Background/Purpose: Dysregulation of the human microbiome has been implicated in the development of several autoimmune diseases, including rheumatoid arthritis and inflammatory bowel disease (IBD). Moreover, antibiotic exposure has been linked with the development of IBD in children. This study aimed to determine whether early antibiotic exposure increases the risk of incident juvenile idiopathic arthritis (JIA) in a general pediatric population.

Methods: A nested case-control study was conducted using The Health Improvement Network, a United Kingdom population-based medical records database with comprehensive diagnostic and outpatient prescription data. Children with incident JIA diagnosed before age 16 were identified by validated diagnostic codes (positive predictive value 86%). Age- and sex-matched control subjects were randomly selected with incidence density sampling in a 10:1 ratio from general practices taking care of at least 1 child diagnosed with JIA. Eligible subjects needed to be registered within 3 months of their birthdate. Individuals with prior IBD, immunodeficiency, autoimmune connective tissue disease, or vasculitis were excluded. The association between antibiotic prescriptions and JIA diagnosis was determined by conditional logistic regression.

Results: There were 153 children diagnosed with JIA in the study population (table 1). Any antibiotic exposure was associated with an increased risk of developing JIA after adjusting for confounders (adjusted OR 2.6, 95% CI 1.5,4.6) (table 2). This risk increased with the number of prescriptions in a dose-dependent manner. These results did not significantly change when adjusting for the number or type of infections. Age of exposure did not significantly modify this association. The relationship between antibiotics and incident JIA was similar across different antibiotic classes, although use of non-bacterial antimicrobial agents (e.g., antifungal, antiviral) was not associated with JIA. In sensitivity analyses excluding data up to 12 months before the index date, the association between antibiotics and incident JIA did not substantively change.

Conclusion: Antibiotic exposure was associated with an increased incidence of JIA in a dose-dependent fashion in a large pediatric population. This study implicates a role for antibiotic exposure in disease pathogenesis, perhaps mediated through alteration in the microbiome.