Objective: To analyze the relationship between the baseline glycemic control and antihyperglycemic therapy with the severity of the SARS-COV-2 infection (COVID-19) in ethnic minority patients with diabetes.

Methods: We conducted a retrospective chart review of 179 consecutive patients at an academic hospital in Chicago, Illinois. A sample of patients who tested positive for COVID-19 between March 21st, 2020 and January 12th, 2021were divided in 3 groups according to the clinical severity of COVID: 1) Mild (outpatient treatment only); 2) Moderate (hospital admission); 3) Severe (ventilatory support and/or death).

Results: Mean hemoglobin A1c (HbA1c) before COVID-19 was 8.43 (±2.0) %. Mean age was 59.6 (±11.9) years, 57.54% were male, 51.9% Hispanic and 41.9% African Americans. Sixty-four patients (35.8%) had mild disease (outpatient treatment, group 1), 93 (52%) moderate (hospital admission without the need for ventilatory support, group 2), and 22 (12.3%) severe disease (intubation and/or death, group 3). Baseline glycemic control was not associated with the clinical outcomes of COVID. Among various therapeutic agents, only metformin use was significantly associated with mild disease versus moderate and severe forms of COVID, controlling for GFR. Fifty-four patients (84.4%) in group 1 were taking metformin, compared with 63 (35%) and 11 (50%) in the moderate and severe groups, respectively.

Conclusion: Here we report that use of metformin was associated with a lower chance of developing moderate (hospitalization) or severe forms (ventilatory support and/or death) of COVID-19 in ethnic minority patients with diabetes. Prospective investigation is required to make precise estimates of the impact of metformin in COVID-19 outcomes.


J. Berardo: None. D. Mihailescu: None. B. Tahsin: None. L. Fogelfeld: Research Support; Self; Bayer Healthcare Pharmaceuticals Inc., Lilly Diabetes, Sanofi.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.