The first breath-based COVID-19 test has been issued an emergency use authorization by the U.S. Food and Drug Administration (FDA). The device takes three minutes to return results and has been found to be over 90 percent accurate in detecting positive COVID-19 cases.
The new breath-testing device is from a company called InspectIR and uses gas chromatography mass-spectrometry (GC-MS) to analyze breath samples. Like prior COVID breathalyzer prototypes, the device doesn’t identify the presence of specific viral particles. Instead it’s designed to pick up patterns of volatile organic compounds (VOCs) that have been found to correspond with SARS-CoV-2 infection. In this case the company says a pattern of five specific VOCs can effectively identify positive cases of COVID-19.
A statement from the FDA indicates the emergency use authorization was influenced by the results of a large study composed of almost 2,500 subjects. The research found the test picked up 91.2 percent of positive COVID-19 cases, a metric known as test sensitivity. The specificity of the test was even higher, correctly identifying 99.3 percent of negative cases.
“Today’s authorization is yet another example of the rapid innovation occurring with diagnostic tests for COVID-19,” said Jeff Shuren, from the FDA’s Center for Devices and Radiological Health. “The FDA continues to support the development of novel COVID-19 tests with the goal of advancing technologies that can help address the current pandemic and better position the U.S. for the next public health emergency.”
While the FDA’s authorization of InspectIR’s breathalyzer is inarguably a milestone – marking the first approval for a breath-based COVID-19 testing technology in the United States – this particular device is unlikely to be a game-changer in the real world.
The FDA authorization requires the device to be used by trained operators under the supervision of health care providers. Plus, although InspectIR have cleverly engineered the bulky GC-MS equipment into a small case, the device is far from a portable object. The current iteration of the breathalyzer comes in a large luggage-style case.
No price has yet been revealed for the device but the FDA’s authorization statement does indicate very limited production over the coming months. The company reports a production capacity of around 100 units a week. Each unit is said to be able to process around 160 samples a day.
So realistically, the InspectIR breathalyzer is likely to only be sporadically seen at hospitals or mobile COVID-19 testing sites over the next few months. There are, however, plenty of other COVID breath-testing systems in the works looking to make this kind of technology more accessible.
In February, a team of researchers in Singapore demonstrated a small portable COVID-19 breathalyzer that delivers highly accurate results in minutes. That device also relies on detecting unique patterns of VOCs in breath, but instead of using bulky GC-MS systems to analyze samples it looks to a technology called Raman spectroscopy. This kind of sensor technology can be engineered into small, relatively affordable portable devices.
Looking beyond VOC breath patterns to catch COVID-19 cases, a team of researchers in Australia recently presented a prototype device designed to specifically detect SARS-CoV-2 particles in breath samples. Clinical trials are currently underway to validate its accuracy and if successful it could be available later this year.
Source: FDA
That's the intuitive explanation, the exact numbers can be found using Bayes' Theorem, or an online specificity-sensitivity calculator. Unless the true COVID prevalence in the tested population is quite high, several percent, then there will be more than a 10% chance that a positive test result is a false positive. So positive breath test results will need to be confirmed by PCR tests in many situations, particularly for asymptomatic subjects in populations without ongoing serious outbreaks. (The PCR tests have their own sensitivity and specificity rates which depend largely on the number of PCR cycles the test uses for amplification - a higher cycle number makes the test more sensitive but less specific.)