Do Estrogens Effectively Prevent Osteoporosis-Related Fractures?

Since Albright, [1] some 60 years ago, reported the benefi-cial effects of estrogens for decreasing urinary calcium ex-cretion and suggested that these harmones might be usefulin preventing postmenopausal osteoporosis, estrogen re-placement therapy (ERT) has been consistently regarded asthe first choice for prevention of trabecular and corticalbone loss in postmenopausal women [2–5]. However, seri-ous controversies remain over the cost/effectiveness oftreating every woman at the time of menopause [6], theoptimal timing for starting ERT [5], the minimal effectivedose of ERT acting on bone [7], and the duration of ERTneeded to prevent osteoporotic fractures [8]. The effective-ness of ERT for preventing osteoporosis-related fractures isundisputed and requirements for marketing authorizationfor ERT products have lightened compared with currentrequests for other therapeutic medications developed in thisfield [9, 10]. However, although the skeletal benefits ofERT for preventing trabecular or cortical bone loss canhardly be challenged, one might be wary of published evi-dence that prolonged ERT use unequivocally reduces therisk of hip fracture. Controlled clinical trials and systematicreviews were located using Medline 1970–1999 andEMBASE 1980–1999. Since 1985 we have searched scien-tific journals on bone and bibliographies of review articles.All prospective controlled trials were included for evalua-tion of the effects of hormone replacement therapy (HRT)on bone loss. A total of 57 prospective controlled trials wereidentified, 46 of which were randomized clinical trials(RCTs) and 15 were double blinded. All clinical trials as-sessing the effects of HRT on fracture rates were consid-ered. Two RCTs and one systematic review were identified[11].In the 46 randomized controlled trials comparing estro-gen (with or without progestins) (HRT) with placebo orcalcium on bone loss prevention, the study population var-ied from 14 to 875 women and the duration was from .5 to10 years. In general, they drew similar conclusions, i.e, thatestrogen intervention reduces the rate of postmenopausalbone loss at trabecular and cortical sites. An early double-blind trial [12] reported the preventive effects of HRT oncortical (metacarpal) bone loss for up to 10 years. Morerecent double-blind, placebo-controlled, randomized clini-cal trials confirmed these findings for oral [13], percutane-ous [14, 15], or transdermal [16] estrogens at the spine [13,14, 16], the forearm [16], and/or the hip [13, 16] for up to3 years. Two prospective open studies [17, 18] showedsimilar results for estrogen implants after 1 year. Whenstandardized for technique used for bone mineral density(BMD) assessment, the magnitude of the point estimatedifferences between the HRT and the control group variedgreatly from one study to another, depending upon the doseof HRT used (dose-related effect on bone mass in most