| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1997;63:721-727
© 1997 The Society of Thoracic Surgeons
Institute for Biodiagnostics, National Research Council of Canada, Winnipeg, Canada; Service de Chirurgie Cardiovasculaire, Hôpital Européen de Paris "La Roseraie," Paris, France; and Division of Cardiovascular and Thoracic Surgery, State University of New York at Buffalo, Buffalo, New York
Accepted for publication October 11, 1996.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods and Results. In the first protocol, 8 perfused rat hearts were subjected to 20, 60, 90, and 120 minutes of esmolol arrest, after which positive and negative first derivative of pressure, heart rate, left ventricular developed pressure, and left ventricular end-diastolic pressure were evaluated. Arrest was achieved 45 to 60 seconds after beginning the infusion of esmolol. Mechanical arrest was achieved before electrical arrest. In the second protocol, dose-response curves were obtained using isolated (Langendorff) rat and rabbit (n = 6) hearts. The concentrations of esmolol varied from 0.084 to 6.7 mmol/L and from 0.12 to 1.45 mmol/L in the rat and rabbit heart experiments, respectively. In the third protocol, the effects of 20 minutes of normothermic (37°C) ischemia on the function of isolated rat hearts perfused with esmolol-containing Krebs solution were compared with those using high-potassium (25 mmol/L) Krebs solution. Group A subjects (n = 9) received the ischemic injury after being perfused (and arrested) for 20 minutes with either esmolol or potassium (KCl, 25 mmol/L). Group B subjects (n = 10) received the same ischemic insult before being perfused with either esmolol or potassium. Esmolol-treated hearts showed better recovery than those receiving potassium, in terms of ±dP/dt (p < 0.01), left ventricular systolic pressure (p < 0.01), and left ventricular developed pressure (p < 0.009). Finally, the fourth protocol was done to evaluate the effects of esmolol in a clinically relevant experimental model. Pigs were divided into esmolol (n = 6) and potassium (n = 5) groups and subjected to normothermic cardiopulmonary bypass and a 1-hour period of cardiac arrest. Twenty minutes after stopping infusion of the cardioplegic agents, all animals were weaned off bypass. There were no statistically significant differences between the groups.
Conclusions. Esmolol hydrochloride can be used as effectively as potassium for inducing and maintaining predictable and reversible cardiac arrest during normothermic cardiac operations.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
Material and Methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Experimental Protocols
All animals received humane care in compliance with the Canadian Council on Animal Care regulations, under supervision of the Institute for Biodiagnostics Animal Care Committee. The animals were acclimatized for 14 days before experimental use, maintained on a normal laboratory diet, and given access to water ad libitum. Four different experimental protocols were performed.
In the first protocol, 8 Sprague-Dawley rats were subjected to different periods of arrest (20, 60, 90, and 120 minutes), after which functional indices were measured (±dP/dt, heart rate, left ventricular systolic pressure [LVSP], left ventricular developed pressure [LVDP], and left ventricular end-diastolic pressure [LVEDP]). Results were compared with the initial values obtained before the induction of arrest.
In the second protocol, dose-response curves were obtained in isolated Langendorff rat and rabbit hearts (n = 6). The final concentrations of esmolol varied from 8.5 x 10-5 to 6.8 x 10-3 mol/L and 1.2 x 10-4 to 1.45 x 10-3 mol/L in the rat and rabbit heart experiments, respectively.
In the third protocol, the effects of ischemia on isolated rat hearts perfused with normothermic (37°C) esmolol-containing Krebs-Henseleit (KH) solution [7] were compared with those using high-potassium (KCl, 25 mmol/L) KH solution. Nineteen rats were divided into two groups and two subgroups. Group A (n = 9) was perfused (and arrested) for 20 minutes with either esmolol or potassium (groups AE and AP in Table 1
) followed by no-flow ischemia for 20 minutes at 37°C. Group B (n = 10) received the same ischemic insult before being perfused with either esmolol or potassium (groups BE and BP).
|
Instrumentation
ISOLATED HEARTS.
The animals were anesthetized with isoflurane at a concentration of 3% to 5% in oxygen (100%) at a rate of 2 L/min and were maintained with 1.5% to 2% at an oxygen flow rate of 1 L/min. After achieving an operative plane of anesthesia, a median thoracotomy was performed and the heart was removed.
The isolated rat and rabbit heart experiments were performed in a Langendorff perfusion apparatus using oxygenated (95% O2, 5% CO2) KH solution under normothermic (37°C) conditions and a constant pressure of 90 mm Hg. Mechanical function was measured using a water-filled compliant balloon inserted into the left ventricle and attached to a pressure transducer.
Perfusion of high-potassium KH solution (25 mmol/L) was provided by a separate reservoir. Esmolol (Brevibloc; Zeneca Pharma Inc, Mississauga, Ont, Canada) was provided in vials containing 10 mL of a 250-mg/mL solution (2.5 g/10 mL). One vial was diluted in 990 mL of oxygenated KH solution, yielding a concentration of 2.5 mg/mL. The solution was infused using an electronic syringe pump attached to a three-way stopcock inserted immediately above the heart, and was mixed with the bulk perfusate. The final concentration of esmolol reaching the heart was determined by the ratio of the rate of drug infusion and the perfusion rate.
For the initial and the ischemic protocols (protocols 1 and 3), the syringe pump infusion rate was set at 5 mL/min (2.5 mg/mL x 5 = 12.5 mg/min) until complete arrest was achieved. At this time, the infusion rate was titrated to the minimum dose necessary to maintain arrest, usually about 1.0 to 1.5 mL/min (2.5 to 3.75 mg/min). The measured coronary flow for a 1-g rat heart was 20 mL/min while beating and approximately 18 mL/min during arrest. For calculation of cardioplegic concentrations of esmolol in the rat heart, we used an average perfusion rate of 22.2 mL/min. Thus, in protocols 1 and 3, the esmolol concentration used for inducing arrest was calculated as 1.7 x 10-3 mol/L, and for maintenance the value was 5.0 x 10-4 mol/L.
LARGE ANIMAL MODEL (PIG) UNDER EXTRACORPOREAL CIRCULATION.
Eleven Yorkshire pigs of either sex, weighing 35 to 50 kg, were subjected to extracorporeal circulation under normothermic conditions (37°C), during which positive and negative first derivative of pressure (±dP/dt), heart rate, LVSP, LVEDP, LVDP, and electrocardiographic (ECG) signal amplitude were monitored.
After overnight fasting, the pigs were preanesthetized with intramuscular ketamine (20 mg/kg), midazolam (0.3 mg/kg), and atropine (0.015 mg/kg). The animals were intubated and ventilated with a mixture of nitrogen (58%), oxygen (40%), and isoflurane (2%) at a flow of 2 to 4 L/min. To maintain muscle paralysis, pancuronium (0.15 mg/kg) was given every hour during the procedure. During extracorporeal circulation, isoflurane was administered (1.0% to 2.0%) directly into the oxygenator. An arterial pressure catheter was placed in the carotid artery. A central venous pressure line was inserted through the jugular vein. A thermocouple was placed in the esophagus for monitoring temperature, which was kept at 37°C by active rewarming with a heat exchanger. The arterial and central venous pressures were monitored continuously, as were the blood gases and electrolytes, which were kept within the physiologic range.
The perfusion system consisted of a membrane oxygenator (COBE Inc, Scarborough, Ontario, Canada), a 40-µm arterial blood filter (Dideco, Mirandola, Italy), and a centrifugal pump (St. Jude Medical Inc, Chelmsford, MA) placed in the arterial line (in the first two experiments, a roller-pump [COBE] was used for one experiment in each group). Cardioplegia was delivered in an antegrade fashion using a roller-pump connected to a cannula (DLP, Grand Rapids, MI) inserted in the ascending aorta. Esmolol was injected in this cardioplegia line by means of an electronic syringe pump through a side connector and a three-way stopcock.
A median sternotomy was performed and heparin (500 IU kg) was given systemically. The activated clotting time was monitored periodically and maintained above 600 seconds. Two cannulas were inserted in the ascending aorta: one for arterial input from the pump and a second for cardioplegia delivery. Venous return was ensured using a two-stage atriocaval cannula (DLP) inserted through the right appendage. Before initiating bypass, a 10-minute period was allowed for acquisition of initial functional data. Normothermic cardiopulmonary bypass was initiated at a flow rate of 60 to 100 mL/kg, and another 10-minute period was allowed for data acquisition. Antegrade cardioplegia was delivered continuously and immediately after aortic cross-clamping at a flow rate of 150 to 250 mL/min.
The animals were divided into esmolol and potassium groups. The potassium group (n = 5) received KCl and magnesium sulfate (MgSO4 at 10%) infused according to nomograms developed by LeHouerou and colleagues [8]. The esmolol group (n = 6) received esmolol as the cardioplegic agent at a concentration in the syringe pump of 10 mg/mL.
Potassium and Esmolol Cardioplegia Concentrations
The potassium concentration in the cardioplegia line was 25 mEq/L for induction and 12 mEq/L for maintenance. Undiluted continuous normothermic blood cardioplegia was administered as described previously [8].
The esmolol concentrations in the cardioplegic line were as follows: induction: syringe-pump rate 15 mL/min (150 mg/min), cardioplegia roller-pump flow 150 mL/min = 0.91 mg·mL-1·min-1; approximately 2.7 x 10-3 mol/L; and maintenance: syringe-pump rate 5 mL/min (50 mg/min), cardioplegia roller-pump flow 150 mL/min = 0.31 mg·mL-1·min-1; approximately 9.0 x 10-4 mol/L.
After 1 hour, the aortic cross-clamp was released, cardioplegic infusion was stopped, and the heart was allowed to beat spontaneously while maintaining total cardiopulmonary bypass. Functional data (±dP/dt, LVSP, LVEDP, LVDP, and heart rate) were acquired during the recovery period. This period was composed of two phases: an initial phase of recovery when the heart was beating empty and the arterial pump maintained the systemic flow, and a second in which the animal was weaned off bypass and the heart resumed the hemodynamic load and maintenance of arterial pressure. Each phase lasted 20 minutes, and no inotropic support was provided. After acquisition of the last data set, the animal was euthanized by circulatory arrest and exsanguination under anesthesia.
Statistical Analysis
Student's t test was used in all analyses with dependent or independent variables when applicable, supported by the Mann-Whitney U test for the independent variables and the Wilcoxon matched-pairs test for the dependent variables. In addition, because of the limited number of animals, the exact permutation test [9] was used in both types of analyses. Differences are expressed as p values.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The LVEDP increased during arrest. The increase in LVEDP was reversible, declining to prearrest values upon cessation of drug infusion. The data indicate that esmolol induces and maintains reversible heart arrest within clinically relevant time frames.
Dose-Response Curves Using Isolated Rat and Rabbit Hearts
Mechanical indices such as ±dP/dt and LVDP reached 0 at a 4-mL/min dose (0.45 mg·mL-1·min-1; approximately 1.36 x 10-3 mol/L; Figs 1 to 3). At 5 mL/min (0.56 mg·mL-1·min-1; about 1.7 x 10-3 mol/L), all indices reached zero except for LVEDP (see Figs 1, 2). At a dose of 4.0 mL/min, the end of the infusion period (fifth minute) was marked by mechanical arrest in the presence of electrical activity (ECG signal). In the experiments using isolated rabbit hearts, mechanical arrest was achieved at 0.04 mg·mL-1·min-1 (approximately 4.2 x 10-4 mol/L), whereas cessation of electrical activity was obtained at 0.2 mg·mL-1·min-1 (approximately 6.0 x 10-4 mol/L).
|
|
|
). During arrest (concentrations >0.42 mmol/L), LVSP was equal to LVEDP measured by the intraventricular balloon. The LVEDP returned to prearrest values after 20 minutes of "reperfusion" in the absence of drug. In both curves, each heart was used to assess the effects of increasing drug concentrations. A cumulative effect on the LVEDP was observed such that the "resting" pressure gradually increased throughout the experiment.
All other indices showed close to 100% recovery 20 minutes after stopping the esmolol infusion (Figs 4, 5). Even at higher doses, the recovery levels were greater than 80% at 20 minutes after "reperfusion," except that 35 minutes of "reperfusion" was required for recovery after the 20-mL/min dose. The duration of the recovery period was directly proportional to the amount of drug infused.
|
|
).
Cardiac Arrest in the Pig In Vivo
Pigs were subjected to esmolol arrest for 1 hour, and hemodynamic functions were evaluated after interruption of cardiopulmonary bypass. The flow rate of antegrade cardioplegia was kept at 150 to 250 mL/min and the aortic pressure was maintained at 75 to 80 mm Hg. Maintaining cardiac arrest with a 12-mEq/L KCl solution was not possible in the pig heart. Therefore, to obtain arrest, we maintained the infusion at 25 mEq/L. Despite a systemic kalemia of 9.5 mmol/L after the hour of arrest, the pigs could be weaned from bypass without the need for inotropic support. In 2 animals treated with potassium to maintain arrest, systemic kalemia reached 17 mmol/L; these animals could not be weaned off bypass 20 minutes after removal of the aortic cross-clamp. These animals were not included in the analysis. The times to arrest and to recovery of cardiac mechanical function were comparable in the esmolol and potassium groups. Cardiac arrest was achieved easily in the esmolol group within 45 to 60 seconds (at experimental doses for induction of 0.91 mg·mL-1·min-1, approximately 2.7 x 10-3 mol/L; and maintenance of 0.31 mg·mL-1·min-1, approximately 9.0 x 10-4 mol/L; Table 2). Except for the two initial potassium experiments, all the pigs were weaned from bypass without the use of any inotropic support at the 20th minute after "reperfusion." The cardioplegic characteristics of esmolol and potassium were similar; however, differences were observed in heart rate (Student's t test, p = 0.042, not supported by the other tests), and in LVDP (Mann-Whitney U test, p = 0.044, not supported by the other tests; Table 3).
|
|
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The mechanisms responsible for potassium arrest have been thoroughly discussed [12]. The critical effect is thought to be the ability of high extracellular potassium to reduce transmembrane potassium gradients and the membrane resting potential. In this way, opening of the sodium channel, which normally allows rapid sodium influx during the upstroke of the action potential, is inactivated because of a reduction in the membrane resting potential to approximately -50 mV.
The mechanism of action of ß-blockers is well known [5, 12]; they inhibit the binding of catecholamines to the ß-adrenergic receptors in the cell membrane. By inhibiting the effects of catecholamines on cell metabolism, they indirectly reduce the number of activated calcium channels, therefore reducing the influx of calcium. The result is a decrease in the heart rate and contractility.
Esmolol is a cardioselective ß-receptor blocker (ß1 antagonist, ß1/ß2 = 35) with a very short duration of action (half-life of 9 minutes [5]. It is normally used in the critical care setting for the treatment of angina pectoris, hypertension, and cardiac arrhythmias, as well as during induction of anesthesia [5, 6]. More recently, it has proven to be effective in the prevention of sudden death after myocardial infarction [13]. Its pharmacologic effects at concentrations greater than 30 mg·kg-1·min-1 are not described in the literature. Severe bradycardia and hypotension have been reported as the major adverse effects of esmolol overdose [5, 14].
Esmolol ((±) -methyl p-[2-hydroxy-3-(isopropylamino) propoxy] hydrocinnamate-C16H26NO6Cl) is soluble in aqueous solutions and is structurally similar to metoprolol. However, esmolol has an ethylene-extended methyl ester group in the para-position of the aryloxypropanolamine structure. The addition of the ester group renders the molecule susceptible to rapid hydrolysis by esterases present in the cytosol of erythrocytes to its acid metabolite (3-[4-[2-hydroxy-3-(isopropylamino) propoxy] phenyl] propionic acid) and methanol. There is evidence that esmolol also may be metabolized by tissue esterases. Because of its ultra–short-acting period, the effects of this drug on heart rate and developed pressure can be monitored closely and dissipate within minutes if the infusion is terminated. The products of esmolol metabolism do not exert any important ß-blocking action. This enzymatic activity is greater in pig and rat blood than in human blood [15, 16].
The acid metabolite of esmolol is an extremely weak ß- blocker with low affinity for ß-adrenergic receptors. It is about 1,500-fold less potent than esmolol. Methanol formed from the metabolism of esmolol is within the range normally found in the untreated population and is at least two orders of magnitude below levels associated with methanol toxicity [5].
The data gathered in this study strongly suggest that drugs other than potassium may be used as cardioplegic agents in normothermic cardiac operations. Esmolol can induce and maintain arrest for periods of as long as 120 minutes with satisfactory values of recovery and within a clinically relevant time frame. The LVEDP increased during arrest, which may be due to the effect of esmolol on L-type calcium channels. Leakage of calcium from the sarcoplasmic reticulum is also possible. The increase in LVEDP was reversible, declining to prearrest values upon cessation of drug infusion.
The data in this study show a great variation in sensitivity to esmolol in the different species studied. Among the species normally used in experimental cardiology, the rat seems to have the most highly developed adrenal system and the highest ß-adrenoceptor density. Thus, in rats, the response to agonists is also much greater than that observed in rabbits and pigs. Conversely, the amount of antagonist required to produce an effect should also be higher. This speculation seems reasonable in terms of our data. In general, pig and rabbit myocardia are considered more similar to human myocardium not only in receptor density, but also in terms of sarcoplasmic reticulum and sarcolemma [17], as well as in conformation of the action potential and conduction patterns. Among mammalian hearts, the rat heart is considered to belong to a separate category. This should be taken into consideration when analyzing these data.
In experimental models of myocardial ischemia, it is generally accepted that ß-receptor density is increased, although a few studies have shown the contrary. It remains controversial whether this increase in receptor density translates into increased adenylyl cyclase activation. The changes observed with ischemia are reversible upon reperfusion, and the duration of coronary artery occlusion or the type of preparation used may affect the findings. The mechanisms clearly involve, however, decreases in high-affinity agonist binding, adenylyl cyclase activation, and depressed functional levels of G-proteins, and will vary with the conditions to which the damaged myocardium is subjected. It seems there is agreement that in patients with cardiac insufficiency, there is at least a decrease in ß-1 receptor function in all types of heart failure. This, however, may not be due exclusively to receptor downregulation, but to what is described as "downstream" events distal to cyclic adenosine monophosphate production, such as defects in calcium handling or sensitivity of the contractile proteins to calcium, or even diminished calcium release from the sarcoplasmic reticulum [18].
Esmolol administered at the concentrations described in the in vivo pig experiments did not abolish cardiac electrical activity, although it maintained continuous mechanical, principally ventricular, arrest. Based on our observations, the sinus node maintains its pacing activity (denoted by the P waves in the ECG) without any ventricular response. The atria demonstrated some motion. In one experiment, the entire heart maintained electrical activity. The ECG showed P waves followed by high-amplitude QRS complexes; however, no ventricular movement was observed (only atrial) and no significant pressure wave was registered.
A possible explanation for the gap found between the concentrations of esmolol that produce mechanical and electrical arrest is that doses capable of inducing electrical arrest have secondary effects on sodium channels. Like other ß-blockers, esmolol has a lidocaine-like effect, although it is much less potent than lidocaine [5]. However, at the nodal cells, the autoexcitatory action potential is mediated by T-type calcium channels [19]. This may be a plausible explanation for the persistence of P waves not followed by QRS complexes as observed in the ECG during the arrest period.
Some concerns can be raised regarding the cardioselectivity and activity of esmolol metabolites at the doses proposed in this study. In the bypass model, there was an apparent increase in pulmonary arterial pressure. Weaning from bypass had to be done more carefully in the esmolol group than in the potassium group. The esmolol-perfused hearts tolerated volume reinfusion poorly during the process of interrupting cardiopulmonary bypass, and volume had to be given very slowly, off bypass, through the aortic line. Despite these alterations, no inotropic support was required and the hearts resumed their initial performance, responding to slow infusion through the aortic cannula of blood accumulated in the reservoir. The hypothesis is that the selectivity of esmolol for various receptors (like other ß blockers) may decrease at higher doses. The ß1/ß2 ratio for esmolol is approximately 35 at the pharmacologic concentrations proposed by the manufacturer. At these concentrations, esmolol may be termed cardioselective [5, 20]. However, at the concentrations used to produce cardiac arrest, esmolol may also interact with ß2 receptors, which may cause the ratio to decrease. Other ß blockers with a higher index of cardioselectivity are under development [21] and should be tested with the same purpose.
Studies using adenosine triphosphate–sensitive potassium channel openers to induce arrest have reported positive results. Dissociation between electrical and mechanical arrest has been observed with Aprikalim (RP 52891; Rhône-Poulenc-Rorer, Antony, France) [22, 23]. Much remains to be elucidated regarding the possible systemic collateral effects of high doses of esmolol, such as hypotension, arrhythmias, and toxicity, to determine its potential clinical utility. Nevertheless, because hyperkalemic cardioplegia is known to have deleterious effects at the subcellular and systemic levels, the search for agents capable of producing prompt, sustainable, and reversible cardiac arrest with little or no sequelae remains timely.
|
Acknowledgments |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Scorsin, A. Mebazaa, N. A. Attar, B. Medini, J. Callebert, R. Raffoul, R. Ramadan, J. M. Maillet, A. Ruffenach, F. Simoneau, et al. Efficacy of esmolol as a myocardial protective agent during continuous retrograde blood cardioplegia J. Thorac. Cardiovasc. Surg., May 1, 2003; 125(5): 1022 - 1029. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Chambers Mechanisms and alternative methods of achieving cardiac arrest Ann. Thorac. Surg., February 1, 2003; 75(2): S661 - 666. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Bessho and D. J. Chambers Myocardial protection: The efficacy of an ultra-short-acting {beta}-blocker, esmolol, as a cardioplegic agent J. Thorac. Cardiovasc. Surg., November 1, 2001; 122(5): 993 - 1003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Yasuda, H. Kamiya, Y. Tanaka, and G. Watanabe Ultra-short-acting cardioselective beta-blockade attenuates postischemic cardiac dysfunction in the isolated rat heart Eur. J. Cardiothorac. Surg., May 1, 2001; 19(5): 647 - 652. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bel, L. P. Perrault, B. Faris, C. Mouas, J.-P. Vilaine, and P. Menasche Inhibition of the pacemaker current: a bradycardic therapy for off-pump coronary operations Ann. Thorac. Surg., July 1, 1998; 66(1): 148 - 152. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
