Maternal Exposure to Housing Renovation During Pregnancy and Risk of Offspring with Congenital Malformation: The Japan Environment and Children’s Study

There have been no large, nationwide, birth cohort studies in Japan examining the effects of house renovation during pregnancy on congenital abnormality. This study examined the impact of (1) prenatal exposure to house renovation and (2) maternal occupational exposure to organic solvents and/or formaldehyde on the incidence of congenital abnormality. The fixed data of 67,503 singleton births from a large national birth cohort study that commenced in 2011 were used to evaluate the presence of congenital abnormalities and potential confounding factors. We employed multiple logistic regression analysis to search for correlations between maternal exposure to house renovation or organic solvents and/or formaldehyde during pregnancy and such congenital abnormalities as congenital heart disease, cleft lip and/or palate, male genital abnormality, limb defect, and gastrointestinal obstruction. After controlling for potential confounding factors, we observed that house renovation was significantly associated with male genital abnormality (OR 1.81, 95% CI 1.03-3.17, P = 0.04) when stratified by congenital abnormality, with no other remarkable relations to house renovation or occupational use of organic solvents and/or formaldehyde during pregnancy. There were also significant correlations for maternal BMI before pregnancy, history of ovulation induction through medication, maternal diabetes mellitus/gestational diabetes mellitus, and hypertensive disorder of pregnancy with an increased risk of congenital abnormality. In conclusion, this large nationwide survey provides important information on a possible association of house renovation during pregnancy with congenital male genital abnormality which needs confirmation in future studies.

Maternal occupational exposure to paints, dyes, glues, and other indoor environmental pollution during renovation may be a risk factor for congenital heart disease (CHD) in offspring 9,10 . The Danish National Birth Cohort (DNBC) study indicated that non-occupational exposure to paint fumes in the home may also be associated with other congenital anomalies in the general population 11 . To our knowledge, no studies have investigated if exposure to materials associated with general house renovation, including organic solvents and VOCs, during pregnancy increases the risk of congenital anomalies in Japan. Accordingly, we conducted a large birth cohort study with the specific objective of examining the impact of (1) prenatal exposure to house renovation and (2) maternal occupational exposure to organic solvents and/or formaldehyde on the incidence of congenital abnormality.

Results
A total of 67,503 (68.7%) mothers with singleton live births who completed the Japan Environment and Children's Study (JECS) questionnaire were available for analysis. Until the second/third trimester, the overall rate of mothers who responded to have house renovation/interior finishing or maternal occupational use of organic solvents and/or formaldehyde was 3.1% (2,058) and 9.4% (6,196), respectively. The prevalence of CHD, cleft lip and/or palate, male genital abnormality, limb defect, and gastrointestinal obstruction was 756 (1.1%), 160 (0.24%), 253 (0.74% among male infants), 176 (0.26%), and 45 (0.07%) cases, respectively. Table 1 summarizes the participants' characteristics and exposure to house renovation or organic solvents categorized by congenital malformations (no congenital malformations [controls], CHD, cleft lip and/or palate, male genital abnormality, limb defect, and gastrointestinal obstruction). No statistical significances were seen for exposure to house renovation/interior finishing or occupational exposure to organic solvents and/or formaldehyde.
In multivariate logistic regression analysis after adjustment for covariates, we observed a significant association between house renovation during pregnancy and male genital abnormality (OR 1.81, 95% CI 1.03-3.17, P = 0.04) ( Table 2). There were no significant relationships for exposure to house renovation/interior finishing or occupational exposure to organic solvents and/or formaldehyde with other congenital abnormalities.

Discussion
We herein describe the first large scale study in Japan by nationwide birth cohort study to determine the effects of maternal exposure to house renovation/interior finishing or organic solvents during pregnancy on such major congenital malformations as CHD, cleft lip and/or palate, male genital abnormalities including hypospadias and cryptorchidism, limb defect, and gastrointestinal obstruction. Our results indicate that house renovation during pregnancy increases the risk of male genital abnormality.
Housing decoration materials typically contain oil paints, dyes, laminate board, solid wood, wallpaper, resin glue, and plywood, among which various kinds of environmental pollutants have been reported. For example, organic solvents, heavy metals, and VOCs such as benzene, toluene, xylene, styrene, and aldehyde may be emitted from paints and dyes, and formaldehyde, trichloroethylene, and VOCs can be found in boards and plywood 8,17,18 . These contaminants may be released into the indoor air during or after house renovations. A previous study showed that formaldehyde concentrations were significantly higher in post-renovated than in pre-renovated homes, which reduced indoor air quality. Moreover, low-quality decoration materials may release greater amounts of pollutants into indoor air 19 .
Several case-control studies have revealed that mothers exposed to organic solvents during pregnancy exhibited an increased risk of giving birth to a child with a cleft lip or palate 20,21 . However, other reports did not observe a positive association between maternal exposure to organic solvents and orofacial defects, which was in agreement with our findings 22 . A case control study in China described that maternal exposure to house renovations increased the risk of CHD. This relationship was stronger for women who had moved into a newly decorated house 23 . Meanwhile, maternal exposure to organic dyes, lacquers, pigments, and paints during the first trimester of pregnancy was related to a higher incidence of cardiac malformations in the fetus 9,10 . The risk of CHD in offspring was also increased when the mother was exposed to organic solvents during pregnancy 24 . The limited number of studies focusing on an association between maternal occupational exposure and limb defects and gastrointestinal obstructions is insufficient for the assessment of risk. Moreover, such case-controlled studies are retrospective in design and may contain recall bias. Even prospective cohort investigations, such as the DNBC study, might not have fully evaluated causality due to the scarcity of outcomes.
Despite the rarity of this pregnancy outcome, our findings indicated a possible adverse association of house renovation during pregnancy on the outcome of male genital abnormalities. Earlier studies have also demonstrated a significant link for maternal occupational exposure to solvents 25,26 , but the mechanism of this phenomenon is not fully understood. As male reproductive organ development relies on androgens and a balanced androgen-estrogen ratio, it has been suggested that exposure to high levels of chemicals interfering with the production or action of sex hormones may disturb male reproductive tract formation 27 . Indeed, specific ingredients contained in house building materials have suspected hormone-disrupting properties that potentially interfere with the fetal development of reproductive organs 27,28 . Welsh et al. found that androgen blocking at the onset of fetal testosterone production in early gestation induced hypospadias in 64% and cryptorchidism in 30% of rats 29 . These findings may at least partially explain the mechanism of male genital abnormality by house renovation during pregnancy. www.nature.com/scientificreports www.nature.com/scientificreports/ Although our main result may have been influenced by the large number of subjects, i.e., mass significance, increased sample sizes are often necessary to evaluate such rare outcomes as congenital abnormalities. The JECS sample size was confirmed to contain sufficient analytical power. For instance, to test a hypothesis concerning a disorder with a prevalence of 0.1%, such as Down syndrome, a relative risk of 2.0, and an alpha error of 0.05 while using a cohort in which the proportion of individuals with a high level of exposure to the chemical substance of interest is 25%, a sample of 67,503 participants was required to provide a statistical power of 80% 30 .
This study has several limitations. First, the data regarding house renovations and occupational exposure to organic solvents and/or formaldehyde after having noticed pregnancy were collected from self-reported questionnaires in the second/third trimester of pregnancy and therefore subjective. An additional limitation was an inability to evaluate the precise timing of exposure during the sensitive organogenesis period. Accordingly, we could not associate the onset of congenital abnormalities with the timing of maternal exposure to house renovations and organic solvents and/or formaldehyde. Because our investigation contained very few cases of frequent occupational exposure to organic solvents and/or formaldehyde among responders, we could not assess a dose-response effect on the development of congenital abnormalities. In addition, although housing decoration materials likely produced various kinds of chemical exposure, we could not ascertain precisely which substances caused the congenital abnormalities. Finally, we used the data on abnormalities diagnosed until 1 month after birth, and so congenital disorders diagnosed afterwards were not included. Especially for cryptorchidism, the true number may have been overestimated as some cases descend by 1 year of age.
Despite the above limitations, this is the first study using a large dataset from a Japanese nationwide birth cohort study to examine the influence of house renovation and occupational exposure to organic solvents and/ or formaldehyde during pregnancy that controlled for confounders identified by previous reports. It provides important information on a possible association of house renovation during pregnancy with male genital abnormality that will need confirmation in future studies.  Table 1. Characteristics of participants with or without congenital abnormality. a Differences in maternal age and BMI were assessed with one-way repeated measures of ANOVA followed by post-hoc (Bonferroni) testing.
outcomes, exposure, and covariates. The main outcomes of interest were congenital malformations diagnosed by the subjects' obstetricians up to 1 month after birth. We selected diseases that were identifiable by the first month of life and had a relatively high incidence rate among congenital abnormalities: CHD, cleft lip and/or cleft palate, male genital abnormalities such as hypospadia and cryptorchidism, limb defects such as polydactyly, syndactyly, and cleft finger/foot, and gastrointestinal obstructions such as esophageal, duodenal, and small intestinal atresia and imperforate anus. We excluded participants with such chromosomal abnormalities as trisomy 21, trisomy 18, trisomy 13, and Turner syndrome. Subjects with congenital malformations other than the above 5 outcomes of interest or who were afflicted by 2 or more congenital malformations were excluded as well (Fig. 1) to assess the impact of exposure on each disease group. Infants with no congenital abnormality were employed as controls.
The factors of house renovation/interior finishing and maternal occupational use of organic solvents and/ or formaldehyde after becoming pregnant were assessed in the second/third trimester of pregnancy. Maternal exposure to house renovation/interior finishing during pregnancy was examined by the questionnaire item of "Did your household undertake house renovation/interior finishing after becoming pregnant?" Examples of organic solvents were paint thinner, solvents for examination/analysis/extraction, dry-cleaning detergents, stain-removing agents, paints, nail polish remover, and others. The frequency of maternal occupational exposure to organic solvents and/or formaldehyde during pregnancy was assessed by the questionnaire item "Please choose the frequency that best describes the use or handling of organic solvents and/or formaldehyde during work for more than half a day after becoming pregnant" and was grouped as "never", "once to three times a month", "once to six times a week", or "every day". We subsequently divided mothers into unexposed and exposed groups.  www.nature.com/scientificreports www.nature.com/scientificreports/ Demographic covariates included maternal age, smoking habit of mothers and their partner, maternal alcohol consumption, pre-pregnancy BMI, and socioeconomic status. Socioeconomic status was evaluated by the highest level of education completed by the mother (junior high school, high school, vocational school/junior college, or university/graduate school) and annual household income (<4,000,000, 4,000,000-7,999,999, or 8,000,000 + JPY). Obstetric and medical variables, such as means of pregnancy, maternal infection and medications during pregnancy, were also evaluated. Statistical analysis. All statistical analyses were performed using SPSS statistical software version 24 (SPSS Inc., Chicago, Illinois). Differences in maternal age and pre-pregnancy BMI among the types of congenital malformations were assessed by one-way repeated measures of analyses of variance (ANOVA) followed by post-hoc (Bonferroni) test. We categorized all continuous and ordinal variables, such as maternal age (<20, 20-34, or 35 + years), pre-pregnancy BMI (<18.5, 18.5-24.9, or 25 + kg/m 2 ), and annual household income. Analysis of variance and chi-square tests were conducted to compare covariates between groups stratified by category as well as by the presence of house renovation/interior finishing or maternal occupational use of organic solvents during pregnancy. We employed logistic regression models to calculate adjusted odds ratios (ORs) and their 95% confidence intervals (CIs). To employ logistic regression models on male genital abnormality, we analyzed subjects limited to male singleton infants. The covariates in our models were selected a priori based on previously published literature and biologic plausibility. In model 1, we first estimated the effects of house renovation/interior finishing and maternal occupational use of organic solvents after adjusting for maternal background (age [<35 or 35 + years], pre-pregnancy BMI, smoking and drinking habits. In model 2, we adjusted for the variables in model 1 along with maternal obstetric information (means of current pregnancy), pregnancy complications (maternal infection), and medications (folic acid supplements).