We conducted a cross-sectional epidemiological study to evaluate personal exposure to nitrogen dioxide and its effect on blood antioxidants. Personal exposure of 107 volunteers was assessed for 14 d with passive monitors. We excluded heavy smokers (> 10 cigarettes/d) from the study. Sociodemographic and environmental data, as well as beta-carotene intake, were recorded. We mainly attributed the mean nitrogen dioxide personal exposure (31.9 +/- 12.7 microg/m3 [0.017 ppm or 0.70 microM/m3]) (R2 = 0.75) to residence site in the city, time spent in urban traffic, and use of gas stoves. The correlation between nitrogen dioxide exposure and blood antioxidant concentration was weak; in addition, the correlation coefficients for men and women were inconsistent. Nonetheless, we found some evidence of an interaction between carotene intake and nitrogen dioxide exposure: a significantly lower plasma beta-carotene level was evident among subjects who consumed < or = 4.5 mg/jour of carotene and who were exposed to nitrogen dioxide levels that exceeded 40 microg/m3 (0.021 ppm or 0.87 microM/m3) of nitrogen dioxide.