Assessment of interobserver reliability and predictive values of CT semiquantitative and severity scores in COVID lung disease.

Autor: Andrew, Dhilip, Shyam, Karthik, Cicilet, Soumya, Johny, Jovis
Zdroj: Egyptian Journal of Radiology & Nuclear Medicine; 6/18/2021, Vol. 52 Issue 1, p1-6, 6p
Abstrakt: Background: The coronavirus disease (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and first reported in December 2019 at Wuhan, China, has since then progressed into an ongoing global pandemic. The primary organ targeted by the virus is the pulmonary system, leading to interstitial pneumonia and subsequent oxygen dependency and morbidity. Computed tomography (CT) has been used by various centers as an imaging modality for the assessment of severity of lung involvement in individuals. Two popular systems of scoring lung involvement on CT are CT semiquantitative score (SQ) and CT severity score (CT-SS), both of which assess extent of pulmonary involvement by interstitial pneumonia and are partly based upon subjective evaluation. Our cross-sectional observational study aims to assess the interobserver reliability of these scores, as well as to assess the statistical correlation between the respective CT scores to severity of clinical outcome. Results: Both the SQ and CT-SS scores showed an excellent interobserver reliability (ICC 0.91 and 0.93, respectively, p < 0.05). The CT-SS was marginally more sensitive (99.2%) in detecting severe COVID pneumonia than SQ (86.5%). The positive predictive value of SQ (98.3%) is more than CT-SS (78%) for detecting severe disease. The similarity of interobserver reliability obtained for both scores reiterates the respective cutoff CT scores proposed by the above systems, as 18 for SQ and 19.5 for CT-SS. Conclusion: Both the SQ and CT-SS scores display excellent interobserver reliability. The CT-SS was more sensitive in detecting severe COVID pneumonia and may thus be preferred over the SQ as an initial radiological tool in predicting severity of infection. [ABSTRACT FROM AUTHOR]
Databáze: Supplemental Index