| Chemical signature: Breath contains over 100 organic compounds
Since at least the time of Hippocrates, physicians have recognised that the smell of their patients’ breath can provide clues to what is ailing them. The rotting apple odour of acetone can signal diabetes. A sulfurous smell can mean cirrhosis of the liver. Other odours can hint at lung infection, kidney disease or simple dental decay.
The potential for using breath as a way to peer into the body increased vastly with the rise of modern chemistry. In the 18th century, the French chemist Antoine Lavoisier proved that the body produces carbon dioxide, emitted with each exhalation of breath. And in the early 1970s, Linus Pauling demonstrated that breath is a complex gas, containing well over 100 different organic compounds.
Still, only recently have researchers begun to mine the elaborate chemical signature of breath to explore whether shifting patterns might be used diagnostically ' to detect conditions like lung cancer or tuberculosis. The most advanced test so far is one that can suggest whether heart transplant patients are rejecting their new organs.
The technique, called the Heartsbreath test, was developed recently by Dr Michael Phillips, a professor of internal medicine at the New York Medical College in Valhalla, New York, and the founder of Menssana Research, a for-profit company. The test received approval from the Food and Drug Administration last year.
It bears a superficial resemblance to the breathalyzer tests used by the police to test drivers for drunkenness, but it is “a billion times more sensitive,” Phillips said.
In the test, a patient breathes through the mouth into a long, stainless steel tube for two minutes. A breath sample is collected and later analysed chemically, using techniques called gas chromatography and mass spectroscopy. A sample is also collected from room air, for purposes of comparison.
If a patient’s breath test is negative, he may be able to skip his next one or possibly two biopsies, although no protocol has yet been developed. But even a small reduction in the number of heart biopsies ' which are generally performed more than a dozen times in the first year after a transplant ' would hold appeal for patients.
“You’re not tickling the heart. You’re not making it quiver,” said Mark Blackowski, 47, who underwent a heart transplant at Newark Beth Israel Medical Center in 2003, and tried the breath test on two occasions. “You walk in, they clamp your nose, you breathe into a tube for two minutes, and you’re done.”
A heart biopsy can also cost thousands of dollars, while a Heartsbreath test will probably cost only a few hundred. (Medicare and other insurers are still working out details of reimbursements ' one reason the test is not yet widely available.)
Dr Mark Zucker, director of the Heart Failure Treatment and Transplant Center at Newark Beth Israel, noted that the Heartsbreath test was currently less sensitive than a biopsy, meaning that it may not pick up on organ rejection until the condition is slightly more advanced.
However, since doctors generally monitor but do not treat the mildest forms of rejection, this is not an issue clinically, he said. (Zucker participated in clinical trials for the Heartsbreath test, financed by the National Institutes of Health, but he said he had no financial interest in the device.)
The Heartsbreath test is also less specific than a biopsy, meaning that it is more likely to yield a positive result, even if the patient’s immune system is not reacting against the transplanted organ ' one reason the FDA approved the test as an adjunct to a biopsy but not a replacement for it.
Phillips and others are also exploring breath tests for the early diagnosis of lung cancer ' another arena in which a noninvasive and relatively inexpensive screening technique would be welcome. Lung cancer can now be detected using CT scans, which expose patients to small doses of radiation. Positive results are followed up with biopsies, in which tissue is removed from the lung for analysis.
Phillips has received a grant to study whether breath testing may have forensic applications: to detect, for example, whether a person has handled any of a wide variety of explosives.
Clinical trials of breath tests for tuberculosis are being financed by the NIH. Dr Antonino Catanzaro, a professor of medicine at the University of California, San Diego, who is involved in this research, called traditional tests for TB slow and sometimes unreliable. Breath testing, he said, “takes a very different approach and may overcome these problems.” Such a test may be of particular value for children and for patients also infected with HIV, he added. Breath testing seems to be aligned with a general trend in medicine toward more noninvasive testing, said Zucker of Newark.
Terence Risby, a professor at Johns Hopkins Bloomberg School of Public Health, agreed that breath analysis held great allure and that it was likely to prove popular with patients, especially children. Breath samples, unlike blood samples, he noted, do not need to be refrigerated and so may someday be valuable in developing countries. (NYTNS)