MON-470 Real-World Data of Adrenal Crisis using a Large-scale Administrative Claims Database

Abstract Disclosure: M. Katsumata: None. Y. Nishioka: Novo Nordisk, Sanofi, Daiichi Sankyo, DeSC Healthcare. H. Nakajima: None. F. Kamitani: None. Y. Kurematsu: None. T. Myojin: None. T. Noda: None. T. Imamura: None. Y. Takahashi: Novo Nordisk, Otsuka, Recordati, Sumitomo Dainippon, Eli Lilly & Company, ONO-Pharma, Novartis Pharmaceuticals, Boehringer Ingelheim, AstraZeneca, Kyowa Kirin. Background: Although Adrenal Crisis (AC) is life-threatening emergency in patients with chronic adrenal insufficiency (AI) and its epidemiology and pathophysiology are important, previous studies have limitations including several patients and the definition of AC. To overcome these limitations, we analyzed a severe AC defined by hospitalization using a large-scaleadministrative claims database. Methods: We used the DeSC Japanese administrative claims database with 13 million patients. Chronic AI was defined by a related disease name and continuous steroid replacement therapy. AC was defined by a hospitalization and intravenous hydrocortisone (HC) administration for at least 2 days. Results: In total, 9523 patients with chronic AI were identified and the underlying conditions were primary AI (9%), secondary AI (59%), and others (32%), respectively. Among these patients, 581 patients experienced 747 AC during a mean follow-up period of 3.7 years. Twenty-one % patients developed AC more than once and the incidence rate increased exponentially with the first time of 1.7 (100 person/year), the second time of 12.2, the third time of 17.8, the fourth time of 22.0, and the fifth time of 25.7, respectively. There was no difference in the first-time incidence rate between men and women (P = 0.68). The odds ratio of developing AC for age showed a U-shaped curve with OR of 1.3 100 person/year (P = 0.30) for 18-39 of age, 1.0 for 40-59 years of age (reference), 1.4 (P = 0.03) for 60-79 years of age, and 1.6 (P = 0.003) for 80 years or older, respectively. The monthly AC incidence rate were lowest in May (0.09%; reference) and significantly increased in January (RR 2.0; P < 0.001), February (2.4; P = 0.03), July (1.8; P = 0.002), and August (1.8; P < 0.001), respectively, indicating that it was more prevalent in winter and summer season. As the underlying diseases that triggered AC, respiratory (24.6%), gastrointestinal (14.6%) infections were the most common. Interestingly, Influenza infection (3.9%) was prevalent in January, demonstrating that AC risk was affected by seasonal infectious diseases. Finally, the acute phase mortality within 30 days during the hospitalization was 7.3% (mean age was 81.8 years, 67% male). There was no difference between the single and multiple AC groups (7.1% and 7.6%, respectively). Multivariate analysis associated with the acute-phase mortality adjusted by age, sex, steroids, and underlying diseases and conditions showed that the primary AI (OR 3.2, P = 0.02) was the significant risk factor. Conclusion: In this study, we demonstrated the epidemiology of AC including risk factors and acute phase mortality using a large-scale administrative claims database. We first showed a seasonal difference in the risk and associated factors of AC. These data suggest the importance of appropriate precautional approaches such as a vaccination for influenza, especially in elderly AI patients. Presentation: Monday, July 14, 2025