Development and evaluation of a physiologically based pharmacokinetic model of fentanyl in dogs

The author developed and evaluated a physiologically based pharmacokinetic model of fentanyl in dogs. The model was developed utilizing tissue partitioning information derived from published values for the rat and included radiolabeled microsphere blood flow and tissue weights measured in experimental subjects. In order to validate the physiologically based model during a variety of blood flow states, one half of the dogs received a continuous rate intravenous infusion (1.5 μg/kg/h) of the alpha-2 adrenergic agonist, medetomidine during the fentanyl sampling period. A bolus intravenous dose fentanyl (15 μg/kg of fentanyl base) was administered to both unmedicated conscious dogs and dogs sedated with a continuous rate intravenous infusion of medetomidine. Arterial blood fentanyl concentrations were measured at 2, 5, 7.5, 10, 15, 30, 45, 60, 90, 120, 360, 480, and 720 min following fentanyl administration by high pressure liquid chromatography with tandem mass spectroscopy. Model simulations of fentanyl arterial blood concentrations were performed for the same sampling period using individual dog data for tissue blood flow and organ volume as model inputs, for comparison. Area under the fentanyl blood concentration verses time curve from injection until the concentrations fell below the assay limit of quantification (AUC_(0-LOQ)) for simulated and measured concentrations were compared using a test for average bioequivalence.; Results of the bioequivalence analysis indicated the model was not equivalent. The model was examined using time averaged inputs of tissue blood flows and using stepped changes in tissue blood flow representing measured changes in individual experimental subjects. Analysis of partial areas under the curve suggested the model was better at predicting the elimination than the distribution of fentanyl, although the partial areas under the curve were still not equivalent.