Potential effect of early blood sampling schedule on calculated pharmacokinetic parameters of drugs after intravenous administration

Performance evaluation of paediatric propofol pharmacokinetic models in healthy young children

BACKGROUND

The performance of eight currently available paediatric propofol pharmacokinetic models in target-controlled infusions (TCIs) was assessed, in healthy children from 3 to 26 months of age.

METHODS

Forty-one, ASA I-II children, aged 3-26 months were studied. After the induction of general anaesthesia with sevoflurane and remifentanil, a propofol bolus dose of 2.5 mg kg(-1) followed by an infusion of 8 mg kg(-1) h(-1) was given. Arterial blood samples were collected at 1, 2, 3, 5, 10, 20, 40, and 60 min post-bolus, at the end of surgery, and at 1, 3, 5, 30, 60, and 120 min after stopping the infusion. Model performance was visually inspected with measured/predicted plots. Median performance error (MDPE) and the median absolute performance error (MDAPE) were calculated to measure bias and accuracy of each model.

RESULTS

Performance of the eight models tested differed markedly during the different stages of propofol administration. Most models underestimated propofol concentration 1 min after the bolus dose, suggesting an overestimation of the initial volume of distribution. Six of the eight models tested were within the accepted limits of performance (MDPE<20% and MDAPE<30%). The model derived by Short and colleagues performed best.

CONCLUSIONS

Our results suggest that six of the eight models tested perform well in young children. Since most models overestimate the initial volume of distribution, the use for TCI might result in the administration of larger bolus doses than necessary.

[Target-controlled infusion. Clinical relevance and special features when using pharmacokinetic models]

Since its commercial introduction in 1996, target-controlled infusion (TCI) has become an established technique for administration of intravenous anaesthetics. Modern TCI systems, however, are characterized by an increasing number of additional options and features, such as the choice between different pharmacokinetic models and modes of application, which may confuse the less experienced user. This review describes the differences between pharmacokinetic models, modes of application and the effect of covariates as well as the consequences for dosing. The aim is to explicate for the user of modern TCI systems the underlying scientific concepts and the relevance for clinical practice.

Development and validation of a recirculatory physiological model of the myocardial concentrations of lignocaine after intravenous administration in sheep

A recirculatory physiological model of the determinants of the myocardial concentrations of lignocaine after intravenous administration was developed in sheep and validated with the intention of analysing and predicting the outcome of altered dose regimens and various pathophysiological states on the initial myocardial concentrations of lignocaine. The structure and parameters of the model were determined by hybrid modelling of the time-courses of the pulmonary artery, arterial and coronary sinus concentrations of lignocaine after the intravenous administration of 100 mg of lignocaine over 5 min to 5 chronically instrumented sheep. The model accounted for the determinants of the myocardial concentrations via compartments for venous mixing, the lung (a single-compartment model with a first-order loss) and the heart (a single flow-limited compartment). Recirculation and the remainder of the body were represented as a single tissue pool with a clearance term. The distribution volume of the heart was 0.42+/-0.009 L, which gave a half-time of myocardium:blood equilibration of 2.37 min. The distribution volume of the lungs was 5.40+/-0.23 L, with an apparent first-order loss of 1.02 L min(-1) representing deep distribution or metabolism. The validity of the model was tested by comparing the predictions of the model with the equivalent data collected in 6 sheep when lignocaine (89 mg) was administered via a complex dose regimen with a faster initial rate of infusion (39.1 mg min(-1)), declining exponentially to basal infusion rate (7.02 mg min(-1)) over 8 min. The predictions of the model were in general agreement with these data. It is concluded that the model was sufficient to account for the effect of altered dose regimens of lignocaine on the time-course of its myocardial concentrations.

Do plasma concentrations obtained from early arterial blood sampling improve pharmacokinetic/pharmacodynamic modeling?

In pharmacokinetic/pharmacodynamic (PK/PD) modeling the first blood sample is usually taken 1 to 2 min after drug administration (late sampling). Therefore, investigators have to extrapolate the plasma concentration to Time 0. Extrapolation, however, erroneously assumes instantaneous and complete mixing of drug in the central volume of distribution. We investigated whether plasma concentrations obtained from early arterial blood sampling would improve PK/PD modeling. In 14 pigs, one of five neuromuscular blocking agents (NMBAs) was administered into the right ventricle within 1 sec and arterial sampling was performed every 1.2 sec (1st min). The response of the tibialis muscle was measured mechanomyographically. The influence of inclusion of data from early arterial sampling on PK/PD modeling was determined. Furthermore, the concentrations in the effect compartment at 50% block (EC50) derived from modeling were compared to the measured concentration in plasma during a steady state 50% block. A very high peak in arterial plasma concentration was seen within 20 sec after administration of the NMBA. Extensive modeling revealed that plasma concentrations obtained from early arterial blood sampling improve PK/PD modeling. Independent of the type of modeling, the EC50 and KeO based on data sets that include early arterial blood sampling were, for all five NMBAs, significantly higher and lower respectively, than those based on data sets obtained from late sampling. Early arterial sampling shows that the mixing of the NMBA in the central volume of distribution is incomplete. A parametric PD (sigmoid Emax) model could not describe the time course of effect of the NMBAs adequately.

Intravenous verapamil kinetics in rats: marked arteriovenous concentration difference and comparison with humans

The pharmacokinetics of verapamil, a calcium channel blocker, were studied in male Sprague-Dawley rats following i.v. administration at a dose of 1 mg kg-1. Both arterial and venous blood were collected and the plasma drug concentrations were determined by reversed-phase high-performance liquid chromatography. Verapamil was distributed to the extravascular tissues very rapidly as indicated by the large Vdss (2.99 +/- 0.57 l kg-1) and Vd beta (5.08 +/- 0.54 l kg-1). The apparent terminal plasma T1/2, MRTiv, and CLp were 1.59 +/- 0.46, 1.26 +/- 0.12 h, and 40.4 +/- 9.73 ml min-1 kg-1, respectively. Marked arterial/venous differences were found with a considerable influence on the MRT and Vdss, and the terminal phase venous levels were higher than arterial levels by 103, 69, and 90%, respectively, for the three rats studied. The distribution of verapamil between plasma and erythrocytes occurred very rapidly and was identical in vitro and in vivo. The average blood to plasma and plasma to blood cell concentration ratios were 0.85 and 1.47, respectively. In contrast to propranolol, blood data rather than plasma data should be used to predict the hepatic extraction ratio of verapamil (0.87). The plasma protein binding of verapamil in humans (90%) and rats (95%) were quite similar and constant over the wide concentration range studied. A comparison of some pharmacokinetic parameters between rats and humans is presented and the potential shortcomings of using T1/2 or CLp and the advantage of using CLu (unbound plasma clearance) in interspecies scaling is also discussed.

An assessment of methods for sampling blood to characterize rapidly changing blood drug concentrations

The accuracy of different blood sampling methods used to characterize rapidly changing blood drug concentrations was examined both in vitro and in vivo. It was shown in vitro that blood sampling methods based on the fraction collection principle failed to characterize a "square wave" change in drug concentration, and there was a 9-16-s delay before achieving 95% of the expected drug concentration. Varying the catheter size and length did not improve the response. This observation is consistent with laminar and/or turbulent flow producing dispersion and mixing of blood of different drug concentrations in the catheter. A sampling method (flush and withdrawal) was developed to minimize these effects. In vivo studies showed that peak blood drug concentrations obtained using this method after an iv bolus of a drug were approximately 25-28% higher than those simultaneously obtained by methods based on fraction collection principles. It is concluded that blood sampling methods based on fraction collection principles can produce significant errors in measured blood drug concentrations. The error is greater the greater the rate of change of the blood drug concentrations.

Preanalytical considerations in drug assays in clinical pharmacokinetic studies

Many hospital pharmacy laboratories undertake drug analysis in biological fluids for the production of pharmacokinetic data. The success of such an undertaking very much depends on the selection of a suitable analytical method and a proper approach to sample collection and handling. This paper surveys the main types of biological specimens taken from the patients for pharmacokinetic drug analysis and discusses factors that affect them during or subsequent to their removal. Guidelines are provided in specimen handling and dealing with many problems which could arise prior to actual analysis. By its very nature this paper brings in many disciplines, the full details of which are well beyond its scope, however, some discussion on pharmacokinetic and bioavailability methods in relation to sampling procedures is included to put the matter into a proper perspective.

N-acetylprocainamide kinetics during intravenous infusions and subsequent oral doses in patients with coronary artery disease and ventricular arrhythmias

The kinetics of N-acetylprocainamide (NAPA) were studied in 5 patients (all men, mean age = 62) with coronary artery disease and ventricular arrhythmias during loading infusions of 0.22-0.45 mg/kg/min, prolonged (19-48 hrs) intravenous infusions 2.5-5.2 mg/min, and in 4 of the patients, during subsequent oral doses 1.5-3 g every 8 hrs. Serum, concentrations of NAPA were determined by high-performance liquid chromatography. The individual concentration-time profiles could, with one exception, be described by a two-compartment, open, kinetic model with apparent first-order elimination. The kinetic variables were: initial distribution volume (Vc) 0.20 +/- 0.11 l/kg (mean +/- SD); steady-state distribution volume (Vss) 1.58 +/- 0.55 l/kg; distributional clearance (Cle) 133 +/- 23 ml/(kg X hr); absorption rate constant (Ka) 0.354 +/- 0.173 hr-1; and fraction of dose reaching systemic circulation (F) 1.00 +/- 0.14. The data for one patient who had received increasing oral dosages of 1.5, 2, 2.5 and 3 g every 8 hours resulted in systematic underprediction of observed concentrations at the two highest oral dosing rates. This suggests the possibility of some degree of nonlinearity or time-dependent change in the kinetic behavior of NAPA. Only low concentrations of procainamide, less than 1 mg/L, were found at the end of the infusions.

Is volume of distribution at steady state a meaningful kinetic variable?

Pharmacokinetic volumes of distribution (Vd) are commonly calculated either by the steady-state method (Vdss) or the area method (Vdarea). Vdss is traditionally perceived as the least biased and most reliable indicator of the extent of distribution, but Vdss in fact has far greater practical and theoretical limitation than does Vdarea. After single doses or multiple discrete doses of a drug, Vdarea correctly relates plasma concentration to amount of drug in the body at all times after distribution equilibrium is attained. Vdss, on the other hand, is a correct proportionality constant only during continuous intravenous infusion or at a single instant in time after discrete dosing. Furthermore, calculated values of Vdss are strongly dependent on the precise configuration of the initial distributional phase of the plasma concentration curve, which may be difficult or impossible to delineate because of variance arising from methodologic artefacts or unexplained causes. Such variance can lead to large nonphysiologic within- and between-individual variability in Vdss. Vdarea, on the other hand, is relatively independent or artefactual changes in the initial distribution profile. Finally, experimental observations indicate that elimination depends physiologically on distribution in the absence of changes in clearance, not the reverse. The relation of distribution and elimination holds whether the steady-state method or the area method is used to calculate Vd. Thus, Vdarea is a more reliable and generally valid descriptor of the extent of drug distribution than is Vdss.

Modelling of initial distribution of drugs following intravenous bolus injection

Based on a recirculatory pharmacokinetic model, a physiologically realistic definition of the initial distribution volume has been developed to characterize the overall distribution process occurring shortly after rapid bolus injection of a drug. This apparent volume of distribution, which refers to the peak right atrial blood concentration, depends on the cardiac output and basic pharmacokinetic parameters usually derived from the whole blood concentration vs time curve. The initial distribution process appears to be affected by changes in the variance of the distribution of residence times of the drug in the body. The influence of the site and time of early blood sampling on the estimated initial distribution volume is discussed. This relatively simple a priori model should prove useful in predicting to a first approximation the principal characteristics of the initial distribution process.

Arterial and venous blood sampling in pharmacokinetic studies: griseofulvin

The pharmacokinetics of griseofulvin were evaluated simultaneously using both arterial and venous plasma in three dogs and one rabbit after a rapid bolus intravenous dosing. Initial arterial-venous ratios 20 sec after injection were the highest and ranged from 15- to 752-fold for dogs; the ratio was 3240-fold for the rabbit. Both curves decayed paralleling each other at the terminal phase with the venous levels higher than arterial levels by 14-43 and 8.4% for the dogs and the rabbit, respectively. The use of the instantaneous input principle was found to overestimate the total area under the plasma level-time curve by as much as 166%. An exponential term with a negative coefficient was used to account for the short and steep rising phase of plasma levels after injection. Detailed analyses showed significant differences in various calculated pharmacokinetic parameters based on arterial or venous data. The present study exemplifies the need for careful assessment and interpretation of classical pharmacokinetic parameters. It appeared that short intravenous infusion rather than the instantaneous or rapid bolus intravenous injection should be preferred for routine pharmacokinetic studies.

Pharmacokinetics and tissue distribution of chlorpheniramine in rabbits after intravenous administration

Intravenous studies of chlorpheniramine (CPM) were conducted in six New Zealand White male rabbits (mean wt. 3.88 kg). CPM and its two demethylated metabolites in arterial serum and urine were assayed by HPLC. Triexponential equations were needed to fit the i.v. CMP serum data in three rabbits, while biexponential equations were required in the other three rabbits. Harmonic mean of V1, Mss, Varea, CL, and terminal t 1/2 were 2.84, 10.8, and 15.5 liters/kg, and 4.14 liters/kg/hr and 2.57 hr, respectively. The average serum protein binding was 44%. The average blood to plasma concentration ratio was 1.85. Estimated mean hepatic blood extraction ratio based on i.v. studies was 0.88. Tissue distribution studies showed rapid and extensive uptake of CPM by various organs such as lung, kidneys, and brain after i.v. bolus injection, and their concentrations were 160-, 80-, and 31-fold higher than the plasma level. The amount of CPM in the muscle was calculated to represent about 50% of CPM present in the body near the steady state. Variation in plasma protein and tissue binding was postulated to be an important factor for the observed marked interspecies difference in the apparent volume of distribution of CPM. Only 2% of the dose was excreted unchanged in the urine.

Instantaneous input hypothesis in pharmacokinetic studies

Observed venous plasma concentrations of furosemide, propranolol, griseofulvin, and theophylline at 0.33 and 0.66 min after intravenous bolus injections to unanesthetized dogs were compared with those extrapolated using the instantaneous input hypothesis. At 0.33 min, extrapolated/observed plasma level ratios as high as 20.5, 65.5, 226, and 1.17 were found for these four drugs, respectively. Venous plasma levels peaked at 1 min postinjection in all studies. Total plasma areas (AUC0-->infinity) estimated using the instantaneous input principle were higher by as much as 6.0, 6.8, and 19.6% for propranolol, griseofulvin, and furosemide, respectively, when compared with experimental data. The effect on theophylline was negligible. These results suggest the need for cautious interpretation of some venous pharmacokinetic data. More studies in animals and humans are required to assess the magnitude of deviation from the instantaneous input hypothesis for drugs in general.