AB1026 ASSOCIATION BETWEEN TRABECULAR BONE SCORE, 10-YEAR PROBABILITY RISK FOR FRACTURE, AND VERTEBRAL FRACTURES IN RHEUMATOID ARTHRITIS

Abstract

BackgroundTrabecular bone score (TBS) utility in vertebral fracture prediction is still questionable.ObjectivesTo evaluate the association of trabecular bone score (TBS), the fracture risk assessment tool (FRAX), and TBS adjusted FRAX with prevalent vertebral fractures (VFs) in Rheumatoid Arthritis (RA).MethodsThis is a cross-sectional study of RA patients diagnosed according to ACR/EULAR 2010 criteria and recruited in our rheumatology department. Patients with an age under 40 or above 90, diabetes, thyroid disease, hyperparathyroidism, cancer, cirrhosis, renal clearance < 70 mL/min, and those using osteoporosis drugs or having a body mass index (BMI) of more than 36kg/m2 were excluded. TBS, bone mineral density (BMD), and vertebral fracture assessment (VFA) were obtained using dual-energy X-ray absorptiometry (DXA). We used the FRAX tool to assess the 10-year probability of major osteoporotic fracture (MOF-FRAX) and hip fracture (HF-FRAX) with and without BMD. These parameters were further adjusted for TBS (FRAX/TBS-MOF and FRAX/TBS-HF). Prevalent VF was evaluated by two readers using VFA. We defined patients with prevalent VFs as patients who had moderate-to-severe degree VFs of the thoracic and lumbar spine (T4 to L4) according to Genant criteria. Osteoporosis was defined as a T-score < -2.5 in the lumbar spine, femoral neck, or 1/3 of the non-dominant forearm location. Clinical data, results of radiological and laboratory tests performed on the same day of the DXA scan were collected. Receiver operating characteristic curves were generated by SPSS.20 to study the predictability of each parameter for VF.ResultsSixty-nine patients were enrolled, and 66.7% of them were menopausal. Their mean lumbar spine, hip, and forearm BMD were 0.817 ± 0.155, 0.681 ± 0.676, and 0.572 ± 0.119, respectively. The mean TBS was 1.328 ± 1.041. Thirty-nine patients (56.5 %) had osteoporosis. Nine patients (13%) had moderate to severe VFs with predominance at the thoracic spine (77.8%). There was a negative correlation between MOF-FRAX score and lumbar spine BMD (r = -0.473; p < 0.001), and also hip BMD (r = - 0.530; p < 0.001), while there was no correlation between MOF-FRAX score, HF-FRAX and TBS. The areas under curves were 0.824, 0.800, 0.792, and 0.443 for the FRAX-MOF score with BMD, FRAX-MOF score, FRAX-MOF score adjusted for TBS, and TBS, respectively.ConclusionIn this study, MOF-FRAX score showed a better correlation with BMD than TBS. It was also superior to TBS and TBS adjusted FRAX in identifying prevalent VFs in RA patients.Disclosure of InterestsNone declared