Main content area

Air pollution and diabetes-related biomarkers in non-diabetic adults: A pathway to impaired glucose metabolism?

Lucht, Sarah, Hennig, Frauke, Moebus, Susanne, Führer-Sakel, Dagmar, Herder, Christian, Jöckel, Karl-Heinz, Hoffmann, Barbara
Environment international 2019 v.124 pp. 370-392
C-reactive protein, adiponectin, adults, air pollutants, air pollution, antagonists, biomarkers, blood sampling, chemistry, cohort studies, diabetes, fibrinogen, glucose, interleukin-1, metabolism, models, nitrogen dioxide, obesity, particulates, women, Germany
While prior studies have linked air pollution (AP) to diabetes prevalence and incidence, few have investigated whether AP exposure is also associated with alterations in diabetes-related biomarkers in metabolically healthy adults.To evaluate the associations between short-, medium-, and long-term AP and diabetes-related biomarkers (adiponectin, interleukin-1 receptor antagonist [IL-1RA], high sensitivity C-reactive protein [hsCRP], fibrinogen) in persons without diabetes.Adiponectin, IL-1RA, hsCRP, and fibrinogen were measured in blood samples collected at the baseline (t0; 2000–2003) and first follow-up (t1; 2006–2008) examinations of the prospective Heinz Nixdorf Recall (HNR) cohort study in Germany. Participants' residential mean exposures to PM10, PM2.5, NO2, and accumulation mode particle number concentration (PNAM) were estimated for several time windows (1- to 365-day) prior to examination using a dispersion and chemistry transport model. We fitted covariate-adjusted linear mixed effects models using a random participant intercept and investigated effect modification by obesity status.We analyzed 6727 observations (nt0 = 3626, nt1 = 3101) from 4052 participants of the HNR study (52% women; ages 45–76 years at t0). For all air pollutants, medium-term exposures (60- to 120-day) were negatively associated with adiponectin (e.g., 91-day PNAM: −2.51% change [−3.40%, −1.53%] per interquartile [IQR] increase). Several short-, medium-, and long-term AP exposures were positively associated with IL-1RA (e.g., 365-day PM10: 2.64% change [1.25%, 4.22%] per IQR increase). Long-term exposures were positively associated with hsCRP level while no consistent patterns were observed for fibrinogen. Stronger associations for adiponectin were observed among non-obese participants.In persons without diabetes, we observed differing patterns of association between AP and diabetes-related biomarkers across a range of exposure windows, supporting the hypothesis that AP may play a role in the development of diabetes.