Comparison of SARS-CoV-2 Detection from Saliva Sampling and Oropharyngeal Swab.

We examined the detection rate of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) using reverse transcription-PCR (RT-PCR) of side-by-side saliva and oropharyngeal swab (OPS) samples from 639 symptomatic and asymptomatic subjects, of which 47 subjects were found to be positive for SARS-CoV-2 in the OPS or saliva sample or both. It was found that the detection rate (93.6% for both OPS and saliva) as well as the sensitivity and specificity were comparable between the two sampling methods in this cohort. The sensitivity was 0.932 (95% confidence interval [CI], 0.818 to 0.977) and the specificity was 0.995 (95% CI, 0.985 to 0.998), both for saliva when OPS sampling was used as the reference and for OPS when saliva was used as the reference. Furthermore, the Cohen's kappa value was 0.926 (95% CI, 0.868 to 0.985), indicating strong agreement between the two sampling methods. In addition, the viral RNA stability in pure saliva and saliva mixed with preservation buffers was examined following storage at room temperature and at 4°C. It was found that pure saliva kept the viral RNA stable for 9 days at both temperatures and that the type of preservation buffer can either enhance or reduce the stability of the RNA. We conclude that self-administered saliva sampling is an attractive alternative to oropharyngeal swabbing for SARS-CoV-2 detection, and it might be useful in large-scale testing.