Residential greenness exposure has been reported to positively impact health mainly by reducing overweight/obesity risk, improving mental health and physical activity. Less is known on biological pathways involved in these health benefits. We aimed to investigate the association between multisite greenness exposure and oxidative stress in children and explore the potential mediating role of physical activity in this association. This cross-sectional study involved 323 healthy subjects (8-11 y) from five schools in Asti (Italy). Children's parents filled a questionnaire providing the residential address, parental education, and physical activity frequency. Oxidative stress was quantified in spot urine by isoprostane (15-F2t-IsoP) using ELISA technique. Residential and scholastic greenness were defined by the Normalized Difference Vegetation Index (NDVI) in buffers with 100, 250, 300, 500 and 1000 m radii, and vegetated portion was also estimated. Multisite exposures were derived accounting for NDVI around home and school, weighted for time spent in each location. Linear mixed models, age-adjusted, with schools as random intercept, tested the association between 500 m-radius buffer multisite grenness variables and log (15-F2t-IsoP), reporting decreased oxidative stress per interquartile range (IQR) increase in multisite NDVI (β: 012, 95%CI -0.240 to 0.004) and multisite vegetated portion (β: 0.14, 95%CI -0.270 to -0.006). The mediation analysis did not support the hypothesis that physical activity frequency could mediate these associations. Multisite greenness exposure is associated with decreased oxidative stress in children and our data did not support the mediating role of physical activity.
