Maternal Exposure to PM2.5 and the Risk of Congenital Heart Defects in 1.4 Million Births: A Nationwide Surveillance-Based Study

Abstract

Background: Evidence remains limited about the association of maternal exposure to ambient fine particulate matter (airborne particles with an aerodynamic diameter ≤2.5 µm [PM_2.5]) with fetal congenital heart defects (CHDs) in highly polluted regions, and few studies have focused on preconception exposure.

Methods: Using a nationwide surveillance-based case-control design in China, we examined the association between maternal exposure to PM_2.5 during periconception (defined as 3 months before conception until 3 months into pregnancy) and risk of CHD in offspring. The study included 1 434 998 births involving 7335 CHDs from 2014 through 2017 on the basis of the National Population-Based Birth Defects Surveillance System, covering 30 provinces, municipalities, or municipal districts in China. We assigned maternal PM_2.5 exposure during the periconception period to each participant using satellite-based PM_2.5 concentrations at 1-km spatial resolution. Multilevel logistic regression models were used to calculate the multivariable-adjusted odds ratio and 95% CI for CHDs in offspring associated with maternal PM_2.5 exposure, and the exposure-response association was investigated using restricted cubic spline analysis. Subgroup or sensitivity analyses were conducted to identify factors that may modify the association.

Results: The average maternal exposure to PM_2.5 levels across all participants was 56.51 μg/m³ (range, 10.95 to 182.13 μg/m³). For each 10 μg/m³ increase in maternal PM_2.5 exposure, the risk of CHDs in offspring was increased by 2% (odds ratio, 1.02 [95% CI, 1.00 to 1.05]), and septal defect was the most influenced subtype (odds ratio, 1.04 [95% CI, 1.01 to 1.08]). The effect of PM_2.5 on CHD risk was more pronounced during the preconception period. Mothers <35 years of age, those living in northern China, and those living in low-income areas were more susceptible to PM_2.5 exposure than their counterparts (all P<0.05). PM_2.5 exposure showed a linear association with total CHDs or specific CHD types.

Conclusions: High maternal PM_2.5 exposure, especially during the preconception period, increases risk of certain types of CHD in offspring. These findings are useful for CHD prevention and highlight the public health benefits of improving air quality in China and other highly polluted regions.

Keywords: congenital; environment; heart defects; pregnancy.

Figure 1.. Flowchart of participant selection.

CHD indicates congenital heart defect.

Figure 2.. Risk of congenital heart defects in offspring per increase of 10 μg/m³ in PM_2.5 level according to exposure period.

Odds ratios (ORs) and 95% CIs were calculated using multilevel logistic regression. The periconception period was defined as the interval from 3 months before conception to 3 months into pregnancy. The preconception period was defined as the 3 months before conception. The first trimester was defined as the first 3 months of pregnancy. AVR indicates anomalous venous return; CTD, conotruncus defect; LVOTO, left ventricular outflow tract obstruction; Other, cardiac structure abnormalities not classified into the other types; Overall, all types of CHD combined; PDA, patent ductus arteriosus; PM_2.5, airborne particles with an aerodynamic diameter ≤2.5 μm; RVOTO, right ventricular outflow tract obstruction; and SPD, septal defect.

Figure 3.. Association between maternal exposure to PM_2.5 and risk of congenital heart defect in offspring according to exposure period.

The preconception period was defined as the 3 months before conception. The first trimester was defined as the first 3 months of pregnancy. Restricted cubic spline analysis was used to evaluate the shape of the association between airborne particles with an aerodynamic diameter ≤2.5 μm (PM_2.5) and the prevalence of congenital heart defects. ORs were estimated by comparing with a reference value of 35 μg/m³, which is the China class II standard of the annual mean of PM_2.5. Percentages of births were calculated over all births during the period from 2014 to 2017 per unit of PM_2.5. The solid line represents point estimates, and the gray area represents 95% CI.