Chemical Profiling Of Exhaled Breath from Cystic Fibrosis Subjects Using Comprehensive Two-Dimensional Gas Chromatography

Publication Type:
Thesis
Issue Date:
2021
Full metadata record
Chronic lung infections are the leading cause of death in subjects with cystic fibrosis (CF). To date, sputum culture is the most common technique for the diagnosis of lung infections in adult CF subjects. However, it requires several days or longer to obtain culture results. Therefore, a rapid diagnostic technique for lung infections would significantly improve CF healthcare. During recent decades, exhaled breath analysis has attracted interest as a rapid and non-invasive tool for the diagnosis of non-communicable diseases such as cancers and heart diseases. However, there is limited progress in the diagnosis of infectious diseases such as lung infections in CF subjects using volatile organic compounds (VOCs) as biomarkers of infection. In this study, sputum and breath samples were collected from CF subjects and healthy controls (only breath) and profiled for VOCs using comprehensive two-dimensional gas chromatography time-of-flight mass spectrometry (GC×GC–TOFMS). Multivariate analyses (e.g. principal component analysis and linear discriminant analysis (LDA)) were then performed to allow differentiation between: (i) CF subjects and healthy controls and (ii) CF subjects with/without 𝘗𝘴𝘦𝘶𝘥𝘰𝘮𝘰𝘯𝘢𝘴 𝘢𝘦𝘳𝘶𝘨𝘪𝘯𝘰𝘴𝘢 infections and those with no known lung infections as confirmed using their sputum culture results. This study identified a set of 16 VOCs which allowed differentiation between CF subjects and healthy controls. In particular, healthy controls were classified with 98% accuracy, while CF subjects were classified with 92% accuracy. It is important to note that all of the CF subjects that participated in this study are significantly different from control groups, not only in terms of their lung infection status but also in terms of numerous other factors (e.g. diet, lifestyle, medications, and other health complications). These factors can also impact the breath profiles obtained from the study group (CF subjects). The analysis of matching sputum and breath samples collected from CF subjects provided a set of 24 core VOCs common between both sample types. LDA performed using these VOCs provided accurate classification of CF subjects according to their lung infection status (i.e. CF subjects with/without 𝘗𝘴𝘦𝘶𝘥𝘰𝘮𝘰𝘯𝘢𝘴 𝘢𝘦𝘳𝘶𝘨𝘪𝘯𝘰𝘴𝘢 infection). The outcome of LDA also showed that these common VOCs have better classification accuracy than the entire profile of the VOCs detected in sputum and breath samples. Finally, the comparison of breath profiles between CF subjects with/without 𝘗𝘴𝘦𝘶𝘥𝘰𝘮𝘰𝘯𝘢𝘴 𝘢𝘦𝘳𝘶𝘨𝘪𝘯𝘰𝘴𝘢 infection and those with no known lung infection showed that it is also possible to allow differentiation between these contrasting groups using breath VOCs profiles.
Please use this identifier to cite or link to this item: