Utility of Available Methods for Diagnosing SARS-CoV-2 in Clinical Samples

The laboratory diagnosis of SARS-CoV-2 should be done to confirm coronavirus disease 2019 (COVID-19) in suspected patients. Although several diagnostic methods have been developed in this regard, their accuracy for clinical application is not very clear yet. To compare the diagnostic value of laboratory tests for the detection of COVID-19 infection, this study provides an upcoming review of the newly developed detection methods. Sensitivity, specificity, detection limit, and turn-around-time of these methods are compared and challenges for their application in clinical settings are reviewed. PubMed and Google Scholar web sites were used for the systematic search until April 9, 2020 to identify the published studies based on the following keywords: "Detection", "Coronavirus 2019", "SARS-CoV-2", and "Sensitivity". Out of 526 results, a total of 54 articles, including 46 studies on detection methods, were considered eligible for the review. The results showed that most of the proposed tests focused on molecular methods, while immunological and point-of-care tests were investigated in 13 studies. There were also a few commercial automated methods for the qualitative detection of SARS-CoV-2 in clinical samples, most of which are not examined in the current review, as no data about their sensitivity and specificity were presented. Although the assessment of publication biases showed that 64% sensitivity and nearly 100% specificity for RT-PCR are close to reality, most of the related reports for serological methods are not valid and further studies are needed to confirm their utility in clinical settings. Moreover, the RT-PCR test alone cannot act as a gold standard because of bias in measurements. Therefore, antibody tests and other proposed methods could be used as supplementary diagnostic tests to improve RT-PCR accuracy. Although clinical findings are invaluable, in many cases, they can provide more valuable supportive data than serological tests.