A multicenter cross sectional study on bone mineral density in rheumatoid arthritis

Objective. To determine the frequency of osteoporosis in a large cohort of women with rheumatoid arthritis (RA) and to investigate the main determinan ts of bone mineral density (BMD) and risk factors for vertebral fractures i n this population. Methods. We recruited 925 consecutive female patients with RA at 21 Rheumat ology Centers in Italy. For each patient pre-registered demographic, diseas e, and treatment-related variables were collected. BMD was measured at lumb ar spine and proximal femur by dual x-ray absorptiometry technique. Collect ed variables underwent a univariate and multivariate statistical procedure. Osteoporosis was defined as BMD > -2.5 T score. Results. The frequency of osteoporosis in the whole sample was 28.8% at lum bar spine and 36.2% at femoral neck and increased linearly from Steinbrocke r's functional stage I to IV (p = 0.0001). Patients with spinal or femoral osteoporosis were significantly older (p = 0.0001), had a lower body mass i ndex (BMI) (p < 0.02), a significantly longer disease duration (p < 0.02) a nd a significantly higher Health Assessment Questionnaire (HAQ) score (p = 0.0001). These differences were significant, even after adjusting for age. Steroid use was associated with significantly lower lumbar and femoral BMD (p = 0.0001) even after adjusting for the main confounding covariates. Anal ysis of lateral spine radiographs revealed 74 women with at least one verte bral fracture. These women had a significantly lower lumbar and femoral BMD (p = 0.0001). The generalized linear model showed that steroid use, menopa use, BMI, age, and HAQ were all significant independent predictors of lumba r and femoral BMD. The logistic procedure showed that age (OR 1.05, 95% CI 1.03-1.07), HAQ (OR 1.3, 95% CI 1.07-1.7), menopause (OR 1.9, 958 CI 1.1-3. 2), use of steroids (OR 1.5, 95% CI 1.07-2.1), and BMI (OR 0.8, 95% CI 0.8- 0.9) were significantly associated with the risk fur osteoporosis. The only variables associated with an increased risk for vertebral fracture were ag e (OR 1.04, 95% CI 1.01-1.08), HAQ (OR 1.7, 95% CI 1.08-2.09), and cumulati ve steroid intake (OR for 1 g of prednisone 1.03. 95% CI 1.006-1.07). Conclusion. To prevent osteoporosis and its dramatic complications in RA th e therapeutic challenge is to preserve functional capacity using the lowest possible dosage of corticosteroids.