Lu Zhe, Tan Li-Wen, Xu Hong, Xia Zheng-Kun, Jiang Xiao-Yun, Wu Xiao-Chuan, Wang Fang, Liu Xiao-Rong, Zhao Cheng-Guang, Li Xiao-Zhong, Mao Jian-Hua, Wang Xiao-Wen, Huang Wen-Yan, Shao Xiao-Shan, Zhang Jian-Jiang, Feng Shi-Pin, Yang Jun, Li Qiu, Zhang Ai-Hua, Wang Mo
Department of Nephrology Children's Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, China International Science and Technology Cooperation Base of Child Development and Critical Disorders, Chongqing Key Laboratory of Pediatric Metabolism and Inflammatory Diseases, Key Laboratory of Children's Vital Organ Development and Diseases of Chongqing Health Commission, Zhongshan 2nd Rd. 136, Chongqing, 400014, China.
Department of Nephrology, Children's Hospital of Fudan University, National Pediatric Medical Center of China, Shanghai, China.
World J Pediatr. 2025 Apr;21(4):372-385. doi: 10.1007/s12519-025-00899-2. Epub 2025 May 10.
Anti-neutrophil cytoplasmic antibody-associated vasculitis (AAV) is a type of necrotizing vasculitis with poor prognosis, which is more severe in children. Classifying AAV patients may be helpful for diagnosis and management. However, present classification criteria for pediatric AAV are developed mainly based on adults, which have limitations in clinical practice. In this study, we introduced an updated algorithm based on the European Medicines Agency (EMA) algorithm in conjunction with the American College of Rheumatology (ACR)/European Alliance of Associations for Rheumatology (EULAR) criteria. This new approach aims to resolve the issue of duplicate classification present in the 2022 ACR/EULAR criteria and to refine the existing EMA algorithm.
This study included 179 pediatric patients diagnosed with AAV across 17 centers in China. Patients were classified using the EMA algorithm, the ACR/EULAR criteria, and the EMA-ACR/EULAR algorithm. The Kappa value and Net Reclassification Index (NRI) were used to evaluate the classification performance of these criteria.
According to the EMA algorithm, 136 (76.0%) patients were classified with microscopic polyangiitis (MPA) and 14 (7.8%) with granulomatosis with polyangiitis (GPA), while 29 (16.2%) remained unclassifiable. According to the ACR/EULAR criteria, 145 (81.0%) patients were classified with MPA, 14 (7.8%) with GPA, 2 (1.1%) with eosinophilic granulomatosis with polyangiitis (EGPA), and 4 (2.2%) with both MPA and GPA, while 14 (7.8%) remained unclassifiable. The EMA-ACR/EULAR algorithm classified 124 patients (69.3%) as MPA, 26 (14.5%) as GPA, and 2 (1.1%) as EGPA, while 27 (15.1%) were unclassified. The Kappa values between the EMA algorithm and ACR/EULAR criteria for GPA and MPA were 0.225 [95% confidence interval (CI) 0.000-0.456, P = 0.003] and 0.357 (95% CI 0.196-0.518, P < 0.001). Compared to these two criteria, the EMA-ACR/EULAR algorithm demonstrated positive NRIs in the classification of both GPA (0.702, 95% CI 0.258-1.146, P = 0.002; 0.547 95% CI 0.150-0.944, P = 0.007) and MPA (0.425, 95% CI 0.209-0.642, P < 0.001; 0.519, 95% CI 0.305-0.733, P < 0.001).
The EMA-ACR/EULAR algorithm addresses the limitations of the 1990 ACR criteria within the EMA framework and resolves the issue of duplicate classification in the 2022 ACR/EULAR criteria. However, further research is necessary to validate the superiority of the EMA-ACR/EULAR algorithm in the clinical classification of pediatric AAV patients.
抗中性粒细胞胞浆抗体相关性血管炎（AAV）是一种预后较差的坏死性血管炎，在儿童中病情更为严重。对AAV患者进行分类可能有助于诊断和管理。然而，目前的儿童AAV分类标准主要基于成人制定，在临床实践中存在局限性。在本研究中，我们引入了一种基于欧洲药品管理局（EMA）算法并结合美国风湿病学会（ACR）/欧洲风湿病联盟（EULAR）标准的更新算法。这种新方法旨在解决2022年ACR/EULAR标准中存在的重复分类问题，并完善现有的EMA算法。
本研究纳入了中国17个中心诊断为AAV的179例儿科患者。使用EMA算法、ACR/EULAR标准和EMA-ACR/EULAR算法对患者进行分类。采用Kappa值和净重新分类指数（NRI）评估这些标准的分类性能。
根据EMA算法，136例（76.0%）患者被分类为显微镜下多血管炎（MPA），14例（7.8%）为肉芽肿性多血管炎（GPA），29例（16.2%）仍无法分类。根据ACR/EULAR标准，145例（81.0%）患者被分类为MPA，14例（7.8%）为GPA，2例（1.1%）为嗜酸性肉芽肿性多血管炎（EGPA），4例（2.2%）同时患有MPA和GPA，14例（7.8%）仍无法分类。EMA-ACR/EULAR算法将124例患者（69.3%）分类为MPA，26例（14.5%）为GPA，2例（1.1%）为EGPA，27例（15.1%）未分类。EMA算法与ACR/EULAR标准在GPA和MPA分类上的Kappa值分别为0.225[95%置信区间（CI）0.000-0.456，P = 0.003]和0.357（95%CI 0.196-0.518，P < 0.001）。与这两个标准相比，EMA-ACR/EULAR算法在GPA（0.702，95%CI 0.258-1.146，P = 0.002；0.547 95%CI 0.150-0.944，P = 0.007）和MPA（0.425，95%CI 0.209-0.642，P < 0.001；0.519，95%CI 0.305-0.733，P < 0.001）分类中均显示出正的NRI。
EMA-ACR/EULAR算法在EMA框架内解决了1990年ACR标准的局限性，并解决了2022年ACR/EULAR标准中的重复分类问题。然而，需要进一步研究来验证EMA-ACR/EULAR算法在儿童AAV患者临床分类中的优越性。
