Chest computed tomography in COVID-19 pneumonia: a retrospective study of 155 patients at a university hospital in Rio de Janeiro, Brazil

OBJECTIVE: To define diagnostic criteria for coronavirus disease 2019 (COVID-19) on computed tomography (CT); to study the correlation between CT and polymerase chain reaction (PCR) testing for infection with severe acute respiratory syndrome coronavirus 2; and to determine whether the extent of parenchymal involvement and the need for mechanical ventilation are associated with the CT findings and clinical characteristics of patients with COVID-19. MATERIALS AND METHODS: This was a retrospective study of 155 patients with COVID-19 treated between March and May 2020. We attempted to determine whether the CT findings correlated with age and clinical variables, as well as whether the need for mechanical ventilation correlated with the extent of the pulmonary involvement. RESULTS: On average, the patients with COVID-19 were older than were those without (mean age, 54.8 years vs. 45.5 years; p = 0.031). The most common CT finding (seen in 88.6%) was ground-glass opacity, which correlated significantly with a diagnosis of COVID-19 (p = 0.0001). The CT findings that correlated most strongly with the need for mechanical ventilation were parenchymal bands (p = 0.013), bronchial ectasia (p = 0.046), and peribronchovascular consolidations (p = 0.012). The presence of one or more comorbidities correlated significantly with more extensive parenchymal involvement (p = 0.023). For the diagnosis of COVID-19, CT had a sensitivity of 84.3%, a specificity of 36.7%, and an accuracy of 73.5% (p = 0.012 vs. PCR). CONCLUSION: The patterns of CT findings are useful for the diagnosis of COVID-19 and the evaluation of disease severity criteria. The presence of any comorbidity is associated with greater severity of COVID-19.