Unsuspected vertebral fracture in older women undergoing screening bone mineral density assessment: Prevalence, impact, and agreement with lateral spine radiography

Abstract

© 2016 Elsevier Masson SAS and European Union Geriatric Medicine Society Objectives To define the prevalence of unsuspected vertebral fracture in older women undergoing screening dual X-ray absorptiometry (DXA) using vertebral fracture assessment (VFA), to assess the impact of VFA on patient management by providing additional information to BMD and FRAX®, and to assess the agreement between VFA and standard lateral spine radiography. Materials and methods The IRB approved the study protocol. Medical records and VFA images of consecutive women aged ≥ 60 years referred for screening DXA during August 2009 to July 2010 were retrospectively reviewed. Those with known conditions potentially affecting bone metabolism or severe scoliosis were excluded. The fracture risk was estimated using FRAX®. An agreement on the presence and type of vertebral fracture was conducted on a subset of 30 subjects. Results Of 634 eligible women, unsuspected vertebral fractures were identified in 153, resulting in an overall prevalence (95% confidence interval [CI]) of 24.1% (20.8–27.6%) and 17.4% (20.8–27.6%) for fractures of any grade and of grade 2–3, respectively. Among non-osteoporotic, low-risk women, the detection of vertebral fracture changed the diagnosis in 17.3% and treatment in 11.1%, demonstrating the impact of VFA. The agreements between VFA and radiography for T11-L4 vertebrae were 0.64–1. Conclusions The prevalence of unsuspected vertebral fracture as defined by VFA was significant in women aged ≥ 60 years referred for screening DXA. By showing vertebral fractures in low-risk non-osteoporotic subjects, VFA appeared to have clinical impact. Its agreement with radiographs for the detection of vertebral fracture was moderate to good.