Related Experiment Videos
Does osteoporosis classification using heel BMD agree across manufacturers?
M Grigorian1, J A Shepherd, X G Cheng
1Osteoporosis and Arthritis Research Group, Department of Radiology, University of California at San Francisco, San Francisco, CA, USA. Mikayel.Grigorian@oarg.ucsf.edu
Summary
This study assessed agreement between two heel bone mineral density (BMD) devices, PIXI and Apollo. While highly correlated, differences in measurement and T-scores highlight the need for standardization in peripheral dual X-ray absorptiometry (pDXA) assessments.
Area of Science:
- Osteoporosis research
- Medical imaging technology
- Biomedical engineering
Background:
- Standardization of bone mineral density (BMD) measurements is a persistent challenge.
- Dual X-ray absorptiometry (DXA) devices, both axial and peripheral (pDXA), require cross-calibration.
- Recent introduction of new pDXA densitometers necessitates agreement evaluation.
Purpose of the Study:
- To evaluate the agreement between two specific heel pDXA devices (PIXI and Apollo) for BMD and T-score measurements.
- To identify sources of disagreement between the two pDXA scanners.
- To assess the impact of normative database differences on T-score agreement.
Main Methods:
- Ninety-nine females (aged 21-78 years) underwent non-dominant heel BMD measurement using PIXI and Apollo pDXA scanners.
- Analysis included correlation, slope, and offset comparisons for BMD and T-scores.
- Diagnostic agreement was assessed using weighted kappa statistics.
Main Results:
- Both PIXI and Apollo devices showed high correlation for BMD and T-scores.
- Significant non-identity slopes and non-zero offsets were observed between the devices.
- Diagnostic agreement on T-scores was 86% (weighted kappa = 0.550).
- Normalizing reference data reduced T-score disagreement, indicating database influence.
Conclusions:
- PIXI and Apollo pDXA scanners demonstrate high correlation but exhibit systematic differences in BMD and T-scores.
- Discrepancies are attributed to variations in region of interest (ROI) definitions and normative databases.
- Standardization efforts are crucial for accurate osteoporosis diagnosis using pDXA technology.