Cross-calibration of a GE iDXA and Prodigy for Total and Regional Body Bone Parameters: The Importance of Using Cross-calibration Equations for Longitudinal Monitoring After a System Upgrade

We aimed to determine if cross-calibration equations could be applied to convert GE Lunar Prodigy total and regional bone measurements to the GE iDXA model to support longitudinal monitoring of subjects. The cross-calibration group comprised 63 adults (age 45.1 [12.8] yr; body mass index: 25.6 [3.7] kg/m2) and the validation group comprised 25 adults (age 40.5 [11.5] yr; body mass index: 25.7 [3.5] kg/m2). The parameters reported were total and regional bone mineral density (BMD), bone mineral content, and bone area. There were significant differences between densitometers for all anatomical regions and reported bone parameters (p < 0.0001); iDXA reported lower BMD than the Prodigy apart from the ribs. Linear regression indicated good agreement for all measurements. Bland-Altman analyses indicated significant bias for all measurements and that cross-calibration equations were required. The derived cross-calibration equations were effective in reducing differences between predicted and measured results for each parameter and at each region apart from leg BMD, where the difference remained significant (0.013 g/cm2; p < 0.05). Our results indicate that cross-calibration is important to maintain comparability of total body-derived regional bone measurements between the Lunar Prodigy and iDXA.