OP X – 4 Multipollutant models for assessing particle number concentration exposure and changes in glucose metabolism in the heinz nixdorf recall study

Background/aim Prior studies have shown possible links between medium-term air pollution (AP) exposure and glucose metabolism markers. We investigated whether associations between accumulation mode particle number concentration (PN AM ) and glucose metabolism measures are robust to adjustment for other potential co-pollutants. Methods This analysis included observations from non-diabetic participants (n obs =7,108) of the population-based Heinz Nixdorf Recall study at baseline (2000–2003) and follow-up examination (2006–2008). Particulate matter (PM), accumulation mode particle number (PN AM ), and NO 2 exposures were estimated for each examination at every participant’s residence using the spatiotemporal EURopean Air pollution Dispersion (EURAD) chemistry transport model. Associations between a range of short- and medium-term PN AM exposures (1- to 182 day average prior to blood draw) and glucose metabolism measures (blood glucose, HbA1c) were assessed for robustness to co-adjustment for PM 2.5 , PM 10 , and NO 2 using a mixed effects linear regression model Results PN AM exposure showed low to medium correlation with PM 2.5 , PM 10 , and NO 2 (e.g., 0.33–0.53 for 28 day exposures). In the model with only PN AM and blood glucose, associations were seen for 14- to 91 day mean exposures windows with strongest associations observed for the 60 day window (1.05 mg/dL per 1580 n/mL (95% CI: 0.46 to 1.63)). Co-adjustment for PM 2.5 attenuated associations at all time points (e.g., 0.67 mg/dL per 1580 n/mL (95% CI: −0.22 to 1.55) for the 60 day window) whereas associations for PN AM remained largely unchanged upon adjustment for PM 10 and NO 2 . For HbA1c, strongest associations were apparent for the 105 day exposure window (0.096 p.p. per 1257 n/mL (95% CI: 0.075 to 0.116)). HbA1c model estimates were largely unchanged upon adjustment for co-pollutants. Conclusion The associations between exposure to PN AM and glucose metabolism measures were robust to adjustment for co-pollutants, possibly suggesting that exposure to small particles plays an independent role in influencing glucose regulation.