医学
临床实习
梅德林
心理学
疾病
精神科
家庭医学
儿科
医学诊断
普通外科
作者
You-Ri Kang,K. Y. Kim,Jae-Won Hyun,Ho Jin Kim,Su-Hyun Kim
标识
DOI:10.3988/jcn.2025.0529
摘要
BACKGROUND AND PURPOSE: Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) has been increasingly recognized, yet concerns remain regarding the overuse and misinterpretation of MOG-IgG testing. The recently published MOGAD diagnostic criteria emphasize selective testing and rigorous interpretation in the context of compatible clinical syndromes and supportive features. We investigated, in real-world practice, the extent to which these criteria are appropriately applied and the frequency and characteristics of their misapplication at a referral center. METHODS: We retrospectively reviewed patients referred to the National Cancer Center with externally reported MOG-IgG positivity between January 2021 and December 2024. External assay results were based on laboratory-specific cutoffs, which had not been verified against clinically validated thresholds. Final diagnoses were determined by expert consensus after comprehensive evaluation, applying the 2023 international MOGAD diagnostic criteria. RESULTS: Fifty-seven patients with external MOG-IgG positivity were referred, of whom 48 (84.2%) had been labeled as MOGAD. Upon re-evaluation, only 39 patients (68.4%) met the diagnostic criteria. Of the 18 non-confirmed patients, 5 (27.8%) manifested nonspecific symptoms incompatible with core demyelinating events, 9 patients (50.0%) received an alternative diagnosis, and 4 (22.2%) presented core events but lacked supportive features. CONCLUSIONS: Over one-third of patients with externally reported positive results did not meet diagnostic criteria, illustrating the risk of non-targeted MOG-IgG testing and overinterpretation of antibody results. Selective test utilization and strict adherence to diagnostic criteria in the context of appropriate clinical syndromes and supportive findings are essential to prevent overdiagnosis and inappropriate treatment.
科研通智能强力驱动
Strongly Powered by AbleSci AI