Abstract: Objective　To evaluate the classification ability of EULAR/ACR criteria and B/P criteria in a retrospective cohort of juvenile idiopathic inflammatory myopathy（JIIM）. Methods　Totally 189 children suspected with JIIM who were treated at Children’s Hospital of Chongqing Medical University form January 2010 to April 2020 were recruited. The children’s clinical manifestations，laboratory examinations，serological and histopathological data were collected，and with physician’s diagnosis as standard，the sensitivity of the EULAR/ACR and the B/P criteria was calculated. Results　There were 189 children，86 females and 103 males. The age of onset was 6 months to 17 years and 7 months （7.0±3.6）years. Among them，140 patients were finally diagnosed with JIIM ［JDM 130 cases（92.9%），ADM 5 cases（3.6%），JPM 5 cases（3.6%）］，and 49 patients were diagnosed as non-JIIM；the B/P criteria classified 117 patients （83.6%）as possible/definite JIIM，the EULAR/ACR criteria classified 121 Cases（86.4%）as possible/definite JIIM， and there was no statistically significant difference in the consistency test between the two classification criterias（κ＝0.497，P＜0.001）；in the subgroup classification of JIIM，the EULAR/ACR criteria could accurately classify 96 patients of JDM，sensitivity being 71.1%，and the B/P criteria could accurately classify 80 JDM patients with a sensitivity of 59.2%. The consistency between the two criteria was poor（κ＝0.366，P＜0.001）；EULAR/ ACR criteria was better than the B/P criteria and there was statistical difference（κ ＝0.366，P＜0.001）；25 children obtained complete muscle biopsy data and were positive，of which 24 cases （96.0%） could be classified according to the B/P criteria，while the EULAR/ACR criteria could only classify 22 of them（88.0%）；the sensitivity of the two criteria was compared by Chi-square Fisher’s exact test，and the difference was not statistically significant（P＞0.05）. Conclusion　The 2017 EULAR/ACR criteria shows higher sensitivity and classification rate in children with JIIM in China，especially in the JDM subgroup classification；in the cases of muscle biopsy，there was no statistical difference in the sensitivity of the two classification criteria.

Key words: juvenile idiopathic inflammatory myopathy, EULAR/ACR criteria, B/P criteria, classification 

摘要： 目的　评估EULAR/ACR标准和B/P标准在青少年特发性炎性肌病（JIIM）回顾性队列中的分类能力。方法　对2010年1月至2020年4月在重庆医科大学附属儿童医院就诊的189例疑似JIIM的住院患儿进行回顾性研究，收集患儿的临床表现、实验室检查、血清学和组织病理学数据，计算EULAR/ACR标准和B/P标准的敏感度。结果　189 例患儿，女86例，男103例，发病年龄6月至17岁7月［（7.0±3.6）岁］，其中140例最终临床诊断为JIIM患儿［JDM 130例（92.9%）、ADM 5例（3.6%）、JPM 5例（3.6%）］，另49名被诊断为非JIIM；B/P标准将117例患儿（83.6%）分类为可能/确定的JIIM，EULAR/ACR标准将121例（86.4%）患儿分类为可能/确定的JIIM，两个分类标准一致性检验差异无统计学意义（κ＝0.497，P＜0.001）；进行JIIM亚组分类时，EULAR/ACR标准可准确分类出96例JDM患儿，敏感度为71.1%，B/P标准可准确分类出80例JDM患儿，敏感度为59.2%，两组标准之间的一致性差，EULAR/ACR标准优于B/P标准，有统计学意义（κ＝0.366，P＜0.001）；25例患儿可获得肌活检完整数据且阳性，其中24例（96.0%）根据B/P标准可进行分类，而EULAR/ACR标准仅可对其中22例（88.0%）患儿进行分类，两种标准的敏感度经卡方fisher精确检验比较，差异无统计学意义（P＞0.05）。结论　2017年EULAR/ACR标准在中国JIIM患儿中显示出更高的敏感度和分类率，尤其在进行JDM亚组分类时更有优势；在有肌活检情况下，两种分类标准的敏感度差异无统计学意义。

关键词: 幼年特发性炎性肌病, EULAR/ACR标准, B/P标准, 分类