Visual Abstract
Hyperglycemic crises, diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS), are being increasingly reported in patients with COVID-19. To date, studies have not directly compared hyperglycemic crises in patients with and without COVID-19. In this retrospective cross-sectional study, 55 patients admitted to an urban academic medical center since 3/1/2020 meeting laboratory criteria for DKA, HHS, or mixed DKA/HHS were grouped by COVID-19 status per RT-PCR testing and compared (see table). Data were collected by electronic and manual chart abstraction. Diabetes type was adjudicated by an endocrinologist. The whole cohort (mean age 55 ± 17 years) was 53% male, 55% black, 13% white, and 20% Hispanic. There were similar proportions of DKA, HHS, and mixed cases between patients with and without COVID. Compared to those without COVID, patients with COVID had lower anion gap, higher pH and beta-hydroxybutyrate, were more likely to have type 2 diabetes (T2D, 76% vs. 53%) and less likely to have T1D (8% vs. 33%), were 4-times more likely to receive glucocorticoids (88% vs. 20%), had nearly double the length of stay (LOS, 14.1 ± 14.9 vs. 7.4 ± 7.0, p=0.03) and 4-fold higher odds of mortality (OR 4.42 [0.81-24.28], p=0.09). These data suggest there are differences in hyperglycemic emergencies between patients hospitalized with and without COVID, most notably a more frequent history of T2D, longer LOS and greater mortality.
A. A. Shah: None. E. Silfani: None. A. Deak: None. C. Koppin: None. S. Allen: None. Y. Zisman-ilani: None. C. Rose: None. I. Sirisena: None. D. J. Rubin: Research Support; Self; AstraZeneca.
