Sniffing Out Infection: Smell Test Speedier Than Urinalysis

Kate Johnson

April 21, 2014

STOCKHOLM — An "electronic nose" can sniff the difference between healthy urine specimens and those infected with Escherichia coli, according to a pilot study.

"The electronic nose can identify not just the organism but the particular strain of organism. It has a specificity of almost 100%, which is very encouraging," said Emmanuel Johnson, MD, from the Royal Liverpool and Broadgreen University Hospital NHS Trust in the United Kingdom.

"We started with E coli because 80% of urinary tract infections in the community and 50% in hospitals are from E coli, but the potential is limitless," Dr. Johnson told Medscape Medical News.

The results of the pilot study were presented here at the European Association of Urology (EAU) 29th Annual Congress.

The odor of urine is produced by volatile organic compounds that can be detected with gas chromatography and mass spectrometry, explained Dr. Johnson.

Bacteria and other pathogens produce "signature" volatile organic compounds that can be detected the same way, he said.

The study looked at midstream urine samples from 52 healthy individuals with negative urine microscopy (23 women and 29 men with a median age of 56 years) and from 36 patients with E coli–positive urine (31 women and 5 men with a median age of 68 years).

Using carboxen/polydimethylsiloxane solid-phase microextraction fiber, the "headspace" above the urine samples was analyzed with gas chromatography/mass spectrometry for volatile organic compound content.

'We obtained over 200 peaks, each representing a unique volatile. From those, we found 10 to be discriminating specifically for E coli (P < .001)," Dr. Johnson reported.

The method correctly classified volatile organic compounds in 90.9% of cases (88.8% on cross-validation), with a sensitivity of 80.5%, a specificity of 98%, a positive predictive value of 96.7%, and a negative predictive value of 87.9%.

"The EAU guidelines stipulate that, ideally, we should have a positive urine culture before prescribing antibiotics, but this takes time and skilled personnel," said Dr. Johnson. "We are working on transferring our method to a palm-held bedside device. That way we could select the correct treatment in about 20 minutes."

Dr. Johnson and his colleagues have previously investigated the role of volatile organic compounds in urine, stool, breath, and blood samples when diagnosing gastrointestinal and liver diseases (J Gastrointestin Liver Dis. 2009;18:337-343) and bladder cancer (PLoS One. 2013;8:e69602).

"We do need a point-of-care test that will identify pathogens," said session chair Robert Pickard, MD, professor of urology at Newcastle University in Newcastle-upon-Tyne, United Kingdom. However, "more important, we need some form of prognostication, perhaps by measuring virulence factors, to guide treatment."

The significance of the bacterial presence has to be determined, he explained. "Is the person going to clear the bacteria without any symptoms? Are they going to develop a mild self-limiting infection such as cystitis, or are they going to develop a severe infection that could potentially be life-threatening?"

"Otherwise, it's not going to change anything," he told Medscape Medical News. "It might even make things worse because you might give antibiotics to people who aren't destined to develop anything that needs it," Dr. Johnson said.

Dr. Johnson and Dr. Pickard have disclosed no relevant financial relationships.

European Association of Urology (EAU) 29th Annual Congress: Abstract 676. Presented April 13, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.