Congenital heart diseases (CHDs), commonly defined as a structural abnormality of the heart or great vessels present at birth, are the most common congenital disease among newborns. The average prevalence rate of CHD worldwide is approximately 9.4‰, with the highest prevalence rate in Asia (1). In China, there were 12.7 cases of CHD per 1,000 live births in 2019, with a trend of increasing prevalence year by year (2). The incidence in Suzhou was consistent with the national trend in recent years. In 2022, the incidence of CHDs in Suzhou was 70/10,000, accounting for 45.14% of total perinatal birth defects and recognized as a major cause of neonatal death due to birth defects.

Embryonic development is one of the critical windows in the growth and development process, and maternal exposure to air pollutants during pregnancy cannot be overlooked in the development of CHDs. Various health hazards associated with air pollution have become increasingly apparent. Since the late 1990s, the impact of air pollution on adverse pregnancy outcomes has garnered increasing attention (3, 4).

Ambient particulate matter, especially PM with an aerodynamic diameter of 10 μm (PM₁₀) and PM with an aerodynamic diameter of 2.5 μm (PM_2.5), is characterized by its small size, large specific surface area, and long residence time in the air. These particles tend to accumulate various toxic and harmful substances, making them a primary threat to human health (5). Animal experiments have indicated that particulate matter may induce heart defects through oxidative stress, DNA damage, and changes in molecular signals or epigenetic events (6, 7). However, the conclusions from existing population studies were not consistent, potentially due to the complex composition of air pollutants, such as different particle sizes, differences in demographic characteristics among study populations, age distribution, and varying levels of population susceptibility. Although the existing research findings were not entirely consistent, a growing body of epidemiological literature has suggested that PM₁₀ and PM_2.5 exposure may result in adverse birth outcomes, including CHD (8, 9). An increasing number of scholars have focused their attention on critical windows for maternal air pollutant exposure and adverse birth outcomes (10). Over the last decade, numerous domestic and international studies on susceptibility windows have divided pregnancy into several relatively long exposure periods, such as the first, second, and third trimester (11, 12). Other epidemiological and clinical evidence suggested that weeks 3–8 of pregnancy were the developmental period for most human organs and were susceptible to environmental exposure or teratogens; consequently, many studies have selected weeks 3–8 of pregnancy as the exposure window (13–15). Subsequently, an increasing number of researchers considered that the division of longer exposure periods may overlook potential critical windows, and they are beginning to attempt to further refine the temporal scale of pollutant exposure to more accurately determine the critical exposure window period (16, 17). However, few studies in China have effectively evaluated the correlation between weekly average ambient particulate matter exposure throughout pregnancy and CHDs. Therefore, it is urgent to clarify the effects of maternal ambient particulate matter exposure on CHDs.

With the accelerating urbanization process in Suzhou Industrial Park, air pollution issues are expected to persist for a considerable period. Currently, eugenics and quality-oriented childbearing are particularly important in this new fertility situation. The main objective of this study is to estimate the association between maternal exposure to ambient particulate matter and CHDs, as well as to identify specific windows using distributed lag models (DLMs) incorporated in Cox proportional-hazards models. The findings of this study could provide scientific evidence for air pollution prevention and control in Suzhou City and simultaneously offer higher-quality perinatal care for pregnant women at the same time.

Although genetic factors play an important role in the development of CHD, studies have shown that only approximately 15% of CHD cases can be attributed to genetic factors, while the remaining 85% are influenced by the interaction of multiple factors, including genetics and the environment (18). Our study demonstrated that maternal exposure to low-level ambient air pollution (46.5 μg/m³ for PM₁₀ and 27.3 μg/m³ for PM_2.5) during the entire pregnancy in Suzhou Industrial Park was lower than annual average maternal exposure concentrations in other cities like Tianjin, Shanghai, Hefei, and Guangzhou (16, 17, 19, 20). However, this exposure concentration was significantly higher than that in some cities abroad, such as Brisbane and Victoria (21, 22). Individual exposure to PM₁₀ (mean ± SD) was highest in the third trimester, while individual exposure to PM_2.5 (mean ± SD) was highest in the first trimester. Compared to our previous study on the analysis of maternal exposure to air pollutants during pregnancy from 2015 to 2019 (23), the overall exposure concentration of particulate matter during pregnancy in this study has decreased, indicating that with the implementation of various measures to combat haze, the air pollution situation in Suzhou City has improved to some extent, and there has been a general trend of reduction in the concentrations of PM₁₀ and PM_2.5.

In the present study, we observed that weekly exposure to PM₁₀ increased the risk of CHDs. The susceptible windows were identified as the 5th–12th weeks and 18th–22nd weeks of pregnancy, with the strongest association observed during the 5th and 6th gestational weeks. We also evaluated the association by trimester, but no significant association between either PM₁₀ or PM_2.5 and the risk of CHDs was observed across all the aforementioned exposure windows. Our study results further revealed that the sensitivity calculated based on exposure levels during different pregnancy trimesters was lower than that calculated on a weekly specific scale. The first susceptibility window was identified during early pregnancy (weeks 5–12). This finding strongly suggests that exposure to ambient PM10 during the initial and active phases of morphological heart development may be a significant risk factor. The second susceptibility window emerged during mid-pregnancy (weeks 18–22), indicating that pollutant exposure might still exert disruptive effects during the late-shaping processes of the heart. Previous studies on the association between environmental pollutants and adverse birth outcomes have reached such conclusions (17, 19). At present, a significant amount of studies have focused on exposure windows measured on a monthly or trimester scale, and some results are similar to ours. For example, birth cohort studies in Foshan and Wuhan found no association between PM_2.5 and PM₁₀ exposure, and the risk of CHD (15, 24). To the best of our knowledge, only a few studies have evaluated weekly specific associations between maternal ambient particulate matter exposure and CHDs. Moreover, the results of the sensitive exposure gestational weeks obtained from these studies were not consistent. A study in Hefei evaluated the associations of weekly air pollution exposure and CHDs, results showed that the susceptibility windows of PM_2.5 (weeks 20–26), PM₁₀ (weeks 0–2 and weeks 25–29 of pregnancy), while the strongest effects on CHDs were observed in week 22 (RR = 1.034, 95% CI: 1.007–1.062) and week 0 (RR = 1.081, 95% CI: 1.02–1.146), respectively (16). Zhang et al. in Wuhan reported a significant association between PM_2.5 exposure during weeks 7–10 of pregnancy and the risk of VSD¹⁴. Many early studies suggested that weeks 0–8 were a sensitive window period for the risk of CHDs associated with PM₁₀, which partially aligned with the sensitive window period we derived from our study (25–28). Regrettably, our study found no significant correlation between PM_2.5 and the risk of CHDs; one possible reason was that we did not classify CHDs into specific subtypes. A considerable number of studies have suggested that the association between particulate matter and specific congenital heart subtypes varies significantly. Padula et al. clarified that PM₁₀ was associated with pulmonary stenosis (PS) and ventricular septal defect (VSD), while PM_2.5 was associated with transposition of the great arteries (TGA) and inversely associated with VSD and atrial septal defect (ASD) (28). The inconsistency in associations and sensitive windows across different study results may be due to the following reasons: First, the exposure concentrations of pollutants varied from place to place, and evidence suggests that lower exposure levels may limit the strength of the corresponding associations (29). Second, demographic characteristics, immunity factors, and other variables may differ among the study population groups. Finally, there was heterogeneity in the assessment methods for environmental pollutant exposure and the diagnosis of CHDs in different studies.

Environmental factors, another crucial risk factor for CHDs, have always been a focus of attention. A study in California highlighted that individuals were susceptible to the interaction effects of genetics and the environment (30). However, the specific mechanisms of ambient particulate matter impacts on CHDs remain to be fully elucidated. In recent years, advancements in genomic detection technologies and model animal experimentation have propelled studies on gene–environment interactions, providing valuable insights into the mechanisms underlying CHD development. Many studies have demonstrated that placental DNA methylation and RNA expression are closely linked to numerous prenatal exposures, such as air pollution, metals, and maternal smoking, as well as infant and childhood health outcomes (31). This highlights the potential role of epigenetic modifications in mediating the effects of environmental factors on fetal development. Animal experiments showed that particulate matter induced cardiac defects through pathways involving oxidative stress, DNA damage, and changes in molecular signaling or epigenetic events (6). Zebrafish embryo experiments have shown that PM_2.5 can cause cardiac developmental malformations by activating the aromatic hydrocarbon receptor (AHR), mediating EOM-induced oxidative stress, and resulting in DNA damage and apoptosis (32). These findings underscore the potential mechanisms by which particulate matter exposure can disrupt normal cardiac development. Embryology studies suggest that 3–8 weeks during pregnancy are critical for organ development and are susceptible to environmental exposure or teratogens, indicating that early pregnancy may represent a critical window for exposure to environmental pollutants. Further experimental studies are needed to provide mechanistic support for the epidemiological study results.

Our study has several strengths. To the best of our knowledge, this study is one of the few to evaluate weekly specific associations between PM_2.5 and PM₁₀ exposure, and the risk of congenital heart defects. Therefore, our findings provide new evidence to identify more refined susceptible exposure windows during pregnancy. Despite these strengths, this study has some limitations. Because of the inability to obtain information such as the distance between study subjects and air monitoring stations, exposure levels were reliant on fixed-site monitoring data, which may not capture true individual-level exposure variation. We estimate that this likely results in exposure misclassification, biasing the effect estimates (e.g., hazard ratios) for the associations between PM2.5 and health outcomes toward the null hypothesis and potentially leading to an underestimation of the true health effects. Furthermore, existing studies have found that high temperatures and other environmental factors can enhance the adverse effects of air pollutants on heart development; however, environmental confounding factors, such as temperature, humidity, and other types of air pollutants, were not included in this study. At the same time, CHDs encompass a wide variety of subtypes, and there is heterogeneity in the causal relationship between different subtypes and air pollutants. Regrettably, CHDs were not subcategorized into specific subtypes in this study. In future studies, it is necessary to increase the number of experiments and population studies on exposure to air pollutants and specific subtypes of CHDs to gain a deeper understanding of the mechanisms. Because of the limitations in our data collection, we were unable to account for several potential confounding factors, such as genetic issues in parents and the fetus, maternal complications, exposure to secondhand smoke, and alcohol consumption history. Furthermore, collecting individual-level behavioral data (e.g., mobility trajectories and mask-wearing practices) is crucial for enabling a more refined personal exposure assessment in the context of future major public health events.

In summary, distributed lag models (DLMs) incorporated in Cox proportional-hazards models were applied in this study to refine exposure periods and more precisely identify susceptible windows for the effects of PM_2.5 and PM₁₀ exposure on CHDs. The analysis showed that a high level of outdoor PM₁₀ was associated with an increased risk of CHDs. Weeks 5–12 and 18–22 during pregnancy were identified as susceptible windows, with the strongest effects of PM₁₀ on CHDs observed during weeks 5 and 6 of gestation. This study has potentially important public health implications for formulating intervention measures to reduce CHDs caused by environmental particulate matter pollution.