Smells like tuberculosis? Sensor study suggests breath samples might reveal disease

A device that detects a pattern of chemicals in the breath was both sensitive and specific for TB in a small pilot study, according to Amandip Sahota, MD, of the University Hospitals of Leicester in England.

In the study, the device was able to detect both pulmonary and extrapulmonary forms of the disease, Sahota reported at the Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC).

The idea of using breath samples to detect disease is not new, Sahota noted, but his study employed a technology -- Field Asymmetric Ion Mobility Spectrometry (FAIMS) -- -- that has the potential to be cheaper, faster, and more widely available than earlier methods.

But he cautioned that the study, involving just 25 patients and 19 healthy controls, needs to be validated in a larger cohort.

In the U.S., TB incidence continues to fall, according to the CDC. The National TB Surveillance System recorded 9,412 cases in 201, a 2.2% decline from 2013. But worldwide, the disease still exacts a stunning toll -- about 9 million new cases and 1.5 million deaths a year.

Despite the advent of new technologies, Sahota said, most TB diagnosis worldwide is still done using culture methods, which are time-consuming and require significant expertise. A simple rapid point-of-care test would speed treatment, he said.

That's "very important," commented Shruthi Ravimohan, PhD, of the University of Pennsylvania in Philadelphia, who was not involved in the study.

"The longer patients wait for their results," she told MedPage Today, "the more likely is it that they will be lost to follow-up or the test results will be lost in the meantime."

As well, she noted, delayed treatment is likely to have other adverse consequences, including advancing illness.

"The simpler the test, the better," she said. But the challenge with such technologies is how well the machines work over time, especially in poor rural areas where the ability to maintain equipment might be limited.

The FAIMS technology, used in some military and security applications, works by ionizing the volatile organic compounds in a sample and passing the ions between electrodes.

A detector generates a pattern of chemicals and the question for Sahota and colleagues was whether the patterns seen in the breath of TB patients would be significantly different from those seen in healthy people.

Sahota and colleagues enrolled 25 TB patients and 19 controls known not to have the disease. The participants breathed into a special 3-L bag and the samples were processed using a commercially available FAIMS device.

A multivariate analysis showed that the patterns generated by the patient were significantly different from those of the controls, Sahota said.

Indeed, the area under the receiver operating characteristic curve -- a measure of how accurate a test is -- was 0.96, he reported. A result of 1.0 would mean it was perfectly accurate.

Put another way, the sensitivity of the test was 93% and the specificity was 94%.

Importantly, Sahota said, the 25 patients had varying forms of TB, with only 11 having pulmonary disease. Also, six had lymph node disease, four had spinal involvement or psoas abscess, two had joint disease, and one patient each had testicular and skin TB.

The method "is not limited to the lung," he said.

One disadvantage of the method is that those being tested have to fast for at least 2 hours before the test to avoid volatile compounds from food that might compromise the result, Sahota said.

The next step, he said, is to repeat the analysis with more participants and to test the approach in TB patients versus other patients, including those with latent TB or HIV.

Sahota and co-authors disclosed no relevant relationships with industry.

Ravimohan disclosed no relevant relationships with industry.

Primary Source
Interscience Conference on Antimicrobial Agents and Chemotherapy
Source Reference: Sahota A, et al "Breath analysis to diagnose pulmonary and extra-pulmonar tuberculosis using ion mobility spectrometry" ICAAC 2015.

Source: MedPage Today