175-LB: Stress Hyperglycemia Predicts Mechanical Ventilation and Death in COVID-19 Infected Adults

Abstract

People with diabetes (DM) hospitalized with COVID-19 infection have a 2-3 fold higher risk of death compared with those without DM, and mechanical ventilation (MV) has been identified as a major risk factor for death. While stress hyperglycemia (StH) has been established as a risk factor for death in some critically ill cohorts, it is not a well-established risk factor for MV in COVID-19. The COVIDEastDM consortium pooled data from 5 academic hospitals on the East Coast of the US to study the relationship between hyperglycemia and COVID-19 outcomes. Data were obtained retrospectively from electronic records of adults with COVID-19 and either DM or StH (defined in this cohort as day-1 admission blood glucose >180 mg/dl and A1c <6.5%). This analysis included 3,435 individuals, of which 1,001 (29.1%) required MV and 748 (21.8%) died. The mean age was 67 ± 15 yrs., BMI 30.4 ± 7.7 kg/m2, A1c 7.98 ± 2.21 % and glucose 184 ± 104 mg/dl. Additionally, 57% were male, 4 % Asian,19% Black, 52% White, and 28% Hispanic. In a univariate analysis, risk factors for MV included younger age (OR 0.98 [95% CI 0.97, 0.99] per year older), male sex (OR 1.4 [1.2, 1.7]), BMI (OR 1.013 [1.003, 1.023] per kg/m2), Hispanic ethnicity (OR 1.16 [1.07, 1.28]) and the presence of diabetic ketoacidosis (n=38) at admission (OR 3.9 [2.0, 7.5]). Patients with StH were more likely to require MV (OR 1.93, [1.10, 3.39]). In a multivariate analysis, this relationship was continuous, with both lower A1c and higher glucose increasing risk of MV (p< 0.01 for higher glucose, p<0.001 for lower A1c). Patients who required MV were more likely to die than those who did not require MV (OR 5.1, [4.3, 6.1]) and StH predicted mortality in the multivariate analysis. Thus, in patients hospitalized with COVID19 infection, StH is a strong predictor of both MV and death in COVID-19 infected adults. While it is unclear if StH is a cause or effect of severe COVID-19, its presence early in the hospital course identifies higher risk patients and could potentially impact management. Disclosure M. E. Mcdonnell: Stock/Shareholder; Spouse/Partner; Abbott Diabetes. N. E. Palermo: None. R. Radhakrishnan: None. G. P. Westcott: None. R. S. Weinstock: Research Support; Self; Boehringer Ingelheim International GmbH, Diasome Pharmaceuticals, Inc., Eli Lilly and Company, Insulet Corporation, Kowa Research Institute, Inc., Medtronic, Tolerion, Inc. R. Garg: None. D. C. Simonson: Stock/Shareholder; Spouse/Partner; Phase V Technologies, Inc. G. Cromwell: None. G. Gopalakrishnan: None. M. Greenfield: None. M. Johnson: None. S. R. Katta: None. J. Lebastchi: None. J. Mitri: Consultant; Self; L-Nutra. Funding TechFoundation; Brigham Education Institute