使用骨密度仪进行病情监测.ppt

* Continued example of precision experiment in 15 subject scanned 3 times each. Used with the advance calculator on the ISCD web site to calculate precision as RMS-%CV Although the calculator calculates the CV and % CV using the individual CV, you can calculate it for the entire group by dividing RMS SD by the Mean of the BMD for the group to get the CV of the group and multiply by 100 to get %CV of the group. * Least significant change is determined by multiplying precision error by 2.77. This number comes from the 95% confidence interval – which is typically used in medical studies. This is the confidence interval that ISCD recommend when monitoring patients with DXA. * Example of precision errors. Next slide will show corresponding LSC. * Continued form previous slide. LSC calculations using one center’s precision values. * Review of previous slides. Ask if there are any questions! * Example of calculating LSC at the lumbar spine using the stated precision values. * Using absolute difference of LSC (0.028 g/cm2) in this example for arriving at significant change. Use of RMS-%CV to determine LSC for purposes of % for reporting of error in % difference. If the amount does NOT exceed the LSC – state in the report that the BMD appears stable. To avoid confusing the referring physician, avoid stating that “there IS a difference, but that the difference is NOT SIGNIFICANT”. * ANIMATED This is a very common scenario where the BMD increases at one site and decreases at another. The vast majority of these cases have changes in BMD that are below the LSC. The teaching point, don’t over interpret small changes in BMD. Report as no significant change in this case. * * Comparing different manufacturers (error 5%) if far worse than same manufacturer (error~5%). This is a MAJOR problem with patients moving or changing insurance. Ideal solution is to measure BMD at initial center for current change, then another scan at the new site to use for baseline in the future – bu