Initial Results of the Use of a Standardized Diagnostic Criteria for Chest Computed Tomography Findings in Coronavirus Disease 2019

J Comput Assist Tomogr. 2020 Jul 7. doi: 10.1097/RCT.0000000000001054. Online ahead of print.

Abstract

Objective: The objective of this study was to assess the initial results of chest computed tomography (CT) standardized diagnostic criteria proposed by the Radiological Society of North America in coronavirus disease 2019 (COVID-19) compared with reverse transcription-polymerase chain reaction (RT-PCR).

Methods: Seventy-one patients who underwent RT-PCR test for COVID-19 and chest CT within an interval of 4 days or less were included. Seventy-five CTs were reviewed and classified as typical, indeterminate, or atypical appearance or negative for pneumonia by 2 radiologists. With RT-PCR as reference standard, the performance of the CT diagnostic criteria in diagnosing COVID-19 was assessed.

Results: The prevalence of positive RT-PCR was 45.1%. It was obtained a sensitivity of 83% (95% confidence interval [CI], 78%-89%), a specificity of 97% (95% CI, 92%-99%), an accuracy of 91% (95% CI, 85%-96%), a positive predictive value of 97% (95% CI, 91%-99%), and a negative predictive value of 86% (95% CI, 80%-92%). The diagnostic performance was excellent, considering the area under the curve of 0.92 (95% CI, 0.84-0.99).

Conclusions: Chest CT standardized diagnostic criteria had high specificity and positive predictive value for the diagnosis of COVID-19 when presenting a typical appearance.