Objectives: In 2017, a new set of criteria was proposed by EULAR/ACR to classify idiopathic inflammatory myopathies. Our aim was to validate the EULAR/ACR 2017 classification criteria in juvenile dermatomyositis (JDM) patients.
Methods: This study was carried out at Hacettepe University Children's Hospital Department of Paediatrics, Divisions of Rheumatology, Neurology and Paediatric Pathology Unit. Control patients included inborn errors of metabolism presenting with myopathy and/or rhabdomyolysis, idiopathic rhabdomyolysis, dystrophinopathies, neuromyotonia and systemic rheumatic disorders. Patients' data were collected retrospectively from patient files.
Results: Fifty-eight JDM patients (60.3% female) and 40 controls (32.5% female) were included in this study. When the probability cut-off was set at 55% as recommended, the sensitivity/specificity of the new criteria to diagnose JDM were 96.5%/85% in the total cohort, 95.8%/84.6% without the muscle biopsy data and 97%/85.7% with biopsy data. With the ROC curve analysis, the optimal probability cut-off for the whole cohort was found to be >62%; providing a sensitivity/specificity of 96.6% (95% CI: 88.1% to 99.6)/90% (95% CI: 76.3% to 97.2%), and >68.5% for the patients with muscle biopsy providing sensitivity/specificity of 97% (84.7-99.9%)/100% (76.8-100%), respectively. The new EULAR/ACR criteria were the most sensitive, however, the least specific compared to the Tanimoto criteria (sensitivity/specificity 64%/97.5%) and Bohan-Peter criteria (sensitivity/ specificity 74.1%/92.5%).
Conclusions: The new EULAR/ACR criteria performed favourably in our JDM cohort especially with the probability cut-off of >62%.