Calculating in-vivo short-term precision error of dual-energy x-ray absorptiometry in human and animal: a technical report

Abstract

Bone mineral density (BMD) measurement by dual-energy X-ray absorptiometry (DXA) is important in diagnosing osteoporosis. Precision error of DXA is a useful measure to determine a true change in BMD value. This study aimed to investigate the short-term coefficient of variance of Hologic Discovery QDR Wi DXA machine. The lumbar spine and hip BMD of fifteen healthy volunteers (mean age: 30.67 + 10.41 years) and the whole body BMD of fifteen female Sprague-Dawley rats (aged three months old) were scanned using Hologic Discovery QDR Wi DXA machine. Each volunteer and rat underwent triplicate scans to assess the reproducibility of BMD values. The interval between the scans ranged within 1 to 12 weeks for human subjects. For animal samples, the scans were repeated on the same day after repositioning. The precision error was expressed as a percentage coefficient of variance (%CV). The %CV obtained for lumbar spine and hip BMD for human were 1.8% and 1.2%, respectively. The %CV for whole body BMD of rats was 1.4%. The short-term CV demonstrated for both human and animal in the present study were comparable. The precision error of DXA must be monitored to ensure optimal performance of the device.