Clincal and chest computed tomography characteristics from 58 Patients with COVID-19 pneumonia and correlations with disease length and severity

Background: This study aimed to review computed tomography (CT) findings in COVID-19 patients, and establish correlations between CT findings in patients with a short vs. long disease course, and in those with mild vs. severe disease. Materials and Methods: From February 2020 to March 2020, 58 patients with SARS-CoV-2 infections were retrospectively included. Clinical, laboratory, and CT findings were compared between patients with a short vs. long disease course, and in subgroups with mild vs. severe disease. Correlation analyses were performed to determine factors correlated to greater disease severity in patients with short/long disease courses, respectively. Results: Fifty-eight patients were included; 29 in the short disease course and 29 in the long disease course group. CT findings were similar between patients with a short and a long disease course (all, P > 0.05). Among the short disease course group, severe disease patients had significantly higher rates of right upper lobe involvement, 5 lobes affected, pericardial effusion, pleural involvement and bilateral pleural thickening, grid shadow, higher-density vascular shadows, crazy-paving appearance, lung consolidation, an air bronchogram sign, and fibrous foci than those with mild disease. Among the long disease course group, severe disease patients had significantly higher rates of right upper lobe and middle lobe involvement, 5 lobes affected, pleural effusion and thickening, grid shadow, higher-density vascular shadows, crazy-paving appearance, lung consolidation, an air bronchogram sign, and atelectasis. Conclusions: CT imaging findings may help to predict disease severity in COVID-19.