The early pattern of joint involvement predicts disease progression in children with oligoarticular (pauciarticular) juvenile rheumatoid arthritis

Objective To evaluate features during the first 6 months of disease that may be associated with a poor outcome as measured principally by extension to a polyarticular disease course in patients with oligoarticular‐onset juvenile rheumatoid arthritis (oligo‐JRA). Methods This study was a retrospective review of patients who fulfilled the American College of Rheumatology criteria for oligo‐JRA, were followed up for at least 5 years, and did not have juvenile psoriatic arthritis, spondylarthropathy‐like disease, or rheumatoid factor positivity. Data from the first 6 months of disease were collected. Continuous variables were dichotomized and then screened by univariate analysis for association with poor outcome at the last followup visit, as measured by extension of involvement (>4 accumulated involved joints) and by “clinically meaningful” extension (≥10 accumulated joints). Variables significantly associated with this latter outcome, with the addition of disease duration as a confounding independent variable, were included in a multiple logistic regression analysis. The same variables were then examined in separate multiple logistic regression models to look at other measures of outcome, including use of disease‐modifying antirheumatic drugs (DMARDs) at any time, erosive disease on radiographs, any remission of disease ever occurring, physician's global assessment of disease activity at the last visit, and disability as measured by the Childhood Health Assessment Questionnaire (C‐HAQ)/HAQ. Results Of the 205 patients (160 of whom were female) studied for a median of 10.8 years (range 5–26.6 years), 39.5% developed extension to >4 joints and 17.6% developed arthritis in ≥10 joints. Using the logistic regression model, symmetric disease was predictive of all measures of poor outcome: extension to ≥10 joints (odds ratio [OR] 19.2), the need to use DMARDs (OR 11.5), radiographic demonstration of erosive disease (OR 4.73), inflammatory activity at last followup visit (OR 3.23), no remission of disease (OR 4.73), and disability as measured by a C‐HAQ score >0.12 (OR 2.95). Ankle and/or wrist disease was predictive of extension (OR 6.61) and erosions (OR 3.59). Wrist disease alone was predictive of the need to use DMARDs (OR 5.87) and of inflammatory disease activity at the last followup visit (OR 4.01). An elevated erythrocyte sedimentation rate (ESR) was predictive of extension (OR 3.76), the need to use DMARDs (OR 6.47), and no remission of disease (OR 2.30). D