Background Osteoporosis is recognized as the main extra-articular complication of RA and results in an increased risk of fractures. It is important to detect vertebral fractures because they have been associated with increased risk of future vertebral and non-vertebral fractures, reduced quality of life, and increased morbidity and mortality. Vertebral morphometry using dual-energy X-ray absorption (DXA) also known as VFA is a convenient method of assess vertebral deformities, with low radiation and cost than morphometric radiography (MRX

Objectives To validate VFA in the diagnosis of prevalent vertebral fractures in RA as compared to MRX and to evaluate availability of VFA in detecting vertebral fracture in RA patients according to the history of previous vertebral fractures or low back pain.

Methods We prospectively enrolled 100 women with RA of 50 years or older and high risk of osteoporotic fracture. All participants had lateral imaging of thoracolumbar spine by VFA and MRX on the same day or within a few days. Also they completed questionnaires via interview about demographics, previous medical history, and medication. We calculated percent values for sensitivity and specificity of VFA with 95% confidence intervals in detecting vertebral fractures. The inter-rater reproducibility to vertebral fractures on VFA was evaluated by kappa statistics. Also the availability of VFA was evaluated in clinical practice according to the history of previous vertebral fracture or low back pain

Results The prevalence of vertebral fracture identified by MRX was 47 patients (47%). Using VFA, 63 (VFA reader A) and 60 patients (reader B) had at least one vertebral fracture, respectively. The proportion of unreadable vertebrae was higher in VFA (5.3% in reader A and 12.2% in reader B) than in MRX (1.2%) and in upper thoracic vertebrae (T4-T8). Sensitivity and specificity of the diagnosis of fractures using VFA were 60-100% and 75-91% from T12 to L4 respectively. Interrater agreement for per vertebra by two VFA readers was moderate agreement at T9-L4 (k=0.4-0.7).All patients who had a history of vertebral fractures (N=13) were detected by both of MRX and VFA. However, 34 (34%) among 87 patients without vertebral fracture history were also detected occult fracture with MRX. Specificity of VFA for latter group by each readers were high in 94.0% (95% CI: 88.4-97.3%) and 91.4% (95% CI: 85.1-95.6%), while its sensitivity by each readers were low in 55.5% (95% CI: 38.1-72.1%) and 41.2% (95% CI: 29.4-53.8%), respectively. The presence of low back pain was not reliable history for detecting occult vertebral fracture.

Conclusions Vertebral fracture assessment is a useful screening method for detecting occult vertebral fractures in RA patients

Disclosure of Interest None Declared