Predictors of arthritis in pediatric patients with lupus

BACKGROUND Arthritis is one of the most common manifestations of systemic lupus erythematosus (SLE). Although typically non-erosive and non-deforming, children with SLE arthritis can have significant morbidity with decreased quality of life. Our goal was to identify potential clinical and laboratory predictors of arthritis in a cohort of pediatric patients with SLE. METHODS We performed a cohort study of incident and prevalent patients with SLE aged ≤ 19 years. In cross sectional analysis, we compared demographic and clinical characteristics at initial clinic presentation between patients with arthritis noted at any time during follow-up and those without arthritis. We performed time to event analysis using Cox proportional hazard ratios to identify predictors of arthritis, clustering for repeated measures. RESULTS Forty seven children and adolescents with SLE were followed in the cohort, 91 % female and 68 % Black. In cross-sectional analyses, presence of malar rash was associated with arthritis. In longitudinal analyses, controlling for gender and race, increased age (HR: 1.4, 95 % CI: 1.1-1.7), malar rash (HR: 2.1, 95 % CI: 1.1-3.6), and presence of RNP antibodies (HR: 1.9, 95 % CI: 1.1-3.4) were predictive of arthritis. When controlling for gender, race, and medication use, anemia (HR: 8.5, 95 % CI: 2.9-24.2) and thrombocytopenia (HR: 6.1, 95 % CI: 2.4-15.6) were associated with increased risk of arthritis. CONCLUSIONS We identified markers predictive of arthritis in a longitudinal cohort of children with SLE. The recognition of these markers may help clinicians identify patients at risk for arthritis before its onset thus improving quality of life in children with SLE.