Background/Purpose: Pharmachild is an international registry involving over 100 the Paediatric Rheumatology INternational Trials Organisation (PRINTO)/ the Paediatric Rheumatology European Society (PRES) centres in 38 countries. The registry was set up to evaluate long term safety and efficacy of treatments in children with JIA. The purpose of the study was to evaluate the occurrence of infections in JIA children and their relationship with biologics ± methotrexate (MTX) and other immunosuppressants.

Methods: Retrospective and prospective evaluation of moderate to very severe and serious infections. We investigated, by means of bivariate and multivariate modeling, the relationship between infections and the drugs used in the 6 month-history prior to the infection and, for patients without infections, in the 6 months prior to the last follow-up visit. The patients receiving only NSAIDs or intra-articular steroid injections were used as a reference.

Results: We analyzed 6969 patients, for a total of 689 (10%) patients with infections. Patients were systemic 639 (9%), polyarticular course 4587 (66%) and persistent oligoarthritis 1473 (25%). Patients with infections were treated in the previous 6 months primarily with MTX (82%), corticosteroids (32%), anti-TNF (52%), anti IL-1/IL-6 (8%), rituximab (2%) or other biologics (2%).Earlier age at diagnosis (≤7.7 years), ANA positivity, and systemic JIA category significantly increased the risk for infection (OR 2.3, 1.7 and 2.2, respectively). The analysis of the impact of the single immunosuppressive drugs showed that the risk for infection is increased by corticosteroids (OR 3.7), MTX (OR 2.8), cyclosporine (OR 5.3) and thalidomide (OR 13.7, 3/5 patients with infections). The same results were observed for biologics, in particular rituximab (OR 16.3, 14/22 patients with infections), IL-1 inhibitors as anakinra (OR 3.3) and, among TNF-α inhibitors, infliximab (OR 1.6). Conversely, abatacept resulted protective (OR 0.4). The multivariate analysis (table) revealed that the addition of steroids to both MTX and biologics (groups MTX + steroids and MTX + steroids + biologics) increased the risk of infections more significantly (OR 11.9 and 10.5, respectively) than only MTX (5.1) or MTX+Biologics (OR 3.7). A great increase in the risk for infection was associated with rituximab ± steroids ± DMARDs (OR 112, 95% CI 11-1000). The multivariate results were confirmed by removing rituximab patients from the model. Table. Risk factors for infections in JIA patients by multivariate analysis. (REF: reference item)

Conclusion: Pharmachild showed that MTX and biologics increase the risk of infection in JIA patients. This risk is significantly enhanced by the addition of steroids to immunosuppressive therapy. We recommend monitoring for infections in JIA patients on immunosuppressive therapy.