[Association between residential green space exposure in maternal early pregnancy and the risk of congenital heart defects in offsprings in Zhejiang province].

Objective

To evaluate the association between residential green space exposure in maternal early pregnancy and the risk of congenital heart disease (CHD) in offsprings, and to explore the potential mediating role of PM2.5.

Methods

A registry-based case-control study was conducted including 7 094 newborns with CHD and 28 376 non-CHD control newborns in Zhejiang Province between July 2019 and June 2022. Sociodemographic characteristics and birth outcomes were compared by Chi-square test or Mann-Whitney U test. Using data from the included CHD and control newborns and their maternal prenatal exposures, Spearman rank correlation analysis was used to assess the relationship between maternal residential green space exposure, PM2.5 exposure and road traffic indicators during gestational weeks 3-8. Generalized linear models were applied to examine the association between normalized difference vegetation index (NDVI) and CHD risk. Sensitivity analysis using different buffer sizes were conducted to assess robustness and stratified analysis were performed by sociodemographic characteristics and CHD phenotypes. Dose-response relationship between NDVI and CHD risk were explored. Causal mediation analysis was used to estimate the mediating effect of PM2.5.

Results

Compared with control group, CHD group had higher proportions of female infants, extreme-season conception, low birth weight and preterm birth, as well as higher prenatal PM2.5 exposure, lower NDVI, higher road density, higher proportion of rural residence and lower maternal education (all P<0.05). Road density was strongly correlated with the number of road intersections (r=0.75). Regression analysis indicated that each interquartile range increase in NDVI during gestational weeks 3-8 was associated with a 28% reduction in CHD risk (adjusted OR=0.72, 95%CI 0.69-0.75, P<0.001). Sensitivity analysis demonstrated consistent inverse associations across different buffer sizes (OR=0.53-0.72). Stratified analysis indicated protective effects of NDVI across multiple CHD subtypes (all OR<0.87). Dose-response analysis suggested an NDVI 1 000 protective threshold above 0.26, with variations by infant sex, region and conception season. Mediation analysis indicated that early-pregnancy PM2.5 exposure partially mediated the association between NDVI and CHD risk.

Conclusions

Higher residential green space coverage during early pregnancy is associated with a reduced risk of CHD in offspring, partially mediated by lower PM2.5 exposure. These findings highlight residential green space as a modifiable environmental factor with potential importance of the primary prevention of birth defects and the promotion of maternal and child health.