Novel breath test shows promise for diagnosing and monitoring bacterial infections

A new, non-invasive breath test has emerged as a potential breakthrough for rapidly diagnosing bacterial infections and tracking treatment effectiveness.¹ Developed by researchers from the University of California, San Francisco (UCSF) and St. Jude Children’s Research Hospital, the test leverages pathogen-specific metabolic tracers and a laser-based detection platform to identify infections in real time.

Lead researcher Dr. Marina Lopez-Alvarez presented findings at ESCMID Global 2025, demonstrating the test’s feasibility in preclinical models and highlighting its potential for future clinical applications.

READ MORE: Bridging the gap: mouse breath research to advance infectious disease diagnostics

READ MORE: Smart mask monitors breath for signs of disease and infection

The breath test is based on the principle that pathogenic bacteria, but not mammalian cells, metabolise certain 13C-enriched compounds, producing [13C]CO2, which can be detected in exhaled breath. The researchers evaluated five bacterial species—Staphylococcus aureus, Escherichia coli, Salmonella typhimurium, Enterococcus faecalis, and Enterobacter cloacae—and found that they metabolised 13C-maltose and 13C-mannitol, as well as 13C-sorbitol, 13C-xylose, and 13C-arabinose in the case of E. coli.

Answers in the breath

Among the 13C-enriched metabolites tested,13C-maltose and 13C-mannitol produced no background [13C]CO2 in healthy mice, confirming their specificity for bacterial metabolism. In contrast, 13C-glucose and 13C-sorbitol were also metabolised by mammalian cells, making them less suitable for infection detection.

To validate the approach, S. aureus-infected mice (myositis, osteomyelitis, and pneumonia models) received intravenous 13C-maltose, while E. coli-infected mice were administered 13C-mannitol. Their exhaled breath was then analysed for [13C]CO2. Healthy mice did not produce [13C]CO2 after receiving 13C-maltose, 13C-mannitol, 13C-arabinose, 13C-xylose, or 13C-maltotriose, confirming that these substrates are metabolised exclusively by bacteria in the animals tested.

Bacteria-specific metabolites

Emphasising the importance of these findings, Dr. Lopez-Alvarez said, “Current imaging tools reflect the host-immune response rather than the causative pathogens themselves. This can lead to an incorrect or delayed diagnosis when the patient has, for example, a sterile inflammatory disease. By using bacteria-specific metabolites, such as mannitol, that are not metabolised by mammalian cells but are metabolised by bacteria, we can rapidly detect infections with greater accuracy.”

In an additional experiment, E. coli-infected mice treated with the antibiotic ceftriaxone for 24 hours exhibited a considerable decrease in [13C]CO2 levels, correlating with a reduced bacterial burden. This result suggests that the breath test could be useful not only for diagnosing infections but also for monitoring the effectiveness of antibiotic treatments in real-time.

While this study did not directly assess the sensitivity of the laser-based detection method, prior research suggests that it offers advantages in cost and portability compared to traditional isotope ratio mass spectrometry (IRMS)-based methods.

Cost and size

Professor David M. Wilson, principal investigator on this study, stated that the major advantages of the integrated cavity output spectroscopy (ICOS) system platform are the cost and size of the instrument. “For approximately $100,000 you could have an ICOS instrument in the emergency room or another acute care setting, and the system is smaller than a piece of carry-on luggage,” he explained.

Before testing in individuals with suspected bacterial infections, researchers must first confirm that healthy humans do not produce [13C]CO2 from these metabolites. This is the crucial next step to ensure the test’s specificity for bacterial infections in clinical settings.

While breath testing is already used clinically for Helicobacter pylori infections,2 Dr. Lopez-Alvarez and Dr. Wilson aim to expand this diagnostic approach to a broader range of bacterial infections. “This study represents an important step towards non-invasive, rapid infection detection, with potential applications in emergency medicine, intensive care, and antimicrobial stewardship programs,” Dr. Lopez-Alvarez concluded.

About the study author

Dr. Marina Lopez-Alvarez is a postdoctoral researcher at the University of California San Francisco (UCSF) specialising in medical microbiology and biomedical imaging. Her work focuses on developing imaging tools for diagnosing infectious diseases. She conducted the research in the laboratory of Dr. David M. Wilson at UCSF.

References

1. Lopez-Alvarez, M., Lee, S., Wadhwa, A., et al. (2025). [13C]CO2 breath testing for detecting and monitoring bacterial infection. Oral presentation. ESCMID Global 2025, Vienna, Austria.
2. Peeters, M. (1998). Urea breath test: a diagnostic tool in the management of Helicobacter pylori-related gastrointestinal diseases. Acta gastro-enterologica Belgica, 61(3), pp. 332–335.