Discriminatory ability of quantitative ultrasound parameters and bone mineral density in a population-based sample of postmenopausal women with vertebral fractures: results of the Basel Osteoporosis Study

Abstract

The discriminatory potential to classify subjects with or without vertebral fractures was tested cross-sectionally with different methods for the measurement of bone status in a population-based sample of postmenopausal women. Quantitative ultrasound (QUS) measurement at the calcaneus (Lunar Achilles, Hologic Sahara), the proximal phalanges (Igea Bone Profiler), and measurement of bone mineral density (BMD) with dual-energy X-ray absorptiometry (DXA; Lunar Expert) at several anatomic sites was performed in 500 postmenopausal women (aged 65-75 years) randomly selected from the population. In addition, 50 young female subjects (20-40 years old) had QUS measurements and served as controls to express QUS results as T-score values. Radiographs of the lumbar and thoracic spine were performed in the elderly women. Two independent radiologists reviewed the X-rays for the presence of vertebral fractures. Of 486 eligible study participants, no fracture was seen in 396 participants. Single vertebral fractures were observed in 71 subjects; 19 individuals presented multiple fractures. The overall prevalence of vertebral fractures was 18.5%. Participants without vertebral fractures were compared with subjects with vertebral fractures. Normal statistical distributions were found for all bone measurement results. Risk of vertebral fracture in subjects with no and multiple vertebral fracture was estimated using age adjusted odds ratios (ORs) for QUS and dual-energy X-ray absorptiometry (DXA) values. Each SD decrease in bone measurement increased the risk of multiple vertebral fracture by 3.0 (95% CI, 1.6-5.6) for the Achilles stiffness, by 3.8 (95% CI, 1.8-8.2) for the Sahara QUI, 2.1 (95% CI, 1.3-3.4) for the Bone Profiler amplitude-dependent speed of sound (AD-SOS), and 2.1 (95% CI, 1.2-3.9) and 2.4 (95% CI, 1.3-4.3) for DXA lumbar spine and for DXA total hip, respectively. Results of a discriminant analysis showed sensitivities between 84% and 58% and specificities between 72% and 58% for the respective DXA and QUS parameters. Optimum fracture thresholds for QUS measurements derived from this analysis were calculated also. Optimum T-score threshold values for QUS measurements tended to be higher than those for DXA measurements. However, the performance of QUS measurements is at least comparable with DXA measurements in identifying subjects with multiple vertebral fractures randomly selected from the population.