The discriminatory capacity of BMD measurements by DXA and dual X-ray and laser (DXL) at the calcaneus including clinical risk factors for detecting patients with vertebral fractures.

UNLABELLED: Osteoporotic fracture risk depends on bone mineral density (BMD) and clinical risk factors (CRF). Five hundred and eighty-eight untreated female and male outpatient subjects were evaluated, 160 with vertebral fractures. BMD was measured both by using calcaneal dual X-ray and laser (DXL) and dual-energy X-ray absorptiometry (DXA), and CRF were evaluated. Detection frequencies for different BMD methods with or without CRF are presented.
INTRODUCTION: Osteoporotic fracture risk depends on bone mineral density and clinical risk factors. DXA of the spine/hip is considered a gold standard for BMD assessment, but due to degenerative conditions, particularly among the older population, assessment of BMD at the lumbar spine has been shown to be of limited significance. Portable calcaneal dual X-ray technology and laser can be an easily obtainable alternative.
METHODS: Vertebral fractures were evaluated in a baseline analysis of 588 females and males (median age 64.4, range 17.6-93.1 years), comparing BMD measurements by using DXL and DXA and CRF with/without BMD. One hundred and sixty subjects had radiological verified vertebral fractures. Area under receiver-operating characteristic curves (AUROCC) and univariate and multiple logistic regressions were calculated.
RESULTS: AUROCC for detection of vertebral fractures was comparable for DXL at calcaneus and DXA at femoral neck (DXL 0.665 and DXA 0.670). Odds ratio for prevalent vertebral fracture was generally weak for DXA femoral neck (0.613) and DXL (0.521). Univariate logistic regression among CRF without BMD revealed age, prevalent fragility fracture, and body mass index significantly associated with prevalent vertebral fracture (AUROCC = 0.805). Combining BMD and CRF, a prognostic improvement in case of DXA at femoral neck (AUROCC 0.869, p = 0.02), DXL at calcaneus (AUROCC 0.869, p = 0.059), and DXA at total hip (AUROCC 0.861, p = 0.06) was observed.
CONCLUSIONS: DXL was similarly sensitive compared with DXA for identification of subjects with vertebral fragility fractures, and combination of CRF with BMD by DXL or DXA further increased the discriminatory capacity for detection of patients susceptible to vertebral fracture.