In utero exposure to polychlorinated biphenyls (PBCs) and related compounds has been linked to a specific pattern of developmental deficits, including growth retardation, poorer short-term memory, and less efficient cognitive processing, in preschool-aged children born to women who consume Lake Michigan fish. Lactation exposure is associated with reduced activity level in these and two other cohorts of Michigan children. The proposed study is designed (1) to evaluate levels of specific PCB congeners and related compounds in these children and changes in total PCB body burden between the ages of 4 and 10 years; (2) to determine whether the early deficits in cognitive functioning continue to be evident in middle childhood and whether they affect the acquisition of reading and quantitative skills in school; and (3) to determine whether these exposures are associated with reduced activity or delays in growth or sexual maturation during puberty. Two cohorts of Michigan children at risk for PCB exposure, previously seen as infants and/or at 4 years, will be evaluated at 10 years. Blood samples will be obtained and evaluated for total PCBs by packed column gas chromatography (GC), for specific PCB congeners by dual capillary column GC, and for polychlorinated dibenzofurans and dibenzo-p-dioxins by capillary column GC/mass spectrometry. Assessments of intellectual functioning will include standardized tests of IQ and academic achievement and four tasks focusing on specific aspects of memory and processing efficiency. Activity level will be assessed by direct behavioral observation and standardized parent and teacher rating scales. Weight, height, and head circumference will be measured and pubertal stage evaluated using Tanner's criteria. A broad range of potential confounders will be assessed, including socioeconomic status, quality of parental intellectual stimulation, perinatal medical complications, maternal alcohol consumption and smoking during pregnancy, and exposure to other contaminants, including lead and mercury. Toxic effects will be inferred only where exposure is significantly related to outcome after statistical adjustment for all relevant potential confounders. Developmental outcomes will be assessed in relation to both current body burden and previously ascertained cord blood and maternal milk PCB levels to evaluate the relative contributions of intrauterine and postnatal exposures.
