ACHIEVING FREEDOM FROM GLUCOCORTICOIDS MIGHT DECREASE RISK OF CLINICAL FRACTURES IN PATIENTS WITH RHEUMATOID ARTHRITIS: FIVE-YEAR FINDINGS OF THE TOMORROW STUDY

Background: Patients with rheumatoid arthritis (RA) who have muscle weakness and stiff or painful joints might be at increased risk of falls and fractures. Objectives: The present study prospectively investigates correlations between decreasing doses of glucocorticoid (GC) and the incidence of clinical fractures in patients with RA based on the five-year findings of the TOMORROW study (UMIN000003876) that started in 2010. Methods: We evaluated anthropometric parameters, bone mineral density, disease activity, RA medication, and the incidence of clinical fractures over a period of five years in 202 patients with RA (mean age, 58.6 years; mean disease duration, 14.0 years). We also assessed the effects of GC doses on the incidence of clinical fractures over the same period in patients with RA using multivariate regression analysis. Results: The incidence of clinical fractures in patients with RA was 0.042/person-years (py). There were 87 RA patients (41.8%) treated with GC whose incidence rate and number of clinical fractures were significantly higher than those without GC treatment (27.4% vs. 11.9%; p=0.008; 0.063 vs. 0.012 py; p=0.012, respectively). After adjusting for fracture risk factors including age, sex, smoking, and body mass index, multivariate logistic regression analysis revealed that GC administered within the five-year period was a significant risk factor for clinical fractures (odds ratio [OR], 2.58; 95% confidence interval [CI], 1.22–5.44, p=0.013). A mean GC dose during the 5-year period of ≥2 mg/day increased risk for fractures in patients with RA (OR 3.48; Table 1). Although only reducing the GC dose did not decrease the risk of clinical fractures in patients with RA, risk was significantly lowered when the dose was reduced to zero within the five-year period (OR 0.21; Table 1). Table 1. Odds ratios for fractures determined by multivariate logistic regression analysis in 87 patients with RA treated by GC OR95% CIp History of fractures1.990.95–4.160.069 Average GC dose (≥2 mg/day/5 y)3.481.19–10.210.023 Only reducing GC within 5 y0.610.21–1.780.362 Reducing GC to zero within 5 y0.210.07–0.650.006 Bisphosphonate medication3.371.51–7.540.003 Low BMD at thoracic vertebrae (<0.7 g/cm)4.441.69–11.64<0.001 RA, rheumatoid arthritis; GC, glucocorticoid; OR, odds ratio; CI, confidence interval; BMD, bone mineral density. Conclusions: Medication with GC was a significant risk factor for clinical fractures, and low doses are apparently significantly associated with an increased frequency of fractures among patients with RA. However, reducing the GC dose to zero among RA patients within five-years could lower the risk for clinical fractures. We concluded that GC medication should be tapered to zero over a period of five years in patients after RA activity is controlled. Disclosure of Interest: None declared DOI: 10.1136/annrheumdis-2016-eular.4142Citation: Annals of the Rheumatic Diseases, volume 75, supplement 2, year 2016, page 713Session: Rheumatoid arthritis - comorbidity and clinical aspects (Poster Presentations )