ABSTRACT
COVID-19 is diagnosed by detecting SARS-CoV-2 by nasopharyngeal swab (NPS) using real-time quantitative reverse transcriptase polymerase chain reaction (qRT-PCR). Emerging evidences have shown the utility of saliva, although conflicting results have been reported regarding viral loads between NPS and saliva. We conducted a study to compare the viral loads in 42 patients with COVID-19. Both NPS and saliva specimens were simultaneously obtained at a median of 6 days (range, 1-12) after symptom onset. SARS-CoV-2 was detected in 34 (81%) using NPS (median Ct value [IQR]=27.4 [21.3, 35.6]) and 38 (90%) using saliva (median Ct value [IQR]= 28.9 [23.1, 33.6]). There was no significance difference between them (Wilcoxon signed rank test: P=0.79) and Kendalls W was 0.82, showing a high degree of agreement, indicating equivalent viral loads in NPS and saliva. After symptom onset, the Ct values of both NPS and saliva continued to increase over time, with no substantial difference. Self-collected saliva has a detection sensitivity comparable to that of NPS and is a useful diagnostic tool with mitigating uncomfortable process and the risk of aerosol transmission to healthcare workers.
