Cardiovascular collapse can occur following inadvertent intravascular injection of bupivacaine . Amide local anaesthetics depress myocardial contractility in proportion to their local anaesthetic potency . Death appears to occur as a result of cardiac dysrhythmias rather than myocardial depression [3,4].
Bupivacaine, the amide local anaesthetic with greatest cardiotoxicity, causes dysrhythmias by producing a unidirectional conduction block that facilitates re-entry . Ropivacaine is an amide local anaesthetic agent, similar in structure to bupivacaine, but prepared as a pure S-enantiomer. The cardiodepressant and electrophysiological toxicity ratios for bupivacaine to ropivacaine are 4:3 and 15:6.7, respectively . Ropivacaine has an alkyl side chain, which is intermediate in length (C3H7) between mepivacaine (CH3) and bupivacaine (C4H9), and the length of the side chain appears to determine the affinity of these drugs for both the Na+ and K+ channels .
Amide local anaesthetics bind inactivated and open Na+ channels in cardiac tissue in a similar manner to neural tissue, producing a block to Na+ conductance . Bupivacaine has an affinity for inactivated channels and produces a conduction block, which is both use and voltage dependent (i.e. block increases as heart rate (HR) increases and can be reversed by hyperpolarization) . Blockade of the Na+ channel is important in cardiac tissue - both in contractile and Purkinje fibres - and this mechanism is presumably responsible for the electrophysiological and contractility effects of bupivacaine . Nodal tissue differs in that the slow calcium current is more prominent in these cells . In a previous study , we consistently observed decreased left ventricular contractility and ST segment depression following the administration of intracoronary ropivacaine in anaesthetized dogs. We selected three agents, on the basis of potential for resuscitation, following ropivacaine, namely nicorandil (a potassium channel 'opener' in view of the evidence that bupivacaine inhibits early and late inwards potassium-rectifier currents [9-11]), calcium for inotropy  and nitroglycerin (glyceryl trinitrate) for a potential anti-ischaemic effect - we observed in a previous study  that ropivacaine produced ST depression following intracoronary administration. The effects of i.v. and intracoronary nicorandil, i.v. calcium chloride and i.v. nitroglycerin were studied following the administration of intracoronary ropivacaine.
These experiments were carried out by licensed investigators in compliance with the Cruelty to Animals Act 1876 (Department of Health, Ireland). Six labradors (five females and one male) were studied. The dogs were premedicated with subcutaneous morphine 10 mg and 30 min later were anaesthetized with i.v. pentobarbital (induction 30 mg kg−1, maintenance 3 mg kg−1 every 30 min) through a cannula inserted into the long saphenous vein under local anaesthesia. After tracheostomy, positive pressure ventilation of the lungs was initiated using a volume-cycled Palmer animal ventilator. Ventilation with oxygen-enriched air was adjusted to an end-tidal carbon dioxide partial pressure of 4.7-5.3 kPa (Morgan capnograph®; Morgan Medical Ltd., Rainham, Kent, UK) and the FiO2 adjusted to maintain PaO2 at or above 12.0 kPa.
Electrocardiographic (ECG) leads II and V5 (SE 40001®, electromedical multichannel amplifier; Southern Electric Laboratories, Middlesex, UK) were recorded and rectal temperature (Harvard homeothermic blanket control unit and probe; Harvard Apparatus Ltd., Edenbridge, Kent, UK) was measured continuously during each experiment. Rectal temperature was maintained at 37-39°C. The left femoral vein and both femoral arteries were cannulated to facilitate sampling and continuous measurement of aortic pressure. The right internal jugular vein was cannulated to facilitate rapid central administration of the resuscitation drugs. A left thoracotomy through the sixth left intercostal space was performed and a 5-F, 70 cm catheter-tipped manometer (MPC-500®; Millar Instruments, Houston, Texas, USA) was placed via the cardiac apical dimple. Cardiac output (CO) was measured using a transit time perivascular flowprobe (Transonic® flow probe; Linton Instruments, Norfolk, UK) placed around the ascending aorta. A 24-G plastic catheter (Insyte, Vialon, Spain) was introduced into the left circumflex coronary artery under direct vision.
A dose-finding study was initially performed in each animal to establish the dose of ropivacaine that produced measurable and reproducible effects on myocardial contractility (reduction of dP/dt of approximately 20%). In each animal, a placebo injection (2 mL 0.9% sodium chloride solution) into the circumflex artery was performed prior to ropivacaine injection. Both placebo and ropivacaine injections were performed over 6 s in a volume of 2 mL at 37°C. After each injection was completed, the haemodynamic indices (mean arterial pressure, left ventricular contractility and ST segments) were allowed to return to baseline before proceeding to the next injection. The three resuscitation drugs were administered in random order. Administration of the resuscitation drug was commenced immediately on observing the effects of intracoronary ropivacaine (within 20 s of the start of the ropivacaine injection).
Nicorandil was administered both by the central venous (one dose: 50 μg kg−1) and intracoronary route (three doses: 5, 30 and 100 μg kg−1); nitroglycerin and calcium chloride were administered via the central venous cannula (5 μg kg−1 and 1-8 mmol, respectively). Intracoronary injections of nicorandil were performed over 6 s in a volume of 2 mL at 37°C. Arterial blood gases were analysed from samples withdrawn from the femoral arterial cannula before the intracoronary injection of ropivacaine and after administration of the resuscitation drug.
Measured variables and derived indices were arterial pressure, HR, left ventricular peak pressure (LVP) and end-diastolic pressure (LVEDP), the first derivatives of LVP (positive and negative LVdP/dtmax), CO, QRS and QT intervals on the V5 ECG lead. These parameters were continuously displayed on an oscilloscope screen and recorded using the K2G Astro-Med Grass Polygraph System® (Astro-Med, Slough, UK). Acidbase status and oxygenation were measured throughout the experiment using arterial blood gas analysis (pH/blood gas analyser IL 1306®; Instrumentation Laboratory, Milan, Italy). This device performed an automatic two-point calibration prior to and at 2 hourly intervals throughout the experiment.
The changes produced by intracoronary ropivacaine were assessed and compared with those when ropivacaine administration was followed by each resuscitation drug. The data were analysed using paired t-tests for normally distributed data and Fisher's exact test for categorical data (P < 0.05 indicates statistical significance).
Six labrador dogs were studied: one male and five females (mean 21.1 kg, range 17-27 kg). Arterial pH ranged from 7.35 to 7.49. The doses of ropivacaine administered (based on the initial dose-finding study) ranged from 1 to 8 mg.
None of the placebo injections (saline) produced haemodynamic or ECG changes. Changes in left ventricular dP/dt and ST segment depression in the V5 ECG lead occurred 6 s after the start of the 6 s injection of ropivacaine, reached a maximum at 15 (±4.9) s; changes in dP/dtmax normalized at 198.4 (±139) s, ST segment changes at 205.8 (±93) s - times are expressed as mean (standard deviation).
The effects of four doses of nicorandil were compared with respect to the magnitude and duration of ropivacaine-induced changes in dP/dt and ST segments using t-test with Bonferroni's correction (Table 1). Intracoronary nicorandil (100 μg kg−1) decreased the magnitude of ST segment changes (P = 0.006) and the magnitude of ropivacaine-induced decrease in left ventricular dP/dtmax, but this did not reach statistical significance. The effects of i.v. nitroglycerin (5 μg kg−1), i.v. calcium chloride (1 mmol) and i.v. nicorandil (50 μg kg−1) on ropivacaine-induced cardiotoxicity were similarly assessed using t-test with post hoc Bonferroni's correction (Table 2). Calcium chloride reduced the magnitude of ropivacaine-induced decrease in left ventricular dP/dtmax (P = 0.006). Calcium chloride also decreased the duration of ropivacaine-induced decrease in left ventricular dP/dtmax and the magnitude of ST segment depression, but these did not reach statistical significance.
The minimum dose of calcium chloride (up to a maximum dose of 8 mmol), which produced immediate and complete reversal of ropivacaine-induced change in dP/dt and ST segment depression, was sought; reversal of dP/dt changes was achieved in 6/6 and of ST segment changes in 3/6 (Table 3). Calcium chloride was more effective than either nitroglycerin (Fisher's exact test, P < 0.005) or nicorandil 100 μg kg−1 (Fisher's exact test, P < 0.005) in reversing ropivacaine-induced changes in contractility. ST segment depression was reversed to a greater extent (3/6) by calcium chloride than by either nitroglycerin (0/6) or nicorandil 100 μg kg−1 (2/6), but these did not reach statistical significance (Fisher's exact test, P > 0.05 for all three comparisons).
The fatal dose was ascertained in each animal: these were 1.5 times the studied dose in one animal, eight times the initial dose in three animals and 16 times the initial dose in two animals.
The most important finding of this study is that calcium chloride consistently reduced the magnitude of ropivacaine-induced changes in left ventricular contractility to a greater extent than either nitroglycerin or nicorandil. Intracoronary nicorandil (100 μg kg−1) decreased the magnitude of ropivacaine-induced ST segment changes to a greater extent than either calcium chloride or nitroglycerin. However, the systemic administration of an equivalent dose would produce profound systemic vasodilation .
The negative inotropic effects due to intracoronary ropivacaine observed in this study are consistent with those reported by Reiz et al.. Differences in the observed effects on QRS duration may be due to the different sites of ropivacaine injection in the two studies (left anterior descending artery (Reiz's study) and circumflex artery in our study).
A number of studies have attempted to identify effective pharmacological agents to reverse long-acting amide local anaesthetic toxicity. Bupivacaine binds inactivated and open Na+ channels in cardiac conducting systems in a similar manner to neural tissue, producing a block to Na+ conductance, which is both use and voltage dependent . This block develops during the upstroke and plateau phases of the action potential and dissipates during the diastolic interval between beats. Recovery from this block is slow and substantial block accumulates at 50-160 beats min−1. Drugs, which increase HR , have been demonstrated to be ineffective or deleterious in the setting of bupivacaine-induced toxicity (slowing of intraventricular conduction facilitates dysrhythmias) [6,15]. Hyperpolarization will partially reverse this block; hypoxia or hyperkalemia, which induce depolarization of cardiac cell membranes, will enhance blockade . Amrinone, a bipyridine phosphodiesterase inhibitor (PDI), which increases intracellular calcium release , has been found to be superior to epinephrine in reversing bupivacaine-induced cardiotoxicity in anaesthetized dogs .
Repolarization is largely controlled by currents flowing through voltage-dependent K+ channels . Bupivacaine has been demonstrated to inhibit both the early- and late-rectifier K+ currents in different species [9-11]. Drugs or conditions, which shorten the action potential duration, will reduce this block to Na+ conductance (e.g. lidocaine or hyperkalemia, but the latter also produces depolarization ).
Activation of the ATP potassium current decreases the duration of action potentials, and hence, offers the potential to attenuate bupivacaine-induced prolongation of the action potential, which in turn enhances toxicity. Agents that open potassium channels have the opposite in vitro electrophysiological effects to bupivacaine because they shorten action potential duration and restore maximal upstroke velocity (Vmax) in cardiac cells . These drugs may reverse the electrophysiological effects of bupivacaine and have been shown to normalize altered QRS duration following bupivacaine administration . Nicorandil, a nicotinamide-nitrate ester, produces a dose-dependent increase in potassium conductance through the ATP-dependent inward-rectifier current in vascular and myocardial smooth muscle . The consequent hyperpolarization of the cell membrane inhibits calcium influx, and thus, reduces intracellular calcium concentration resulting in vasodilation .
Bupivacaine has been shown to depress late-peaking contractile responses, attributed to calcium release from the sarcoplasmic reticulum , and to depress the calcium-mediated inwards current of the slow action potential (slow inwards current) of the cardiac action potential at higher doses . Calcium chloride partially restored inotropic function in isolated atrial tissue following exposure to bupivacaine . However, the addition of calcium failed to reverse bupivacaine-induced negative inotropic and chronotropic effects in an isolated guinea pig preparation .
In a previous study , we observed consistent ST segment depression following ropivacaine administration. The improvement in ST segment changes following administration of calcium chloride in the current study and the lack of efficacy of nitroglycerin in this setting suggests that the mechanism underlying ropivacaine-induced ST changes is not ischaemia. As in our previous study , a reproducible animal model was used in which both electrophysiological and cardiodepressant effects of ropivacaine could be studied. Ropivacaine produced significant changes in positive dP/dt and in ST segments of the V5 lead of the ECG. We administered a dose of ropivacaine, which produced approximately 20% reduction in left ventricular dP/dt but did not alter conduction indices. The use of the intracoronary route allowed repeated administration of the drug, as doses were small and, therefore, unlikely to produce systemic effects or accumulate. Normalization of the indices occurred within minutes. Each dog served as its own control with respect to the individual dose chosen and with respect to the cardiotoxic changes observed following ropivacaine alone.
The doses selected (1-8 mg) were delivered directly to the area of the left ventricular myocardium supplied by the left circumflex artery. This does not precisely simulate the clinical setting, where large doses of local anaesthetic (e.g. ropivacaine 150 mg) could be inadvertently administered into the systemic circulation. The high concentrations of ropivacaine, which would result from inadvertent intravascular injection, would produce vasodilation . This effect and consequent compensatory cardiovascular reflexes are not reproduced by intracoronary injection. The wide range of doses employed probably resulted from the position of the tip of the cannula within the left circumflex coronary artery. If the cannula tip was placed proximal to the bifurcation of the aorta, the degree of dilution of injected agent was greater and a larger dose was required to produce measurable effects. Conversely, if the cannula tip was lodged more distally within the vessel, the degree of dilution was smaller and a smaller dose was required to produce a similar effect. Individual anatomical variation in the relationship of the great coronary vein to the circumflex artery is determined by the site at which the artery was cannulated: this accounted for the variety of cannula positions, and consequently, the range of doses required.
In conclusion, of the drugs studied, calcium chloride appears to have the greatest potential for treatment of ropivacaine-induced cardiotoxicity. Further investigation will be required before these results can be extrapolated to the clinical setting.
We express our gratitude to Rhone-Poulenc-Rorer for their financial assistance with this work.
1. Albright GA. Cardiac arrest following regional anaesthesia with etidocaine or bupivacaine. Anesthesiology
2. Reiz S, Haggmark G, Nath S. Cardiotoxicity of ropivacaine - a new amide local anaesthetic agent. Acta Anaesthesiol Scand
3. De Jong RH, Davis NL. Treating bupivacaine arrhythmias: preliminary report. Reg Anesth
4. Kotelko DM, Shnider SM, Dailey PA, et al.
Bupivacaine-induced cardiac arrhythmias in sheep. Anesthesiology
5. Moller R, Covino GG. Cardiac electrophysiological properties of bupivacaine and lidocaine compared to those of ropivacaine, a new amide local anesthetic. Anesthesiology
6. Clarkson CW, Hondeghem LM. Mechanism for bupivacaine depression of cardiac conduction: fast block of sodium channels during the action potential with slow recovery from block during diastole. Anesthesiology
7. De La Cousaye JE, Masse C, Bassoul BP, Eledjam JJ, Gagnol JP, Sasine A. Bupivacaine-induced slow-inward current inhibition: a voltage clamp study on frog atrial fibres. Can J Anaesth
8. Porter JM, Markos F, Snow HM, Shorten GD. Effects of respiratory and metabolic pH changes and hypoxia on ropivacaine-induced cardiotoxicity in dogs
. Br J Anaesth
9. Castle NA. Bupivacaine inhibits the transient outward K+
current but not the inward rectifier in rat myocytes. J Pharmacol Exper Therap
10. Courtney KR, Kendig JJ. Bupivacaine is an effective potassium channel blocker in the heart. Biochim Biophys Acta
11. Sanchez-Chapula J. Effects of bupivacaine on membrane currents of guinea-pig ventricular myocytes. Eur J Pharmacol
12. Eledjam JJ, de La Coussaye JE, Brugada J, et al. In vitro
study on mechanisms of bupivacaine-induced depression of myocardial contractility
. Anesth Analg
13. Gherardini G, Samuelson U, Jernbeck J, Aberg, Sjostrand N. Comparison of vascular effects of ropivacaine and lidocaine on isolated rings of human arteries. Acta Anaesthesiol Scand
14. Saitoh K, Hirabayashi Y, Shimizu R, Fukuda H. Amrinone is superior to epinephrine in reversing bupivacaine-induced cardiovascular depression in sevoflurane-anesthetized dogs
15. De La Coussaye JE, Eledjam JJ, Peray P, et al.
Mechanisms of the putative cardioprotective effect of hexamethonium in anesthetized dogs
given a large dose of bupivacaine. Anesthesiology
16. Kasten GW, Martin ST. Successful cardiovascular resuscitation after massive intravenous bupivacaine overdosage in anesthetized dogs
. Anesth Analg
17. De La Coussaye, Eledjam JJ, Peray P, et al.
Lemakalim, a potassium channel agonist, reverses electrophysiological impairments induced by a large dose of bupivacaine in anaesthetized dogs
. Br J Anaesth
18. Purcell H, Mulcahy D, Fox K. The clinical profile of nicorandil. Rev Contemp Pharmacother
19. Surtapranata H, MacLeod D. Nicorandil and cardiovascular performance in patients with coronary artery disease. J Cardiovasc Pharmacol
20. Tanz RD, Heskett T, Loehning RW, Fairfax CA. Comparative cardiotoxicity of bupivacaine and lidocaine in the isolated perfused mammalian heart. Anesth Analg