POS1108 THE ACCURACY OF OSTEOPOROTIC FRACTURE RISK PREDICTION IN PATIENTS WITH RHEUMATOID ARTHRITIS USING THE PREDICTIVE MODEL DEVELOPED AT V.A. NASONOVA REASEARCH INSTITUTE OF RHEUMATOLOGY (RUSSIA) AND FRACTURE RISK ASSESSMENT TOOL (FRAX)

Abstract

Background:FRAX is a computer-based algorithm that calculates the 10-year probability of a major osteoporotic fracture and the 10-year probability of hip fracture. However, FRAX has several limitations in assessing the risk of fracture in patients with rheumatoid arthritis (RA).In 2013 V.A. Nasonova Reasearch Institute of Rheumatology (Russia) developed a predictive mathematical model for assessing the risk of osteoporotic fractures in RA, which includes 2 main risk factors: cumulative glucocorticoid dose (GC), decrease in BMD in the femoral neck to osteoporosis, and 2 additional factors: for patients under 65 years of age - the presence of ischemic heart disease, and for people over 65 - a history of gastric ulcer or duodenal ulcer.Objectives:To compare accuracy of osteoporotic fracture risk prediction in patients with RA using the predictive model developed at V.A. Nasonova Reasearch Institute of Rheumatology (IR) and FRAX.Methods:This monocentric (single-center) prospective study included 70 patients with RA, aged 40 to 80 years. The follow-up period - 8.0 ± 1.2 years; mean age at the baseline was 55.4±7.8 years old; the mean disease duration at the baseline - 14,7±10,2 years. All patients retrospectively calculated the 10-year probability of fractures and prognostic model developed by the IR.Results:According to the Fracture Risk Assessment Tool, 32 (46%) patients had a low risk of osteoporotic fractures, 38 (54%) had a high risk. According to the predictive model of IR 33 (47%) patients had a low risk of osteoporotic fractures, 37 (53%) had a high risk. During the follow-up period, osteoporotic fractures were occurred in 18 (26%) patients: 14 (78%) of them had a high risk of fractures according to the predictive IR model, and 13 (72%) patients - according to the Fracture Risk Assessment Tool. Positive and negative predictive value of the Fracture Risk Assessment Tool was 34% and 84%, respectively, of the predictive model of IR - 38% and 88%, respectively. Prognosis of the predictive model of IR in 73% cases coincided with assessing the 10-year probability of fracture.Conclusion:The predictive model developed at V.A. Nasonova Reasearch Institute of Rheumatology (Russia) showed a higher sensitivity and specificity in determining the risk of osteoporotic fractures in RA patients vs FRAX algorithm.Disclosure of Interests:None declared.