As electronic cigarettes (e-cigarette) use has been grown, more concerns have appeared about the exposure of bystanders to secondhand aerosol (SHA) from these devices. This exposure results when the aerosol inhaled by users (firsthand aerosol) is exhaled into the air where it may be breathed by non-users.1 This exposure is worrying because several studies have found toxic and carcinogenic substances in the aerosol (both firsthand and secondhand) generated by e-cigarettes.1,2

Tobacco-specific nitrosamines have been found in the aerosol generated by e-cigarettes3,4 and also in some brands of e-cigarette liquids,5,6 although not all of the studies have detected tobacco-specific nitrosamines in the samples studied,7 a finding that indicates that important differences between brands may exist. Despite the potential presence of toxic substances in these aerosols, exposure to SHA from e-cigarettes has received scant attention according to a recent systematic review.2

For these reasons, we conducted the present study to assess the levels of NNAL, nitrosamine associate with lung cancer,8 in urine samples in a group of non-smokers exposed to SHA from e-cigarette users in their homes under real-life conditions. Moreover, we compared these concentrations with those of non-smokers passive exposed to conventional cigarettes in their homes and with those of non-smokers not exposed to aerosol from e-cigarettes neither conventional cigarettes in their homes.

We conducted a study of passive exposure to electronic and conventional cigarettes in real-use conditions. A tobacco-specific nitrosamine (NNAL) was determined in urine samples from a group of volunteers. We recruited a convenience sample of 55 non-smoking volunteers from different homes, distributed as follows: 25 living at home with conventional smokers, 6 living with e-cigarette users, and 24 from control homes (without the presence of either conventional smokers or e-cigarette users). We also enrolled the 6 e-cigarette users who lived with the non-smoking volunteers. Participants living with e-cigarette users or smokers, and the e-cigarette users, provided self-reported data affirming that their only source of exposure to SHA from e-cigarette or tobacco smoke during the one-week study period was in their home and also confirmed that they did not use any tobacco products. These were the conditions for study inclusion that volunteers (non-smokers and e-cigarette users) agreed to at study enrolment. The self-reported lack of exposure in settings other than their homes (work, leisure time, and transport) was confirmed by a personal interview. In addition, all volunteers, including the 6 e-cigarettes users, declared that they did not use other tobacco products or nicotine replacement therapy during the study period. This pilot study is part of a large project which the main objective is to assess the impact of Spanish legislations and the passive exposure to secondhand smoke of conventional cigarettes at home. We conducted this pilot study due to the increasing popularity of the e-cigarettes and the lack of evidence worldwide about passive exposure to SHA from e-cigarettes. The fieldwork was conducted in Barcelona, Spain, in 2012.

The characteristics of the e-liquids and e-cigarettes used in this study are described in Table 1.

We obtained urine samples for NNAL analysis by liquid chromatography coupled with tandem mass spectrometry with multiple reaction monitoring (LC/MS/MS). We measured the NNAL concentration in picograms per milligram of creatinine (pg/mg) because the NNAL concentrations were adjusted for creatinine. The limit of quantification (LOQ) of NNAL was 0.25 pg/mg in 5mL of urine for a 1mg/mL of creatinine excretion.

Statistically significant differences were found among the three groups of homes in the percentage of volunteers with quantifiable levels of NNAL in urine: 29.2%, 66.7%, and 76% (Table 2). After adjusting for multiple comparisons, differences in NNAL urine concentration were on the borderline of statistical significance between non-smokers in control homes and those exposed to tobacco smoke at home (Table 2).

As shown in Table 3, we quantified traces of NNAL in the urine samples of four out of the six volunteers exposed to SHA from e-cigarette users. Spearman's correlation of the NNAL levels in urine of the e-cigarettes users and the volunteers exposed to SHA from e-cigarette users was 0.943 (p=0.005).

From six e-cigarette users, NNAL was quantifiable in the urine sample (median: 2.86 pg/mg; IQR: 0.36-6.92 pg/mL) (data not shown).

Our results show that there are quantifiable levels of NNAL in urine among bystanders exposed to SHA from e-cigarette users at home. Moreover, we also found a strong correlation between NNAL levels in urine among e-cigarette users and those of the non-smokers exposed to SHA at home from e-cigarette users. A previous study conducted in 698 individuals (532 of them non-smokers) also found levels of NNAL in saliva among non-smokers exposed to secondhand smoke.9

Vogel et al.10 found that the levels of NNAL in urine (without adjusting for creatinine) ranged from 6.27 to 12.54 pg/mL, among non-smokers exposed to secondhand smoke from conventional cigarettes. In our study, the levels were lower in participants who were passively exposed to e-cigarettes (median: 0.47 pg/mL; range: 0.125-3.2 pg/mL) and to conventional cigarettes (median: 0.52 pg/mL; range: 0.125-8.3 pg/mL). These differences could be explained partially by the differences on the sample size of both studies and the inclusion criteria of our study (at home was the only source of exposure of our non-smokers volunteers and e-cigarette users in the last week). Moreover, other potential explanation could be the difference in the duration and intensity of exposure among studies, and the measurement bias inter-laboratory. Although the levels of NNAL among non-smokers exposed passively to e-cigarettes are very low8 there is no safety level of exposure and the risk increases with the intensity and duration of exposure.11,12

The average NNAL concentration in urine (unadjusted for creatinine) of smokers is around 300 pg/mL.13 We also found levels of NNAL in users of electronic cigarettes (median without adjusting for creatinine: 2.6 pg/mL; range: 0.33-9.7 pg/mL). Compared to smokers, these results are orders of magnitude lower10,13 but this magnitude was similar to those found in e-cigarette users in a previous study.14

Our results should be interpreted with caution due to the small number of volunteers exposed to e-cigarette users at home and the differences brands of e-cigarettes. In addition, although one inclusion criterion was that participants’ only permitted exposure to smoking during the study period was at home, we cannot be sure that participants’ did not receive any additional unreported exposure. In an effort to avoid this limitation all participants did not use nicotine replacement therapy and they were instructed to avoid exposure to e-cigarettes and conventional cigarettes except at home.

In conclusion, we found quantifiable levels of NNAL in urine samples of non-smokers passively exposed to SHA from e-cigarette users. However, these results could be confirmed with more studies.

Cristina Linares Gil.

The corresponding author on behalf of the other authors guarantee the accuracy, transparency and honesty of the data and information contained in the study, that no relevant information has been omitted and that all discrepancies between authors have been adequately resolved and described.

M. Ballbè and J.M. Martínez-Sánchez conceived the study, conducted the fieldwork, prepared the database, analysed the data and drafted the manuscript. M. Ballbè, J.M. Martínez-Sánchez, X. Sureda, M. Fu, R. Pérez-Ortuño, J.A. Pascual, A. Peruga and E. Fernández contributed substantially to the conception, design, and interpretation of data. All authors contributed to the manuscript and approved its final version.

The authors wish to thank the 61 volunteers who kindly collaborated in this study. We also thank Bradley Londres for editing the manuscript.

This project was co-funded by the Instituto de Salud Carlos III, Subdirección General de Evaluación, Government of Spain (PI12/01114, PI12/01119, PI15/00291, PI15/00434) co-funded by ISCIII-Subdirección General de Evaluación and by FEDER funds/European Regional Development Fund (ERDF) — a way to build Europe —. J.M. Martínez-Sánchez is supported by the Ministry of Universities and Research, Government of Catalonia (grant 2017SGR608). M. Fu, M. Ballbè and E. Fernández are supported by the Ministry of Universities and Research, Government of Catalonia (2017SGR139). J.A. Pascual and R. Pérez Ortuño are supported by the Ministry of Universities and Research, Government of Catalonia (grant 2017SGR138). E. Fernández is supported by the Instituto de Salud Carlos III, Government of Spain (INT16/00211 and INT17/00103), co-funded by the European Regional Development Fund (FEDER).

None.