Taking a breath for better health

A DCU research team is developing a system that could diagnose a disease simply by patients blowing into a device

What does your breath say about you? It might tell the world (or at least anyone at close quarters) that you didn’t brush your teeth this morning, or that you ate garlic last night. But there might also be other, subtle and important clues in there about your state of health – if you know how to look for them. These volatile compounds could be signals of disease, and there’s a push towards developing convenient tests and arrays of sensors that would allow patients to simply blow into a device to get their analysis.

It’s an emerging field, and there is plenty about breath analysis that is still at the research or early testing stage, but initial and relatively small studies in humans are suggesting that the approach could pick up lung disease and problems in other parts of the body, such as cancer or diabetes.

"There's growing evidence that you can detect markers in the breath that are systemic in their origins and they can be linked into disease, or the early appearance of certain disease states," says Dr Conor Burke, associate director (commercialisation) at the Science Foundation Ireland-funded Biomedical Diagnostics Institute (BDI) in Dublin City University.

Of course, breath is not a new place to go looking for information about a person’s lung function or health: people have long been measuring oxygen and carbon dioxide in the breaths of athletes and people with conditions such as asthma and chronic obstructive pulmonary disease. Those are tried and trusted methods, and the measurements link back in a well-defined way to lung function, respiration and metabolism, according to Dr Burke.

Picking up the more subtle markers in breath that could help to diagnose disease or monitor a patient can be a little trickier, he notes, in part because those chemical signals may be present in only vanishingly small amounts.

“A lot of the work in the area of breath analysis for disease diagnosis or monitoring is focused on gases such as ammonia as well as volatile organic compounds or VOCs in breath, such as ethane, isoprene, isopropanol, ethanol, pentanes and formaldehyde,” he says. “These can be linked into different disease states, but they are typically present in very low concentrations, single parts per billion, or sometimes less.”

You also have to catch the right part of the breath so that you reduce the risk of picking up confounding signals from the mouth or gut, adds Dr Burke. “If the VOCs are oral in their origin then you might just be detecting bad breath.”

So the trick is to collect the sample from the last gasps of a really big puff. “If you are detecting the end-stage of the tidal volume of the breath, then the gases are more likely to be systemic,” he says. “You are emptying the alveoli or air sacs in the lungs, and capturing the compounds that are transported from the blood into the alveoli. This is what you want to detect.”

The BDI has been focusing efforts on detecting ammonia in breath in order to assess a person’s kidney and liver function. “If there is a problem with either of those organs you would expect to see elevated levels of ammonia in a person’s breath,” says Dr Burke.

The BDI's approach, led by Prof Tony Killard (who is now at the University of the West of England) uses a polymer called polyaniline, which conveniently changes the way it conducts electricity when it meets ammonia.

With funding from Enterprise Ireland, Killard and colleagues figured out a way to print polyaniline out on a sensor – making it cheap to produce – which then goes into a device. When a patient blows into the device, the sensor gives a measurement of ammonia levels.

While it sounds simple, that last sentence has required quite a bit of engineering. The BDI’s prototype stacks up well against other methods, but its analysis can be done at a fraction of the cost, according to Dr Burke.

And when they put the system to the test with people on kidney dialysis, they found that the breath-ammonia measurements correlated well with the blood test that is generally used to monitor the patient’s progress.

“We showed that on a person-to-person basis there was a very strong correlation between the breath and blood tests,” says Dr Burke. “And because a breath test is less invasive than a blood test, we would see that this has the potential to be a personal monitoring device, so we are doing more kidney studies and we are also moving into studies of people with liver dysfunction.”