Effective Performance of the 2022 American College of Rheumatology/EULAR Classification Criteria for Antineutrophil Cytoplasmic Antibody–Associated Vasculitis in Pediatric Patients: An ARChiVe Study

Abstract

Objective: To assess the 2022 American College of Rheumatology (ACR)/EULAR classification criteria for antineutrophil cytoplasmic antibody–associated vasculitis (AAV) in children with chronic small-to-medium vessel vasculitis. Methods: A cohort of 574 patients, identified by physician's diagnosis (MD-diagnosis) in A Registry of Childhood Vasculitis, was classified by computation of registry data as having granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), or eosinophilic GPA after applying (1) ACR/EULAR AAV criteria and (2) pediatric-adapted European Medicines Agency (Ped-EMA) classification algorithm (incorporating Ankara GPA criteria). Venn diagrams compared the resulting GPA and MPA cohorts with MD-diagnosis. Sensitivity and specificity of criteria for GPA were evaluated against MD-diagnosis. Fisher exact test evaluated differences in the frequencies of individual clinical features in GPA versus MPA. Results: Comparing ACR/EULAR criteria against the Ped-EMA algorithm for classifying AAV, more patients were classified as GPA or MPA (n = 396 vs 360, respectively), fewer had GPA (n = 261 vs 288, respectively), more had MPA (n = 135 vs 72, respectively), and fewer GPA cases coclassified as MPA (12% vs 28%, respectively); there were more differences between GPA and MPA in Pediatric Vasculitis Activity Score–defined clinical features (n = 14 vs 10, respectively). When classifying GPA by ACR/EULAR or Ankara criteria, sensitivity (74.5% vs 72.1%, respectively) was comparable, and specificity for ACR/EULAR criteria (93.9% vs 79.9%, respectively) was improved. Conclusion: The 2022 ACR/EULAR classification criteria for AAV perform at least as well as previous pediatric criteria and provide categorical MPA criteria where none existed previously; the criteria for GPA and MPA now specifically differentiate each other, with more differences between them in the frequencies of clinical features. Our findings support the preferential use of ACR/EULAR over Ankara criteria for GPA in pediatrics. (Figure presented.).