Key points are not available for this paper at this time.
Aim: To investigate the interobserver and intraobserver agreement and performance of the Radiological Society of North America Expert agreement declaration in assessing chest computed tomography (CT) findings related to new COVID-19 pneumonia. Materials and Methods: In this retrospective study, conducted from March 15 to April 1, 2020, 113 patients with suspected COVID-19 infection were enrolled. All patients underwent investigation using real-time reverse transcription polymerase chain reaction (RT-PCR) and chest CT scans. Chest CT features were categorized by three radiologists following the North American Consensus Statement. Characteristic and indeterminate features were considered as Group A, while atypical and negative features were considered as Group B. The interobserver and intraobserver agreement of the imaging features were evaluated, along with the sensitivity and specificity of the consensus statement. Results: The study population comprised 113 consecutive patients. Out of the 113 patients, 61 tested positive for RT-PCR. Group A (categories 3 and 4) consisted of 89 patients, while Group B (categories 1 and 2) included 24 patients. The ICC score for intraobserver and interobserver agreement was 0.996 (95% CI) and 0.971 (95% CI), respectively. Using RT-PCR as a reference standard, the sensitivity, specificity, negative predictive value, and positive predictive value of CT findings (group A, B) for COVID-19 pneumonia were 82%, 25%, 56.1%, and 54.1%, respectively. Conclusion: The expert agreement declaration on reporting new COVID-19 pneumonia tomography findings is a well-designed, reliable, and reproducible standardized CT reporting language. It demonstrates excellent intra- and interobserver agreement.
Özdemir et al. (Tue,) studied this question.