Can Chest CT be the first-line diagnostic test in COVID-19? A cross-sectional study

Abstract

ABSTRACT Background/aim: Early in the COVID-19 pandemic, chest computerized tomography (CCT) was shown to be more sensitive than real-time reverse transcription polymerase chain Reaction (rRT-PCR) in detecting the disease, especially in regions with high disease prevalence. In this study, we aimed to determine the sensitivity and specificity of chest CT in the diagnosis of COVID-19. Materials and methods: Between 17 March, 2020 and 25 April, 2020, 2170 patients who were admitted to the hospital for any reason and underwent chest CT scanning were retrospectively analyzed. Patients with a pre-diagnosis of COVID-19 and a positive rRT-PCR test, two consecutive negative rRT-PCR tests, or with a first negative rRT-PCR test and a positive follow-up rRT-PCR test were included. The day of the rRT-PCR test for these patients was counted as day “0,” and 200 patients whose CCT was performed within +/- three days were included in the study. Results: Of the 200 patients included in the study, 118 were rRT-PCR-positive, and 82 were rRT-PCR-negative. Of the 118 patients with positive rRT-PCR results, 62/118 (52.5%) had positive CCT scans. With the rRT-PCR results as the reference, the sensitivity, specificity, and accuracy of CCT in indicating COVID-19 infection were 52.5% (95% CI 43–61%), 67% (95% CI 56–77%), and 58.5% (95% CI 51-65%), respectively. The negative predictive value of CCT in diagnosing COVID-19 was greater for women than for men (p = 0.01). Conclusions: In regions where the prevalence of COVID-19 is not high, the use of CCT in the diagnosis of the disease is not an alternative to the rRT-PCR test due to its low sensitivity.