Background/Purpose: Rheumatoid Factor positive polyarthritis (RF+ poly) is the JIA subtype that resembles adult seropositive RA. However, the ILAR classification criteria for RF+ polyarthritis do not capture all children with childhood onset RA due to specific exclusion criteria: lack of 2 (+) RF tests, <5 active joints in the first 6 months, family history of psoriasis, and (+) HLA-B27 in boys with onset after age 6. The ACR/EULAR criteria used for diagnosing adult RA do not have these exclusions, and they include the highly specific anti-CCP antibodies (ACPA). The current ILAR classification system does not include ACPA. Hence, children who are RF (-) but ACPA (+) may be treated less aggressively and develop complications secondary to undertreated disease. Our objectives are to 1) determine whether RF and/or ACPA (+) children meet ILAR criteria for RF+ poly JIA and 2) assess for significant differences between children who meet RF+ poly criteria and those who are classified as other subtypes.

Methods: Demographic and disease-related data were collected from charts of RF and/or ACPA (+) children. Each child was classified using ILAR criteria, and the ACR/EULAR classification was used to determine whether each child met criteria for adult RA. Children with RF+ poly JIA were compared to those with other subtypes. Nominal variables were compared using Chi square or Fisher’s exact tests, and continuous variables were compared using Student’s T test.

Results: Of 49 children with RF and/or ACPA (+) JIA, 29 (59%) met criteria for RF+ poly JIA (Table 1). Of the 20 who did not, 9 (45%) met criteria for undifferentiated JIA, 6 (30%) for RF- polyarthritis, 3 (15%) for persistent oligoarthritis, and 2 (10%) for extended oligoarthritis. All children with undifferentiated JIA were ACPA (+); 7 had presentations consistent with oligoarthritis, but had 2 positive RF tests; 1 had a father with psoriasis; 1 was an HLA-B27 (+) boy with onset after age 6. Comparison of children who met criteria for RF+ poly JIA to those who did not revealed significant differences in subtype distribution, number meeting ACR/EULAR criteria for RA, and use of steroids. All other features were not significantly different. The ACR/EULAR criteria for RA captured more children with RF and/or ACPA (+) JIA than the ILAR RF+ poly classification (92% vs. 59%).

Conclusion: A significant number of children (41%) with RF and/or ACPA (+) JIA did not meet criteria for RF+ poly JIA, though many of their demographic features and disease measures were similar to children who did. The ACR/EULAR criteria allow a positive RF or ACPA to qualify as positive serology, and these criteria capture more children with RF and/or ACPA (+) JIA. We propose the inclusion of ACPA in future revisions of the JIA classification criteria to improve the specificity of diagnosing childhood onset RA, and we suggest replacing RF+ polyarthritis with RF/ACPA+ JIA.