Association between maternal exposure to environmental endocrine disruptors and the risk of congenital heart diseases in offspring: a systematic review and meta-analysis

Congenital heart disease (CHD) is the most common type of congenital malformation and the leading cause of death in newborns. Some observational studies have investigated the relationship between exposure to environmental endocrine disruptors (EEDs) and CHD in pregnant women. However, the findings of epidemiological studies in different countries and regions remain controversial and exhibit significant variations. This meta-analysis aimed to explore the relationship between exposure to EEDs and CHD in pregnant women, hoping to provide some insights into related research in different regions and further demonstrate the relationship between the two. Three databases (PubMed, Embase, and Web of Science) were searched, and 17 studies with 1373 117 participants were selected, including 3 on polycyclic aromatic hydrocarbons (PAHs), 5 on pesticides/insecticides, 4 on phthalates, 4 on alkylphenolic compounds, and 7 on heavy metals. The Newcastle–Ottawa Scale was used to evaluate the quality of the studies. Begg’s and Egger’s tests were used to determine the publication bias of the studies, and the I 2 statistics to evaluate the statistical heterogeneity among the studies. The adjusted estimates were pooled using the random-effects and fixed-effects models to explore the association between EEDs and CHD and its subtypes. Maternal exposure to PAHs [odds ratio (OR) = 1.34, 95% confidence interval (CI): 1.17–1.53)] (e.g. PAHs and tetralogy of Fallot, septal defects, and conotruncal defects)], pesticides/insecticides (OR = 1.32, 95% CI: 1.20–1.46), alkylphenolic compounds (OR = 1.46, 95% CI: 1.14–1.86), and heavy metals (arsenic, cadmium, mercury, and lead) (OR = 2.09, 95% CI: 1.53–2.86) during pregnancy was positively associated with CHD in offspring. This study found that exposure to EEDs in pregnant women was positively associated with CHD in offspring. These findings are of great significance for researchers to further study the relationship between the two.


Introduction
CHD is the most common congenital disorder leading to infant mortality from birth defects, affecting approximately 0.8% of all newborn infants (Bouma and Mulder 2017).Based on the recently published Global Burden of Disease Aggregate for CHD, it is speculated that the global prevalence of CHD is 18/1000 in newborns and about 20/1000 in sub-Saharan Africa (Lawrenson 2020).The prevalence of CHD varies by subtype, with significant geographical differences in the prevalence of severe and very severe CHD (Pérez-Lescure et al 2018).In addition, a report from the Chinese Ministry of Health has shown that China has a high prevalence of congenital abnormalities, with the incidence closely aligning with the world average for middle-income countries.The incidence rate is approximately 5.6%, and the number of new birth defects is approximately 900 000 cases per year, of which approximately 250 000 are clinically evident at birth.The incidence of perinatal and neonatal CHD increased in urban, rural, and even national areas during 2000-2011 (Ministry of Health of the People's Republic of China 2012).The incidence of CHD in newborns remains high.
Since the twentieth century, the impact of environmental factors on CHD has received much attention owing to accelerated modern industrialization and the release of large amounts of pollutants.EEDs are defined as exogenous substances capable of interfering with the endocrine system and disrupting hormonal physiological functions, such as synthesis and metabolism (Di Nisio andForesta 2019, Rumph et al 2020).EEDs are commonly found in various common household products, such as plastic bottles, cosmetics, and toys.They have bioactive properties similar to those of estrogens and are diverse (Heindel et al 2017).Common EEDs include dioxins and dioxin-like compounds, polychlorinated biphenyls (PCBs), insecticides, pesticides, herbicides, bisphenol A (Darbre 2017, Heindel et al 2017, Brehm and Flaws 2019, Piazza and Urbanetz 2019), nonylphenols (NP, Park et al 2020, Bhandari et al 2021), alkylphenolic compounds (Darbre 2018, Walker et al 2021), PAHs (Hýžd'alová et al 2018), and heavy metals such as Pb (Wei and Zhu 2019), As (Jocsak et al 2019, Meakin et al 2019, Sargis and Simmons 2019), Cd (Bimonte et al 2021), and Hg (Kirtana and Seetharaman 2022).Among these, phthalates have high lipophilicity and are readily absorbed by humans and animals (Wittassek et al 2011).In summary, EEDs are commonly present in the environment and can affect human life with a high exposure probability.
Many studies have shown that maternal exposure to EEDs before and during pregnancy is strongly associated with the development of CHD in offspring.Wang et al (Wang et al 2022a) used ICP-MS to measure maternal plasma concentrations of As, Cd, Hg, Pb, and manganese (Mn) in eight hospitals in China.They discovered that, although As, Cd, and Pb concentrations were not statistically associated, the concentrations of Cd, Hg, and Mn in plasma were positively associated with CHD, and the Hg concentration was positively associated with SD and CD.Similarly, a case-control study in China found varied associations between maternal exposure to Cd and As and different CHD subtypes (Jin et al 2016).For instance, maternal As exposure was positively associated with RVOTO but had no statistically significant association with Cd (Jin et al 2016).The case-control study by Patel et al (2020) showed that maternal exposure to PAHs might increase the risk of CHD in infants, particularly TOF.Furthermore, Rappazzo et al (2016) discovered an association between mother's exposure to pesticides/insecticides and ASD and hypoplastic left heart syndrome in a case-control study using singleton live birth records from the North Carolina State Center for Health Statistics from 2003 to 2005 and data from the North Carolina (NC) Birth Defects Surveillance Program.Rocheleau et al (2015) conducted a case-control study using birth data from the National Birth Defects Prevention Study conducted from 1997 to 2002.They found that maternal occupational exposure to pesticides/insecticides was associated with specific CHD subtypes, such as secondary ASD, pulmonary valve stenosis, hypoplastic left heart syndrome, and TOF.In addition, a case-control study conducted in Sichuan Province, China, found that PVSD and PDA in offspring were positively associated with maternal occupational exposure to phthalates and alkylphenolic compounds.In contrast, pulmonary valve stenosis and secundum ASD in offspring were positively associated with maternal occupational exposure to phthalates (Wang et al 2015).Besides populationbased observational studies, several animal studies have reported an association between EEDs and CHD.Tang et al (2018) explored the effects of maternal rat exposure to di-(2-ethylhexyl) phthalate (DEHP) on fetal heart development and found a significantly higher rate of fetal cardiac malformations, including VSDs and cardiac hypoplasia, suggesting that DEHP was metabolized in vivo by activating peroxisome proliferator-activated receptor γ.This inhibited the expression of GATA4/Mef2c/Chf1 and the differentiation of mouse bone marrow stromal cells into cardiomyocytes, leading to the metabolic remodeling of cardiomyocytes.Li et al (2014) exposed zebrafish to PCBs and observed cardiac abnormalities, including pericardial edema and cardiac circulation defects, during zebrafish embryonic development.Thus, maternal exposure to EEDs increased the risk of CHD in offspring.
An increasing number of environmental pollutants associated with EEDs have been identified in recent years.Despite many studies on EEDs and CHD, their association remains unclear.Therefore, this systematic review and meta-analysis was conducted on the existing literature.It summarized the existing evidence on the association between maternal exposure to EEDs during pregnancy and the risk of CHD in offspring, thus providing a scientific basis for the prevention and treatment of CHD.

Search strategy
This study was conducted based on the PRISMA guidelines (Moher et al 2009).Three medical databases (PubMed, Embase, and Web of Science) were retrieved, and the keyword combinations were searched using the Boolean logical operators 'OR' , 'AND' , and 'NOT' .The following keywords were used: 'congenital heart disease,' 'congenital heart malformation,' 'congenital disease,' 'congenital disorder,' 'congenital heart defect,' 'lead,' 'mercury,' 'cadmium,' 'arsenic,' 'polycyclic aromatic hydrocarbons,' 'pesticides/insecticides,' 'environmental endocrine disruptors,' 'phthalates,' 'endocrine disruptors,' 'endocrine disrupting chemicals,' and 'alkylphenolic compounds.'Detailed search strategies were shown in table S1 of the supplementary material.All studies were restricted to the English language.The guidelines outlined by the PRISMA statement (Page et al 2021) were adopted for this study.

Literature inclusion and exclusion criteria
The inclusion criteria were as follows: (1) studies measuring EED concentrations and assessing pregnant women's exposure during pregnancy; (2) the outcome of the studies being CHD or its subtype, such as CD, TGAs, hypoplastic left heart syndrome, and pulmonary stenosis; (3) observational studies (e.g.case-control or cohort study); (4) studies describing measurement methods, such as ultrahigh-performance liquid chromatography coupled with tandem mass spectrometry, and job-exposure matrix, and assessing EED exposure in pregnant women; (5) studies reporting quantitative risk estimates, such as OR with 95% CI; (6) studies published from January 2003 to June 2023.
The exclusion criteria were as follows: (1) nonobservational studies; (2) animal/cell experiments, not epidemiological studies; (3) review or meta-analysis, conference abstracts, reviews, editorials, or case reports; (4) full text not available; (5) studies not in English; (6) exposed substances and research outcomes inconsistent with this study.

Literature screening and quality assessment
All retrieved studies were imported into the EndNoteX9 software and screened independently by two researchers (Dr Jie Yu and Kai Pan).Duplicate studies not identified by the software were manually removed.The remaining studies from the initial screening were read for titles and abstracts to further remove the studies that did not meet the inclusion criteria.Finally, the included studies were identified, and the full text was read in detail.The NOS (Peterson et al 2011) was used to evaluate the quality of the included studies.It was divided into three main sections: selection, comparability, and exposure, and included eight entries.A maximum of one point was awarded for each numbered entry in both the selection and exposure sections, and a maximum of two points was awarded for the comparability section.Studies with scores greater than 6 were considered as medium to high quality and those with scores less than 5 as low quality, with a maximum score of 9.

Data extraction
Data extraction included general characteristics, such as first author, year of publication, country, study period, study design, sample size, exposed substances, and exposure assessment method.When crude and adjusted estimates were provided in the studies, we chose adjusted estimates to reduce the influence of confounding factors and better reflect the 'association between EED exposure in pregnant women and CHD in offspring.'When the studies provided the effect estimates (OR/RR) of the third and fourth quantiles, we preferentially extracted the effect estimates of the third and fourth quantiles with the first quantile as a reference.The main outcomes observed in this study were CHD and several subtypes of CHD, including PVSD, VSD, ASD, SD, TOF, AVSD, CA, PDA, TGA, CD, RVOTO, and LVOTO.Data extraction was performed independently by two investigators.Any disagreements or issues that arose during this period were resolved through discussion or by the judgment of a third researcher (Dr Jie Xu).

Statistical analysis
A meta-analysis was performed using Stata version 16.0, and data were pooled when two or more studies were available for each EED and CHD subtype.The I 2 statistics were used to evaluate the statistical heterogeneity among studies.P < 0.05 or I 2 ⩾ 50% indicated statistical significance, and heterogeneity among the various included studies was considered.When overall I 2 was ⩾50%, we used a random-effects model, and when I 2 was ⩽50%, we used a fixedeffects model.Begg's test (Begg et al 1994) and Egger's test (Egger et al 1997) were used to determine the publication bias in the studies.

Literature screening results
A total of 2425 studies were retrieved by searching three databases: PubMed, Embase, and Web of Science.Through automatic and manual checks using EndNoteX9 software, 429 studies were excluded, leaving 1996 studies.The excluded studies included 15 meta-analyses, 44 reviews, 11 case reports and guidelines, 52 non-English papers, and 1 conference abstract.After reading the titles, abstracts, and full texts to exclude irrelevant 1816 studies, the remaining 57 studies were examined in detail, and 17 studies were finally included (Snijder et

Quality assessment
The NOS scores of the studies included in this metaanalysis ranged from 6 to 7, with an average score of 6.647 (See table S2 of the supplementary materials).

Pesticides/insecticides
Six studies were included to analyze the association between maternal pesticide/insecticide exposure during pregnancy and CHD subtypes in offspring.The pooled results showed that the maternal exposure to pesticides/insecticides during pregnancy was positively associated with VSD (OR = 1.24, 95% CI: 1.00-1.54),ASD (OR = 1.68, 95% CI: 1.37-2.07),and PDA (OR = 1.52, 95% CI: 1.25-1.85) in offspring.However, other CHD subtypes were not significantly associated with maternal exposure to pesticides/insecticides.The overall pooled results showed that maternal exposure to pesticides/insecticides during pregnancy was positively associated with CHD subtypes (OR = 1.32, 95% CI: 1.20-1.46) in offspring (figure 3).

Phthalates
Four studies were included to analyze the association between phthalates and CHD subtypes.The results showed that phthalates were positively associated with PVSD (OR = 2.73, 95% CI: 1.02-7.27).However, the overall pooled results suggested no significant association between the two (OR = 1.31, 95% CI: 0.83-2.05;figure 4).

Alkylphenolic compounds
Four studies were included to analyze the association between alkylphenolic compounds and CHD subtypes.The pooled results showed that maternal exposure to alkylphenolic compounds during pregnancy was positively associated with PVSD (OR = 1.72, 95% CI: 1.07-2.77)and PDA (OR = 1.89, 95% CI: 1.19-3.00) in offspring.The overall pooled results showed that maternal exposure to alkylphenolic compounds during pregnancy was positively associated with CHD subtypes (OR = 1.46, 95% CI: 1.14-1.86;figure 5) in offspring.

Heavy metals
Five studies were included in this meta-analysis to explore the association between heavy metals and CHD subtypes.The pooled results showed that the maternal exposure to Cd (OR = 2.34, 95% CI: 1.53-3.59)and Pb (OR = 2.92, 95% CI: 1.36-6.27)during pregnancy was positively associated with CHD (See figure S10 of the supplementary material), maternal exposure to Hg (OR = 2.36, 95% CI: 1.11-5.03)and Pb (OR = 2.86, 95% CI: 1.10-7.43)during pregnancy was positively associated with SD (See figure S11 of the supplementary material), and maternal exposure to Pb (OR = 3.52, 95% CI: 1.77-6.99)during pregnancy was positively associated with RVOTO

Association between maternal exposure to EEDs during pregnancy and CHD in offspring 3.3.2.1. Pesticides/insecticides, phthalates, and alkylphenolic compounds
Five studies were included to examine the association between maternal exposure to EEDs during pregnancy and CHD in offspring.The overall pooled results had no significant association (OR = 1.17, 95% CI: 0.89-1.55;figure 6).

Heavy metals
Seven studies were included to examine the association between maternal exposure to heavy metals during pregnancy and CHD in offspring.The pooled results showed that the maternal exposure to Cd (OR = 1.83, 95% CI: 1.30-2.57)and Pb (OR = 3.21, 95% CI: 2.08-4.97)during pregnancy was positively associated with CHD in offspring.The overall pooled results also showed that maternal exposure to heavy metals during pregnancy was positively associated with CHD (OR = 2.09, 95% CI: 1.53-2.86,figure 7) in offspring.

Sensitivity analysis and publication bias
The results of the meta-analysis were examined for changes by deleting the studies individually.No significant changes were observed, indicating that the results of the sensitivity analysis were stable (See figure S1-S9 of the supplementary material).Egger's test and Begg's test were used for detecting publication bias.No significant publication bias was found in our results (Egger's test and Begg's test, P > 0.05, see table S3 of the supplementary material).

Discussion
CHD is defined as structural malformations involving the heart and great vessels (Diab et al 2021).The accuracy of CHD diagnosis has increased with improved medical care, and 3D echocardiography can calculate ventricular volumes comparable to cardiovascular magnetic resonance imaging (Simpson and van den Bosch 2019).Despite significant advances in diagnosis and care, congenital heart malformations remain one of the leading causes of infant mortality (Bravo-Valenzuela et al 2018).Even if children with CHD survive into adulthood, late complications, such as heart failure and arrhythmia, increase mortality and morbidity (Bouma et al 2020).Therefore, protection during pregnancy is crucial to the occurrence and development of CHD in offspring.Exposure to adverse environmental factors, such as chemical toxicants, during pregnancy can cause mutagenic and teratogenic effects, leading to congenital anomalies (Baldacci et al 2018).Therefore, the mothers in this study were exposed to at least one EED during pregnancy.PAHs are commonly found in the environment, primarily originating from coal, diesel/gasolinedriven vehicle emissions, and so forth (Yu et al 2019).The risk of population exposure to PAHs increases due to the increasing number of vehicles.Similar to many studies, we found that maternal exposure to PAHs during pregnancy was positively associated with TOF, SD, and CD in offspring.However, the heterogeneity of the pooled results on the association between PAHs and SD, and CD was relatively large and needs to be treated with caution.Benzo(a)pyrene (BaP), a typical representative of PAHs, can adversely affect the female reproductive system (Zhang et al 2018).Zhao et al (2014) showed that BaP exposure in pregnant mice resulted in an imbalance between estrogen and progesterone levels, affecting the expression of their receptors and downstream related genes.This led to an altered endometrial tolerance and a reduction in the number of implantation sites.Huang et al (2021) found that BaP induced oxidative stress through aryl hydrocarbon receptor activation, leading to mitochondria-mediated intrinsic apoptosis and causing cardiac malformations in zebrafish embryos.In summary, PAH exposure might result in embryonic congenital malformations by disrupting hormonal imbalances, affecting related gene expression, and inducing oxidative stress.
Maternal exposure to pesticide/insecticide exposure during pregnancy was strongly associated with CHD subtypes in offspring, especially ASD, PDA, and VSD, in this study.The results of sensitivity analysis were stable, indicating that the pooled results were stable and reliable.However, the pooled results of PAHs-PDA were unstable, which might be due to the different methods used for assessing pesticide exposure.Previous conclusions on the relationship between exposure to pesticides/insecticides and CHD subtypes were not consistent.For example, Carmichael et al (2014) found that TOF was associated with exposure to pesticides/insecticides.However, Rocheleau et al (2015) showed no significant association between the two, which was consistent with our findings.The exposure to insecticides/insecticides was associated with certain CHD subtypes, but the conclusions were not consistent.This inconsistency might be related to the study population, exposure assessment, exposure concentration, exposure time, geographical differences, and some other unavoidable biases, such as unavoidable recall bias of study participants filling out questionnaires and inadequate self-protection measures during pregnancy.
No association was found between maternal exposure to phthalates and CHD in offspring.However, previous studies showed that phthalate exposure might increase the risk of PVSD and PDA in offspring.When both parents were assessed for phthalate exposure separately, the exposure of either parent was associated with an increased risk of PVSD in offspring (Wang et al 2015).Similar to the results of this study, it was found that phthalate exposure in pregnant women was positively associated with PVSD.However, this was contrary to the findings of Wijnands et al (2014), which might be due to unadjusted confounding factors in the latter study.Studies have also shown that phthalates are pollutants in indoor air and dust (Rudel et al 2003).Pregnant women in late pregnancy often stay indoors, have a small range of activities, and have a high risk of exposure, which increases the risk of CHD in offspring.The evidence of their association is still insufficient due to the lack of studies on the association between phthalates and CHD.At the same time, the OR values of the studies vary significantly, which affects the overall pooled results.Therefore, the results of this meta-analysis need to be treated with caution.More epidemiological evidence and animal experiments  are needed to further explore their association in the future.
The typical representative of alkylphenolic compounds is NP.However, the epidemiological evidence related to NP is lacking due to limited relevant studies.Currently, it has been shown that NP can cause damage to endocrine, reproductive, and neurological functions (Jie et al 2013).Our previous experiments also demonstrated that perinatal NP exposure in pregnant rats resulted in myocardial mitochondrial damage in offspring, indicating that NP might have genotoxic effects (Ni et al 2023).Our study found that maternal exposure to alkylphenolic compounds increased the risk of PVSD and PDA in offspring, confirming that NP might have genotoxicity.However, more epidemiological studies are needed to address the substantial heterogeneity in results and the instability observed in sensitivity analyses.Many recent studies are primarily focused on animal and cellular experiments, leading to a lack of large-scale epidemiological studies on the relationship between typical alkylphenolic compounds such as NP and CHD.Further studies in this area are necessary in the future.Previous studies have confirmed that some endocrine-disrupting heavy metals, such as Cd and As, have metalloestrogenic effects; they can bind to estrogen receptors and play an estrogen-like role (Takiguchi and Yoshihara 2006).A systematic review and meta-analysis conducted by Li et al showed a positive association between maternal heavy metal exposure and CHD and its subtypes (Li et al 2022).Similar to previous studies, this study also found  aluminum (Al) in umbilical cord blood and found that Pb and Al exposure led to a decrease in superoxide dismutase activity and an increase in malondialdehyde content, which promoted the occurrence of fetal CHD.Oxidative stress may be related to CHD.In summary, maternal exposure to heavy metals during pregnancy increases the risk of CHD in offspring, and oxidative stress induced by heavy metals may be a possible mechanism to increase the risk of CHD in offspring.
We did not find a statistically significant association between the three EEDs (pesticides/insecticides, phthalates, and alkylphenolic compounds) and CHD (CHD is a general term without specifying which subtype it is).However, when analyzing these three EEDs separately with specific CHD subtypes, a significant association was detected (except for phthalates).The CHD in each study might represent different CHD subtypes.Consequently, in conducting the pooled analysis, pooling different CHD subtypes together was inevitable, resulting in a large difference in the results.Therefore, this result should be treated with caution.However, some of the pooled results were highly heterogeneous (I 2 ⩾ 50%).The possible reasons for this high heterogeneity might be as follows: (1) the types of EEDs included were diverse, with different measurement methods and maternal exposure assessments for each specific type.(2) Although the study population included all pregnant women, differences were found with respect to age, ethnicity, geography, culture, exposure concentration, and duration.Although the substantial heterogeneity might affect the stability of the conclusions, the results of the sensitivity analysis remained consistent, indicating the reliability of our findings.
However, the different covariates adjusted between the studies might also have had some impact on our results.Pregnancy-related CHD has long been associated with the age of a mother.For instance, children of older mothers may be more susceptible to aortic coarctation and pulmonary stenosis compared with children of younger mothers (Cedergren et al 2002).A case-control study showed that maternal age was not associated with the incidence of CHD.However, it was observed that age might increase the risk of different CHD subtypes after the age of 35 years.For instance, infants born to mothers aged less than 35 years were more likely to have an ASD, whereas infants born to mothers aged more than 35 years were more likely to have VSD and PDA (Hashim et al 2020).The findings of populationbased studies conducted in various countries and races on the association between maternal age and CHD in offspring were inconsistent, but the increase in maternal age might increase the risk of CHD in infants.Among the 17 studies included in this metaanalysis, the age of pregnant women in all studies was nearly 35 years, and most of the studies (15/17) adjusted an important covariate, which was the age of pregnant women.Therefore, in our pooled results, we believe that the potential impact of the age factor of pregnant women on the results is acceptable.Maternal pre-pregnancy body mass index (BMI) is another significant factor thought to play a role in the development of CHD in babies.For instance, a metaanalysis of the association between maternal BMI and CHD in offspring showed that both maternal overweight (BMI 25.0-29.9kg m 2 ) and maternal obesity (BMI ⩾ 30 kg m −2 ) increased the risk of CHD in offspring (Zhu et al 2018).The possible underlying mechanism was that obese pregnant women might experience lipotoxicity during pregnancy.Also, the combination of excessive fatty acids and oxidative stress leads to the production of oxidized lipids, which can act as ligands for nuclear receptors, display cytotoxicity, and affect gene expression.An excess of lipids and oxidative stress can result in endothelial dysfunction.Obese pregnant women are more likely to experience unfavorable pregnancy outcomes because lipotoxicity can affect placental and maternal endothelial functions (Jarvie et al 2010).Some of the included studies failed to adjust the BMI of pregnant women, which might have had an impact on our pooled results.Unhealthy BMI of pregnant women may increase the risk of CHD in offspring, indicating that pregnant women should maintain a healthy BMI before pregnancy.A large number of included studies adjusted the covariate of folic acid content.Therefore, the potential impact of folic acid content on our research results was acceptable, with some covariates such as hereditary diseases such as hypertension and family history of CHD.The participants with related diseases were excluded from the beginning and some studies also adjusted the aforementioned covariates.Therefore, the impact of these covariates might be small.Moreover, perinatal smoking (Bolin et al 2022), drinking (Wang et al 2022b), and use of drugs (Fisher et al 2017) during pregnancy might increase the chance of CHD in the offspring.In the studies included in this meta-analysis, many only adjusted one of the covariates, and the adjusted covariates were inconsistent between different studies, which inevitably affected our research results.In conclusion, with regard to our research results, even if the covariates adjusted by the included studies were inconsistent, the overall results still reflected the association between maternal EED exposure and CHD in offspring.However, the impact of other confounding factors, such as race, family economic status, and so forth, cannot be ignored.The inconsistency of the covariates adjusted by each study implied that the association between maternal EED exposure and CHD in offspring was affected by other confounding factors, and the association between the two inevitably included the results obtained using other confounding factors.Future studies should further adjust the covariates having the significant impact on CHD to clarify the relationship between the two.
In addition, when we extracted CHD subtypes, at least two or more English papers were available, and we used this CHD subtype as the study outcomes of this study.Therefore, the association between EEDs and CHD subtypes shown in this study only reflects the association between some CHD subtypes and EEDs, and does not represent all CHD subtypes.Therefore, we observed a statistically significant or non-statistically significant association between the two in our pooled results, referring to the CHD subtypes included in this study.Due to the lack of published findings on other CHD subtypes, we did not include them.Therefore, this study only explored the association between EED exposure in pregnant women and a few common CHD subtypes.First, the information on other CHD subtypes remains limited, which was also a limitation of this metaanalysis.More relevant evidence is needed to demonstrate the association between the two.Second, the included studies only included infants diagnosed with CHD at birth but lacked data on infants diagnosed with CHD after birth, leading to some discrepancies.We found some cases diagnosed as CHD after birth.For example, Gong et al (2017) recruited pregnant women admitted to Panyu District Maternity Hospital in Guangzhou, China, and found that harmful substances in factories, especially organic solvents, were identified as potential risk factors for CHDs.This study did not meet our inclusion criteria and was not related to the harmful substances used in the present meta-analysis, and hence was excluded.In addition, a population study in Denmark found that the offspring of pregnant women exposed to As in drinking water had a higher probability of being diagnosed with CHD in the first year after birth (Richter et al 2022).The aforementioned studies suggested a diagnosis of CHD for some time after birth.At present, the most common participants in studies on the association between EED exposure and CHD are mainly infants diagnosed with CHD at birth, whereas the studies on infants diagnosed with CHD after birth is rare.We believe that some cases in real life are detected with CHD for some time after birth.Future studies can consider children with CHD in both cases and supplement this part of the data.We should also reveal the association between EED exposure and CHD in pregnant women.In addition, many studies suggest that the window period of cardiac malformation vulnerability should be set at 3-8 weeks of pregnancy (Strickland et al 2009).However, the exposure duration in our meta-analysis ranged from 3 months before pregnancy to a certain time at the end of pregnancy.Also, we were not sure whether they were in the window period (3-8 weeks of pregnancy).Therefore, EED exposure in different stages of pregnancy may lead to different incidence rates of CHD in offspring.Further, the exposure data in some studies are mainly from the monitoring sites near the residence of pregnant women.The studies conducted in different countries have different pollution concentrations in different regions.The monitoring methods used in different regions may also have deviations in measurements.Different research designs and ethnic differences may lead to different incidence rates of CHD in offspring.The differences in the incidence of CHD in offspring caused by these factors might have affected the overall pooled results of this meta-analysis.Future studies should extract biological samples, such as umbilical cord blood, of pregnant women or offspring as much as possible, so as to more truly reflect the association between EED exposure in pregnant women and CHD in offspring.

Strengths and weaknesses
The strengths of this study were as follows: (1) different ethnicity, large sample size, and wide study area improved the robustness of the results.(2) The included studies were of medium/high quality.
(3) This meta-analysis comprehensively and systematically analyzed the relationship between multiple EEDs and offspring CHD, conducted subgroup analysis for each EED and CHD subtype, and summarized the relationship between each EED and CHD subtype, thus providing a new scientific basis for preventing CHD.
However, the study had some shortcomings.(1) We could not include maternal age as a factor in the subgroup analysis due to the age difference of mothers, and ignored that advanced maternal age might also be a risk factor for CHD.(2) We explored only the association between maternal exposure to EEDs and CHD in offspring and ignored the effect of paternal exposure to EEDs on CHD in offspring.(3) Some typical EED representatives, such as bisphenol A, NP, and PCBs, were not included due to the lack or limited number of studies on the association of these typical representatives with CHD.(4) The results of some subgroups were highly heterogeneous, such as PAHs-SD, PAHs-CD, and so forth.Therefore, the results obtained need to be treated with caution, as detailed in table 2.

Conclusions
This study showed a statistically significant association between several specific EEDs and CHD/CHD subtypes.However, no significant association was observed between CHD and exposure to EEDs such as phthalates, alkylphenolic compounds, and so forth.This association might be related to different exposure assessment methods, study regions, and races.The results of this study require additional and larger population-based epidemiological evidence to demonstrate the association between CHD and exposure to EEDs.

Figure 6 .
Figure 6.Forest plot ORs (95% CI) for the association between three EEDs and CHD.Note: CHD is a general term without specifying which subtype it is.

Figure 7 .
Figure 7. Forest plot ORs (95% CI) for the association between Heavy metals and CHD.(a): Maternal blood lead; (b): umbilical cord blood lead; (c): highest tertile in maternal blood and umbilical cord blood.
that maternal exposure to Pb, Hg, Cd, and As during pregnancy was positively associated with CHD/CHD subtypes in offspring.In addition, it was observed that the interference of metal ions might lead to oxidative stress (Paithankar et al 2021).Oxidative stress can lead to changes in gene expression in embryos, causing oxidative damage to proteins and disrupting normal developmental processes (Barrozo et al 2021).Liu et al (2018) detected the concentrations of Pb and

Figure 8 .
Figure 8. Countries included in the studies incorporated into this meta-analysis.
68 3-33 Huang L, Mao B, Li J, Nan N, He L, Qiu J,Yi B and Liu Q 2023   Associations between the lead level in maternal blood and umbilical cord blood and congenital heart diseases in offspring Biol.Trace Elem.Res.201 2191-9 Huang Y, Zhang J, Tao Y, Ji C, Aniagu S, Jiang Y and Chen T 2021 AHR/ROS-mediated mitochondria apoptosis contributes to benzo[a]pyrene-induced heart defects and the protective effects of resveratrol Toxicology 462 152965 Hýžd'alová M, Pivnicka J, Zapletal O, Vázquez-Gómez G, Matthews J, Neca J, Pencíková K, Machala M and Vondrácek J 2018 Aryl hydrocarbon receptor-dependent metabolism plays a significant role in estrogen-like effects of polycyclic aromatic hydrocarbons on cell proliferation Toxicol.Sci.165 447-61 Jarvie E, Hauguel-de-mouzon S, Nelson S M, Sattar N, Catalano P M and Freeman D J 2010 Lipotoxicity in obese pregnancy and its potential role in adverse pregnancy outcome and obesity in the offspring Clin.Sci.119 123-9 Jie X, Jianmei L, Zheng F, Lei G, Biao Z and Jie Y 2013 Neurotoxic effects of nonylphenol: a review Wien.Klin.Wochenschr.125 61-70 Jin X, Tian X, Liu Z, Hu H, Li X, Deng Y, Li N and Zhu J 2016 Maternal exposure to arsenic and cadmium and the risk of congenital heart defects in offspring Reprod.Toxicol.59 109-16 Jocsak G, Ioja E, Kiss D S, Toth I, Barany Z, Bartha T, Frenyo L V and Zsarnovszky A 2019 Endocrine disruptors induced distinct expression of thyroid and estrogen receptors in rat versus mouse primary cerebellar cell cultures Brain Sci. 9 359 Kirtana A and Seetharaman B 2022 Comprehending the role of endocrine disruptors in inducing epigenetic toxicity Endocr.Metab.Immune.Disord.Drug Targets 22 1059-72 Lawrenson J 2020 The prevalence of congenital heart disease: we need to work towards getting more data Cardiovasc.J. Afr. 31 225-6 Li M, Wang X, Zhu J, Zhu S, Hu X, Zhu C, Guo X, Yu Z and Han S 2014 Toxic effects of polychlorinated biphenyls on cardiac development in zebrafish Mol.Biol.Rep. 41 7973-83 Li N, Mu Y, Liu Z, Deng Y, Guo Y, Zhang X, Li X, Yu P, Wang Y and Zhu J 2018 Assessment of interaction between maternal polycyclic aromatic hydrocarbons exposure and genetic polymorphisms on the risk of congenital heart diseases Sci.Rep. 8 3075 Li S et al 2022 Relationship between maternal heavy metal exposure and congenital heart defects: a systematic review and meta-analysis Environ.Sci.Pollut.Res.Int. 29 55348-66 Liu Z et al 2015 Maternal lead exposure and risk of congenital heart defects occurrence in offspring Reprod.Toxicol.51 1-6 Liu Z et al 2018 The effects of lead and aluminum exposure on congenital heart disease and the mechanism of oxidative stress Reprod.Toxicol.81 93-98 Lupo P J et al National Birth Defects Prevention Study 2012 Maternal occupational exposure to polycyclic aromatic hydrocarbons and congenital heart defects among offspring in the National Birth Defects Prevention Study Birth Def.Res.A 94 875-81 Meakin C J, Martin E M, Szilagyi J T, Nylander-French L A and Fry R C 2019 Inorganic arsenic as an endocrine disruptor: modulation of the glucocorticoid receptor pathway in placental cells via CpG methylation Chem.Res.Toxicol.32 493-9 Moher D, Liberati A, Tetzlaff J and Altman D G PRISMA Group 2009 Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement BMJ 339 b2535 Ni C, Pan K, Xu J, Long X, Lin F, Nie Y, Yang Y and Yu J 2023 Effects and mechanism of perinatal nonylphenol exposure on cardiac function and myocardial mitochondria in neonatal rats Ecotoxicol.Environ.Saf.258 114977 Nicoll R 2018 Environmental contaminants and congenital heart defects: a Re-Evaluation of the evidence Int.J. Environ.Res.Public Health 15 2096 Ou Y, Bloom M S, Nie Z, Han F, Mai J, Chen J, Lin S, Liu X and Zhuang J 2017 Associations between toxic and essential trace elements in maternal blood and fetal congenital heart defects Environ.Int.106 127-34 Page M J et al 2021 The PRISMA 2020 statement: an updated guideline for reporting systematic reviews BMJ 372 n71 Paithankar J G, Saini S, Dwivedi S, Sharma A and Chowdhuri D K 2021 Heavy metal associated health hazards: an interplay of oxidative stress and signal transduction Chemosphere 262 128350 Park H J, Zhang M, Lee W Y, Hong K H, Do J T, Park C and Song H 2020 Toxic effects of nonylphenol on neonatal

Table 1 .
Basic details of the included studies.

Table 2 .
Summary of meta-analysis results for the association between maternal exposure to EEDs and risk of CHD in offspring.
Note: (/): no value.Three EEDs: Pesticides/insecticides, phthalates, and alkylphenolic compounds.Heavy metals: As, Cd, Hg, and Pb. a : Removal of a study identified as a source of heterogeneity.
Liang Q, Zheng D, Zhong R, Wen Y and Wang X 2017 Congenital heart defects of fetus after maternal exposure to organic and inorganic environmental factors: a cohort study Oncotarget 8 100717-23 Hashim Jr S T, Alamri R A, Bakraa R, Rawas R, Farahat F and Waggass R 2020 The association between maternal age and the prevalence of congenital heart disease in newborns from 2016 to 2018 in single cardiac center in Jeddah, Saudi Arabia Cureus 12 e7463 Heindel J J et al 2017 Metabolism disrupting chemicals and metabolic disorders Reprod.Toxicol.