Association of Computer Tomography Features Of COVID-19 with Outcomes, Clinical and Laboratory Parameters

Abstract

Aim: to assess the results of chest computer tomography (CT) of patients with novel coronavirus infection in correspondence with their outcomes, clinical and laboratory data. Methods: retrospective analysis of 962 chest CT scans, outcomes, clinical and laboratory data of all 354 COVID-19 patients hospitalized from April to June 2020. Results: Sensitivity and specificity of CT with polimerase chain reaction (PCR) as a reference were: 98.0 % and 5.7 % respectively; for PCR with CT as a reference: 54.6 % and 70.7 % respectively. Patients with positive and negative PCR tests had no significant differences in mean CT score and CO-RADS score. Cumulative survival was better in patients with lower CT score (significant only for maximal, not baseline scores). CT score changed during hospitalization in survived patients clinically insignificant (from 2 (1-2) to 2 (1-2), p=0.001), and increased in dead (from 2 (1,5-3) to 4 (4-4), p <0.001). Lower CT score and better survival was in females, patient younger than 59 years, with NEWS score <3, without atrial fibrillation. Diabetes mellitus and obesity was associated with higher CT score, but not with survival. Chronic obstructive pulmonary disease, coronary heart disease and chronic heart failure was associated with lower survival, but not CT score. Conclusion: chest CT significantly increases diagnostic accuracy and assessment of the prognosis in COVID-19 patients.