Partition coefficients for the trihalomethanes among blood, urine, water, milk and air.

Chloroform, bromodichloromethane, chlorodibromomethane, and bromoform comprise the trihalomethanes, a group of widespread and mildly lipophilic compounds that result from water chlorination and other sources. Many animal studies show the chronic toxicity and carcinogenicity of these compounds, and recent work has demonstrated the importance of both ingestion and inhalation exposure pathways. This study presents partition coefficients describing the equilibrium among biological compartments (air, water, blood, milk, urine) for the four THMs based on results of headspace gas chromatographic analyses performed under equilibrium conditions and at 37 degrees C. The calculated partition coefficients ranged from 2.92 to 4.14 for blood/water, 1.54-2.85 for milk/blood, and 3.41-4.93 for blood/urine, with the lowest being chloroform and the highest being bromoform. Both human and cow milk were tested, with similar results. The available samples of human milk may not fully account for differences in lipid content and possibly other factors that affect estimates of partition coefficients. Simultaneous measurements of milk and blood in exposed individuals are suggested to confirm laboratory results. Partition coefficients are predicted using the octanol-air partition coefficient, also measured in this study, and the octanol-water partition coefficient. Results are similar to literature estimates for liquid/air partitioning of chloroform and chlorodibromomethane, but they differ from predictions based on hydrophobicity and lipid content. High correlations between the derived partitioned coefficients and the molecular structure (number of Br atoms) and physical properties (molecular weight and boiling point) are found for these analogous chemicals. In humans, THMs are both stored and metabolized with relatively rapid clearance rates. The derived partition coefficients can help to interpret results of biological monitoring and predict the potential for the accumulation and transfer of chemicals, specifically by the application of physiologically-based pharmacokinetic models. THM exposures to potentially susceptible populations, e.g. nursing infants, can be predicted using either such models.