Introduction
Coronavirus disease (COVID-19) has been spreading worldwide since December 2019 from Wuhan, Hubei Province.1 The virus called SARS-Cov-2 belongs to the Betacoronavirus genus, is composed of a positive single-stranded RNA, and has a 50-200 nm diameter.2,3 Bats are the natural reservoir for coronaviruses.4 On April 26th 2020, more than 2 355 853 cases (of which 947 693 in Europe) and 164 656 deaths worldwide were reported due to this pandemic.5On March 16th 2020 in France, 124 114 cases were confirmed with 22 614 deaths (i.e. 18,6 % of mortality).6 Fever, cough, fatigue, diarrhoea, dyspnoea and myalgia are the most common symptoms but the patients’ condition can deteriorate rapidly, requiring intensive medical care.7,8Real time reverse-transcription polymerase chain reaction (RT-PCR) is the standard reference to detect viral nucleic acid but false-negative results have been reported and have led to a 66-80 % sensitivity due to multiple factors, notably the quality of sampling.9Compared to RT-PCR, chest computed tomography (CT) is an easy-to-use and a faster method to diagnose and assess an early pulmonary COVID-19 infection.10 Chest CT can also help to monitor the evolution and diagnose complications of COVID-19. Indeed, some authors have demonstrated the diagnostic value of chest CT in COVID-19 pneumonia describing compatible radiological signs.11–13 Ai.et al. have evaluated chest CT diagnostic value in 1014 cases of suspected COVID-19 using RT-PCR as a reference, and determined a sensitivity of 97 % and negative predictive value of 83 %.10 Few authors have studied the prognostic value of chest CT. Yuan et al . have found in a small sample of 27 patients (of which 10 died) higher rates of consolidation and air bronchogram in the dead patients’ group.14 In a retrospective study, Zhao et al. found pleural effusions and architectural distorsions to be CT signs potentially correlated with severe deterioration in a small population of severe or deceased patients (14/101 patients).15
In this study, our main objective was to assess whether some chest CT signs were associated with pejorative evolution (defined as requiring intensive care unit hospitalization, invasive endotracheal ventilation or a fatal outcome), for patients with COVID-19. Our secondary objective was to evaluate the diagnostic value of chest CT versus RT-PCR.