O'Sharkey Karl, Chow Ting, Mitra Sanjali, Thompson Laura, Su Jason, Cockburn Myles, Liew Zeyan, Ritz Beate
Department of Epidemiology, Fielding School of Public Health, University of California, Los Angeles, CA 90095, USA.
Department of Population and Public Health Sciences, University of Southern California, Los Angeles, CA 90032, USA.
Environ Int. 2025 Jun;200:109526. doi: 10.1016/j.envint.2025.109526. Epub 2025 May 9.
Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder with increasing prevalence. While genetics play a strong causal role, among environmental factors, air pollution (AP) exposure in pregnancy and infancy has been strongly endorsed as a risk factor. However, potential multigenerational impacts through the exposure of the grandmother during her pregnancy remain unexplored.
Using a multigenerational, population-based cohort from California spanning three decades (1990-2019), we examined the association between grandmother's gestational AP exposure (PM, NO, O) and ASD risk in grandchildren using logistic regression per 1-IQR unit increase, adjusting for maternal exposure in pregnancy or infant's exposure in the first year of life. We used continuous AP exposure as well as a categorical variable representing high and/or low (above or below median) exposure levels for both the grandmaternal and maternal pregnancies. Pregnancy and first year of life AP exposures were assigned using a land-use regression model with advanced machine-learning approaches.
We observed associations between PM (OR = 1.07, 95 % CI: 1.05, 1.10) and NO (OR = 1.09, 95 % CI: 1.05, 1.13) exposure during the grandmaternal pregnancy and increased ASD risk in the grandchild. However, only for PM did the increased effect estimates persist after adjusting for maternal pregnancy exposure (OR = 1.05; 95 % CI: 1.02, 1.08). High compared to low exposures in both grandmaternal and maternal pregnancies to PM (OR = 1.16, 95 % CI: 1.11, 1.23) and NO (OR = 1.12, 95 % CI: 1.06, 1.17) showed the strongest joint effects for ASD in the grandchild.
Multigenerational exposure to air pollution, particularly PM exposure in grandmaternal pregnancy, may influence ASD risk. Our study also suggests that ASD risk due to air pollution may be compounded by multigenerational exposures.
自闭症谱系障碍(ASD)是一种患病率不断上升的神经发育障碍。虽然遗传因素起着重要的因果作用,但在环境因素中,孕期和婴儿期接触空气污染(AP)已被强烈认为是一个风险因素。然而,祖母孕期接触空气污染对后代潜在的多代影响仍未得到探索。
我们使用了一个来自加利福尼亚州、跨越三十年(1990 - 2019年)的多代人群队列,通过逻辑回归分析每增加1个四分位间距单位,研究祖母孕期AP暴露(颗粒物、氮氧化物、臭氧)与孙辈患ASD风险之间的关联,并对母亲孕期暴露或婴儿出生第一年的暴露情况进行了调整。我们使用了连续的AP暴露以及一个分类变量,分别代表祖母和母亲孕期的高暴露和/或低暴露水平(高于或低于中位数)。孕期和出生第一年的AP暴露通过使用先进机器学习方法的土地利用回归模型进行赋值。
我们观察到祖母孕期接触颗粒物(比值比[OR]=1.07,95%置信区间[CI]:1.05,1.10)和氮氧化物(OR = 1.09,95% CI:1.05,1.13)与孙辈患ASD风险增加有关。然而,仅在调整母亲孕期暴露后,颗粒物增加的效应估计值仍然显著(OR = 1.05;95% CI:1.02,1.08)。祖母和母亲孕期高暴露与低暴露相比,接触颗粒物(OR = 1.16,95% CI:1.11,1.23)和氮氧化物(OR = 1.12,95% CI:1.06,1.17)对孙辈患ASD的联合效应最强。
多代暴露于空气污染,尤其是祖母孕期接触颗粒物,可能会影响ASD风险。我们的研究还表明,空气污染导致的ASD风险可能会因多代暴露而加剧。