Hyperglycemia in the diabetic range, but not previous diagnosis of diabetes mellitus, is an independent indicator of poor outcome in patients hospitalized for severe COVID-19

Aims Diabetes mellitus (DM) and hyperglycemia are associated with poor outcome(s) in COVID-19 hospitalized patients, but their independent impact on prognosis remains unclear. We aimed to assess the impact of DM and hyperglycemia on COVID-19 outcomes. Methods Clinical data/records from COVID-19 patients admitted to the Parma University-Hospital (February 23rd to March 31st, 2020) were retrieved and analysed (NCT04550403). Fasting plasma glucose (FPG), inflammatory markers and the main biochemical variables were collected at admission. Patients underwent chest high-resolution CT and arterial blood gas analysis to determine the PaO2/FiO(2) ratio (P/F ratio). The primary outcome was a composite of intensive care unit admission and/or death. Results Among 756 subjects, 143 (19%) had DM. These patients were older with higher comorbidity rates. The primary outcome occurred in 61.5% DM patients versus 43.4% without DM (p < 0.001). In multivariable analysis (accuracy UC = 0.93), older age, cardiovascular and kidney diseases, FPG >= 126 mg/dl, C-reactive protein, and P/F ratio, but not previous DM, were independent risk indicators. Conclusions DM indicated poor COVID-19 outcomes, but not when adjusted for other clinical variables/comorbities, suggesting that its impact was mostly driven by concomitant factors. The independent role of fasting hyperglycemia points to the need for further research on its contribution to COVID-19.