Hyperglycemia as a predictor of mortality in adult patients with COVID-19 hospitalized in a public hospital in Peru

Abstract

Objective

To investigate the association between glycemic levels and mortality in patients without diabetes hospitalized for COVID-19 in Perú.

Methods

In a retrospective study conducted from April to June 2020 in Cayetano Heredia hospital, 529 patients were admitted with a positive SARS-CoV-2 laboratory result or a computed tomography chest scan with suggestive images of COVID-19 pneumonia. Patients were classified into three groups according to their first blood glucose measure. Group 1: glucose level lower than 100 mg/dL; Group 2: glucose level between 100 mg/dL and 126 mg/dL, and Group 3: glucose level over 126 mg/dL. Demographical characteristics, concomitant diseases, laboratory data, and treatment received during hospitalization were also described. Regression-adjusted models were used to analyze association of interest.

Results

The number of patients who met inclusion criteria was 289. Mortality occurred in 137 cases (47 %). Group 1, group 2 and group 3 had 29/77 (38 %), 58/120 (48 %), and 50/92 (54 %) mortality/severe cases, respectively. After all available confounding factors were adjusted, the group of patients with blood glucose levels over 126 mg/dL had a 73 % increased mortality hazard compared to the ones lower than 100 mg/dL (aHR: 1.73 [95%CI: 1.05–2.84]; p = 0.032).

Conclusion

Hyperglycemia (≥ 126 mg/dL) at baseline in patients without a previous history of diabetes is associated with mortality in admitted patients with COVID-19. Routine laboratory testing should never miss a baseline measure of glycemia as this allows for timely blood glucose management, thereby minimizing its negative impact on COVID-19 patients' outcomes.