Chest CT features in COVID-19 patients: A systematic review and meta-analysis

Objective: To derive the pooled estimate of chest computed tomography (CT) findings in coronavirus disease 2019 (COVID-19) patients. Methods: A comprehensive systematic search was conducted according to the PRISMA checklist from January 2020 to September 2020 in electronic databases including PubMed, Google Scholar, and Scopus based on search terms in title and texts. Original descriptive studies with epidemiological parameters of interest were included into the systematic review and meta-analysis. Results: Totally 54 articles comprised of 4 879 patients with a mean age of 49.05 years were eligible for this study. The pooled prevalence for abnormal CT images was 86.0%. Pooled prevalence for ground-glass opacity was 68.0%, 71.0% for bilateral abnormalities, 47.0% for mixed ground-glass opacity and consolidation and 29.0% for consolidation. In addition, 64.0% of lesions were peripheral, and 12.0% were central while 28.0% were both central and peripheral. Furthermore, 61.0% of lower lungs were involved, and 7.0% and 5.0% of the cases presented with pleural effusion and pericardial effusion, respectively. Besides, 11% of the cases showed lymphadenopathy, and 37% had air broncho gram sign. The pooled prevalence of other chest CT findings ranged from 8.0% to 65.0%. Conclusions: Chest CT can be used as predictive tools for the detection of COVID-19 disease along with clinical manifestations and the RT-PCR method.