Ces radiol. 2020, 74(3):171-179 | DOI: 10.55095/CesRadiol2020/027

Chest CT: A Valuable Tool in Discrimination of COVID-19 Pneumonia, Community Acquired Pneumonia and the Other Pathologies in Slow Epidemic PhaseOriginal article

Jiří Ferda¹, Martin Vítovec¹, Jan Baxa¹, Dalibor Sedláček², David Havel³, Ondřej Topolčan⁴, Radek Kučera⁴, Jan Beneš⁵, Martin Matějovič⁶, Hynek Mírka¹

¹ Department of the Imaging, University Hospital Pilsen, Czech Republic

² Department of Infectious Diseases, University Hospital Pilsen, Czech Republic

³ Department of Respiratory Diseases, University Hospital Pilsen, Czech Republic

⁴ Division of the Immunodiagnostics, University Hospital Pilsen, Czech Republic

⁵ Department of the Anesthesiology, Resuscitation and Intensive Care, University Hospital Pilsen, Czech Republic

⁶ 1^st Department of Internal Medicine, University Hospital Pilsen, Czech Republic

Aim: SARS-CoV-2 infection manifests like COVID-19 pneumonia, computed tomography (CT) has been reported as its valuable imaging. Frequent arguments against are discussed: the problem of the possible further spread of the infection to medical staff and/or other patients, and also insufficient SARS-CoV-2 detectability based solely on CT. The aim of the study was to examine the usefulness of CT for personalized, precision and predictive diagnostics of COVID-19-suspected.

Method: Chest CT was performed due to suspected COVID-19 pneumonia in 310 patients during slow epidemic in Czech Republic, during March and April 2020. All patients underwent sampling for RT-PCR to confirm SARS-CoV-2 viral presence. When all RT-PCR tests had been completed and after one week of follow-up, the assessment of the sensitivity, specificity, positive and negative predictive value, and accuracy were calculated, a role of the discrimination using leukocytosis in suspected chest CT findings was evaluated respectively. One-week-follow-up was performed to assess the re-admission of the patient, and/or the COVID-19 transmission within patients or staff.

Results: The overall sensitivity of 93.75%, specificity of 94.60%, positive predictive value of 66.67%, negative predictive value of 99.25%, and accuracy 94.52 were reached; adding leukocytosis in suspected cases, specificity of 99.64%, positive predictive value of 96.77%, negative predictive value of 99.28%, and accuracy 99.03% were improved respectively. Only one re-admission with COVID-19 in formerly CT-negative was recorded. There were no noted SARS-CoV-2 transmission neither patient-to-patient, nor patient-to-staff.

Conclusion: The high negative predictive value of chest CT significantly contributed to the early identification of a large group patients with low risk for COVID-19, thus allowing them to be safely admitted outside the reserved COVID-19 units, the additional strength lies in the ability to identify an alternative non-COVID-19 diagnosis to explain the patients' clinical symptoms. The CT based approach is enabling personalized, precise and predictive approach to COVID-19 pneumonia detection.

Keywords: chest CT, SARS-CoV-2 infection, COVID-19 pneumonia

Grants and funding:

Supported by the projects Conceptual development of the Research institution 00669806 - FN Plzeň and Charles University Project Progress q39.

Accepted: September 15, 2020; Published: September 1, 2020  Show citation