The popularity of the electronic cigarette (e-cigarette) has increased significantly in the past decade. In Belgium, the implementation of the revised European Tobacco Products Directive (TPD) in 2014 marked a turning point for this phenomenon. Prior to this, the use of the e-cigarette was not yet mainstream, as only nicotine-free e-cigarettes were allowed on the market. However, as a result of the legislative changes, nicotine-containing e-cigarettes became freely available on the market and ‘vapeshops’ skyrocketed since then. As the number of e-cigarette users increased, so did the media coverage about the benefits and dangers of the e-cigarette, whether scientifically substantiated or not. To assure public health, scientific research about the safety and quality of the e-cigarette is of great importance. The objective of this PhD thesis was therefore to first determine the chemical composition of the liquids used in e-cigarettes (e-liquid) and next to investigate toxicological aspects of flavouring substances present in e-liquid refills.
First, a comprehensive literature search was performed to obtain an overview about the chemical composition of the e-liquids and the analytical methods used for their detection. Although the analytical methods used in these studies have not always been well validated and thus the results of these studies need to be critically examined; three main problems could be uncovered: (1) the content of nicotine in the e-liquids often does not correspond to the claimed concentration, (2) there is a significant presence of hazardous impurities and contaminants in the e-liquids and (3) food flavourings (diacetyl and acetylpropionyl) which are toxic when inhaled are also present.
In order to analyse the composition of the e-cigarettes and to check the e-liquids for the abovementioned issues, alternative methods were developed for the quantitative determination of nicotine and its related impurities (HPLC-DAD) and for the flavours diacetyl and acetylpropionyl (HS/GC-MS). Also, screening methods were developed for the identification of volatile organic compounds (HS/GC-MS) and the additives taurine (LC-MS/MS) and caffeine (GC-MS). Subsequently, the influence of the revised TPD on the quality of e-liquids available on the Belgian market was investigated using these developed methods. A total of 246 e-liquids were purchased before (2013-2016), during (2016) and after (2017-2018) the implementation of the revised TPD. The samples were examined for the presence of nicotine, nicotine-related impurities, volatile organic compounds, caffeine, taurine and the harmful flavours diacetyl and acetylpropionyl. In general, the legislative changes have had a positive effect on the quality of e-liquids: the results of our study show that the quality of e-liquids has improved following the implementation of the revised TPD. Indeed, in recent years, there have been significantly fewer discrepancies between the effective nicotine content and the claimed concentration. No hazardous volatile organic compounds were found in the 2017-2018 samples compared to the samples before the TPD. In 2018, 5% of the samples contained caffeine, compared to 16% in 2017. The food flavours diacetyl and acetylpropionyl were still present in e-liquids with a sweet, buttery taste such as cake, caramel, popcorn (55% in 2017 compared to 27% in 2018).
Next, as a test case, the risk of inhaling diacetyl present in e-liquids was investigated. An adapted risk assessment methodology for intentional inhalation of substances through the e-cigarette was applied. This exercise showed that there is no risk for systemic toxicity related to diacetyl vapours. However, the risk for local lung toxicity (lung tissue lesions associated with chronic pulmonary bronchiolitis obliterans) could not be excluded in case of repeated exposure to diacetyl through e-cigarette use.
In the final experimental part of the thesis, we focused on the identification of potential genotoxic flavouring substances in e-liquids through the use of non-animal methodologies. As such, 807 flavouring substances were identified in 129 e-liquids using complementary HS-GC MS methods. In a first step, all these substances were screened for genotoxicity using qualitative and quantitative in silico models. In total, potential genotoxicity activity was predicted for 44 flavourings. Based on information from European databases, genotoxicity could be confirmed for five of these flavourings (estragole, safrole, 2-furylmethylketone, 2,5-dimethyl-4-hydroxyl-3(2H)-furanone and transhexanal). Genotoxicity could be excluded for 23 flavourings. For the remaining 16 flavourings, insufficient information on their genotoxicity was present. For four of these flavourings, a commercial standard was available and thus could be tested in vitro using an Ames- and micronucleus test. One of the four substances was only slightly positive in the micronucleus test (b-hellandrene), while for isoledene, 2,3-butanedione and 2,3-pentanedione a clear positive result was obtained in at least one of the two in vitro tests.
Finally, in order to minimise potential health risks imposed by the use of e-cigarettes, some recommendations are suggested to further amend the current e-cigarette legislation.
