| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anesthesiology, University of Utah School of Medicine; and
Department of Pharmaceutics and Anesthesiology, University of Utah Schools of Pharmacy and Medicine, Salt Lake City, Utah
Address correspondence and reprint requests to Talmage D. Egan, MD, Department of Anesthesiology, University of Utah Health Sciences Center, Room 3C444, 30 N. 1900 E., Salt Lake City, UT 84132-2304. Address e-mail to talmage.egan{at}hsc.utah.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
IMPLICATIONS: A modified cyclodextrin-based formulation of propofol has been developed that may mitigate some of the problems associated with propofol in lipid emulsion. However, reformulation of propofol may change its clinical characteristics. This study in a pig model showed that the novel propofol formulation was substantially similar to the lipid emulsion propofol formulation.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The currently marketed propofol formulation has a number of undesirable properties that are, in part, a function of the lipid emulsion formulation. This formulation often produces pain on injection (1) and has also been associated with serious allergic reactions (2,3). In addition, because the lipid formulation supports rapid microbial growth (4,5), inadvertent contamination of the formulation can be a cause of potentially lethal postoperative sepsis (6). Finally, because propofol is often infused over many days as a sedative to facilitate the mechanical ventilation of intensive care unit patients, the lipid-based formulation has been associated with hyperlipidemia in the intensive care unit setting (7). There is, therefore, substantial interest in the development of new formulations of propofol that are devoid of some or all of the undesirable features of the current formulation.
A modified cyclodextrin (sulfobutyl ether-ß-cyclodextrin; Captisol®)-based formulation of propofol has been developed that may have some advantages over the current formulation by reducing the incidence of these formulation-dependent adverse effects. However, reformulation of propofol may alter its pharmacokinetic and pharmacodynamic characteristics (8), because at least some of propofol’s clinical pharmacologic behavior is thought to be dependent on the formulation. For example, an earlier experimental hydroxypropyl-ß-cyclodextrin-based propofol formulation was associated with more adverse hemodynamic effects (bradycardia and hypotension) compared with the lipid formulation (Diprivan) in an animal model (9). Another animal study comparing the lipid-based formulation (Diprivan) with a lipid-free formulation (propofol in glycerol, ethanol, dextrose, and water) suggested that the lipid formulation was critical in producing propofol’s typical rapid-onset/rapid-offset clinical profile (10).
The aim of this study was to compare the pharmacokinetics and pharmacodynamics of the two propofol formulations when administered by infusion in a porcine model. On the basis of pilot studies, we hypothesized that the cyclodextrin-based formulation (propofol in Captisol) would exhibit pharmacokinetic and pharmacodynamic characteristics similar to the those with the lipid-based formulation (Diprivan).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The animals were fasted except for ad libitum water for 12 h before anesthetic induction. Anesthesia was induced with a modification of the technique described by Ko et al. (11) for pigs (tiletamine/zolazepam, xylazine, ketamine, and atropine given as an IM injection). After the animals became recumbent, their tracheas were intubated and mechanically ventilated with isoflurane (1%) in oxygen (100%), keeping the PaCO2 between 35 and 40 mm Hg. An IV catheter was placed in an ear vein, and normal saline was infused at a rate of 70 mL/h with an IV infusion pump. Neuromuscular blockade was provided with pancuronium bromide.
A femoral artery was cannulated to measure mean arterial blood pressure (MAP) and to collect blood samples for pharmacokinetic assay and arterial blood gas analysis. A pulmonary artery catheter was placed via a jugular vein to measure cardiac output (CO) by using the continuous thermodilution method. Lead 2 of the electrocardiogram was used to measure heart rate (HR) and to assess cardiac rhythm. Oxygen saturation was monitored with a pulse oximeter placed on the ear. Temperature was measured in the pulmonary artery and was maintained between 36°C and 37.5°C.
Bipolar electroencephalograph (EEG) leads with low-impedance surface electrodes were placed over the frontal and occipital regions of the cerebral hemispheres, approximately 50 mm apart and 20 mm from the midline. A ground electrode was placed in the midline between the frontal and occipital regions. The Bispectral Index (BIS) was measured continuously by using the Aspect Medical A1000 (Natick, MA). Thirty minutes after the initial instrumentation was completed, baseline values of BIS, HR, MAP, CO, and arterial blood gases were recorded.
Because the study was intended to provide the data necessary for the building of combined pharmacokinetic/pharmacodynamic models, the dosage scheme was designed to elicit near-maximal EEG effects (as a surrogate for therapeutic effects), to achieve propofol concentrations that would be measurable for many hours after termination of the infusion, and to achieve the maximal EEG effects slowly enough that the equilibration of peak plasma and effect-site concentrations could be adequately characterized. On the basis of computer simulations with pharmacokinetic parameters obtained from the literature normalized to a 35-kg pig (12), an infusion rate of 500 µg · kg-1 · min-1 was selected. Because many of the animals who received the 500 µg · kg-1 · min-1 dose (8 animals for each formulation) did not exhibit a near-maximal EEG effect (see Results), a second group of animals received 750 µg · kg-1 · min-1 (8 animals for each formulation). Thus, dose assignment was not randomized.
The propofol (in the two different formulations) was administered IV as a 10-min infusion after the stabilization period through the peripheral IV catheter. Both formulations contained propofol at 10 mg/mL.
BIS, HR, MAP, and CO were the drug-effect measures of interest. BIS, HR, MAP, and CO were measured continuously and recorded at the time of each blood sample. The wave form of the electrocardiogram was also observed throughout the study.
Arterial blood samples (5 mL each) were taken from the abdominal aorta at 2, 4, 6, 8, 10, 11, 12, 13, 14, 15, 17.5, 20, 25, 30, 45, 60, 90, 120, and 180 min after the start of the infusion. The samples were collected in heparinized tubes and stored on ice. Within 30 min of collection, the samples were centrifuged, and the red blood cells and plasma were separated. The plasma was stored at -70°C until assay.
Propofol concentrations were determined by using a gas chromatography/mass spectrometer technique with selected ion monitoring, as described by Ibrahim et al. (13). The detection limit was 50 ng/mL. Coefficients of variation were 8%, 5%, and 4% for interday and 6%, 5%, and 4% for intraday quality control samples at 0.02, 0.5, and 4 µg/mL.
Applying the theory of statistical moments to pharmacokinetics (14), a model-independent moment analysis was performed to calculate the clearance (CL), mean residence time (MRT), and apparent volume of distribution at steady state (VDss) for both formulations in each animal. The area under the concentration-time curve was calculated for each plasma concentration-time plot by using the trapezoidal rule. The terminal slope for each data set was estimated by log-linear regression after the terminal portion of each curve was visually identified.
CL, MRT, and VDss were calculated by using standard equations (14). Individual moment analysis parameters for each formulation were compared by using an unpaired, two-tailed Student’s t-test, assuming equal variances. A P value of <0.05 was considered significant. The model-independent analysis was considered the primary technique for comparing the pharmacokinetics of the two formulations.
To obtain estimates of the pharmacodynamic parameters for comparison between groups, it was first necessary to estimate the compartmental pharmacokinetic parameters for each individual animal so that apparent effect-site concentrations could be predicted as part of the pharmacodynamic analysis. By using nonlinear regression techniques implemented on WinNonlin (Pharsight, Mountain View, CA), a three-compartment mamillary model parameterized in terms of central distribution volume, exponents, and microrate constants was fit to the raw concentration versus time data for each animal. The goodness of fit was assessed by visual inspection of each fit, by plotting the predicted versus observed concentrations, and by examining the residuals and SE values of the parameters.
To perform pharmacokinetic simulations from which clinical inference could be drawn, population pharmacokinetic models were also constructed for the two propofol formulations. By using nonlinear regression techniques implemented on WinNonMix (Pharsight), a three-compartment mamillary mixed-effects model parameterized in terms of central distribution volume, exponents, and microrate constants was fit to the raw concentration versus time data for the two propofol formulation groups. Population model performance was assessed by visual inspection of each population model when applied to individual animals, by examination of the residuals and SE values of the parameters, and by computation of the median absolute prediction error and median prediction errors (15).
Computer simulations using the population pharmacokinetic parameters obtained from the mixed-effects compartmental analysis were performed to provide an illustration of the predicted time course of propofol plasma concentration after the administration of clinically relevant doses of the two propofol formulations. The dosage scheme, consisting of both a bolus injection and a continuous infusion (with some washout time after the bolus injection before commencing the infusion, to illustrate the time course of bolus-only dosing), was designed to achieve concentrations near 5 µg/mL (i.e., the approximate BIS 50% effective concentration for the two formulations; see Results). Assuming administration to a 40-kg animal, the dosage scheme simulated was 3 mg/kg (i.e., 120 mg) for the bolus dose (at Time 0) and 150 µg · kg-1 · min-1 (6 mg/min) for the continuous infusion (beginning at 15 min from Time 0 and lasting for 60 min). The simulations were based on Euler’s solution to the three-compartment model with a step size of 1 s.
A two-way analysis of variance was undertaken to test for formulation, dose, and formulation x dose interactions of CO, BIS, HR, and MAP. To compress the 26 data points for each animal, 2 epochs were chosen: before drug infusion (7 times (min): -75, -65, -55, -45, -30, -15, and 0) and the first 15 min after drug infusion commenced (10 times (min): 2, 4, 6, 8, 10, 11, 12, 13, 14, 15). An area under the effect curve for each response variable was determined for each epoch and normalized for the epoch duration; the area under the effect curve was used to reduce the high dimensionality of this repeated-measures experiment and represents an integrated average value during the epoch. The before-drug-infusion epoch was tested to compare baseline similarity for the treatment groups. The 15-min-after-infusion epoch represented the period of drug onset, maximum effect, and initial recovery; this infusion epoch allowed the detection of formulation and dose effects. Additionally, the smallest value for each animal during the 15 min after commencement of drug infusion was selected to compare the maximum effect for each variable of interest (CO, BIS, HR, and MAP). No testing for change in value with onset of infusion was performed because the immediate interest was a comparison of formulation effect.
Analyses were accomplished with S-PLUS Version 5.1 (Insightful, Seattle, WA) running under Linux OS 6.1 (Red Hat, Raleigh, NC). A P value of <0.05 was considered significant. Results are presented as mean ± SE. This model-independent analysis was considered the primary technique for comparing the pharmacodynamics of the two formulations.
The concentration-effect relationship for the EEG effect was characterized by using an effect compartment model (16). The pharmacokinetic parameters for each individual animal were fixed according to their best fit from the individual compartmental pharmacokinetic analysis. The pharmacodynamic parameters of an inhibitory, sigmoidal, maximal-effect model and ke0 (a first-order rate constant characterizing effect-site equilibration kinetics) were then simultaneously estimated for each animal by using nonlinear regression techniques implemented with WinNonlin.
Only animals that achieved a profound pharmacodynamic effect (i.e., a BIS value of <40) were included in the EEG pharmacodynamic analysis (this was guided by the raw data; animals that reached a BIS value <40 exhibited enough data to characterize the entire concentration-effect relationship adequately). The goodness of fit was assessed by visual inspection of each fit and by examination of the residuals and SE values of the parameters. The pharmacodynamic parameters for each formulation were then compared by using an unpaired, two-tailed Student’s t-test, assuming equal variances. A P value of <0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The shape of the concentration-time curves for both formulations was typical of those observed for brief IV infusions of anesthetic induction drugs, exhibiting a steep distribution phase after infusion termination followed by a flatter elimination phase. Figure 1 shows the mean concentration-time curves for the 500 µg · kg-1 · min-1 and the 750 µg · kg-1 · min-1 dosage groups for both formulations. On visual inspection, the raw pharmacokinetic data were substantially similar for the two formulations, although the cyclodextrin formulation appeared to exhibit slightly lower peak propofol concentrations for the 500 µg · kg-1 · min-1 dosage group.
|
|
|
|
|
|
|
|
|
|
|
(the steepness of the concentration-effect relationship) were statistically different. The smaller Ezero value for the lipid-based formulation probably represents, at least in part, an artifact of the modeling process, because two more animals from the lipid group reached less than the minimum BIS value of 40 (near-maximal effect) to be included in the analysis. Both groups of animals reached profound levels of drug effect as measured by the BIS. The difference in the variable, for practical purposes, is probably not significant because for both formulations, the concentration-effect relationship is extremely steep (i.e., nearly a step function). |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The raw data are perhaps the most compelling evidence of the two formulations’ similarity. Simple inspection of the raw pharmacokinetic and pharmacodynamic data presented in Figures 1 and 4 strongly suggests that the two formulations are substantially similar.
The model-independent pharmacokinetic and pharmacodynamic analysis, summarized in Tables 1 and 3–5, provides the strongest statistical evidence of the substantial similarity of the two formulations. We viewed these model-independent analysis approaches as the primary technique for comparing the pharmacokinetics and pharmacodynamics of the two formulations because these approaches are not encumbered with the numerous assumptions of a compartmental kinetic model linked to a theoretical effect-compartment pharmacodynamic model.
However, the parameterized, combined kinetic/dynamic models were helpful in identifying some of the possible subtle differences between the formulations through the use of computer simulation, although it is unclear whether these subtle differences are real and clinically relevant. The simulations suggest that the cyclodextrin formulation may be associated with slightly smaller peak propofol concentrations after bolus dosing. The pharmacodynamic models also suggest that the lipid-based formulation may exhibit a slightly steeper concentration-effect relationship and a slightly greater maximal effect, although these observations are, at least in part, an artifact of the model-building process (as noted in Results).
Several limitations of the study deserve emphasis. The study may be underpowered to detect more subtle differences in the clinical pharmacology of the two formulations. Also, the presence of other anesthetics, most notably isoflurane, certainly had some influence on the pharmacodynamic measurements, although this influence was presumably the same for both formulations. In addition, in examining the concentration-time curves, it is apparent that perhaps an additional sample several hours later would have been optimal in characterizing the terminal portion of the curve. Finally, it is important to recognize that these results from a porcine model are obviously not fully translatable to human pharmacology.
There is controversy in the literature as to whether modifications in the propofol formulation ought to be expected to produce changes in propofol’s pharmacokinetic or pharmacodynamic behavior. Comparing a lipid-free formulation (propofol in glycerol, ethanol, dextrose, and water) with the currently marketed formulation (Diprivan), Dutta and Ebling (8,10,17) provided data from a rat model showing that propofol’s evanescent clinical effects may be due in part to the characteristics of the lipid-based formulation. Trapani et al. (18) made similar observations in a rat model, comparing a cyclodextrin-based propofol formulation with the current intralipid formulation (Diprivan) and concluding that the cyclodextrin-based formulation resulted in shorter sleep induction times and longer total sleep times than the intralipid formulation. This is consistent with the study in rats by Bielen et al. (9), who reported an increased incidence of severe bradycardia and hypotension associated with a cyclodextrin-based propofol formulation. However, other investigators have drawn opposite conclusions. For example, using a rabbit model comparing a cyclodextrin-based propofol formulation with the currently marketed lipid-based formulation, Viernstein et al. (19) concluded that there were no differences in the onset, duration, or maximal effects of the two formulations.
In summary, the clinical pharmacology of the two propofol formulations appears to be very similar. On the basis of the results of this study, further investigation of the novel, cyclodextrin-based propofol formulation is warranted to examine whether the new formulation may have clinical advantages over the currently marketed formulation in terms of undesirable adverse effects, such as pain on injection, the support of microbial growth, allergic reactions, and lipid load, among others. Because the adverse hemodynamic effects associated with cyclodextrin-based propofol formulations have been observed with large bolus doses, a study comparing the cardiovascular toxicity of bolus doses of the lipid- and cyclodextrin-based formulations is a logical next step in preclinical development.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Ravenelle, P. Vachon, A. E. Rigby-Jones, J. R. Sneyd, D. Le Garrec, S. Gori, D. Lessard, and D. C. Smith Anaesthetic effects of propofol polymeric micelle: a novel water soluble propofol formulation Br. J. Anaesth., August 1, 2008; 101(2): 186 - 193. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Bovill Anesthetic Pharmacology: Reflections of a Section Editor Anesth. Analg., November 1, 2007; 105(5): 1186 - 1190. [Full Text] [PDF]
|
|
|
|
|
|
M. Naguib, A. F. Kopman, and J. E. Ensor Neuromuscular monitoring and postoperative residual curarisation: a meta-analysis Br. J. Anaesth., March 1, 2007; 98(3): 302 - 316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Naguib Sugammadex: Another Milestone in Clinical Neuromuscular Pharmacology Anesth. Analg., March 1, 2007; 104(3): 575 - 581. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Hunter and E. A. Flockton The doughnut and the hole: a new pharmacological concept for anaesthetists. Br. J. Anaesth., August 1, 2006; 97(2): 123 - 126. [Full Text] [PDF]
|
|
|
|
|
|
J. Fechner, H. Ihmsen, C. Schiessl, C. Jeleazcov, J. J. Vornov, H. Schwilden, and J. Schuttler Sedation with GPI 15715, a Water-Soluble Prodrug of Propofol, Using Target-Controlled Infusion in Volunteers Anesth. Analg., March 1, 2005; 100(3): 701 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Sneyd Recent advances in intravenous anaesthesia Br. J. Anaesth., November 1, 2004; 93(5): 725 - 736. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
