Maternal-infant biomonitoring of environmental chemicals: the epidemiologic challenges

Abstract

There is growing concern about the potential health effects of exposure to various environmental chemicals during pregnancy and infancy. One of the key limitations of past epidemiologic research in this field has been the potential for exposure misclassification to lead to biases in the health risk estimate. The use of biomarkers in pregnancy cohort or case-control studies has significantly advanced the field; however, this is true only if the biomarker is a true measurement of exposure for the relevant time period of interest. There are a number of theoretical and practical constraints to their use, including difficulty interpreting biomonitoring data, high analytical and collection costs, potential participant selection biases, and ethical challenges in reporting results to study subjects. Identifying a representative sample and collecting biospecimens in the developmental window of interest can be problematic. Various strategies for identifying pregnant women range from the more representative but least efficient sampling of the general population to recruitment through early ultrasound clinics and local advertising. Whereas measurement of xenobiotic chemicals in cord blood, amniotic fluid, or meconium provides unequivocal evidence that the chemical has entered the fetal environment, analysis of maternal blood and urine can be used as a surrogate for fetal exposure. Use of stored midpregnancy serum collected for fetal screening and of large-cohort biobanks offer unique opportunities for biomonitoring data for birth defects studies. Future research is needed to explore less invasive matrices for biomonitoring of infants and to develop less costly analytical methods that require smaller sample volumes.