Инфекция и иммунитет (Feb 2021)
To the question regarding accuracy of COVID-2019 laboratory diagnostics
Abstract
Issues of accuracy (sensitivity and specificity) of PCR-analysis depending on features of performing preanalytical and analytical stages of laboratory diagnostics of COVID-19, as well as comparing PCR and lung computed tomography (CT) results have been analyzed in the study. Currently, a molecular genetic test based on polymerase chain reaction (PCR) is used for diagnostics of a new coronavirus infection (COVID-19). As of November 1, 2020, more than 750 million PCR tests have been conducted globally. Evidence accumulated by now allows to estimate diagnostic sensitivity and specificity of the SARS-CoV-2-specific PCR as high as 82—91% and 99—100%, respectively. In addition, increased PCR sensitivity may be noted upon performing repeated testing of the upper respiratory tract samples comprising 82.2% during the primary analysis that was further elevated up to 90.6% after two consecutive tests. A whole set of factors affect the PCR accuracy. In particular, false negative data might result from insufficient amount of virus-coupled genetic material in the sample, timeframe and mistakes made upon selecting biological samples. It was found that SARS-CoV-2 virus RNA was detected at the maximum diagnostic sensitivity in the upper respiratory tract 1—3 days before the onset of symptoms and sustained within the 5—6 days after disease onset. Such period of time is associated with the peak risk of SARS-CoV-2 transmission. On week 2 after disease onset, there have been noted elevated rate of detecting viral RNA in bronchopulmonary samples. The duration of detecting virus-related markers (including those found in the absence of viable virus forms) correlates with disease severity and may last for as long as 1—2 months. Another real-world issue related to PCR analysis is posed by an opportunity of obtaining false positive data, which solution requires high level organized laboratory research, especially in case large-scale studies. Upon that, it is worth noting that positive PCR results may account for detecting solely certain RNA-related fragments present in any sample, rather than a viable virus. It was noted that PCR in comparison to CT analysis demonstrates higher specificity, but does not allow to distinguish pneumonia caused by SARS-CoV-2 from pneumonia caused by other etiological agents (up to 25% false positive results). However, the diagnostic CT sensitivity was 97.2% that exceeds such parameter for PCR by 10—15%. It was concluded that the approach combining use of both PCR and CT by taking into account their own features as well as factors affecting the accuracy of the data obtained, allows us to correctly interpret the diagnostical results.
Keywords
