COMPARISON OF RATES OF BONE LOSS FROM THE SPINE MEASURED USING 2 MANUFACTURERS DENSITOMETERS

Measurements of bone mineral density of the lumbar spine (LS-BMD) by d ual-energy X-ray absorptiometry (DXA), made using different manufactur ers' densitometers, differ by approximately 15%. The aim of this study was to compare the rate of bone loss from the lumbar spine measured u sing two manufacturers' densitometers. LS-BMD (vertebrae L2-L4) was me asured using both a Lunar DPX and a Hologic QDR 1000/W in 17 postmenop ausal women at baseline and 2 years, The mean rate of change of LS-EMD made using the two densitometers did not differ (DPX -0.54% per year; QDR -0.17% per year), Although rates of change of LS-BMD measured usi ng the two densitometers correlated (r = 0.47, p = 0.05), the variance of the rates of loss differed such that, in individual subjects, the difference between the rates of loss of LS-BMD, measured using the two densitometers, was related to the average rate of loss (p = 0.61, p = 0.009), For both densitometers, as bone mineral content (BMD). increa sed, the measured area also increased, leading to an underestimate of change in bone mineral density (BMD), This relation differed between t he two densitometers such that the change in BMD was underestimated to a greater extent by QDR We conclude that in the context of multicente r studies it may not be possible to directly combine data on the rate of bone loss acquired using different manufacturers' densitometers.