Exposure to air pollution is associated with adipokines in midlife women: The Study of Women's Health Across the Nation

Abstract

This study examined the associations between ambient air pollution exposure, including fine particulate matter (PM_2.5), nitrogen dioxide (NO₂), and ozone (O₃), with serum levels of high molecular weight (HMW) adiponectin, leptin, and soluble leptin receptors (sOB-R) in midlife women. The analysis included 1551 participants from the Study of Women's Health Across the Nation (median age = 52.3 years) with adipokine data from 2002 to 2003. Annual air pollution exposures were assigned by linking residential addresses with high-resolution machine learning models at a 1-km² resolution. Multivariable linear regression and Bayesian kernel machine regression (BKMR) were used to evaluate the associations for individual pollutants and pollutant mixtures. After adjusting for confounders in linear regression models, an interquartile range increase in PM_2.5 (2.5 μg/m³) was associated with a 4.6 % lower HMW adiponectin level (95 % CI: -8.8 %, -0.3 %). Exposure to air pollutant mixtures showed negative associations with HMW adiponectin and positive associations with leptin levels in BKMR models. These findings suggest that exposures to PM_2.5, NO₂, and O₃ are associated with adverse levels of adipokines, which may contribute to obesity-related outcomes. Further research is needed to confirm these findings and explore the underlying biological mechanisms.

Keywords: Adipokine; Adiponectin; Leptin; NO(2); O(3); PM(2.5).

Figure 1.

Associations between air pollutant mixtures and high molecular weight (HMW) adiponectin by Bayesian kernel machine regression. Dose-response relationships and 95% confidence interval (95% CI) bands for A. PM_2.5, B. NO₂, and C. O₃ with HMW adiponectin while holding all other air pollutants at median exposure levels. D. shows the estimated exposure-response relations and 95% (95% CIs when all air pollutant levels are held at a certain percentile, compared to when they are held at the 10th percentile. The model was adjusted for age, race/ethnicity, study site, education, financial hardship, smoking status, exposure to secondhand smoke, alcohol consumption, physical activity, total energy intake, waist circumference, and menopause transition stage. The posterior inclusion probabilities are 0.57 for PM_2.5, 0.11 for NO₂, and 0.17 for O_3.

Figure 2.

Associations between air pollutant mixtures and leptin by Bayesian kernel machine regression. Dose-response relationships and 95% confidence interval (95% CI) bands for A. PM_2.5, B. NO₂, and C. O₃ with leptin while holding all other air pollutants at median exposure levels. D. shows the estimated exposure-response relations and 95% (95% CIs when all air pollutant levels are held at a certain percentile, compared to when they are held at the 10th percentile. The model was adjusted for age, race/ethnicity, study site, education, financial hardship, smoking status, exposure to secondhand smoke, alcohol consumption, physical activity, total energy intake, waist circumference, and menopause transition. The posterior inclusion probabilities are 0.26 for PM_2.5, 0.12 for NO₂, and 0.11 for O_3.