Aim: The aim of this study was to compare the ability of Hologic(QDR-4500C), Lunar(DPX-MD)and Norland (XR-46) to reproduce Dual Energy Absorptiometry (DXA) results using measurements obtained by scanning a phantom designed to simulate a spine with low bone mass . Background: Osteoporosis is characterized by low bone mass and structural deterioration. The measurement of bone mineral density (BMD) using
DXA might improve predictions of bone strength.
DXA is the most frequently used technique to measure
BMD and therefore assess the risk of fracture . Reproducibility or Coefficient of Variation (CV%) determines the ability of
DXA systems to differentiate small changes in BMD. Although
DXA systems perform reproducible results when scanning healthy young subjects of average size, a variation in the bone mass could introduce fluctuations of
CV% particularly in subjects with low bone mass . Material and Methods: A phantom containing K2HPO4 powder set in Perspex to simulate a low bone mass spine was manufactured. This was done in collaboration with both the Research Center of Science and Technology in Medicine (RCSTIM) and the Endocrinology and Metabolism Research Center (EMRC( . The phantom was scanned ten times on each system without repositioning. The acquisition and analysis was performed by a qualified
DXA technologist. The fast array mode was used for the Hologis scanner and medium mode was used for the Lunar system., The Norland system does not produce an image using standard mode, so the high precision mode was used in this instance .
The CV%, was calculated as the SD/Mean ×100, The
CV% was then used to compare the reproducibility of the three scanners .
Results: We found
CV% equal 0.81, 1.46 and 0.88 for Hologic, Lunar and Norland, respectively .
The Mean of the BMD, SD and %CV of systems
DXA machine Model
Scan Mode
Mean (mg/cm2 (
SD
CV %
Hologic QDR-4500C Fast array 705 5 0.81
Lunar DPX-MD Medium 813 12 1.46
Norland XR-46 High precision 682 6 0.88
Conclusion: Although the same mode must be used for follow-up scans, but the most appropriate scan mode should be selected, based on BMI of the subject. This is important, especially in osteoporotic patients to maintain to the CV .% According to our results the Hologic has better reproducibility when measuring subjects with low BMD. This could be due to Hologic’s internal reference wheel that provides real time calibration during scanning, and the use of an anthropomorphic spine phantom for quality control .