A simple swab may have a role to play in testing for tuberculosis. (Photo: Thomas G/Pixabay)News & Features
14th April 2025 | Elri Voigt
While likely years away from widespread use, a test for TB that relies on a simple tongue swab instead of a person having to cough up sputum is showing promise. Spotlight takes stock of how the test works and asks experts what its future role might be.
Most tuberculosis (TB) tests are conducted on sputum samples. The difficulty with this is that many people, particularly children and people living with HIV, struggle to cough up the thick mucus, also sometimes called phlegm, from their lungs.
While sputum-based tests are likely to remain the standard for TB diagnosis for some time, researchers are hard at work looking for more convenient alternatives, especially for children and people living with HIV. One leading candidate is something that has been around for years, a simple oral swab, also known as a tongue swab.
How it works
Using a swab to test for a disease should be fresh in the memory of anyone who had a COVID-19 test. A tongue swab for TB works in a similar manner, but is far less invasive than a COVID-19 test because it only involves swabbing the tongue, rather than the deep nasal passages or the back of the throat.
“A tongue swab test for TB works by collecting bacteria from the surface of the tongue using a soft, flocked swab. The swab is gently rubbed over the back of the tongue for about 30 seconds, ensuring as much bacteria as possible is gathered without causing discomfort,” explains Anura David, a senior medical scientist at the Wits Diagnostic Innovation Hub.
“The swab is then sent to a laboratory, where it is tested for the presence of Mycobacterium tuberculosis (MTB), the bacteria that causes TB.”
David says the specimen is tested using the same diagnostic methods used for sputum. “However, we are hoping that tongue swabs can eventually be tested in the field as close as possible to the patient so that the patient can be diagnosed at the same time without the need to return on a different day for their results (as is currently the case),” she says.
With current technology, TB is easier to detect in sputum samples than those taken from the tongue. That is because sputum comes directly from the lungs, which is typically where the TB bacterium is most active. That said, tongue swabs are still being researched, and it is not yet clear exactly how accurate they are.
Since tongue swabs are not approved as a TB diagnostic tool, David explains that anyone who gets a positive result from a tongue swab must be tested again with a sputum sample.
“We’re making significant progress on using tongue swabs for TB diagnosis, but there are still key questions to address before they can be widely implemented,” she says.
How TB is normally diagnosed
Dr Christopher Ealand, a senior researcher and molecular biologist at the School of Pathology in the Faculty of Health Sciences at Wits University, describes sputum as “a viscus, thick, messy, ugly type of sample” that someone coughs up into a collection vial. This sample can be tested for TB using one of three methods.
He explains it can be run through a nucleic acid amplification test, the one used in South Africa is called the Gene Xpert Ultra, which can identify whether TB DNA is present, and whether the bacteria is resistant to rifampicin – one of the first-line drugs for TB. Alternatively, a sputum sample can be put into a Mycobacteria Growth Indicator Tube (MGIT), and if the sample grows over the period of about 50 days, it means TB is present. TB can also be diagnosed by looking at the sample under a microscope after it has been treated with a special stain. The bacteria will light up in bright green if present.
One TB diagnostic that does not depend on sputum is chest X-rays. X-rays can show the lung damage caused by TB disease, but it is usually not possible to tell from an X-ray if there is ongoing disease or whether the damage is from disease that has already resolved. For this reason, X-rays suggestive of TB have to be followed up by sputum-based tests.
The problem with sputum-based tests, as stressed by Ealand, is that not everyone is able to cough the stuff up.
For a child who is unable to cough up sputum, one alternative is a gastric wash. This involves inserting a tube down a child’s throat to collect fluids from the stomach after a night of fasting. This works because kids with TB can have the bacteria in their stomachs if they’ve swallowed sputum that has come up from the lungs.
Another option for both children and adults who struggle to produce sputum is induction. This involves inhaling a nebulised saline solution which typically causes a person to cough.
“There is a clear need to improve TB diagnostics,” says David. The bacterium that causes TB was first isolated in 1882, she says, and for nearly 130 years, only four diagnostic tests were available. Since 2010, David says significant advancements in TB diagnostics have been made, “particularly with molecular tests that enable faster and more accessible TB diagnosis”.
LEADING KILLER | While TB is curable, diagnosing and treating it can be tricky. In this week’s newsletter, we focus on the centuries-old TB bacteria: https://t.co/SoeZxVDIpJ
Join our mailing list: https://t.co/ya957ntqMY pic.twitter.com/PkMcKCxCO8
— Spotlight (@SpotlightNSP) March 28, 2025
Limitations of tongue swabs
Using a tongue swab may provide a potential, and much more convenient alternative to a sputum sample or even a chest x-ray, but it’s not without limitations.
For instance, David says, tongue swabs perform better when testing people with high TB bacterial loads. But overall, the swabs are less sensitive than sputum when it comes to diagnosing TB.
“An ideal test has both high sensitivity and high specificity, meaning it correctly identifies sick individuals while avoiding misdiagnosis of healthy ones,” she says. (Sensitivity refers to a test’s ability to correctly indicate the presence of a bug, while specificity refers to a test’s ability to correctly indicate its absence.)
Studies on tongue swabs show varying results on how accurately the test can identify people with TB, Dr Ryan Dinkele, a research officer and epidemiologist in the School of Public Health at the University of Cape Town, tells Spotlight.
A 2024 systematic review published in The Lancet Global Health journal looked at 20 studies – 15 involving adults and 5 involving children. It found that the sensitivity of oral swabs ranged from 36% to 91% in adults and from 5% to 42% in children. The specificity of the tests did not vary as much, with most studies reporting specificity greater than 90%.
This variation, says Dinkele could be due to several factors, like who collected the sample, the type of swab used or how good the person in the laboratory is at extracting DNA. Another factor is that a tongue swab typically contains fewer TB bacteria than a sputum sample, and having too few bacteria in the sample can make DNA extraction difficult.
Commenting in The Lancet on the systematic review, the authors note: “Oral swabs hold promise but the reported sensitivity is below the WHO [World Health Organization] cutoff for diagnostic tools. Hopefully sensitivity can be improved through optimising specimen collection, including self-collection, and testing methods.”
Related Posts
David says further research is needed to optimise sample collection, improve diagnostic accuracy, and ensure consistency across different populations and settings.
Ealand argues that since there’s no standard method for using tongue swabs to test for TB, the first step should be to create a standardised protocol. This includes what buffer liquid is used, when samples are collected, the type of swab used, and time between collection and processing. This will make it easier to compare results between different laboratories and studies.
In addition, Ealand says, a way to potentially improve the sensitivity of the test is to pool results. Instead of collecting just one swab sample, three or four could be collected and those can be tested together. This will likely increase the bacterial load and make detection easier.
Where might this test fit in?
While it is unlikely that tongue swab TB tests will be implemented at a population level anytime soon, and it seems improbable that it will ever replace sputum-based testing, the experts agree that it has potential in certain settings.
Dinkele is sceptical of the usefulness of tongue swabs as a TB diagnostic in its current state, but he says that it may have a place as a screening tool. Or as a supplementary test for an individual where no other diagnostic method is working.
Ealand agrees, saying that it could be a useful tool for screening households for TB. He explains that everyone in a household could provide a tongue swab sample, which could then be tested all together as a pooled sample. If the result is negative, it can be safely assumed no TB is present in that household. But if TB is detected then sputum samples can be collected for confirmatory tests.
He adds that it could also be useful in detecting TB in people who have subclinical (asymptomatic) TB – where they have TB disease and may be infectious, but aren’t showing any symptoms.
TBAC is calling for increased testing, better care and stronger accountability to #EndTB. In this in-depth Spotlight article, we write that South Africa’s progress against TB depends on diagnosing more people more quickly: https://t.co/Q7UicKtBGZ https://t.co/UO79gtB1Fs
— Spotlight (@SpotlightNSP) February 4, 2025
Part of the case for such wider use as a screening test, is that the greater convenience of a tongue swab could enable wider and more frequent testing. The arguments here get quite technical, but one idea is that the wider and more frequent testing enabled by tongue swabs could to some extent make up for their lower sensitivity.
Along such lines, one avenue being pursued is the potential for TB self-testing using tongue swabs.
One recently published study, of which David was an author, looked at the accuracy of self-collected tongue swab tests in detecting TB. In the study, conducted at the Hillbrow Community Health Centre in Johannesburg, 399 people being investigated for TB were asked to take their own tongue swabs. They were also asked for sputum and urine samples. The self-collected tongue swabs were tested in the laboratory using the Gene Xpert MTB/RIF Ultra test. This was compared to people’s sputum specimens, which were tested using liquid culture, another name for the MIGIT test described before.
Overall, the sputum in liquid culture test had higher sensitivity than the tongue swabs, 95% versus 78%. Meaning the tongue swabs weren’t as good at detecting TB in people who had the bacteria as the sputum test was – though 78% sensitivity is nevertheless high compared to other tongue swab studies.
The researchers also found the performance of the self-collected tongue swabs were comparable to tongue swabs that had been collected by healthcare workers in other studies. David says this implies that self-collected tongue swabs are just as reliable as healthcare worker collected tongue swabs.
In a survey done afterwards, all the study participants said they were happy with the swab, and 74% said they preferred the use of a tongue swab over a urine or sputum sample for TB testing. Interestingly, 52% of the participants said they would prefer a healthcare worker perform the swabbing, while 48% preferred to do it themselves.
