Background: Juvenile idiopathic arthritis (JIA) is the most common rheumatic disease in children, with lower limb involvement highly prevalent. Recent evidence has highlighted the lack of specific lower limb physical examination (PE) tools for clinicians assisting the paediatric rheumatology team in identifying lower extremity disease in patients with JIA. Early clinical detection may lead to more prompt and targeted interventions to reduce lower limb problems in children with JIA. The aim of this pilot study is to provide preliminary data on the diagnostic accuracy of a lower limb PE tool in JIA. Methods: Children with JIA requiring magnetic resonance imaging (MRI) on their lower limb joints per their usual care were eligible. Lower limb joint counts were conducted clinically by a podiatrist and paediatric rheumatologist using the proposed twenty joint per side, PE tool. The PE were compared to MRI assessments completed by two independent paediatric radiologists. Data were analysed using agreement (observed, positive and negative) and Cohen’s kappa with 95% CIs. Results: Fifteen participants were recruited into the study in which 600 lower limb joints were clinically examined. Statistical analysis showed excellent inter-rater reliability between podiatrist and paediatric rheumatologist for both joint swelling and tenderness. Results of the intra-rater reliability of the podiatrist using the PE tool indicated excellent percentage agreements (98.5–100%) and substantial kappa coefficients (0.93–1). The inter-rater reliability between radiological assessments contrasted the PE results, showing low agreement and poor reliability. Comparisons between PE and MRI resulted in poor kappa coefficients and low agreement percentages. The most agreeable joint between MRI and PE was the ankle joint, while the worst performing joint was the sub-talar joint. Conclusion: Results indicate potential clinical reliability; however, the validity and diagnostic accuracy of the proposed PE tool remains unclear due to low kappa coefficients and inconsistent agreements between PE and MRI results. Further research will be required before the tool may be used in a clinical setting.

No abstract