Effects of Pericardial Lidocaine on Hemodynamic Parameters and Responses in Dogs Anesthetized With Midazolam and Fentanyl

Effects of Pericardial Lidocaine on Hemodynamic Parameters and Responses in Dogs Anesthetized With Midazolam and Fentanyl Motoshi Takada, MD,* Shuji Dohi, MD, PhD,* Shigeru Akamatsu, MD, PhD,† and Akira Suzuki, MD, PhD‡ Objective: Tachycardia during anesthesia should be avoided, especially during off-pump coronary artery bypass graft surgery. Decreasing heart rate without reducing cardiac contractility is an ideal goal. To achieve this, the authors attempted to block the cardiac nerves by pericardial administration of local anesthetic. Design: A prospective study. Setting: A laboratory. Participants: Anesthetized, mechanically ventilated dogs (n ⴝ 69). Interventions: The pericardial space was infused with 2.5 or 5 mL of 1% lidocaine, 5 mL of 2% lidocaine, or normal saline solution. The hemodynamic changes and the cardiac responses to atropine or isoproterenol were measured during cardiac nerve blockade. To examine the inhibitory action of pericardial lidocaine on arrhythmias, an electrical fibrillator was installed. Furthermore, the blood level of lidocaine was measured. Measurements and Main Results: Pericardial injection of lidocaine significantly decreased heart rate without a change in stroke volume. Under pericardial lidocaine, the

tachycardia response to isoproterenol was similar to that observed without pericardial lidocaine, but response to atropine was significantly reduced. Pericardial lidocaine increased the voltage thresholds for inducing arrhythmias and ventricular fibrillation. Intravenous injection of lidocaine elevated the plasma concentration of lidocaine immediately, whereas the plasma concentration peaked at 10 minutes after pericardial administration. Conclusions: Pericardial lidocaine (1) decreased heart rate without affecting stroke volume, (2) preserved the tachycardiac response to isoproterenol but completely blocked the response to atropine, and (3) increased the voltage thresholds for arrhythmias and ventricular fibrillation induced by an electrical fibrillator. These results suggest that pericardial lidocaine may be useful for controlling heart rate during off-pump coronary artery bypass graft surgery. © 2007 Elsevier Inc. All rights reserved.

P

travenous and pericardial administration. The HR responses to isoproterenol and atropine were also observed in the presence of pericardial lidocaine, and the effectiveness of pericardial lidocaine against arrhythmias was assessed.

ATIENTS WITH ISCHEMIC HEART DISEASE can pose particular problems for anesthesiologists. To decrease myocardial oxygen consumption, meticulous control of heart rate (HR) is desirable during anesthesia and surgery. However, this can be very difficult to achieve without decreasing stroke volume, especially during off-pump coronary artery bypass graft (OPCAB) surgery. Although the use of a stabilizer during OPCAB surgery has become popular in recent years, the authors have often experienced difficulty in achieving HR control. Several potential strategies exist for this purpose, including the use of a short-acting ␤-blocker,1 deepening general anesthesia, use of an ␣2-adrenergic agonist,2,3 application of thoracic epidural anesthesia,4-6 and use of a newly improved stabilizer.7,8 Although these may be useful for controlling HR and for easing surgical manipulation, some problems remain during clinical anesthesia, such as a decrease in stroke volume and the appearance of an unexpected bradycardia and/or serious arrhythmias. Although there have been only a few studies of cardiac nerve blockade from the physiologic standpoint,9-12 administration of a local anesthetic into the pericardial sac could be a potential way to control HR during OPCAB surgery. Lidocaine, one of the most commonly used local anesthetics, has been widely used for its antiarrhythmic action13 and its protection of ischemic cardiac muscle.14 Furthermore, lidocaine has been included in cardioplegia solutions used to protect the nonbeating heart.15-17 Hence, when administered into the pericardial sac, lidocaine could be beneficial in controlling HR, as well as in protecting the heart, during OPCAB surgery. To test this hypothesis, the present study was performed by using the technique of pericardial administration of local anesthetic in dogs anesthetized with midazolam and fentanyl. Various doses and concentrations of lidocaine were administered into the pericardial sac, and the hemodynamic changes were observed. Blood samples were collected to verify that the pattern of change in the plasma lidocaine concentration was different between in-

KEY WORDS: local anesthetics, lidocaine, dogs, hemodynamic phenomena, heart rate, stroke volume, infusions, pericardium

METHODS With the institutional animal investigation committee’s approval, 69 (50 ⫹ 6 ⫹ 6 ⫹ 7) dogs (weight 12.0 ⫾ 3.3 kg) were used for the present study. Fifty dogs were used for the observation of hemodynamic changes (10 dogs per group), 6 dogs were used for the measurement of the blood lidocaine concentration, 6 dogs were used for the observation of the cardiac responses to atropine or isoproterenol after the pericardial injection of 5 mL of 1% lidocaine or 5 mL of normal saline, and 7 dogs were used to examine the possible antiarrhythmic effect of pericardial lidocaine. In each dog, anesthesia was induced with pentobarbital, 200 mg, and an endotracheal tube was inserted. Anesthesia was maintained with a continuous intravenous infusion of midazolam, 1 mg/h, fentanyl, 100 ␮g/h, and vecuronium, 1 mg/h, and each dog was mechanically ventilated with air and oxygen, with PCO2

From the *Department of Anesthesiology and Pain Medicine, Gifu University Graduate School of Medicine, Gifu, Japan; †Department of Intensive Care Medicine, Matsunami General Hospital, Gifu, Japan; and ‡Department of Anesthesia and Intensive Care Medicine, Daiyukai General Hospital, Aichi, Japan. Supported in part by a grant-in-aid for scientific research (14207059) from the Ministry of Education, Science, Sports and Culture, Tokyo, Japan. Presented in part at the 2005 ASA Annual Meeting, Atlanta, GA, October 22-26, 2005. Address reprint requests to Motoshi Takada, MD, Department of Anesthesiology and Pain Medicine, Gifu University School of Medicine, 1-1 Yanagido, Gifu City, Gifu 501-1194, Japan. E-mail: [email protected] © 2007 Elsevier Inc. All rights reserved. 1053-0770/07/2103-0014$32.00/0 doi:10.1053/j.jvca.2006.02.004

Journal of Cardiothoracic and Vascular Anesthesia, Vol 21, No 3 (June), 2007: pp 393-399

393

394

controlled at 35 to 45 mmHg. A catheter was inserted into the right femoral artery to measure arterial pressure, and a pulmonary artery catheter (outer diameter 7F) was inserted via the right femoral vein to measure pulmonary artery pressure and cardiac output. An electrocardiogram was used to assess the rhythm. A left thoracotomy was performed in the fifth intercostal space. A polyethylene catheter (inner diameter 0.6 mm, outer diameter 1.0 mm) was inserted into the pericardial sac in the vicinity of the apex, and the pericardial sac was sutured. Temperature was controlled at 36° to 37°C by warming during the study. Fifty dogs were used for the observation of hemodynamic changes (10 dogs per group). The control data for hemodynamic and arterial blood gas analyses were collected after the animal’s hemodynamic variables had stabilized. Then, 2.5 mL of 1% lidocaine, 5 mL of 1% lidocaine, or 5 mL of 2% lidocaine were injected into the pericardial space to anesthetize the cardiac nerves, or 5 mL of 0.9% saline was injected as a control. All solutions were warmed to 37°C before injection. To control for the possibility that lidocaine was absorbed into the blood during its injection into the pericardial space, an identical protocol was followed except that 5 mL of 1% lidocaine were injected intravenously. HR, mean arterial pressure (MAP), mean pulmonary arterial pressure, central venous pressure, and stroke volume were continuously measured in each animal, and values were noted every 5 minutes for 30 minutes. Six dogs were used for the measurement of the blood lidocaine concentration. First, 5 mL of 1% lidocaine were injected into the pericardial space, and a blood sample was collected every 5 minutes for 30 minutes. Then, the remaining lidocaine was removed from the pericardial sac, which was washed out with normal saline, and the authors waited long enough for lidocaine to disappear from the blood (about 60 minutes). Second, 5 mL of 1% lidocaine were injected intravenously, and blood was sampled every 5 minutes for 30 minutes. The samples were centrifuged immediately, and the collected plasma samples were frozen until needed. Six dogs were used for the observation of the cardiac responses to atropine or isoproterenol after the pericardial injection of 5 mL of 1% lidocaine or 5 mL of normal saline. First, normal saline was placed into the pericardial space, and HR and MAP baseline values were obtained. Then, 0.01 mg of isoproterenol were injected intravenously, and HR and MAP were recorded after the injection. The authors waited for 30 minutes for the effect of isoproterenol to disappear, and baseline data for intravenous atropine were recorded. Then, atropine, 0.5 mg, was intravenously injected, and HR and MAP recorded again. After HR had returned to baseline and the effects of atropine had worn off over 60 minutes, the HR and MAP responses to intravenous isoproterenol and atropine in the presence of pericardial lidocaine were measured. Seven dogs were used to examine the possible antiarrhythmic effect of pericardial lidocaine. The setup was the same as that used to observe hemodynamic changes. An electrical fibrillator (model s-9343; Nihon Kohden, Tokyo, Japan) was installed between the pericardium and chest wall. First, as a control, 5 mL of normal saline were injected into the pericardial space, and hemodynamic data were collected. The fibrillator was then fired for 5 seconds at 0.5 V, and the voltage was increased gradually in 0.5-V steps while the voltages needed to induce an arrhythmia (eg, premature ventricular contraction) and ventricular fibrillation (VF) were recorded. After defibrillation, the normal saline was removed from the pericardial space and the hemodynamics stabilized (about 30 minutes). Next, 5 mL of 1% lidocaine were injected into the pericardial space, and the voltages needed to induce arrhythmias were determined. After defibrillation, the pericardial lidocaine was washed out. The method described previously was repeated 3 times for each solution; the average of the 3 values was used for the statistical analysis. The statistical significance of time-dependent hemodynamic changes

TAKADA ET AL

was tested by means of a paired t test. Data were presented as mean ⫾ standard deviation (SD). The blood lidocaine concentrations were analyzed by using an unpaired t test. Data were presented as means ⫾ SD. The cardiac response to isoproterenol was analyzed by using an unpaired t test. An analysis of variance with repeated measures within each subject was used to identify significant differences in the data for the cardiac response to atropine. If the analysis of variance indicated significance, a Fisher protected least significant difference was used to identify which means were significantly different. Data were presented as mean change from baseline ⫾ SD. The effectiveness of the pericardial lidocaine injection against arrhythmias was analyzed by using a paired t test. Data were presented as mean change from baseline ⫾ SD. Differences were considered statistically significant at the level of p ⬍ 0.05. RESULTS

Table 1 shows time-dependent hemodynamic changes. Pericardial lidocaine induced a decrease in HR (versus baseline). When 2.5 mL of 1% lidocaine were administered into the pericardial space, HR did not change significantly, although it showed a tendency to increase. Administration of 5 mL of either 1% or 2% lidocaine into the pericardial space caused HR and MAP to decrease significantly (p ⬍ 0.05). Thereafter, they gradually returned toward baseline in the 30 minutes after the injection. Pericardial saline and intravenous lidocaine (5 mL of 1%) decreased HR at the end of the observation period. Although pericardial lidocaine (5 mL of 1% or 2%) and intravenous lidocaine (5 mL of 1%) increased central venous pressure significantly, they were small changes unlikely to be clinically significant (Fig 1). Mean pulmonary artery pressure did not change significantly in any group. When 5 mL of 2% lidocaine were injected into the pericardial space, stroke volume decreased significantly (p ⬍ 0.05) at 10 minutes after the injection, and when 5 mL of 1% lidocaine were given intravenously, stroke volume decreased significantly (p ⬍ 0.05) at 5 minutes after the injection. Whether lidocaine was injected into the pericardial space or intravenously, its plasma concentration showed an abrupt increase, but there were differences in the timing and magnitude of the peak concentrations with the different administrations (Fig 2). After intravenous injection, the highest blood level was seen at the first blood-sampling point. At that point (5 minutes after the injection), the blood level was 6.4 ⫾ 3.0 ␮g/mL, and it decreased gradually thereafter. In contrast, the plasma concentration was highest 10 to 15 minutes after the pericardial injection (4.1 ⫾ 2.0 ␮g/mL at 10 minutes and 3.9 ⫾ 1.6 ␮g/mL at 15 minutes). A significant difference between the 2 types of administration was shown at 5 minutes after the injection. Isoproterenol caused HR to increase and MAP to decrease; the responses were not significantly different with pericardial lidocaine (Fig 3A and B). Although intravenous atropine, 0.5 mg, increased HR significantly from baseline, the presence of pericardial lidocaine almost completely blocked the HR increase (Fig 3C). There were no statistically significant differences in MAP (versus normal saline) in response to atropine (Fig 3D). Pericardial lidocaine (5 mL of 1%) caused a significant rise in the threshold voltage for inducing arrhythmias (by an average of 0.2 ⫾ 0.2 V, p ⬍ 0.05) and for inducing VF (by an average of 0.6 ⫾ 0.3 V, p ⬍ 0.01) (Table 2).

EFFECTS OF PERICARDIAL LIDOCAINE

395

Table 1. Hemodynamic Changes After Injection (minutes)

Heart rate beats/min PC saline IV lidocaine PC lidocaine, 1%, 2.5 mL PC lidocaine, 1%, 5 mL PC lidocaine, 2%, 5 mL Mean arterial pressure (mmHg) PC saline IV lidocaine PC lidocaine, 1%, 2.5 mL PC lidocaine, 1%, 5 mL PC lidocaine, 2%, 5 mL Central venous pressure (mmHg) PC saline IV lidocaine PC lidocaine, 1%, 2.5 mL PC lidocaine, 1%, 5 mL PC lidocaine, 2%, 5 mL Mean pulmonary artery pressure (mmHg) PC saline IV lidocaine PC lidocaine, 1%, 2.5 mL PC lidocaine, 1%, 5 mL PC lidocaine, 2%, 5 mL Stroke volume (mL) PC saline IV lidocaine PC lidocaine, 1%, 2.5 mL PC lidocaine, 1%, 5 mL PC lidocaine, 2%, 5 mL

Baseline

5

10

15

20

25

30

127 ⫾ 33 109 ⫾ 22 113 ⫾ 36 132 ⫾ 38 137 ⫾ 41

126 ⫾ 34 108 ⫾ 23 122 ⫾ 31 111 ⫾ 29* 107 ⫾ 28†

126 ⫾ 33 104 ⫾ 20 123 ⫾ 31 107 ⫾ 33* 106 ⫾ 35†

124 ⫾ 33 101 ⫾ 18 121 ⫾ 29 108 ⫾ 36* 102 ⫾ 25†

123 ⫾ 33 99 ⫾ 18* 117 ⫾ 30 115 ⫾ 46 101 ⫾ 30†

122 ⫾ 33* 97 ⫾ 18* 114 ⫾ 32 121 ⫾ 49 108 ⫾ 36*

121 ⫾ 33* 94 ⫾ 20* 112 ⫾ 34 123 ⫾ 44 116 ⫾ 38*

96 ⫾ 10 97 ⫾ 12 114 ⫾ 25 120 ⫾ 20 122 ⫾ 17

93 ⫾ 11 93 ⫾ 17 117 ⫾ 32 112 ⫾ 24† 114 ⫾ 14*

93 ⫾ 9S 96 ⫾ 15 115 ⫾ 28 111 ⫾ 26* 110 ⫾ 18†

94 ⫾ 9 97 ⫾ 12 114 ⫾ 22 111 ⫾ 27* 112 ⫾ 18*

93 ⫾ 8 98 ⫾ 11 110 ⫾ 23 110 ⫾ 31 111 ⫾ 17*

94 ⫾ 7 96 ⫾ 10 111 ⫾ 20 114 ⫾ 31 113 ⫾ 18*

94 ⫾ 8 95 ⫾ 10 110 ⫾ 20 114 ⫾ 26 112 ⫾ 18

3⫾2 3⫾2 6⫾2 2⫾1 3⫾2

3⫾2 4 ⫾ 2† 6⫾2 3 ⫾ 1† 4 ⫾ 2*

3⫾2 4⫾2 6⫾2 4 ⫾ 1† 5 ⫾ 2†

3⫾2 4 ⫾ 2* 6⫾2 4 ⫾ 1† 5 ⫾ 2†

3⫾2 4 ⫾ 2* 6⫾2 3 ⫾ 2* 4 ⫾ 2†

3⫾2 4⫾2 6⫾1 3⫾2 4 ⫾ 2†

3⫾2 4 ⫾ 2* 5⫾2 3⫾2 4⫾2

21 ⫾ 5 17 ⫾ 5 27 ⫾ 16 16 ⫾ 5 17 ⫾ 5

21 ⫾ 4 16 ⫾ 5 27 ⫾ 15 16 ⫾ 4 16 ⫾ 4

20 ⫾ 4 16 ⫾ 5 26 ⫾ 15 16 ⫾ 5 16 ⫾ 4

21 ⫾ 4 16 ⫾ 5 27 ⫾ 15 16 ⫾ 4 16 ⫾ 4

21 ⫾ 4 16 ⫾ 4 26 ⫾ 15 17 ⫾ 5 16 ⫾ 4

21 ⫾ 4 16 ⫾ 5 27 ⫾ 15 17 ⫾ 5 16 ⫾ 4

21 ⫾ 4 16 ⫾ 4 26 ⫾ 15 27 ⫾ 5 26 ⫾ 5

10.7 ⫾ 2.8 9.7 ⫾ 2.7 9.6 ⫾ 4 12.1 ⫾ 4.6 13.6 ⫾ 6

10.4 ⫾ 2.3 8.6 ⫾ 2.9* 9.1 ⫾ 4.2 12.1 ⫾ 5.7 13.5 ⫾ 6.2

10.7 ⫾ 2.9 8.7 ⫾ 2.8 8.6 ⫾ 4.3 12.3 ⫾ 5.5 12.6 ⫾ 6.5*

10.4 ⫾ 2.5 9.4 ⫾ 3.1 8.8 ⫾ 4.2 12.7 ⫾ 6.4 12.5 ⫾ 6.9

10.3 ⫾ 2.9 9.6 ⫾ 2.7 9.1 ⫾ 4.7 12 ⫾ 4.9 13.6 ⫾ 6.4

10.6 ⫾ 2.7 9.9 ⫾ 2.7 9.4 ⫾ 4.4 12 ⫾ 4.7 13.2 ⫾ 5.1

10.5 ⫾ 2.9 10.2 ⫾ 3 9.3 ⫾ 3.8 12.6 ⫾ 5.2 13.2 ⫾ 5.5

NOTE. The data are presented as mean values ⫾ SD. Abbreviations: PC, pericardial; IV, intravenous. *p ⬍ 0.05 (significant compared with baseline). †p ⬍ 0.01 (significant compared with baseline).

DISCUSSION

In the present study, it was shown that pericardial lidocaine decreased HR without affecting stroke volume, and at its highest dose, it decreased MAP and increased central venous pressure significantly. Pericardial lidocaine preserved the HR response to isoproterenol but almost completely blocked that to atropine. After injection of 5 mL of 1% lidocaine into the pericardial space, the electrical inducibility of arrhythmias and VF were significantly suppressed. In theory, pericardial lidocaine could affect cardiac function in at least 5 different ways: (1) a direct action on cardiac cells, both on the Na⫹ and/or Ca2⫹ channels of myocytes; (2) a direct action on the autonomic nerves innervating the heart; (3) a systemic effect (ie, acting on the heart and/or centrally on autonomic nervous activity via a pharmacologically active plasma concentration achieved by local absorption); (4) an indirect effect via changes in coronary blood flow because of lidocaine’s action on coronary vessels; and/or (5) a passive effect because of the volume injected into the pericardial space and/or to the pH of the solution. However, the plasma concentration achieved when lidocaine was administered into the

pericardial space was lower than that achieved after its intravenous administration (at a time point at which significant effects were seen after the former administration). It is, therefore, unlikely that lidocaine absorbed from the pericardial space affected the observed cardiovascular parameters in any important way. Similarly, a potential passive effect because of the injectate per se is unlikely because the same volume of normal saline solution had no effect and because the pH and temperature of all solutions injected into the pericardial space were around 6.5 and 37°C, respectively. Although the significance of the other factors under these in vivo experimental conditions is not easy to evaluate, a direct effect of pericardial lidocaine on coronary arteries, and thus on coronary blood flow, needs to be considered. Because lidocaine has been reported to cause coronary arteries to constrict (endothelium independent) at a low dose (10 ␮g/mL) but to dilate them at a high dose (2,000 ␮g/mL),18 it is unlikely that changes in coronary blood flow secondary to pericardial lidocaine contributed in any important way in the present experiments. Thus, the observed cardiac effects (especially HR effects) of pericardial lidocaine would seem most likely to be caused by a blockade of the autonomic

396

TAKADA ET AL

Fig 1. Maximal changes in (A) HR, (B) MAP, and (C) central venous pressure in each of 5 groups. Data were expressed mean ⴞ SD. CVP, central venous pressure; PC, pericardial.

Fig 2. Mean (ⴞ SD) changes in blood concentration of lidocaine every 5 minute for 30 minutes after either pericardial injection of 5 mL of 1% lidocaine (solid line with filled circles) or intravenous bolus injection of 5 mL of 1% lidocaine (solid line with filled squares). *p < 0.05 between groups.

nervous system at the cardiac level and/or a blockade of ion channels (such as Na⫹, Ca2⫹, and K⫹ channels) in cardiac cells. Several investigators have studied the potential actions of pericardially administered local anesthetic from a physiologic standpoint, and pericardial procaine has been used by several groups to block cardiac afferent nerves to study effects of cardiogenic reflexes.19 Dorward et al,10 found that pericardial procaine induced bradycardia in conscious rabbits and speculated that the bradycardia is because of a direct action of procaine on the sinoatrial pacemaker. In conscious dogs, however, pericardial procaine injection consistently caused an increase in HR,11 and it was assumed that this response resulted from the blockade of an ascending sympathoinhibitory pathway of cardiac origin. The explanation for the tendency for an increase in HR with a low volume (2.5 mL) of 1% lidocaine, whereas a decrease was observed with a larger volume (5 mL), is not clear. A relevant observation, made by Dorward et al,10 may be that when a lower concentration of procaine was given for a longer time, vagal efferent function was blocked more rapidly than sympathetic efferent function. Moreover, subepicardial vagal ganglion cells are more susceptible to blockade than postganglionic sympathetic nerve fibers.12 Therefore, a high volume of lidocaine might induce a more intense, although not necessarily complete, blockade of both subepicardial vagal

EFFECTS OF PERICARDIAL LIDOCAINE

397

Fig 3. Changes in (A) HR and (B) MAP 2 minutes after intravenous injection of 0.01 mg of isoproterenol. There were no statistical differences between the groups. Changes in (C) HR and (D) MAP 2 minutes and 5 minutes after intravenous injection of 0.5 mg of atropine. Significant difference between groups, *p < 0.05, **p < 0.01.

ganglion cells and postganglionic sympathetic nerve fibers and thus cause a decrease in heart rate. The anatomic distribution of sympathetic and parasympathetic nerves within the heart is complex, with regional variations; in dogs, sympathetic afferent fibers travel in the superficial subepicardium in an apex-to-base direction, whereas vagal afferent fibers travel deeper in the myocardium until they approach the atrioventricular groove, where they ascend to the Table 2. Hemodynamics and Voltage of Inducing Arrhythmia or VF

Voltage of inducing arrhythmia (V) Voltage of inducing VF (V) Heart rate (beats/min) Mean arterial pressure (mmHg) Central venous pressure (mmHg) Mean pulmonary arterial pressure (mmHg) Stroke volume (mL)

PC Saline

PC Lidocaine

1.2 ⫾ 0.3 1.6 ⫾ 0.4 126 ⫾ 26 102 ⫾ 15 4⫾2 21 ⫾ 5

1.4 ⫾ 0.3* 2.2 ⫾ 0.6† 105 ⫾ 20† 97 ⫾ 22 7 ⫾ 3† 21 ⫾ 5

9⫾4

7⫾3

NOTE. The data are presented as mean values ⫾ SD. Abbreviations: VF, ventricular fibrillation; PC, pericardial. *p ⬍ 0.05 (significant compared with control). †p ⬍ 0.01 (significant compared with control).

superficial subepicardium.20 Furthermore, their anatomic distributions seem not to be uniform throughout the heart. Although the efferent vagal and sympathetic innervations of the right ventricle resemble those of the left ventricle, regional differences in the anatomic distribution exist in the right ventricle, efferent sympathetic fibers to the right ventricular outflow tract being located not only in the subepicardium but also in the subendocardium.21 Selective vagal denervation of the sinus and atrioventricular nodes and atria reportedly decrease heart rate variability and eliminate baroreflex sensitivity, even though the ventricular innervation is preserved.22 Possibly, the previously mentioned opposing effects on HR with differing dosages of pericardial lidocaine might be caused by different receptors located in the ventricles, atria, and/or lungs being affected. After interruption of cardiac vagal afferents by pericardial lidocaine (4-6 mL of 4% solution), efferent sympathetic nerve activity and left atrial pressure after blood volume expansion seem to be markedly attenuated in anesthetized dogs. These results indicate that the heart provides the primary source of afferent input for the control of sympathetic outflow by the vagal cardiopulmonary reflex during changes in thoracic blood volumes and pressures.23 Because the authors did not confirm that the distribution of lidocaine injected was uniform within

398

TAKADA ET AL

the pericardial space or whether the sympathetic and parasympathetic denervations induced by pericardial lidocaine occurred uniformly in these experiments, the possibility that changes in cardiac function would differ depending on which region of the heart was affected by the pericardial lidocaine cannot be excluded. In the present study, the HR response to atropine was completely blocked by pericardial lidocaine, whereas the HR response to isoproterenol was well preserved. This suggests that pericardial lidocaine, used at the present concentration (5 mL of 1%), is unlikely to affect ␤1-adrenergic receptor stimulation at the heart level. The HR response to atropine seems to be more profoundly blocked by pericardial lidocaine than by the cardiac sympathectomy produced with cervical or thoracic epidural anesthesia induced using lidocaine.6 Thus, pericardial lidocaine may provide a more potent blockade of cardiac sympathetics, mainly sympathetic efferents to the heart, than epidural anesthesia. It is well known that intravenous lidocaine raises the arrhythmic threshold and that its antiarrhythmic action is related to its interaction with a specific receptor associated mainly with the cardiac sodium channel.24,25 Thus, the observation that pericardial lidocaine raised the voltages needed to induce arrhythmias and VF was predictable in view of (1) the blood lidocaine level achieved via absorption, and (2) lidocaine’s direct action on the heart. Lidocaine administered into the pericardial space might anesthetize the surface of the myocardium and thus might inhibit the spread of stimulation over the myocardium. Furthermore, because many factors affect autonomic nervous activity and thus dispose the heart to ventricular arrhythmias during and after anesthesia and surgery26 and because vagal withdrawal and/or adrenergic hyperactivity may help to precipitate ventricular arrhythmias, the cardiac sympathectomy produced by pericardial lidocaine may itself reduce arrhythmias and increase the threshold for VF.27 It is well known that enhanced sympathetic activity predisposes the heart to arrhythmias and increases the myocardial vulnerability to VF, and so the balance of autonomic control is more important than the absolute levels of sympathetic and parasympathetic activity. Moreover, the electrical inducibility of VF differs with the basal anesthetic agents used.28 Indeed, anesthetic agents may suppress, to some extent, cardiovascular responses and also the electrical inducibility of ventricular tachycardia.29 Although fentanyl and midazolam, as used in the present experiments, are widely thought to be two of the best choices of anesthetics for patients undergoing cardiac surgery,30,31 it remains to be established how the combi-

nation of pericardial lidocaine with fentanyl and midazolam affects arrhythmogenicity. The present results indicate that pericardial lidocaine can decrease HR without decreasing cardiac function, preserve the HR response to isoproterenol, and increase the voltage thresholds for arrhythmias and VF. These effects would seem to be very important for controlling HR during cardiac surgery, and OPCAB surgery may be a situation for which this new means of drug administration is indicated. However, there are still some issues to resolve before clinical use. Pericardial administration of a large dose of lidocaine may cause significant decreases in MAP and stroke volume, as reported in previous studies with procaine.19 Indeed, it was noted that giving 2% lidocaine into the pericardial space decreased MAP and stroke volume at 10 minutes after injection. Furthermore, there is a narrow margin between the doses of intrapericardial procaine that block cardiac nerves and those that can produce confounding effects from phrenic nerve blockade or absorption into the bloodstream. Indeed, respiratory depression because of phrenic nerve blockade has been reported when this agent is given into the intrapericardial space.19 In addition, intrapericardial block by a local anesthetic may eliminate the heart as a reflexogenic organ9 and might affect cardiac function via neurohumoral changes. Because many receptors located in the ventricle, atria, and great vessels could serve as the primary source of the afferent input for the vagal cardiopulmonary reflex, reflexes arising from the manipulation required for OPCAB surgery might be affected or eliminated by the local anesthesia. Conceivably, potential blockade of this input might be beneficial for patient management during the anesthesia needed for OPCAB. In this study, lidocaine that was given into the closed pericardial space would spread to the surface of the whole heart by its beating motion. However, the pericardial cavity is wide open during CABG surgery. Moreover, further studies are needed in the ischemic heart and to determine interactions with coronary vasodilators (eg, nitroglycerin). In summary, in anesthetized dogs, pericardial lidocaine decreased HR without affecting stroke volume and preserved the HR response to isoproterenol, but completely blocked that to atropine. Moreover, it raised the voltage thresholds for the induction of arrhythmias and VF by an electrical fibrillator. Taking previous findings together with the present data suggest that selective autonomic nerve blockade using pericardial lidocaine may be beneficial for a clinical situation that requires meticulous HR control, such as anesthesia for OPCAB surgery.

REFERENCES 1. Booth JV, Spahn DR, McRae RL, et al: Esmolol improves left ventricular function via enhanced beta-adrenergic receptor signaling in a canine model of coronary revascularization. Anesthesiology 97:162169, 2002 2. Stuhmeier KD, Mainzer B, Cierpka J, et al: Small, oral dose of clonidine reduces the incidence of intraoperative myocardial ischemia in patients having vascular surgery. Anesthesiology 85:706712, 1996 3. Myles PS, Hunt JO, Holdgaard HO, et al: Clonidine and cardiac surgery: Haemodynamic and metabolic effects, myocardial ischaemia and recovery. Anaesth Intensive Care 27:137-147, 1999

4. Chakravarthy M, Jawali V, Patil TA, et al: High thoracic epidural anesthesia as the sole anesthetic for performing multiple grafts in off-pump coronary artery bypass surgery. J Cardiothorac Vasc Anesth 17:160-164, 2003 5. Kessler P, Neidhart G, Bremerich DH, et al: High thoracic epidural anesthesia for coronary artery bypass grafting using two different surgical approaches in conscious patients. Anesth Analg 95:791-797, 2002 6. Dohi S, Nishikawa T, Ujike Y, et al: Circulatory responses to airway stimulation and cervical epidural blockade. Anesthesiology 57:359-363, 1982

EFFECTS OF PERICARDIAL LIDOCAINE

7. Scott NA, Knight JL, Bidstrup BP, et al: Systematic review of beating heart surgery with the Octopus Tissue Stabilizer. Eur J Cardiothorac Surg 21:804-817, 2002 8. Jansen EW, Lahpor JR, Borst C, et al: Off-pump coronary bypass grafting: How to use the Octopus Tissue Stabilizer. Ann Thorac Surg 66:576-579, 1998 9. Arndt JO, Pasch U, Samodelov LF, et al: Reversible blockade of myelinated and nonmyelinated cardiac afferents in cats by instillation of procaine into the pericardium. Cardiovasc Res 15:61-67, 1981 10. Dorward PK, Flaim M, Ludbrook J: Blockade of cardiac nerves by intrapericardial local anaesthetics in the conscious rabbit. Aust J Exp Biol Med Sci 61:219-230, 1983 11. O’Donnell CP, Scheuer DA, Keil LC, et al: Cardiac nerve blockade by infusion of procaine into the pericardial space of conscious dogs. Am J Physiol 260:R1176-R1182, 1991 12. Samodelov LF, Pohl M, Arndt JO: Reversible blockade of cardiac efferents with procaine instilled into the pericardium of cats. Cardiovasc Res 16:187-193, 1982 13. Lown B, Verrier RL: Neural activity and ventricular fibrillation. N Engl J Med 294:1165-1170, 1976 14. Hinokiyama K, Hatori N, Ochi M, et al: Myocardial protective effect of lidocaine during experimental off-pump coronary artery bypass grafting. Ann Thorac Cardiovasc Surg 9:36-42, 2003 15. Dobson GP, Jones MW: Adenosine and lidocaine: A new concept in nondepolarizing surgical myocardial arrest, protection, and preservation. J Thorac Cardiovasc Surg 127:794-805, 2004 16. Asano M, Inoue K, Ando S, et al: Optimal temperature of continuous lidocaine perfusion for the heart preservation. Jpn J Thorac Cardiovasc Surg 51:1-9, 2003 17. Baraka A, Hirt N, Dabbous A, et al: Lidocaine cardioplegia for prevention of reperfusion ventricular fibrillation. Ann Thorac Surg 55:1529-1533, 1993 18. Perlmutter NS, Wilson RA, Edgar SW, et al: Vasodilatory effects of lidocaine on epicardial porcine coronary arteries. Pharmacology 41:280-285, 1990 19. Evans RG, Hayes IP, Ludbrook J, et al: Factors confounding blockade of cardiac afferents by intrapericardial procaine in conscious rabbits. Am J Physiol 264:H1861-H1870, 1993

399

20. Barber MJ, Mueller TM, Davies BG, et al: Phenol topically applied to canine left ventricular epicardium interrupts sympathetic but not vagal afferents. Circ Res 55:532-544, 1984 21. Ito M, Zipes DP: Efferent sympathetic and vagal innervation of the canine right ventricle. Circulation 90:1459-1468, 1994 22. Chiou CW, Zipes DP: Selective vagal denervation of the atria eliminates heart rate variability and baroreflex sensitivity while preserving ventricular innervation. Circulation 98:360-368, 1998 23. Minisi AJ: Vagal cardiopulmonary reflexes after total cardiac deafferentation. Circulation 98:2615-2620, 1998 24. Hill RJ, Duff HJ, Sheldon RS: Class I antiarrhythmic drug receptor: Biochemical evidence for state-dependent interaction with quinidine and lidocaine. Mol Pharmacol 36:150-159, 1989 25. Jia H, Furukawa T, Singer DH, et al: Characteristics of lidocaine block of sodium channels in single human atrial cells. J Pharmacol Exp Ther 264:1275-1284, 1993 26. Feeley TW: Management of perioperative arrhythmias. J Cardiothorac Vasc Anesth 11:10-15, 1997 27. Schwartz PJ, Snebold NG, Brown AM: Effects of unilateral cardiac sympathetic denervation on the ventricular fibrillation threshold. Am J Cardiol 37:1034-1040, 1976 28. Hunt GB, Ross DL: Comparison of effects of three anesthetic agents on induction of ventricular tachycardia in a canine model of myocardial infarction. Circulation 78:221-226, 1988 29. Hanouz JL, Yvon A, Flais F, et al: Propofol decreases reperfusion-induced arrhythmias in a model of “border zone” between normal and ischemic-reperfused guinea pig myocardium. Anesth Analg 97: 1230-1238, 2003 30. Rivenes SM, Lewin MB, Stayer SA, et al: Cardiovascular effects of sevoflurane, isoflurane, halothane, and fentanyl-midazolam in children with congenital heart disease: An echocardiographic study of myocardial contractility and hemodynamics. Anesthesiology 94:223229, 2001 31. Zickmann B, Hofmann HC, Pottkamper C, et al: Changes in heart rate variability during induction of anesthesia with fentanyl and midazolam. J Cardiothorac Vasc Anesth 10:609-613, 1996