Aims: The aim was to report the prevalence of diabetes status in patients hospitalized with COVID-19 and assess the association between the glucometabolic status at admission and 90-day mortality. 

Methods: Consecutive patients hospitalized with COVID-19 were included in the study. All participants included had an HbA_1c measurement 60 days prior to or within 7 days after admission. We studied the association between diabetes status, the glycemic gap (difference between admission and habitual status), admission plasma-glucose, and mortality using Cox proportional hazards regression. 

Results: Of 674 patients included, 114 (17%) had normal glucose level, 287 (43%) had pre-diabetes, 74 (11%) had new-onset, and 199 (30%) had diagnosed diabetes. No association between diabetes status, plasma-glucose at admission, and mortality was found. Compared to the 2nd quartile (reference) of glycemic-gap, those with the highest glycemic gap had increased mortality (3rd (HR 2.38 [1.29–4.38], p = 0.005) and 4th quartile (HR 2.48 [1.37–4.52], p = 0.002). 

Conclusion: Abnormal glucose metabolism was highly prevalent among patients hospitalized with COVID-19. Diabetes status per se or admission plasma-glucose was not associated with a poorer outcome. However, a high glycemic gap was associated with increased risk of mortality, suggesting that, irrespective of diabetes status, glycemic stress serves as an important prognostic marker for mortality.