Determinants of children's exposure to pyrethroid insecticides in western France.

Pyrethroids are insecticides frequently used in agriculture and in the home; exposure occurs through dietary and non-dietary pathways, including indoor and outdoor environmental contamination. Our objective was to study the potential determinants of pyrethroid metabolite concentrations measured in children's urine samples and in the dust of their homes. Specifically, we measured urinary metabolites from morning spot samples of 245 six-year-old children living in Brittany (France) in 2009-2012 and from dust vacuumed from the floor of their homes. Mothers reported home insecticide use, dietary habits, sociodemographic data; residential and school proximity to agricultural crops was assessed with spatialized data. The metabolites cis-DBCA, trans-DCCA, cis-DCCA, 3-PBA, and F-PBA were detected in 84, 95, 64, 63, and 16% of the urine samples, respectively. Permethrin, cypermethrin, cyfluthrin, deltamethrin, and tetramethrin pyrethroids were detected in 100, 56, 9, 15, and 26% of the dust samples, respectively. Multiple regression analysis suggested diet plays a role in children's exposure, in particular, the food groups "pasta, rice or semolina" (for cis-DCCA and F-PBA), fruit (3-PBA), "breakfast cereals and whole grain bread" (cis-DBCA), and the global proportion of organic food in diet (for cis-DBCA, trans-DCCA). Children with a parent occupationally exposed to pesticides were about 3-times more likely to have higher urinary concentrations of 3-PBA (OR=2.8, 95% CI [1.2; 6.5]). Dust content was correlated mainly with household insecticide use: higher mean concentrations of permethrin (β=0.8 [0.3; 1.3], in μg/g) and an increased risk of a detectable level of cyfluthrin (OR=4.7 [1.7; 12.9]) were observed in home dust, for indoor use of at least twice a year. Outdoor insecticide use at least once a year was associated with detection in dust of cypermethrin (OR=3.0 [1.3; 6.7]) and tetramethrin (OR=3.7 [1.6; 8.3]). Three positive and one negative correlations (out of 11) between urinary metabolite concentrations and home dust contents of their possible corresponding parent compounds were observed. The strength of this study lies in its concurrent use of biomarkers, environmental measurements, and potential sources of exposure. Its limitations include the use of a single urine sample and imprecise data about pyrethroid use in local agriculture.