Residential environmental and biomarker measurements were made of multiple pollutants during two seasons (spring and summer, 1993) in order to assess human exposure for a purposeful sample of 18 nonsmoking adults residing within nine homes (a primary and secondary subject in each home) in the Lower Rio Grande Valley (LRGV) near Brownsville, TX. Pesticides, metals, PAHs, VOCs, and PCBs were measured in drinking water, food, air, soil, and house dust over a one-to two-day period in each season. Biomarker measurements were made in blood, breath, and urine. A total of 375 measurements across five pollutant classes (227 pesticides, 44 trace elements, 78 VOCs, 18 PAHs, and 8 PCBs) was possible for each home in one or more media. A large percentage of the measurements was below the method limit of detection ranging from 0-37% for pesticides, 22-61% for metals, 6% and 90% for VOCs in water and air, respectively, and 0-74% for PAHs. The total number of analytes measurable in blood, urine, or breath was considerably less, i.e., 58 (21 pesticides, 1 PCB, 4 metals, 31 VOCs, and 1 PAH) with the percentage above the method limit of detection for pesticides and metals ranging from 40 to 100%, while for VOCs, PAHs, and PCBs, this percentage ranged from 2 to 33%. A significant seasonal difference (p less than or equal to 0.10) was found in the biomarker levels of two of seven nonpersistent pesticides (3,5,6-trichloro-2-pyridinol and 2,5-dichlorophenol) and 3 of 3 metals (arsenic, cadmium, and mercury) and the pyrene metabolite, 1-hydroxypyrene, measured in urine. In all cases, levels were higher in the summer relative to the spring. For the persistent pesticides and PCBs in blood serum, a seasonal effect could be evaluated for 5 of 10 analytes; a significant difference (p less than or equal to 0.10) was observed only for hexachlorobenzene, which like the urine biomarkers, was higher in the summer. In contrast to the urine metals, blood-Pb concentrations did not change significantly (p less than or equal to 0.05) from spring to summer. Biological results from the current study are compared to the reference range furnished by the Third National Health and Nutrition Examination Survey (NHANES III). Comparisons are only suggestive due to limitations in comparability between the two studies. Based on the percentage of measurements above the detection limit, a significant elevation (p less than or equal to 0.10) in 2 of 12 nonpersistent pesticides (4-nitrophenol and 2,4-D) was observed for the LRGV study subjects. The VOC carbon tetrachloride was found in the blood (monitored only in spring) with greater prevalence (p less than or equal to 0.10) than would be expected from NHANES III results. Blood serum levels of two persistent pesticides (4,4'-DDE, and trans-nonachlor) and PCB exceeded median and/or 95th percentile reference levels as did arsenic in urine. Where seasonal differences were identified or for compounds exceeding reference levels, environmental monitoring results were investigated to identify potential contributing pathways and sources of exposure. However, because environmental sampling did not always coincide with the biological sampling and because of the high frequency of analytes measured below the limit of detection, sources and pathways of exposure in many cases could not be explained. Chlorpyrifos was an exception where urine metabolite (3,5,6-TCP) levels were found to be significantly correlated with air (R-2=0.55; p less than or equal to 0.01) and dust (R-2=0.46; p less than or equal to 0.01) concentrations. Based on the results of biomarker and residential environmental measurements over two seasons, this scoping study shows a seasonal effect for some analytes and suggests where exposures may be high for others. This information may be useful in considering future studies in the region. Published by Elsevier Science Ltd.
