P169 Exhaled volatile organic compound breath analysis in Inflammatory Bowel Disease
L. Hicks*1, J. Huang2, S. Kumar2, S. Powles1, T. Orchard1, G.B. Hanna2, H. Williams1
1Imperial College London, Gastroenterology, London, United Kingdom, 2Imperial College London, Surgery and Cancer, London, United Kingdom
Distinguishing between Crohn's disease (CD) and ulcerative colitis (UC) is important for determining management and prognosis. Selected ion flow tube mass spectrometry (SIFT-MS) may be used to analyse volatile organic compounds (VOCs) in exhaled breath: these may be altered in disease states, and distinguishing breath VOC profiles can be identified . A recent paediatric study used SIFT-MS to distinguish IBD patients from healthy controls . The aim of this pilot study was to identify, quantify and analyse VOCs present in the breath of adult IBD patients and controls, potentially providing insights into disease pathogenesis and complementing current diagnostic algorithms.
SIFT-MS breath profiling of 56 individuals (20 UC, 18 CD and 18 healthy controls) was undertaken. Multivariate analysis included principal components analysis and partial least squares discriminant analysis with orthogonal signal correction (OSC-PLS-DA). Receiver Operator Characteristic (ROC) analysis was performed for each comparison using statistically significant (as calculated by Mann Whitney U test p ≥ 0.05) VOCs.
OSC-PLS-DA modelling was able to distinguish both CD and UC from healthy controls and from one other with good sensitivity and specificity.
“Figure 1. Cross-validated OSC-PLS-DA score plot for 18 CD patients and 20 healthy controls showing the separation achieved with the one-component model”
ROC analysis using combinations of statistically significant VOCs (dimethyl sulphide, hydrogen sulphide, hydrogen cyanide, ammonia, butanal and nonanal) gave integrated areas under the curve (AUC) of 0.86 (CD vs healthy controls), 0.74 (UC vs healthy controls) and 0.83 (CD vs UC).
“Figure 2. Concentrations of (a) dimethyl sulphide and (b) hydrogen sulphide; median concentration and inter-quartile ranges are shown.”
SIFT-MS breath profiling was able to distinguish IBD patients from controls, as well as separate UC from CD, using both multivariate and univariate statistical techniques. The specific VOCs characterising the breath in IBD relate to bacterial dysbiosis and oxidative stress  - both mechanisms implicated in disease pathogenesis.
Predictive abilities of OSC-PCS-DA models created
|Sensitivity (%)||Specificity (%)||Q2 (cross validation parameter) value|
|CD vs HC||94.4||94.4||0.78|
|UC vs HC||90.5||94.4||0.66|
|CD vs UC||88.9||90.0||0.69|
ROC analysis using significant VOCs
|AUC||95% confidence limits|
|CD vs HC||0.864||0.749–0.980|
|UC vs HC||0.742||0.581–0.902|
|CD vs UC||0.828||0.699–0.956|
 Popov TA, (2011), Human exhaled breath analysis, Ann Allergy Asthma Immunol, 451-6
 Patel N, Alkhouri N, Eng K, Cikach F, Mahajan L, Yan C, Grove D, Rome ES, Lopez R, Dweik RA, (2014), Metabolomic analysis of breath volatile organic compounds reveals unique breathprints in children with inflammatory bowel disease: a pilot study, Aliment Pharmacol Ther
 Bos LD, (2013), Volatile metabolites of pathogens: a systematic review, PLoS Pathog