Bupivacaine is widely used for regional anesthesia, although it is undoubtedly more cardiotoxic than most related drugs, especially lidocaine, as demonstrated in the isolated perfused heart [1,2] as well as in the heart in situ [3-6]. The disorders elicited by bupivacaine consist of initial depression of intraventricular conduction, followed by reentrant arrhythmias [6,7]. Myocardial contractility is affected only later . Bupivacaine remains devoid of significant action on left ventricular dP/dtmax at plasma concentrations responsible for serious conduction disturbances, ectopic beats, and even ventricular tachycardias.
Actually, there is normally a large safety margin between bupivacaine plasma concentrations arising from resorption after regional anesthesia (peak concentration below 1.5 micro gram/mL) and those inducing the earliest disorders (above 4.0 micro gram/mL). Consequently, severe accidents generally are observed only after inadvertent intravascular injection. But various factors, such as hypothermia , hyponatremia, hyperkalemia , and drug interactions [10,11], sensitize the heart to the effects of bupivacaine and lower the concentrations sufficient to produce accidents after regional anesthesia. Likewise, the cardiac toxicity of bupivacaine is enhanced by myocardial ischemia , and the onset of fibrillation triggered by a given experimental ischemia is substantially shortened by bupivacaine at moderate plasma concentrations.
The aim of this work is to further elucidate the mechanism for this increase in ischemic ventricular fibrillation by the concurrent study of conduction and excitability in the course of ischemia.
After approval of the protocol by the institutional animal investigation committee, eight domestic pigs of either sex, approximately 6 wk of age and weighing 18-24 kg, were used for the study. The animals, premedicated with droperidol 1.25 mg/kg intramuscularly 1 h before the experiment, were anesthetized with 0.05 mg/kg flunitrazepam and 80 mg/kg chloralose intravenously injected via the marginal ear vein. They were ventilated with a respirator through a tracheotomy tube. This respirator delivered an air-oxygen mixture (40% and 60% respectively), and PaO2, PaCO2, and pHa were checked several times to be within the ranges of 80-200 mm Hg, 32-42 mm Hg, and 7.35-7.50, respectively. An electronic esophageal thermometer enabled core temperature to be monitored throughout the experiment. Because of the sensitivity of intraventricular conduction to the combined action of bupivacaine and hypothermia , core temperature was kept constant (37.7-39.0 degrees C). This was done by an infrared heater placed at a variable distance from the animals. Finally, mean arterial blood pressure was recorded continuously through a catheter inserted in a carotid artery and connected to a polygraph.
Each pig, lying on its right side, had a left thoracotomy with resection of the fourth and fifth ribs to expose the heart. After opening the pericardium, the left anterior descending coronary artery was dissected close to its origin and a snare was passed around it for complete but temporary occlusion.
To exclude variations of vulnerability to fibrillation linked with the slowing or acceleration of heart rate, all measurements of electrical fibrillation threshold (EFT) were achieved at the same heart rate, 180 bpm. With this relatively high rate, the ventricles could be electrically driven in all the animals, in spite of accelerations likely to occur during the experiments. Values of approximately 7 mA were obtained prior to ischemia with the chosen duration of impulses, 100 ms (instead of 2 ms for the usual pacing). The impulses, delivered at regular intervals (333 ms period corresponding to the rate of 180 bpm) by an S1 stimulator (Hugo Sachs, Freiburg im Brisgau, Germany), were transmitted to the left ventricle by a bipolar electrode, the tip of which had been introduced into the subepicardial layer at a point located in the area intermittently subjected to ischemia (or at a border point between the ischemic and nonischemic zones in order to measure conduction time in these two zones simultaneously, as indicated below). The EFT was determined by applying to the ventricles 100 ms stimuli of increasing intensity, by steps of 1.0 or 0.5 mA, until fibrillation occurred, according to the method previously described .
The establishment of control EFT was checked in each animal in the absence of ischemia. The reproducibility of the EFT time course with ischemia was also checked. This time course implied the establishment of several coronary occlusions of increasing duration (30, 60, 120, 180 s), and the repeated ischemias were liable to produce increasingly slighter or increasingly stronger effects as discussed below. In fact, similar values were found for EFT in successive determinations, provided that the periods of ischemia were not too long (shorter than 4-5 min) and were separated from each other by intervals long enough for sufficient recovery (2-10 min according to the ischemia duration). The effects of the preceding ischemia did not completely disappear when ischemia was repeated and the repetition resulted in sensitization.
As the gradual decrease of EFT during coronary occlusion approaches 0 mA, pacing stimuli will then trigger fibrillation instead of causing synchronous contractions. The time of the onset of fibrillation under ventricular pacing was also taken into account when evaluating vulnerability to fibrillation. Its diminishing confirms the rapidity of EFT decrease; its lengthening confirms the delay.
In all cases, coronary occlusion was discontinued as soon as the onset of fibrillation and synchronous activity was reestablished when one or, if necessary, several countershocks from a Sirecard F defibrillator (Siemens, Erlangen, Germany) were delivered on the thoracic wall.
Monophasic action potential (MAP) was recorded through a Catronic ORX electrode 6 F (Plastimed, Saint-Leu-La-Foret, France) advanced within the epicardium, in ischemic and nonischemic areas; MAP recording electrodes were placed at an equal distance (approximately 2 cm) on either side of the pacing electrode which was on the line separating the two areas, as shown in Figure 1. MAP duration, correlated with polarization of the fibers , enabled the decrease in polarization due to ischemia to be followed when the variations due to heart rate were prevented by pacing the ventricles at a constant rate. Therefore, MAP duration was measured under electrical driving, but with impulses of 2 ms duration (see above), at 90% repolarization, just before each determination of EFT.
Conduction time between the pacing electrode and the MAP recording electrodes was thus concurrently obtained. Conduction time is given by the interval spike of stimulation-steep upstroke of MAP. Given the respective positions of the pacing and recording electrodes mentioned above, conduction velocity might be comparatively studied in ischemic and nonischemic areas throughout the ischemic periods.
Just after the onset of fibrillation, MAP electrodes registered fibrillation waves at a rate that appeared to be different in ischemic and nonischemic areas. The number of fibrillation waves per time unit, regardless of their amplitude , were then counted in each of the two areas.
Finally, a surface electrocardiogram was recorded in standard and precordial leads with a Mingograf 34 electrocardiograph (Elema-Schonander, Stockholm, Sweden) and, in the intervals of the recordings, monitored with an EM 531 oscilloscope (Siemens, Erlangen, Germany).
Two control measurements of EFT, MAP duration, conduction time, and fibrillation rate were performed 10 and 5 min prior to ischemia and at the end of ischemic periods of 30, 60, 120, and 180 s obtained by tightening the snare previously passed around the coronary artery. The first four measurements (before ischemia and after 30- and 60-s ischemic periods) were separated by a 150- to 300-s interval and the last two measurements (120- and 180-s ischemic periods) by a 5- to 10-min interval from the preceding one. All the measurements were repeated according to the same procedure, after intravenous injection of a 1.00-mg/kg initial dose of bupivacaine, followed by a constant infusion of 0.04 mg centered dot kg-1 centered dot min-1 over 25 min. The measurements of the bupivacaine series were started 5 min after the last measurement of the control series and, therefore, bupivacaine was administered 15 min after this measurement. Reverse order (first bupivacaine series then control series), although theoretically desirable, was not possible since, as the effects of bupivacaine do not disappear for 20-40 min, the time elapsing between the two series would have been too long to maintain reproducibility of the values.
Blood sampling for bupivacaine assay was performed via an arterial catheter before each electrophysiologic recording (starting 7 min after the initial dose). Plasma concentrations of bupivacaine were determined by a high-performance liquid chromatography method  with a limit of sensitivity of 20 ng/mL and a 4.2% coefficient variation over the concentration range of 0.1-5.0 micro gram/mL.
After an analysis of variance to confirm the absence of significant differences between initial values of the control series and the bupivacaine series, a paired Student's t-test was used to compare the values obtained at the same time under the two circumstances, particularly during ischemia.
Results are arithmetic means +/- SE and the differences are considered significant when P < 0.05.
Bupivacaine plasma concentrations were found between 1.4 and 1.8 micro gram/mL during the whole test period in all the animals. These plasma concentrations increased the EFT significantly, from 7.1 +/- 0.4 to 9.5 +/- 0.4 mA, when the ventricles were normally perfused. In contrast, as soon as the coronary artery was occluded, bupivacaine did not attenuate, but enhanced, the decrease of EFT resulting from the occlusion (2.2 +/- 0.5 vs 4.8 +/- 0.5 mA, P < 0.05, at 30 s). Then, EFT remained below control values (0.3 +/- 0.1 vs 2.0 +/- 0.5 mA, P < 0.05, at 60 s), so that the relationship of EFT-ischemia duration was shifted to the left Figure 2.
However, a strict comparison of EFT in the absence and presence of bupivacaine became impossible before the end of the test period. Fibrillation occurred at 120 s of ischemia in all the animals given bupivacaine. EFT, which was considerably higher (about 7 mA) than the pacing threshold (0.3-0.4 mA), then decreased to the pacing threshold level and triggered fibrillation. Thus the highest number of fibrillations observed were in the presence, rather than in the absence, of bupivacaine. This is another proof of ischemia sensitization Figure 3.
Rapid EFT decrease leading to an earlier fibrillation after bupivacaine cannot be linked with an aggravation of ischemic depolarization, since the shortening of MAP duration produced by ischemia was not as marked after bupivacaine administration Figure 4, which significantly increased MAP duration before ischemia or in the nonischemic area. But a relationship could be established with the lengthening of conduction time (41 +/- 2-79 +/- 4 ms, P < 0.05, with bupivacaine alone, and up to 99 +/- 4 ms under the combined influence of bupivacaine and ischemia) Figure 5, as well as with the decrease of the fibrillation rate from 2920 +/- 120 to 1920 +/- 90 waves/min or from 47.0 +/- 2.0 to 32.0 +/- 1.5 Hz, P < 0.05, with bupivacaine alone, but decreasing to 1160 +/- 110 waves/min or to 19.3 +/- 1.8 Hz under the combined influence of bupivacaine and ischemia Figure 6.
Mean arterial blood pressure, significantly diminshed by ischemic episodes involving a large myocardial area (102 +/- 6-84 +/- 2 mm Hg), was also moderately decreased by bupivacaine (92 +/- 4 mm Hg) and more considerably by the combination of ischemia and bupivacaine (78 +/- 5 mm Hg).
Bupivacaine has been shown by the present work, and in past experiments  to enhance ventricular fibrillation which results from myocardial ischemia. This is the case when ischemia is sufficiently severe. Fibrillation occurs earlier when bupivacaine is added, even though plasma concentration is not very large but is similar to those usually reached after regional anesthesia. The present experiments, based on the simultaneous measurement of EFT, MAP duration, and intraventricular conduction time, describes what happens before the beginning of fibrillation, especially concerning excitability and conduction, the alterations of which are responsible for fibrillation.
A strict reproducibility of EFT decrease produced by successive ischemias of increasing duration in the absence and presence of bupivacaine was necessary to demonstrate bupivacaine time effects. This reproducibility can be doubted since brief ischemia episodes have been claimed to protect against arrhythmogenic effects of subsequent ischemias [17,18]. Consequently, the repeated ischemias used to determine EFT change are likely to give rise to increasingly slighter effects. The second series of measurements to determine EFT decrease with bupivacaine is even more debatable. As a matter of fact, the effects of "preconditioning" by prior ischemias are conflicting. Increased incidence of ventricular fibrillation has been reported in preconditioned dogs  and this finding is consistent with the reduction of MAP duration in the rabbit isolated myocardium .
According to our personal observations, the values of EFT, MAP duration, and intraventricular conduction time do not significantly vary in successive ischemias under the conditions of our experimental protocol, e.g. ischemia periods shorter than 4 min separated by intervals allowing the disappearance of electrical and biochemical disorders induced by ischemia, 2 to 10 min depending on the duration of ischemia. In particular, if the recovery period is too short, prior ischemias do not protect against fibrillation, but sensitize to the phenomenon. This possible error was carefully avoided in the present experiments.
In this kinetic study on the effects of ischemia, EFT gradual decline first leads one to think that fibrillation depends on an excess of excitability of the myocardial fibers. Other factors such as depression of conduction are likely to intervene in the strength of the depolarizing current necessary to trigger fibrillation. This force is considerably lowered, to near 0 mA, with natural stimuli then triggering fibrillation instead of synchronous contractions, as usual.
Depression of intraventricular conduction plays a major role in the triggering of fibrillation by ischemia because it delays activation of the fibers farthest from the pacemaker. Stimuli thus ends by falling in a refractory period of these fibers  which, in the absence of regular excitation, tend to lose their resting membrane potential. When this potential approaches the threshold potential for depolarization, the increased excitability reaches spontaneous excitability, i.e., automaticity and rhythmic depolarization of each fiber appears. Such depolarization remains independent of surrounding fibers. The maximal level of polarization is not sufficient for the activation of fast sodium channels [22,23] and, consequently, for impulse propagation. This spontaneous isolated activity of the ventricular fibers corresponds to fibrillatory activity.
Depolarization of the fibers by ischemia is responsible for the defect of conduction according to the relationship demonstrated by Weidmann . It is also responsible for a general enhancement of myocardial excitability. This property varies in inverse ratio to the differences between resting potential and threshold potential for depolarization . Depolarization is reflected by the shortening of MAP duration, since a correlation of this duration with resting membrane potential has been established . The shortening of MAP duration is essentially secondary to cellular depletion in potassium [26,27] which is no longer reintroduced into the cell during diastole when energy availability is reduced by lack of oxygen. The deficiency in ion pumps concurrently affects sodium and calcium extrusion , which participate in depolarization by accumulating in the cellular medium. However, when accumulation of sodium, and especially calcium, is too large because of the severity or duration of ischemia, excitability increase stops and then decreases and excess calcium becomes detrimental to excitability, according to a "bell-shaped curve" relationship  between intracellular calcium concentration and this property.
Bupivacaine increases EFT from approximately 7.0 to 9.5 mA in the intact heart. It produces antifibrillatory properties under these conditions, as expected from the well known blockade of the sodium channel [29,30], associated with blocking the calcium inward current and potassium outward current . Conceivably, excitability is depressed by bupivacaine (as intraventricular conduction).
Thus the profibrillatory effect of bupivacaine, peculiar to ischemic conditions, is probably related to depolarization caused by ischemia. Bupivacaine, a blocking drug of passive ion fluxes, is incapable of counteracting the reduction of active ion transfers depending on the deficiency of energetic resources. It can even aggravate the situation by inhibiting production of cyclic adenosine monophosphate . Bupivacaine does not prevent ischemic depolarization since it does not attenuate substantial shortening of MAP duration, correlated with the decrease in resting potential . Under the dual depressant influence of ischemia and bupivacaine, intraventricular conduction is sufficiently impeded to unmask potential automaticity of the myocardial fibers according to the process pointed out above. The role of the conduction disorder in the triggering of fibrillation prevails. An argument for this prevalence is provided by the decrease of the fibrillation rate which appeared to be related to conduction velocity. This decrease, already substantial after bupivacaine, becomes considerable when ischemia is added. However, depression of conduction in these experiments was dependent on the relatively high heart rate of stimulation, 180 bpm.
A high driving rate is known to enhance the effects of the drugs inhibiting sodium channel, class I antiarrhythmic drugs and bupivacaine [5,6,33,34]. This enhancement may be interpreted by the modulated receptor theory proposed by Hondeghem and Katzung , which postulates a greater affinity for the sodium channel in an inactivated state corresponding to systoles. As the time spent in systolic periods increases with the rate, the binding and the action of bupivacaine is favored by tachycardia.
But tachycardia also implies retention of sodium and calcium in the myocardial fibers [36,37] as a result of the increase of passive ion fluxes related to the greater number of systoles per time unit, while the shortening of the diastolic periods opposes the active ion transfers [6,33]. Because of the reduction of the difference of extra- and intracellular concentrations, overload in sodium and calcium is detrimental to depolarization (rate-dependent effect) and conduction is increasingly slower when intracellular concentration increases in the first accelerated systoles (use-dependent effect). At the same time, tachycardia determines a potassium leakage from the cell [38,39] which tends to depolarize the fibers, with a decrease of EFT from 10-12 mA to 5-6 mA when heart rate increases from 100-120 to 180-200 bpm . Depolarization, which results from tachycardia, sensitizes the unwelcome effects of bupivacaine on conduction.
The anesthetic drugs used in this study may have influenced the rate dependency of the bupivacaine effects, but only slightly.
In conclusion, bupivacaine even at moderate plasma levels induces, when associated with myocardial ischemia, a decrease of the electrical ventricular fibrillation threshold. The present study suggests that this effect is explained mainly by depression of ventricular conduction and is enhanced by tachycardia.
1. Tanz RD, Heskett T, Loehning RW, Fairfax CA. Comparative cardiotoxicity of bupivacaine and lidocaine in the isolated perfused mammalian heart. Anesth Analg 1984;63:549-56.
2. Wheeler DM, Bradley FI, Woods WT Jr. The electrophysiologic actions of lidocaine and bupivacaine in the isolated, perfused canine heart. Anesthesiology 1988;68:201-12.
3. De Jong RH, Gamble CA, Bonin JD. Bupivacaine causes arrhythmias in normokalemic cats, lidocaine does not. Reg Anesth 1982;7:84-5.
4. Kotelko DM, Shnider SM, Dailey PA. Bupivacaine-induced cardiac arrhythmias in sheep. Anesthesiology 1984;60:10-8.
5. Hotvedt R, Refsum H, Helgesen KG. Cardiac electrophysiologic and hemodynamic effects related to plasma levels of bupivacaine in the dog. Anesth Analg 1985;64:388-94.
6. Timour Q, Freysz M, Lang J, et al. Electrophysiological study in the dog of the risk of cardiac toxicity of bupivacaine. Arch Int Pharmacodyn Ther 1987;287:65-77.
7. Freysz M, Timour Q, Mazze RI, et al. Potentiation by mild hypothermia of ventricular conduction disturbances and reentrant arrhythmias induced by bupivacaine in dogs. Anesthesiology 1989;70:799-804.
8. Beal JL, Freysz M, Timour Q, et al. Haemodynamic effects of high plasma concentrations of bupivacaine in the dog. Eur J Anaesthesiol 1988;5:251-60.
9. Timour Q, Freysz M, Mazze RI, et al. Enhancement by hyponatremia and hyperkalemia of ventricular conduction and rhythm disorders caused by bupivacaine. Anesthesiology 1990;72:1051-6.
10. Timour Q, Freysz M, Couzon P, et al. Possible role of drug interactions in bupivacaine-induced problems related to intraventricular conduction disorders. Reg Anesth 1990;15:180-5.
11. Bertrix L, Timour Q, Mazze R, et al. Adverse interaction between bupivacaine and halothane on ventricular contractile force and intraventricular conduction in the dog. Anesth Analg 1991;73:434-40.
12. Freysz M, Timour Q, Bertrix L, et al. Enhancement by ischemia of the risk of cardiac disorders, especially fibrillation, in regional anesthesia with bupivacaine. Acta Anaesth Scand 1993;37:350-6.
13. Aupetit JF, Timour Q, Chevrel G, et al. Attenuation of the ischemia-induced fall of electrical ventricular fibrillation threshold by a calcium antagonist, diltiazem. Naunyn Schmiedeberg's Arch Pharmacol 1993;348:509-14.
14. Janse MJ, Wit AL. Electrophysiological mechanisms of ventricular arrhythmias resulting from myocardial ischemia and infarction. Physiol Rev 1989;69:1049-1169.
15. Farges JP, Faucon G, Lievre M, Ollagnier M. Relationship between atrial and ventricular rates of fibrillation and cardiac contractile tissue refractory periods in dog. Br J Pharmacol 1978;63:587-91.
16. Lindberg RIP, Pihlajamaki KK. High-performance liquid chromatographic determination of bupivacaine in human serum. J Chromatogr 1984;309:369-74.
17. Hagar JM, Hale SL, Kloner RA. Effect of preconditioning ischemia on reperfusion arrhythmias after coronary artery occlusion and reperfusion in the rat. Circ Res 1991;68:61-8.
18. Vegh A, Komori S, Szekeres L, Parratt JR. Antiarrythmic effects of preconditioning in anaesthetised dogs and rats. Cardiovasc Res 1992;26:487-95.
19. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 1986;74:1124-36.
20. Tan HL, Mazon P, Verberne HJ, et al. Ischemic preconditioning delays ischemia-induced cellular electrical uncoupling in rabbit myocardium by activation of ATP-sensitive potassium channels. Circ Res 1993;27:644-51.
21. Timour Q, Aupetit JF, Loufoua-Moundanga J, et al. Role of asynchronous activation of the ventricular fibres by an ectopic pacemaker in the accidents, especially fibrillation, caused by Ic antiarrhythmic drugs. Fundam Clin Pharmacol 1992;6:159-67.
22. Coraboeuf E. Ionic basis of electrical activity in cardiac tissues. Am J Physiol 1978;234:H101-16.
23. Reuter H. Ion channels in cardiac cell membranes. Annu Rev Physiol 1984;46:473-84.
24. Weidmann S. The effect of the cardiac membrane potential on the rapid availability of sodium carrying system. J Physiol (Lond) 1955;127:213-24.
25. Lyons CJ, Burgess MJ, Abildskov JA. Effects of acute hyperkalemia on cardiac excitability. Am Heart J 1977;94:755-63.
26. Gettes LS, Cascio WE, Johnson T, Flect WF. Local myocardial biochemical and ionic alterations during myocardial ischaemia and reperfusion. Drugs 1991;42(1 Suppl):7-13.
27. Heijnis JB, Coronel R, Van Zwieten PA. The effects of calcium antagonists on extracellular potassium accumulation during global ischemia in isolated perfused rat hearts. Cardiovasc Drugs Ther 1991;5:1035-41.
28. Lang J, Timour Q, El Chebly M, Faucon G. Effect of gradual rise in plasma calcium concentration on the impairment of atrioventricular conduction due to verapamil. J Cardiovasc Pharmacol 1986;8:6-13.
29. Hondeghem IM, Katzung BG. Time- and voltage-dependent interactions of antiarrhythmic drugs with cardiac sodium channels. Biochim Biophys Acta 1977;472:373-98.
30. Clarkson CW, Hondeghem IM. 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.
31. Butterworth JF IV, Strichartz GR. Molecular mechanisms of local anesthesia: a review. Anesthesiology 1990;72:711-34.
32. Butterworth JF IV, Brownlow RC, Leith JP, et al. Bupivacaine inhibits cyclic -3 prime,5-adenosine monophosphate production. A possible contributing factor to cardiovascular toxicity. Anesthesiology 1993;78:88-95.
33. Lang J, Timour Q, Aupetit JF, et al. Frequency- and time-dependent depression of ventricular distal conduction by two novel antiarrhythmic drugs, cibenzoline and flecainide. Arch Int Pharmacodyn Ther 1988;293:97-108.
34. De La Coussaye JE, Brugada J, Allessie MA. Electrophysiologic and arrhythmogenic effects of bupivacaine. Anesthesiology 1992;77:132-41.
35. Hondeghem IM, Katzung BG. Antiarrhythmic agents: the modulated receptor mechanism of action of sodium and calcium channel-blocking drugs. Annu Rev Pharmacol 1984;24:387-423.
36. Boyett MR, Jewell BR. Analysis of the effects of changes in rate and rhythm upon electrical activity of the heart. Prog Biophys Mol Biol 1980;36:1-52.
37. Franz MR, Schaefer J, Schottler W, et al. Electrical and mechanical restitution of the human heart at different rates of stimulation. Circ Res 1983;53:815-22.
38. Kline RP, Morad M. Potassium efflux in heart muscle during activity. Extracellular accumulation and its implications. J Physiol (Lond) 1978;280:537-58.
39. Boyett MR, Fedida D. A simple model of ion concentration changes in the heart during repetitive activity. J Physiol (Lond) 1983;342:52-3 P.
40. Timour Q, Aupetit JF, Loufoua J, et al. Heart rate and vulnerability to fibrillation in the acutely ischaemic myocardium [abstract]. Fundam Clin Pharmacol 1992;6:214.