A comparison of Juvenile Idiopathic Arthritis classification systems with the Research in Arthritis in Canadian Children, Emphasizing Outcomes (ReACCh-Out) cohort

To compare International League of Associations for Rheumatology (ILAR) and Paediatric Rheumatology INternational Trials Organisation (PRINTO) classifications for juvenile idiopathic arthritis (JIA) using data from the Research in Arthritis in Canadian Children emphasizing Outcomes cohort (ReACCh-Out). The aim of the PRINTO classification, still in development, is to identify homogeneous groups of patients with JIA. We used data recorded within 7 months of diagnosis in 1228 patients to assign patients according to the two classification systems by computation. We compared the proportions of patient classified and the alignment of classified categories with clinical-biological subtypes and adult types of arthritis. The PRINTO system classified 244 patients (19.9%) as Early-onset ANA-positive (EOANA), 157 (12.8%) as Enthesitis/Spondylitis-related, 38 (3.1%) as Systemic, and 10 (0.8%) as RF-positive JIA. Twelve percent of patients were unclassifiable using ILAR while 63.3% were unclassifiable using PRINTO criteria (777 Other JIA and 2 Unclassified JIA). In sensitivity analyses, >50% remained unclassifiable by PRINTO. ILAR categories aligned better with clinical-biological subtypes in 131 patients (χ =44, p=0.005 versus χ =15, p=0.07 for PRINTO), and with adult types of arthritis in 389 evaluable patients. Current identified PRINTO disorders classify a minority of patients with JIA, leaving a large proportion with Other disorders requiring further characterization. Current PRINTO classifications do not appear to improve alignment with clinical-biological subtypes or adult forms of arthritis relative to the older ILAR system.