Does maternal exposure during pregnancy to higher ambient temperature increase the risk of congenital heart defects (CHDs)? Significant associations were found between maternal exposure during pregnancy to higher ambient temperature and CHDs risk especially during the cold season. From rodents to non-human primates, a teratogenic effect of hyperthermic insult has been demonstrated. There are fewer data regarding the effect on the human fetus and specifically the association between maternal exposures during pregnancy to higher ambient temperature and CHDs. This population registry-based cohort study included 135 527 live and stillbirths in the Tel-Aviv region of Israel in 2000–2006. Two clinical diagnostic groups of isolated cardiac defects (atrial septal defects and ventricular septal defects: n = 542 and 481, respectively) and one group of multiple cardiac defects (defined by the presence of two or more cardiac malformations, n = 607) were studied. Temperature measurements were constructed from ambient stations and used to assess the impact of maternal exposure to average ambient temperature and extreme heat events (daily average temperature above the 90th percentile) during Weeks 3–8 of pregnancy on risk of CHDs. Logistic models, adjusted for sociodemographic covariates, were used to evaluate the associations between temperature and CHDs. Overall, we found no significant associations between ambient temperature and CHDs throughout the year, with one exception for multiple CHDs. After stratifying by season of conception, continuous exposure to average ambient temperature and maximum peak temperature (1°C increase) during the cold season increased the risk for multiple CHDs [odds ratio (OR) 1.05, 95% confidence interval (CI): 1.00, 1.10 and OR 1.03, 95% CI: 1.01, 1.05, respectively]. A 1-day increase in extreme heat events increased the risk for multiple CHDs (OR 1.13, 95% CI: 1.06, 1.21) and also for isolated atrial septal defects (OR 1.10, 95% CI: 1.02, 1.19). Information both on CHD cases and on ambient temperature was based on registries and it is possible that this may cause some misclassification. In urban areas, pregnant women may be exposed to higher temperatures than recorded by ambient monitors because of the ‘heat island effect’. Furthermore, data for the amount of time spent indoors were unavailable and this could have resulted in exposure misclassification. The findings are important within the context of global climate change, which may have implications for public health in countries with mild winters and hot summers. This study was supported by the Israeli Ministry of Environmental Protection (research grant-7-2-7) and by the Environment and Health Fund (PhD Fellowship Program). There are no competing interests.

Background: Natural environments may have beneficial impacts on pregnancy outcomes. However, longitudinal evidence is limited and the associations with variance in surrounding greenness is unknown. Our objective was to evaluate these associations among 73 221 live births in Tel Aviv, Israel. Methods: Longitudinal exposure to mean of greenness during pregnancy and trimesters were calculated using satellite-based Moderate Resolution Imaging Spectroradiometer (MODIS) Normalised Difference Vegetation Index (NDVI) data. In addition, exposure to mean and variation of NDVI from high-resolution satellite and percentage of tree cover [Vegetation Continuous Fields (VCF)] at 300-m buffer were evaluated in a cross-sectional approach. Generalized linear models were used to estimate the crude and adjusted associations. We explore the possible mediating role of ambient exposures and distance to ‘outdoor gyms’ located in parks. Results: Crude beneficial associations between exposure to higher mean NDVI during pregnancy and pregnancy outcomes were observed [for birthweight, 3rd/1st tertile exposure increased the mean by 25.5 g, 95% confidence intervals (CIs): 15.4, 35.5] and decreased the odds of low birthweight, small for gestational age, preterm deliveries (PTD) and very PTD. Adjustment for individual and neighbourhood-level markers of socio-economic status (SES) attenuated all the associations. Strongest associations were observed during the first and second trimesters. Cross-sectional associations for mean greenness were similar with narrower CIs, and associations with NDVI were stronger than with tree cover and stronger for mean compared with variance of greenness. Associations were consistent for term births, different buffer sizes and for further adjustment to maternal education. Stronger associations were observed for lowest SES. Distance to ‘outdoor gyms’ and variance of greenness had the largest estimates of mediation. Conclusion: This study adds to the limited information on when exposure to greenness is most beneficial, on the association with variance of greenness and the possible pathways. These observations require confirmation in other populations.