| Abstract [eng] |
Hyperglycemia in COVID-19 patients leads to higher mortality, need for artificial lung ventilation and longer hospitalization (5,6). Knowledge on how to apply this in clinical practice and especially in the intensive care unit is lacking. Aim: to evaluate the prognostic value of the blood plasma glucose concentration test performed during the hospitalization in intensive care unit COVID-19 patients. Methods: retrospective data analysis of COVID-19 patients treated in the Intensive care units of Vilnius University Hospital "Santaros Klinikos" between 03/01/2020 - 04/01/2021. Glucose concentration on arrival at the hospital and on arrival at the intensive care unit was analysed, patients were divided into normoglycemic, moderate and severe hyperglycemic groups. Diabetic patients were analysed separately. Results: the sample consisted of 247 patients, 63.16% were men. The average age of the subjects was 61.21±13.27 years. 79 (32%) of them had previously diagnosed diabetes. In non-diabetic patients, there were statistically significant differences between glycemic groups in terms of mortality (p<0.001), duration of stay at the intensive care unit (p=0.032), ISARIC 4C (p=0.027), SAPS (p<0.01), SOFA (p<0.01) scales, PaO2/FiO2 ratio (p<0.01), need for artificial lung ventilation (p<0.001), and vasopressors (p=0.001). In patients with diabetes, such associations were observed only when assessing the duration of hospitalization in the intensive care unit (p=0.032), PaO2/FiO2 ratio (p=0.024) and the need for high-flow oxygen therapy (p=0.02). It was found that if the prognostic scales include the question "Is the patient's arrival glycemia higher than 10 mmol/l?", their accuracy increases by about 3%, recall by 9%. Conclusions: Glycemia on hospital and intensive care unit admission is reliably associated with outcomes in non-diabetic patients. Adding the question "Is the patient's glycemia higher than 10mmol/l?" to prognostic scales increases their accuracy. |
