Bone mineral density and its change in pre-and perimenopausal white women: the Michigan Bone Health Study.

There is a need to better understand potential bone mineral density (BMD) loss during the menopausal transition since this period may include the initiation of interventions. The study purpose was to determine if there was BMD loss at the femoral neck, lumbar spine, or total body bone sites in a population-based study of women approaching or transitioning the midlife. The 583 enrollees were 25-45 years of age at the first of four annual measurements from 1992 through 1996. Bone mineral content and bone width were measured using dual-energy X-ray absorptiometry. Considering all enrollees collectively, there was a significant 3-year decline (1%) in BMD at the femoral neck over the 3-year period (p = 0.076). There was no significant annual change in the lumbar spine (p = 0.11), and a significant annual increase in the total body BMD (p = 0.0003). Within subgroups and cross-sectionally, BMD values of the femoral neck were 5% lower in women classified as perimenopausal compared with premenopausal enrollees; BMD was 3% and 1% lower at the lumbar spine and total body site, respectively. Longitudinally, among perimenopausal women, a double oophorectomy was associated with BMD loss in the spine (p = 0.0003), even though 75-85% of these women had a hormone replacement prescription at some time during the study period. In summary, the site with evidence of loss was the femoral neck, specifically among perimenopausal women. There was little evidence of substantial total body or lumbar spine BMD loss in premenopausal women with ovaries who maintained follicle-stimulating hormone levels < 20 mIU/l in the early follicular period. Double oophorectomy, even with hormone replacement, was associated with bone loss.