Prenatal Exposure to Organophosphorous Pesticides and Fetal Growth: Pooled Results from Four Longitudinal Birth Cohort Studies

Abstract

Background: Organophosphorous (OP) pesticides are associated with reduced fetal growth in animals, but human studies are inconsistent.

Objectives: We pooled data from four cohorts to examine associations of prenatal OP exposure with birth weight (n = 1,169), length (n = 1,152), and head circumference (n = 1,143).

Methods: Data were from the CHAMACOS, HOME, Columbia, and Mount Sinai birth cohorts. Concentrations of three diethyl phosphate (ΣDEP) and three dimethyl phosphate (ΣDMP) metabolites of OP pesticides [summed to six dialkyl phosphates (ΣDAPs)] were measured in maternal urine. Linear regression and mixed-effects models were used to examine associations with birth outcomes.

Results: We found no significant associations of ΣDEP, ΣDMP, or ΣDAPs with birth weight, length, or head circumference overall. However, among non-Hispanic black women, increasing urinary ΣDAP and ΣDMP concentrations were associated with decreased birth length (β = -0.4 cm; 95% CI: -0.9, 0.0 and β = -0.4 cm; 95% CI: -0.8, 0.0, respectively, for each 10-fold increase in metabolite concentration). Among infants with the PON1192RR genotype, ΣDAP and ΣDMP were negatively associated with length (β = -0.4 cm; 95% CI: -0.9, 0.0 and β = -0.5 cm; 95% CI: -0.9, -0.1).

Conclusions: This study confirms previously reported associations of prenatal OP exposure among black women with decreased infant size at birth, but finds no evidence of smaller birth weight, length, or head circumference among whites or Hispanics. Contrary to our hypothesis, we found stronger inverse associations of DAPs and birth outcome in infants with the less susceptible PON1192RR genotype. The large pooled data set facilitated exploration of interactions by race/ethnicity and PON1 genotype, but was limited by differences in study populations.

Citation: Harley KG, Engel SM, Vedar MG, Eskenazi B, Whyatt RM, Lanphear BP, Bradman A, Rauh VA, Yolton K, Hornung RW, Wetmur JG, Chen J, Holland NT, Barr DB, Perera FP, Wolff MS. 2016. Prenatal exposure to organophosphorous pesticides and fetal growth: pooled results from four longitudinal birth cohort studies. Environ Health Perspect 124:1084-1092; http://dx.doi.org/10.1289/ehp.1409362.

Figure 1

Association of 10-fold increase in maternal prenatal urinary total DAP, DMP, and DEP metabolites with infant birth weight (A), length (B), and head circumference (C) in each cohort and in the pooled data set. Models were adjusted for sex, race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, other), country of origin (USA, other), marital status (married/living as married, single), maternal education (< high school, high school graduate), smoking during pregnancy, parity (nulliparous, multiparous), maternal age at delivery, and gestational age (spline). Interaction p-values from Wald tests on cross-product terms.

Figure 2

Association of 10-fold increase in maternal prenatal urinary total DAP, DMP, and DEP metabolites with infant birth weight (A), length (B), and head circumference (C) in the pooled data set, stratified by race/ethnicity. Models adjusted for cohort, sex, country of origin (USA, other), marital status (married/living as married, single), maternal education (< high school, high school graduate), smoking during pregnancy, parity (nulliparous, multiparous), maternal age at delivery, and gestational age (spline). Participants of “other” race/ethnicity not included in figure. Interaction p-values from Wald tests on cross-product terms.

Figure 3

Association of 10-fold increase in maternal prenatal urinary total DAP, DMP, and DEP metabolites with infant birth weight, length, and head circumference in the pooled data set, stratified by infant PON_–108 genotype (A–C) and infant PON₁₉₂ genotype (D–F). Models adjusted for cohort, sex, race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, other), country of origin (USA, other), marital status (married/living as married, single), maternal education (< high school, high school graduate), smoking during pregnancy, parity (nulliparous, multiparous), maternal age at delivery, and gestational age (spline). Interaction p-values from Wald tests on cross-product terms.