Volatile Compound Analysis of Breath from Patients with Clostridium difficile Infection for Biomarker Discovery and Biological Insight

Volatile Compound Analysis of Breath from Patients with Clostridium difficile Infection for Biomarker Discovery and Biological Insight

1 Teny John, 2 Leen Hasan, 3 Kirk Pappan, 3 Owen Birch, 2 David Grove, 2 Gary Procop, 2 Raed Dweik, 3 Billy Boyle, 3 Max Allsworth, 2 Nabin Shrestha

1 The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, United States; 2 Cleveland Clinic, 9500 Euclid Avenue, Cleveland, Ohio, United States; 3 Owlstone Medical, 183 Cambridge Science Park, Milton Road, Cambridge CB4 0GJ, UK

Poster PDF

Abstract:

As an opportunistic pathogen, Clostridium difficile remains a leading source of hospital acquired infection and is a troublesome cause of co-morbidity and death. Following earlier research by the investigators using a method that targeted a specific group of breath-borne volatile compounds, this project used an untargeted biomarker discovery approach to identify on breath volatile organic compounds (VOCs) associated with C. difficile infection (CDI).

Within the current study, insights of potential biological and epidemiological importance to CDI were suggested by the VOC compounds that differed on breath between CDI patients compared to a control population. These range from biomarkers of plastics exposure to derivatives of tobacco smoke, vehicle exhaust, and others. We propose steps to extend or confirm these observations, but also note the potential bias of VOC literature, compared to other non-VOC biomarker studies, toward exposure and toxicological studies.

Sixty-five volatile compounds were included in the final curated data set. Of these volatile compounds, 14 were assigned a tentative identification based on matching of their spectral features to compounds in the National Institute of Standards and Technology (NIST) database. Lipid peroxidation, as potentially indicated by statistically significant increases of the alkanes n-hexane and 3-methylundecane, appeared to be higher in CDI patients than in controls, but other frequently detected markers of lipid peroxidation, such as lipid aldehydes, were not among the tentatively identified compounds possibly reflecting the greater stability alkanes. 4-heptanone (known to occur in the body as a result of exposure to routinely encountered plastics) appears as a CDI biomarker candidate. A decrease of  tetrachloroethylene in breath samples was also observed, potentially linked to the ability of some Clostridium species to metabolize tetrachloroethylene, suggesting this change could be due to degradation by C. difficile.

Nine tentatively identified VOCs were selected by a quadratic discriminant analysis modeling method. The model was able to distinguish between CDI cases and controls with 0.74 accuracy, sensitivity of 0.71, a specificity of 0.76, and a mean area under the receiver operating characteristic curve of 0.72.