Association between incident and baseline vertebral fractures in European women: vertebral fracture assessment in the Osteoporosis and Ultrasound Study (OPUS)

Abstract

Vertebral fracture assessment (VFA) with densitometric devices uses less radiation than spinal radiography. We assessed risk of new vertebral fracture (VF) in women with baseline fracture identified on VFA using algorithm-based qualitative diagnosis. Women with VF had significantly greater risk of VF after 6 years compared to those without baseline fracture. Prevalent VFs predict future fracture and are identifiable on vertebral fracture assessment (VFA) using bone densitometry devices. We have previously performed cross-sectional, but not longitudinal, VFA using the algorithm-based qualitative method (ABQ). We aimed to examine the prevalence and incidence of VF and test the association between prevalent and incident VF identified by ABQ VFA. We used ABQ to assess vertebral images obtained at baseline and 6 years (Hologic devices) in 674 women at ages 39 to 80 years participating in the Osteoporosis and Ultrasound Study. Criteria for prevalent and incident VF were endplate fracture, with/without cortical fracture. We compared proportions (chi-squared test) and characteristics (two-sample t tests and analysis of variance) of women with and without VF and calculated odds ratios for incident VF in women with prevalent VF (logistic regression). Prevalent VF was identified in one premenopausal woman and 41 postmenopausal women. Incident VF was identified in 18 postmenopausal women. Odds ratios (95% CI) for incident VF in postmenopausal women with prevalent VF were 7.8 (2.8, 22.1) (unadjusted) and 4.3 (1.4, 13.7) (adjusted for age and bone mineral density, BMD). Women with prevalent or incident VF were older (P < 0.01), with lower hip BMD (P < 0.001) compared to women without VF. Population-based postmenopausal women had relatively low prevalence and incidence of VF analysed with the ABQ method applied to VFA. Women with prevalent fracture had a significantly greater risk of incident VF than women without prevalent fracture.