Abstract
OBJECTIVE: To investigate metabolic pathways linking air pollution exposure to osteoarthritis (OA) development and quantify their mediating role in disease pathogenesis.
METHODS: This prospective cohort study utilized UK Biobank data from 220,872 participants. Air pollution exposure was assessed using land use regression models for PM₂.₅, PM₁₀, PM₂.₅ ₋ ₁₀, NO₂, and NOₓ, with composite scores constructed. Nuclear magnetic resonance metabolomics profiling quantified 251 circulating metabolites. Elastic net regression identified air pollution-related metabolic signatures. Cox proportional hazards models assessed associations of air pollution related metabolic profiles with incident OA. Causal mediation analysis quantified metabolic pathway mediation using counterfactual methods.
RESULTS: During follow-up, 40,399 participants (18.3%) developed incident OA. Elastic net regression identified 50 metabolites associated with air pollution scores, encompassing lipoprotein subclasses (26%), fatty acids (16%), amino acids (12%), and inflammatory biomarkers. Air pollution-related metabolic signatures showed stronger associations with OA risk (HR 1.095, 95% CI: 1.082-1.108 per IQR increase) than air pollution scores alone (HR 1.030, 95% CI: 1.018-1.042). Effects were most pronounced for knee OA (HR 1.140, 95% CI: 1.118-1.162). Causal mediation analysis revealed that metabolic signatures mediated 21.04% (95% CI: 16.52%-41.95%) of the air pollution-OA association.
CONCLUSION: Metabolic pathways significantly mediate air pollution-OA associations, providing novel insights into environmental contributions to musculoskeletal health and identifying potential therapeutic targets for prevention strategies.