DEXA management for the diagnosis of osteoporosis – a worldwide perspective: Part 2
In a separate calculation differences in prediction of hip fracture were studied (O. Johnell and J.A. Kanis, personal communication), whether lumbar spine BMD measurement had the same predictive ability as femoral neck BMD measurement and if the combination of femoral neck BMD and lumbar spine BMD could increase the predictive ability. The data show that for hip fracture BMD at the hip had a better predictive ability than the lumbar spine measurements and that combination of lumbar spine BMD and femoral neck BMD did not contribute to an increased predictive ability.
The problem with BMD is also that the absolute values differ between the types of scanner but an algorithm has been created to get a standardized BMD for the different scanners (Kanis et al., 2002b). This is necessary to make an easy calculation for the T score. As mentioned earlier, the diagnosis and the intervention threshold should be divided (Kanis et al., 2002b). Therefore focus on risk assessment for intervention threshold with BMD measurements and other risk factors should be done. The prevalence of osteoporosis is somewhat different in the published studies in that there are more osteoporotic patients in some countries, partly depending on the reference population, whether it is from the local area or from the manufacturer. Therefore we should shift focus from the diagnosis to risk assessment. From the studies quoted this seems to work worldwide (Johnell et al., 2005).










