Radiologic damage at baseline predicts patient-related outcomes 18 years after the initiation of methotrexate therapy in patients with severe rheumatoid arthritis.

Objective We aimed to assess the association of the degree of radiologic damage at baseline with long-term patient-related outcomes (PRO) in patients with severe rheumatoid arthritis (RA) Methods This prospective observational single-centre study (Ratingen, Germany) included all RA patients starting treatment with methotrexate (MIX) between 1980 and 1987. Standardised clinical evaluations and radiographs of hands and feet were obtained at baseline and during the following years. About 18 years later, patients were invited for a re-assessment. PRO were assessed in three dimensions according to the International Classification of Functioning and Disability (ICF). Statistical analyses comprised multivariable models using baseline values for radiologic damage of hands and feet, age, gender, disease duration, rheumatoid factor positivity, measures of disease activity, and response to MTX as covariates. Results At baseline, the mean disease duration was 8.5 years. The disease was active with a mean number of swollen joints of 18 (out of 32) and a mean erythrocyte sedimentation rate of 55 mm/hour. Radiologic damage was present in 95% of the patients. After 18 years, patient-related outcomes could be assessed in 78/271 patients (29%). Among chosen covariates, only the degree of baseline radiologic damage as measured by the Ratingen score was predictive of all long-term PRO (p<0.016). Conclusion In this cohort including patients with severe RA, baseline radiologic damage was a good long-term predictor of PRO related to all three ICF dimensions.