Prospective 10-year study of the determinants of bone density and bone loss in normal postmenopausal women, including the effect of hormone replacement therapy.

OBJECTIVE

To prospectively assess bone density and the factors determining the rate of bone loss over a 10-year period of postmenopausal life.

DESIGN

Prospective, observational study.

METHODS

One hundred and four normal White postmenopausal women, baseline mean age 59 years (range 47-71 years) completed the study (mean duration of follow-up 10.2 years, range 9.4-10.6 years). None had diseases or were taking medications affecting bone metabolism at entry to the study. Information was collected on medical, fracture and smoking history, alcohol use, dietary calcium intake and physical activity. Body composition and bone density were measured by dual-energy X-ray absorptiometry at baseline and at 10 years. Biochemical, haematological and hormonal analyses were performed.

RESULTS

Twenty-four percent of the women started hormone replacement therapy (HRT) during the study period; most of these remained on therapy at follow-up. The mean duration of therapy was 6.6 years (range 2.8-10.4 years). The use of HRT was associated with significant gains in bone density (total body + 3.0%, trochanter + 4.2%, Ward's triangle + 4.4%, spine + 10.5%) and a significant reduction in vertebral fracture risk [standardized risk ratio compared with non-HRT users 0.42 (confidence interval 0.18-0.83)]. HRT use was not associated with greater weight gain than that occurring in other members of the cohort. The baseline and follow-up bone densities in the non-HRT users were highly correlated (0.82 < or = r < or = 0.91, P < or = 0.0001) and baseline bone density accounted for the majority of the variance in the 10-year results. Multivariate analyses showed that the independent correlates of rate of change of bone density were weight and fat mass (both baseline values and changes during follow-up), time after menopause, sex hormone concentrations, urinary calcium loss, PTH levels and haemoglobin concentration (which may reflect nutrition and health).

CONCLUSIONS

Bone density is highly predictable over an extended period of time in normal postmenopausal women. Maintenance of body weight and good health reduce bone loss. HRT is effective for treating osteoporosis, with improvement in bone density and reduction in vertebral fractures. Good compliance with HRT long-term is achievable. These findings are relevant to deciding the frequency of bone density measurement, and in advising women regarding prevention and treatment of postmenopausal bone loss.