Cross-calibration of prodigy and horizon a densitometers and precision of the horizon a densitometer

Authors: REITSHAMER, ELISE - Jean Mayer Human Nutrition Research Center On Aging At Tufts University; VIERA, KELSEY - Jean Mayer Human Nutrition Research Center On Aging At Tufts University; SHEA, KYLA - Jean Mayer Human Nutrition Research Center On Aging At Tufts University; DAWSON-HUGHES, BESS - Jean Mayer Human Nutrition Research Center On Aging At Tufts University

Submitted to: Journal of Clinical Densitometry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/17/2021
Publication Date: 2/21/2021
Citation: Reitshamer, E., Viera, K.B., Shea, K., Dawson-Hughes, B. 2021. Cross-calibration of prodigy and horizon a densitometers and precision of the horizon a densitometer. Journal of Clinical Densitometry. https://doi.org/10.1016/j.jocd.2021.02.003.
DOI: https://doi.org/10.1016/j.jocd.2021.02.003

Interpretive Summary: Dual-energy X-ray absorptiometry (DXA) is widely used to assess bone mineral density (BMD) because DXA measures predict fracture risk and because they are reproducible, rapid, and involve low radiation exposure. DXA is also increasingly used to assess body composition. Over time, it is necessary to acquire a new scanner because of aging equipment and/or advancing technology. In clinical research laboratories with ongoing studies involving DXA measurements, it is important not only to cross-calibrate the old to the new scanner but also to determine the reproducibility of measurements on the new scanner. The purpose of this study was to cross-calibrate measurements of bone and body composition of older adults on a GE Lunar Prodigy and a newer Hologic Horizon A scanner and to assess the short-term in vivo precision of the Horizon A scanner in our laboratory. Forty-five healthy older adults came to the Center once each for a set of DXA scans on the old scanner and two sets (with repositioning between sets) on the new scanner. Equations were developed to cross-calibrate measurements on the older to the newer scanner. The precision of the new scanner was in the range of 1 to 2% at the spine and hip and in the same range for total body fat mass, lean mass, and bone mass. The findings in this study facilitate a smooth and reliable transition from one DXA scanner to another for ongoing clinical studies and establish that the precision of the new scanner is excellent.

Technical Abstract: Introduction/background: We performed this study to enable a reliable transition for clinical study participants and patients from a GE Lunar Prodigy to a Hologic Horizon A dual-energy X-ray absorptiometry (DXA) scanner and to assess the reproducibility of measurements made on the new DXA scanner. Methodology: Forty-five older adults had one spine, hip and total body scan on a Prodigy dual-energy X-ray absorptiometry (DXA) scanner and two spine, hip, and total body scans, with repositioning, on a new Hologic Horizon A DXA scanner. Linear regression models were used to derive cross calibration equations for each measure on the two scanners. Precision (group root-mean-square average coefficient of variation) of bone mineral density (BMD) of the total hip, femoral neck, and lumbar spine (L1-L4), and total body fat, bone, and lean mass, appendicular lean mass, and trabecular bone score (TBS) was assessed using the International Society of Clinical Densitometry's (ISCD's) Advanced Precision Calculation Tool. Results: Correlation coefficients for the BMD and body composition measures on the two scanners ranged from 0.94 to 0.99 (p<0.001). When compared with values on the Prodigy, mean BMD on the Horizon A was lower at each skeletal site (0.136 g/cm2 lower at the femoral neck and 0.169 g/cm2 lower at the lumbar spine (L1-4)), fat mass was 0.47 kg lower, and lean mass was 4.50 kg higher. Precision of the Horizon A scans was 1.60% for total hip, 1.94% for femoral neck, and 1.25% for spine (L1-4) BMD. Precision of TBS was 1.67%. Precision of total body fat mass was 2.16%, total body lean mass was 1.26%, appendicular lean mass was 1.97%, and total body bone mass was 1.12%. Conclusions: The differences in BMD and body composition values on the two scanners illustrate the importance of cross-calibration to account for these differences when transitioning clinical study participants and patients from one scanner to another.