Effects of maternal smoking during pregnancy on congenital anomalies: Results from the Japan Environment and Children's Study (JECS)


 Objective
Tobacco smoke contains over 7,000 chemical compounds, tobacco smoking is a possible risk factor of congenital anomalies (CAs). Therefore, we investigated the relationship between maternal smoking during pregnancy and CAs in children. Drawing data from the Japan Environment and Children’s Study (JECS), gathered between January 2011 and March 2014, 91,626 pregnant Japanese women were included.
 
Results
Of 91,626 subjects examined, 2.4% (2,183) of their infants had “any CAs.” By smoking behavior; 1,256 (57.5%), 523 (23.9%), 271 (12.4%), and 136 (6.2%) subjects were in the “never smoked,” “quit before pregnancy,” “quit after pregnancy,” and “still smoking” groups, respectively. The odds ratios (OR) for any CAs were not significant when “quit before pregnancy,” or “quit after pregnancy,” groups were compared to the “never smoked” group. However, upon comparing “never smoked” group with the “still smoking” group regarding “any CAs,” the OR was significant (OR 1.38, 95% confidence interval, 1.13-1.69, p=0.002). This study showed association between smoking continuation during pregnancy and an increased risk of CAs. This study results may be used as the basis for recommending anti-smoking education for pregnant women, and younger yet-to-become pregnant women.

The principal component of tobacco smoke (nicotine) is a nervous system teratogen [2], and carbon monoxide has also been indicated as a possible teratogen [3].
Furthermore, tobacco smoke contains over 7,000 chemical compounds, hundreds of which have harmful effects in humans [4]. Therefore, the possible associations between CAs and smoking during pregnancy and/or exposure to tobacco smoke must be investigated.
In a Danish register-based birth cohort study including 838,265 singleton liveborn babies, there was a significantly higher rate of CAs, including oral clefts and respiratory and cardiovascular abnormalities, in infants of women who had smoked while pregnant [5]. Moreover, among 1,413,811 infants registered at the Swedish Health Registries, there were significantly higher rates of "any malformations" among children of women who had smoked while pregnant [6]. However, no largescale birth cohort study investigating the association between CAs and maternal smoking while pregnant has been conducted in Japan or in the general Asian region.
In 2012, when the recruitment of the study subjects for the present study began, the average smoking rate among Japanese women in their twenties and thirties was 12.3% and 11.9%, respectively [7]. These were relatively high rates of smoking in women of reproductive age in Japan. The American Cancer Society revealed that the female smoking rate has not decreased as suggested by the global epidemic model [4].
In the present research, we investigated the associations between CAs and maternal smoking behavior in the early stages of pregnancy, using data that included approximately 100,000 pregnant women who participated in the nationwide birth cohort Japan Environment and Children's Study (JECS).

Study design
The JECS is a birth cohort study (conducted principally by the Ministry of the Environment) investigating the associations between environmental factors and childhood health and development. Recruitment for the study occurred across 15 regional centers within Japan from 2011 to 2014. Subject recruitment involved a face-to-face explanation of the survey to mothers, and self-administered informed consent was obtained. The details of the JECS design are well documented [8,9].
The present research analyzed data from the "jecs-ag-2016042" and the "allbirth_revice001_ver001" datasets. Both datasets included 104,102 fetuses and their mothers. We included data on 91,626 subjects in our analysis, excluding the following: those with missing information about smoking during the early stages of pregnancy, 29 mothers who withdrew total consent, those with multiple consents for multiple participations (after the second instance), those with multiple births, or those with missing information transcribed from medical records at birth and at onemonth-old (Fig 1). Questionnaires about exposure to tobacco smoke Self-administered smoking habits questionnaires were distributed to and collected from participants by the study research staff during the early pregnancy. Smoking habits were categorized into four possible responses as follows: "never smoked," "quit before pregnancy (QBP)," "quit after pregnancy (QAP)," and "still smoking." Those with a smoking history were asked about the age at which they started smoking and the approximate number of cigarettes they smoked per day on average, while those who had quit were asked when they quit. Based on these responses, the number of years smoked, and the packs-per-year were calculated.
The following questions on second-hand smoke exposure were asked: "What is the smoking behavior of the child's father (the subject's husband or partner)?," "Before the present pregnancy, how many times per week did you encounter tobacco smoke from others, either within buildings outside the home, in the home, or at the workplace?." "How many people in your family smoke?," and "How many people smoke in your room at home?"

Main outcomes
Using hospital chart histories recorded during childbirth and at one-month follow-up examinations, categorical data on 61 types of CAs were recorded in the transcriptional sheets [10]. In the event of a contradiction between the two time points (at birth and one month later), a CA was accepted as being present if the CA had been indicated at either time point. We determined the odds ratio (OR), to investigate the association with smoking by using the main CA groups according to the ICD-10 classification system. Among genital organ CAs, since cryptorchidism and hypospadias afflicted only male children, the analysis was limited to the 46,893 male children included in the study.

Statistical analysis
To identify the association between CAs and maternal smoking, we performed a logistic regression analysis and determined the 95% confidence intervals (CI). In the multivariable logistic regression analysis, we adjusted for maternal age at delivery, pre-partum body mass index (BMI), diabetes mellitus, marital status, education history, annual household income, alcohol intake, folic acid intake, anticancer drug intake, antihypertensive intake, anti-convulsant intake, and retinoic acid intake during early pregnancy. We used SAS version 9.4 software (SAS Institute Inc., Cary, NC) for all statistical analyses.

Results
Out of the 91,626 subjects included in our analysis, 2,186 (2.4%) had CAs (any CAs).
Categorized by smoking behavior, "still smokers" had lower socioeconomic status indexes, such as education history and income, than the other groups (Table 1). An additional table file shows the number of CAs per 10,000 population according to maternal smoking behaviors during pregnancy in more detail [see Additional file 1]. On the other hand, we could not show the risk of CAs for each organ category unlike in a previous research, the Danish register-based cohort study [5]. It is also possible, despite the large sample size of 91,626 subjects in this study, this may not have been sufficient to give interpretive weight to an investigation on an ailment like CAs, which has a relatively low incidence rate. Therefore, a study with a larger cohort might be necessary.
The present study's strength is that it is based on a nationwide survey of Japan with a very high response rate from the start of the study (during pregnancy) to the time of delivery. Many responses relating to smoking behaviors were obtained. Multiple covariates (confounding factors), thought to be optimal for analysis, were available in the study, enabling many different effects to be considered and discarded.
In summary, this study showed that smoking continuation during pregnancy was associated with an increased risk of CAs. Therefore, this study may serve as justification for anti-smoking education for pregnant women and younger women who are yet to be pregnant.

limitations
A limitation of the present study, therefore, is that the data relating to smoking behaviors and quantity are self-reported, and therefore could not be analyzed objectively. Further, the present study used the 30 ailments and abnormalities defined by the JECS and, therefore, conducted its analysis based on these categories. Moreover, due to small numbers, further statistical analysis of the QAP or the still smoking groups could not be performed.

Ethics approval and consent to participate
The JECS protocol was approved by the Review Board of the Ministry of the Environment for epidemiological studies, and by the Ethics Committees of all participating institutions. The JECS is conducted in accordance with the Declaration of Helsinki and other nationally valid regulations, and written informed consent was obtained from all participants.

Availability of data and materials
The data used to derive our conclusions are unsuitable for public deposition due to ethical restrictions and specific legal framework in Japan. The Ethical Guidelines for Epidemiological Research enforced by the Japan Ministry of Education, Culture, Figure 1 Flow diagram of the enrollment and exclusion process in this study.

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download.