Exhaled Volatile Organic Compounds of Infection - ACS Publications

Subscriber access provided by University of Colorado Boulder

Review

Exhaled Volatile Organic Compounds of Infection: A Systematic Review Waqar Ahmed, Oluwasola Lawal, Tamara M Nijsen, Royston Goodacre, and Stephen J Fowler ACS Infect. Dis., Just Accepted Manuscript • DOI: 10.1021/acsinfecdis.7b00088 • Publication Date (Web): 04 Sep 2017 Downloaded from http://pubs.acs.org on September 7, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Infectious Diseases is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Infectious Diseases

Exhaled Volatile Organic Compounds of Infection: A Systematic Review Waqar M. Ahmed1,2, Oluwasola Lawal1,2, Tamara M. Nijsen2, Royston Goodacre3, Stephen J. Fowler1,4 1. Division of Infection, Immunity & Respiratory Medicine, School of Biological Sciences, The University of Manchester, Oxford Road, Manchester, M13 9PL, United Kingdom 2. Philips Research, Royal Philips B.V, High Tech Campus 34, Eindhoven, 5656 AE, The Netherlands 3. School of Chemistry, Manchester Institute of Biotechnology, The University of Manchester, 131 Princess Street, Manchester, M1 7DN, United Kingdom 4. Manchester Academic Health Science Centre, University Hospital of South Manchester NHS Foundation Trust, Southmoor Road, Wythenshawe, Manchester, M23 9LT, United Kingdom Correspondence: [email protected] Correspondence mailing address: Education and Research Centre, University Hospital South Manchester, Southmoor Road, Wythenshawe, Manchester, M23 9LT, UK

With heightened global concern of microbial drug resistance, advanced methods for early and accurate diagnosis of infection are urgently needed. Analysis of exhaled breath volatile organic compounds (VOCs) towards detecting microbial infection potentially allows a highly informative and non-invasive alternative to current genomics and culture-based methods. We performed a systematic review of research literature reporting human and animal exhaled breath VOCs related to microbial infections. In this review, we find that a wide range of breath sampling and analysis methods are used by researchers, which significantly affects inter-study method comparability. Furthermore, studies either target single breath VOCs in comparison to in vitro microbial culture headspace experiments,

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

or extract a global VOC fingerprint, with the latter focusing on a range of distinctive chemical groups to provide better indication of microbial infection, or host inflammatory response. In general, the field of breath analysis is still relatively immature, and there is much to be understood about the metabolic production of breath VOCs, particularly in a host where both commensal microflora as well as pathogenic microorganisms may be manifested in the airways. We anticipate that measures to standardise high throughput sampling and analysis, together with an increase in large scale collaborative international trials, will bring routine breath VOC analysis to improve diagnosis of infection closer to reality. Keywords Breath, Metabolomics, Mass Spectrometry, Infectious Disease, Respiratory Infection, Volatile Organic Compounds, VOCs, Microbial Metabolites [Table of Contents Graphic] Key Concepts •

Interest in breath research for the detection of microbial infection is on the rise

•

Volatile organic compounds from the breath can discriminate an individual with microbial infection from healthy controls

•

A wide variety of sampling and analysis methods for breath VOCs are available

•

Detection of distinguishable VOC fingerprints or chemical groups may give indication of infection pathogenesis

Global health organisations and government agencies have expressed the need for accurate and timely diagnosis of microbial infection, and reduced occurrence of antibiotic resistance1,2. Analysis of exhaled breath has the potential to provide a highly informative and non-invasive alternative to


Page 2 of 52

Page 3 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


current methods, many of which are based on invasive procedures paired with subsequent culture of the pathogen3. Inert gases and water vapour are the major constituents of breath. Non-volatile compounds are predominantly found in exhaled breath condensate4,5. Volatile organic compounds (VOCs) and other low molecular weight organic compounds (such as hydrogen and ammonia) can provide a fingerprint of metabolite production, and are released by various bodily routes – through urine, skin, faeces and breath6. Exhaled breath VOCs have been studied extensively for pollutant exposure and disease diagnosis7–9. Hundreds of VOCs have been identified from breath, which have both endogenous and exogenous origins10,11. There are many advantages with analysing breath VOCs, which include the following: •

Non-invasive and less strenuous sampling for subjects

•

Theoretically unlimited samples of breath gas

•

Free from biological and cellular material

•

Direct access to the lungs, and indirect access to the systemic circulation

In the last decade, several breath tests have been approved by the FDA for clinical use12. For instance, exhaled nitric oxide is related to asthmatic airway inflammation and has been subsequently approved for clinical use13,14. Other such breath tests include diagnosis of Helicobacter pylori15, and detection of a group of alkanes to screen patients with heart transplant rejection16. A range of instruments are used to analyse breath VOCs, and are discussed in more detail elsewhere17,18. One of the most commonly used is gas chromatography-mass spectrometry (GC-MS), usually coupled with a pre-concentration method such as thermal desorption or solid phase microextraction. Selected ion flow tube MS (SIFT-MS) and proton-transfer reaction MS (PTR-MS) instruments are also used to perform highly sensitive breath VOC analysis without the need for prior breath capture and pre-concentration19,20. Both instruments use soft ionisation and, in comparison to electron impact used in GC-MS, have the advantage that they analyse the intact product ion, which



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

can then undergo controlled product ion fragmentation from collision reactions with reagent ions. Researchers have also applied secondary electrospray ionisation-MS (SESI-MS) to breath analysis, avoiding pre-concentration and requirement of specified precursor ions21. Ion Mobility Spectrometry (IMS) and Differential Mobility Spectrometry (DMS) instruments offer a more versatile alternative to bulky MS instruments22,23. Here, ions are separated in the gas phase, where the drift time is proportional to the ion’s mobility through a controlled electrical field. Alternatively, electronic nose (E-nose) devices provide quick and simple classification between gaseous samples, and have proven useful in differentiating disease from healthy subjects via exhaled breath24,25. Volatile organic compounds are prone to high temporal and diurnal variation, and are inherently unstable depending on their chemical characteristics and the sampling method employed. Therefore, as for the majority of small molecule studies, enhanced data processing methods and standards are required to ensure robust and unbiased analysis26,27. Technical standards and best practices are available and is continually developing within the breath research community28. Data pre-processing and machine learning techniques within breath research have also been reviewed29. Scope of this Systematic Review A range of VOCs are emitted by microbes in vitro30,31. It is therefore assumed that exhaled breath VOC profiles can be linked to volatile metabolite fingerprints emitted by microbes during infection pathogenesis within the host environment. The aim of this systematic review is to assess research where breath VOCs are sampled and analysed, towards diagnosis of microbial infections and infectious diseases. We will evaluate and compare breath VOCs related to infection, and their sampling and analysis methods.

Bibliographic Search Methods Database Search


Page 4 of 52

Page 5 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Studies were chosen based on their description to discriminate infected subjects from control samples using breath VOC analysis. Specifically, this was based upon match with relevant keywords, syntaxes and article type. This global search was performed using three bibliographic databases, namely BIOSIS, Embase and MEDLINE. Figure 1 shows an overview of the bibliometric process used including article attrition and study selection process. [Figure 1] Some article types, (specifically patents, letters, conference abstracts, and reviews) were excluded as they did not provide complete information about VOCs or methods, and therefore were considered not relevant towards the scope of this review. However, relevant articles were archived, and reviewed as part of the quality assessment later on. In most cases, databases listed similar articles, therefore duplicated articles were removed during the automated search. Key words were grouped into three tiers and were based on the scope of this review: •

The first tier consisted of mandatory key words within the title (‘breath’ or ‘exhaled’).

•

The second tier included key words mandatory in the title or abstract (‘volatile(s)’ or ‘VOC(s)’ or ‘marker(s)’ and ‘infection’ or ‘infectious’ or ‘microbe(s)’ or ‘bacterial’ or ‘fungal’ or ‘viral’ or ‘pathogen’).

•

The third and final tier included search terms that may have been present within the title, abstract or text body (‘chromatography’ or ‘spectrometry’, or ‘e-nose’ or ‘spectrum’ or ‘sensor’). After the bibliographic search, the resultant research articles were further filtered, in order to match the inclusion and exclusion criteria.

Article Screening and Selection Criteria For this step, the resulting list of articles from the bibliographic database search were manually selected based on the scope of this review. This was accomplished by determining whether an article’s title and abstract description fit the predefined exclusion criteria: studies which sampled



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

exhaled breath condensate only; studies using gas-specific analysers (such as nitric oxide or CO2); targeted upper respiratory tract disorders (for example, rhinitis or halitosis), or where breath VOC analysis was not performed. Studies using animal subjects were included. Specifically, where studies performed investigations of zoonosis, or used animal models for pre-clinical translational investigations. Similarly, studies investigating microbial infection with a chronic underlying condition, and descriptive feasibility studies, were also included. During the database search, several articles were not assigned with an article type, or were duplicated because of grammatical or punctuation differences. These articles were removed or assessed as part of the manual filtration process. Quality Assessment and Additional Searches To reduce article selection bias, the final set of articles to be reviewed was agreed upon by two authors (WMA and TMN). The inclusion and exclusion criteria were developed with agreement between all authors. To ensure only high-quality articles were assessed, the search was limited to peer-reviewed journals. In addition, only original research articles with descriptive methods (sufficient for replication of experiments) were chosen. Where patients were sampled, adherence to ICH-GCP guidelines was mandatory. A high standard of quality was maintained during database searches and study selection. The database search was performed using the STN platform (Chemical Abstracts Service, American Chemical Society) by two specialists (Center for Information Services, Royal Philips B.V.). Keywords were subjected to syntax rules (e.g. keyword order, keyword distance, suffixes, and plurals, amongst others) within the English language. Both American and British spelling differences were considered.


Page 6 of 52

Page 7 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Additional manual searches were performed through the literature search engine Scopus (Elsevier B.V., version 2016). Here, each article from the search results was entered, and the article’s reference list and citation list were searched for additional relevant articles, which may have not been included from the original database search.

Review of Extracted Studies Principle Findings The bibliographic database search extracted 1526 articles. After article screening, study selection, and quality assessment methods were applied, the list reduced to 51 articles. The reviewing authors mutually agreed on the final set of articles to be reviewed. The resultant articles had a publication date from January 2004 to July 2017, and the majority of studies were performed in European and North American countries. Fewer studies were performed in Asian and African countries. Several studies investigated breath VOCs of infection (n = 24), including pneumonia and nosocomial respiratory infection (n = 9), Pulmonary Aspergillosis (n = 5), Pulmonary Tuberculosis (n = 5), airborne influenza viral infection (n = 2), and infection of the circulatory and gastrointestinal systems (n = 3). A list of these studies are summarised in Table 1. Studies where subjects had an infection with an underlying non-communicable disease were also reviewed (n = 17). Studies also investigated zoonotic infection or used animal models as pre-clinical in vivo experiments (n = 14). Across these categories, bacterial infections were the most studied (n = 26), followed by fungal (n = 7), viral (n = 2), and protozoan (n = 1). Others investigated mixed microbial infection. Some of these studies focused on the pathogenesis of the underlying disease, and performed experiments on a subgroup with infection. [Table 1] Respiratory Tract Infections in Humans Pneumonia and Nosocomial Respiratory Infection



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Respiratory infections are a common occurrence within hospitals due to the compromised nature of patients and their exposure to the clinical microbial community. Sampling from patients within a clinical setting requires investigators to adhere to regulations that govern the safety of clinical staff and patients. Therefore, the development and adaptation of devices to safely and efficiently sample patient breath is mandatory. With IMS-based and e-nose devices, it is possible to analyse immediately after sampling; for example, Ruzsanyi et al. and Rabis et al. utilise IMS-based technology with an overall test accuracy of above 82 % from the latter study.32,33. Similarly, Hockstein et al. showed the ability of a commercial e-nose sensor to differentiate exhaled breath gas from mechanically ventilated patients diagnosed with ventilator-associated pneumonia (VAP). The authors compared breath analysis to computerised tomography scans, and pneumonia scores, revealing an overall test accuracy of 60 % for the latter34,35. A number of studies have investigated the capability of breath analysis to diagnose VAP and other respiratory infections within patients under mechanical ventilation (n = 6). Filipiak et al. linked VOCs extracted from the headspace to patient breath, using cultured swabs as a reference36. For instance, 4-heptanone was found in the headspace of Candida albicans, and breath samples from patients with confirmed C. albicans infection. Similarly, Gao et al. found distinctive VOCs in the headspace of Acinetobacter baumannii culture, and linked the findings to VAP patients with A. baumannii infection37. Interestingly, VAP patients with and without A. baumannii infections could be differentiated. Both studies use a targeted approach where VOCs are known prior to in vivo analysis, thereby increasing confidence in chemical identification. In contrast to targeted analysis, other studies performed non-targeted analysis. This type of study design theoretically allows for maximum coverage (depending on the sampling and analysis methods) of both known and unknown compounds within a given sample, usually hypothesis-free and with the use of multivariate analysis and statistical validation38,39. Fowler et al. associated a range of VOCs with lower respiratory tract infection of patients in the intensive care unit39. They discussed


Page 8 of 52

Page 9 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


that the influence on breath VOC profile might not solely associated with pathogens but also the inflammatory response. Van Oort et al. discovered 1-propanol as a potential biomarker of bacterial growth in patients with pneumonia40. With regard to confirmed VAP diagnosis breath profiles, some similar VOCs were found and linked to metabolic pathways38. Between the VAP or VAP-like studies described above, commonly reported VOCs included ethanol, heptane, nonanal, and carane. Pulmonary Aspergillosis Patients with a weakened immune system may also be prone to infection by Aspergillus spp., the causative agents of pulmonary aspergillosis. All studies extracted from the search investigated invasive aspergillosis (IA), a severe form of pulmonary aspergillosis. Terpene-based compounds were found in vitro and compared to breath samples from patients with confirmed IA41,42. Koo et al. reported additional terpene-based VOCs, supported by reference standards and argued β-transbergamotene was misidentified as β-farnesene in an older study41. Gerritsen et al. also compared in vitro headspace to breath samples and aligned three VOCs between the two VOC profiles, however the number of breath samples were low43. Chambers et al. performed targeted analysis of 2pentylfuran from culture headspace experiments, and showed its potential as a biomarker for IA44. However, 2-pentylfuran was not identified in other studies, although other low molecular weight chemical species were found41–43. One study made use of e-nose technology and together with robust feature classification methods, were able to show a distinction between the breath of prolonged chemotherapy-induced neutropenia patients with IA infection and healthy controls (cross validated accuracy of 90.9 %)45. Pulmonary Tuberculosis As mentioned earlier, current diagnosis of microbial infectious diseases involves sampling methods which can be invasive. Samples are then cultured and typed (such as using proteomics-based diagnostics), and/or analysed directly for specific genetic loci. It is important to note that microbial pathogens may remain dormant for a long period of time before symptoms of infection emerge, and



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

some may become latent, therefore the presence of DNA alone does not mean that there is pathogenicity. Early diagnosis of pulmonary tuberculosis (pTB) infection is important to prevent progressive illness and drug resistance, therefore several studies have investigated breath VOCs of subjects infected with Mycobacterium tuberculosis. Philips et al. found a range of VOCs associated with pTB, notably methylated and non-methylated hydrocarbons, with 1-methyl naphthalene, and 1,4-dimethyl cyclohexane aligned between both in vitro culture headspace and in vivo patient breath samples46. The same authors found methylated and non-methylated hydrocarbons (including derivatives of benzene) from larger international studies and with one study analysing samples on-site (point-ofcare; POC)47,48. Where targeted analysis was conducted, researchers found methyl nicotinate increased in patient breath, compared to routine sputum smear tests49. This was based upon cited in vitro studies where nicotinic acid found in the headspace of cultured M. tuberculosis. Another study by Zetola et al. reported distinguishable pTB breath VOC profiles using an e-nose device50. Interestingly, these authors reported unique profiles of pTB with and without drug treatment. They were also able to show differences between patients with human immunodeficiency virus and pTB, compared to pTB alone. Although this could include VOCs related to the response of the immune system, it also indicates a VOC profile can depend on the concentration and diversity of a microbial community within the host environment. Influenza An increased interest for effective treatment of airborne influenza virus following an influenza subtype A(H1N1) pandemic in 200951 led to studies measuring the change in a subject’s breath signature over time, after vaccine inoculation52,53. It was found that the concentration of breath VOCs, especially those emitted following an immune response (nitrogen oxide and oxidative stressassociated VOCs), were correlated with vaccine metabolism over time. Philips et al. showed 2,8dimethyl-undecane had a positive correlation with vaccine response over time52.


Page 10 of 52

Page 11 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Infection within Underlying Chronic Lung Disease Non-communicable disease can be exacerbated by microbial pathogens, particularly for pulmonary diseases. Studies focused on obtaining breath VOC profiles of infected subjects together with an underlying chronic disease. A summary of these studies are listed in Table 2. [Table 2] Few studies profiled breath VOCs from subjects with chronic obstructive pulmonary disease and mixed microbial infection, and subsequently were able to distinguish between infected and noninfected subjects54,55. It is worth mentioning that Shafiek et al. also found significantly different ‘breathprints’ for subjects with pneumonia as a co-morbidity (sensitivity 85 %, specificity 86 %)55. Pulmonary infection in subjects with cystic fibrosis (CF) was found to be of particular interest amongst researchers. Feasibility studies have described efficient sampling and methods for high throughput and real-time analysis56,57. In CF the majority of studies focused on profiling pulmonary infection by Pseudomonas aeruginosa, and targeted highly volatile compounds. One study reported an increased concentration of pentane and decrease of methanol, however they also showed high variance in VOC intensity between breath samples58. Cyanide compounds have been linked to P. aeruginosa infection within CF subjects, specifically hydrogen cyanide and methyl thiocyanate59–61, using SIFT-MS together with in vitro experiments. To show hydrogen cyanide as a potential breath biomarker of CF patients with infection, a follow up study with a larger sample size was performed. The study confirmed that HCN was indeed an early breath biomarker in children with CF, however with low sensitivity, routine diagnostic use is unlikely to be useful62. Hydrogen cyanide was also found not to be characteristic of patients with chronic supportive lung disease and infection with P. aeruginosa63. The authors of the latter study also emphasised the importance of controlled sampling from the mouth, as hydrogen cyanide had high inter-sample variation when collected from the mouth.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Whiteson et al. demonstrated the presence of 2,3-butadione in both breath gas and microbial metabolism in sputum, an important outcome for further validating the use of breath biomarkers64. Scott-Thomas et al. used GC-MS and detected 2-aminoacetophenone (a larger and less volatile compound than the cyanide compounds mentioned earlier) as a potential breath biomarker for CF subjects with P. aeruginosa infection (sensitivity 93.8 %, specificity 69.2 %)65. Using similar techniques, two studies alternatively took advantage of non-targeted metabolomic methods, and found distinguishing VOC features from the breath, characteristic of CF subjects with infection66,67. Notably, Neerincx et al. detected a distinctive VOC profile for Staphylococcus aureus infection in children with CF (sensitivity 100 %, specificity 80 %)67. By contrast, Joensen et al. instead used an e-nose sensor and reported a significant difference for breath profiles of subjects with CF and P. aeruginosa infection (sensitivity 71.4 %, specificity 63.3 %)68. Interestingly, the authors found no such separation for patients with primary ciliary dyskinesia and P. aeruginosa infection. However, separation was found for other colonized pathogens such as Stenotrophomonas maltophilia, although the study sample size was comparably low and detection by the e-nose was most likely masked by the excessive signal from VOCs associated with an inflammatory response. The influence of multiple microbial co-infection contributing to CF has also been investigated. One study targeted three volatile sulfur compounds69. As for the cyanide compounds mentioned before, the chosen sulfur compounds are highly volatile, and similarly would show high inter-sample variation. Another study identified 2-pentylfuran from the breath of CF patients with A. fumigatus infection70. However, 2-pentylfuran was also identified in some patients with no microbial infection. Additionally, some patients infected with A. fumigatus also had confirmed infection of Haemophilus influenzae and P. aeruginosa. Infectious Diseases of the Circulatory and Gastrointestinal Systems in Humans


Page 12 of 52

Page 13 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Aside from researching respiratory infection, studies have also investigated the potential of breath analysis in infectious diseases of the circulatory or gastrointestinal system (see Table 1 for a list of relevant studies). Urea breath tests for Helicobacter pylori infection are routine practice, where urease intensity is measured (the urease enzyme pathway is used by H. pylori) after digesting a labelled carbon isotope of urea, which after urease treatment generates labelled CO2. With regard to breath VOCs, increased concentrations of hydrogen nitrate and hydrogen cyanide (by SIFT-MS), and ethanol and butane (SPME-GC-MS) were suggested as potential biomarkers of H. pylori infection71,72. However, these compounds are very volatile, are not species specific, are commonly found in breath, and therefore difficult to translate into distinguishable biomarkers for H. pylori infection. In addition, they may not be a suitable cost-effective alternative to established methods. As endogenous exhaled breath VOCs are directly linked to the circulatory system, bloodstream infections have also been investigated73. Here, concentration of some VOCs, most notably thioesters, have shown to increase after colonisation by the malaria causing protozoan Plasmodium falciparum. After treatment with antimalarial chemotherapy, the concentration of thioesters reduced significantly. Although thioesters were not found in the headspace of P. falciparum, they have been linked to many metabolic processes including lipid degradation. However, increased concentration of thioesters in breath may represent the VOC originating from the interaction between P. falciparum and the host immune system. Studies Investigating Zoonoses Early detection of zoonosis is important to both animal and human health. Zoonotic infections can ultimately spread between animals, and from animals to humans. A list of studies investigating zoonosis, or using animal models, are shown in Table 3. [Table 3] Bovine respiratory infections were predominantly researched compared to other zoonoses (60%)74– 77

. An early pilot study focused on efficient sampling methods74. One study focused on brucellosis, a



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

highly contagious zoonosis, and reported breath VOCs linked to Brucella abortus infection in bison77. Early detection of bovine tuberculosis within cattle would prevent infection spreading across dairies. Studies were able to discriminate VOC profiles of Mycobacterium bovis from cattle breath samples75,76. Interestingly, Peled et al. were able to show variation in breath VOCs for infected cattle and non-infected cattle, within dairies categorised as infected or infection-free75. Another livestock infection, Mycobacterium avium subspecies paratuberculosis, was studied within ruminants78,79. Tentative identification of a range of breath VOCs, or a group of VOC features78, have been associated to both infected and non-infected ruminants. Studies using Animal Models Pre-clinical in vivo models provide fundamental information about infection pathogenesis, the host response, and organ function, in pharmaceutical and medical device development studies. Researchers predominantly used murine animals for infection inoculation and subsequent breath analysis (see Table 3 for selected articles)80–83, although in a recent study Mellors et al. used a macaque animal model to perform non-targeted analysis investigating differences in breath profiles before and after TB infection84The authors reported several volatiles, of which dodecane, tridecane, and hexylcyclohexane have been associated with TB in previous human studies47,48. In other studies by Zhu et al. using murine models, distinct breath VOC profiles between bacterial strains were reported80. It is noteworthy that only an estimated 25 % to 34 % of VOC fragments can be directly linked to in vitro bacterial culture83, illustrating that the metabolism of the bacterium inside the host is very different compared with when it is cultivated in an artificial environment. A further study by the same group with an increased subject cohort found lung infection could be robustly discriminated depending on the infected pathogen (between S. aureus and P. aeruginosa)85. Bean et al. sampled from mice inoculated with isogenic strains of methicillin-resistant S. aureus were analysed, and it was found that VOC fragments, shared with methicillin-sensitive S. aureus, showed different concentrations between the two strains81. In another study by the same group, induced S.


Page 14 of 52

Page 15 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


aureus and P. aeruginosa lysates provoked an immune response and lung damage after 48 hours, which correlated with the breath VOC fingerprint82. This is important as VOCs could be used trace early onset and progress of infection. Where GC was used instead of SESI-MS, researchers sampled breath of mice inoculated with Borrelia hermsii, and found an increased concentration in carbon monoxide86, which reduced after antibiotic treatment. Similarly, increased levels of carbon monoxide (normalised to carbon dioxide levels) were found in rats injected with Lipopolysaccharide endotoxin purified from Escherichia coli to induce sepsis87. Several VOCs relating to an inflammatory response (as a result of endotoxin injection) were identified in another study, also investigating sepsis in murine models88.

Review of Breath Sampling and Analysis Methods Breath Sampling Within the scope of this review, we found that a wide range of breath sampling methods were used, as summarised in Table 4. This was however dependant on the target disease and the environment around the subject. Sampling devices were only used when transportation or storage of breath VOCs was necessary. [Table 4] The portion of breath collected is important in disease diagnosis. For example, sampling alveolar breath may include a higher proportion of blood-borne VOCs. In some cases, breath was sampled from the nose to reduce influence of VOCs originating from the mouth61,63. Studies using animal subjects had to develop bespoke breath sampling systems depending on the animal’s anatomy75,77,83. Breath from mechanically-ventilated patients was sampled from the ventilator circuit limb36–38, or by inserting a catheter into the endotracheal tube39. Volatile organic compounds can be linked to other factors not directly associated with disease. These confounding factors were generally taken into account by most studies using various methods,



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

including restriction of diet or excessive movement immediately prior to sampling, and smoking restrictions, amongst others. Diet is an important consideration as VOCs of interest may change in concentration and therefore yield false-positive results42,89. Where reported, studies defined dietary restrictions between one to three hours before sampling. The total volume of sampled breath ranged from a single breath60; multiple breaths49,50, to predefined sample gas volumes39. One study sampled a total of 100 mL per subject and concluded this volume was sufficient for discriminating CF subjects with infection (for SPME-GC-MS)56. Gas sampling bags varied in total volume between studies, up to 10 L45,54. Some studies stored and transported breath in gas sampling bags61, whilst others defined a time limit between sampling and analysis67. Depending on the target infection, a portion of the breath may be sampled as opposed to total breath sampling. This is mandatory for blood-borne infection, where many VOCs may be found in the alveolar breath portion. Gas sampling bags are known to introduce contaminants into breath gas, and can change VOC concentration during storage as polar VOCs can adhere to the plastic surface. In addition, some gas sampling bags are known to be sensitive to amines and sulfur compounds. Some researchers described a sampling system where breath was sampled into evacuated steel containers thereby avoiding adsorption onto sorbent-packed tubes58,69,86. This method was similarly used with a preconcentration system, however was limited to the number of samples that could be analysed at any one time. Storing breath gas in evacuated steel containers has advantages over gas sampling bags as there is less risk of contamination and VOC degradation. In comparison, the purge and trap method (where breath gas is immediately purged onto sorbent tubes) would allow prolonged storage of breath VOCs (at least several days) and high-throughput analysis90,91. With regard to sorbent tube use, some studies used sorbent material that were not appropriate for semi-quantitative analysis of highly volatile compounds, but were identified and reported anyway. Not all VOCs absorb to all sorbents. For instance, ethanol, with a carbon number of two, cannot be


Page 16 of 52

Page 17 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(semi-)quantitatively adsorbed onto many sorbent materials37. Additional sample preparation may be required depending on the type of sorbent bed combination used. Volatiles with low vapour pressure require sorbent beds with a wide carbon absorption range, such as porous polymers. Sample Analysis Breath VOCs can be measured in real-time or off-line. The latter encompasses off-site analysis (where measurement was performed in a laboratory environment), or at the point-of-care (measurement in the same environment as the subject, but not in real-time). The choice of instrument determined the type of analysis that could be performed. Sample analysis instruments are summarised in Table 5. [Table 5] Gas chromatography coupled to MS is a popular analysis method throughout extracted studies, but requires that breath gas be pre-concentrated prior to analysis. Many desorption methods are used for pre-concentrating breath. Solid phase micro extraction (SPME) is the most frequently used preconcentration method. However, SPME is highly biased towards hydrophobic VOCs, and use of multisorbent tubes, or directly purging breath gas (as possible with evacuated steel containers), is more suitable for most VOCs. Instrument settings varied between research groups, depending on the study design and if VOCs were targeted. Limited versatility and measurement time restricted GC-based methods to off-site analysis in most cases, although point-of-care analysis was possible in one study that used GC coupled with a surface acoustic wave detector48. Here, it was reported that a breath sample could be measured within six minutes. With regard to MS instruments, a range of mass analysers were used. Studies using non-targeted profiling used time-of-flight mass analysers38,39,66. Many studies described the use of a single quadrupole mass analyser (see Tables 1-3). In addition, studies took advantage of selected ion monitoring mode with triple quadrupole mass analysers for targeted quantitative analysis44,49,65,70.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Where MS was used, the mass-to-charge (m/z) acquisition range varied between studies. A broad m/z range was set by some studies thereby reducing the number of scans/second39, whilst others set a smaller range (approximately from m/z 20 to 200) more suited to breath VOCs36,37. To increase sensitivity of detection by GC for very volatile compounds, one study used a flame ionisation detector58, whilst another study additionally used a chemical ionisation source with MS65. The mechanism behind PTR-MS and SIFT-MS allows highly sensitive real-time analysis of targeted breath VOCs57,59,60,63. In some circumstances, lack of suitable sampling inlets or study design meant these instruments were used at the point-of-care, but not in real-time53,62,71. Similarly studies using SESI-MS, where developed for real-time analysis, measured breath VOCs at the point-of-care. Mass spectrometry-based approaches are known to have high operational costs and lack portability, when compared to IMS-based detectors and e-nose sensors. To this end, some researchers measured breath VOCs using portable multi-capillary column IMS instruments. The differences between IMS-based platforms used within studies are described elsewhere23. Researchers were able to perform point-of-care or real-time measurements with IMS-based instruments, although some were proof-of-concept studies with a low sample size33,78. In comparison, e-nose devices were the most commonly used instrument for real-time analysis of breath VOCs. This is because they are comparatively simple to operate and adaptable depending on study requirements. There are many variations in e-nose device sensors, which included the commercial Cyranose 32034,35,45,54,55,68, coated quartz microbalance sensors50, and a nanomaterialbased sensor array75,77. Consequently, specificity of volatiles and detection sensitivity will differ between sensing mechanisms. Quality Assurance and Data Analysis Quality assurance procedures are important for small molecule analysis92,93. Compared to serum or urine metabolomics, quality assessment and control methods in breath research are relatively underdeveloped.


Page 18 of 52

Page 19 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Research of sampling standardisation have raised some important issues94–97. With regard to breath sample quality, some studies sampled ambient air parallel to breath samples. Theoretically, VOCs present in environment air samples are exogenous to breath and therefore could be subtracted from breath sample VOCs. The diverse use and interpretation of reference standards and materials was observed. Analytical standards included use of target analyte standards65,67, permeation tubes63, retention indexing mixtures48, and certified reference mixtures39. However, not all studies described the use of analytical standards, depending on the instrument used. For example, reagent ions used in soft ionisation methods were used to measure instrument precision. Reporting standards have been published for metabolomic studies26,27. These standards allow necessary chemical and data analysis recommendations for studies, and these are particularly important in breath VOC research. As breath VOC samples can have a high number of variables and confounding factors, most studies performed various univariate and multivariate analysis with the acquired VOC data. In most cases, exploratory unsupervised analysis was performed to gain an insight into how the data were correlated. Several unsupervised multivariate methods were employed, such as principal component analysis (PCA), fuzzy c-means, or hierarchical clustering analysis (HCA). To counteract high intersubject variability and reduce the dominant effect of high concentration VOCs, alternative data processing methods have been described39,98. In addition to unsupervised analysis, some researchers additionally employed supervised learning methods to further assess breath VOC data based on pre-defined hypotheses. Where data were deemed to have a linear relationship, linear discriminant analysis (LDA) and partial least squaresdiscriminant analysis (PLS-DA) was used. Studies using non-linear methods included k-nearest neighbour, random forest (RF), support vector machines (SVM), and weighted digital analysis (WDA). Further information on the use of these chemometric approaches are described in detail



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

elsewhere99. However, such multivariate methods, which are based on training a computational algorithm, risk data over-correction and it is important to avoid false discovery100. Studies reduced the likelihood of statistical bias by robust validation of their statistical methods. In some studies, metabolite pathway databases were used to evaluate VOCs for possible network linkage with host metabolism38,76. Where targeted analysis of known VOCs was performed, multivariate methods were not used. However in some non-targeted studies, acquired data was processed without additional statistical or hypothesis-free classification36,41,42. Here, the authors explain that VOC profiles of subjects should be treated on an individual bases, thereby eliminating the effect of data over-correction and reducing interference associated with confounding factors36,42.

Discussion Breath contains a complex mixture of VOCs comprising many different chemical species from endogenous and exogenous sources. The host immune response to microbial invasion and pathogenesis can result in an enhanced VOC profile, thereby masking VOCs emitted directly from microbes. As perhaps expected, authors have reported the absence of VOCs from in vitro headspace when compared to breath samples37,73, which is because of biochemical differences due to hostpathogen interactions that are missed in artificial culture. For example, a study found only 4 out of 8 volatiles were aligned between in vitro and in vivo breath samples for A. baumannii37. It has been reported that less than half of a VOC profile may be linked directly to microbial pathogens83. This is evident when aligning bacterial metagenomics and breath VOC analysis101. Additionally, co-cultures have shown release of distinctive VOCs in headspace102,103. Notably, P. aeruginosa has shown to exacerbate the growth of A. fumigatus when co-cultured together103. This suggests distinctive VOCs may be produced for each pathogen-pathogen, pathogen-host, and host-disease interaction. An observation made between some studies was an increase in terpene-based compounds found in breath samples linked to invasive aspergillosis41,42. Bacterial infection may also emit distinctive chemical groups depending on their individual virulence mechanisms. For example, many studies


Page 20 of 52

Page 21 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


targeting P. aeruginosa reported nitrogen-containing compounds59,60,89. Furthermore, branched hydrocarbons were found across studies investigating tuberculosis in both humans and animals46,49,75,76. Branched hydrocarbon VOCs, together with increased concentration of inert gases, were reported across many non-targeted studies and could indicate a host’s inflammatory response to infection rather than the presence of a particular microbial species37,39,86. Breath samples in studies investigating influenza vaccination were not taken from subjects without prior vaccine treatment. Breath VOCs were therefore associated with oxidative stress and inflammatory response caused by the vaccine, which included branched hydrocarbons52,53. Bean et al. grouped similar mass spectral fragmentation patterns into one VOC identity, to reduce incorrect identification82. Rounding exact mass to the nearest integer from high resolution instruments is a useful technique for small molecules104, although we would stress that many metabolomics studies use accurate mass to infer molecular formulae of an analyte under analysis26. The ideal m/z range for breath analysis depends on the research hypothesis. Many VOCs characteristically have a mass spectral molecular ion peak below m/z 200, although this molecular weight is somewhat arbitrary. Some volatile and semi-volatile chemicals in breath have distinguishable molecular ion peaks above m/z 200 (for example, pentadecane molecular ion peak m/z = 212), consequently reducing accuracy of identification within non-targeted experiments. In addition, the percentage probability of tentative identification may be lowered or less accurate. Matching VOCs were identified across studies. Hydrogen cyanide was associated with both H. pylori infected subjects, and P. aeruginosa infection in CF subjects62,71. Two studies investigating bovine tuberculosis using the same analytical methods, however found only two matching VOCs75,76. Concentration levels for ethanol and heptane were contradictory between studies investigating respiratory infection within the intensive care unit36–39. Hydrogen cyanide was well characterised in the breath of patients with infection and CF, but was not suitable for diagnosis due to low sensitivity62. 2-pentylfuran has been proposed as a biomarker for invasive aspergillosis105, although



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

this has not been replicated by other groups in either in vitro headspace experiments, or patient breath samples41,42,106. 2-pentylfuran was also quantified in foodstuffs by the same researchers107. Similarly, 2-aminoacetophenone, a proposed biomarker of P. aeruginosa infection in CF patients, was also found in foodstuffs89. This raises the concern that breath VOC profiles are influenced by the environment or activity before or during sampling. Preference of sampling methods by research groups primarily relied upon the analysis capabilities, the target infection being studied and the portion of breath collected (and this is reviewed in depth elsewhere108). For example, access to intensive care units, site inclusion in underdeveloped countries, or facilities for animal studies, may be difficult to organise. This diversity makes breath sampling standardisation efforts difficult across the breath research community. For example, if sampling breath from the nose, the presence of upper respiratory pathogens must be considered, for example P. aeruginosa nasal infection often precedes lung infection in cystic fibrosis. Where breath is sampled from a mechanical ventilator, It is important to consider exogenous VOCs from ventilator tubing and inhaled anaesthetics contamination during sampling and analysis within the intensive care unit40,109. Another key concern is the lack of consistency of proposed breath biomarkers that were not found in other studies investigating the same infection. For example, out of three studies investigating lung infection in intensive care patients37–39, only 4/26 potential breath biomarkers were matched between two or more studies (tetradecane, nonanal, ethanol, and heptane). Several breath biomarkers were also found for other diseases, for instance, nonanal was identified in the breath of patients with VAP, and in patients with lung cancer from other studies110. Another example from within the review is ethanol, which was also found in patients with H. pylori infection, and tuberculosis infection46,72. Although we have discussed the potential origin of these VOCs, the quest for finding highly selective breath biomarkers is without doubt very challenging, especially with the lack of defined chemical identification workflows and standardised methodologies in breath


Page 22 of 52

Page 23 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


research. However, breath analysis alone will not constitute a complete diagnosis for a disease, yet may prove useful when combined with other routine diagnostic methods, symptoms, and subject activity and medical history. Clinical breath samples may include additional VOCs linked to drug treatment, for example, anaesthetic agents found in patient breath after recovering from surgery11. In addition, when VOC profiles are associated with an particular infection, the influence of non-infectious underlying conditions must also be taken into account39. It is clear that sampling humans is very complex as they are exposed to many other factors – both intrinsic and extrinsic – as reviewed in Goodacre et al.111, which might affect the microbiome, and hence the ability to predict infection status accurately. With many studies, the accuracy of breath VOC analysis is relied upon by the positive identification of infection by comparative invasive diagnosis. If false-positive or negative identification is given by these methods, the overall accuracy of breath VOC analysis may be reduced and hence unreliable. The most widely used statistical validation method was the leave-one-out technique, however it is often more appropriate to use resampling methods such as bootstrapping112,113. This is especially important as many studies use small sample sizes and statistically processed data can be interpreted in different ways114. By way of example, studies have processed data with and without statistical validation and have found differences in overall test accuracy34,35, which suggests that the data generated are perhaps not appropriate for predictive analyses. It is worth stressing that these invasive diagnostic methods are used as the primary reference data and if such data contain errors then these errors will be propagated throughout analysis; the adage 'garbage in – garbage out', coined by Zupan and Gasteiger, is particularly relevant here for chemical data calibrated by ‘gold standard’ data that are not so ‘golden’115. Breath VOC analysis for detecting infection requires further research and development. For cases where VOCs are well-characterised, applications for fast on-site diagnosis are being developed116. However, no current VOC biomarkers of infection have been clinically approved as of yet. Increased



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

research output and understanding of VOC metabolic pathways would increase confidence diagnostic approval. With regard to breath VOC analysis for diagnosis, cost-effectiveness is very important. In order for a test to be cost effective, a compromise between diagnostic accuracy, the prevalence of an infection, and monetary cost per sample, is required. This is especially important when aiming to replace current practices or propose new breath biomarkers. Current low cost E-nose devices perform binary classification between multiple samples, limiting their VOC measurement ability and requiring prior sample training. However, some researchers have argued a binary classification is all that is needed for diagnosis35, and sensors selective for volatiles are under development. The use of e-nose and IMS technology within the clinical environment is particularly interesting, as devices can be tailor made and easy-to-use for clinical staff. In addition, sensor electronics can be mass-produced making integration into routine diagnosis less demanding than other technologies, resulting in low cost infection disease diagnostics117. International multicentre studies remove experimental bias from tentative biomarkers identifications, and increase confidence for application in clinical diagnosis25,47,48,118. We have noticed that fewer studies were performed in continents generally considered to have higher occurrence of infectious diseases. Future research could include sites in countries with a higher prevalence of a specific infection. Investigators new to breath analysis may find recent technical standards useful, for a review of standardisation efforts across sampling and analysis of breath28. The continued interest in breath analysis for non-invasive and fast identification of infection is evident by the exponential rise in research publications over the last decade. With concerns of antimicrobial resistance and invasiveness of current methods, breath analysis is proving valuable and advantageous for early detection of microbial infection.

Conclusions


Page 24 of 52

Page 25 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Studies show that diagnosis of infection and infectious diseases using breath VOC biomarkers is possible. External validation is required for reliable measurements, before biomarkers can be tested within clinical protocols. However, sampling and analysis methods are heavily dependent on the targeted disease, surrounding environment, and instruments used, which can affect standardisation efforts, and consequently a ‘one size fits all’ analysis is not yet apparent. Metabolic pathways of VOC production are still not yet fully understood and this is important as this would enhance infection diagnostics and help understand the complex host-pathogen crosstalk, and that microbial infection and associated inflammation may produce distinguishable VOC fingerprints or chemical groups. Increased high-throughput research of breath VOCs associated with infection are needed to further validate biomarkers and to fully realise the potential of non-invasive diagnostic methods.

Author Information W.M.A, T.M.E, O.L, R.G and S.J.F planned and carried out the systematic search, and reviewed the studies. W.M.A wrote the manuscript. The authors declare no competing financial interest.

Acknowledgements We would like to thank our funding body (European Commission FP7 Marie Curie Actions), under the Industry-Academia Partnerships and Pathways (IAPP) programme (MC-IAPP BreathDx 611951). We are also grateful for fruitful collaborations within the BreathDx Consortium (University of Manchester, Royal Philips B.V., Academic Medical Centre Amsterdam, and University of Twente), Center for Information Services, Chronic Disease Management Department, (Philips Research B.V.), the Laboratory of Bioanalytical Spectroscopy (University of Manchester), and the Manchester Breathomics Group (University of Manchester).

References



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(1)

World Health Organization. Antimicrobial resistance: global report on surveillance. WHO 2014, 2, 3–6.

(2)

Davies, S. C.; Fowler, T.; Watson, J.; Livermore, D. M.; Walker, D. Chief Medical Officers Report: Infections and the Rise of Antimicrobial Resistance; 2013; Vol. 2.

(3)

Rattray, N. J. W.; Hamrang, Z.; Trivedi, D. K.; Goodacre, R.; Fowler, S. J. Taking your breath away: metabolomics breathes life in to personalized medicine. Trends Biotechnol. 2014, 32 (10), 538–548 DOI: 10.1016/j.tibtech.2014.08.003.

(4)

Hunt, J. Exhaled breath condensate: An evolving tool for noninvasive evaluation of lung disease. J. Allergy Clin. Immunol. 2002, 110 (1), 28–34 DOI: 10.1067/mai.2002.124966.

(5)

Zanconato, S.; Carraro, S.; Corradi, M.; Alinovi, R.; Pasquale, M. F.; Piacentini, G.; Zacchello, F.; Baraldi, E. Leukotrienes and 8-isoprostane in exhaled breath condensate of children with stable and unstable asthma. J. Allergy Clin. Immunol. 2004, 113 (2), 257–263 DOI: 10.1016/j.jaci.2003.10.046.

(6)

Amann, A.; De Lacy Costello, B.; Miekisch, W.; Schubert, J.; Buszewski, B.; Pleil, J.; Ratcliffe, N.; Risby, T. The human volatilome: volatile organic compounds (VOCs) in exhaled breath, skin emanations, urine, feces and saliva. J. Breath Res. 2014, 8 (3), 34001 DOI: 10.1088/17527155/8/3/034001.

(7)

Miekisch, W.; Schubert, J. K.; Noeldge-Schomburg, G. F. E. Diagnostic potential of breath analysis—focus on volatile organic compounds. Clin. Chim. Acta 2004, 347 (1–2), 25–39 DOI: 10.1016/j.cccn.2004.04.023.

(8)

Lindstrom, A. B.; Pleil, J. D.; Berkoff, D. C. Alveolar breath sampling and analysis to assess trihalomethane exposures during competitive swimming training. Environ. Health Perspect. 1997, 105 (6), 636–642 DOI: 10.1289/ehp.97105636.

(9)

Bos, L. D.; Sterk, P. J.; Fowler, S. J. Breathomics in asthma and COPD. J. Allergy Clin. Immunol.


Page 26 of 52

Page 27 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2016 DOI: 10.1016/j.jaci.2016.08.004. (10)

de Lacy Costello, B.; Amann, A.; Al-Kateb, H.; Flynn, C.; Filipiak, W.; Khalid, T.; Osborne, D.; Ratcliffe, N. M. A review of the volatiles from the healthy human body. J. Breath Res. 2014, 8 (1), 14001 DOI: 10.1088/1752-7155/8/1/014001.

(11)

Pleil, J. D.; Stiegel, M. A.; Risby, T. H. Clinical breath analysis: discriminating between human endogenous compounds and exogenous (environmental) chemical confounders. J. Breath Res. 2013, 7 (1), 17107 DOI: 10.1088/1752-7155/7/1/017107.

(12)

Paschke, K. M.; Mashir, A.; Dweik, R. A. Clinical applications of breath testing. F1000 Med. Rep. 2010, 2 (56), 1–6 DOI: 10.3410/M2-56.

(13)

Kharitonov, S. A.; Yates, D.; Robbins, R. A.; Logan-Sinclair, R.; Shinebourne, E. A.; Barnes, P. J. Increased nitric oxide in exhaled air of asthmatic patients. Lancet 1994, 343 (8890), 133–135 DOI: 10.1016/S0140-6736(94)90931-8.

(14)

Silkoff, P. E.; Carlson, M.; Bourke, T.; Katial, R.; Ögren, E.; Szefler, S. J. The Aerocrine exhaled nitric oxide monitoring system NIOX is cleared by the US Food and Drug Administration for monitoring therapy in asthma. J. Allergy Clin. Immunol. 2004, 114 (5), 1241–1256 DOI: 10.1016/j.jaci.2004.08.042.

(15)

Graham, D. Y.; Evans, D. J.; Alpert, L. C.; Klein, P. D.; Evans, D. G.; Opekun, A. R.; Boutton, T. W. Campylobacter Pylori Detected Noninvasively By the 13C-Urea Breath Test. Lancet 1987, 329 (8543), 1174–1177 DOI: 10.1016/S0140-6736(87)92145-3.

(16)

Phillips, M.; Boehmer, J. P.; Cataneo, R. N.; Cheema, T.; Eisen, H. J.; Fallon, J. T.; Fisher, P. E.; Gass, A.; Greenberg, J.; Kobashigawa, J.; et al. Heart allograft rejection: Detection with breath alkanes in low levels (the HARDBALL study). J. Hear. Lung Transplant. 2004, 23 (6), 701–708 DOI: 10.1016/j.healun.2003.07.017.

(17)

Beale, D.; Jones, O.; Karpe, A.; Dayalan, S.; Oh, D.; Kouremenos, K.; Ahmed, W.; Palombo, E. A



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Review of Analytical Techniques and Their Application in Disease Diagnosis in Breathomics and Salivaomics Research. Int. J. Mol. Sci. 2016, 18 (1), 24 DOI: 10.3390/ijms18010024. (18)

Lourenço, C.; Turner, C. Breath Analysis in Disease Diagnosis: Methodological Considerations and Applications. Metabolites 2014, 4 (2), 465–498 DOI: 10.3390/metabo4020465.

(19)

Smith, D.; Španěl, P. Selected ion flow tube mass spectrometry (SIFT-MS) for on-line trace gas analysis. Mass Spectrom. Rev. 2005, 24 (5), 661–700 DOI: 10.1002/mas.20033.

(20)

Lindinger, W.; Hansel, A.; Jordan, A. On-line monitoring of volatile organic compounds at pptv levels by means of proton-transfer-reaction mass spectrometry (PTR-MS) medical applications, food control and environmental research. Int. J. Mass Spectrom. Ion Process. 1998, 173 (3), 191–241 DOI: 10.1016/S0168-1176(97)00281-4.

(21)

Zhu, J.; Bean, H. D.; Kuo, Y. M.; Hill, J. E. Fast detection of volatile organic compounds from bacterial cultures by secondary electrospray ionization-mass spectrometry. J. Clin. Microbiol. 2010, 48 (12), 4426–4431 DOI: 10.1128/JCM.00392-10.

(22)

Basanta, M.; Jarvis, R. M.; Xu, Y.; Blackburn, G.; Tal-Singer, R.; Woodcock, A.; Singh, D.; Goodacre, R.; Paul Thomas, C. L.; Fowler, S. J. Non-invasive metabolomic analysis of breath using differential mobility spectrometry in patients with chronic obstructive pulmonary disease and healthy smokers. Analyst 2010, 135 (2), 315 DOI: 10.1039/b916374c.

(23)

Fink, T.; Baumbach, J. I.; Kreuer, S. Ion mobility spectrometry in breath research. J. Breath Res. 2014, 8 (2), 27104 DOI: 10.1088/1752-7155/8/2/027104.

(24)

Wilson, A. Advances in Electronic-Nose Technologies for the Detection of Volatile Biomarker Metabolites in the Human Breath. Metabolites 2015, 5 (1), 140–163 DOI: 10.3390/metabo5010140.

(25)

Nakhleh, M. K.; Amal, H.; Jeries, R.; Broza, Y. Y.; Aboud, M.; Gharra, A.; Ivgi, H.; Khatib, S.; Badarneh, S.; Har-Shai, L.; et al. Diagnosis and Classification of 17 Diseases from 1404 Subjects


Page 28 of 52

Page 29 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


via Pattern Analysis of Exhaled Molecules. ACS Nano 2017, 11, 112–115 DOI: 10.1021/acsnano.6b04930. (26)

Sumner, L. W.; Amberg, A.; Barrett, D.; Beale, M. H.; Beger, R.; Daykin, C. A.; Fan, T. W.-M.; Fiehn, O.; Goodacre, R.; Griffin, J. L.; et al. Proposed minimum reporting standards for chemical analysis. Metabolomics 2007, 3 (3), 211–221 DOI: 10.1007/s11306-007-0082-2.

(27)

Goodacre, R.; Broadhurst, D.; Smilde, A. K.; Kristal, B. S.; Baker, J. D.; Beger, R.; Bessant, C.; Connor, S.; Capuani, G.; Craig, A.; et al. Proposed minimum reporting standards for data analysis in metabolomics. Metabolomics 2007, 3 (3), 231–241 DOI: 10.1007/s11306-007-00813.

(28)

Horváth, I.; Barnes, P. J.; Högman, M.; Olin, A.; Amann, A.; Antus, B.; Baraldi, E.; Bikov, A.; Boots, A. W.; Bos, L. D.; et al. A European Respiratory Society technical standard : exhaled biomarkers in lung disease. Eur Respir J 2017, 49 DOI: 10.1183/13993003.00965-2016.

(29)

Smolinska, A.; Hauschild, A.-C.; Fijten, R. R. R.; Dallinga, J. W.; Baumbach, J.; van Schooten, F. J. Current breathomics—a review on data pre-processing techniques and machine learning in metabolomics breath analysis. J. Breath Res. 2014, 8 (2), 27105 DOI: 10.1088/17527155/8/2/027105.

(30)

Bos, L. D. J.; Sterk, P. J.; Schultz, M. J. Volatile Metabolites of Pathogens: A Systematic Review. PLoS Pathog. 2013, 9 (5), e1003311 DOI: 10.1371/journal.ppat.1003311.

(31)

Lemfack, M. C.; Nickel, J.; Dunkel, M.; Preissner, R.; Piechulla, B. mVOC: a database of microbial volatiles. Nucleic Acids Res. 2014, 42 (1), 744–748 DOI: 10.1093/nar/gkt1250.

(32)

Rabis, T.; Sommerwerck, U.; Anhenn, O.; Darwiche, K.; Freitag, L.; Teschler, H.; Bödeker, B.; Maddula, S.; Baumbach, J. I. Detection of infectious agents in the airways by ion mobility spectrometry of exhaled breath. Int. J. Ion Mobil. Spectrom. 2011, 14 (4), 187–195 DOI: 10.1007/s12127-011-0077-6.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(33)

Ruzsanyi, V.; Baumbach, J. I.; Sielemann, S.; Litterst, P.; Westhoff, M.; Freitag, L. Detection of human metabolites using multi-capillary columns coupled to ion mobility spectrometers. J. Chromatogr. A 2005, 1084 (1–2), 145–151 DOI: 10.1016/j.chroma.2005.01.055.

(34)

Hockstein, N. G.; Thaler, E. R.; Torigian, D.; Miller, W. T.; Deffenderfer, O.; Hanson, C. W. Diagnosis of pneumonia with an electronic nose: correlation of vapor signature with chest computed tomography scan findings. Laryngoscope 2004, 114 (10), 1701–1705 DOI: 10.1097/00005537-200410000-00005.

(35)

Hockstein, N. G.; Thaler, E. R.; Lin, Y.; Lee, D. D.; Hanson, C. W. Correlation of pneumonia score with electronic nose signature: A prospective study. Ann. Otol. Rhinol. Laryngol. 2005, 114 (7), 504–508.

(36)

Filipiak, W.; Beer, R.; Sponring, A.; Filipiak, A.; Ager, C.; Schiefecker, A.; Lanthaler, S.; Helbok, R.; Nagl, M.; Troppmair, J.; et al. Breath analysis for in vivo detection of pathogens related to ventilator-associated pneumonia in intensive care patients: a prospective pilot study. J. Breath Res. 2015, 9 (1), 16004 DOI: 10.1088/1752-7155/9/1/016004.

(37)

Gao, J.; Zou, Y.; Wang, Y.; Wang, F.; Lang, L.; Wang, P.; Zhou, Y.; Ying, K. Breath analysis for noninvasively differentiating Acinetobacter baumannii ventilator-associated pneumonia from its respiratory tract colonization of ventilated patients. J. Breath Res. 2016, 10 (2), 27102 DOI: 10.1088/1752-7155/10/2/027102.

(38)

Schnabel, R.; Fijten, R.; Smolinska, A.; Dallinga, J.; Boumans, M.-L.; Stobberingh, E.; Boots, A.; Roekaerts, P.; Bergmans, D.; van Schooten, F. J. Analysis of volatile organic compounds in exhaled breath to diagnose ventilator-associated pneumonia. Sci. Rep. 2015, 5 (17179), 1–10 DOI: 10.1038/srep17179.

(39)

Fowler, S. J.; Basanta-Sanchez, M.; Xu, Y.; Goodacre, R.; Dark, P. M. Surveillance for lower airway pathogens in mechanically ventilated patients by metabolomic analysis of exhaled


Page 30 of 52

Page 31 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


breath: a case-control study. Thorax 2015, 70 (4), 320–325 DOI: 10.1136/thoraxjnl-2014206273. (40)

Van Oort, P. M. P.; De Bruin, S.; Hans, W.; Knobel, H. H.; Schultz, M. J.; Bos, L. D. Exhaled breath metabolomics for the diagnosis of pneumonia in intubated and mechanicallyventilated intensive care unit (ICU)-patients. Int. J. Mol. Sci. 2017, 18 (2), 1–14 DOI: 10.3390/ijms18020449.

(41)

Lin, J.; Li, M.; Xu, W.; Peng, B.; Guo, Z.; Shui, W.; Xin, Y.; Zhang, C. Techniques in Infectious Diseases Identification of Unique Volatile Compounds of Aspergillus fumigatus for Potential Diagnostic Breath Test by HS-SPME and GC-MS. J Immunol Tech Infect Dis 2013, 2 (3), 9–12.

(42)

Koo, S.; Thomas, H. R.; Daniels, S. D.; Lynch, R. C.; Fortier, S. M.; Shea, M. M.; Rearden, P.; Comolli, J. C.; Baden, L. R.; Marty, F. M. A Breath Fungal Secondary Metabolite Signature to Diagnose Invasive Aspergillosis. Clin. Infect. Dis. 2014, 59 (12), 1733–1740 DOI: 10.1093/cid/ciu725.

(43)

Gerritsen, M. G.; Brinkman, P.; Escobar, N.; Bos, L. D.; de Heer, K.; Meijer, M.; Janssen, H.-G.; de Cock, H.; Wösten, H. A. B.; Visser, C. E.; et al. Profiling of volatile organic compounds produced by clinical Aspergillus isolates using gas chromatography–mass spectrometry. Med. Mycol. 2017, 1–4 DOI: 10.1093/mmy/myx035.

(44)

Chambers, S. T.; Syhre, M.; Murdoch, D. R.; McCartin, F.; Epton, M. J. Detection of 2pentylfuran in the breath of patients with Aspergillus fumigatus. Med. Mycol. 2009, 47 (5), 468–476 DOI: 10.1080/13693780802475212.

(45)

De Heer, K.; Van Der Schee, M. P.; Zwinderman, K.; Van Den Berk, I. A. H.; Visser, C. E.; Van Oers, R.; Sterk, P. J. Electronic nose technology for detection of invasive pulmonary aspergillosis in prolonged chemotherapy-induced neutropenia: A proof-of-principle study. J. Clin. Microbiol. 2013, 51 (5), 1490–1495 DOI: 10.1128/JCM.02838-12.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(46)

Phillips, M.; Cataneo, R. N.; Condos, R.; Ring Erickson, G. A.; Greenberg, J.; La Bombardi, V.; Munawar, M. I.; Tietje, O. Volatile biomarkers of pulmonary tuberculosis in the breath. Tuberculosis 2007, 87 (1), 44–52 DOI: 10.1016/j.tube.2006.03.004.

(47)

Phillips, M.; Basa-Dalay, V.; Bothamley, G.; Cataneo, R. N.; Lam, P. K.; Natividad, M. P. R.; Schmitt, P.; Wai, J. Breath biomarkers of active pulmonary tuberculosis. Tuberculosis 2010, 90 (2), 145–151 DOI: 10.1016/j.tube.2010.01.003.

(48)

Phillips, M.; Basa-Dalay, V.; Blais, J.; Bothamley, G.; Chaturvedi, A.; Modi, K. D.; Pandya, M.; Natividad, M. P. R.; Patel, U.; Ramraje, N. N.; et al. Point-of-care breath test for biomarkers of active pulmonary tuberculosis. Tuberculosis 2012, 92 (4), 314–320 DOI: 10.1016/j.tube.2012.04.002.

(49)

Syhre, M.; Manning, L.; Phuanukoonnon, S.; Harino, P.; Chambers, S. T. The scent of Mycobacterium tuberculosis - Part II breath. Tuberculosis 2009, 89 (4), 263–266 DOI: 10.1016/j.tube.2009.04.003.

(50)

Zetola, N. M.; Modongo, C.; Matsiri, O.; Tamuhla, T.; Mbongwe, B.; Matlhagela, K.; Sepako, E.; Catini, A.; Sirugo, G.; Martinelli, E.; et al. Diagnosis of pulmonary tuberculosis and assessment of treatment response through analyses of volatile compound patterns in exhaled breath samples. J. Infect. 2016, 67 (4), 367–376 DOI: 10.1016/j.jinf.2016.12.006.

(51)

Centers for Disease Control and Prevention. The 2009 H1N1 Pandemic: Summary Highlights, April 2009-April 2010 https://www.cdc.gov/h1n1flu/cdcresponse.htm.

(52)

Phillips, M.; Cataneo, R. N.; Chaturvedi, A.; Danaher, P. J.; Devadiga, A.; Legendre, D. a; Nail, K. L.; Schmitt, P.; Wai, J. Effect of influenza vaccination on oxidative stress products in breath. J. Breath Res. 2010, 4 (2), 26001 DOI: 10.1088/1752-7155/4/2/026001.

(53)

Mashir, A.; Paschke, K. M.; van Duin, D.; Shrestha, N. K.; Laskowski, D.; Storer, M. K.; YenLieberman, B.; Gordon, S. M.; Aytekin, M.; Dweik, R. A. Effect of the influenza A (H1N1) live


Page 32 of 52

Page 33 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


attenuated intranasal vaccine on nitric oxide (FENO) and other volatiles in exhaled breath. J. Breath Res. 2011, 5 (3), 37107 DOI: 10.1088/1752-7155/5/3/037107. (54)

Sibila, O.; Garcia-Bellmunt, L.; Giner, J.; Merino, J. L.; Suarez-Cuartin, G.; Torrego, A.; Solanes, I.; Castillo, D.; Valera, J. L.; Cosio, B. G.; et al. Identification of airway bacterial colonization by an electronic nose in Chronic Obstructive Pulmonary Disease. Respir. Med. 2014, 108 (11), 1608–1614 DOI: 10.1016/j.rmed.2014.09.008.

(55)

Shafiek, H.; Fiorentino, F.; Merino, J. L.; López, C.; Oliver, A.; Segura, J.; De Paul, I.; Sibila, O.; Agustí, A.; Cosío, B. G. Using the electronic nose to identify airway infection during COPD exacerbations. PLoS One 2015, 10 (9), 1–16 DOI: 10.1371/journal.pone.0135199.

(56)

Kramer, R.; Sauer-Heilborn, A.; Welte, T.; Guzman, C. A.; Höfle, M. G.; Abraham, W.-R. A rapid method for breath analysis in cystic fibrosis patients. Eur. J. Clin. Microbiol. Infect. Dis. 2014, 34 (4), 745–751 DOI: 10.1007/s10096-014-2286-5.

(57)

White, I. R.; Willis, K. a; Whyte, C.; Cordell, R.; Blake, R. S.; Wardlaw, A. J.; Rao, S.; Grigg, J.; Ellis, A. M.; Monks, P. S. Real-time multi-marker measurement of organic compounds in human breath: towards fingerprinting breath. J. Breath Res. 2013, 7 (1), 17112 DOI: 10.1088/1752-7155/7/1/017112.

(58)

Barker, M.; Hengst, M.; Schmid, J.; Buers, H.-J.; Mittermaier, B.; Klemp, D.; Koppmann, R. Volatile organic compounds in the exhaled breath of young patients with cystic fibrosis. Eur. Respir. J. 2006, 27 (5), 929–936 DOI: 10.1183/09031936.06.00085105.

(59)

Shestivska, V.; Nemec, A.; Dřevínek, P.; Sovová, K.; Dryahina, K.; Spaněl, P. Quantification of methyl thiocyanate in the headspace of Pseudomonas aeruginosa cultures and in the breath of cystic fibrosis patients by selected ion flow tube mass spectrometry. Rapid Commun. Mass Spectrom. 2011, 25 (17), 2459–2467 DOI: 10.1002/rcm.5146.

(60)

Enderby, B.; Smith, D.; Carroll, W.; Lenney, W. Hydrogen cyanide as a biomarker for



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Pseudomonas aeruginosa in the breath of children with cystic fibrosis. Pediatr. Pulmonol. 2009, 44 (2), 142–147 DOI: 10.1002/ppul.20963. (61)

Gilchrist, F. J.; Bright-Thomas, R. J.; Jones, A. M.; Smith, D.; Španěl, P.; Webb, A. K.; Lenney, W. Hydrogen cyanide concentrations in the breath of adult cystic fibrosis patients with and without Pseudomonas aeruginosa infection. J. Breath Res. 2013, 7 (2), 26010 DOI: 10.1088/1752-7155/7/2/026010.

(62)

Gilchrist, F. J.; Belcher, J.; Jones, A. M.; Smith, D.; Smyth, A. R.; Southern, K. W.; Spanel, P.; Webb, A. K.; Lenney, W. Exhaled breath hydrogen cyanide as a marker of early Pseudomonas aeruginosa infection in children with cystic fibrosis. ERJ Open Res. 2015, 1 (2), 44-2015-44– 2015 DOI: 10.1183/23120541.00044-2015.

(63)

Dummer, J.; Storer, M.; Sturney, S.; Scott-Thomas, A.; Chambers, S.; Swanney, M.; Epton, M. Quantification of hydrogen cyanide (HCN) in breath using selected ion flow tube mass spectrometry—HCN is not a biomarker of Pseudomonas in chronic suppurative lung disease. J. Breath Res. 2013, 7 (1), 17105 DOI: 10.1088/1752-7155/7/1/017105.

(64)

Whiteson, K. L.; Meinardi, S.; Lim, Y. W.; Schmieder, R.; Maughan, H.; Quinn, R.; Blake, D. R.; Conrad, D.; Rohwer, F. Breath gas metabolites and bacterial metagenomes from cystic fibrosis airways indicate active pH neutral 2,3-butanedione fermentation. ISME J. 2014, 8 (6), 1247– 1258 DOI: 10.1038/ismej.2013.229.

(65)

Scott-Thomas, A. J.; Syhre, M.; Pattemore, P. K.; Epton, M.; Laing, R.; Pearson, J.; Chambers, S. T. 2-Aminoacetophenone as a potential breath biomarker for Pseudomonas aeruginosa in the cystic fibrosis lung. BMC Pulm. Med. 2010, 10 (1), 56 DOI: 10.1186/1471-2466-10-56.

(66)

Robroeks, C. M. H. H. T.; van Berkel, J. J. B. N.; Dallinga, J. W.; Jobsis, Q.; Zimmermann, L. J. I. U. C. J. I.; Hendriks, H. J. E. A. N. J. E.; Wouters, M. F. M.; van der Grinten, C. P. M.; van de Kant, K. D. G.; van Schooten, F.-J. J.; et al. Metabolomics of Volatile Organic Compounds in


Page 34 of 52

Page 35 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Cystic Fibrosis Patients and Controls. Pediatr. Res. 2010, 68 (1), 75–80 DOI: 10.1203/PDR.0b013e3181df4ea0. (67)

Neerincx, A. H.; Geurts, B. P.; van Loon, J.; Tiemes, V.; Jansen, J. J.; Harren, F. J. M.; Kluijtmans, L. A. J.; Merkus, P. J. F. M.; Cristescu, S. M.; Buydens, L. M. C.; et al. Detection of Staphylococcus aureus in cystic fibrosis patients using breath VOC profiles. J. Breath Res. 2016, 10 (4), 46014 DOI: 10.1088/1752-7155/10/4/046014.

(68)

Joensen, O.; Paff, T.; Haarman, E. G.; Skovgaard, I. M.; Jensen, P.; Bjarnsholt, T.; Nielsen, K. G. Exhaled breath analysis using electronic nose in cystic fibrosis and primary ciliary dyskinesia patients with chronic pulmonary infections. PLoS One 2014, 9 (12), 1–15 DOI: 10.1371/journal.pone.0115584.

(69)

Kamboures, M. A.; Blake, D. R.; Cooper, D. M.; Newcomb, R. L.; Barker, M.; Larson, J. K.; Meinardi, S.; Nussbaum, E.; Rowland, F. S. Breath sulfides and pulmonary function in cystic fibrosis. Proc. Natl. Acad. Sci. U. S. A. 2005, 102 (44), 15762–15767 DOI: 10.1073/pnas.0507263102.

(70)

Syhre, M.; Scotter, J. M.; Chambers, S. T. Investigation into the production of 2-Pentylfuran by Aspergillus fumigatus and other respiratory pathogens in vitro and human breath samples. Med. Mycol. 2008, 46 (May), 209–215 DOI: 10.1080/13693780701753800.

(71)

Lechner, M.; Karlseder, A.; Niederseer, D.; Lirk, P.; Neher, A.; Rieder, J.; Tilg, H. H. pylori infection increases levels of exhaled nitrate. Helicobacter 2005, 10 (5), 385–390 DOI: 10.1111/j.1523-5378.2005.00345.x.

(72)

Ulanowska, A.; Kowalkowski, T.; Hrynkiewicz, K.; Jackowski, M.; Buszewski, B. Determination of volatile organic compounds in human breath for Helicobacter pylori detection by SPMEGC/MS. Biomed. Chromatogr. 2011, 25 (3), 391–397 DOI: 10.1002/bmc.1460.

(73)

Berna, A. Z.; McCarthy, J. S.; Wang, R. X.; Saliba, K. J.; Bravo, F. G.; Cassells, J.; Padovan, B.;



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Trowell, S. C. Analysis of Breath Specimens for Biomarkers of Plasmodium falciparum Infection. J. Infect. Dis. 2015, 212 (7), 1120–1128 DOI: 10.1093/infdis/jiv176. (74)

Spinhirne, J. P.; Koziel, J. A.; Chirase, N. K. Sampling and analysis of volatile organic compounds in bovine breath by solid-phase microextraction and gas chromatography-mass spectrometry. J. Chromatogr. A 2004, 1025 (1), 63–69 DOI: 10.1016/j.chroma.2003.08.062.

(75)

Peled, N.; Ionescu, R.; Nol, P.; Barash, O.; McCollum, M.; Vercauteren, K.; Koslow, M.; Stahl, R.; Rhyan, J.; Haick, H. Detection of volatile organic compounds in cattle naturally infected with Mycobacterium bovis. Sensors Actuators, B Chem. 2012, 171–172, 588–594 DOI: 10.1016/j.snb.2012.05.038.

(76)

Ellis, C. K.; Stahl, R. S.; Nol, P.; Waters, W. R.; Palmer, M. V.; Rhyan, J. C.; VerCauteren, K. C.; McCollum, M.; Salman, M. D. A pilot study exploring the use of breath analysis to differentiate healthy cattle from cattle experimentally infected with Mycobacterium bovis. PLoS One 2014, 9 (2) DOI: 10.1371/journal.pone.0089280.

(77)

Bayn, A.; Nol, P.; Tisch, U.; Rhyan, J.; Ellis, C. K.; Haick, H. Detection of Volatile Organic Compounds in Brucella abortus-Seropositive Bison. Anal. Chem. 2013, 85 (22), 11146–11152 DOI: 10.1021/ac403134f.

(78)

Purkhart, R.; Köhler, H.; Liebler-Tenorio, E.; Meyer, M.; Becher, G.; Kikowatz, A.; Reinhold, P. Chronic intestinal Mycobacteria infection: discrimination via VOC analysis in exhaled breath and headspace of feces using differential ion mobility spectrometry. J. Breath Res. 2011, 5 (2), 27103 DOI: 10.1088/1752-7155/5/2/027103.

(79)

Bergmann, A.; Trefz, P.; Fischer, S.; Klepik, K.; Walter, G.; Steffens, M.; Ziller, M.; Schubert, J. K.; Reinhold, P.; Köhler, H.; et al. In vivo volatile organic compound signatures of Mycobacterium avium subsp. paratuberculosis. PLoS One 2015, 10 (4), 1–20 DOI: 10.1371/journal.pone.0123980.


Page 36 of 52

Page 37 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(80)

Zhu, J.; Bean, H. D.; Jimenez-Diaz, J.; Hill, J. E. Secondary electrospray ionization-mass spectrometry (SESI-MS) breathprinting of multiple bacterial lung pathogens, a mouse model study. J. Appl. Physiol. 2013, 114 (11), 1544–1549 DOI: 10.1152/japplphysiol.00099.2013.

(81)

Bean, H. D.; Zhu, J.; Sengle, J. C.; Hill, J. E. Identifying methicillin-resistant Staphylococcus aureus (MRSA) lung infections in mice via breath analysis using secondary electrospray ionization-mass spectrometry (SESI-MS). J. Breath Res. 2014, 8 (4), 41001 DOI: 10.1088/17527155/8/4/041001.

(82)

Bean, H. D.; Jimenez-Diaz, J.; Zhu, J.; Hill, J. E. Breathprints of model murine bacterial lung infections are linked with immune response. Eur. Respir. J. 2015, 45 (1), 181–190 DOI: 10.1183/09031936.00015814.

(83)

Zhu, J.; Bean, H. D.; Wargo, M. J.; Leclair, L. W.; Hill, J. E. Detecting bacterial lung infections: in vivo evaluation of in vitro volatile fingerprints. J. Breath Res. 2013, 7 (1), 16003 DOI: 10.1088/1752-7155/7/1/016003.

(84)

Mellors, T. R.; Blanchet, L.; Flynn, J. L.; Tomko, J.; O’Malley, M.; Scanga, C. A.; Lin, P. L.; Hill, J. E. A new method to evaluate macaque health using exhaled breath: A case study of M. tuberculosis in a BSL-3 setting. J. Appl. Physiol. 2017, 122 (3), 695–701 DOI: 10.1152/japplphysiol.00888.2016.

(85)

Zhu, J.; Jiménez-Díaz, J.; Bean, H. D.; Dapthary, N. A.; Aliyeva, M. I.; Lundblad, L. K. A.; Hill, J. E. Robust detection of P. aeruginosa and S. aureus acute lung infections by secondary electrospray ionization-mass spectrometry (SESI-MS) breathprinting: From initial infection to clearance. J. Breath Res. 2013, 7 (3), 395–401 DOI: 10.1088/1752-7155/7/3/037106.

(86)

Barbour, A. G.; Hirsch, C. M.; Ghalyanchi Langeroudi, A.; Meinardi, S.; Lewis, E. R. G.; Estabragh, A. S.; Blake, D. R. Elevated Carbon Monoxide in the Exhaled Breath of Mice during a Systemic Bacterial Infection. PLoS One 2013, 8 (7), 1–9 DOI: 10.1371/journal.pone.0069802.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(87)

Langeroudi, A. G.; Hirsch, C. M.; Estabragh, A. S.; Meinardi, S.; Blake, D. R.; Alan G. Barbour. Elevated Carbon Monoxide to Carbon Dioxide Ratio in the Exhaled Breath of Mice Treated With a Single Dose of Lipopolysaccharide. Open Forum Infect. Dis. 2014, 1 (2), 1–8 DOI: https://doi.org/10.1093/ofid/ofu085.

(88)

Fink, T.; Wolf, A.; Maurer, F.; Albrecht, F. W.; Heim, N.; Wolf, B.; Hauschild, A. C.; Bödeker, B.; Baumbach, J. I.; Volk, T.; et al. Volatile Organic Compounds during Inflammation and Sepsis in Rats. Anesthesiology 2014, 122 (1), 117–126 DOI: 10.1097/ALN.0000000000000420.

(89)

Scott-Thomas, A.; Pearson, J.; Chambers, S. Potential sources of 2-aminoacetophenone to confound the Pseudomonas aeruginosa breath test, including analysis of a food challenge study. J. Breath Res. 2011, 5 (4), 46002 DOI: 10.1088/1752-7155/5/4/046002.

(90)

van der Schee, M. P.; Fens, N.; Brinkman, P.; Bos, L. D.; Angelo, M. D.; Nijsen, T. M.; Raabe, R.; Knobel, H. H.; Vink, T. J.; Sterk, P. J. Effect of transportation and storage using sorbent tubes of exhaled breath samples on diagnostic accuracy of electronic nose analysis. J. Breath Res. 2013, 7 (1), 16002 DOI: 10.1088/1752-7155/7/1/016002.

(91)

Kang, S.; Paul Thomas, C. L. How long may a breath sample be stored for at −80 °C? A study of the stability of volatile organic compounds trapped onto a mixed Tenax:Carbograph trap adsorbent bed from exhaled breath. J. Breath Res. 2016, 10 (2), 26011 DOI: 10.1088/17527155/10/2/026011.

(92)

Dunn, W. B.; Wilson, I. D.; Nicholls, A. W.; Broadhurst, D. The importance of experimental design and QC samples in large-scale and MS-driven untargeted metabolomic studies of humans. Bioanalysis 2012, 4 (18), 2249–2264 DOI: 10.4155/bio.12.204.

(93)

Fiehn, O.; Robertson, D.; Griffin, J.; van der Werf, M.; Nikolau, B.; Morrison, N.; Sumner, L. W.; Goodacre, R.; Hardy, N. W.; Taylor, C.; et al. The metabolomics standards initiative (MSI). Metabolomics 2007, 3 (3), 175–178 DOI: 10.1007/s11306-007-0070-6.


Page 38 of 52

Page 39 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(94)

Herbig, J.; Beauchamp, J. Towards standardization in the analysis of breath gas volatiles. J. Breath Res. 2014, 8 (3), 37101 DOI: 10.1088/1752-7155/8/3/037101.

(95)

Beauchamp, J. D.; Pleil, J. D. Simply breath-taking? Developing a strategy for consistent breath sampling. J. Breath Res. 2013, 7, 42001 DOI: 10.1088/1752-7155/7/4/042001.

(96)

Beauchamp, J. D.; Pleil, J. D.; Dweik, R. A.; Herbig, J.; Risby, T. H. International Association of Breath Research 10th anniversary conference at the Schoenbrunn Palace in Vienna, Austria. J. Breath Res. 2016, 10 (1), 19001 DOI: 10.1088/1752-7155/10/1/019001.

(97)

Basanta, M.; Ibrahim, B.; Douce, D.; Morris, M.; Woodcock, A.; Fowler, S. J. Methodology validation, intra-subject reproducibility and stability of exhaled volatile organic compounds. J. Breath Res. 2012, 6 (2), 26002 DOI: 10.1088/1752-7155/6/2/026002.

(98)

Xu, Y.; Fowler, S. J.; Bayat, A.; Goodacre, R. Chemometrics models for overcoming high between subject variability: applications in clinical metabolic profiling studies. Metabolomics 2013, 10 (3), 375–385 DOI: 10.1007/s11306-013-0616-8.

(99)

Gromski, P. S.; Muhamadali, H.; Ellis, D. I.; Xu, Y.; Correa, E.; Turner, M. L.; Goodacre, R. A tutorial review: Metabolomics and partial least squares-discriminant analysis – a marriage of convenience or a shotgun wedding. Anal. Chim. Acta 2015, 879, 10–23 DOI: 10.1016/j.aca.2015.02.012.

(100) Broadhurst, D. I.; Kell, D. B. Statistical strategies for avoiding false discoveries in metabolomics and related experiments. Metabolomics 2007, 2 (4), 171–196 DOI: 10.1007/s11306-006-0037z. (101) Bos, L. D. J.; Meinardi, S.; Blake, D.; Whiteson, K. Bacteria in the airways of patients with cystic fibrosis are genetically capable of producing VOCs in breath. J. Breath Res. 2016, 10 (4), 47103 DOI: 10.1088/1752-7163/10/4/047103. (102) Neerincx, A. H.; Geurts, B. P.; Habets, M. F. J.; Booij, J. A.; van Loon, J.; Jansen, J. J.; Buydens, L.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

M. C.; van Ingen, J.; Mouton, J. W.; Harren, F. J. M.; et al. Identification of Pseudomonas aeruginosa and Aspergillus fumigatus mono- and co-cultures based on volatile biomarker combinations. J. Breath Res. 2016, 10 (1), 16002 DOI: 10.1088/1752-7155/10/1/016002. (103) Briard, B.; Heddergott, C.; Latgé, J. P. Volatile compounds emitted by Pseudomonas aeruginosa stimulate growth of the fungal pathogen Aspergillus fumigatus. MBio 2016, 7 (2), 1–5 DOI: 10.1128/mBio.00219-16. (104) Pleil, J. D.; Isaacs, K. K. High-resolution mass spectrometry: basic principles for using exact mass and mass defect for discovery analysis of organic molecules in blood, breath, urine and environmental media. J. Breath Res. 2016, 10 (1), 12001 DOI: 10.1088/17527155/10/1/012001. (105) Chambers, S. T.; Bhandari, S.; Scott-thomas, A. M. Y.; Syhre, M. Novel diagnostics: progress toward a breath test for invasive Aspergillus fumigatus. 2011, 49, 54–61 DOI: 10.3109/13693786.2010.508187. (106) Heddergott, C.; Calvo, A. M.; Latgé, J. P. The Volatome of Aspergillus fumigatus. Eukaryot. Cell 2014, 13 (8), 1014–1025 DOI: 10.1128/EC.00074-14. (107) Bhandari, S.; Chambers, S.; Pearson, J.; Syhre, M.; Epton, M.; Scott-Thomas, A. Determining the limits and confounders for the 2-pentyl furan breath test by gas chromatography/mass spectrometry. J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 2011, 879 (26), 2815–2820 DOI: 10.1016/j.jchromb.2011.08.010. (108) Lawal, O.; Ahmed, W. M.; Nijsen, T. M. E.; Goodacre, R.; Fowler, S. J. Exhaled breath analysis: a review of “breath-taking” methods for off-line analysis. Metabolomics 2017, 13 (110) DOI: 10.1007/s11306-017-1241-8. (109) Bos, L. D. J.; Wang, Y.; Weda, H.; Nijsen, T. M. E.; Janssen, A. P. G. E.; Knobel, H. H.; Vink, T. J.; Schultz, M. J.; Sterk, P. J. A simple breath sampling method in intubated and mechanically


Page 40 of 52

Page 41 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ventilated critically ill patients. Respir. Physiol. Neurobiol. 2014, 191, 67–74 DOI: 10.1016/j.resp.2013.11.001. (110) Fuchs, P.; Loeseken, C.; Schubert, J. K.; Miekisch, W. Breath gas aldehydes as biomarkers of lung cancer. Int. J. Cancer 2010, 126 (11), 2663–2670 DOI: 10.1002/ijc.24970. (111) Goodacre, R. Metabolomics of a superorganism. J. Nutr. 2007, 137, 259S–266S. (112) Efron, B. Bootstrap Methods: Another Look at the Jackknife. Ann. Stat. 1979, 7 (1), 1–26 DOI: 10.1214/aos/1176344552. (113) Efron, B.; Gong, G. A leisurely look at the bootstrap, the jackknife, and cross-validation. Am. Stat. 1983, 37 (1), 36–48 DOI: 10.1080/00031305.1983.10483087. (114) Miekisch, W.; Herbig, J.; Schubert, J. K. Data interpretation in breath biomarker research: pitfalls and directions. J. Breath Res. 2012, 6 (3), 36007 DOI: 10.1088/1752-7155/6/3/036007. (115) Zupan, J.; Gasteiger, J. Neural Networks for Chemists: An Introduction; John Wiley & Sons, Inc.: New York, NY, USA, 1993. (116) Metters, J. P.; Kampouris, D. K.; Banks, C. E. Electrochemistry provides a point-of-care approach for the marker indicative of Pseudomonas aeruginosa infection of cystic fibrosis patients. Analyst 2014, 139 (16), 3999–4004 DOI: 10.1039/C4AN00675E. (117) Nakhleh, M. K.; Jeries, R.; Gharra, A.; Binder, A.; Broza, Y. Y.; Pascoe, M.; Dheda, K.; Haick, H. Detecting active pulmonary tuberculosis with a breath test using nanomaterialbased sensors. Eur. Respir. J. 2014, 43 (5), 1522–1524 DOI: 10.1183/09031936.00132613. (118) van Oort, P. M. P.; Nijsen, T.; Weda, H.; Knobel, H.; Dark, P.; Felton, T.; Rattray, N. J. W.; Lawal, O.; Ahmed, W.; Portsmouth, C.; et al. BreathDx – molecular analysis of exhaled breath as a diagnostic test for ventilator–associated pneumonia: protocol for a European multicentre observational study. BMC Pulm. Med. 2017, 17 (1), 1 DOI: 10.1186/s12890-016-0353-7.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

Page 42 of 52

Page 43 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Bibliometric assessment including article attrition and study selection flowchart. A diagram showing the systematic review process. Briefly, articles were filtered through an automated bibliographic database search using keywords. The resultant article list was further refined by way of inclusion/exclusion criteria and additional quality assessment procedures. See supporting information for detailed methodology.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 52

Table 1. Studies investigating infectious disease

Study 34

Hockstein et al.

Publication Year

Subject Size

Target Infection

2004

33

VAP

b

Headspace Analysis Performed?

Instrument

Potential Breath Biomarkers

✘

e-nose

N/A

✘

MCC-IMS

N/A

33

2005

40

Pulmonary (candida and pneumonia infections)

Hockstein et al.

35

2005

50

VAP

✘

e-nose

N/A

Lechner et al.71

2005

25

H. pylori

✘

PTR-MS

hydrogen nitrate, hydrogen cyanide*

TD-GC-MS

1-methyl-naphthalene , 1,4-dimethyl-cyclohexane, and derivatives of heptane and benzene (aligned with in vitro results)

Ruzsanyi et al.

++

Phillips et al.46

Chambers et al.

44

49

Syhre et al.

2007

101

TB a

✓

2009

56

IA

c

✘

2009

20

TB

✘

SPME-GCMS/MS SPME-GCMS/MS

52

2010

33

Influenza A

✓

TD-GC-MS

Phillips et al.47

2010

226

TB

✘

TD-GC-MS

Phillips et al.

2-pentylfuran

++

methyl nicotinate 2,8-dimethyl-undecane, 4,8-dimethyl-undecane, and 2-isopropyl-5-methyl- 1heptanol 3-(1-methylethyl)-oxetane, ++ 4-methyl-dodecane , hexylcyclohexane, bis(3,5,5-trimethylhexyl) phthalate, 1,3,5-trimethyl++ benzene , 3,7-dimethyl++

decane, tridecane , 4,6,8trimethyl-1-nonene, 5-ethyl2-methyl-heptane, 4-methyl1-Hexene 31

Rabis et al.

2011

53

P. aeruginosa

✘

MCC-IMS

N/A isobutane, 2-butanone*, ethyl acetate, styrene*, ethylbenzene*, ethanol*, and butane

↑ nitric oxide, isoprene*

Ulanowska et al.

2011

29

H. pylori

✓

SPME-GCMS

52

2011

9

Influenza A

✘

SIFT-MS

54

Mashir et al.

camphene, l-beta-pinene, ++ 1,3,5-trimethyl-benzene , ++ 1-methyl-naphthalene ,

2012

251

TB

✘

TD-GCSAW

Lin et al.40

2013

18

IA

✓

SPME-GCMS

β-farnesene

De Heer et al.44

2013

46

IA

✘

e-nose

N/A

Phillips et al.

47


++

tridecane , 2-butyl-1octanol*, 4-methyl++ dodecane

Page 45 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Koo et al.41

2014

Fowler et al.38

35

Filipiak et al.

Schnabel et al.37

55

Berna et al.

2015

2015

64

IA

✓

TD-GC-MS

46

Lower Respiratory Tract (H. influenza, S. aureus)

✘

TD-GCTOF-MS

28

2015

100

2015

13

VAP (S. aureus, E. coli, Candida spp., P. aeruginosa) VAP (S. aureus, P. aeruginosa, E. coli, K. pneumoniae, H. influenzae, A. baumannii) Malaria

α-trans-bergamotene, βtrans-bergamotene, a βvatirenene–like sesquiterpene, transgeranylacetone ↑ 3-carene, n-butyric acid 2++ ethylhexyl ester, nonanal , 2,6,11,15-tetramethylhexadecane ++ ↓ ethanol *, 2-methyl ++

cyclopentanone, heptane , and N-cyclohexyl-N0(2hydroxyethyl)thiourea TD-GC-MS

✓

✘

TD-GCTOF-MS

✓

TD-GC-MS

dimethyl sulfide (and multiple microbe specific VOCs) ↑ 2-methylbutane, ++ ++ ethanol *, heptane , ethylbenzene*, carane, tetradecane++, tetradecanal ↓ acetone*, tetrahydrofuran, dodecane allyl methyl sulfide, 1methylthio-propane, (Z)-1methylthio-1-propene, and (E)-1-methylthio-1-propene ++

2016

60

VAP (A. baumannii)

✓

TD-GC-MS

1-undecene, nonanal , decanal*, 2,6,10-trimethyldodecane, 5-methyl-5propyl-nonane, longifolene, tetradecane++, 2-butyl-1octanol*

2016

71

TB

✘

e-nose

N/A

Gerritsen et al.

42

2017

3

IA

✓

TD-GC-MS

Van Oort et al39

2017

93

HAP/CAP d

✘

TD-GC-MS

Gao et al.36

49

Zetola et al.

a

b

c

d

1,3-pentadiene, 2-ethyl-1hexanol, 2-methyl-1propanol acetone, carbon disulfide, 1propanol,2-ethoxy-2-methylpropane, cyclohexene, methylisobutylketone

Tuberculosis, Ventilator-associated Pneumonia, Invasive Aspergillosis, Hospital-Acquired Pneumonia/Community++ Acquired Pneumonia, same VOC identified in another study targeting the same infection (not necessarily same pathogen), *same VOC identified in a study targeting a different infection or subject



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 46 of 52

Table 2. Studies investigating infection within underlying chronic disease. Publication Subject Year Size

Study 71

Kamboures et al.

Target Pathogen

Underlying Disease


Instrument


✘

custom-made system

N/A

2005

43

P. aeruginosa S. aureus

CF

2006

40

P. aeruginosa

CF

✘

GC-FID-MS

↑ pentane ↓ methanol

2008

21

A. fumigatus

CF

✓

SPME-GCMS/MS

2-pentylfuran++

2009

34

P. aeruginosa CF and Asthma

✘

SIFT-MS

hydrogen cyanide++*

2010

105

P. aeruginosa

CF

✘

TD- GC-TOF-MS

N/A

2010

46

P. aeruginosa

CF

✓

SPME-GCMS/MS

2Aminoacetophenone

2011

28

P. aeruginosa

CF

✓

SIFT-MS

methyl thiocyanate

2013

14

P. aeruginosa S. aureus A. fumigatus

CF

✘

PTR-TOF-MS

N/A

2013

20

P. aeruginosa

CF

✘

SIFT-MS

hydrogen cyanide++*

2013

26

P. aeruginosa Bronchiectasis

✘

SIFT-MS

hydrogen cyanide++*

COPD b

✘

e-nose

N/A

CF

✓

SPME-GC-TOFMS

N/A

✘

e-nose

N/A

CF

✘

GC-MS

2,3-butanedione

COPD

✓

e-nose

N/A

a

60

Barker et al. 72

Syhre et al. 62

Enderby et al. 68

Robroeks et al. 67

Scott-Thomas et al. 61

Shestivska et al. 59

White et al. 63

Gilchrist et al. 65

Dummer et al.

S. aureus P. aeruginosa S. pneumoniae H. influenzae M. catarrhalis Candida spp. S. viridans S. epidermidis Corynebacterium spp.

56

Sibila et al.

2014

50

P. aeruginosa S. aureus C. albicans A. xylosoxidans

55

Kramer et al.

2014

11

2014

106

70

Joensen et al.

Whiteson et al.

64

2014

57

Shafiek et al.

2015

185

P. aeruginosa P. aeruginosa S. aureus R. mucilaginosa S. aureus P. aeruginosa S. pneumoniae H. influenzae M. catarrhalis Candida spp. S. viridans Neisseria spp. S. epidermidis

CF and PCD

c


Page 47 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Corynebacterium spp. 64

Gilchrist et al.

2015

233

P. aeruginosa

CF

✘

SIFT-MS

hydrogen cyanide++*

TD-GC-MS

↑ 1,4-pentadiene, ethanol*, acetone*, 2-butanone*, undecane*, and 2methyl naphthalene ↓ 3-hydroxy-2butanone, Isopropyl myristate, hexanal

69

Neerincx et al.

a

b

2016

18

S. aureus

CF

✘

c

++

Cystic Fibrosis, Chronic Obstructive Pulmonary Disease, Primary Ciliary Dyskinesia, same VOC identified in another study targeting the same infection (not necessarily same pathogen), * same VOC identified in a study targeting a different infection or subject



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 48 of 52

Table 3. Studies investigating infection within animal subjects

Publication Subject Year Size

Study

Zoonotic or Model Study

Animal Subfamily

Target Infection


Analysis Method


73

Spinhirne et al.

2004

6

Zoonosis

Bovine

Respiratory Tract

✘

SPME-GC-MS

acetaldehyde and decanal*

2011

18

Zoonosis

Caprine

paraTB a

✘

MCC-DMS

N/A

✘

2,3-Dimethyl-1,3TD-GC-MS Pentadiene, and 1,3and e-nose Dimethylbutyl Cyclohexane

77

Purkhart et al.

74

Peled et al.

b

2012

27

Zoonosis

Bovine

Bovine TB

2013

10

Animal Model

Murine

Respiratory Tract

✓

SESI-MS

N/A

2013

12

Animal Model

Murine

Respiratory Tract

✘

SESI-MS/MS

N/A

2013

86

Animal Model

Murine

Respiratory Tract

✘

SESI-MS

N/A

2013

24

Animal Model

Murine

Borrelia hermsii

✓

GC-FID

↑ Carbon monoxide++

✘

TD-GC-MS and e-nose

2-ethyl-1-hexanol, acetophenone*, benzaldehyde*, heptanal, octanal

✘

SESI-MS

7 VOC m/z fragments

82

Zhu et al. 79

Zhu et al.

85

Zhu et al.

85

Barbour et al.

76

Bayn et al.

2013

38

Zoonosis

Bovine

2014

11

Animal Model

Murine

Brucellosis

80

Bean et al.

MRSA

c

71

Ellis et al.

toluene, styrene*, benzaldehyde*, 2ethyl-1-hexanol, αTD-GC-MS acetophenone*, 1,1dimethyl 2-(1methylethyl) cyclopropane

2014

23

Zoonosis

Bovine

Bovine TB

✘

2014

8

Animal Model

Murine

Sepsis-inducing

✘

GC-FID/TCD

2015

42

Zoonosis

Caprine

paraTB

✘

↓ 2-butanone*, NTME-GCbenzene*, 2-methylMS butanal

Langeroudi et 86

al. Bergmann et 78

al.

87

Fink et al.

Mellors et al.84

2015

40

Animal Model

Murine

Sepsis-inducing

2017

5

Animal Model

Cercopithecine

TB

b


✘

✘

↑ Carbon monoxide++

1-propanol, butanal, acetophenone*, 1,2butandiol, 3MCC-IMS pentanone, acetone*, 2hexanone 1,1’-bicyclohexyl, 2,2TD-GCxGC- dimethylheptane, TOF-MS tridecane 2-Heptanone, Allyl heptanoate, 4-

Page 49 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Methylene-1-(1methylethyl)bicyclo[3.1.0]hexane, 2-Methylbutyl ester butanoic acid, nAmyl isovalerate, oCymene, Trans-αOcimene

a

b

c

++

Paratuberculosis, Tuberculosis, Methicillin-resistant Staphylococcus aureus, same VOC identified in another study targeting the same infection (not necessarily same pathogen), * same VOC identified in a study targeting a different infection or subject



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 50 of 52

Table 4. An overview of breath sampling methods from reviewed studies Sampling Method

Illustration

Strengths •

Syringe

• •

Simple and fast sampling method Relatively low cost High potential for clinical use

Limitations •

• • •

Gas Sampling Bag

• • •

Simple sampling method Commonly used for gas sampling Relatively low cost

• •

•

• Evacuated Steel Container

• • •

• Glass Bulb

Breath Partitioning or Continuous (Purge and Trap)

•

• • •

Low material contamination Robust material Evacuated chamber Limited VOC degradation

Low material contamination Limited VOC degradation

Breath partition sampling Flow and volume control Immediate adsorption


Possible contaminants depending on chamber material Limited control of flow Contaminants from material VOC concentration can change VOCs may react with material Depending on type of bag, not all VOCs may be sampled Risk of water condensation

•

Complex sampling and analysis fittings

•

Potentially hazardous material especially within a clinical environment Risk of water condensation

•

• • • •

High maintenance May require power Adaptation for clinical use required Relatively high cost

Page 51 of 52

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 5. An overview of breath sample analysis methods from reviewed studies Analysis Method

Instrument Sensitivity

Strengths

Limitations •

Gas Chromatography

GC-MS GC-TOF-MS GCxGC-TOFMS GC-MS/MS GC-SAW GC-FID/TCD

• •

a

ppb

•

Large reference library Established methods for VOC analysis High sensitivity

• •

• • •

Proton Transfer Reaction

PTR-MS PTR-TOF-MS

Selected Ion Flow Tube

SIFT-MS SIFT-MS/MS

Ion Mobility-based Spectrometry

MCC-IMS MCC-DMS

b

ppb to ppt

• •

• ppb to ppt

c

ppm to ppb

• • • •

• • Electronic Nose Sensors

Cyranose NA-Nose Quartz Crystal

ppm to ppb

• •

• Secondary Electrospray Ionisation

a

SESI-MS SESI-MS/MS

ppb

parts per billion, b parts per trillion, c parts per million


•

No preconcentration High sensitivity High specificity

No preconcentration High sensitivity High specificity High versatility On-line clinical analysis possible Highly portable On-line clinical analysis possible Relatively low cost Highly adaptable to a clinical environment No preconcentration No soft ionisation reaction required

•

Pre-concentration required Relatively high cost Usually large instruments, and requires regular maintenance by a specialist Relatively slow analysis (per sample) Relies on reaction with reagent ion Relatively high cost Usually large instruments, and requires regular maintenance by a specialist

• •

Relatively high cost Relies on reaction with selected reagent ions

•

Chemical characterisation required Maintenance by specialist required

• • • • •

• • •

Binary classification Limited chemical selectivity Needs prior training Selective sensors are still under development Relatively high cost Characterisation required Requires regular maintenance by a specialist


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For Table of Contents Use Only


Page 52 of 52

Exhaled Volatile Organic Compounds of Infection - ACS Publications

Recommend Documents