Among the pregnant women with smoking and pregnancy outcome data, 71.1% never smoked, 21.5% smoked before but not during pregnancy. Among the women who smoked during pregnancy, light smokers (mean 4.8 cigarettes/day) predominated (92.3% of smokers) and only 7.7% of smokers smoked 10 or more cigarettes per day. In this cohort of women receiving prenatal care at a health maintenance organization, 5.0% of infants had LBW, 5.2% were born preterm, and 2.0% were small for gestational age (intrauterine growth restriction, IUGR).
presents maternal characteristics by tobacco smoke-exposure status. This is the overall low-risk population, with the majority of women at their optimal reproductive ages, high education, most having the ideal BMI, blood pressure, and most non-smokers. Smoking during pregnancy was associated with maternal age, education, marital status, and smoking history before pregnancy: the P value of exact test was p < 0.05. Infants of active smokers revealed non-significant reduction in mean birth weight: among non-smokers, the birth weight was 3445 ± 25 g, and light smokers – 3365 ± 59, p = 0.2.
presents variables that were associated with maternal smoking and other known LBW risk factors and provides inferential statistics, that is, odds ratios and 95 percent confidence intervals for the discrete variables. In univariate analyses, increasing number of cigarettes smoked was associated with an increased risk in LBW infants. Smokers of 9 cigarettes and more per day had crude odds ratios 1.97 (95% CI 0.78–5.02) times those of unexposed women; however, a small number of LBW cases were reported among smokers and that had an effect on the statistical significance of the results. Age, marital status and blood pressure had statistically significant effect on LBW risk. These risk factors were incorporated into multivariate logistic regression models. Variables that were associated with IUGR risk were same as LBW.
In terms of GSTT1 and GSTM1 genotype frequency, women in the group exposed to tobacco smoke and the groups not exposed were similar. Table 3
presents the combined association of maternal cigarette smoking and maternal genotypes with LBW controlling for effect of major covariates.
The percentage of GSTT1 absent genotype was 16.9% and that of GSTM1 was 46.6%. As shown in Table 3
, without consideration of genotype, maternal smoking during pregnancy was associated with an adjusted OR of 1.21 (95% CI 0.44 – 3.31) for LBW compared with the non – smokers. When GSTT1 genotype was considered, the association between maternal smoking and LBW increased and the adjusted OR was 2.06 (95% CI 0.67 – 6.37) among mothers with genotype present, but we could not assess the association among mothers with absent genotype because of 0 LBW cases in the smokers group.
When GSTM1 genotypes were considered, the association between maternal smoking and LBW differed: the adjusted OR was 1.11 (95% CI 0.26 – 4.76) among mothers with present but adjusted OR was 1.91 (95% CI 0.43–8.47) among mothers with absent genotypes. However, a test of interaction between smoking and the GSTM1 – null genotype showed that there was no statistically significant evidence for an effect modification adjusted OR 1.54; 95% CI 0.25 – 9.91, p = 0.59. Presence of both GSTT1 and GSTM1 genotypes tended to increase the smoking effect by 1.49, while the GSTT1 – present genotype and GSTM1 – null genotype were associated with 3.31 times higher risk among smokers (OR 3.31 95% CI 0.60–18.4). A test of interaction between maternal smoking and two studied genotypes did not confer a significant adverse effect on LBW risk, adjusted OR 1.45; 95% CI 0.22 – 10.1, p = 0.66.
presents the combined association of maternal smoking and GSTT1 – and GSTM1 genotypes with IUGR. Without considering genotypes, maternal smoking during pregnancy was associated with an adjusted OR of 1.57 (95% CI 0.45–5.55) for IURG compared with the non – smokers.
When we considered genotype GSTT1, the association between tobacco smoke exposure and IURG tended to be higher, and adjusted OR was found to be 2.63 (95% CI 0.65 – 10.6) among the mothers group with GSTT1 genotype present. The estimated smoking effect tendered to be higher among mothers with the GSTM1 – null allele, compared with non-smoking mothers OR was 1.70 (95% CI 0.28 – 10.4). We found some evidence of synergistic effect the GSTT1 and GSTM1 genotypes and active maternal smoking: OR were 2.66 (95% CI 0.38 – 18.5) for both alleles present and OR 2.47 (95% CI 0.31 – 13.1) for GSTM1 absent; nevertheless, there was no statistically significant interaction, adjusted OR 1.10; 95% CI 0.10 – 12.61, p = 0.93.
In this molecular epidemiological study on maternal cigarette smoking and genetic determinants of xenobiotic metabolism, we found some evidence that effects of maternal smoking on LBW risk and infant growth were increased by maternal GSTM1 null genotype. This study used a case-control design to analyze the genetic effects and the gene-environment interaction controlling for major confounding variables. Consistent with previous studies, we found that maternal cigarette smoking was associated with fetal growth restriction and increased risk of LBW risk [5
]. Our findings are consistent with a number of other studies that LBW risk may vary in relation to maternal age, BMI, parity, and other variables of the population in the study [29
]. Some other investigators who have examined the issue, revealed dose-response gradients in relation to the amount smoked [4
Present our findings show the greater LBW risk among light-smoking mothers with GSTM1 null genotype compared to those with GSTM1 present genotype, however the findings do not show a statistically significant results. These results are consistent with previous studies which analysed genetic susceptibility to cigarette smoke in the context of LBW or IUGR risk.
Wang et al.
reported that pregnant women with certain genotypes are susceptible to the adverse pregnancy effects of tobacco smoking, such as an increased risk of LBW [17
]. Without consideration of genotype, maternal smoking during pregnancy was associated with reduction in birth weight and elevated risk of LBW. When GSTT1 genotype was considered, the reduction in birth weight increased and 1.7 (0.9 – 3.2) - fold elevated risk of LBW for those with the genotype present, and 3.5 (1.5 – 8.3) - fold elevated risk of LBW for GSTT1 genotype absent was found among smoking mothers. The corresponding features for IUGR were 3.3 (1.7 – 6.3) and 2.5 (0.9 – 6.4), suggesting an interaction between metabolic genes and maternal smoking.
It has been reported that an individual difference in metabolic activation and detoxification xenobiotics partly depends on the genetic polymorphisms associated with GSTT1 and GSTM1 enzymes [33
]. The interactive effect of exposure to tobacco smoke and the presence of the GSTT1 polymorphism on infant birth weight was found to be significant by multivariate analysis, whereas the interactive effect of the presence the GSTM1 polymorphism did not reach statistical significance (p = 0.21) [25
Sasaki et al.
also reported combined effects between maternal genetic polymorphisms and smoking during pregnancy [35
]. The effects on reduction birth weight were not observed among women with GSTM1 null genotype who had never smoked. The authors conclude that maternal smoking in combination with maternal genetic susceptibility may adversely affect infant birth weight. However, results presented here do not show a statistically significant association between infant birth size and maternal smoking as linked to the GSTT1 genotype, while birth weight and length were significantly lower in subjects with GSTM1 null genotype.
Sram et al.
found that the risk of LBW and prematurity was significantly increased by the genotypes of GSTM1 null and a genotype combination with the CYP1A1*2A genotype [36
]. A survey among pregnant women have shoved that a combination of the GSTM1 null and the GSTT1 null genotypes exacerbate the effect of maternal exposure to tobacco smoke on birth weight more than the presence of either genotype one [24
Different results were presented by some authors [25
]. In the case-control study, controlling for several confounding factors, the authors revealed that the maternal GSTT1 null genotype had a 1.6 – fold reduced risk for small-for-gestational-age births. However, after adjustment for maternal smoking (categories less than 10 cigarettes/day and more than 10 cigarettes/day) the results were not statistically significant.
There is evidence that effect of cigarette smoke exposure depends on population characteristics: among Japanese GSTM1 null genotype decreases fetal growth but this effect is not observed in Caucasians. Moreover, the adverse effect on birth weight did not always accompany fetal growth restriction [37
Previous studies have suggested several plausible gene-smoking interaction explanations. First, tobacco smoke could disturb fetal and placental cellular regulation via elevated PAH-DNA adducts due to the increased activity of enzymes that metabolize cigarette toxins (e.g. CYP1A1) and lower or absent activity of enzymes that detoxify these compounds (e.g. GSTT1 and GSTM1 null genotypes) [15
]. Second, gene-smoking interactions may exert their synergistic effects through oxidative stress that occurs upon tobacco smoke exposure. In response to this stress various inflammatory cytokines are produced in lung tissue increasing inflammatory responses and immune responses [38
]. Moreover, as reported by some authors, maternal exposure to tobacco smoke affects the fetal urine cotinine concentration and also induces production of oxidative stress [26
]. Further, other environmental factors and genetic polymorphism of GSTM1 and GSTT1 may modify the response to oxidative stress and lead to adverse pregnancy outcomes [32
In this study, we demonstrated that there is increase in LBW and IUGR risk among smoking women even after adjusting for maternal age, education, BMI, and marital status; however, these findings suggest that there was no significant association between the GSTT1 and GSTTM1 polymorphism with low-level maternal smoking during pregnancy. The reason may be that the size of our nested case-control study and the proportion of women who smoked during pregnancy were too small to detect any significant difference.
Consistent to previous studies, we found that the effect of tobacco smoke increased LBW risk in the women’s group with combination of GSTT1 present and GSTM1 absent alleles was more than 3 times greater compared with the non-smokers group (OR 3.31; 95% CI 0.6 – 18.4). Similar evidence of the synergic effect of GSTT1 and GSTM1 polymorphism we revealed on fetal growth restriction, adjusted OR 2.47; 95% CI 0.31 – 13.1. The adverse effects of GSTM1 null genotype on IUGR in the presence of cigarette smoke exposure were observed even among light smokers. These data strengthen the previous research findings that indicated that subjects with GSTM1 null genotype have a greater risk of toxic tobacco smoke effects while restricted fetal growth among light smokers provides evidence of unhealthy development in uterus [35
When the results of this study are interpreted, a few conditions should be taken into account. This is a low-risk population with low-level tobacco smoke exposure, and low prevalence of GSTT1 null genotypes and these factors may limit extrapolation of these results to the other populations. The evaluation of exposure to tobacco smoke was indirect; we used self-reported information on smoking during- and before pregnancy, and thus the possibility of reporting bias exists. Because of the subjective measure of smoking exposure, there is a possibility of random exposure classification errors. However, in this study, we controlled for the main variables that might confound the association between smoking, genetic polymorphism, and birth outcomes, among them age, BMI, education, and family status, therefore, the residual confounding of results by smoking is expected to be small.
Our findings stress the need for appropriate policy and programs aimed at cessation of tobacco use among pregnant women. The evidence of increased risk of adverse birth outcomes in presence of genetic polymorphism reinforces the motivation argument for quitting smoking. This could help in directing smoking cessation interventions toward pregnant women and prevent adverse birth outcomes since smoking prevalence rate and effectiveness of tobacco control programs mostly depend not only on legislative recourses, but also on the individual perceiving that smoking is harmful to health [39