Arsenic exposure and oxidative damage to lipid, DNA, and protein among general Chinese adults: A repeated-measures cross-sectional and longitudinal study.

Arsenic-related oxidative stress and resultant diseases have attracted global concern, while longitudinal studies are scarce. To assess the relationship between arsenic exposure and systemic oxidative damage, we performed two repeated measures among 5236 observations (4067 participants) in the Wuhan-Zhuhai cohort at the baseline and follow-up after 3 years. Urinary total arsenic, biomarkers of DNA oxidative damage (8-hydroxy-2'-deoxyguanosine (8-OHdG)), lipid peroxidation (8-isoprostaglandin F2alpha (8-isoPGF2α)), and protein oxidative damage (protein carbonyls (PCO)) were detected for all observations. Here we used linear mixed models to estimate the cross-sectional and longitudinal associations between arsenic exposure and oxidative damage. Exposure-response curves were constructed by utilizing the generalized additive mixed models with thin plate regressions. After adjusting for potential confounders, arsenic level was significantly and positively related to the levels of global oxidative damage and their annual increased rates in dose-response manners. In cross-sectional analyses, each 1% increase in arsenic level was associated with a 0.406% (95% confidence interval (CI): 0.379% to 0.433%), 0.360% (0.301% to 0.420%), and 0.079% (0.055% to 0.103%) increase in 8-isoPGF2α, 8-OHdG, and PCO, respectively. More importantly, arsenic was further found to be associated with increased annual change rates of 8-isoPGF2α (β: 0.147; 95% CI: 0.130 to 0.164), 8-OHdG (0.155; 0.118 to 0.192), and PCO (0.050; 0.035 to 0.064) in the longitudinal analyses. Our study suggested that arsenic exposure was not only positively related with global oxidative damage to lipid, DNA, and protein in cross-sectional analyses, but also associated with annual increased rates of these biomarkers in dose-dependent manners.