Anesth Analg 2002;94:203-207
© 2002 International Anesthesia Research Society
REGIONAL ANESTHESIA
Bupivacaine-Induced QRS Prolongation is Enhanced by Lidocaine and by Phenytoin in Rabbit Hearts
Lionel Simon, MD*
,
Nobutaka Kariya, MD*
,
Emilie Pelle-Lancien, MD*, and
Jean-Xavier Mazoit, MD PhD*
*Laboratoire d’Anesthésie UPRES EA 392, Université Paris-Sud, Faculté de Médecine du Kremlin-Bicêtre; Hôpital Saint-Vincent de Paul, Paris, France; and Osaka City University, Osaka, Japan
Address correspondence to Jean Xavier Mazoit, MD, PhD, Laboratoire d’Anesthésie Faculté de Médecine du Kremlin-Bicêtre, F-94276, Le Kremlin-Bicêtre, France. Address e-mail to jean-xavier. mazoit{at}kb.u-psud.fr
 |
Abstract
|
|---|
Lidocaine, phenytoin, and bupivacaine are sodium channel blockers. Lidocaine displaces bupivacaine from its receptor on the sodium channel. However, lidocaine does not seem to decrease bupivacaine toxicity. Phenytoin also has been used to treat bupivacaine cardiotoxicity. To test the hypothesis that lidocaine or phenytoin might be used for the treatment of bupivacaine overdose, we compared the effects of bupivacaine on intraventricular conduction in the isolated heart of rabbits with bupivacaine and with either phenytoin or lidocaine added to bupivacaine. Twenty-four rabbit hearts were retrogradely perfused in a nonrecirculating Langendorff apparatus. The duration of QRS was measured without any drug and 10 min after infusion of 3 µM bupivacaine. Saline (control group) or increasing concentrations of either lidocaine or phenytoin was then added by 10-min-step increments. QRS duration was measured for each concentration at the end of a 10-min step. It was also determined 10 min after discontinuation of bupivacaine and after a period of washout for all drugs. QRS duration was significantly increased by adding phenytoin or lidocaine to bupivacaine. These drugs should not be used to treat the manifestations of bupivacaine toxicity.
IMPLICATIONS: The effects of lidocaine and phenytoin on bupivacaine-related increases in cardiac conduction time have been studied in an isolated heart preparation. Both drugs increased the QRS widening induced by bupivacaine. We conclude that none of these drugs should be used for treating bupivacaine intoxication.
|
Introduction
|
|---|
Long-acting local anesthetics like bupivacaine impair intraventricular conduction (1). Thus, bupivacaine can be responsible for life-threatening arrhythmias after either an accidental IV administration or an enhanced absorption from the site of injection (2–4). Other Na+ channel blockers, such as lidocaine and possibly phenytoin, are less cardiotoxic than bupivacaine. Clarkson and Hondeghem (5) showed that lidocaine could displace bupivacaine from its receptor site, and some authors proposed to treat the manifestations of bupivacaine cardiotoxicity with lidocaine or phenytoin, or both (2,3,6). However, because few researchers studied the interactions between lidocaine or phenytoin and bupivacaine, and because the subject remains controversial, we assessed the influence of lidocaine and phenytoin on the intraventricular conduction block induced by bupivacaine in isolated hearts of rabbits.
|
Methods
|
|---|
Twenty-four male New-Zealand rabbits, weighing 1520–2130 g, were allocated randomly to three groups. This experiment was approved by our institutional animal care committee. Care of the animals conformed to the recommendation of the Helsinki Declaration. The rabbits were anesthetized with pentobarbital 0.6 mg/kg given intraperitoneally. A tracheotomy was performed and the animals were mechanically ventilated. The chest was opened, and, after IV heparin injection, the heart was removed and quickly mounted on a nonrecirculating Langendorff apparatus. The hearts were retrogradely perfused at a constant flow of 40 mL/min with a modified Krebs-Henseleit buffer bubbled with a mixture of 95% O2 and 5% CO2 and maintained at 37°C. The buffer had the following composition (millimolar): NaCl 118, KCl 4.7, CaCl2 2.5, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25, glucose 5.5, and Na pyruvate 2.0. The pH of the perfusate was between 7.37 and 7.42. Reagents for chromatography and salts for buffer were purchased from Prolabo (Paris, France). We used the following exclusion criteria for the preparation: 1) the presence of aortic valve regurgitation, 2) a rhythm (before pacing) <120 bpm or >210 bpm, and 3) the presence of arrhythmias. The hearts were paced atrially throughout the study with a bipolar electrode at 270 bpm by using a Grass S88 stimulator (Astro-Med, Trappes, France). QRS was recorded by using surface electrodes and computerized by using Chartsoftware (ADInstruments, Les Ulis, France). QRS duration was measured from the average of three consecutive beats. After a 10-min stabilization period, drug infusion was begun. The hearts were divided into three groups. All hearts received bupivacaine 3 µM (1 µg/mL) for 40 min (from T0 to T40) ( Fig. 1, top). In addition to bupivacaine, Group One hearts (n = 8) received saline 10 mL/h from T10 to T50; Group Two hearts (n = 8) received phenytoin 17 µM (4.5 µg/mL) from T10 to T20, 35 µM (9 µg/mL) from T20 to T30, and 116 µM (30 µg/mL) from T30 to T50; and Group Three hearts (n = 8) received lidocaine 8.5 µM (2 µg/mL) from T10 to T20, 17 µM (4 µg/mL) from T20 to T30, and 43 µM (10 µg/mL) from T30 to T50. QRS duration was mea-sured at T0, T10, T20, T30, T40, T50, T60, and T70. The outflow perfusate was sampled at frequent intervals between T0 and T70, and the corresponding concentrations of bupivacaine, lidocaine, and phenytoin were assayed by using our usual gas chromatographic method (7).
View larger version (18K):
[in this window]
[in a new window]
|
Figure 1. Top, experimental design. Bupivacaine was infused from T0 to T40 in all groups. Phenytoin, lidocaine, or saline (depending on the group considered) was infused from T10 to T50. X = measure of QRS duration (min). Bottom, model used for analysis of pharmacokinetic/pharmacodynamic variables. Linear pharmacokinetics were assumed, and the heart was described by a one-compartment open model. Each drug was considered to distribute in a separate compartment, i.
|
|
Pharmacokinetics and Pharmacodynamics
We used a simplification of our usual model (1,8). Briefly, linear pharmacokinetics were assumed and the heart was described by a one-compartment open model (with the assumption of venous equilibrium) (Fig. 1, bottom). Each drug was considered to distribute in a separate compartment, i. If C0i is drug concentration in the inflow perfusate, the outflow perfusate concentration (Ci) can be expressed as a function of time (t) by the following relationship:
equation
where ki0 is the exit rate constant of drug i from the central compartment, Ai is the amount of drug i in the central compartment at time t, and Q is perfusate flow. Each drug compartment was unidirectionally linked to a common effect compartment. The rate constant from bupivacaine compartment to the effect compartment was set to 0.001 times the exit rate constant from this compartment, thus leading to a negligible loss of drug outside the kinetic compartment. Mass transfer outside the effect compartment was allowed by a unique rate constant to the bupivacaine compartment.
The increase in QRS duration (E) was comparatively fitted to a linear model and to the Emax model:
equation
where E0 is basal QRS duration, Emax is the maximal increase in QRS duration, AE is drug amount in the effect compartment, and AE50 is the drug amount in the effect compartment producing half the maximal increase in QRS duration at steady state. The steady-state bupivacaine perfusate concentration producing half the maximal increase in QRS duration (Css50) was calculated as Css50 = AE50/k10. To test the effect of phenytoin and lidocaine on QRS widening produced by bupivacaine, we compared this model with the three drugs participating to the resulting effect [QRS widening] to models without effect arising when phenytoin or lidocaine or both drugs are added.
Pharmacokinetic analysis was performed by using the procedure ADVAN 5 from NONMEM version V level 1.1 (9). A two-step approach was used. We first calculated structural kinetic variables (i.e., the ki0s) for each drug, and used these variables fixed at their values estimated at the first step in the final dynamic model. An intersubject variability (
) with mean zero and variance 
2 was associated to each of the structural variables assuming a log-normal distribution. Intraindividual variability caused by assay error or model misspecification was modeled by using a proportional error model. No covariance between the two sources of error was considered. The two different dynamic models (linear and Emax models) and the significant effect of phenytoin and/or lidocaine on QRS duration were compared by using the Akaike information criterion applied to NONMEM Objective Function (10). P < 0.05 was considered statistically significant.
|
Results
|
|---|
QRS duration at baseline (T0) was 25 ± 3, 26 ± 2, and 22 ± 3 ms, in Control, Lidocaine, and Phenytoin groups, respectively. Bupivacaine infusion induced a significant QRS widening in all groups (Fig. 2). QRS duration remained stable from T10 to T40 in the Control group whereas it increased during the same period in the two other groups. Ventricular arrhythmias were observed only at T40 (i.e., at the largest concentration of lidocaine or phenytoin) in four of eight rabbits in the Phenytoin group and in two of eight rabbits in the Lidocaine group. After discontinuation of bupivacaine infusion (T40), QRS widening reversed quickly in all groups. QRS duration at T60 was comparable to baseline value (T0) in all rabbits.
View larger version (33K):
[in this window]
[in a new window]
|
Figure 2. QRS duration expressed as percentage change from basal value at T0. At T10, hearts had only received bupivacaine regardless of their group. At T20, T30, and T40, they had received bupivacaine with either saline or increasing concentrations of phenytoin or lidocaine. Bupivacaine was stopped at T40 in all groups. The other drug (i.e., saline, phenytoin, or lidocaine) was stopped at T50.
|
|
Bupivacaine, lidocaine, and phenytoin concentration-versus-time data were adequately fitted by a simple one-compartment model. The increase in QRS duration-versus-time data were best fitted by using the Emax model. Elimination of the three drugs was rapid (Table 1). Lidocaine and phenytoin significantly increased the QRS duration when added to the preparation. The objective function was -797.2 when the three drugs were considered acting on QRS, -865.1 when only bupivacaine and phenytoin was considered active (P < 0.001), -868.3 when only bupivacaine and lidocaine was considered active (P < 0.001), and -893.4 when both phenytoin and lidocaine were considered inactive (P < 0.001).
|
Discussion
|
|---|
We found that QRS widening produced by bupivacaine was not decreased by phenytoin and by lidocaine in isolated rabbit hearts. Moreover, both drugs significantly enhanced bupivacaine-induced intraventricular block. After bupivacaine discontinuation, intraventricular conduction impairment reversed faster in the group receiving bupivacaine alone than in the Lidocaine and Phenytoin groups.
Pharmacokinetic and pharmacodynamic variables obtained by our model were similar to those we have already described (8) (Table 1). We used a 3-µM bupivacaine concentration in the inflow perfusate to induce a significant widening of QRS without inducing any dysrhythmia. Depending on the subjects studied or on the model used, this concentration of bupivacaine is either just at or markedly above the upper limit of bupivacaine unbound concentration leading to the first manifestations of cardiovascular toxicity (4,8,11).
Lidocaine is the chief class 1B antiarrhythmic drug in the Vaughan Williams classification and it is extensively used for the treatment of acute ventricular arrhythmias. Despite a similar site of action on the Na+ channel, lidocaine and bupivacaine exert different effects on the heart. Because its effects on intraventricular conduction and on the effective refractory period are moderate compared with those of bupivacaine (12), lidocaine exhibits only a small arrhythmogenic potency (13). In isolated rabbit hearts, the increase in QRS duration induced by bupivacaine is much greater than the one induced by lidocaine. These pharmacokinetic differences are not related to differences in myocardial uptake and disposition kinetics (8).
Phenytoin is used as an antiepileptic. Its action is related to the block of fast sodium channels at the central nervous system. However, this drug also exerts an effect on intracardiac conduction. It is therefore considered to be a class IB antiarrhythmic drug (14). Phenytoin both enhances and delays atrioventricular conduction, shortens the refractory period, and suppresses automaticity in ventricular muscle (14). Moreover, some authors showed that phenytoin could block the cardiac Na+ channel in Purkinje cells in a frequency-dependent manner (15), whereas other authors failed to demonstrate any use-dependence of that block (16). Severe conduction blocks have been described with IV phenytoin (14,17). These blocks were mainly described after IV bolus of phenytoin or rapid infusions, justifying recommendations to perform only slow infusions of phenytoin (17).
Because ventricular conduction only depends on fast sodium channels, QRS widening that directly reflects the slowing of ventricular conduction velocity allows the accurate measure of the effect of drugs on ventricular conduction in the Langendorff preparation (1,5,18). Class Ib antiarrhythmics such as lidocaine and bupivacaine increase the ventricular conduction time and therefore widen QRS (1,19). Because of a decreased use dependence, lidocaine is less cardiotoxic than bupivacaine. Lidocaine competitively displaces bupivacaine from its sodium channels binding site in vitro (5). This concept has led to the proposal of using lidocaine to treat bupivacaine intoxication (2,6). In a similar way, phenytoin has been used to treat manifestations of bupivacaine cardiotoxicity (2,3). However, in these observations, phenytoin and/or lidocaine were administered after several other drugs and this treatment was often delayed after the cardiac manifestations of bupivacaine overdose. We cannot exclude that the plasma bupivacaine concentrations had decreased below the toxic threshold when the successful treatment by phenytoin or by lidocaine or by both drugs was administered to these patients. The most important part of the treatment of bupivacaine-induced cardiac arrest is a sustained resuscitation to ensure a correct coronary perfusion that allows an efficient bupivacaine washout from the heart (8). Although lidocaine intoxication has less arrhythmogenic potencies and a better prognosis compared with those of bupivacaine intoxication (4,8,13), we showed, in accordance with another experimental study, either no effect or deleterious effects of lidocaine to treat bupivacaine cardiotoxicity (6). Therefore, the hypothetical displacement effect does not seem to occur in vivo. This is the reason we wanted to study the ex vivo effect of both drugs on isolated hearts receiving bupivacaine. Another theoretical reason for not using lidocaine to treat bupivacaine-induced cardiotoxicity is the fact that most studies aiming to evaluate the toxicity of local anesthetic mixtures demonstrated an additivity (20–22). In this study, we have clearly demonstrated that adding either phenytoin or lidocaine to the inflow perfusate increased the intraventricular conduction block induced by bupivacaine. These results suggest that these drugs should not be used to treat patients with bupivacaine overdose.
|
Acknowledgments
|
|---|
Supported by the French Ministry of Research and association MAPAR (Mises au Point en Anesthésie et Réanimation). NK was supported by a grant from Osaka City University.
The authors gratefully acknowledge Mrs. Régine le Guen for her technical assistance.
|
Footnotes
|
|---|
Presented in part at the European Society of Anaesthesiologists Annual Meeting, Gothenburg, Sweden, April 7–10, 2001.
|
References
|
|---|
-
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: 784–92.[Web of Science][Medline]
-
Maxwell LG, Martin LD, Yaster M. Bupivacaine-induced cardiac toxicity in neonates: successful treatment with intravenous phenytoin. Anesthesiology 1994; 80: 682–6.[Web of Science][Medline]
-
Davis NL, de Jong RH. Successful resuscitation following massive bupivacaine overdose. Anesth Analg 1982; 61: 62–4.[Free Full Text]
-
Groban L, Deal DD, Vernon JC, et al. Cardiac resuscitation after incremental overdosage with lidocaine, bupivacaine, levobupivacaine, and ropivacaine in anesthetized dogs. Anesth Analg 2001; 92: 37–43.[Abstract/Free Full Text]
-
Clarkson CW, Hondeghem LM. Evidence for a specific receptor site for lidocaine, quinidine, and bupivacaine associated with cardiac sodium channels in guinea pig ventricular myocardium. Circ Res 1985; 56: 496–506.[Abstract/Free Full Text]
-
Kasten GW, Martin ST. Bupivacaine cardiovascular toxicity: comparison of treatment with bretylium and lidocaine. Anesth Analg 1985; 64: 911–6.[Abstract/Free Full Text]
-
Theissen O, Boileau S, Cornet C, et al. Analgésie après thoracotomie par administration extrapleurale de bupivacaïne à débit continu. Ann Fr Anesth Réanim 1993;12:265–72.
-
Mazoit JX, Orhant EE, Boico O, et al. Myocardial uptake of bupivacaine. I. Pharmacokinetics and pharmacodynamics of lidocaine and bupivacaine in the isolated perfused rabbit heart. Anesth Analg 1993; 77: 469–76.[Abstract/Free Full Text]
-
Sheiner LB, Beal SL, Boeckmann A. NONMEM users guides 1–6. San Francisco: Regents of the University of California, 1979–1999.
-
Yamaoka K, Nakagawa T, Uno T. Application of Akaike’s information criterion (AIC) in the evaluation of linear pharmacokinetic equations. J Pharmacokinet Biopharm 1978; 6: 165–75.[Web of Science][Medline]
-
Bardsley H, Gristwood R, Baker H, et al. A comparison of the cardiovascular effects of levobupivacaine and rac-bupivacaine following intravenous administration to healthy volunteers. Br J Clin Pharmacol 1998; 46: 245–9.[Web of Science][Medline]
-
Kasten GW. Amide local anesthetic alterations of effective refractory period temporal dispersion: relationship to ventricular arrhythmias. Anesthesiology 1986; 65: 61–6.[Web of Science][Medline]
-
Groban L, Deal DD, Vernon JC, et al. Ventricular arrhythmias with or without programmed electrical stimulation after incremental overdosage with lidocaine, bupivacaine, levobupivacaine, and ropivacaine. Anesth Analg 2000; 91: 1103–11.[Abstract/Free Full Text]
-
Randazzo DN, Ciccone A, Schweitzer P, Winters SL. Complete atrioventricular block with ventricular asystole following infusion of intravenous phenytoin. J Electrocardiol 1995; 28: 157–9.[Web of Science][Medline]
-
Spinelli W, Rosen MR. Frequency-dependent actions of phenytoin in adult and young canine Purkinje fibers. J Pharmacol Exp Ther 1986; 238: 794–801.[Abstract/Free Full Text]
-
Mestre M, Djellas Y, Carriot T, Cavero I. Frequency-independent blockade of cardiac Na+ channels by riluzole: comparison with established anticonvulsants and class I anti-arrhythmics. Fundam Clin Pharmacol 2000; 14: 107–17.[Web of Science][Medline]
-
Berry JM, Kowalski A, Fletcher SA. Sudden asystole during craniotomy: unrecognized phenytoin toxicity. J Neurosurg Anesthesiol 1999; 11: 42–5.[Web of Science][Medline]
-
de La Coussaye J, Brugada J, Allessie MA. Electrophysiologic and arrhythmogenic effects of bupivacaine: a study with high-resolution ventricular epicardial mapping in rabbit hearts. Anesthesiology 1992; 77: 32–41.
-
Mazoit JX, Kantelip JP, Orhant EE, Talmant JM. Myocardial uptake of lignocaine: pharmacokinetics and pharmacodynamics in the isolated perfused heart of the rabbit. Br J Pharmacol 1990; 101: 843–6.[Web of Science][Medline]
-
de Jong RH, Bonin JD. Mixtures of local anesthetics are no more toxic than the parent drugs. Anesthesiology 1981; 54: 177–81.[Web of Science][Medline]
-
Mets B, Janicki PK, James MF, et al. Lidocaine and bupivacaine cardiorespiratory toxicity is additive: a study in rats. Anesth Analg 1992; 75: 611–4.[Abstract/Free Full Text]
-
Brecher MJ, Greenberg RA, Greene NM. Drug interactions (toxicity) associated with drug combinations. Acta Anaesthesiol Scand 1972; 16: 22–6.[Web of Science][Medline]
Accepted for publication September 4, 2001.
This article has been cited by other articles:
|
|
|
|
 
D. Schwartz, T. VadeBoncouer, G. Weinberg, A. Tung, B. J. Sweitzer, and T. W. Cutter
Was Case Report a Case of Unrecognized Local Anesthetic Toxicity? * Response
Anesth. Analg.,
June 1, 2003;
96(6):
1844 - 1845.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
