Abstract
Chest-computer tomography (CT) is a crucial factor in the clinical course and evaluation of patients with COVID-pneumonia. In the initial phase of the COVID-19 pandemic little information was known on the prognostic value of the initially taken thoracic CTs. The purpose of this study was to determine predictive values for clinical outcome based on CT classification of the pulmonary pathologies in patients with COVID-pneumonia. This single center study included 51 non-immunized patients during the first COVID-19 outbreak in Germany. The patients underwent a clinically indicated chest-CT. Using the radiological society of North America (RSNA)-report template, chest-CTs were classified into 4 categories (typical, atypical, indeterminate, and no changes). We analyzed the outcomes based on these imaging classifications and relevant comorbidities. Among the 51 patients of our study population 14 (27.5%) patients had a lethal outcome. Typical radiological COVID-19 pattern was found in 92.9% of the deceased patients and in 59.5% of the surviving patients (P = .022). The lethal group showed a significant higher proportion of diabetes mellitus (50% vs 10.8%; P = .003) and arterial hypertension (aHTN) (85.7% vs 54.1%; P = .037). Male sex, higher age and coronary heart disease (CHD) were also seen more often in the lethal group. In patients with clinically proven COVID-19 pneumonia, typical chest CT findings show a negative outcome. A classification system used in this study is helpful for classifying imaging features and is recommended as a standardized CT reporting tool. It could also help in triaging of the therapy of patients with COVID-19 pneumonia. Especially the comorbidities, diabetes and arterial hypertonia triggered a negative outcome in our study population.