Value of CT findings in predicting transformation of clinical types of COVID-19

Objective To investigate the value of CT findings in predicting transformation of clinical types of COVID-19. Methods From January 24 to February 6, 2020, the clinical and chest CT data of patients with common covid-19 were analyzed retrospectively. A total of 64 patients were enrolled, 32 males and 32 females, ranging in age from 18 to 76 years, with an average age of (45 ± 15) years. Based on the fact whether patients’ conditions had deteriorated into severe type, all the cases were divided into common type group (51 cases) and deteriorated type group (13cases). Differences of CT findings in two groups of patients were analyzed, and visual semi-quantitative scores were introduced to evaluate the pneumonia. Results Compared with the common type group, the heavy type group was more likely to involve the left upper lobe, the right middle lobe and the part far away from the pleura. The difference between the two groups was statistically significant (χ2= 5.897, P = 0.027; χ2= 8.549, P= 0.005; χ2= 10.169,P= 0.002). There were 2 (1,5) medians of the involved lobes in the general type group and 5 (4,5) medians of the involved lobes in the heavy type group. The difference between the two groups was statistically significant (Z = -3.303, P = 0.001). Taking the involved lobes (n=4) as the boundary value, the sensitivity and specificity of the diagnosis of the general type to the heavy type patients were the highest, 76.9% and 74.5% respectively, and the area under the ROC curve was 0.787. Pneumonia score 10 (4,16) of the severe group was higher than that of the common group 4 (1,13), the difference was statistically significant (Z=-4.040, P<0.001); the sensitivity and specificity of the general severe group were the highest, 69.2% and 86.3% respectively, and the area under ROC curve was 0.863. Conclusions CT imaging has a profound value in early prediction of deterioration in clinical type. It can help evaluate the severity of pneumonia in early stage. Range of lesions might be an important indicator for prognosis of common type COVID-19. Key words: COVID-19; Tomography, X-ray computed