Prenatal current-use pesticide exposure and children's neurodevelopment at one year of age in the Infants' Environmental Health (ISA) birth cohort, Costa Rica

Abstract

Background: Pesticide exposure may affect young children's neurodevelopment, but only few cohort studies have addressed possible effects of non-organophosphate pesticides.

Objective: We evaluated associations between prenatal current-use pesticide exposure and neurodevelopmental outcomes among 1-year-old children from the Infants' Environmental Health (ISA) birth cohort.

Methods: To determine prenatal pesticide exposure, we measured biomarkers of pyrimethanil, chlorpyrifos, synthetic pyrethroids, and 2,4-D in urine samples among 355 women, 1-3 times during pregnancy. One-year post-partum, we evaluated children's neurodevelopment with the Bayley Scales of Infant and Toddler Development 3rd edition (BSID-III). We assessed associations between exposures and neurodevelopmental outcomes (composite and z-scores) using single-chemical linear regression models adjusted for possible confounders (maternal education, parity, sex, gestational age at birth, child age, HOME-score, location of assessment, biomarkers of mancozeb), and studied effect-modification by sex. We evaluated non-linear associations of multiple pesticide exposures with Bayesian kernel machine regression (BKMR).

Results: We found higher prenatal urinary 2,4-D concentrations were associated with lower language (β_{per ten-fold increase} = -2.0, 95 % confidence interval (CI) = -3.5, -0.5) and motor (β_{per ten-fold increase} = -2.2, 95 %CI = -4.2, -0.1) composite scores among all children. Also, higher chlorpyrifos exposure [measured as urinary 3,5,6-trichloro-2-pyridinol (TCPy)] was associated with lower cognitive composite scores (β_{per ten-fold increase} = -1.9, 95 %CI = -4.7, 0.8), and lower motor composite scores among boys (β_{per ten-fold increase} = -3.8, 95 % CI = -7.7, 0.1) but not girls (β_{per ten-fold increase} = 2.3, 95 %CI = -1.6, 6.3, pINT = 0.11). Finally, higher pyrimethanil was associated with lower language abilities among girls, but not boys. Pyrethroid metabolite concentrations did not explain variability in BSID-III composite scores. Associations were similar for BSID-III z-scores, and we found no evidence for non-linear associations or mixture effects.

Discussion: Prenatal exposure to common-use pesticides may affect children's neurodevelopment at 1-year of age, some effects may be sex-specific.

Keywords: 2,4-D; Chlorpyrifos; Epidemiology; In-utero; Pesticides; Pyrimethanil.