Reproductive
Using cycle-timed air pollutant estimates to model fecundability Gabrielle S. Johnson* Gabrielle Johnson Johnson Johnson Johnson Johnson Johnson Johnson Johnson Johnson Johnson Johnson Department of Environmental Health Sciences, School of Public Health and Health Sciences, University of Massachusetts Amherst
Background: Accounting for sensitive windows of reproduction is crucial for understanding how timing of underlying biologic processes impacts the overall effect of environmental exposures, such as air pollution, on odds of pregnancy.
Methods: We included 1001 women from the Effects of Aspirin in Gestation and Reproduction (EAGeR) trial who lived in the Intermountain West and had residential georeferenced pollutant estimates. We abstracted pollutant data from the Air Quality Data for Health-Related Applications (AQDH) and Community Multiscale Air Quality Modeling System (CMAQ), including ozone (O3), particulate matter 2.5 (PM2.5), nitrogen dioxide (NO2), sulfur dioxide (SO2), and carbon monoxide (CO). We estimated biologically relevant sensitive windows to air pollution timed to cycle start date including pre-ovulation and implantation as well as the corresponding broader follicular and luteal phases, respectively. We constructed a multi-pollutant discrete-time Cox proportional hazards model to predict fecundability odds ratio (FOR) per inter-quartile range (IQR) increase in pollutants. Models adjusted for age, treatment status, income, education, and smoking.
Results: Multi-pollutant models using CMAQ data found that during pre-ovulation, O3 decreased odds of pregnancy by 20% (FOR 0.80, 95% CI 0.67, 0.96 per 15.6 ppb) and NO2 by 20% (FOR 0.8, 95% CI 0.56, 1.13 per 7.1 ppb). During implantation, O3, PM2.5, NO2, and SO2 were associated with trends of lower odds of pregnancy while CO increased odds of pregnancy by 43% (FOR 1.43, 95% CI 1.08, 1.91 per 80.6 ppb). Findings during pre-ovulation and implantation were consistent with follicular and luteal phases, respectively. AQDH data produced similar results.
Conclusion: Our findings account for sensitive windows of reproduction and suggest that air pollution may impact fecundability. Next steps include evaluating pollutant mixtures to further characterize pollutants associated with pregnancy.