Optimization of human breath sampling for VOC analysis

Hugo Knobel, Philips (NL)

In literature many different volatile organic compounds (VOCs) have been associated with diseases [1]. The particular VOCs range from very volatile molecules, such as ethane, acetone, acetaldehyde [2], to very heavy molecules, such as octadecane [3]. In setting up new clinical trials one would like to maximize the chances on finding associated VOC biomarkers. Therefore one would like to design a sampling method which quantitatively captures a large range of VOCs.

To trap over a wide volatility range, multi-bed adsorbent tubes need to be used. Trapping and recovery for a broad volatility range of VOCs was studied for various polymer and carbon adsorbents. The influence of humidity on trapping was evaluated, since quantitative sampling is severely reduced at higher relative humidity (RH).
Break through volumes for individual compounds were determined for various sorbents under dry (0% RH) and wet (100% RH at RT) conditions using calibration gases and subsequently tested using breath.
Thermal desorption parameters were optimized for trapping/releasing n-C2 to n-C15.
Since acetone and isoprene in breath have a much higher concentration than most volatiles, separation was optimized to separate acetone and isoprene from other volatiles.
Finally, sensitivity of the method and repeatability for calibration standards and breath VOCs were determined.

This resulted in the selection of a multi-bed sorbent tube optimized for breath samples with the following parameters:
§ Quantitative Trapping and Release of nC4-nC15 § Tested for RH 100% at 220C (including breath samples from Tedlar bag) § Maximum Safe Sample Volume § Separation of other volatiles from acetone and isoprene § Acetone and Isoprene cannot be quantitatively determined with this method due to overloading MS and sorption tube. Acetone and Isoprene are amongst the most abundant human VOCs. They are often studied and generally not the target of new clinical trials.
With the optimized sorbents tubes and chromatographic method, a storage experiment was conducted using breath (from a volunteer). Samples were stored under wet and dry conditions for 0, 1, 2, 4, 8 and 26 weeks respectively. The results show a clear trend on the influence of moisture on the stability of breath samples in a sorbent tube and thus starting points to obtain an optimal sampling method for a clinical study. Drying immediately after sampling is preferred, with storage ≤ 60 days. The approach using unbiased clustering can also be used to test for other parameters on the stability of breath samples during storage (e.g. temperature).
Future work will include identification of unstable compounds and devising optimal sampling strategies.

[1] Biomed Chromatogr. 2007 Jun; 21(6):553-66. Human exhaled air analytics: biomarkers of diseases. Buszewski B, Kesy M, Ligor T, Amann A.
[2] Chem. Rev.2012, 112, 5949-5966, Volatile Organic Compounds of Lung Cancer and Possible Biochemical Pathways, Hakim, M. et al.
[3] J Breath Res. 2014 Sep; 8 (3):034001. The human volatilome: volatile organic compounds (VOCs) in exhaled breath, skin emanations, urine, faeces and saliva. Amann A, Costello Bde L, Miekisch W, Schubert J, Buszewski B, Pleil J, Ratcliffe N, Risby T.


Pen Profile

Hugo Knobel received his Masters degree in Chemical Engineering from the Eindhoven University of Technology in the Netherlands in 1994, and is currently a Scientist at Philips Lighting. His research interests relate to separation techniques coupled to mass spectrometry (GC-MS & LC-MS) especially for exhaled breath metabolomics with respect to respiratory disease management, in particular unobtrusive, non-invasive monitoring techniques.


+31 (0)15 2 690 448