Bone densitometry in clinical practice: longitudinal measurements at three sites in postmenopausal women on five...

We report sequential changes in bone mineral density (BMD) at the forearm, hip and spine in 340 consecutive postmenopausal women referred by 103 general practitioners and six specialists, and who were either untreated or being treated with calcium, estrogen, norethisterone or calcitriol for a median period of 25 months (range 11-52). The mean annual rate of change in BMD at the three sites was: 1.39% in 44 women on norethisterone; 0.94% in 107 women on estrogen (both p < 0.001); 0.24% (not significant) in 52 women on calcitriol; -0.53% in 92 women on calcium; and -1.06% in 45 women on no treatment (both p < 0.01). The mean annual rate of change at the three sites in the 295 treated women was 0.43%, which was significantly positive (p < 0.001) and was 1.49 percentage points more positive than in the untreated women (p < 0.001). The greatest mean difference between treated and untreated patients was seen at the forearm, where it was 2.16 percentage points (p < 0.001). This was significantly greater than the difference at the femoral neck (1.21 percentage points (p = 0.037)) and lumbar spine (1.10 percentage points (p = 0.044)). The data did not change significantly after correction for age, years since menopause or baseline BMD. Those who started the treatments at baseline gained bone faster than those who were continuing on existing therapies, but this difference was not significant at any site. In the hormone- and calcitriol-treated groups, there was no significant difference between those who had a calcium supplement and those who did not. We conclude that the effects of treatment on BMD in clinical practice are comparable to those predicted from clinical trials, that there are significant differences between the responses to treatment at different sites and that the forearm appears to be the most sensitive site, in this series at least.