| ||||||||||||||
|
|
|||||||||||||
Department of Anesthesiology, Wake Forest University School of Medicine, Winston-Salem, North Carolina
Address correspondence and reprint requests to Leanne Groban, MD, Department of Anesthesiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157-1009. Address e-mail to lgroban{at}wfubmc.edu
| Abstract |
|---|
|
|
|---|
45 mm Hg). Hypotension and arrhythmias were treated with epinephrine, open-chest massage, and advanced cardiac life support protocols, respectively. Outcomes were defined as the following: successful (stable rhythm and MAP
55 mm Hg for 20 min), successful with continued therapy (stable rhythm and MAP <55 mm Hg after 20 min), or death. Continued therapy was required in 86% of LIDO dogs compared with only 10%30% of the other dogs (P < 0.002). Mortality from BUP, LBUP, ROP, and LIDO was 50%, 30%, 10%, and 0%, respectively. Myocardial depression was primarily responsible for the profound hypotension, as the occurrence of lethal arrhythmias preceding resuscitation was not different among local anesthetics. Epinephrine-induced ventricular fibrillation occurred more frequently in BUP-intoxicated dogs than in dogs given LIDO or ROP (P < 0.05). The unbound plasma concentrations at collapse were larger for ROP, 19.8 µg/mL (1039 µg/mL), compared with BUP, 5.7 µg/mL (311 µg/mL); whereas the concentrations of LBUP, 9.4 µg/mL (518 µg/mL) and BUP were not significantly different from each other.
Implications: There were consistent differences among the local anesthetics, the sum of which suggests that larger doses and blood concentrations of ropivacaine (ROP) and lidocaine will be tolerated as compared with bupivacaine (BUP) and levobupivacaine (LBUP). Lidocaine intoxication results in myocardial depression from which resuscitation is consistently successful but will require continuing drug support. After BUP, LBUP, or ROP, resuscitation is not always successful, and the administration of epinephrine may lead to severe arrhythmias. The unbound plasma concentrations at collapse were larger for ROP compared with BUP, whereas the concentrations of LBUP and BUP were not significantly different from each other. Furthermore, larger plasma concentrations of ROP than BUP are present after resuscitation, suggesting a wider margin of safety when large volumes and large concentrations are used to establish upper or lower extremity nerve blocks for surgical anesthesia and during long-term infusions for pain management.
| Introduction |
|---|
|
|
|---|
LBUP and ROP are S(-) enantiomers of two different molecules, l-butyl-2',6'-pipecoloxylidide and l-propyl-2',6'-pipecoloxylidide, respectively. Only LBUP is the S(-) enantiomer of BUP. Both molecules possess physicochemical and LA properties similar to racemic BUP. These single enantiomeric preparations were developed based on the observation that S(-) isomers possess equal anesthetic potency to the R(+) isomers, but appear less cardiotoxic than the R(+) isomers (5). Animal studies demonstrate that ROP depresses cardiac electrophysiology and function less than BUP (6,7). LBUP produces fewer and less deleterious arrhythmias than equivalent concentrations of BUP (8). The lower affinity of the S(-) isomer for the inactivated state of the cardiac sodium channel (9) and for the cardiac potassium channel (10) than the R(+) isomer may explain some of these stereoselective differences in cardiotoxicity.
Clinical and laboratory accounts of acute intoxication from intravascular injections of ROP and, more recently, one LBUP report, suggest favorable outcomes (1115). There is scant literature about cardiovascular toxicity that might occur from systemic absorption and drug accumulation during long-term epidural or regional infusions. Moreover, the clinically relevant questions as to whether the newer LAs provide a greater margin of safety, or are less frequently associated with unsuccessful resuscitation than BUP, have not been answered.
We compared the success of resuscitation after incremental infusions of ROP, LBUP, BUP, and lidocaine (LIDO) to the point of cardiovascular collapse in dogs. The LAs were infused at rates chosen to simulate the plasma concentrations that might be achieved from cumulative effects of systemic absorption during various regional anesthetic techniques (16,17) and the levels that might be expected from an accidental intravascular injection (11,14). To eliminate the confounding effects of seizures and their metabolic consequences, including hypoxia, hypercarbia, and acidosis, the animals were anesthetized and ventilated for the study. Comparisons of the arrhythmogenic effects of these drugs have been shown in a companion report (18).
| Methods |
|---|
|
|
|---|
Forty heartworm-free, nonpregnant, purebred hounds (Harlan Sprague Dawley, Inc., Indianapolis, IN), weighing between 18 and 21 kg, were used. Anesthesia was induced with 20 mg/kg of IV thiopental and maintained with a continuous infusion of fentanyl (0.3 µg · kg-1 · min-1) and midazolam (0.005 µg · kg-1 · min-1). The dogs were tracheally intubated and ventilated to normocapnic values before starting the experiment with oxygen-enriched room air using a volume-cycled ventilator. Both femoral arteries and veins were cannulated for arterial pressure monitoring and sampling, and for drug and fluid infusion, respectively. An 8.5F introducer catheter was inserted into the left external jugular vein for the administration of resuscitation drugs. Body temperature was measured rectally and maintained at 37° to 38°C during the experiment by using a heating blanket. A continuous infusion of 0.9% NaCl at 3 mL · kg-1 · h-1 was given throughout the experiment. The urinary bladder was catheterized. A left thoracotomy and pericardotomy were performed for placement of left-ventricular pacing wires and to allow open-chest heart massage and direct cardiac countershock. Dogs were assigned randomly to receive incremental infusions of vehicle (saline; n = 3) or one of four LAs until cardiovascular collapse: BUP (n = 10), LBUP (n = 10), ROP (n = 10), LIDO (n = 7). The principal investigator, who gathered the data, remained blinded to the LA until all experiments had been completed. The dosing regimens for BUP, LBUP, and ROP were based on drug clearance rates in dogs provided by AstraZeneca Research and Development (Södertälje, Sweden) and the target plasma concentrations were 2, 4, 8, 16, and 19 mg/L. A clearance rate of 0.9 L · h-1 · kg-1 was assumed for BUP and LBUP and a clearance rate of 1.16 L · h-1 · kg-1 was assumed for ROP. The clearance rate assumed for LIDO was 1.71 L · h-1 · kg-1, and the target plasma concentrations to be achieved were 8, 16, 32, and 64 mg/L, assuming a BUP-to-LIDO potency ratio of 4:1. Drug infusions consisted of a 12-min initial infusion (defined as 3 times the calculated steady-state rate) followed by a 12-min maintenance infusion (defined as: clearance rate (L · h-1 · kg-1) x desired target concentration [mg/L] = mg · h-1 · kg-1). From previous pharmacokinetic data (AstraZeneca R&D, Södertälje, Sweden), we assumed that 88% of the desired plasma concentration would be achieved after the initial 12-min loading infusion, and that 92% of the desired target concentration would be achieved after the steady-state infusion. The infusion protocol for vehicle was set at a rate that corresponded to the rate used for the BUP-treated dogs.
A surface lead II electrocardiogram and femoral arterial blood pressure were continuously monitored on an oscilloscope and recorded on a polygraph interfaced with a computer for data retrieval. Two insulated silver-coated wires were inserted 1 cm apart into the epicardial surface of the left ventricle to allow for pacing. Standard, programmable electrical stimulation (PES) protocols were also used, as described previously (18), to determine the arrhythmogenic potential of each LA.
Arterial blood samples were collected in heparinized syringes for blood gas analysis at baseline and at the end of each maintenance infusion until cardiovascular collapse. Minute ventilation was adjusted or a sodium bicarbonate solution (8.4%) was administered to maintain arterial blood chemistry within the following ranges: pHa 7.35 to 7.45; PaCO2 30 to 40 mm Hg; and PaO2 150 to 300 mm Hg. Ten milliliters of arterial blood was collected for determination of free and total LA concentrations at baseline, cardiovascular collapse, and after 20 min of successful resuscitative maneuvers. The samples were stored on ice until they were centrifuged to obtain plasma. Plasma samples were stored at -20°C pending batch gas chromatographic analysis by AstraZenecas reference laboratory.
Cardiovascular collapse was defined as asystole, sustained ventricular tachycardia (spontaneous or PES-induced), ventricular fibrillation, or a mean blood pressure
45 mm Hg persisting for 15 consecutive s. Resuscitative procedures consisted of the following:
55 mm Hg was achieved. If hypotension was accompanied with bradycardia (
50 beats/min, ventricular pacing at a rate of 120130 bpm was initiated.
55 mm Hg).
Outcomes after 20 min of resuscitation were defined as: successful (stable rhythm and MAP
55 after 20 min without need for continued drug therapy), successful with continued need for pharmacologic support (stable rhythm and MAP
55 after 20 min unless additional EPI was given), or death. The animals were euthanized with IV KCl, 100 mEq, after 20 min had elapsed.
BUP HCl and ROP HCl were commercially available 0.5% preservative-free solutions and LIDO HCl was a commercially available 2.0% preservative-free solution provided by AstraZeneca (Philadelphia, PA). LBUP was provided by AstraZeneca (Södertälje, Sweden) in dry form. Before its use, it was reconstituted in normal saline with pH adjustment to provide a 0.5% solution. EPI (American Regent, Inc., Shirley, NY) and atropine (American Regent Inc.) solutions were prepared from commercially available multiple-dose vials. Sodium bicarbonate and bretylium (Abbott, North Chicago, IL) were from single-dose commercially available ampules.
Data were analyzed by using SAS Version 6.1 (SAS Institute, Cary, NC). Analysis of variance was used to analyze hemodynamic measurements and plasma concentration data. Log transformation was used when this maneuver improved the "normality" of distributions. Post hoc tests were corrected for multiple comparisons when the overall effects or interactions were not significant (Fisher Protected LSD). Exact
2 was used to test for differences among groups with respect to the ability to resuscitate. Values of P
0.05 were considered significant.
| Results |
|---|
|
|
|---|
|
|
|
|
The median cumulative doses (mg/kg) and median LA plasma concentrations required to produce cardiac collapse are shown in Table 2. Pair-wise contrasts comparing cardiac collapse concentration ratios and their 95% confidence limits found LIDO to differ significantly from the other three LAs (P < 0.005). The unbound plasma concentrations at collapse for ROP-treated dogs were significantly larger than for BUP-treated animals (P < 0.01). In all groups, plasma concentrations of LA 20 min after a successful resuscitation were dramatically decreased from the concentrations that resulted in collapse.
|
| Discussion |
|---|
|
|
|---|
The cause of cardiovascular collapse for both the long-acting (e.g., BUP) and the short-acting (e.g., LIDO) amide LAs was typically hypotension from myocardial depression rather than from arrhythmias (18). Although others have reported an increased incidence of arrhythmias with BUP than with LIDO intoxication (6,20), these studies were conducted in conscious, spontaneously ventilating animals in which seizures preceded the cardiac effects, resulting in a hyperdynamic circulation preceding circulatory collapse (4). Such confounding factors associated with seizure activity, including hypoxia, hypercarbia, acidemia, and hyperkalemia (21,22), were eliminated by use of our normoventilated, anesthetized canine model.
Our findings are consistent with the work of Liu et al. (2) in anesthetized dogs, in which "pump failure" was the cause of death in both BUP- and LIDO-treated dogs. In fact, the hypotension that resulted was often more severe with LIDO than BUP overdosage. In our study, 86% of LIDO-treated dogs required continued EPI dosing during resuscitation because of continuing hypotension whereas continued EPI dosing was required in only 10%30% of the dogs given BUP, LBUP, and ROP. One explanation for this increased need for vasoactive support after LIDO may be that the absence of lethal arrhythmias (18) permits a relatively larger amount of LIDO (compared with the other drugs) to be infused, approaching that required for contractile failure. Interestingly, concentrations of LIDO necessary to produce depression of cardiac contractility have been 10 times larger than BUP in in vitro studies (23,24). Nevertheless, similar to the findings of Chadwick (25), all of the LIDO animals survived whereas only half of the BUP-intoxicated animals survived, despite the increased need for continued vasoactive support after LIDO.
The S(-) ROP enantiomer appears to be potentially less cardiotoxic than either its S(-) butyl homologue LBUP or racemic BUP. Unsuccessful resuscitation, or death, occurred in only 10% of the ROP-treated dogs as compared with 50% and 30% of BUP- and LBUP-treated dogs, respectively. A power analysis using the observed mortality rates indicates that 30 dogs per group would have been required to detect significant differences between BUP and ROP, and 140 dogs per group would have been required to detect differences between BUP and LBUP (power >0.8,
= 0.05). ROP was also associated with fewer deaths and a decreased incidence of electrocardiographic QRS prolongation, ventricular arrhythmias, and ventricular tachycardia than LBUP or BUP in conscious rats (26). Consistent with our findings, Huang et al. (8) reported a decreased cardiotoxicity with IV LBUP than with BUP in conscious sheep. However, Kasten and Martin (27) found that sheep were more sensitive than dogs to cardiotoxic doses of BUP. Nevertheless, our study is the first to compare the systemic toxicity and the success of resuscitation of all three long-acting amide LAs after incremental overdosage to the point of cardiovascular collapse in the intact animal.
The pharmacokinetic profiles of the BUP enantiomers and ROP appear to be different. The unbound plasma concentrations at the time of collapse were larger for ROP compared with BUP, whereas the cumulative dose and unbound plasma concentrations of LBUP and BUP were similar. In conscious sheep, however, the differences between the unbound drug concentrations of ROP 7.44 ± 1.64 µg/mL, compared with BUP 3.44 ± 1.78 µg/mL, were similar to the differences observed between LBUP 6.82 ± 2.45 µg/mL and BUP after a constant rate IV infusion until circulatory collapse (28). Similarly, Morrison et al. (29) recently reported that LBUP was less toxic than BUP, but not different from ROP in anesthetized swine receiving intracoronary bolus injections of LA until death. In contrast, Feldman et al. (13) found no significant difference in peak plasma concentrations between ROP (5.75 ± 0.23 µg/mL) and BUP (5.04 ± 0.13 µg/mL) after rapid IV injection to the point of collapse in conscious dogs. Inconsistencies in these results may be attributable to species differences (27) and/or differences in racemate kinetics after prolonged versus short (bolus) infusions (30).
We observed an increased incidence of EPI-induced ventricular fibrillation in the BUP-intoxicated dogs than in dogs given LIDO or ROP. EPI was chosen for treatment of hypotension from LA-induced myocardial depression because of its inotropic, chronotropic, and vasopressor properties, and because it is routinely used for cardiac arrest (19). Large doses of EPI (
30 µg/kg) are effective for LA toxicity in dogs (31), cats (25), and rats1. The use of EPI for LA cardiotoxicity is not, however, without drawbacks. In anesthetized rats, Heavner et al. (32) observed an increased incidence of arrhythmias with EPI and isoproterenol compared with amrinone, dopamine, or norepinephrine for treatment of BUP-induced asystolic arrest. Similarly, EPI-induced ventricular arrhythmias have been reported in dogs (13,33), sheep (22), and pigs (34) treated for BUP-induced circulatory collapse. One explanation for EPI-induced ventricular arrhythmias in the setting of BUP toxicity is the rapid increase in ventricular pressure that occurs, increasing myocardial oxygen demand and predisposing the heart to ventricular ectopy. This mechanism is unlikely, however, in our study because both the LIDO- and ROP-treated animals had rapid, hyperdynamic responses after EPI therapy that were similar to the BUP-intoxicated animals. Alternatively, there could be a differential modulatory effect of ß stimulation on the slow inward calcium currents in the heart. BUP exhibits a potency about 5 times greater than LIDO in blocking slow calcium channels (35). EPI increases the frequency and duration of calcium channel opening (36), which can potentiate BUP-induced oscillations in the membrane potential (37), possibly triggering extrasystoles and uncovering reentrant pathways. Finally, LIDO and ROP may inhibit reentrant arrhythmias from EPI. Whether LA stereoselectivity or structure plays a role in this effect is unclear, because EPI-induced ventricular arrhythmias also occurred in dogs intoxicated with LBUP.
In summary, we found that the pattern of LA cardiovascular toxicity and the LA concentration that produces cardiovascular toxicity differs among LAs. There were consistent differences among the LAs, the sum of which suggests that larger doses and blood concentrations of ROP and LIDO will be tolerated as compared with BUP and LBUP. LIDO intoxication results in myocardial depression from which resuscitation is consistently successful but will require continuing drug support. After BUP, LBUP, or ROP, resuscitation is not always successful, and the administration of EPI may lead to severe ventricular arrhythmias. Nevertheless, continuing resuscitative therapy will not usually be required in our model of gradual IV overdosage of these longer-acting amide LAs. Furthermore, larger plasma concentrations of ROP than BUP are present after resuscitation, suggesting a greater margin of safety when larger volumes and concentrations are used to establish upper and lower extremity nerve blocks for surgical anesthesia and during long-term infusions for pain management.
| Acknowledgments |
|---|
| Footnotes |
|---|
1 Wu SJ, Bircher NG, Safar P. Epinephrine reverses bupivacaine-induced cardiac arrest in rats [abstract]. Anesthesiology 1990;73:A303. ![]()
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. E. Copeland, L. A. Ladd, X.-Q. Gu, and L. E. Mather The Effects of General Anesthesia on the Central Nervous and Cardiovascular System Toxicity of Local Anesthetics Anesth. Analg., May 1, 2008; 106(5): 1429 - 1439. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Heid, N. Muller, T. Piepho, M. Bares, M. Giesa, P. Drees, A. Rumelin, and C. Werner Postoperative Analgesic Efficacy of Peripheral Levobupivacaine and Ropivacaine: A Prospective, Randomized Double-Blind Trial in Patients After Total Knee Arthroplasty Anesth. Analg., May 1, 2008; 106(5): 1559 - 1561. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. D. Mayr, L. Mitterschiffthaler, A. Neurauter, C. Gritsch, V. Wenzel, T. Muller, G. Luckner, K. H. Lindner, and H.-U. Strohmenger A Comparison of the Combination of Epinephrine and Vasopressin with Lipid Emulsion in a Porcine Model of Asphyxial Cardiac Arrest After Intravenous Injection of Bupivacaine Anesth. Analg., May 1, 2008; 106(5): 1566 - 1571. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Warren, R. B. Thoma, A. Georgescu, and S. J. Shah Intravenous Lipid Infusion in the Successful Resuscitation of Local Anesthetic-Induced Cardiovascular Collapse After Supraclavicular Brachial Plexus Block Anesth. Analg., May 1, 2008; 106(5): 1578 - 1580. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. N. Stehr, T. Christ, B. Rasche, S. Rasche, E. Wettwer, A. Deussen, U. Ravens, T. Koch, and M. Hubler The Effects of Levosimendan on Myocardial Function in Ropivacaine Toxicity in Isolated Guinea Pig Heart Preparations Anesth. Analg., September 1, 2007; 105(3): 641 - 647. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Ganesh and G. Cucchiaro Multiple simultaneous perineural infusions for postoperative analgesia in adolescents in an outpatient setting Br. J. Anaesth., May 1, 2007; 98(5): 687 - 689. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Corcoran, J. Butterworth, R. S. Weller, J. C. Beck, J. C. Gerancher, T. T. Houle, and L. Groban Local Anesthetic-Induced Cardiac Toxicity: A Survey of Contemporary Practice Strategies Among Academic Anesthesiology Departments Anesth. Analg., November 1, 2006; 103(5): 1322 - 1326. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. G. B. Chedid, R. T. Sudo, M. I. S. Aguiar, M. M. Trachez, M. O. Masuda, and G. Zapata-Sudo Regulation of intracellular calcium by bupivacaine isomers in cardiac myocytes from wistar rats. Anesth. Analg., March 1, 2006; 102(3): 792 - 798. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. F. Royse and A. G. Royse The Myocardial and Vascular Effects of Bupivacaine, Levobupivacaine, and Ropivacaine Using Pressure Volume Loops Anesth. Analg., September 1, 2005; 101(3): 679 - 687. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Gielen, R. Slappendel, and N. Jack Successful defibrillation immediately after the intravascular injection of ropivacaine: [Defibrillation reussie immediatement apres l'injection intravasculaire de ropivacaine] Can J Anesth, May 1, 2005; 52(5): 490 - 492. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-T. Kim, C.-W. Jung, and K.-H. Lee The Effect of Insulin on the Resuscitation of Bupivacaine-Induced Severe Cardiovascular Toxicity in Dogs Anesth. Analg., September 1, 2004; 99(3): 728 - 733. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Weinberg, C. Paisanthasan, D. Feinstein, and W. Hoffman The Effect of Bupivacaine on Myocardial Tissue Hypoxia and Acidosis During Ventricular Fibrillation Anesth. Analg., March 1, 2004; 98(3): 790 - 795. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Klein, T. Pierce, Y. Rubin, K. C. Nielsen, and S. M. Steele Successful Resuscitation After Ropivacaine-Induced Ventricular Fibrillation Anesth. Analg., September 1, 2003; 97(3): 901 - 903. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. S. Liu and F. V. Salinas Continuous Plexus and Peripheral Nerve Blocks for Postoperative Analgesia Anesth. Analg., January 1, 2003; 96(1): 263 - 272. [Full Text] [PDF] |
||||
![]() |
M. Senard, J. L. Joris, D. Ledoux, P. J. Toussaint, B. Lahaye-Goffart, and M. L. Lamy A Comparison of 0.1% and 0.2% Ropivacaine and Bupivacaine Combined with Morphine for Postoperative Patient-Controlled Epidural Analgesia After Major Abdominal Surgery Anesth. Analg., August 1, 2002; 95(2): 444 - 449. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Maurer, G. Ekatodramis, K. Rentsch, and A. Borgeat Interscalene and Infraclavicular Block for Bilateral Distal Radius Fracture Anesth. Analg., February 1, 2002; 94(2): 450 - 452. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Simon, N. Kariya, E. Pelle-Lancien, and J.-X. Mazoit Bupivacaine-Induced QRS Prolongation is Enhanced by Lidocaine and by Phenytoin in Rabbit Hearts Anesth. Analg., January 1, 2002; 94(1): 203 - 207. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Ohmura, M. Kawada, T. Ohta, K. Yamamoto, and T. Kobayashi Systemic Toxicity and Resuscitation in Bupivacaine-, Levobupivacaine-, or Ropivacaine-Infused Rats Anesth. Analg., September 1, 2001; 93(3): 743 - 748. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||