You are here

  • Home
  • Archive
  • Volume 22, Issue 3
  • Exposure to different sources of second-hand smoke during pregnancy and its effect on urinary cotinine and tobacco-specific nitrosamine (NNAL) concentrations

Article Text

Article menu
Research paper
Exposure to different sources of second-hand smoke during pregnancy and its effect on urinary cotinine and tobacco-specific nitrosamine (NNAL) concentrations
  1. Constantine I Vardavas1,2,
  2. Eleni Fthenou3,
  3. Evridiki Patelarou1,
  4. Emmanouil Bagkeris1,4,
  5. Sharon Murphy5,
  6. Stephen S Hecht5,
  7. Gregory N Connolly2,
  8. Leda Chatzi1,
  9. Manolis Kogevinas1,4,6,7
  1. 1Department of Social Medicine, Faculty of Medicine, University of Crete, Heraklion, Crete, Greece
  2. 2Center for Global Tobacco Control, Department of Society, Human Development and Health, Harvard School of Public Health, Boston, Massachusetts, USA
  3. 3Department of Histology, Faculty of Medicine, University of Crete, Heraklion, Crete, Greece
  4. 4Center for Research in Environmental Epidemiology (CREAL), Barcelona, Spain
  5. 5Masonic Cancer Center, University of Minnesota, Minneapolis, MN, USA
  6. 6Municipal Institute of Medical Research (IMIM-Hospital del Mar), Barcelona, Spain
  7. 7CIBER Epidemiologia y Salud Pública (CIBERESP), Spain
  1. Correspondence to Dr Constantine I Vardavas, Center for Global Tobacco Control, Department of Society, Human Development and Health, Harvard School of Public Health, Boston, Massachusetts, USA; vardavas{at}hsph.harvard.edu

Abstract

Background To date, no research exists on the role that different sources of exposure to second-hand smoke (SHS) have on 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and nicotine uptake, assessed via urinary 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and cotinine concentrations of non-smoking pregnant women, nor the differences in NNAL concentrations among pregnant women who quit smoking in comparison to those who do not.

Methods As part of the ‘Rhea’ mother childbirth cohort in Crete, Greece, 1317 mother–child pairs were followed-up until delivery, while among a subsample, maternal urine was assessed for its NNAL (n=117) and cotinine concentrations (n=377).

Results Pregnant women who continued to smoke during pregnancy were found to have geometric mean urinary NNAL concentrations of 0.612 pmol/ml, in comparison to the 0.100 pmol/ml of ex-smokers and 0.0795 pmol/ml of non-smokers exposed to SHS. Exposure to SHS in the home was associated with a 4.40 ng/ml increase in urinary cotinine levels, while reported exposure to SHS in cars was associated with an even higher (8.73 ng/ml) increase in cotinine concentrations and was strongly related to NNAL concentrations. Exposure to SHS in the workplace and in public places was also shown to increase cotinine and NNAL concentrations. The NNAL:cotinine ratio was found to be higher among pregnant women who were exposed to SHS but did not smoke (p<0.001).

Conclusions Using cotinine levels as an indicator of NNK, exposure due to SHS during pregnancy leads to an underestimation of exposure to NNK uptake. Moreover, each source of exposure contributed to the increase in cotinine levels, indicating the importance of avoiding SHS exposure from any source.

  • Smoking
  • cessation
  • pregnancy
  • fetal health
  • passive smoking
  • SHS
  • toxicology
  • prevalence
  • environmental tobacco smoke
  • advertising and promotion
  • packaging and labelling
  • electronic nicotine delivery devices
  • public policy
  • cotinine
  • second-hand smoke
  • smoking-caused disease
  • carcinogens
  • harm reduction
  • taxation and price
  • advocacy

https://doi.org/10.1136/tobaccocontrol-2011-050144

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Active and passive smoking during pregnancy are associated with a number of negative outcomes in the developing fetus and young child, including low birth weight, fetal growth restriction, spontaneous abortion and also possibly respiratory regulation.1–3 Tobacco use and exposure to second-hand smoke (SHS) are significant sources of chemicals, toxicants and carcinogens, including nicotine, polycyclic aromatic hydrocarbons, heavy metals and tobacco-specific nitrosamines. Among the latter, the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) has been classified as an International Agency for Research on Cancer (IARC) carcinogen and demonstrated to be a lung carcinogen in every animal model tested.4 ,5 Within the human body, NNK is metabolised by carbonyl reduction to 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL). NNK has been shown to damage DNA, mainly through DNA adduct formation resulting in mutations in growth control genes such as K-ras. Treatment of pregnant hamsters and mice with NNK causes tumours in their offspring in a variety of tissues, demonstrating its transplacental carcinogenicity.4 ,6 ,7

    With the above in mind, it is biologically plausible that children exposed to active and passive smoking in utero may be at an elevated risk for the development of currently acknowledged health outcomes and also may be predisposed to develop chronic diseases and possibly cancer in later life. Research has indicated that within cohort and case–control studies, higher cotinine and NNAL levels among smokers are associated with a higher risk for developing lung cancer in the future, indicating the use of cotinine and NNAL as important biomarkers of lung cancer risk among smokers.8 Moreover, smoking cessation has been shown to lead to a reduction in measured NNAL concentrations within the human body.9–11 To date, little research exists on the relationship between smoking cessation or the role that different sources of exposure to SHS may have on NNAL concentrations, while no research has investigated into the above hypotheses among pregnant women.9–16

    Consequently, the aims of this study were threefold: (1) to evaluate the relationship between urinary cotinine and NNAL levels during pregnancy, (2) to examine the role of smoking, smoking cessation and SHS exposure during pregnancy on cotinine and NNAL concentrations and (3) to assess the individual role of different sources of exposure to SHS on cotinine and NNAL urinary concentrations during pregnancy.

    Methods

    Design of the study

    The ‘Rhea’ study is a mother–child cohort of pregnant women, residents of the prefecture of Heraklion in Crete. All women who became pregnant for 1 year from February 2007 until February 2008 were contacted and asked to participate in the study. During the study period, 1765 eligible women were approached, 1610 (91%) agreed to participate, out of which 1317 (82%) were followed-up until delivery. The first contact was performed around the 12th week of gestation in a face-to-face interview, while the second contact was performed during the third trimester approximately around the 30th week of gestation. During these interviews, information on socio-demographic characteristics, pre-pregnancy body mass index and exposure data was collected. Due to financial restrictions, NNAL concentrations were calculated from urine samples of 117 participants (57 active/ex-smokers and 60 non-smokers), and cotinine was analysed from the urine of 367 non-smoking participants. These participants were selected (convenience sample) from the Rhea cohort based on if they had provided a urine sample and if they had provided other biological matrixes for the analysis of other xenobiotic substances (ie, heavy metals) or biological outcome indexes (ie, DNA damage) that could be subsequently correlated with cotinine concentrations within the framework of the RHEA birth cohort study. A comparison of the socio-demographic and exposure characteristics between the current substudy participants and the entire RHEA birth cohort population is available online (supplementary table 1), through which it was identified that the substudy participants were higher educated and more likely to be Greek. Further information on factors influencing active smoking, passive smoking and socio-demographic characteristics of the entire RHEA cohort can be found elsewhere.17 The study was approved by the corresponding ethical committee, and all participants provided written informed consent.

    Sample handling—cotinine and NNAL analysis

    The majority of the non-smoking participants in our substudy (n=326) provided two urine samples (with inter-cotinine levels strongly correlated, Spearman's ρ=0.78, p<0.001). Based on the cotinine concentrations of these two samples, the geometric mean cotinine levels during pregnancy were calculated for each participant. Among these participants, average cotinine levels were not found to differ by trimester (first trimester cotinine average=18.5 ng/ml vs second trimester cotinine average=18.4 ng/ml vs third trimester cotinine average=17.8 ng/ml). For participants who provided only one sample, this sample was used to assess the women's exposure to SHS during pregnancy (n=41).

    Total cotinine (free plus cotinine N-glucuronide) and total NNAL (NNAL plus its glucuronides) were assessed by liquid chromatography/mass spectrometry and gas chromatography-thermal energy analysis, respectively, as described previously.14 Limits of quantification for total NNAL and total cotinine were 0.01 pmol/ml and 0.5 ng/ml, respectively, while all samples were above the level of quantification in the current analysis. Data on precision (cumulative variance = <10%) and accuracy of these assays have been previously described.14

    Self-reported definition of smoking status and exposure to SHS

    During the analysis, three groups were created based on reported smoking status. Pregnant women were either categorised as a ‘smoker’ if they currently reported any tobacco use at the time of the first interview (around the 12th week of gestation), ‘ex-smokers’ were the women who reported that they smoked during the first weeks of pregnancy but have since quit, while ‘non-smokers’ were defined as the women who reported that they had not smoked cigarettes any time during pregnancy and at least 3 months prior to conception. Women were asked to report SHS exposure during both the first and second interview, while any SHS exposure in one or the other interviews classified them as exposed to that specific source.

    SHS exposure was assessed by a battery of questions. Specifically, participants were requested to report their exposure to SHS in the house by answering yes or no to the following question: “Is there anyone in your house who smokes in your presence?”. Moreover, women were asked to clarify who smoke at home, the number of cigarettes they usually smoke inside and outside the house. Moreover, SHS exposure at work was assessed by answering yes or no to the question “Were you exposed to passive smoking in your workplace during your pregnancy?”. SHS exposure in public places was assessed by asking whether they had gone to cafes, bars or restaurants where they were exposed to SHS during pregnancy. Exposure to SHS in the car was assessed by the question “During this pregnancy have you been in a car where a person was smoking?”, while other sources of SHS exposure were assessed by asking whether they were exposed to SHS during pregnancy from any other source other than those noted above (ie, at a friend's house).

    Among the women who were self-reported as non-smokers, their urinary cotinine levels were used to validate their self-report and a cut-off of 120 ng/ml was designated, based on the distribution of the cotinine. Only women who reported to be non-smokers and whose cotinine levels were at all times below 120 ng/ml were classified in the analyses as non-smokers. Eight self-reported non-smokers were excluded from the analysis and re-categorised as active smokers based on their cotinine levels that were distinctly above the cut-off (cotinine levels of these eight participants were not compatible with SHS exposure levels and were as follows: subject A: 2841 ng/ml; subject B: 2467 and 6096 ng/ml; subject C: 1281 and 920 ng/ml; subject D: 1306 and 686 ng/ml; subject E:508 ng/ml; subject F: 1204 and 392 ng/ml; subject G: 173 ng/ml and subject H: 128 ng/ml).

    Statistical analysis

    For the purpose of our analysis, cotinine and NNAL geometric mean values were calculated and used as their original values did not follow a normal distribution. Statistical hypothesis testing between the geometric means of our biomarkers and maternal and husband's educational status, maternal and husband's ethnicity, parity, place of residence, husband's smoking status, maternal cigarette consumption during pregnancy and self-reported exposure to SHS were performed using Mann–Whitney and Kruskal–Wallis non-parametric tests. Similarly, correlations between continuous variables were performed using the Spearman's Rho correlation coefficient.

    Moreover, NNAL:continine ratios were calculated. The procedure followed was to first calculate the ratio for each woman separately and then to average the ratios. Comparisons between smoker and non-smoker NNAL:cotinine ratios were also performed.

    Based on a recent publication by Lumley et al,18 who supported the validity of using linear regression models even for non-normally distributed data based on the ‘Central Limit Theorem’ (which states that the average of a large number of independent random variables is approximately normally distributed), linear regression analyses were also performed. With the use of linear regression models, we calculated the β-coefficients and the 95% CIs for the associations between self-reported SHS exposure and the geometric means of urinary cotinine and NNAL concentrations among non-smoking pregnant women. Furthermore, all linear regression models were adjusted for socio-demographic factors that were based a priori on the literature review and found, in previous analyses among Rhea study participants, to affect exposure. These included maternal education, husband's education, maternal age, husband's age, place of residence, parity, maternal origin, husband's origin and pre-pregnancy maternal body mass index.1 ,17 All hypothesis testing was conducted assuming a 0.05 significance level and a two-sided alternative hypothesis. The analyses were performed using the statistical package PASW V.19.0.

    Results

    NNAL and cotinine by smoking status

    Urinary NNAL concentrations during pregnancy were strongly dependent on smoking status, with urinary NNAL levels found to be significantly higher among smokers in comparison to non-smokers and ex-smokers (supplementary figure 1). Specifically, pregnant women who continued to smoke during pregnancy were found to have geometric mean urinary NNAL concentrations of 0.612 pmol/ml (95% CI 0.365 to 0.860), in comparison to 0.100 pmol/ml (95% CI 0.056 to 0.1443) among ex-smokers and 0.0795 pmol/ml (95% CI 0.048 to 0.111) among non-smokers. The NNAL:cotinine ratio also differed by smoking status, which was found to be much higher among pregnant women who did not smoke (NNAL:cotinine ratio=0.0076) in comparison to those who did (NNAL:cotinine ratio=0.0013), and this difference was statistically significant (p<0.001). Moreover, as assessed among non-smoking pregnant women (n=112), geometric mean NNAL and cotinine levels were strongly correlated (Spearman's ρ=0.819, p<0.001).

    Socio-demographic/household characteristics and urinary NNAL and cotinine

    Table 1 provides the socio-demographic characteristics of the non-smoking pregnant women of the RHEA study that participated in this substudy, in relation to their urinary geometric mean cotinine and NNAL concentrations. Having a spouse who smokes and the number of cigarettes smoked by the husband in the house per day were associated with higher urinary geometric mean cotinine and NNAL concentrations. According to the analysis, women with lower educated spouses were found to have higher geometric mean cotinine and NNAL concentrations than those with higher educated spouses (geometric mean: 15.38 vs 8.93 ng/ml, p<0.001 and 0.068 vs 0.056 pmol/ml, p=0.012, respectively). Within the applied linear regression analysis (table 2), maternal age was found to statistically effect cotinine levels during pregnancy as each increase in maternal age by 1 year reduced their geometric mean cotinine levels by 0.5 ng/ml (p=0.009).

    View this table:
    Table 1

    Socio-demographic and household factors that influence geometric mean cotinine (n=367) and NNAL (n=60) concentrations of non-smoking pregnant women

    View this table:
    Table 2

    Results of linear regression analysis for the association between socio-demographic factors and geometric mean urinary cotinine (n=362) and geometric mean NNAL (n=58) concentrations among non-smoking pregnant women

    Self-reported exposure and urinary cotinine and NNAL concentrations

    The univariate association between self-reported exposure to SHS and urinary NNAL (n=60) and cotinine (n=367) among non-smoking pregnant women is depicted in table 3. Both NNAL and cotinine concentrations were found to increase in line with the number of sources to which the pregnant woman was exposed to SHS (p<0.001 for cotinine and p=0.009 for NNAL). Our results indicated that those exposed to SHS from more than two sources had higher urinary NNAL concentrations than those exposed to SHS from two or less sources (geometric mean: 0.072 vs 0.049 pmol/ml, p=0.003). Exposure to SHS within a car was strongly related to NNAL and cotinine concentrations (geometric mean: 0.073 vs 0.052 pmol/ml, p=0.033 and 18.12 vs 7.95 ng/ml, p<0.001, respectively), while exposure to SHS in public places was also associated with increased cotinine concentrations (geometric mean: 12.12 vs 8.4 ng/ml, p=0.005).

    View this table:
    Table 3

    The association between self-reported exposure to SHS among non-smoking pregnant women and geometric mean NNAL (n=60) and geometrical mean cotinine (n=367) concentrations

    Based on the above findings, we assessed the alteration in cotinine and NNAL concentrations according to their self-reported exposure to each source, while controlling for their socio-demographic characteristics. As viewed in table 4, self-reported exposure to SHS in the home was associated with a 4.40 ng/ml (95% CI 1.40 to 7.40) increase in urinary cotinine levels, while reported exposure to SHS in cars during pregnancy was associated with an even higher (8.73 ng/ml (95% CI 5.81 to 11.65)) increase in cotinine concentrations. When controlling for socio-demographic characteristics, the associations did not alter to a large extent, with exposure to SHS in the family car and at home found to cause the highest increase in maternal cotinine levels (by 7.01 and 6.02 ng/ml, p<0.001, respectively). Furthermore, it is of interest to note that exposure to SHS in public places and the workplace was associated with an increase in cotinine levels by 3.51 ng/ml (95% CI −0.06 to 7.09, p=0.052) and 2.84 ng/ml (95% CI −0.03 to 5.72, p=0.053), respectively. Similar results were found in regard to NNAL concentrations among those exposed to SHS in public places and the workplace.

    View this table:
    Table 4

    Increase in geometric mean cotinine (n=367) and geometric mean NNAL (n=60) concentrations by self-reported exposure to second-hand smoke among non-smoking pregnant women, before and after controlling for their socio-demographic characteristics

    Discussion

    Main findings

    To our knowledge, this is the first study to assess tobacco-specific carcinogen uptake during pregnancy attributable to SHS exposure and also to investigate the role of different sources of exposure to SHS on NNAL concentrations during pregnancy. Moreover, we identified that women who did quit smoking once pregnant had urinary NNAL concentrations close to that of non-smokers exposed to SHS, confirming that the decrease in NNAL concentrations that is noted after smoking cessation in adults also applies to the period of pregnancy. Moreover, our findings indicate that using cotinine pregnancy concentrations to estimate NNK uptake during pregnancy would lead to a significant underestimation of gestational NNK uptake, and this fact should be taken into account during risk and exposure assessment.

    Experimental evidence has indicated that if daily cigarette smoking is reduced, NNAL concentrations gradually decrease.19 Out of a number of tobacco carcinogens that have been assessed, the washout of NNAL is gradual and has been found to decrease by 92% after 42 days of cessation conversely to other xenobiotics, such as monohydroxybutyl mercapturic acid (MHBMA), s-phenyl mercapturic acid (SPMA), a metabolite of benzene and 2-hydroxyethyl mercapturic acid (HEMA), a metabolite of ethylene oxide, which have much faster washout rates.10 This decrease after smoking cessation during pregnancy is of pertinent interest as animal-based research has identified an association between NNK exposure during pregnancy and the likelihood for the infant to develop paediatric lung disorders, such as bronchopulmonary dysplasia, due to fetal pulmonary neuroendocrine cell regulation.20 However, our results indicated that the ratio between NNAL:cotinine concentrations was much higher among pregnant women exposed to SHS in comparison to pregnant women who were active smokers, similar to the results presented by Benowitz et al 16 among non-pregnant adults, thus confirming that cotinine levels underestimate exposure to the tobacco-specific carcinogen NNK in pregnant women.

    SHS exposure and NNAL–cotinine

    Self-reported exposure to SHS may be inaccurate, and as seen within our study, even non-smokers who reported not to be exposed still had detectable levels of both cotinine and NNAL, a result also found in previous research among Moldovan children who reported not to be exposed to SHS and, however, still had detectable levels of urinary NNAL. It is interesting to note that similar NNAL concentrations among both studies were found in regard to being exposed to household SHS as both identified that household exposure to SHS leads to urinary NNAL concentrations of approximately 0.06 pmol/ml.13

    Exposure to SHS within cars has been a field of growing interest over the past few years; however, to our knowledge, this is the first study to assess a direct increase in cotinine and NNAL concentrations in pregnant women, which can be attributed to reported exposure to SHS in the family car.21 ,22 According to our research, car exposure caused the highest increase in cotinine levels, indicating that the brief but intense exposure to SHS in the car has a strong impact on overall exposure. Such an increase in maternal cotinine levels when exposed to SHS in the car could be attributable to both the intensity of SHS exposure in the confined space of the car as possibly also to the existence of residual tobacco smoke pollution (third-hand smoke) in their cars.21–24 In line with the above, a number of countries have moved forward to ban smoking in cars when a child is present, while current guidelines recommend that cars should be smoke-free.25 Based on our findings, such legislation is justified and could be expanded to cover other vulnerable populations, such as pregnant women.

    Moreover, exposure to passive smoke in public places and workplaces was associated with an increase in cotinine and NNAL levels (albeit borderline non-statistically significant), implying the effect that smoke-free legislation has on maternal and fetal contamination with tobacco-specific carcinogens. A population-based study performed in Italy, before and after the implementation of the smoking ban, indicated that the percentage of women exposed to SHS after the smoking ban had dropped.26 Within Greece, and among the RHEA birth cohort study participants before the implementation of the 2010 smoke-free legislation, 64% of pregnant women were exposed to SHS in public venues, 48% in the workplace and 41% from other undefined sources. It will be of interest to assess the above-noted associations between SHS exposure and NNAL/cotinine concentrations after the enforcement and implementation of the comprehensive ban on smoking.17

    The clearance of nicotine and cotinine during pregnancy is higher than usual with the half-life of cotinine estimated at approximately 9 h. Extrahepatic sites such as the placenta have been hypothesised to be involved in cotinine clearance, while levels of enzymes responsible for cotinine metabolism, notably CYP2A6, are markedly increased.27 Naturally, as both cotinine and NNAL are tobacco-related biomarkers, both were found to be strongly correlated among non-smoking pregnant women. The correlation that we identified between urinary NNAL and cotinine was calculated (r=0.82), which is slightly stronger than the correlation that have been previously identified between the two biomarkers among school children in Moldova (r=0.47), however, very similar within children in the USA (r=0.71) and among adults (r=0.79).13–15

    Our study has a number of strengths, as limited research exists on cotinine versus NNAL uptake, especially during pregnancy, while our sample size is adequate and exposure was classified through both self-report and biochemical evaluation. However, the time frame of SHS exposure could have been assessed more appropriately (ie, asking about exposure over the past few days); moreover, our findings might not be generalisable to the Greek population, as the socio-demographic and exposure characteristics of our study sample were slightly different from that of the entire Rhea cohort; however, this was not the aim of our research hypothesis, which was solely to investigate into the internal relationship between SHS exposure, cotinine and NNAL levels during pregnancy.

    Conclusions

    Our findings indicate that among non-smoking women, cotinine concentrations underestimate NNAL uptake during pregnancy, which could have significant implications during exposure and risk assessment of carcinogen uptake. In addition to the above, the fact that household and car exposure to SHS caused increases in cotinine and NNAL levels indicates the imperative need for the adoption of smoke-free homes and cars. Moreover, as other sources of exposure to SHS, such as that from exposure within public places, were also associated with elevated metabolites of tobacco-specific carcinogens during pregnancy, the complete avoidance and elimination of SHS exposure would be ideal.

    What is already known on this subject

    • Exposure to SHS during pregnancy is known to be a threat to the developing fetus.

    • Using cotinine levels as an indicator of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) exposure uptake due to SHS exposure among adults leads to an underestimation of exposure to NNK uptake.

    What this paper adds

    • Active smokers have higher levels of tobacco-specific carcinogens than non-smokers and ex-smokers, with smoking cessation during early pregnancy found to lead to 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) concentrations similar to that of non-smokers.

    • Using cotinine levels as an indicator of NNK uptake due to SHS exposure leads to an underestimation of NNK uptake during pregnancy, due to the different NNAL:cotinine ratios identified between smoking and non-smoking pregnant women.

    • Exposure to SHS in the family car caused the highest increase in cotinine levels, indicating the importance of regulating smoking in cars with other passengers present.

    • Exposure to SHS is multifactorial, with each source of exposure a cause for an increase in cotinine levels, with the latter found to increase in line with the number of sources of SHS exposure one is exposed to.

    Acknowledgments

    We thank Nicole Thomson for performing the cotinine analysis and Steven Carmella for performing the NNAL analysis. We also thank Anthony Kafatos and Antonis Koutis for their help in participant recruitment and study design.

    References

    1. ↵
      1. Vardavas CI,
      2. Chatzi L,
      3. Patelarou E,
      4. et al
      . Smoking and smoking cessation during early pregnancy and its effect on adverse pregnancy outcomes and fetal growth. Eur J Pediatr 2010;169:741–8.
      OpenUrlCrossRefPubMed
    2. ↵
      US Department of Health and Human Services. The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General. Atlanta, G.A: US Department of Health and Human Services, Center for Disease Control and Prevention, Coordinating Center of Health Promotion, National Center for chronic disease Prevention and health Promotion, Office on Smoking and Health, 2006. http://www.surgeongeneral.gov/library/secondhandsmoke/report/index.html
    3. ↵
      1. Fregosi R,
      2. Pilarski J
      . Prenatal nicotine exposure and development of nicotinic and fast amino acid-mediated neurotransmission in the control of breathing. Respir Physiol Neurobiol 2009;169:1–10.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Hecht SS
      . Progress and challenges in selected areas of tobacco carcinogenesis. Chem Res Toxicol 2008;21:160–71.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      International Agency for Research on Cancer. Smokeless tobacco and tobacco-specific nitrosamines. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Vol. 89. Lyon, France: IARC, 2007.
    6. ↵
      1. Anderson LM,
      2. Hecht SS,
      3. Dixon DE,
      4. et al
      . Evaluation of the transplacental tumorigenicity of the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone in mice. Cancer Res 1989;49:3770–5.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Beebe LE,
      2. Kim YE,
      3. Amin S,
      4. et al
      . Comparison of transplacental and neonatal initiation of mouse lung and liver tumors by N-nitrosodimethylamine (NDMA) and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and promotability by a polychlorinated biphenyls mixture (Aroclor 1254). Carcinogenesis 1993;14:1545–8.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Yuan JM,
      2. Koh WP,
      3. Murphy SE,
      4. et al
      . Urinary levels of tobacco-specific nitrosamine metabolites in relation to lung cancer development in two prospective cohorts of cigarette smokers. Cancer Res 2009;69:2990–5.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Hecht SS,
      2. Carmella SG,
      3. Chen M,
      4. et al
      . Quantitation of urinary metabolites of a tobacco-specific lung carcinogen after smoking cessation. Cancer Res 1999;59:590–6.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Carmella SG,
      2. Chen M,
      3. Han S,
      4. et al
      . Effects of smoking cessation on eight urinary tobacco carcinogen and toxicant biomarkers. Chem Res Toxicol 2009;22:734–41.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Goniewicz ML,
      2. Havel CM,
      3. Peng MW,
      4. et al
      . Elimination kinetics of the tobacco-specific biomarker and lung carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol. Cancer Epidemiol Biomarkers Prev 2009;18:3421–5.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Lackmann GM,
      2. Salzberger U,
      3. Töllner U,
      4. et al
      . Metabolites of a tobacco-specific carcinogen in urine from newborns. J Natl Cancer Inst 1999;91:459–65.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Stepanov I,
      2. Hecht SS,
      3. Duca G,
      4. et al
      . Uptake of the tobacco-specific lung carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone by Moldovan children. Cancer Epidemiol Biomarkers Prev 2006;15:7–11.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Hecht SS,
      2. Carmella SG,
      3. Le KA,
      4. et al
      . 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol and its glucuronides in the urine of infants exposed to environmental tobacco smoke. Cancer Epidemiol Biomarkers Prev 2006;15:988–92.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Hecht SS,
      2. Ye M,
      3. Carmella SG,
      4. et al
      . Metabolites of a tobacco-specific lung carcinogen in the urine of elementary school-aged children. Cancer Epidemiol Biomarkers Prev 2001;10:1109–16.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Benowitz N,
      2. Goniewicz ML,
      3. Eisner MD,
      4. et al
      . Urine cotinine underestimates exposure to the tobacco-derived lung carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone in passive compared with active smokers. Cancer Epidemiol Biomarkers Prev 2010;19:2795–800.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Vardavas CI,
      2. Patelarou E,
      3. Chatzi L,
      4. et al
      . Factors associated with active smoking, quitting and secondhand smoke exposure among pregnant women in Greece. J Epidemiol 2010;20:355–62.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Lumley T,
      2. Diehr P,
      3. Emerson S,
      4. et al
      . The importance of the normality assumption in large public health data sets. Annu Rev Public Health 2002;23:151–69.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Hecht SS,
      2. Murphy SE,
      3. Carmella SG,
      4. et al
      . Effects of reduced cigarette smoking on the uptake of a tobacco-specific lung carcinogen. J Natl Cancer Inst 2004;96:107–15.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Schuller HM,
      2. Jull BA,
      3. Sheppard BJ,
      4. et al
      . Interaction of tobacco-specific toxicants with the neuronal alpha (7) nicotinic acetylcholine receptor and its associated mitogenic signal transduction pathway: potential role in lung carcinogenesis and pediatric lung disorders. Eur J Pharmacol 2000;393:265–77.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Vardavas CI,
      2. Linardakis M,
      3. Kafatos AG
      . Environmental tobacco smoke exposure in motor vehicles: a preliminary study. Tob Control 2006;15:415.
      OpenUrlFREE Full Text
    22. ↵
      1. Rees VW,
      2. Connolly GN
      . Measuring air quality to protect children from secondhand smoke in cars. Am J Prev Med 2006;31:363–8.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Fortmann AL,
      2. Romero RA,
      3. Sklar M,
      4. et al
      . Residual tobacco smoke in used cars: futile efforts and persistent pollutants. Nicotine Tob Res 2010;12:1029–36.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Matt GE,
      2. Quintana PJ,
      3. Hovell MF,
      4. et al
      . Residual tobacco smoke pollution in used cars for sale: air, dust, and surfaces. Nicotine Tob Res 2008;10:1467–75.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      Committee on Environmental Health; Committee on Substance Abuse; Committee on Adolescence; Committee on Native American Child. From the American Academy of pediatrics: policy statement–tobacco use: a pediatric disease. Pediatrics 2009;124:1474–87.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Franchini M,
      2. Caruso C,
      3. Perico A,
      4. et al
      . Assessment of foetal exposure to cigarette smoke after recent implementations of smoke-free policy in Italy. Acta Paediatr 2008;97:546–50.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Dempsey D,
      2. Jacob P 3rd,
      3. Benowitz NL
      . Accelerated metabolism of nicotine and cotinine in pregnant smokers. J Pharmacol Exp Ther 2002;301:594–8.
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Funding This work was supported by a Flight Attendant Medical Research Institute grant (Clinical Investigator Award 072058). This work was also partly supported by the EU Integrated Project NewGeneris, 6th Framework Programme (Contract No. FOOD-CT-2005-016320) and by the EU-funded project HiWATE, 6th Framework Programme (Contract No. Food-CT-2006-036224).

    • Competing interests None.

    • Ethics approval University Hospital of Crete, Institutional Review Board (IRB).

    • Provenance and peer review Not commissioned; externally peer reviewed.