High level of glycated hemoglobin (HbA<sub>1c</sub>) in patients with COVID-19 is a marker of the severity of the infection but not an indicator of previous diabetes mellitus

Abstract

BACKGROUND : Less than a year has passed since the start of the new coronavirus infection COVID-19 pandemic caused by SARS-CoV-2. First published research results demonstrate a frequent increase in glycemia in patients without previously diagnosed carbohydrate metabolism disorders. A possible relationship between the carbohydrate metabolism state and the course of COVID-19 is considered. AIM : To identify the incidence of known and newly diagnosed diabetes mellitus (DM) in hospitalized patients with COVID-19 and evaluate the relationship between glycated hemoglobin (HbA 1c ), inflammation markers and infectious disease severity. METHODS : A single-center, cross-sectional, retrospective study included 155 patients with confirmed COVID-19 and bi- lateral polysegmental viral pneumonia hospitalized in the Endocrinology Research Centre, Russian Federation. Diagnosis of carbohydrate metabolism disorders was based on the HbA 1c level, blood glucose level at admission (BGA) and fasting plasma glucose (FPG). Patients were divided into 2 groups: without the previously diagnosed DM (n=129, 83.3%) and with known history of DM (n=26, 16,7%). Patients without previously diagnosed DM were divided into subgroups according to HbA 1c levels: ≤6,0% (Group A), >6,0% HbA 1c , <6.5% (Group B), ≥6,5% (Group C). Additionally, insulin, interleukin 6 (IL-6) and D-dimer levels were measured in all patients. Blood oxygen saturation (SpO 2 ) was measured by pulsoximetry, computerized tomography of lungs with calculation of lung parenchyma damage percentage. RESULTS : Type 2 DM was previously established in 16.7% of all included patients. Among patients without DM history, DM was revealed in 8 patients (5.2%), based on HbA 1c , FPG and BGA. In 66 patients (42.6%) we observed no changes in carbohydrate metabolism. In 55 patients, the interpretation of carbohydrate metabolism state was difficult due to inconsistency of HbA 1c level with FPG and BGA: in particular, the level of HbA 1c ≥ 6.5% (which corresponds to the diagnostic threshold of the DM) was detected in 19 patients (12,2%) with normal FPG and BGA. No true stress hyperglycemia on admission and according to FPG was reported. Despite the presence of confirmed DM, HbA 1c levels positively correlated with inflammatory markers (erythro- cyte sedimentation rate, C-reactive protein, IL-6) and SpO 2 . Patients with a HbA 1c ≥6.5% without DM history had the most severe course of the disease: longest duration of hospitalization, largest damage of the pulmonary tissue, and high lethality. CONCLUSION : The incidence of DM among patients hospitalized with COVID-19 was 21.9% (16.7% had previous diagnosis of DM, 5.2% — newly diagnosed), which is 1.5 times higher than in general population in a comparable age category. It is difficult to assess the carbohydrate metabolism disorders in patients in acute infectious period. An increased level of HbA 1c (≥6.5%), first detected in the acute period of infection, in combination with normal FPG and BGA parameters, cannot be a diagnostic criterion and requires dynamic monitoring. HbA 1c level can be considered as a predictive factor of COVID-19 severity, independent of DM.