Early postmenopausal diminution of forearm and spinal bone mineral density: a cross-sectional study.

Diminution of bone mineral density (BMD) in the spine and forearm was studied cross-sectionally in 363 women who were 6 months to 10 years postmenopausal. BMD was determined by dual-energy X-ray absorptiometry (DXA) (Hologic QDR-2000) in the lumbar spine, in both the supine lateral (LAT) and anteroposterior (AP) projections, and in the distal third of the forearm. The postmenopausal diminution of BMD was best described by an exponential fit. The initial rate of postmenopausal diminution of BMD was highest in the most trabecular sites (LAT > AP > forearm), but 10-year diminution was similar at all sites (12%-13%, corresponding to about 1.0-1.5 SD), and extrapolation suggested reverse order of the rates of diminution thereafter (forearm > AP > LAT). When bone mineral content of the entire L3 vertebra (tBMC) was measured in vivo, AP tBMC could account for only 67% of the variation in LAT tBMC, compared with r2 = 0.997 in vitro. This observation suggests an accuracy problem in vivo in one of the spine measurement methods. We conclude that the initial rate of BMD diminution after the menopause seems to be highest in the spine, especially when measured laterally, but that this rate levels off within the first decade. The lower precision error of a forearm measurement (0.8% v 1.6 for AP and 3.1 for LAT) therefore implies that this method may require a shorter observation period than spine measurements for the detection of bone loss 5-10 years after menopause. Long-term longitudinal spine and forearm measurements are, however, needed to confirm these conclusions.