LAUSR

Introduction/Background In certain circumstances when axial bone mineral density (BMD) cannot be measured using dual-energy X-ray absorptiometry (DXA), DXA BMD of the one-third radius (33% radius) can be used for the diagnosis of osteoporosis. Calcaneal quantitative ultrasound (cQUS) variables may also be used for assessing osteoporosis. The aim of this study was to test the hypothesis that cQUS variables may be better than one-third radius (33% radius) BMD for the identification of axial osteoporosis in women. Material and method Two hundred and nine women aged between 21 and 85 years whose posteroanterior spine, hip, and 33%radius DXA BMD measurements as well as cQUS variables were measured were included in this study. WHO criteria were employed for defining osteoporosis in postmenopausal women and premenopausal women with Z-scores of ≤ −2.0 were considered as having osteoporosis in axial regions. We used receiver operating characteristic (ROC) analysis to evaluate lumbar spine or hip osteoporosis discriminative performance of radius DXA BMD and cQUS parameters. Results While 33% radius BMD and its T-score revealed greater areas under ROC curves (AUCs) (0.760 and 0.759) than those of the cQUS variables [0.707 for quantitative ultrasound index (QUI), 0.705 for QUI T-score, 0.682 for broadband ultrasound attenuation (BUA), 0.718 for speed of sound (SOS), and 0.706 for estimated heel BMD (eBMD)] for identifying lumbar spine osteoporosis, the AUC values for all cQUS variables (0.816 for QUI, 0.815 for QUI T-score, 0.829 for BUA, 0.791 for SOS, and 0. 813 for eBMD) in the discrimination of femoral neck osteoporosis were greater than those of 33% radius DXA BMD and its T-score (0.779 and 0.783). Conclusion The results of this study confirmed our hypothesis for femoral neck osteoporosis, but not for lumbar spine osteoporosis. Our results have implications that cQUS can be used for the prediction of femoral neck osteoporosis in women whose hip DXA BMD cannot be measured.