External validation of pharmacokinetic and pharmacodynamic models of microemulsion and long-chain triglyceride emulsion propofol in beagle dogs

This study aimed at assessing the predictive performance of a target-controlled infusion (TCI) system, which incorporates canine PK-PD models for microemulsion and long-chain triglyceride emulsion (LCT) propofol and at investigating time independency of propofol effect on the observed electroencephalographic approximate entropy (ApEn) in TCI. Using a crossover design with a 7-day washout period, 28 healthy beagle dogs were randomized to receive TCI of both formulations in a stepwise or constant manner. Plasma propofol concentrations and ApEn were measured at preset intervals. Pooled biases, inaccuracies, divergences, and wobbles in pharmacokinetic and pharmacodynamic predictions were 2.1% (95% CI: -0.8 to 4.9), 18.1% (15.6-20.5), 1.9%/h, 7.3% (5.4-9.3), and -0.5% (-2.6 to 1.6), 8.7% (7.3-10.1), 2.5%/h, 6.0% (4.1-7.2) for microemulsion propofol, and -9.3% (-11.6 to -6.9), 20.1% (18.2-22.0), 5.1%/h, 7.6% (6.1-9.1) and 5.6% (4.1-7.1), 8.0% (6.9-9.3), 4.7%/h, 4.1% (3.1-5.1) for LCT propofol. Observed ApEn values over time were statistically not different across all time points in a TCI with constant manner. Canine PK-PD model of microemulsion propofol showed good predictive performances. Propofol effect (ApEn) was time independent as long as time is allowed for equilibration.