Application and optimization of RT-PCR in diagnosis of SARS-CoV-2 infection

BackgroundCoronavirus Disease 2019 (COVID-19) caused by Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has become a global threat to public health. Aiming to construct an efficient screening pattern, we comprehensively evaluated the performances of RT-PCR and chest CT in diagnosing COVID-19. MethodsThe records including demographics, RT-PCR, and CT from 87 confirmed COVID-19 cases and 481 exclusion cases were collected. The diagnostic accuracy of the pharyngeal swab RT-PCR, CT, combination with the second pharyngeal swab RT-PCR or with CT were evaluated individually. Besides, all the stool RT-PCR results were plotted by time to explore the value of stool RT-PCR. FindingsCombination of RT-PCR and CT has the higher sensitivity (91.9%,79/86) than RT-PCR alone (78.2%,68/87) or CT alone (66.7%, 54 of 81) or combination of two RT-PCR tests (86.2%,75/87). There was good agreement between RT-PCR and CT (kappa-value, 0.430). In 34 COVID-19 cases with inconsistent results, 94.1% (n=32) are mild infection, 62.5% of which (20/32) showed positive RT-PCR. 46.7% (35/75) COVID-19 patients had at least one positive stool during the course. Two cases had positive stool earlier than the pharyngeal swabs. Importantly, one patient had consecutive positive stool but negative pharyngeal swabs. InterpretationCombination of RT-PCR and CT with the highest sensitivity is an optimal pattern to screen COVID-19. RT-PCR is superior to CT in diagnosing mild infections. Stool RT-PCR should be considered as an item for improving discovery rate and hospital discharge. This study shed light for optimizing scheme of screening and monitoring of SARS-CoV-2 infection. FundingThis work was supported by the National Natural Science Foundation of China (No. 81502104), National Program on Key Basic Research Project (No. 2018YFC0910600),the Nature Science Foundation of Guangdong Province, China (Grant No: 2017A030313771 and 2020A151501001) and the Young Teachers Nurturing Program of Sun Yat-Sen University (Grant No:17ykpy62)