SO-002

Objective: To investigate the effects of pre-pregnancy and prenatal exposure to six major air pollutants (CO, NO₂, O₃, PM₁₀, PM_2.5, SO₂) on birth weight in newborns and identify critical exposure windows. Methods: This retrospective cohort study analyzed data from 1 561 mother-infant pairs who delivered at Taiyuan Central Hospital between January 2020 and December 2023, combined with contemporaneous air pollution monitoring records. This study used the geographic information system (GIS) technology to evaluate the individual air pollutants exposure level of pregnant women, and calculated the average pollutant concentrations for four stages: preconception (the 12 weeks before conception), first trimester (weeks 1-13), second trimester (weeks 14-27) and third trimester (weeks 28-37). Multiple linear regression models were used to assess the associations between air pollutant exposure and birth weight at different pregnancy stages. The distributed lag nonlinear model (DLNM) was further constructed to characterize the nonlinear exposure-lag-response relationships, identify sensitive windows, and examine sex differences. Results: Preliminary multiple linear regression showed that third-trimester exposures to PM_2.5(β^=-43.00, 95%CI:-79.40- -6.68), PM₁₀(β^=-35.00, 95%CI:-66.00- -4.07), NO₂(β^=-35.50, 95%CI:-66.90- -4.08) and SO₂(β^=-28.80, 95%CI:-52.70- -4.97) were negatively associated with birth weight (all P<0.05), but these associations disappeared after full adjustment for covariates. DLNM analysis revealed exposure-response relationships for CO, O₃, PM₁₀, and SO₂ on birth weight, with distinct critical exposure windows: CO (weeks 2-13, 14-20), O₃ (weeks 4-13, 14-24), PM₁₀ (weeks 22-27, 28-37), and SO₂ (weeks 11-13, 14-27 and 28-37). Among these, the largest effect were observed for CO at gestational week 12, O₃ at week 13, and for both PM₁₀ and SO₂ at week 37. The effect estimates and their 95% confidence intervals were -37.12(-62.78- -11.45), -3.61(-6.45- -0.77), -5.01(-8.98- -1.04) and -8.31(-12.07- -4.56), all P<0.05. Gender differences in pollutant effects were observed: male newborns were more sensitive to SO₂ across multiple stages, to CO in mid-pregnancy, and to PM₁₀ in late-pregnancy; female newborns were more susceptible to PM₁₀ in the preconception and early-to-mid-pregnancy periods, to O₃ in early-to-mid pregnancy, and to SO₂ in late pregnancy. Conclusion: Exposure to air pollutants during pregnancy is associated with reduced birth weight in newborns. Different pollutants have specific exposure windows, and gender differences exist. This finding provides important scientific evidence for preventing adverse pregnancy outcomes and developing intervention strategies to improve neonatal health.

London plane (Platanus × acerifolia (Aiton) Willd.) tree ring chronologies from urban contaminated street and non-contaminated park were compared regarding accumulation of 15 polycyclic aromatic hydrocarbons (PAHs), 10 polychlorinated biphenyls (PCBs) and 6 nitro (-NO₂) PAH derivates (NPAHs), and nitrogen stable isotope ratio (δ¹⁵N). Temporal annual accumulations were tracked from 1970 to 2021 with the goal to reconstruct concentrations of air pollutants (nitrogen dioxide (NO₂), carbon monoxide (CO), sulfur dioxide (SO₂), particulate matter (PM)_2.5 and PM₁₀), using complex generalized additive mixed model (GAMM). Results indicate that tree rings accumulated all 31 targeted pollutants, while trees from contaminated site accumulated two-to three-fold higher amounts of pollutants in their tree rings. Likewise, δ¹⁵N clearly indicate that trees from the contaminated site accumulated more heavier N stable isotope. Trees from park had greater radial increment, while their chronologies did not correlate with air pollutants, unlike the targeted organic pollutants and δ¹⁵N. The GAMM results showed that only air PM₁₀ is weakly correlated with tree rings (Adj. R² < 0.35), while air NO₂ and SO₂ had the best fit with tree rings (Adj. R² = 0.4-0.8). Finally, this study showed that air pollution could be reconstructed by using organic pollutants accumulated in tree rings, while δ¹⁵N can be a key indicator for air pollutant reconstruction by using complex environmental models.