We sought to determine the associations between hemoglobin A1c (A1c) and admission glucose with in-hospital mortality among patients with diabetes mellitus (DM) hospitalized with COVID-19.
Adults hospitalized between 3/5/20 and 12/1/20 in a Connecticut health care system were included if they had prior DM diagnosis, an in-hospital A1c, and a positive RT-PCR nasopharyngeal swab for SARS-CoV-2. A1c was stratified into <7%, 7-<9%, and ≥9%. Both bivariate and multi-variable adjusted logistic regression analyses were performed to determine the association of A1c categories and admission glucose >200 mg/dL with mortality (in-hospital death or transition to hospice) and with intensive care unit (ICU) use. Models were adjusted for demographics and 8 relevant comorbidities.
Among 733 patients (median age 67 years [interquartile range, 56-77], 48.3% female, 43.11% White, 35.47% Black, 24.97% Hispanic, 1.64% Asian), 31.7% had A1c <7%, 40.5% 7-<9%, 27.8% ≥9%, and 38.1% admission glucose >200 mg/dL. During hospitalization, 111 (15.1%) patients died or transitioned to hospice and 230 (31.4%) required ICU care. In 2 multi-variable adjusted analyses, neither A1c category nor high admission glucose were significantly associated with mortality (A1c 7-<9%: OR 0.89, 95% CI 0.53-1.49; A1c >9% OR 1.3, CI 0.72-2.35 compared with A1c <7%; glucose >200 OR 1.34, CI 0.72-2.35) or ICU care (A1c 7-<9% OR 1.30, 95% CI 0.88-1.93; A1c ≥9% OR 1.35, CI 0.86-2.1 compared with A1c <7%; glucose >200 OR 1.26, CI 0.9-1.78). Age (per year OR 1.06, CI 1.04-1.08), male sex (OR 1.78, CI 1.14-2.81), obesity (OR 1.85, CI 1.16-2.96) and CKD (OR 1.90, CI 1.19-3.03) were significantly associated with mortality. Only female sex (OR 0.67, CI 0.48-0.93) was significantly associated with ICU care.
In our retrospective study of hospitalized patients with DM, neither A1c nor admission glucose were prognostic of COVID-19 mortality or ICU care. In those with DM, male sex, obesity and CKD predicted worse outcomes.
J. Brancale: None. A. M. Jastreboff: Consultant; Self; Boehringer Ingelheim Pharmaceuticals, Inc., Eli Lilly and Company, Novo Nordisk, Other Relationship; Spouse/Partner; Elucidata. K. J. Lipska: None. D. Athonvarangkul: None. M. Simonov: None. J. Alausa: None. L. K. Subair: None. A. M. Hajduk: None. Y. Kang: None. T. Arora: None. F. P. Wilson: Other Relationship; Self; Efference LLC, Medscape.
National Institutes of Health (1T32GM136651-01)