BACKGROUND: Accurate dosing of propofol in children requires accurate knowledge of propofol pharmacokinetics in this population. Improvement in pharmacokinetic accuracy may depend on the incorporation of individual patient factors into the pharmacokinetic model or the use of population approaches to estimating the pharmacokinetic parameters. We investigated whether incorporating individual subject covariates (e.g., age, weight, and gender) into the pharmacokinetic model improved the accuracy. We also investigated whether the use of a mixed-effects population model (e.g., the computer program NONMEM) improved the accuracy of the pharmacokinetic model beyond the accuracy obtained with models estimated using two simple approaches. METHODS: We studied 53 healthy, unpremedicated children (28 boys and 25 girls) ranging from 3 to 11 yr of age. Twenty children only received an initial loading dose of 3 mg/kg intravenous propofol. In the remaining 33 children, an initial intravenous propofol dose of 3.5 mg/kg was followed by a propofol maintenance infusion. Six hundred fifty-eight venous plasma samples were gathered and assayed for propofol concentrations. Three different regression techniques were used to analyze the pharmacokinetics: the "standard two-stage" approach, the "naive pooled-data" approach, and the nonlinear mixed-effects modeling approach (as implemented in NONMEM). In both the pooled-data and mixed-effects approaches, individual covariates (age, weight, height, body surface area, and gender) were added to the model to examine whether they improved the quality of the fit. Accuracy of the model was measured by the ability of the model to describe the observed concentrations. RESULTS: The pharmacokinetics of propofol in children were best described by a three-compartment pharmacokinetic model. There were no appreciable differences among the pharmacokinetics estimated using the two-stage, pooled-data, and mixed-effects approaches. Weight was a significant covariate, and the weight-proportional model was supported by all three regression approaches. The pharmacokinetic parameters of the weight-proportional pharmacokinetic model (pooled-data approach) were: central compartment (V1) = 0.52 1 x kg-1; rapid-distribution compartment (V2) = 1.01 x kg-1; slow-distribution compartment (V3) = 8.2 1 x kg-1; metabolic clearance (Cl1) = 34 ml.kg-1 x min-1; rapid-distribution clearance (Cl2) = 58 ml.kg-1 x min-1; and slow-distribution clearance (Cl3) = 26 ml.kg-1 x min-1. The inclusion of age as an additional covariate of V2 statistically improved the model, but the actual improvement in the fit was small. CONCLUSIONS: The pharmacokinetics of propofol in children are well described by a standard three-compartment pharmacokinetic model. Weight-adjusting the volumes and clearances significantly improved the accuracy of the pharmacokinetics. Adjusting the pharmacokinetics for inclusion of additional patient covariates or using a mixed-effects model did not further improve the ability of the pharmacokinetic parameters to describe the observations.
BACKGROUND: Accurate dosing of propofol in children requires accurate knowledge of propofol pharmacokinetics in this population. Improvement in pharmacokinetic accuracy may depend on the incorporation of individual patient factors into the pharmacokinetic model or the use of population approaches to estimating the pharmacokinetic parameters. We investigated whether incorporating individual subject covariates (e.g., age, weight, and gender) into the pharmacokinetic model improved the accuracy. We also investigated whether the use of a mixed-effects population model (e.g., the computer program NONMEM) improved the accuracy of the pharmacokinetic model beyond the accuracy obtained with models estimated using two simple approaches. METHODS: We studied 53 healthy, unpremedicated children (28 boys and 25 girls) ranging from 3 to 11 yr of age. Twenty children only received an initial loading dose of 3 mg/kg intravenous propofol. In the remaining 33 children, an initial intravenous propofol dose of 3.5 mg/kg was followed by a propofol maintenance infusion. Six hundred fifty-eight venous plasma samples were gathered and assayed for propofol concentrations. Three different regression techniques were used to analyze the pharmacokinetics: the "standard two-stage" approach, the "naive pooled-data" approach, and the nonlinear mixed-effects modeling approach (as implemented in NONMEM). In both the pooled-data and mixed-effects approaches, individual covariates (age, weight, height, body surface area, and gender) were added to the model to examine whether they improved the quality of the fit. Accuracy of the model was measured by the ability of the model to describe the observed concentrations. RESULTS: The pharmacokinetics of propofol in children were best described by a three-compartment pharmacokinetic model. There were no appreciable differences among the pharmacokinetics estimated using the two-stage, pooled-data, and mixed-effects approaches. Weight was a significant covariate, and the weight-proportional model was supported by all three regression approaches. The pharmacokinetic parameters of the weight-proportional pharmacokinetic model (pooled-data approach) were: central compartment (V1) = 0.52 1 x kg-1; rapid-distribution compartment (V2) = 1.01 x kg-1; slow-distribution compartment (V3) = 8.2 1 x kg-1; metabolic clearance (Cl1) = 34 ml.kg-1 x min-1; rapid-distribution clearance (Cl2) = 58 ml.kg-1 x min-1; and slow-distribution clearance (Cl3) = 26 ml.kg-1 x min-1. The inclusion of age as an additional covariate of V2 statistically improved the model, but the actual improvement in the fit was small. CONCLUSIONS: The pharmacokinetics of propofol in children are well described by a standard three-compartment pharmacokinetic model. Weight-adjusting the volumes and clearances significantly improved the accuracy of the pharmacokinetics. Adjusting the pharmacokinetics for inclusion of additional patient covariates or using a mixed-effects model did not further improve the ability of the pharmacokinetic parameters to describe the observations.
Authors: Chenguang Wang; Mariska Y M Peeters; Karel Allegaert; Heleen J Blussé van Oud-Alblas; Elke H J Krekels; Dick Tibboel; Meindert Danhof; Catherijne A J Knibbe Journal: Pharm Res Date: 2012-06 Impact factor: 4.200
Authors: S H Kang; M R Poynton; K M Kim; H Lee; D H Kim; S H Lee; K S Bae; O Linares; S E Kern; G J Noh Journal: Br J Clin Pharmacol Date: 2007-02-23 Impact factor: 4.335