Clinical Features and Outcomes of Myxedema Coma in Patients Hospitalized for Hypothyroidism: Analysis of the United States National Inpatient Sample.

Abstract

Background: Hypothyroidism is a common endocrine condition and chronic thyroid hormone deficiency is associated with adverse effects across multiple organ systems. In compensated hypothyroidism, however, patients remain clinically stable due to gradual physiological adaptation. In contrast, the clinical syndrome of decompensated hypothyroidism referred to as myxedema coma (MC) is an endocrine emergency with high risk of mortality. Because of its rarity, there are currently limited data regarding MC. This study analyzes the clinical features and hospital outcomes of MC compared with hypothyroid patients without MC (nonMChypo) in national United States hospital data. Methods: A retrospective analysis of the National Inpatient Sample, a public database of inpatient admissions to nonfederal hospitals in the United States, 2016-2018, including adult patients with primary diagnosis of MC (International Classification of Diseases 10th Revision [ICD-10]: E03.5) or nonMChypo (E03.0-E03.9, E89.0). Patient demographics, relevant clinical features, mortality, length of stay (LOS), and hospital costs were compared. Results: Of 18,635 patients hospitalized for hypothyroidism, 2495 (13.4%) had a diagnosis of MC. Sex distribution and race/ethnicity were similar between patients with MC and nonMChypo, whereas MC was associated with older patient age (p = 0.02), public insurance (p = 0.01), and unhoused status (p = 0.04). More admissions with MC occurred in winter compared with other seasons (p = 0.01). The overall mortality rate for MC was 6.8% versus 0.7% for nonMChypo (p < 0.001), and MC was independently associated with in-hospital mortality after adjusted regression analysis (adjusted odds ratio = 9.92 [CI 5.69-17.28], p < 0.001). Mean LOS ± standard error was 9.64 ± 0.73 days for MC versus 4.62 ± 0.12 days for nonMChypo (p < 0.001), and total hospital cost for MC was $21,768 ± $1759 versus $8941 ± $276 for nonMChypo (p = 0.07). In linear regression analyses, MC was an independent predictor of both increased LOS and total hospital cost. Conclusions: In summary, MC remains a clinically significant diagnosis in the modern era, independently associated with high mortality and health care costs. This continued burden demonstrates a need for further efforts to prevent, identify, and optimize treatment for patients with MC.