This memo is based on rapid searches in PubMed, LitCov and supplementary searches for pre-prints. One researcher assessed the relevance of each reference and summarized the findings. Another researcher supplied information and summarized the findings, read and provided feedback on the review. In the current situation, there is an urgent need for identifying the most important evidence quickly. Hence, we opted for this rapid approach despite an inherent risk of overlooking key evidence or making misguided judgements.
The aim of this rapid report is to investigate if there are studies supporting the use of salivary samples for detection of SARS-CoV-2, and if there are studies that can be used to calculate the diagnostic accuracy of salivary testing compared to nasopharyngeal or oropharyngeal swab samples.
We selected studies focusing on testing for SARS-Cov-2 in saliva. We excluded studies that had not used SARS-CoV-2 testing with samples from nasopharyngeal and/or oropharyngeal swabs as reference standard. This rapid review does not include a formal quality assessment of included papers, nor does it include a grading of the certainty of evidence. The results should therefore be interpreted with caution.
We identified 32 new papers from the database search and by manual searching of reference lists. From these, we included 8 studies comparing SARS-Cov-2 nucleic acid detection from salivary samples compared to nasopharyngeal or oropharyngeal swab samples. Two of the included studies had not been through peer review. The sample sizes range from 4-82 patients. Two of the studies could be used to assess diagnostic accuracy. The results indicate that the correlation between diagnostic accuracy of symptomatic patients tested with reverse transcription polymerase chain reaction (rRT-PCR) using self- collected saliva samples, compared to health care worker administrated nasopharyngeal or oropharyngeal swabs is good, corresponding to a sensitivity of 97-100 % and a specificity of 89-100%. The results should be considered with caution as further studies may change the estimates. Additionally, six of the studies are indicative of various types of salivary sampling as appropriate for detection of SARS-Cov-2 nucleic acid. These findings are highly relevant in the face of shortages of both swabs and personal protective equipment for health care workers. Self-collection of saliva is comfortable for patients as well as being easy, cheap, and non-invasive with minimal equipment required.
In conclusion, the results indicate that the correlation between diagnostic accuracy of symptomatic patients tested with rRT-PCR using self- collected saliva samples compared nasopharyngeal or oropharyngeal swabs is good. The results should be considered with caution as further studies may change the estimates.