Sampling and Analytical Considerations for the Detection of Volatile Organic Compounds in Breath

Abstract

With every exhalation hundreds of volatile organic compounds (VOCs) are released. Patterns of VOCs have been shown to provide information about metabolic states and have been suggested as tools to aid diagnosis and management of a range of diseases. Biomarker discovery in breath holds several key advantages over blood as it is non-invasive, functionally endless and available in real-time. However, one of the major roadblocks faced by the breath research field is the lack of standardisation in sampling and analysis between institutions. Time of day is an important consideration when designing breath sampling campaigns, especially for diseases with diurnal fluctuations such as asthma. We sampled 20 individuals and found nine breath VOCs exhibited a circadian rhythm. Furthermore changes in rhythmicity were observed between the asthmatic and healthy groups. This suggests that VOCs might provide insight into circadian profiling of asthma and reinforces the importance of controlling time of day as a sampling parameter. Sorbent selection and best practices for managing water retention are important considerations to allow reproducible, untargeted, biomarker discovery in water saturated breath samples. This work showed that loading a standard mix in humid gas led to reduced recovery of compounds based on their chemical properties. Additionally we recommend dry purging times and found two compounds that were differently recovered on different sorbent materials during breath sampling. Whilst improving the design of studies and the methodologies employed in sampling is important, there is a need for standardisation across the field to allow comparison and pooling of results. Therefore, we propose a set of benchmarking values based on the washout curves of compounds found in peppermint oil capsules. By assessing the time taken for the signal of several key compounds to return to baseline values the sensitivity of a group’s sampling and analytical approach can be tested. Finally, for point of care devices to become reality, advances in the rapid targeted analysis of VOCs is required. Compact gas chromatography-photo ionisation or flame ionisation detector platforms have potential, but, the data generated is highly convoluted. Here two novel algorithms for extracting peak areas are presented. However, it was not possible for the algorithms to successfully deconvolve the data. Together the research presented in this thesis outlines a development in the effort to standardise VOC breath sampling and improve the quality of the data collected. These are important steps towards the development of clinically relevant tests for a wide range of diseases.

Date of Award	31 Dec 2019
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Royston Goodacre (Supervisor) & Stephen Fowler (Supervisor)