Gastrointestinal absorption of xenobiotics in physiologically based pharmacokinetic models. A two-compartment description.

Gastrointestinal (GI) absorption in physiologically based pharmacokinetic models is typically described as first-order transfer from one compartment directly into the liver. This approach appears to be adequate for water but not for oily vehicles. In this study, a two-compartment description of GI absorption was developed in which the chemical moved from the first compartment to the second (first-order rate constant KT) and was absorbed from both compartments (first-order rate constants KAS and KAD, respectively) into the liver. Rat blood time-course data sets for methylene chloride, chloroform, dichloroethane, and trichloroethylene after oral gavage in water or corn oil obtained from the literature were used for model validation. Optimization of the KAS, KAD, and KT values for each dosing solution allowed accurate simulation of each data set. In general, the KAS values were 3-4 times greater when water rather than corn oil vehicle was used. The KAD and KT values were similar for the two vehicles. By comparison, a one-compartment description resulted in a poor simulation of the oil gavage data. The two-compartment model rate constant values obtained by optimization of the rat blood time-course data set for trichloroethylene after oral gavage in water were used in the model to predict rat exhaled breath concentrations after oral gavage of trichloroethylene in water (data generated in this laboratory). Exhaled breath trichloroethylene concentrations compared favorably with model predictions.