ABSTRACT
A PB-PK model has been developed for nasal nonreactive vapor uptake in the F344 rat which incorporates nasal enzyme distribution as well as nasal airflow patterns. Nasal tissue is separated into respiratory and olfactory mucosal areas with each area containing mucus, epithelial, and submucosal compartments. Metabolic activities are distributed among the compartments in accordance with published histochemical data, and intercompartmental transfer rate constants are based on molecular diffusivity. Two airflow paths are assumed: a lateral/ventral path which passes over only respiratory mucosa and a dorsal medial path which passes over both respiratory and olfactory mucosa. This model was fit to F344 rat nasal uptake data obtained for five vapors: acetone (an unmetabolized vapor), isoamyl alcohol (alcohol dehydrogenase substrate), ethyl acetate (carboxylesterase substrate), and o-xylene and bromobenzene (mixed function oxidase substrates). These vapors cover a wide span of uptake values ranging from a fractional uptake of less than 0.01 for xylene to 0.80 for isoamyl alcohol. The model accurately predicted the fractional uptake of these vapors, the maximal deviation between actual and predicted values being 0.04. The best fit for the data was obtained assuming: (1) an asymmetric blood flow pattern (with the majority of the blood flow being apportioned to the respiratory mucosa), and (2) only a small fraction of the total airflow (8%) following the dorsal medial pathway and penetrating to the olfactory mucosa. Since only a small fraction of the inspired air passes over the olfactory mucosa, a site with high metabolic activity, an inspiratory flow rate dependence on inspired vapor metabolism results. At low flows, olfactory metabolism is limited by low vapor delivery to that site, an effect that does not occur at high flows. This highlights the potential importance of regional nasal airflow and enzymatic activity patterns in influencing uptake and metabolism of inspired vapors and indicates the need for physiologically relevant models incorporating these factors.
