Performance of calcaneus quantitative ultrasound and dual-energy X-ray absorptiometry in the discrimination of prevalent asymptomatic osteoporotic fractures in postmenopausal women

Abstract

Due to its low cost, portability, and nonionizing radiation, quantitative ultrasound (QUS) of the heel is an alternative to the measurement with dual X-ray absorptiometry (DXA) in the evaluation of bone status. The objective of the study is to compare in asymptomatic postmenopausal women the ability of QUS and DXA to discriminate between those with and without prevalent vertebral fractures (VFs). The study cohort consists of a population of 295 postmenopausal women aged between 60 and 84 (mean age, weight and BMI of 66.3 years, 72.0 kg and 29.4 kg/m(2), respectively). Lateral VFA images and scans of the lumbar spine and proximal femur were obtained by two technologists using a GE Healthcare Lunar Prodigy densitometer. VFs were defined using a combination of Genant semiquantitative (SQ) approach and morphometry. All women had a calcaneous QUS examination. The mean age of the women in our sample was 66.3 (+/-5.3) years, ranging from 60 to 84 years. Eighty-seven (29.3%) women had VFs Genant grade 2 and 3. Patients with VFs had an age and a number of years of menopause higher to those without VFs, but showed lower height, weight, and BMI. All densitometric and ultrasonometric measurements were significantly reduced in women with VFs. The intercorrelations of BMD at different sites were high, and the correlations of BUA with BMD were lower. BUA correlated weakly with total hip BMD (r = 0.36), lumbar spine BMD (r = 0.32), and much less with femur BMD (r = 0.30); all correlations were significant (P < 0.01). Analysis of the AUC for the ROC curves showed lumbar spine T-score below -2.5 to provide consistently the highest AUC (0.64). Age-adjusted ORs after correction for confounding variables (years of menopause, weight, height, and BMI) for QUS and BMD measurements showed that only lumbar spine T-score below -2.5 could predict significantly the presence of VFs (OR, 1.94; 95%CI, 1.02-3.41). Lumbar spine BMD (and not QUS) was able to discriminate asymptomatic postmenopausal women with prevalent VFs from women without VFs and independently contributed to determining the association with fracture. The combination of QUS and BMD did not improve the diagnostic ability of either individual technique.