Differentiation of COVID-19 from seasonal influenza: A multicenter comparative study

As coronavirus disease 2019 (COVID-19) crashed into the influenza season, clinical characteristics of both infectious diseases were compared to make a difference. We reported 211 COVID-19 patients and 115 influenza patients as two separate cohorts at different locations. Demographic data, medical history, laboratory findings, and radiological characters were summarized and compared between two cohorts, as well as between patients at the intensive care unit (ICU) andnon-ICU within the COVID-19 cohort. For all 326 patients, the median age was 57.0 (interquartile range: 45.0-69.0) and 48.2% was male, while 43.9% had comorbidities that included hypertension, diabetes, bronchitis, and heart diseases. Patients had cough (75.5%), fever (69.3%), expectoration (41.1%), dyspnea (19.3%), chest pain (18.7%), and fatigue (16.0%), etc. Both viral infections caused substantial blood abnormality, whereas the COVID-19 cohort showed a lower frequency of leukocytosis, neutrophilia, or lymphocytopenia, but a higher chance of creatine kinase elevation. A total of 7.7% of all patients possessed no abnormal sign in chest computed tomography (CT) scans. For both infections, pulmonary lesions in radiological findings did not show any difference in their location or distribution. Nevertheless, compared to the influenza cohort, the COVID-19 cohort presented more diversity in CT features, where certain specific CT patterns showed significantly more frequency, including consolidation, crazy paving pattern, rounded opacities, air bronchogram, tree-in-bud sign, interlobular septal thickening, and bronchiolar wall thickening. Differentiable clinical manifestations and CT patterns may help diagnose COVID-19 from influenza and gain a better understanding of both contagious respiratory illnesses.