Abstract
Air pollution has been associated with a higher kidney failure risk; however, the underlying metabolic mechanisms remain unclear. We retrieved 229,192 participants without kidney failure at baseline from a large cohort. Annual concentrations of air pollution were assessed using a bilinear interpolation approach based on the residential address. Elastic net regression was applied to identify metabolites reflecting air pollutants and establish metabolic profiles. Cox regression models were utilized to evaluate the association of air pollutants, metabolic signatures, and kidney failure risk. Generalized propensity score (GPS) was performed to examine the potential causality. Mediation analyses were used to investigate the role of metabolism in the air pollution-kidney failure relationships. We identified 17,287 cases of incident kidney failure. Elastic net regression identified 87 metabolites for PM2.5, 65 for PM10, 71 for NOx, and 76 for NO2. Air pollution-associated metabolic signatures were associated with elevated risk of kidney failure, with the HRs of 1.08, 1.03, 1.11, and 1.11 for PM2.5, PM10, NO2, and NOx, respectively. The causal associations remained significant when assessed using GPS. Metabolic profiles mediated the air pollutants-kidney failure associations, with the mediation proportions ranging from 7.84 % to 11.51 %. We provided new insights from metabolism to explain air pollution-kidney failure associations.</p>