In the early days of the pandemic, public health officials scrambled to get a read on the new viral outbreak. The CDC covered all its bases, recommending specimens be sent to them in three ways for each person: from the back of their throat (oral-pharyngeal), deep into their nasal passage (nasopharyngeal), and from sputum.
Not a month later, the CDC announced a change -- only the nasopharyngeal test was required. Later, shallower nasal swabs known as mid-turbinate or anterior nare swabs became more common, and the CDC reversed course on its preference for the nasopharyngeal test.
A handful of variants later, scientists are asking once again: Is nasal swabbing really the best way to detect Omicron quickly and accurately?
What's Known So Far
Two preprint studies, not yet peer reviewed and done with small sample sizes, made a preliminary case for testing with saliva by swabbing the mouth or back of the throat.
A preprint from scientists with the COVID-19 Sports and Society Working Group compared nasal rapid antigen tests to saliva PCR tests. Among 30 people with active Omicron infections tested daily with a nasal swab and saliva rapid antigen test, the nasal rapid antigen test did not pick up positive infections until around a day after a saliva PCR test did.
Notably, viral load in the saliva peaked in a subset of five patients in the first 1-2 days -- before the nasal rapid antigen tests even showed a positive test.
Although it has already been established that PCR tests detect COVID-19 at a lower threshold than a rapid antigen test, this preprint suggests that nasal rapid antigen tests may not detect Omicron at its most infectious point.
"We observed people being infectious and confirmed that they were transmitting to other people during that high early saliva viral load" period, before a nasal swab detected the virus, said study author Blythe Adamson, PhD, MPH.
She said that while researchers did not do a direct comparison in this study, and can't recommend off-label use of a rapid antigen test (i.e., with saliva), manufacturers are likely to soon begin comparing rapid antigen tests via nasal or saliva swabs.
Another recent preprint from researchers in South Africa also suggested, based on 67 Omicron and Delta cases, that positive saliva tests were more reliable than positive nasal mid-turbinate swabs for Omicron. This is the opposite of Delta, where nasal swab results were more accurate than saliva swabs. Omicron had relatively more viral shedding in the mouth than in the nose compared with Delta.
For the saliva test, patients coughed three to five times, swabbed the inside of their cheeks, above and below the tongue, and on the gums and hard palate. The standard for comparison was a composite wherein infection was considered present if either the saliva or mid-turbinate swab came up positive.
But this is preliminary data, cautioned Benjamin tenOever, PhD, a microbiologist at NYU Grossman School of Medicine.
And it doesn't show a dramatic difference, he said. Looking over the figure in the study, he noted that "for Omicron, there is the same amount in saliva as it is for the [nasal] swab... I don't really know how you take that result and justify needing to ever do saliva or a mid-turbinate. It's really saying that you could do either."
Anecdotal and Expert Support
Around Christmas time, a flurry of tweets showed a seemingly strange phenomenon: people got a negative rapid test result when they swabbed their nose but got a positive when they swabbed both their throat and nose. One user wrote, "Today (after POSITIVE PCR) tried nose only (per leaflet). NEGATIVE! Same test done same time with nose + throat POSITIVE. How many people out there think they're clear?"
These may only be anecdotes, but leading voices in science think there's more to it. Eric Feigl-Ding, ScD, an epidemiologist with the Federation of American Scientists, tweeted on January 1, "Saliva samples (hence throat swabs) are much better for detecting Omicron coronavirus. We need clear public health messaging to adapt to this new reality."
Michael Mina, MD, PhD, formerly an epidemiologist at Harvard and now the chief science officer at eMed, also tweeted that "there is a chance the virus isn't yet growing in the nose when you first test. Virus may start further down. Throat swab + nasal may improve chances a swab picks up virus."
Mina also pointed out that the rapid antigen test, or rapid lateral flow test, can be used in both the nose and throat in the U.K., where that practice has been "fairly standard" throughout the pandemic. The U.K. Health Security Agency posted an instructional video on how to do both swabs using government-issued test kits in 2020. A Jamaica government hospital appears to swab both too.
Feigl-Ding said that there's no harm in adding a throat swab to a rapid antigen test in the privacy of your own home. "This is not, do B instead of A. This is, do A and B. I'm not saying 'don't swab the nose.' Swab the mouth and the nose. It's a precautionary thing. It's not taking a drug."
Difference in Tests
But to understand why tests from different parts of the body might yield different results, it's important to understand what kinds of specimens these tests gather.
"I'm not sure if we're totally set on which types of cells the virus infects versus where is the virus in high enough amounts for us to detect it?" said Christina Wojewoda, MD, of the University of Vermont and chair of the College of American Pathologists Microbiology Committee.
Throat and mouth, Wojewoda explains, are lined with similar epithelial cells. But those are slightly different from the lining of the nose and nasopharynx. Throat swabs and saliva tests (taken either by spitting in a cup or swabbing the inside of the mouth) gather the virus more from squamous epithelial cells, whereas nasal swabs grab it more from respiratory epithelial cells. Which type of cell a virus "prefers" -- what's called tissue tropism -- appears to be narrower for Omicron than previous variants.
Spokespeople for three leading at-home rapid antigen tests -- Abbott BinaxNOW, Quidel QuickVue, and BD Veritor -- confirmed that their tests don't have emergency use authorization from the FDA for use with a throat swab. And the companies must stick to the exact terms of their emergency use authorizations lest the FDA rescind them, Feigl-Ding said.
Why Would Omicron Be Different?
Omicron, a number of preprint studies in animal models have suggested, operates in a very different way from Delta: it infects the upper respiratory tract but may affect the lungs less. Daniel Rhoads, MD, of the Cleveland Clinic, a board-certified pathologist and vice chair of the College of American Pathologists Microbiology Committee, said that because of this, "it would make sense that maybe the highest load of virus might be in a slightly different anatomic location than previous variants," but that more paired studies are needed.
This lung versus upper respiratory distinction with Omicron, tenOever said, shouldn't be critical for testing samples taken from saliva versus the nose. "It doesn't make a lot of sense that a rapid antigen test in the nose versus the back of your throat would give you different results, because that's all the upper respiratory tract, so I don't really understand why that would be the case," he said. "And I find it doubtful that it is the case."
But Mina and Feigl-Ding theorized that throat tests may be picking up more virus in the throat than in the nose also because of Omicron's differences in replication. There may be more virus in bronchial tubes, Feigl-Ding said, than with previous variants. "The biological plausibility is that there's a greater concentration in your windpipe than ever before. It just replicates differently, and it clearly has a different potential tissue replication," he said.
When to Change Methods
Though none of the experts disputed the possibility of changing how tests are done, they diverged on the burden of proof needed to take action.
Because COVID-19 changes so fast, Feigl-Ding argued, decisions made slowly could mean potentially infecting many more people.
"One thing in this pandemic that we've learned is we've made too many mistakes by sitting on our butts," he said. "Some people are like, 'Well, we should follow the rules.' But the rules have always been many, many steps behind the science and the curve and the emerging evidence."
Rhoads, on the other hand, was willing to wait a little longer to get it right. "We need to just follow the data, and as people continue to study this, if we find empirically that something seems to work better than another then we should be willing to pivot or modify our approach," he said.
Wojewoda took an intermediate tack: "I would like to see it repeated at least once," she said of the early findings. "Another study site or a multi-center study to show that this is really applicable to different patient populations would be great." She added that data on asymptomatic patients would also be valuable.