Coronavirus Infection COVID-19 Clinical Trial
Official title:
Analysis of Laboratory Markers for Severe COVID-19
The course of coronavirus infection was often severe and required hospitalization of patients in the intensive care unit. The new SARS-Cov-2 has been poor studied, so relatively reliable markers are needed to effectively monitor patients and predict complications and outcome. Taking into account the known mechanisms of pathogenesis, the biochemical markers as ferritin, procalcitonin, C-reactive protein and D-dimer were chosen for this purpose. Patients were divided according to the degree of pulmonary infiltration. We hypothesized that the markers would correlate with dynamics, complications, and outcomes.
In the presented study, an analysis of the medical records of 193 patients hospitalized in severe condition to the intensive care unit with a confirmed diagnosis of Coronavirus infection COVID-19 was carried out. Taking into account the volume of pulmonary infiltration according to computer tomography (CT) of the chest organs, patients were divided into 4 groups in accordance with the approved classification: CT 1(up to 25% of lung tissue was infiltrated) - 27 patients, CT 2 (25-50%) - 60 patients, CT 3 (50-75%) - 67 patients, CT 4 (75% and more) - 39 patients. The following biochemical parameters were selected and used to monitor dynamics: procalcitonin (PCT), C-reactive protein (CRP), D-dimer (DD), ferritin (FRT). The duration of observation was 15 days. In order to determine correlations between quantitative and qualitative data at different stages of treatment, a correlation analysis was carried out (Spearman's test was used). ROC analysis was performed to evaluate selected laboratory markers as predictors of outcome. Next, the odds ratio (OR) was assessed taking into account the obtained Youden's J index and the Associated criterion for each of the selected markers in relation to the patient's outcome. Preliminary contingency tables were compiled in relation to laboratory parameters and outcomes (2x2 tables). Data were processed using statistical software jamovi (Computer Software ,Version 2.3.26), MedCalc (MedCalc Software Ltd, Ostend, Belgium), Microsoft Office Excel, 2016. ;