Radiographic outcome of recent-onset rheumatoid arthritis: a 19-year study of radiographic progression

OBJECTIVE: To describe the longitudinal radiographic course of rheumatoid arthritis (RA), and to identify and quantitate predictors of radiographic progression. METHODS: This prospective, longitudinal study of radiographic progression and clinical predictors of RA involved 256 patients with RA who were seen within the first 2 years of disease (mean 0.77 years) and were followed up for up to 19 years. Participants underwent a total of 6,278 clinical assessments (mean 24.5) and 934 paired radiographs (mean 3.1, range 2-6). Clinical assessments at every visit included determination of the erythrocyte sedimentation rate (ESR), grip strength, pain scores, tender joint counts, and anxiety and depression measurements. Regression analyses utilized time-integrated predictors. RESULTS: Overall, radiographic progression rates, as measured by the summary Sharp scores, appeared constant over the course of RA. The strongest correlate of progression was the time-integrated ESR (rho=0.53). This association grew stronger with time. At 0-5 years, 5-10 years, 10-15 years, and 15-20 years, correlations were 0.40, 0.50, 0.65, and 0.74, respectively, and for the period 10-20 years, the correlation was 0.67. In multivariate models, the mean ESR, mean grip strength, rheumatoid factor positivity, and tender joint count were independent predictors of radiographic progression. CONCLUSION: Radiographic damage occurs at a constant rate in RA, and is not greater early in RA or reduced later in the course of the illness. Acute-phase reactants are, by far, the strongest determinants of progression.