Comparison of Trihalomethanes in Tap Water and Blood: A Case Study in the United States

BACKGROUND Epidemiological studies have used various measures to characterize trihalomethane (THM) exposures, but the relationship of these indicators to exposure biomarkers remains unclear. OBJECTIVES We examined temporal and spatial variability in baseline blood THM concentrations and assessed the relationship between these concentrations and several exposure indicators (tap water concentration, water-use activities, multiroute exposure metrics). METHODS We measured water-use activity and THM concentrations in blood and residential tap water from 150 postpartum women from three U.S. locations. RESULTS Blood ΣTHM [sum of chloroform (TCM), bromodichloromethane (BDCM), dibromo-chloromethane (DBCM), and bromoform (TBM)] concentrations varied by site and season. As expected based on variable tap water concentrations and toxicokinetic properties, the proportion of brominated species (BDCM, DBCM, and TBM) in blood varied by site (site 1, 24%; site 2, 29%; site 3, 57%) but varied less markedly than in tap water (site 1, 35%; site 2, 75%; site 3, 68%). The blood-water ΣTHM Spearman rank correlation coefficient was 0.36, with correlations higher for individual brominated species (BDCM, 0.62; DBCM, 0.53; TBM, 0.54) than for TCM (0.37). Noningestion water activities contributed more to the total exposure metric than did ingestion, but tap water THM concentrations were more predictive of blood THM levels than were metrics that incorporated water use. CONCLUSIONS Spatial and temporal variability in THM concentrations was greater in water than in blood. We found consistent blood-water correlations across season and site for BDCM and DBCM, and multivariate regression results suggest that water THM concentrations may be an adequate surro-gate for baseline blood levels.