Secondary Logo

Journal Logo

Original Article

The systemic absorption and disposition of levobupivacaine 0.5% after epidural administration in surgical patients: a stable-isotope study

Simon, M. J. G.*; Veering, B. T.*; Stienstra, R.*; van Kleef, J. W.*; Williams, S. G. P.; McGuire, G. M.; Burm, A. G. L.*

Author Information
European Journal of Anaesthesiology: June 2004 - Volume 21 - Issue 6 - p 460-470
  • Free

Abstract

Levobupivacaine (S(−)-1-butyl-2-piperidylformo-2′,6′-xylidide hydrochloride) is a recently introduced local anaesthetic. In contrast to bupivacaine, which is available as racemate, containing equal amounts of the R(+)- and S(−)-enantiomers, levobupivacaine only contains the pure S(−)-enantiomer. Studies have shown that the R(+)- and the S(−)-enantiomer of bupivacaine have different pharmacokinetic, pharmacodynamic and toxicological characteristics [1-9].

It is important to know the pharmacokinetics of local anaesthetics with regard to their clinical profile, particularly the duration of their action, and to the risk of systemic side-effects and toxicity [10,11]. In this respect, both systemic absorption, i.e. the uptake from the perineural site of administration into the blood, and systemic disposition (distribution and elimination) must be considered. Unfortunately, absorption rates of local anaesthetics generally cannot be derived directly from the concentration-time profiles of a perineurally administered local anaesthetic, because slow absorption limits the rate of elimination of the drug from the body, which complicates the discrimination between absorption and disposition kinetics [12]. However, absorption and disposition kinetics of local anaesthetics can be determined in a single experiment using a stable-isotope method [12]. With this approach a stable-isotope-labelled analogue of the drug to be investigated is administered intravenously (i.v.) shortly after the unlabelled drug has been administered via the perineural route. A prerequisite for the use of this method is that the unlabelled drug and the stable-isotope-labelled analogue have similar distribution and elimination characteristics, i.e. it presumes that labelling of the drug does not influence its pharmacokinetic profile [12,13].

At present, data on the pharmacokinetics of levobupivacaine are scarce [14,15] and absorption kinetics has not been studied in detail. Therefore, we determined the absorption and disposition kinetics of 0.5% levobupivacaine after epidural administration in surgical patients, using a stable-isotope method. To validate the use of deuterium-labelled levobupivacaine (D3-levobupivacaine) in a stable-isotope method, we first compared the disposition kinetics of levobupivacaine and D3-levobupivacaine after rapid simultaneous i.v. administration in healthy male volunteers.

Methods

Volunteers and patients

After approval of the study protocols by the Committee on Medical Ethics of the Leiden University Medical Centre and after obtaining written informed consent, eight healthy male volunteers, aged 18-32 yr and 15 patients, aged 23-85 yr, ASA Grade I-II, were included in the respective studies.

The health of the volunteers was substantiated by medical history and physical examination, haematology, clinical chemistry and 12-lead electrocardiography (ECG). Volunteers with a history of clinically relevant allergy, known hypersensitivity to amide local anaesthetics, adverse events to any drug, or with a history of drug, alcohol or nicotine abuse were excluded. Volunteers who donated blood or lost more than 400 mL or who had been given an investigational drug or vaccine during the 12 weeks preceding the experiment or who had taken any medication during a period of 5 days before the experiment were also excluded.

Patients underwent minor orthopaedic (n = 3), urological (n = 8), or lower abdominal (n = 4) surgery. These procedures were chosen because the duration of surgery is relatively short (this study: <135 min) and associated with a blood loss of less than 250 mL. Patients with a history of known hypersensitivity to amide local anaesthetics, severe respiratory, renal, hepatic or cardiac disease, in particular A-V or intraventricular conduction abnormalities, diabetes mellitus, severe arteriosclerosis, or neurological, psychiatric or seizure disorders were excluded. Patients, whose height was under 150 cm or weighed over 110 kg were also excluded; pregnant women were also excluded.

Procedures

Studies with volunteers. Studies with volunteers were performed in an operating room. The volunteers, positioned supine on an operating table, were attached to a device for ECG-recording and noninvasive blood pressure (BP) measurements (Cardiocap II®; Datex-Ohmeda B.V., Hoevelaken, The Netherlands). ECG rhythm strips were produced pre-infusion, at 5 min intervals during the first 30 min following the start of the infusion, at 45 and 60 min, and thereafter hourly until 3 h post-infusion. Supine diastolic and systolic BP, as well as supine heart rate (HR), were measured at screening, pre-infusion, at 5 min intervals during the infusion, at 5-, 10- and 15 min post-infusion, and thereafter every 15 min until 3 h post-infusion.

Flexible i.v. cannulae (Biovalve® 18G; Laboratories Vygon S.A., Ecouen, France) were inserted bilaterally in suitable veins in the forearm or on the hand, and were used for i.v. infusion of the study drug and for blood sampling, respectively. For each volunteer, a solution was prepared by the pharmacy of our hospital by adding levobupivacaine 10 mL (2.48 mg mL−1) and D3-levobupivacaine 10 mL (2.41 mg mL−1) to sodium chloride 0.9% 30 mL (exact concentrations were derived from high performance liquid chromatography (HPLC) analysis certificates). D3-levobupivacaine differs from levobupivacaine in that one of the methyl groups on the xylidine ring is triple labelled with deuterium (−C2H3). After a short stabilization period (about 15 min), approximately 50 mL of this solution was administered i.v., using a manually controlled pump (Becton Dickinson, Brézins, France).

Study in patients. Patients were premedicated with temazepam 20 mg (<60 yr) or 10 mg (≥60 yr) 45 min before induction of epidural anaesthesia. Dextrose/saline 500 mL was rapidly infused before the epidural injection and the infusion rate was then maintained at 2 mL kg−1 h−1. A 20-G catheter was inserted in the radial artery of the contralateral arm after local infiltration of the skin with lidocaine 0.5%. The epidural puncture was performed at the L3-L4 interspace with the patient in the sitting position using a midline or paramedian approach. After local infiltration of the skin with lidocaine 0.5%, the epidural space was identified by the loss of resistance to saline technique. With the bevel of a 16-G Hustead needle pointing cephalad, a test dose of 3 mL levobupivacaine 0.5% was injected at a rate of 1 mL s−1. Three minutes later, if there was no sign of subarachnoid injection, incremental doses of 5, 5 and 6 mL levobupivacaine 0.5% were administered at a rate of 1 mL s−1 with 1 min interval between doses. The patient was then placed in the horizontal supine position. When satisfactory anaesthetic conditions had been achieved (usually 15-20 min after the epidural injection) an 18-G flexible cannula was introduced into a foot vein. Twenty-five minutes after completion of the epidural administration approximately 50 mL of a solution prepared by the pharmacy of our hospital containing D3-levobupivacaine 0.48 mg mL−1 was administered at a constant rate of 5 mL min−1 into the foot vein, using a manually controlled pump (Becton Dickinson). Precise concentrations were derived from HPLC analysis certificates. If anaesthetic conditions were unsatisfactory after 20 min, D3-levobupivacaine was not administered. Surgery commenced soon after completion of the i.v. infusion of D3-levobupivacaine.

Blood samples and assays

In volunteers, venous blood samples were collected before dosing and at the following target times after start of the infusion: 2, 5, 10, 15, 20, 30, 45 and 60 min, and 1.25, 1.5, 1.75, 2, 2.5, 3, 3.5, 4, 5, 6, 7 and 8 h. In patients, arterial blood samples were collected before dosing and 0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50 and 60 min, and 1.25, 1.5, 2, 3, 4, 6, 8, 10, 12, 16, 20 and 24 h after the end of the epidural administration. All blood samples were temporarily stored on ice. Within 4 h after collection the samples were centrifuged at 1500g for 10 min at 4°C. The plasma was then transferred to clear pre-labelled tubes and these were immediately stored at about −20°C.

Plasma concentrations of levobupivacaine and D3-levobupivacaine were determined by Inveresk Research (Tranent, Scotland, UK), using liquid chromatography-mass spectrometry (LC-MS) with positive ion atmospheric pressure chemical ionization. Details are described in the Appendix.

Data analysis

Total doses administered i.v. were determined by multiplying the infusion rate (5.0 mL min−1) and infusion times. All doses and concentrations are expressed as free base equivalents. Pharmacokinetic data were derived using both compartmental and noncompartmental analysis. In the volunteer study, all time points refer to the start of the infusion. In the patient study, sampling points for determining unlabelled levobupivacaine refer to the actual times after completion of the epidural test dose, which was defined as 0 min. Sampling points for determining D3-levobupivacaine refer to the start of the i.v. infusion.

Non-compartmental analysis. Non-compartmental analysis was performed in a spreadsheet program (Quattro Pro® version 8.0; Corel Corporation, Ottawa, Ont., Canada). The slope of the terminal log-linear part of the curve (kz) was determined from the last 3-8 data points using linear regression. Areas under the curve (AUC) and under the first moment curve (AUMC) from t = 0 to the last sampling time included (tz) were derived using the linear trapezoidal rule when concentrations were increasing and the logarithmic trapezoidal rule when concentrations were decreasing. Subsequently, extrapolated AUCs and AUMCs from tz to ∞ were calculated and added to obtain AUC0→∞ and AUMC0→∞[16]. AUC0→∞, AUMC0→∞ and kz, as derived from concentration-time curves after i.v. administration, were used to derive the disposition parameters terminal half-life (t1/2,z), mean residence time (MRT), total plasma clearance (Cl), and volume of distribution at steady state (Vss) [16-18]. In the patient study, the fraction of the dose absorbed into the general circulation was calculated by comparison of the AUC after epidural and i.v. administration, corrected for the difference in dose:

The mean absorption time (MAT) was calculated using the equation: MAT = MRTepidural − MRTi.v., as described before [16], where MRTepidural and MRTi.v., are the MRT of epidurally administered unlabelled levobupivacaine and i.v. administered D3-levobupivacaine, respectively.

Compartmental analysis. Compartmental analysis was performed using WinNonlin® version 1.1 (Scientific Consulting Inc, Apex, USA). Mono- and bi-exponential functions were fitted to the plasma concentration-time data of levobupivacaine and D3-levobupivacaine, obtained in volunteers, using weighted (1/predicted concentration squared) least-squares non-linear regression analysis. A lag-time (restricted to ≤2 min) was included in the model, because the concentration in the first sample was usually very low, due to the venous sampling. Including a lag-time resulted in better fits in most subjects. Disposition kinetics in surgical patients were derived by fitting bi- and tri-exponential functions to the plasma concentration-time data of D3-levobupivacaine, using weighted (1/predicted concentration squared) least-squares non-linear regression analysis. Absorption rates and the cumulative fractions absorbed were then estimated using a deconvolution method with unequal sampling times, as described by Iga and colleagues [19]. The absorption rate between two time points was constrained to be non-negative. Subsequently, assuming a first-order absorption, the fractions absorbed (F1, F2) and the absorption half-lives (t1/2,a1, t1/2,a2) were derived by fitting a bi-exponential function to the obtained cumulative fraction absorbed-time data, using unweighted least-squares non-linear regression analysis. This assumes that the absorption occurs by two parallel processes [12,20]. The values of the parameters, characterizing the disposition and absorption processes were used to generate (simulate) plasma concentration-time curves after epidural administration of levobupivacaine for all individual patients. These curves were obtained by substituting parameter values into the equation describing a model with two parallel first-order absorption compartments and three disposition compartments [12,20]. The generated values were compared with the measured concentrations of levobupivacaine. The absorption kinetics were also determined by fitting the same aggregated model to the measured plasma concentration-time data of unlabelled levobupivacaine, using weighted (1/predicted concentration squared) least-squares nonlinear regression. In this approach, the disposition parameters were entered as constants. To evaluate whether this last step in the compartmental analysis improved the description of the measured plasma concentrations, the performance error (PE) for each plasma concentration-time pair and the median performance error (MDPE) and median absolute performance error (MDAPE) for each individual were calculated.

Statistical analysis

Data were analysed using the software package SPSS® v8.2.1 (SPSS Inc, Chicago, IL, USA). Parametric general linear models and, when appropriate, nonparametric tests were performed. In all tests, P < 0.05 was considered the minimum level of statistical significance.

In the volunteer study, the most appropriate compartmental model (mono- or bi-exponential) to describe the plasma concentration-time data in each individual was selected by inspection of the scatter of the data points around the fitted curves and comparison of the residual weighted sum of squares, using the F-test. Values for the AUC (AUC0→∞), determined by compartmental and non-compartmental analysis, and peak concentration (Cmax) were normalized to a dose of 25.0 mg of levobupivacaine and D3-levobupivacaine before statistical analysis. These values were then log-transformed and subjected to analysis of variance (ANOVA). Point estimates and the 90% confidence intervals (CIs) for the difference of the levobupivacaine to the D3-labelled analogue were constructed using the error variance obtained from the ANOVA. The point and interval estimates were back transformed to give estimates of the ratio of levobupivacaine relative to the D3-analogue. The two preparations were considered equivalent when the 90% CI of the ratio of AUC0→∞, which is the measure of equivalence, lay within the acceptance range of 0.90-1.125 [21] (the acceptance interval is asymmetrical because of the log-transformation of AUC0→∞). This procedure is equal to the rejection of two one-sided hypotheses concerning bioinequivalence, the primary concern of which is to limit the risk of erroneously accepting equivalence [22].

The pharmacokinetic parameters distribution half-life (t1/2,λ1), elimination half-life (t1/2,el), volume of central compartment (Vc), Vss, MRT and Cl of levobupivacaine and D3-levobupivacaine, as determined by compartmental analysis and, where applicable, by non-compartmental analysis, were subject to ANOVA-techniques without log-transformation.

In the patient study, the most appropriate compartmental model (bi- or tri-exponential) to describe the plasma concentration-time data of D3-levobupivacaine in each individual was selected by inspection of the scatter of the data points around the fitted curves and comparison of the residual weighted sum of squares, using the F-test. The MDPE and MDAPE were subjected to ANOVA. Pharmacokinetic data in patients who did and those who did not receive general anaesthesia were compared using two-sample t-tests.

Results

Table 1 shows the patient characteristics and the dosing of the volunteers and patients, included in the respective studies. The volunteers showed no clinically significant changes in vital signs and ECG and no adverse events occurred during the study. In the patient study, anaesthetic conditions seemed to be satisfactory after 20 min, but five patients ultimately received general anaesthesia later on because they experienced pain during surgery. However, pharmacokinetic data did not differ between patients who did and who did not receive general anaesthesia. Furthermore, mean values and standard deviations of the pharmacokinetic parameters were similar, whether these patients were included or not. Therefore, pharmacokinetic data reported herein are based on all 15 recruited patients.

Table 1
Table 1:
Patient characteristics and total dose administered to volunteers and surgical patients in the corresponding studies.

Disposition kinetics in volunteers

Normalized (to a 25 mg dose) plasma concentrations of levobupivacaine and D3-levobupivacaine (Fig. 1) virtually coincided in all subjects. With few exceptions the concentration ratios were between 0.9 and 1.1 and showed no significant changes over time (Fig. 1). Concentration-time data were best described by a bi-exponential function in six volunteers and a mono-exponential function in one volunteer. One subject was excluded from the compartmental analysis, because the fits of mono-, bi- and tri-exponential functions were considered inadequate. The ratio estimate of the geometric means of AUC0→∞ of levobupivacaine and D3-levobupivacaine, determined by both compartmental and non-compartmental analysis, was 1.02. The corresponding 90% CIs were 1.00-1.04 and 1.00-1.03, respectively. These intervals were well within the acceptance range of 0.90-1.125 and, therefore, the formulations were considered equivalent. Pharmacokinetic data are shown in Table 2. There were no significant differences in the parameters of levobupivacaine and D3-levobupivacaine, except for a small (2.3%) difference in the total plasma clearance, determined by compartmental analysis. Concentrations of R(+)-bupivacaine and its D3-labelled analogue (the optical antipodes of levobupivacaine and D3-levobupivacaine) were below the limit of quantification (<10 ng mL−1) in all collected samples.

Figure 1
Figure 1:
Plasma concentrations of levobupivacaine (upper panel) and D3-levobupivacaine (middle panel) in individual volunteers (n = 8). Although sampling was continued for 8 h in all volunteers, plasma concentrations were below the detection limit (10 ng mL−1) after 7 h in one patient and at 8 h in four patients. In the lower panel, the plasma concentration ratios of levobupivacaine to D3-levobupivacaine in individual subjects are shown. All curves are normalized to a dose of 25.0 mg of both compounds and ratios were calculated after normalizing concentrations to a dose of 25.0 mg of both compounds. Limits of the acceptance range (0.90, 1.125) are shown.
Table 2
Table 2:
Pharmacokinetic data of levobupivacaine and D3-levobupivacaine in healthy volunteers.

Disposition kinetics in surgical patients

Plasma D3-levobupivacaine concentrations were measurable in individual patients over periods, ranging from 336 to 1409 min (Fig. 2). Tri-exponential functions fitted the concentration-time curves better than bi-exponential functions for all patients. The values of the disposition parameters of D3-levobupivacaine are shown in Table 3. The values, derived by compartmental analysis, were similar to the results obtained by non-compartmental analysis.

Figure 2
Figure 2:
Plasma concentrations of D3-levobupivacaine after rapid i.v. infusion (upper panel) and of levobupivacaine after epidural administration (lower panel) in individual surgical patients (n = 15). Although blood samples were collected for 24 h in all patients, plasma concentrations of D3-levobupivacaine dropped below the detection limit (10 ng mL−1) at earlier points in time in most patients. In one patient, this was also the case for unlabelled levobupivacaine.
Table 3
Table 3:
Pharmacokinetic parameters characterizing the disposition of D3-levobupivacaine after rapid i.v. infusion in patients under epidural anaesthesia.

Absorption kinetics in surgical patients

Plasma levobupivacaine concentrations were measurable over the entire 24 h period, with one exception (Fig. 2). The maximum concentration of levobupivacaine (1086 ± 296 ng mL−1) was reached after 10.4 ± 4.4 min. In three patients it was reached at 5 min after completion of the test dose, i.e. at the completion of the fractional administration of the epidural dose. Individual cumulative fraction absorbed-time curves are shown in Figure 3. These curves were adequately described by bi-exponential functions, reflecting two parallel absorption processes, in all patients. Results of these fits are presented in Table 4, along with the results of the fits of the aggravated model, including two parallel absorption processes. In one patient F1 and t1/2,al were highly correlated and, therefore, the values of these parameters could not be estimated with confidence and were not included in the results and statistical evaluation. Concentrations, predicted by the aggravated models adequately fitted the measured levobupivacaine concentrations (Fig. 4), but the fits, judged from the PE, were slightly better when absorption kinetics were determined by fitting the aggravated model directly to the data (MDPE = −0.9%; MDAPE = 7.6%) rather than from the fraction absorbed-time data (MDPE = 2.5%, P < 0.05; MDAPE = 7.6%, P > 0.05). Systemic availability (F) and MAT, determined by non-compartmental analysis did not differ from those estimated by compartmental analysis (Table 4).

Figure 3
Figure 3:
Cumulative fractions absorbed of levobupivacaine vs. time in individual surgical patients (n = 15). Absorption-time data were obtained by deconvolution of the measured concentrations against the i.v. unit impulse response curve.
Table 4
Table 4:
Pharmacokinetic parameters characterizing the absorption of levobupivacaine in patients under epidural anaesthesia.
Figure 4
Figure 4:
Graphs showing the patients in whom the measured concentration-time data were best (a) and worst (b) described by the aggravated pharmacokinetic models. Dotted line: concentrations predicted by the model, whereby absorption parameters were derived by fitting a bi-exponential function to the fraction absorbed-time data. Unbroken line: concentrations predicted by the model, whereby absorption data were derived by fitting the aggravated model (assuming bi-phasic absorption) directly to the data. Disposition functions were tri-exponential and identical with both modelling procedures.

Discussion

Stable isotopes are powerful research tools to study the pharmacokinetics of extravascularly administered drugs (see below) [12]. To be useful in this respect, the disposition kinetics of the stable-isotope-labelled analogue should be representative of those of the unlabelled regularly used drug. As demonstrated in the study in volunteers, D3-levobupivacaine meets this requirement, since the ratios of the AUC0→∞ of levobupivacaine and D3-levobupivacaine, determined by compartmental and non-compartmental analysis, and the corresponding 90% CIs were well within the predefined acceptance range. Even though compartmental analysis showed a significant difference in the total plasma clearance, this difference was very small (2.3%), and not confirmed by non-compartmental analysis.

In the volunteers, blood sampling was continued until 8 h after the start of the infusion, i.e. for approximately four times the terminal (elimination) half-life. In the patient study, blood sampling was continued until 24 h after the epidural injection, because the terminal half-life after epidural administration is considerably longer than the half-life after i.v. administration, due to the slow secondary absorption rate. When the absorption rate constant is smaller than the elimination rate constant observed after i.v. administration the elimination rate constant after extravascular (e.g. epidural) administration will approximate the absorption rate constant (a drug cannot be removed from the blood before it has been absorbed into it). This complicates the discrimination between the absorption and disposition kinetics after epidural administration of a local anaesthetic agent. However, with the stable-isotope method, disposition kinetics can be readily derived from the concentration-time profiles of the labelled drug and subsequently used to derive the absorption characteristics of the unlabelled drug.

Although plasma concentrations of D3-levobupivacaine dropped below the detection limit after 6-16 h in most patients, the concentrations were measurable over a time period of at least two to three times the elimination half-life, which is generally considered sufficient to characterize the pharmacokinetics accurately. Nevertheless, the fact that concentrations of D3-levobupivacaine could not be determined over the entire 24 h period may have influenced the estimation of the systemic availability F slightly (see below).

For ethical reasons venous rather than arterial blood samples were collected in volunteers. When compared to the data that would have been obtained with arterial sampling, the volume of the central compartment and the distribution half-life are probably overestimated. However, the sampling site probably has minimal effect on the elimination half-life, total plasma clearance, MRT and steady-state volume of distribution. In any case, the main objective of this study was to compare the pharmacokinetics of levobupivacaine and D3-levobupivacaine and in this respect the sampling site is of minor importance.

The pharmacokinetics of levobupivacaine in volunteers, as observed in the present study differ somewhat from the pharmacokinetics of S(−)-bupivacaine after i.v. administration of racemic bupivacaine [3]. In that study, the total plasma clearance of S(−)-bupivacaine (317 ± 67 mL min−1) was lower and the elimination half-life (157 ± 77 min) and MRT (172 ± 55 min) were longer than found in the present study. On the other hand, a review on levobupivacaine reports a higher total plasma clearance (651 mL min−1) and a shorter MRT (85 min) [23]. The discrepancies between these studies and the present study emphasize the large inter-individual variability and the relatively small sample sizes of studies in volunteers. Eight volunteers participated in this study, but it must be emphasized that the main objective of this study was to compare the pharmacokinetics of D3-levobupivacaine and unlabelled levobupivacaine. From this perspective the number of patients included was more than sufficient, as is illustrated by the detection of negligible and irrelevant difference in the clearance of the two compounds. Also, it cannot be excluded that the pharmacokinetics of S(−)-bupivacaine after i.v. administration of racemic bupivacaine are influenced by the presence of the R(+)-enantiomer. The absence of both R(+)-bupivacaine and its D3-labelled analogue in this study indicates that no racemization of levobupivacaine occurs in human beings [24].

In the patient study, the i.v. infusions of D3-levobupivacaine were started 25 min after the epidural injection. The reason for this is that we wanted to start the infusion after a satisfactory epidural block had developed in order to avoid the administration of very expensive D3-levobupivacaine to patients in whom a satisfactory block would not develop and that would a priori have been considered as dropouts. Theoretically, the later administration of D3-levobupivacaine might affect the estimation of absorption parameters. However, the feasibility of the administration of the labelled analogue after full development of the epidural block has been demonstrated in previous studies [12,20] and is illustrated by the fact that systemic availability (total fractions absorbed) of lidocaine (F = 0.97 [12]) and bupivacaine (F = 0.94 [12] and F = 0.97 [20]) and levobupivacaine (F = 1.06) were close to unity.

The study in surgical patients demonstrated that the systemic absorption of levobupivacaine is bi-phasic. This is in keeping with observations from previous studies examining the absorption kinetics of other local anaesthetics, including lidocaine, ropivacaine and racemic bupivacaine [12,20,25,26]. The fraction of levobupivacaine absorbed during the fast absorption process (F1 = 0.22 ± 0.06) was somewhat less than that reported for bupivacaine (F1 = 0.29 ± 0.09) [20]. This may be attributed to a greater vasoconstrictive action of levobupivacaine [27,28], as vasoconstriction of the epidural vessels may decrease the absorption rate of the local anaesthetic from the epidural space into the systemic circulation. A recent study showed that the vasoactive effect of levobupivacaine is bi-phasic, i.e. levobupivacaine is a vasoconstrictor at low concentrations and a vasodilator at high concentrations [29]. As exact concentrations of local anaesthetics at different sites within the epidural space are not known it is not possible to determine which mechanism (vasoconstriction or vasodilatation) prevails in clinical practice, but, in any case, the evidence suggests that levobupivacaine is likely to have a greater vasoconstrictive (or a less vasodilatatory) action than bupivacaine.

The fraction of levobupivacaine absorbed during the slow absorption process (F2 = 0.84 ± 0.14) was higher than that reported for bupivacaine (F2 = 0.68 ± 0.11) [20]. This can in part be explained by the higher estimated total fraction absorbed (F) in this study, which exceeded 1 with both the compartmental and non-compartmental analysis. Theoretically, this means that the amount of levobupivacaine absorbed in the systemic circulation exceeded the amount that was administered epidurally, which is impossible. This finding might be related to the deconvolution method applied in this study. In this procedure, the absorption rates between two time points in this study were constrained to be non-negative. On the other hand, it has been demonstrated that this deconvolution method slightly underestimates the cumulative fractions absorbed [19,26]. This is also in keeping with our observation that the total fraction absorbed, estimated by fitting biexponential functions to the fraction absorbed-time data resulted in somewhat smaller estimates (F = 1.06 ± 0.14) than those obtained by directly fitting a model with two parallel first-order absorption compartments and three disposition compartments to the measured plasma concentration-time data (F = 1.15 ± 0.14) or by non-compartmental analysis (F = 1.16 ± 0.14). However, a more likely explanation for the overestimation of F is that the AUCs of D3-levobupivacaine were slightly underestimated. This was because plasma concentrations of D3-levobupivacaine could not be determined over the full 24 h study period, but dropped below the limit of quantification earlier. This probably resulted in an underestimation of the terminal half-life.

Values for the MAT differed between compartmental and non-compartmental analysis. These differences can be clarified, because MAT is calculated by subtracting MRTi.v. from MRTepidural and in either case the MRT is very sensitive to small discrepancies in estimation of the AUC and AUMC [16].

The disposition parameters t1/2,el and MRT derived in the patients were longer and clearance was lower than the corresponding values in the volunteers. This might be related to differences in the populations. Alternatively, changes in regional blood flows and possibly cardiac output, that are associated with epidural anaesthesia, may also contribute to the 'slowing' of the pharmacokinetics in surgical patients compared to healthy volunteers.

Peak plasma concentrations of levobupivacaine after epidural administration have been found to be higher than those of bupivacaine, measured as mixed enantiomers [15], although others did not find a difference between levobupivacaine and racemic bupivacaine [30]. Peak plasma concentrations of S(−)-bupivacaine have also been shown to be higher than those of R(+)-bupivacaine after epidural administration of racemic bupivacaine [4]. This reflects mainly enantioselective disposition, secondary to differences in the protein binding of the S(−)-bupivacaine and R(+)-bupivacaine, rather than enantioselective absorption [4]. This can be explained, because the absorption is largely dependent on the partitioning between epidural (fat) tissue and the blood draining the epidural space and because fat tissue can be considered an achiral environment [4]. However, the absorption of levobupivacaine, given as a single agent, may very well differ from the absorption of S(−)-bupivacaine, because the vasoactive properties of levobupivacaine and bupivacaine may differ. As explained above, the available evidence suggests that levobupivacaine is more vasoconstrictive/less vasodilatatory than bupivacaine. Thereby, slower absorption of levobupivacaine into the blood may compensate the slower disposition from the blood of levobupivacaine compared to mixed bupivacaine enantiomers. This might explain why peak plasma concentrations of levobupivacaine and bupivacaine (reported as mixed enantiomers) in the study of Bader and colleagues did not differ.

Fitting the model with two parallel first-order absorption compartments and three disposition compartments directly to the measured plasma concentration-time data, instead of first estimating the absorption characteristics and then construct an aggravated model to predict the plasma concentrations, reduced (by definition) the weighted sum of squares and the bias, expressed as the MDPE for all blood samples. However, the inaccuracy, expressed as MDAPE value did not change.

In this study, five of 15 patients were given general anaesthesia because they experienced pain during the operation. The rate of successful block (67%) in our patients undergoing various types of surgery is comparable with the 62% in a study by Cox and colleagues [31], in which patients underwent lower limb surgery after administration of levobupivacaine 0.5% 15 mL. To improve the success rate in patients undergoing lower limb surgery, the use of a more concentrated solution (0.75%) may be recommended.

In conclusion, this study demonstrated that the disposition kinetics of levobupivacaine and D3-levobupivacaine are similar and that, therefore, D3-levobupivacaine can be used in a stable-isotope method to study the absorption and disposition of levobupivacaine in a single experiment. Using the stable-isotope method we determined the absorption kinetics of levobupivacaine after epidural administration in surgical patients. In keeping with previous observations with other local anaesthetics, levobupivacaine showed a bi-phasic absorption pattern, whereby the smaller fraction of the dose was rapidly absorbed into the systemic circulation whereas the remainder was absorbed at a much slower rate. A comparison with earlier publications suggested that the fraction of levobupivacaine absorbed during the rapid primary absorption phase is smaller than the corresponding fraction of racemic bupivacaine.

Acknowledgements

The authors thank Stefan Maartense, Joost Rothbarth, Koen Peeters, Litska Onderwater, Arlette Odink and Assaf Senft for their valuable clinical assistance. We thank Ms K. Clydesdale and Mrs K. Davis of Inveresk Clinical Research Ltd. for their valuable assistance in data management. The study was supported and study medications, including D3-levobupivacaine, were provided by Celltech Chiroscience Ltd, Cambridge, UK.

References

1. Mather LE. Disposition of mepivacaine and bupivacaine enantiomers in sheep. Br J Anaesth 1991; 67: 239-246.
2. Mather LE, Rutten AJ, Plummer JL. Pharmacokinetics of bupivacaine enantiomers in sheep: influence of dosage regimen and study design. J Pharmacokinet Biopharm 1994; 22: 481-498.
3. Burm AGL, Van der Meer AD, Van Kleef JW, Zeijlmans PWM, Groen K. Pharmacokinetics of the enantiomers of bupivacaine following intravenous administration of the racemate. Br J Clin Pharmacol 1994; 38: 125-129.
4. Groen K, Mantel M, Zeijlmans PWM, et al. Pharmacokinetics of the enantiomers of bupivacaine and mepivacaine after epidural administration of the racemates. Anesth Analg 1998; 86: 361-366.
5. Lee-Son S, Wang GK, Concus A, Crill E, Strichartz G. Stereoselective inhibition of neuronal sodium channels by local anesthetics. Anesthesiology 1992; 77: 324-335.
6. Kanai Y, Katsuki H, Takasaki M. Comparisons of the anesthetic potency and intracellular concentrations of S(−) and R(+) bupivacaine and ropivacaine in crayfish giant axon in vitro. Anesth Analg 2000; 90: 415-420.
7. Vanhoutte F, Vereecke J, Verbeke N, Carmeliet E. Stereoselective effects of the enantiomers of bupivacaine on the electrophysiological properties of the guinea-pig papillary muscle. Br J Pharmacol 1991; 103: 1275-1281.
8. Denson DD, Behbehani MM, Gregg RV. Enantiomer-specific effects of an intravenously administered arrhythmogenic dose of bupivacaine on neurons of the nucleus tractus solitarius and the cardiovascular system in the anesthetized rat. Reg Anesth 1992; 17: 311-316.
9. Mazoit JX, Decaux A, Bouaziz H, Edouard A. Comparative ventricular electrophysiologic effect of racemic bupivacaine, levobupivacaine, and ropivacaine on the isolated rabbit heart. Anesthesiology 2000; 93: 784-792.
10. Thomas JM, Schug SA. Recent advances in the pharmacokinetics of local anaesthetics. Long-acting amide enantiomers and continuous infusions. Clin Pharmacokinet 1999; 36: 67-83.
11. Gennery B, Mather LE, Strichartz G. Levobupivacaine: new preclinical and clinical data. Semin Anesth 2000; 19: 132-148.
12. Burm AGL, Vermeulen NPE, Van Kleef JW, De Boer AG, Spierdijk J, Breimer DD. Pharmacokinetics of lignocaine and bupivacaine in surgical patients following epidural administration. Simultaneous investigation of absorption and disposition kinetics using stable isotopes. Clin Pharmacokinet 1987; 13: 191-203.
13. Burm AGL, De Boer AG, Van Kleef JW, et al. Pharmacokinetics of lidocaine and bupivacaine and stable isotope labelled analogues: a study in healthy volunteers. Biopharm Drug Dispos 1988; 9: 85-95.
14. Bader AM, Tsen LC, Camann WR, Nephew E, Datta S. Clinical effects and maternal and fetal plasma concentrations of 0.5% epidural levobupivacaine versus bupivacaine for cesarean delivery. Anesthesiology 1999; 90: 1596-1601.
15. Kopacz DJ, Allen HW, Thompson GE. A comparison of epidural levobupivacaine 0.75% with racemic bupivacaine for lower abdominal surgery. Anesth Analg 2000; 90: 642-648.
16. Riegelman S, Collier P. The application of statistical moment theory to the evaluation of in vivo dissolution time and absorption time. J Pharmacokinet Biopharm 1980; 8: 509-534.
17. Benet LZ, Galeazzi RL. Noncompartmental determination of the steady-state volume of distribution. J Pharm Sci 1979; 68: 1071-1074.
18. Perrier D, Mayersohn M. Noncompartmental determination of the steady-state volume of distribution for any mode of administration. J Pharm Sci 1982; 71: 372-373.
19. Iga K, Ogawa Y, Yashiki T, Shimamoto T. Estimation of drug absorption rates using a deconvolution method with nonequal sampling times. J Pharmacokinet Biopharm 1986; 14: 213-225.
20. Veering BT, Burm AGL, Vletter AA, Van den Heuvel RPM, Onkenhout W, Spierdijk J. The effect of age on the systemic absorption, disposition and pharmacodynamics of bupivacaine after epidural administration. Clin Pharmacokinet 1992; 22: 75-84.
21. Pabst G, Jaeger H. Review of methods and criteria for the evaluation of bioequivalence studies. Eur J Clin Pharmacol 1990; 38: 5-10.
22. Schuirmann DJ. A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability. J Pharmacokinet Biopharm 1987; 15: 657-680.
23. McClellan KJ, Spencer CM. Levobupivacaine. Drugs 1998; 56: 355-362.
24. Gristwood RW, Greaves JL. Levobupivacaine: a new safer long acting local anaesthetic agent. Expert Opin Invest Drug 1999; 8: 861-876.
25. Tucker GT, Mather LE. Pharmacology of local anaesthetic agents. Pharmacokinetics of local anaesthetic agents. Br J Anaesth 1975; 47: 213-224.
26. Emanuelsson BM, Persson J, Alm C, Heller A, Gustafsson LL. Systemic absorption and block after epidural injection of ropivacaine in healthy volunteers. Anesthesiology 1997; 87: 1309-1317.
27. Aps C, Reynolds F. An intradermal study of the local anaesthetic and vascular effects of the isomers of bupivacaine. Br J Clin Pharmacol 1978; 6: 63-68.
28. Khodorova AB, Strichartz GR. The addition of dilute epinephrine produces equieffectiveness of bupivacaine enantiomers for cutaneous analgesia in the rat. Anesth Analg 2000; 91: 410-416.
29. Newton DJ, Burke D, Khan F, et al. Skin blood flow changes in response to intradermal injection of bupivacaine and levobupivacaine, assessed by laser Doppler imaging. Reg Anesth Pain Med 2000; 25: 626-631.
30. Bader AM, Tsen LC, Camann WR, Nephew E, Datta S. Clinical effects and maternal and fetal plasma concentrations of 0.5% epidural levobupivacaine versus bupivacaine for cesarean delivery. Anesthesiology 1999; 90: 1596-1601.
31. Cox CR, Faccenda KA, Gilhooly C, Bannister J, Scott NB, Morrison LM. Extradural S(−)-bupivacaine: comparison with racemic RS-bupivacaine. Br J Anaesth 1998; 80: 289-293.

Appendix

Assay of levobupivacaine

One millilitre aliquot of plasma was transferred to a test tube and 10 μL added to the internal standard solution containing prilocaine 200 ng. Subsequently 1 mL of a saturated sodium bicarbonate solution was added and the contents of the test tubes were mixed on a vortex mixer. Then methyl-tertiary-butyl-ether 6 mL was added and the test tube was capped and shaken on a rotating action shaker for 10 min. After centrifugation for 10 min at 3000 rpm the upper organic phase was transferred to a clean test tube and evaporated to dryness under a stream of nitrogen at 35°C. Finally the sample was reconstituted in 150 μL of the mobile phase and 40 μL were injected into the liquid chromatograph.

The analytical apparatus consisted of a Fisons Instruments VG Platform® mass spectrometer (Micromass (formerly Fisons Instruments), Manchester, UK), a Waters 510® HPLC pump (Waters Corporation, Milford, USA) and a Gilson 231® autosampler (Gilson Medical Electronics (France) S.A., Villiers-le-Bel, France), and was equipped with a 250 mm long 4.6 mm inner diameter analytical column, filled with Hichrom Chiral L-PGC, CHI-L-PGC(B)-250Å and a 10 mm long 4.6 mm inner diameter guard column, filled with Hichrom Chiral L-PGC, CHI-L-PGC(B)-10C5. The mobile phase consisted of hexane and ethanol (85 : 15 v/v) and the flow rate was 1 mL min−1. The column temperature was 40°C. Chemical ionization occurred in the positive ion mode. The corona and cone voltages were 3.5 kV and 15 V, respectively; source and probe temperatures were 150°C and 400°C, respectively. The following ions were monitored: m/z = 292 (D3-levobupivacaine), m/z = 289 (levobupivacaine and R(+)-bupivacaine) and m/z = 221 (prilocaine). Data were quantified following peak integration using peak area internal standardization with weighted (1/x) linear regression analysis for the calibration lines.

Retention times of D3-levobupivacaine, levobupivacaine, R(+)-bupivacaine and prilocaine were approximately 10, 10, 8.5 and 7.0 min, respectively, with small differences between the analyses of the two studies. In the volunteer study, the interday accuracies of the quality control samples at concentrations of 30, 200 and 400 ng mL−1 were 101.5%, 104.4% and 99.7%, respectively, for levobupivacaine and 103.5%, 107.1% and 101.3%, respectively, for D3-levobupivacaine. The interday precisions for these samples were 8.5%, 4.3% and 5.8% for levobupivacaine, and 6.3%, 4.5% and 4.5% for D3-levobupivacaine. In the patient study, the interday accuracies of the quality control samples at concentrations of 30, 200 and 400 ng mL−1 were 103.3%, 103.5% and 103.9%, respectively, for levobupivacaine and 101.1%, 100.6% and 99.8%, respectively, for D3-levobupivacaine. The interday precisions for these samples were 4.3%, 5.9% and 3.4% for levobupivacaine, and 3.6%, 3.3% and 3.3% for D3-levobupivacaine. The limits of quantification were 10 ng mL−1 for D3-levobupivacaine, levobupivacaine and R(+)-bupivacaine in both studies.

Keywords:

ANAESTHETIC TECHNIQUES, epidural; ANAESTHETICS, LOCAL, levobupivacaine; HUMAN EXPERIMENTATION; PHARMACOKINETICS, levobupivacaine

© 2004 European Academy of Anaesthesiology