Advances in Canine Anesthesia: Physiologically Based Pharmacokinetic Modeling for Predicting Propofol Plasma Profiles in Canines with Hepatic Impairment

Abstract

Background: A PBPK model allows the prediction of the concentration of drug amounts in different tissues and organs over time and can be used to simulate and optimize different therapeutic protocols in healthy and sick individuals. The objective of this work was to create a PBPK model to predict propofol doses for healthy canines and canines with hepatic impairment. Methods: The study methodology was divided into two major phases, in which the first phase consisted of creating the PBPK model for healthy canines, and in the second phase, this model was adjusted for canines with hepatic impairment. The model for healthy canines presented good predictive performance, evidenced by the value of the performance measure of the geometric mean fold error that ranged from 0.8 to 1.25, meeting the double error criterion. The simulated regimen for healthy canines, i.e., of 5 mg/kg (administered as a bolus) followed by a continuous infusion at a rate of 0.13 mg/kg/min, was sufficient and ensured that all simulated subjects achieved the target plasma concentration. Canines with 60% and 40% liver function had infusion rate adjustments to ensure that individuals did not exceed the therapeutic window for maintenance of anesthesia. Results: The results presented in this manuscript are suggestive of the effectiveness and practicality of a PBPK model for propofol in canines, with a particular focus on hepatic impairment.

Keywords: PK model; dose individualization; precision anesthesia.

Figure 1

Adjustment of observed and predicted data by the pharmacokinetic model [9,10,11,12]. (a–d) for construction and [13,14,15,16,17] (e–i) for validation. Blue, venous blood; red, arterial blood. Population simulations (n = 1000) are shown as solid lines with shaded areas (geometric mean and geometric standard deviation). The observed data are shown in circles ± standard deviation.

Figure 1

Adjustment of observed and predicted data by the pharmacokinetic model [9,10,11,12]. (a–d) for construction and [13,14,15,16,17] (e–i) for validation. Blue, venous blood; red, arterial blood. Population simulations (n = 1000) are shown as solid lines with shaded areas (geometric mean and geometric standard deviation). The observed data are shown in circles ± standard deviation.

Figure 2

Observed and predicted values. (a) Plasma concentration; (b) area under the curve from the first to the last point; and (c) maximum plasma concentration.

Figure 3

Sensitivity analysis demonstrating the impact of changes in different parameters under the model parameters.

Figure 4

Median plasma concentration of various simulated populations 12 h after a 5 mg/kg bolus administration over 30 s followed by a 0.13 mg/kg/min infusion over 3 h. Dashed lines represent the target concentration range of 2.5 to 4.7 μg/mL. Solid black line represents the target concentration for anesthetic recovery.

Figure 5

The median plasma concentration of various simulated populations 12 h after the administration of the protocol adjusted for each group. Dashed lines represent the target concentration range of 2.5 to 4.7 μg/mL. Solid black line represents the target concentration for anesthetic recovery.

Figure 6

Population distribution of maximum plasma concentrations achieved after dose adjustment in canines with different levels of hepatic impairment, including healthy canines. Dashed lines represent the target concentration range of 2.5 to 4.7 μg/mL.