Optimizing data acquisition and analysis of morphometric X-ray absorptiometry

Abstract

Morphometric X-ray absorptiometry (MXA) uses dual-energy X-ray absorptiometry (DXA) scanners to perform vertebral morphometric measurements of the vertebrae. In this study we evaluated the four available MXA scan modes--single-energy (SE) and dual-energy fast (F), array (A) and high definition (HD)--on a commercial bone densitometer (Hologic QDR-4500A). Sixty postmenopausal women (mean age 59 years, range 40-73 years) were recruited and split into two groups matched for body mass index (BMI, kg/m2). Three MXA scans, covering 13 vertebrae from T4 to L4, were acquired on each subject; all subjects were scanned in SE and A modes, while the third scan was performed in F mode in group 1 and in HD mode in group 2. Subjects were invited to return 6 months after the commencement of the study to repeat their scans. The HD mode produced the most reliable image, with 97% of all scans analyzable to T7 and the fewest vertebrae being lost to analysis (1.5/13 vertebrae lost per scan). A SE + HD combination (using whichever image allows the analysis of more vertebrae) further decreased the number of vertebrae lost to 0.8 of 13 vertebrae, i.e. a typical scan was analyzable up to and including T5. BMI had a noticeable and scan-mode-dependent effect on MXA image quality, an increase in the number of vertebrae lost to analysis occurring once BMI exceeded 30. BMD had a far smaller effect on image quality and no effect at all using the SE + HD combination. Precision (CV%) was similar for all three dual-energy modes at around 3.5% without the scan 'compare' facility and 2.6% with it. The best precision was obtained with SE scan (2.7%/2.2%). BMI and BMD had little or no effect on precision. We conclude that optimal results are obtained by the acquisition of both SE and HD scans. However, for rapid assessment by trained operators SE scans alone offer almost equal utility.