Open AccessPrognostic factors for radiographic damage in early rheumatoid arthritis: A multiparameter prospective study
Author(s) -
Combe B.,
Dougados M.,
Goupille P.,
Cantagrel A.,
Eliaou J. F.,
Sibilia J.,
Meyer O.,
Sany J.,
Daurès J.P.,
Dubois A.
Publication year - 2001
Publication title -
arthritis & rheumatism
Language(s) - English
Resource type - Journals
eISSN - 1529-0131
pISSN - 0004-3591
DOI - 10.1002/1529-0131(200108)44:8<1736::aid-art308>3.0.co;2-i
Subject(s) - medicine , rheumatoid factor , rheumatoid arthritis , erythrocyte sedimentation rate , anti nuclear antibody , univariate analysis , prospective cohort study , cohort , rheumatology , gastroenterology , autoantibody , antibody , immunology , multivariate analysis
Abstract Objective To determine prognostic factors of radiologic damage and radiologic progression in early rheumatoid arthritis (RA). Methods A cohort of 191 patients with RA whose disease duration was shorter than 1 year were prospectively followed up for 3 years. Radiologic scores (as determined by Sharp's method, modified by van der Heijde) and radiologic progression were used as outcome measures. Numerous baseline clinical, laboratory, genetic, and radiographic data were obtained. Results The change in the total radiologic score for the patients followed up over 3 years was a mean ± SD increase of 6.1 ± 6.2. Radiologic progression was observed in 71 of the 172 patients for whom there were data at the end of the study. By univariate analysis with Fisher's exact test, radiologic scores and progression at followup were closely correlated with the baseline values of the erythrocyte sedimentation rate (ESR), C‐reactive protein level, IgM and IgA rheumatoid factor positivity, antiperinuclear antibody positivity, radiologic scores, duration of morning stiffness, and RA‐associated HLA–DRB1∗04 genes. No correlation was demonstrated with sex, age, Disease Activity Score, swollen or tender joint counts, extraarticular manifestations, Health Assessment Questionnaire score, Ritchie Articular Index, patient's assessment of pain, positivity for anti–heat‐shock protein 90‐kd antibodies, anticalpastatin antibodies, anti‐RA33 antibodies, antinuclear antibodies, YKL‐40, or antikeratin antibodies, and HLA–DRB1∗01 genes. The logistic regression analysis revealed that the only baseline values that were predictive of the 3‐year radiologic scores were IgM rheumatoid factor positivity, DRB1∗04 genes, pain score, and total radiologic score. Progression of joint damage was predicted by the ESR, IgM rheumatoid factor positivity, DRB1∗04 genes, and erosions score at baseline. Conclusion Prognostic factors for radiographic damage in early RA were identified. A combination of these baseline values allowed us to draw up a predictive arithmetic score that could be used to predict radiologic damage at 3 years and radiologic progression in individual patients.