postmenopausal women. However, to date, no large randomized clinical trials have evaluated the antifracture efficacy of estrogen. The current image of osteoporosis has broadened over the past few years. The disorder is now viewed as a generalized skeletal disease characterized by both low bone mass and changes in bone architecture that predispose an individual to fracture. Based on the strong relationship between BMD and fracture risk, osteoporosis is now defined by the World Health Organization (WHO) solely in terms of BMD, and recent clinical guidelines recommend therapeutic intervention at specific BMD values. Because of this focus on bone density as the definition of osteoporosis, preventing fractures and their consequences as the main treatment objective sometimes escapes clinicians. Low bone mass does not cause symptoms or impaired function and is, itself, not a significant clinical issue, other than its relationship to fracture risk. It is the fractures themselves that result in physical deformity, acute and chronic symptoms, impaired physical and emotional function, decreased enjoyment of life, and reductions in longevity. In this issue of THE JOURNAL, Ettinger and colleagues report the results of the Multiple Outcomes of Raloxifene Evaluation (MORE) study, a large, well-designed, randomized controlled trial that demonstrates the effectiveness of raloxifene for reducing the incidence of vertebral deformities and fractures in postmenopausal women with osteoporosis. Raloxifene is a tissue-specific estrogen agonist that possesses estrogenlike effects in bone and on serum lipid levels but does not stimulate the endometrium or breast. Instead, raloxifene therapy is associated with a decreased incidence of estrogen-responsive malignancies in the breast and endometrium. The study by Ettinger and coauthors is the first large randomized trial documenting that an estrogenlike drug is effective in reducing fractures related to osteoporosis. This suggests that similar salutary effects might be observed with estrogen therapy. Both the control and treatment groups in this study received calcium and cholecalciferol supplements, therapy that reduces fracture incidence in older adults or in women with osteoporosis. The effects of raloxifene therapy occur in addition to the beneficial effects of calcium and cholecalciferol. Adequate calcium and cholecalciferol intake is the starting point for treating osteoporosis, but pharmacologic therapy has additional benefit. The women enrolled in this trial were selected on the basis of high current fracture risk, either with BMD values defined as osteoporosis by the WHO criteria, with previous vertebral fractures, or both. The study results were analyzed separately for the women with and without prevalent fractures. The absolute risk of fracture in the control groups during the 3 years of observation was 4.5 times higher in the women with preexisting fractures (21%) compared with the women with osteoporosis but no previous fractures (4.5%), confirming the powerful influence of previous fractures on subsequent fracture risk. In both groups, the relative risk of new vertebral fractures was substantially and significantly reduced with raloxifene therapy. The number of patients needed to treat with 60 mg/d of raloxifene to prevent a fracture was substantially smaller (42) in the women with previous spine fractures than in women without (113). Furthermore, the incidence of fractures in the raloxifenetreated groups with prevalent fractures was higher than the incidence of fractures in the control group with no prior frac-