Confirming Accuracy of Saliva Testing for COVID-19

Saliva nucleic acid amplification tests (NAAT) for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are accurate, with sensitivity and specificity similar to that of the nasopharyngeal NAAT, according to a new study.

The systematic review and meta-analysis, published in JAMA Internal Medicine, included 8 peer-reviewed studies and 8 preprints involving a total of 5922 samples.

“Biggest takeaway is that whether you perform a test using a nasopharyngeal swab (the nose swab) or a saliva sample, you can trust the results equally,” corresponding author Guillaume Butler-Laporte, MD, of McGill University Health Center told Contagion®. “However, saliva tests are simpler, more comfortable, and don't required a trained nurse. Hence, they represent an attractive alternative to our current best tests.”

Pooled sensitivity for the saliva NAAT was 83.2% (95% credible interval [CrI], 74.7%-91.4%) compared with a sensitivity of 84.8% (95% CrI, 76.8%-92.4%) for the nasopharyngeal swab NAAT, according to the study. Pooled specificity was 99.2% (95% CrI, 98.2%-99.8%) for the saliva NAAT and 98.9% (95% CrI, 97.4%-99.8%) for the nasopharyngeal swab NAAT.

“Our biggest surprise was actually our main result,” Butler-Laporte said. “Nasopharyngeal tests have proven to be very accurate in many other viral infections in the past, and were really the standard test for upper respiratory tract infections when the pandemic started. So initially we thought we'd see that saliva samples would be worse, but we were pleasantly surprised.”

The studies included in the analysis were from the United States, Asia, Europe and Australia. Ambulatory patients were included in 15 of the studies, and 9 studies included only outpatient participants with no symptoms or mild symptoms. In 5 of the studies, only those who showed COVID-19-like symptoms were enrolled, and 1 study enrolled hospitalized patients.

Of the 5922 samples, 941 yielded positive results and 4981 had negative results.

“There's now a well-researched alternative to the usual test,” Butler-Laporte said. “This is of most value to laboratory medicine and public health professionals who can now provide these tests to clinicians knowing that they're still offering what's best for the patient.”

The study authors urged testing centers to consider saliva testing as a first choice, particularly for mass screenings.

“Next step is to make it more widely available,” Butler-Laporte said. “Another step would be to use the fact that the test is simpler to perform and perhaps deploy it as part of larger testing campaign in harder-to-reach populations like children, college students, homeless populations, etc. But this would likely require dedicated research to make sure it's done properly. Hence, our study is really just opening the door for this possibility.”

Limitations of the study included variations in such things as study population and timing of testing and underrepresentation of testing of hospitalized and critically ill patients. Assays also may have varied slightly among studies.

The Infectious Diseases Society of America’s (IDSA) recently updated guidelines for COVID-19 testing to include saliva tests.

“There’s just been an explosion of papers looking at the performance of these different sample types, and especially saliva,” IDSA board member Kimberly Hanson, MD, MHS, an associate professor of internal medicine at the University of Utah School of Medicine, said at the time. “And what we learned was that saliva [testing] actually looks pretty good.”

New testing strategies are in the works, including a rapid test developed at the University of California, Santa Cruz, and one by Decision Diagnostics that uses a test strip, among others. In July, the US Food and Drug Administration granted Emergency Use Authorization to Color for COVID-19 testing without clinical monitoring, and in November, the first at-home COVID-19 self-test received EUA for the Lucira COVID-19 All-In-One Test Kit.