OBJECTIVES: To investigate the usefulness of severity classification for predicting outcomes in patients with adult-onset Still's disease (AOSD). METHODS: This was a multi-centre retrospective cohort study. AOSD patients were classified into mild, moderate, and severe groups based on severity classification (Japanese Ministry of Health, Labour and Welfare) during the initial treatment, and clinical features were compared among these groups. The primary endpoints were the AOSD-related mortality and drug-free remission rate. For comparison, the same analysis was performed in parallel for patient groups stratified by the modified Pouchot systemic score. RESULTS: According to severity classification, 49 (35%), 37 (26%), and 56 patients (39%) were classified into mild, moderate, and severe groups, respectively. Patients in the severe group showed higher frequency of severe complications and the use of biological agents. Although AOSD-related survival was not significantly different (p = .0776), four of the five fatal cases were classified into the severe group. The severe group showed a reduced rate of drug-free remission (p = .0125). Patient groups classified by systemic score did not correlate with survival or drug-free remission. CONCLUSIONS: Severity classification is useful for predicting outcomes in patients with AOSD.
All Science Journal Classification (ASJC) codes