Tobacco cheats caught by smoking gun of VOCs

Skip to Navigation


  • Published: Jan 1, 2018
  • Author: Ryan De Vooght-Johnson
  • Channels: Gas Chromatography
thumbnail image: Tobacco cheats caught by smoking gun of VOCs

Smokers have particular VOCs in their breath

Smoking is a major cause of serious ill health and early death. Unfortunately, patients are not always entirely honest about whether or not they are smoking, so objective methods are required to determine smoking habits; one technique is to examine the VOCs in smokers’ breath. The breath of smokers tends to be higher in compounds such as carbon monoxide, toluene, xylene and benzene (the latter is known to be a human carcinogen); however, all of these compounds may be present for other reasons, such as nearby sources of pollution or (in the case of toluene and xylene) the use of solvent-based products in the home or workplace. It is therefore hard to select one particular compound that is an unambiguous marker for smoking.

The Lecce researchers decided to compare the overall GC profiles of smokers and non-smokers using statistical methods, such as the Mann–Whitney test. A group of 16 smokers and 10 non-smokers was examined, the smokers being asked to exhale into bags first thing in the morning (before any cigarettes), 1 hour after their first cigarette and 1 hour after their last cigarette of the day. The non-smokers exhaled into bags first thing in the morning to give control samples.

SPE and GC used to give profiles of smokers’ breath

Samples of exhaled breath in bags were placed on a 38 °C heating plate in order to evaporate condensation; the compounds present were then taken up onto a Supelco carboxen/polydimethylsiloxane (CAR/PDMS) solid-phase extraction (SPE) fibre. Bromobenzene was added as internal standard.

GC was carried out using an Agilent 6890N instrument, fitted with an Agilent DB-WAX capillary column. The oven temperature was programmed as follows: 40 °C for 5 min, 40–140 °C at 3 °C/min, 140 °C for 10 min, 140–230 °C at 10 °C/min and 230 °C for 3 min. An Agilent 5973 mass spectrometer was employed with electron ionisation (EI). The instrument was used in full scan mode, with the peaks being identified by reference to the NIST 98 spectral library.

In all, 83 compounds were identified in the breath samples, both hydrocarbons and various species containing oxygen or nitrogen (alcohols, ketones, heterocycles, etc.). The breath of smokers was rich in organic compounds after their first or last cigarette, while their breath first thing in the morning contained fewer such species, being closer to that of non-smokers. A Mann–Whitney analysis was implemented to detect markers for smoking (regardless of whether or not the subjects had already smoked that day). This statistical method showed that three compounds, toluene, pyridine and pyrrole, were statistically significant signs of smoking. However, carrying out a probit regression using these three compounds showed that they were not in practice very accurate markers for distinguishing smokers and non-smokers since the analysis gave many false positives, incorrectly identifying non-smokers as smokers.

A Mann–Whitney analysis was also carried out to compare the difference between smokers who had not yet smoked that day to those who had. Seven compounds were found at statistically significantly higher levels in the latter group: pyridine, pyrrole, toluene, benzene, 2-pentanone, 2-butanone and 1-methyldecylamine. A probit regression was carried out using these seven compounds and this time the predictive power was good. It could thus be determined whether or not a smoker had smoked during a particular day.

New GC method helps determine compliance with ‘no smoking’ regime

The new method using GC-MS and probit regression enables medical staff to determine whether a smoker had recently smoked or not. However, further work is needed to give a clearer distinction between those who have truly quit smoking and those who are merely pretending to have done so. The authors state that the work has to be extended to larger samples of people in order to obtain more statistically powerful methods.

Related Links

Biomedical Chromatography, 2017, Early View Paper. Capone et al. Chromatographic analysis of VOC patterns in exhaled breath from smokers and nonsmokers.

Wikipedia, Mann-Whitney U Test

Wikipedia, Probit Model

Article by Ryan De Vooght-Johnson

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.

Follow us on Twitter!

Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Copyright Information

Interested in spectroscopy? Visit our sister site

Copyright © 2019 John Wiley & Sons, Inc. All Rights Reserved