Abstract
BACKGROUND: Prenatal exposures to neurotoxic metals and trace elements are associated with early childhood neurodevelopmental outcomes. However, consequences of simultaneous exposure to mixtures of elements remain unclear. OBJECTIVE: To examine individual and joint effects of prenatal trace element exposure on early childhood neurodevelopment. METHODS: Using a well-established Bangladesh prospective birth cohort (2008-2011), we measured concentrations of 52 trace elements in umbilical cord serum of 569 mother-infant pairs using inductively coupled plasma mass spectrometry. Neurodevelopment was evaluated at 20-40 months of age using Bayley Scales of Infant and Toddler Development, Third Edition. Stability elastic net (ENET) was used to screen elements individually associated with the outcome; candidate exposures were combined by weighted linear combination to form a risk score representing their mixture effect on early childhood neurodevelopment. RESULTS: Stability ENET identified 15 trace elements associated with cognitive composite score and 14 associated with motor composite score, which were linearly combined to form the element risk score (ERS). Children with higher ERS(cognitive) had lower probability of cognitive developmental delay (OR(highest vs lowest): 0.21; 95 %CI: 0.10, 0.40; P < 0.001; P(trend) < 0.001). Children with ERS(motor) in the top quintile had a significantly lower risk of motor developmental delay (OR: 0.16; 95 %CI: 0.09, 0.31; P < 0.001; P(trend) < 0.001) versus the lowest quintile. In Bayesian kernel machine regression analyses, lithium [conditional posterior inclusion probability (cPIP) = 0.68], aluminum (cPIP = 0.83) and iron (cPIP = 1.00) contributed most to the lower cognitive composite score; zinc (cPIP = 1.00), silver (cPIP = 0.81), and antimony (cPIP = 0.65) mainly contributed to the change of motor composite score. CONCLUSION: Co-exposure to lithium/aluminum/iron or zinc/silver/antimony appears to impact children's neurodevelopment. ERS score reflecting maternal exposure could indicate children's risk of neurodevelopmental delay, warranting further studies to explore the underlying mechanism.