Prenatal Bisphenol A Exposure and Cognitive and Behavioural Development in Children

Abstract

Background: Experimental evidence suggests that bisphenol A (BPA) may be a neurodevelopmental toxicant. Several small epidemiological studies concluded that prenatal BPA exposure may affect child behavioural development in a sex-specific manner, but findings are not consistent. Aims: To examine the effects of prenatal BPA exposure on cognitive and behavioural development in children at age 14 months and 4 years from the Spanish INMA birth cohort. Methods: BPA concentrations were measured in two maternal spot urine samples collected in trimester 1 and 3 of pregnancy; the average of the two creatinine-adjusted BPA concentrations was used as exposure variable. Cognitive and psychomotor development were assessed at 14 months with the Bayley Scales of Infant Development and at 4 years with the McCarthy Scales of Children's Abilities. At 4 years, attention deficit hyperactivity disorder (ADHD) symptoms were assessed with the DSM-IV scales and autism symptoms with the Childhood Autism Spectrum Test. Analyses used linear, logistic and negative binomial regression models and included 349 mother-child pairs. Results: BPA concentrations were 2.6 ?g/g creatinine in 1stand 2.0 ?g/g in 3rd trimestersamples (correlation=0.14, p<0.01). BPA exposure was associated with an increased risk of ADHD-hyperactivity symptoms (IRR per log10?g/g creatinine=2.71; 95%CI 1.39-5.26) and a reduction in psychomotor development (b=-4.28; -8.25, -0.32); these associations were strongest in boys. In girls, BPA was related to an increased risk of autism symptoms (OR=13.1; 1.03, 166), based on only 11 girls with symptoms. BPA exposure was not associated with the mental development score at age 14 months, or with the McCarthy general cognition score or its sub-scores at 4 years. Conclusion: Prenatal BPA exposure may affect children’s behavioural development, but does not appear to affect cognitive development. Uncertainties in BPA exposure estimation may have biased results towards the null hypothesis.