Tobacco smoking is a risk factor for neoplasms such as lung cancer, chronic obstructive pulmonary disease and cardiovascular diseases, among others.1

In most of these diseases, a high degree of oxidative stress occurs principally due to the amount of free radicals and reactive oxygen and nitrogen species originating in tobacco smoke. Oxidative stress causes irreversible structural and functional cell damage, including changes in essential proteins, lipid peroxidation, DNA strand breakage and nitrogenous base modification, a large increase in intracellular free calcium and in some cases, apoptosis or necrosis.2,3

Of all the changes caused by the increase in oxidative stress, lipid peroxidation generates a large number of metabolites from oxidation reactions catalyzed by enzyme systems associated with cytochrome P450.4,5 These include, among others, alkanes, alkenes, aldehydes and carboxylic acids, all of low molecular weight and volatile, known as volatile organic compounds (VOC), which can be excreted via the airways. The presence of these substances in exhaled air indicates oxidative stress in the airways and lungs; for example, linear aldehydes, such as hexanal, heptanal and nonanal compounds are end metabolites of the lipid peroxidation process of fatty acids ω3 and ω6, principal components of the phospholipids that form part of the cell membranes.6

The aim of this study was to determine whether smoking influences the production of VOC in exhaled air as a result of oxidative stress in a clinically healthy population. Other compounds producing oxidative stress in biological macromolecules were not studied.

The characteristics of the study population are listed in Table 1. The sample consists of 89 subjects, all healthy volunteers, with a mean age of 49.3 (9.5) years. With regard to the gender distribution, 47% were men and 53% women. In all, 39% were non-smokers, 34% smokers and 27% ex-smokers, with a tobacco consumption of 30.9 (18.6) packs per year for the smokers and 26.5 (21.6) packs per year for ex-smokers.

The values obtained were classified in 3 categories: (a) quantified, (b) unquantified and (c) undetected, according to the criteria defined in the section Patients and methods. The frequency distribution is indicated in Table 2. Standardized values of VOC concentrations deviated from a normal distribution.

Table 3 shows the median values (IQR) for each of the VOC analyzed for each group. Statistically significant differences were observed between the 3 groups for the nonanal compound only: non-smokers 0.160 (0.160–0.215); ex-smokers 0.160 (0.160–5.650); and smokers 2.050 (0.160–10.25) (P=.04). Fig. 1 shows the P-values, based on the Mann Whitney test, comparing the groups two by two.

Fig. 1.

Graph of nonanal median and interquartile range in the three study groups. P-values based on the Mann Whitney test.

(0.04MB).

Data scatter in ex-smokers and smokers is very high, so the values of the first group are included in the results of the second. This was the starting point for pooling tobacco users into a single group (smokers and ex-smokers) for comparison with the group of non-smokers. The comparative study of the data obtained between these two groups also shows statistical significance for the nonanal compound (Mann Whitney test, P=.019). No differences are observed in the other VOC among the groups mentioned.

A qualitative analysis of detection/no detection of VOC was carried out according to the limits of detection and quantification (LOD and LOQ) established and defined above in the section Patients and methods. The frequency distribution in percentage and P-values calculated with a χ2 test are given in Table 4. Once again, the nonanal compound was the only one that showed statistically significant differences among the groups (non-smokers, 20%; smokers, 53%; ex-smokers, 29%), P=.016.

Since the nonanal compound is associated in our sample with tobacco use, we decided to study the relationship between the amount of tobacco smoked (expressed in packs per year) and the amount of nonanal exhaled. After applying Spearman's linear correlation test, no correlation was observed between the amount of tobacco consumed and the amount of nonanal exhaled (P=.823), nor was any relation found between the nonanal compound and age or sex (males: r=0.05; P=.753 and females: r=0.081; P=.588).

This study evaluates the differences in the VOC concentration found in exhaled air according to tobacco smoking in our study population, consisting exclusively of healthy volunteers.

According to the results, the only compound associated with tobacco use is nonanal. It is known to be of biological origin, a result which is supported by the fact that smoking causes oxidative stress, generating cell destruction and the appearance of VOC. This is an important consideration if nonanal compounds are to be used as a marker for this phenomenon. The finding of nonanal is independent of age, gender and amount of tobacco smoked, but is related with the status of being a smoker or ex-smoker or never having smoked. It is important to observe that no quantitative correlation was observed between tobacco smoked and nonanal exhaled. These findings were not found in the literature consulted, i.e. the appearance of nonanal depended on the status of being a smoker, but was independent of the amount of tobacco smoked. This compels us to think about the persistence of the cell damage produced by tobacco use, in which a high level of oxidative stress is sustained, even if the subject has quit smoking a long time previously. This sustained cell damage may lead to inflammatory disease or tumor, depending on the cytochrome P450 polymorphism (CYP2E1, CYP1A1),8 which should be the focus of further research.

A review of the literature shows that the majority of studies on VOC in exhaled air are directed toward the search for markers that allow populations with lung cancer to be differentiated from other non-tumor groups. In these studies, the patient groups were distributed according to smoking habit, pooling, on the one hand, active smokers and on the other, ex-smokers and non-smokers, or without making any clear reference to their status as smokers or ex-smokers. Although VOC determinations were investigated in smokers, ex-smokers and non-smokers in the control group in many of the studies, the results always referred to the findings with regard to lung cancer and chronic obstructive pulmonary disease.9–11 Very few publications however compare VOC in exhaled air in smokers and non-smokers exclusively in a healthy population,12 allowing the evaluation of the importance of exposure to tobacco smoke as a factor in increased oxidative stress. Chambers et al. opened an avenue with the detection of VOC in blood, but restricted themselves to the study of aromatic hydrocarbons in tobacco smoke.13

A wide range of compounds are described in the literature, ranging from alkanes, aromatic hydrocarbons, other hydrocarbons, alcohols, carbonyl compounds and carboxylic acids to many other families of compounds. Studies such as those of Phillips et al.9,14,15 and Poli et al.,10 which showed high sensitivity and specificity using combinations of different hydrocarbons, were later criticized for not being able to explain the metabolic origin of the proposed compounds. A critical analysis of the results of the studies cited above by Hórvath et al.16 is of interest in this regard.

At present, there is a tendency toward the study of compounds oxygenated as a result of the peroxidation process of the cell components. Various authors, in the most recent papers, already include aldehydes and ketones, among others, as possible markers of oxidative stress.17–22 Our study focused on aldehydes and carboxylic acids (as described in the section Patients and methods), due to the high incidence of these 2 groups of compounds in many determinations previously performed.23 These compounds, which are of known origin, are intermediate and end metabolites of lipid peroxidation processes of the fatty acids that form part of the cell membranes. For example, hexanal, heptanal and nonanal are products of the oxidation of fatty acids ω3 and ω6 (polyunsaturated fatty acids [PUFA]), principal components of phospholipids.6 With respect to propanoic acid, this is a natural anti-inflammatory, and nonanoic acid is the result of the complete oxidation of aldehyde in a very oxidizing environment.

It should be remembered that VOC not only have a systemic origin but they may also be exogenous, originating from ambient air or tobacco smoke. Distinguishing compounds of endogenous origin from those of exogenous origin is fundamental for this type of study, in which the collection and handling of the sample before analysis is of utmost importance. Various methodologies have been proposed in this respect. Phillips et al.9,12,14,15 proposed a simple mathematical calculation, subtracting the ambient concentrations from the normalized concentrations in exhaled air. This approach has been questioned, as it does not explain the complexity of the pulmonary absorption and subsequent exhalation of exogenous compounds after these have been metabolized in the form of different VOC.16,24,25 Another method proposed was to have the patient breathe pure, VOC-free air for a period of between 4 and 30min.26 Although this was considered a good method, it has not been followed in most studies.

Several authors, including ourselves, carried out sample collection using a Bio-VOC® (as described in the section Patients and methods), which allows the collection of air from the anatomical dead space, with the aim of avoiding contamination of the sample with non-alveolar air as far as possible. To distinguish exogenous compounds from endogenous compounds, a sample of ambient air was obtained at the same time and both samples were compared. Instead of using the approach of Phillips et al.,9,12,14,15 mentioned above, we standardized the concentration to the internal standard, used as a correction factor for the variability of the tube desorption process. Any compound with an exhaled/ambient ratio greater than 1 was assumed to be endogenous, since this gave us the assurance that we were dealing with a compound of endogenous origin. However, we are aware that with this method, we increase the specificity but reduce the sensitivity.

To conclude, the quantitative or qualitative evaluation of nonanal in exhaled air is associated with tobacco use. The other compounds studied do not show any correlation with smoking. There is a lack of standardization in the methodology for sample collection.

This study was funded by the Instituto de Salud Carlos III (PI07/1116), Neumomadrid 2008 and SEPAR 2010 (Registry No.: 881). The authors declare that there is no conflict of interest with any of the companies whose products or services may be discussed in this article.

The authors declare that there was no conflict of interest.