- Email: [email protected]
Lidocaine cardioplegia for prevention of reperfusion ventricular fibrillation
Lidocaine Cardioplegia for Prevention of Reperfusion Ventricular Fibrillation Anis Baraka, FRCA(Hon), Nada Hirt, MD, Aliya Dabbous, MD, Samar Taha, MD, Corine Rouhana, MD, Nawal El-Khoury, MD, Maroun Ghabash, MD, Mireille Jamhoury, MD, and Abla Sibaii, MS Department of Anesthesiology and Department of Epidemiology and Biostatistics, American University of Beirut, Beirut, Lebanon
Lidocaine addition to crystalloid cardioplegic solution for prevention of reperfusion ventricular fibrillation after the release of the aortic cross-clamp was studied in 50 patients undergoing coronary artery bypass grafting and in 30 patients undergoing mitral or aortic valve replacement. Twenty-six of the patients undergoing coronary artery bypass grafting received lidocaine, 100 mg/L of cardioplegia, whereas a control group of 24 patients received cardioplegia without lidocaine. In the group undergoing valve replacement, 14 patients received lidocaine cardioplegia and 16 patients served as control. In
the coronary artery bypass grafting group, lidocaine cardioplegia reduced significantly the incidence of reperfusion ventricular fibrillation from 100% to 42%. In the valve group, lidocaine cardioplegia also reduced significantly the incidence of reperfusion ventricular fibrillation from 93% to 42%. In both groups, lidocaine cardioplegia decreased the number of direct-current countershocks required to defibrillate the heart, with no significant increase in the incidence of high-grade atrioventricular block. (Ann Thorac Surg 1993;55:1529-33)
R
Material and Methods
eperfusion ventricular fibrillation (VF) is a major concern in patients undergoing open heart operations, despite the improvement in myocardial protection strategies. After the release of the aortic cross-clamp (ACC), VF is reported to occur in 74% to 96% of patients [1-4]. Reperfusion VF may result in increased myocardial wall tension, increased myocardial oxygen consumption, and impaired subendocardial blood flow [5, 61. Furthermore, defibrillation with direct-current (DC)countershock will result in myocardial injury [7]. Thus, it is advantageous to prevent the occurrence of reperfusion VF after release of the ACC. Previous reports have demonstrated in patients undergoing coronary artery bypass grafting (CABG) that addition of lidocaine to cold hyperkalemic crystalloid cardioplegia, in a concentration of 500 mg/L, decreased significantly the incidence of reperfusion VF; however, the incidence of high-grade atrioventricular (AV) block was markedly increased [3, 81. Using a lower concentration of lidocaine (100 mgL) decreased the incidence of reperfusion VF without increasing the incidence of heart block [4]. The purpose of this report is to assess the efficacy of lidocaine, 100 mgiL of hyperkalemic crystalloid cardioplegic solution, in reducing the incidence of reperfusion VF in patients undergoing CABG as compared with patients undergoing mitral or aortic valve replacement.
Accepted for publication Sep 30, 1992 Address reprint requests to Dr Baraka, Department of Anesthesiology, American University of Beirut, Beirut, Lebanon. 0 1993 by The Society of Thoracic Surgeons
Fifty patients undergoing elective CABG and 30 patients undergoing aortic or mitral valve replacement were included in the study. The investigation conformed to the ethical standards as set out in the Declaration of Helsinki, and informed consent was obtained from all patients. Patients were randomly assigned to receive either cold potassium cardioplegia without lidocaine (control group) or cold potassium cardioplegia with lidocaine, 100 mgL (lidocaine group). In patients undergoing CABG, 26 patients received lidocaine cardioplegia, and 24 patients served as controls. In patients undergoing valve replacement, 14 patients received lidocaine cardioplegia, and 16 patients served as controls. Patients were premedicated with morphine, 0.1 mg/kg; promethazine, 25 mg; and scopolamine, 0.3 mg intramuscularly. Anesthesia was induced with midazolam, 0.1 mgkg; fentanyl, 50 pgkg; and a mixture of vecuronium, 0.1 mg/kg, and pancuronium, 0.1 mg/kg. After tracheal intubation, ventilation was controlled with 100% oxygen with no inhalation anesthetic supplementation. Patients were monitored with an electrocardiogram (V5),radial artery cannula, and thermodilution Swan-Ganz pulmonary artery catheter. Cardiopulmonary bypass was initiated using a bubble oxygenator (Bentley 10B). Patients were perfused by a roller pump (Sarns 7000, Ann Arbor, MI) at a flow rate of 2.4 L min-' * m-*. Moderate systemic hypothermia, around 28"C, was induced during cardiopulmonary bypass. Cold potassium cardioplegic solution was infused at the aortic root or in the coronary ostia. The composition of the cardioplegic solution was as follows: KCl, 20 mEq/L; NaHCO,, 20 mEqL; mannitol, 20 g/L; and dextrose, 12 g/L in lactated Ringer's. Lidocaine, 100 mg, was added 0003-4975/93/$6.00
1530
Ann Thorac Surg 1993;55:1529-33
BARAKA ET AL LIDOCAINE CRYSTALLOID CARDIOPLEGIA
Table 1. Hernodynamic Data 20 Minutes After Induction of Anesthesia in Patients Undergoing Coronary Artery Bypass Grafting Variable HR (beatdmin) MAP (mm Hg) PCWP (mm Hg) CO (Wmin) CI (L min-’ . m-’)
-
Control Group Lidocaine Group P (n = 24) Value (n = 26)
-
68.0 f 9.7 93.2 f 9.2 10.2 f 3.7 2.9 f 1.0 1.68 f 0.46
62.7 f 11.6 90.4 f 10.6 10.5 f 5.0 3.0 t 0.87 1.72 k 0.44
NS NS NS NS NS
C1 = cardiac index; CO = cardiac output; HR = heart rate; M A P = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
to each liter of cardioplegia used in the lidocaine group of patients, whereas cardioplegic solution without lidocaine was used in the control group. The volume of cardioplegic solution initially infused was 10 mL/kg, followed by intermittent infusion every 20 minutes or whenever electrical activity reappeared. Patients‘ data consisted of age, sex, weight, and preoperative medications, as well as electrocardiographic, echocardiographic, and coronary angiographic findings. Intraoperative data included number of vessels bypassed in the CABG groups, bypass time, ACC time, degree of hypothermia, and total volume of cardioplegia used. At the time of the release of the ACC, we recorded central venous blood temperature and mean arterial blood pressure. Also, an arterial blood sample was analyzed for electrolytes and blood gases. The cardiac rhythm after unclamping of the aorta was recorded continuously until sinus rhythm took place, and the incidence of VF was compared in the control group versus the lidocaine group. Whenever VF followed the release of the ACC, DC countershocks of 5 J were used for defibrillation. Increments of 5 J were added when further shocks were needed. The need for a pacemaker was noted in each group. After weaning from cardiopulmonary bypass, inotropic support was provided whenever necessary by an infusion of dobutamine at a rate of 5 to 10 pg/kg per minute. The rate and duration of the infusion were subsequently
Table 2. lntraoperative Data in Patients Undergoing Coronary Artery Bypass Grafting Variable Number of grafts Bypass time (min) ACC time (min) Hypothermia (“C) Total cardioplegia volume per weight (mL/kg)
Control Group (n = 24)
Lidocaine Group p (n = 26) Value
2.54 f 0.66 65.5 t 19.1
2.92 f 0.74 69.4 f 20.1
NS NS
38.6 f 14.3 29.1 f 2.26 14.62 f 4.73
43.6 f 14.8 28.8 f 1.82 15.83 f 4.0
NS
ACC = aortic cross-clamp;
NS = not significant
NS
NS
adjusted according to the hemodynamic parameters of the patients. In all patients, the different hemodynamic parameters including cardiac output measurement by thermodilution were monitored 20 minutes after induction of anesthesia and 15 minutes after weaning from cardiopulmonary bypass. Data are expressed as absolute numbers or mean f standard deviation. The results are analyzed using the 2 analysis for discrete variables and by Student’s t test for continuous variables. Values of p less than 0.05 were considered significant.
Results Coronary Artery Bypass Graffing Group The control group and lidocaine group of patients were similar with respect to age, sex, preoperative ECG changes, left ventricular function, and preoperative medications. The mean age was 57.0 f 7.8 years in the control group and 56.6 f 10.9 years in the lidocaine group. Preoperative electrocardiograms showed evidence of old myocardial infarction in 11 patients of the control group and 13 patients of the lidocaine group. Preoperative pblockers were administered to 15 patients of the control group and 21 patients of the lidocaine group, whereas calciumchannel blockers were used in 18 patients of the control group and in 18 patients of the lidocaine group. There was no significant difference between the hemodynamic variables of the two groups determined 20 minutes after induction of anesthesia (Table 1).The number of vessels grafted, the bypass time, ACC time, degree of hypothermia, and the total cardioplegic volume used were not significantly different between the groups (Table 2). At the time of unclamping of the aorta, central venous blood temperature, arterial potassium concentration, oxygen tension, carbon dioxide tension, pH, and mean arterial blood pressure were also not significantly different (Table 3). DEMOGRAPHIC AND INTRAOPERATIVE DATA.
REPERFUSION VENTRICULAR FIBRILLATION. The incidence of reperfusion VF after release of the ACC was significantly reduced in the lidocaine group (Fig 1).Also, the number of DC countershocks required to defibrillate the heart was
Table 3. Conditions at Aortic Cross-Clamp Release in Patients Undergoing Coronary Artery Bypass Grafting Variable Temperature (“C) Potassium (mEqlL) PO, (mm Hg) PCOl (mm Hg) PH MAP (mm Hg)
Control Group Lidocaine Group p (n = 24) (n = 26) Value 33.8 f 1.49 4.56 f 0.65 375 f 119 30.3 f 5.1 7.52 f 0.049 91.0 f 14.1
33.9 f 1.8 4.58 f 0.54 404 f 109 28.7 f 2.8 7.54 f 0.041 88.8 f 13.6
NS = not significant; MAP = mean arterial blood pressure; POz = oxygen tension. = carbon dioxide tension;
NS NS
NS NS
NS NS PCO,
BARAKA ET AL LIDOCAINE CRYSTALLOID CARDIOPLEGIA
Ann Thorac Surg
1993:55:1529.33
1531
Table 5. Hernodynamic Data 20 Minutes After lnduction of Anesthesia in Patients Undergoing Valve Replacement
p=o.ooo1 CONTROL GROUP LIDOCAME CROUP
Variable
Control Group Lidocaine Group p (n = 16) (n = 14) Value
HR (beatslmin) MAP (mm Hg) PCWP (mm Hg) CO (L/min) CI (L * min-’ . m-2)
74.7 f 13.7 83.7 f 6.5 23.1 f 18.2 2.91 f 1.35 1.62 2 0.62
86.7 f 23.0 86.3 f 7.85 26.6 f 23.7 3.27 f 0.99 1.90 f 0.59
NS NS NS NS NS
CI = cardiac index; CO = cardiac output; HR = heart rate; MAP = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
VENTRICULAR FIBRILLATION Fig 1. Incidence of reperfusion ventricular fibrillation in the patients undergoing coronary artery bypass grafting. Ventricularfibrillation followed release of the aortic cross-clamp in all 24 control patients (ZOO%), as compared with 11 of the 26 patients of the lidocaine group (42.3% ).
significantly decreased from 2.0 & 0.75 in the control group to 1.27 f 0.44 in the lidocaine group. ATRIOVENTRICULAR BLOCK. Atrioventricular block requiring epicardial pacing occurred after release of the ACC in 9 of 24 in the control group and in 10 of 26 in the lidocaine group. There was no significant difference between the two groups. Among the patients in whom AV block developed, only 3 patients in the control group and 2 patients in the lidocaine group required continued pacing on discharge from the operating room.
HEMODYNAMIC PARAMETERS. The hemodynamic parameters measured 15 minutes after weaning from bypass were similar in the two groups (Table 4).
left ventricular function, and preoperative medications. The mean age was 44.3 f 11.9 years in the control group and 44.4 ? 15.9 years in the lidocaine group. Atrial fibrillation was present in 5 patients of the control group and 6 patients of the lidocaine group. Preoperative digitilization was used in 13 patients of the control group and 11 patients of the lidocaine group. There was no significant difference between the hemodynamic parameters of the two groups determined 20 minutes after induction of anesthesia (Table 5). Bypass time, ACC time, degree of hypothermia, and total cardioplegia volume were not significantly different between the groups (Table 6). At the time of unclamping of the aorta, central venous blood temperature, arterial potassium concentration, oxygen tension, carbon dioxide tension, pH, and mean arterial pressure were also not significantly different (Table 7). The incidence of reperfusion VF after release of the ACC was significantly reduced in the lidocaine group (Fig 2). Also, the number of DC countershocks required to defibrillate the heart was decreased from 2.79 k 1.69 in the control group to 1.26 k 0.54 in the lidocaine group. REPERFUSION VENTRICULAR FIBRILLATION.
The control group and the lidocaine group of patients were similar with respect to age, sex, preoperative atrial fibrillation,
Atrioventricular block requiring epicardial pacing occurred after release of the ACC in 7 of 16 in the control group and in 5 of 14 patients of the lidocaine group. There was no significant difference between the two groups. Among the patients in whom AV
Table 4. Hernodynamic Data 15 Minutes After Weaning Off Bypass in Patients Undergoing Coronary Artery Bypass Grafting
Table 6. lntraoperative Data in Patients Undergoing Valve Replacement
Valve Group
ATRIOVENTRICULAR BLOCK.
DEMOGRAPHIC AND INTRAOPERATIVE DATA.
Control Group Lidocaine Group p (n = 24) (n = 26) Value
Variable HR (beatdmin) MAP (mm Hg) PCWP (mm Hg) CO (Umin) CI (L * mir-’ m-’)
-
92.7 f 14 89.6 f 7.3 14.0 f 6.8 3.4 f 1.4 2.0 0.64
*
86.2 f 9.7 89.1 f 11.7 12.5 f 5.1 3.2 f 0.96 1.76 f 0.53
NS NS NS NS NS
CI = cardiac index; CO = cardiac output; HR = heart rate; M A P = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
Variable Bypass time Win) ACC time (min) Hypothermia (“C) Total cardioplegia volume per weight (mL/kg)
Control Group (n = 16)
Lidocaine Group (n = 14)
p Value
57.4 f 17.4
62.3
18.1
NS
42.3 f 13.0 28.1 f 0.88 16.25 f 6.6
42.9 f 15.7 28.7 f 1.4 13.4 5 3.3
NS NS NS
ACC = aortic cross-clamp;
NS = not Significant.
f
1532
BARAKAETAL LIDOCAINE CRYSTALLOID CARDIOPLECIA
Ann Thorac Surg 1993;55:1529-33
Table 7 . Conditions at Aortic Cross-Clamp Release in Patients Undergoing Valve Replacement
Table 8 . Hernodynamic Data 15 Minutes After Weaning Bypass in Patients Undergoing Valve Replacement
Control Group Lidocaine Group p (n = 16) (n = 14) Value
Variable
Variable
Control Group Lidocaine Group P (n -- 16) (n = 14) Value
~~~~
Temperature (“C) Potassium (mEq/L) PO2 (mm Hg) PCOz (mm Hg) PH Pressure (mm Hg)
35.2 & 1.7 4.13 2 0.56 344 f 135 32.3 f 7.1 7.51 f 0.08 80.6 2 14.8
34.6 f 1.05 4.15 f 0.603 415 2 108 29.5 f 2.3 7.52 f 0.03 81.0 f 13.6
NS NS NS NS NS NS
~~~~
NS = not significant; oxygen tension.
PCO, = carbon dioxide tension;
PO, =
block developed, only 1 patient in each group required continued pacing on discharge from the operating room. HEMODYNAMIC PARAMETERS. The hemodynamic parameters measured 15 minutes after weaning from bypass were similar in the two groups (Table 8).
Comparison of Bypass Grafting and Valve Operation The incidences of reperfusion VF and high-grade AV block were not significantly different between the control group of patients undergoing CABG and those undergoing valve replacement, or between the lidocaine group of patients undergoing CABG and those undergoing valve replacement. In both the valve and the CABG groups, lidocaine cardioplegia significantly decreased the incidence of reperfusion VF after release of the ACC, with no significant change in the incidence of pacemaker insertion.
loo-. 90 -.
93.77.
p=0.002
80-.
w
CONTROL CROUP
70-1
m L I D O C A I N E GROUP
2W 607: a
50- . 40-.
42.0%
30 _. 20-.
-
10 -.
VENTRICULAR FIBRILLATION Fig 2 . Incidence of reperfusion ventricular fibrillation in patients undergoing valve replacement. Ventricular fibrillation followed release of the aortic cross-clamp in 15 of the 16 control patients (93.7%),as compared with 6 of the 14 patients of the lidocaine group (42.8%).
HR (beatdmin) MAP (mm Hg) PCWP (mm Hg) CO (Umin) CI (L * min-’ m-’)
off
102.7 f 14.6 87.3 f 15.0 15.8 2 5.8 3.91 f 1.4 2.23 f 0.78
102.1 f 16.1 87.2 f 6.6 17.8 f 6.0 4.00 f 0.95 2.34 f 0.530
NS NS NS NS NS
CI = cardiac index; CO = cardiac output; HR = heart rate; MAP = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
Comment Aortic cross-clamping is an important cause of myocardial damage during cardiopulmonary bypass; the magnitude of damage is determined by the duration of ischemia, and the extent of supplyldemand imbalance [9, 101. Suboptimal myocardial protection during ACC may result in a heterogenous population of cells that can be classified as irreversibly damaged, minimally damaged, or stunned [ l l ] . After release of the ACC, the heterogenous recovery of myocardial cells and the subsequent increase in reentry [12] and automaticity [13], as well as the possibility of reperfusion-induced injury [14], can result in a high incidence of reperfusion VF. Ventricular fibrillation has been shown to cause redistribution of coronary blood flow away from the subendocardium, and to result in myocardial ischemia and impaired ventricular performance, in the face of increased myocardial oxygen consumption [5, 61. This will result in intramyocardial acidosis [15] and release of creatine kinase MB postoperatively [16]. In an attempt to decrease the incidence of reperfusion VF after release of the ACC, lidocaine has been added to the cardioplegic solution in patients undergoing CABG (3, 4, 81. The present report demonstrates that lidocaine, 100 mg/L of cardioplegic solution, produces a significant reduction in the incidence of reperfusion VF in patients undergoing CABG as well as in patients undergoing valve replacement. Also, the addition of lidocaine to the cardioplegic solution decreases significantly the number of DC countershocks required to defibrillate the heart in both the CABG and the valve groups. The decrease in number of DC countershocks required to defibrillate is advantageous, because DC countershocks have been shown to cause myocardial necrosis [7], subepicardial contraction abnormalities [171, and marked dehiscence of the intercalated discs between damaged myocytes [18]. Myocardial damage due to direct defibrillation can be avoided by using energy levels lower than 30 Wattseconds, minimizing the number of shocks given, and allowing a few minutes between consecutive shocks [19]. Lidocaine, an amide local anesthetic, belongs to class IB antiarrhythmic drugs. It alters membrane conductance of cations by binding to sodium channels. Lidocaine decreases the slope of phase 4 depolarization and increases
Ann Thorac Surg 1993;55:1529-33
the diastolic electrical current threshold in Purkinje fibers [20, 211. Also, it increases the threshold for ventricular fibrillation, and the increase in threshold is related to plasma lidocaine concentration [22]. These properties can explain our findings showing that the addition of lidocaine to the cardioplegic solution decreases significantly the incidence of reperfusion VF after the release of the ACC, as well as the number of DC countershocks required to defibrillate the heart. Thus, lidocaine cardioplegia will minimize the myocardial damage after release of the ACC by decreasing the incidence of reperfusion VF as well as the energy and number of countershocks required for defibrillation. The increase in high-grade AV block observed by previous investigations when lidocaine, 500 mg/L of cardioplegic solution, has been used [3, 81 is not demonstrated in our study when a lidocaine dose of only 100 mg/L is used. The incidence of AV block in our patients who received lidocaine cardioplegia is not significantly different from that observed in the control group of patients, whether undergoing CABG or valve replacement. The different hemodynamic parameters after weaning from bypass were not significantly different between the lidocaine group and the control group. However, the early postoperative period is a rather unstable period, and hence these findings may not reflect the subsequent hemodynamic performance; recent studies have shown increased levels of creatine kinase MB isoenzymes in patients in whom reperfusion VF develops [16]. In conclusion, the present report demonstrates that lidocaine addition to potassium cardioplegic solution in a dose of 100 mg/L reduces significantly the incidence of reperfusion VF after release of the ACC in patients undergoing CABG as well as in patients undergoing valve replacement. Also, lidocaine cardioplegia decreases in both groups the number of DC countershocks required to defibrillate the heart, without increasing the incidence of high-grade AV block.
References Fall SM, Burton NA, Graeber GM, et al. Prevention of ventricular fibrillation after myocardial revascularization. Ann Thorac Surg 1987;43:1824. Curling PE, Nagle D, Zeidan JR, Kaplan JA. Effects of lidocaine, propranolol and bretylium on reperfusion dysrhythmias. Anesthesiology 1981;55:A52. Tchervenkov CI, Symes JF, Wynands JE, et al. Lidocaine as an adjunct to high potassium cardioplegia: a prospective randomized clinical trial. Surg Forum 1983;34:319-21. Fiore AC, Naunheim KS, Taub J, et al. Myocardial preservation using lidocaine blood cardioplegia. Ann Thorac Surg 1990;50:771-5.
BARAKA ET AL LIDOCAINE CRYSTALLOID CARDIOPLEGIA
1533
5. Hottenrott C, Maloney JV, Buckberg G. Studies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow 111. J Thorac Cardiovasc Surg 1974;68:634-45. 6. Buckberg GD, Hottenrott CE. Ventricular fibrillation: its effect on myocardial flow, distribution and performance. Ann Thorac Surg 1975;20:76-85. 7. Dahl CF, Ewy GA, Warner ED, Thomas ED. Myocardial necrosis from direct current countershock: effect of paddle electrode size and time interval between discharges. Circulation 1974;50:956-61. 8. Tchervenkov CZ, Wynands JE, Symes JF, Malcolm ID, Dobell ARC, Morin JE. Electrical behavior of the heart following high-potassium cardioplegia. Ann Thorac Surg 1983;36: 3149. 9. Iyengar SRK, Ramchand S, Charrette EJP, lyengar CKS, Lynn RB. Anoxic cardiac arrest: an experimental and clinical study of its effects. J Thorac Cardiovasc Surg 1973;66:722-30. 10. Hearse DJ, Garlick PB, Humphrey SM. Ischemic contracture of the myocardium: mechanisms and prevention. Am J Cardiol 1977;39:98&93. 11. Braunwald E, Honer RA. The stunned myocardium: prolonged, postischemic ventricular function. Circulation 1982; 66114G9. 12. Murdock D, Loeb JM, Euler DE, Randall WC. Electrophysiology of coronary reperfusion and mechanism for reperfusion arrhythmias. Circulation 1980;61:175-82. 13. Kaplinsky E, Ogawa S, Michelson EL, Dreifus LS. Instantaneous and delayed ventricular arrhythmias after reperfusion of acutely ischemic myocardium: evidence for multiple mechanisms. Circulation 1981;63:333-40. 14. Hearse DJ. Reperfusion of the ischemic myocardium. J Mol Cell Cardiol 1977;9:605-16. 15. Khuri SF, Maiston WA, Josa M, et al. Observations on 100 patients with continuous intraoperative monitoring of intramyocardial pH: the adverse effects of ventricular fibrillation and reperfusion. J Thorac Cardiovasc Surg 1985;89: 170-82. 16. Lockerman ZS, Rose DM, Cunningham JN, Lichstein E. Reperfusion ventricular fibrillation during coronary artery bypass operations and its association with postoperative enzyme release. J Thorac Cardiovasc Surg 1987;93:247-52. 17. Kerber RE, Martins JB, Gasho ]A, Marcus ML, Grayzel J . Effect of direct current countershocks on regional myocardial contractility and perfusion. Circulation 1981;63:32332. 18. Doherty PW, McLaughlin PR, Billingham M, Kemoff R, Goris ML, Harrison DC. Cardiac damage produced by direct current countershock applied to the heart. Am J Cardiol 1973;43:22532. 19. Lake CL, Sellers TD, Nolan SP, Crosby IK, Wellons HA Jr, Crampton RS. Energy dose and other variables possibly affecting ventricular defibrillation during cardiac surgery. Anesth Analg 1984;63:743-51. 20. Weld FM, Bigga JT Jr. The effect of lidocaine on diastolic transmembrane currents determining pacemaker depolarization in cardiac Purkinje fiber. Circ Res 1976;38:203-8. 21. Carmeliet E, Saikawa T. Shortenings of the action potential and reduction of pacemaker activity by lidocaine, quinidine and procainamide in sheep cardiac purkinge fibers. An effect on Na and K currents? Circ Res 1982;50257-72. 22. Schinittger I, Griffin JC, Hall RJ, Meffin PJ, Winkle RA. Effects of tocainide on ventricular fibrillation threshold. Comparison with lidocaine. Am J Cardiol 1978;42:7@1.
Lidocaine addition to crystalloid cardioplegic solution for prevention of reperfusion ventricular fibrillation after the release of the aortic cross-clamp was studied in 50 patients undergoing coronary artery bypass grafting and in 30 patients undergoing mitral or aortic valve replacement. Twenty-six of the patients undergoing coronary artery bypass grafting received lidocaine, 100 mg/L of cardioplegia, whereas a control group of 24 patients received cardioplegia without lidocaine. In the group undergoing valve replacement, 14 patients received lidocaine cardioplegia and 16 patients served as control. In
the coronary artery bypass grafting group, lidocaine cardioplegia reduced significantly the incidence of reperfusion ventricular fibrillation from 100% to 42%. In the valve group, lidocaine cardioplegia also reduced significantly the incidence of reperfusion ventricular fibrillation from 93% to 42%. In both groups, lidocaine cardioplegia decreased the number of direct-current countershocks required to defibrillate the heart, with no significant increase in the incidence of high-grade atrioventricular block. (Ann Thorac Surg 1993;55:1529-33)
R
Material and Methods
eperfusion ventricular fibrillation (VF) is a major concern in patients undergoing open heart operations, despite the improvement in myocardial protection strategies. After the release of the aortic cross-clamp (ACC), VF is reported to occur in 74% to 96% of patients [1-4]. Reperfusion VF may result in increased myocardial wall tension, increased myocardial oxygen consumption, and impaired subendocardial blood flow [5, 61. Furthermore, defibrillation with direct-current (DC)countershock will result in myocardial injury [7]. Thus, it is advantageous to prevent the occurrence of reperfusion VF after release of the ACC. Previous reports have demonstrated in patients undergoing coronary artery bypass grafting (CABG) that addition of lidocaine to cold hyperkalemic crystalloid cardioplegia, in a concentration of 500 mg/L, decreased significantly the incidence of reperfusion VF; however, the incidence of high-grade atrioventricular (AV) block was markedly increased [3, 81. Using a lower concentration of lidocaine (100 mgL) decreased the incidence of reperfusion VF without increasing the incidence of heart block [4]. The purpose of this report is to assess the efficacy of lidocaine, 100 mgiL of hyperkalemic crystalloid cardioplegic solution, in reducing the incidence of reperfusion VF in patients undergoing CABG as compared with patients undergoing mitral or aortic valve replacement.
Accepted for publication Sep 30, 1992 Address reprint requests to Dr Baraka, Department of Anesthesiology, American University of Beirut, Beirut, Lebanon. 0 1993 by The Society of Thoracic Surgeons
Fifty patients undergoing elective CABG and 30 patients undergoing aortic or mitral valve replacement were included in the study. The investigation conformed to the ethical standards as set out in the Declaration of Helsinki, and informed consent was obtained from all patients. Patients were randomly assigned to receive either cold potassium cardioplegia without lidocaine (control group) or cold potassium cardioplegia with lidocaine, 100 mgL (lidocaine group). In patients undergoing CABG, 26 patients received lidocaine cardioplegia, and 24 patients served as controls. In patients undergoing valve replacement, 14 patients received lidocaine cardioplegia, and 16 patients served as controls. Patients were premedicated with morphine, 0.1 mg/kg; promethazine, 25 mg; and scopolamine, 0.3 mg intramuscularly. Anesthesia was induced with midazolam, 0.1 mgkg; fentanyl, 50 pgkg; and a mixture of vecuronium, 0.1 mg/kg, and pancuronium, 0.1 mg/kg. After tracheal intubation, ventilation was controlled with 100% oxygen with no inhalation anesthetic supplementation. Patients were monitored with an electrocardiogram (V5),radial artery cannula, and thermodilution Swan-Ganz pulmonary artery catheter. Cardiopulmonary bypass was initiated using a bubble oxygenator (Bentley 10B). Patients were perfused by a roller pump (Sarns 7000, Ann Arbor, MI) at a flow rate of 2.4 L min-' * m-*. Moderate systemic hypothermia, around 28"C, was induced during cardiopulmonary bypass. Cold potassium cardioplegic solution was infused at the aortic root or in the coronary ostia. The composition of the cardioplegic solution was as follows: KCl, 20 mEq/L; NaHCO,, 20 mEqL; mannitol, 20 g/L; and dextrose, 12 g/L in lactated Ringer's. Lidocaine, 100 mg, was added 0003-4975/93/$6.00
1530
Ann Thorac Surg 1993;55:1529-33
BARAKA ET AL LIDOCAINE CRYSTALLOID CARDIOPLEGIA
Table 1. Hernodynamic Data 20 Minutes After Induction of Anesthesia in Patients Undergoing Coronary Artery Bypass Grafting Variable HR (beatdmin) MAP (mm Hg) PCWP (mm Hg) CO (Wmin) CI (L min-’ . m-’)
-
Control Group Lidocaine Group P (n = 24) Value (n = 26)
-
68.0 f 9.7 93.2 f 9.2 10.2 f 3.7 2.9 f 1.0 1.68 f 0.46
62.7 f 11.6 90.4 f 10.6 10.5 f 5.0 3.0 t 0.87 1.72 k 0.44
NS NS NS NS NS
C1 = cardiac index; CO = cardiac output; HR = heart rate; M A P = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
to each liter of cardioplegia used in the lidocaine group of patients, whereas cardioplegic solution without lidocaine was used in the control group. The volume of cardioplegic solution initially infused was 10 mL/kg, followed by intermittent infusion every 20 minutes or whenever electrical activity reappeared. Patients‘ data consisted of age, sex, weight, and preoperative medications, as well as electrocardiographic, echocardiographic, and coronary angiographic findings. Intraoperative data included number of vessels bypassed in the CABG groups, bypass time, ACC time, degree of hypothermia, and total volume of cardioplegia used. At the time of the release of the ACC, we recorded central venous blood temperature and mean arterial blood pressure. Also, an arterial blood sample was analyzed for electrolytes and blood gases. The cardiac rhythm after unclamping of the aorta was recorded continuously until sinus rhythm took place, and the incidence of VF was compared in the control group versus the lidocaine group. Whenever VF followed the release of the ACC, DC countershocks of 5 J were used for defibrillation. Increments of 5 J were added when further shocks were needed. The need for a pacemaker was noted in each group. After weaning from cardiopulmonary bypass, inotropic support was provided whenever necessary by an infusion of dobutamine at a rate of 5 to 10 pg/kg per minute. The rate and duration of the infusion were subsequently
Table 2. lntraoperative Data in Patients Undergoing Coronary Artery Bypass Grafting Variable Number of grafts Bypass time (min) ACC time (min) Hypothermia (“C) Total cardioplegia volume per weight (mL/kg)
Control Group (n = 24)
Lidocaine Group p (n = 26) Value
2.54 f 0.66 65.5 t 19.1
2.92 f 0.74 69.4 f 20.1
NS NS
38.6 f 14.3 29.1 f 2.26 14.62 f 4.73
43.6 f 14.8 28.8 f 1.82 15.83 f 4.0
NS
ACC = aortic cross-clamp;
NS = not significant
NS
NS
adjusted according to the hemodynamic parameters of the patients. In all patients, the different hemodynamic parameters including cardiac output measurement by thermodilution were monitored 20 minutes after induction of anesthesia and 15 minutes after weaning from cardiopulmonary bypass. Data are expressed as absolute numbers or mean f standard deviation. The results are analyzed using the 2 analysis for discrete variables and by Student’s t test for continuous variables. Values of p less than 0.05 were considered significant.
Results Coronary Artery Bypass Graffing Group The control group and lidocaine group of patients were similar with respect to age, sex, preoperative ECG changes, left ventricular function, and preoperative medications. The mean age was 57.0 f 7.8 years in the control group and 56.6 f 10.9 years in the lidocaine group. Preoperative electrocardiograms showed evidence of old myocardial infarction in 11 patients of the control group and 13 patients of the lidocaine group. Preoperative pblockers were administered to 15 patients of the control group and 21 patients of the lidocaine group, whereas calciumchannel blockers were used in 18 patients of the control group and in 18 patients of the lidocaine group. There was no significant difference between the hemodynamic variables of the two groups determined 20 minutes after induction of anesthesia (Table 1).The number of vessels grafted, the bypass time, ACC time, degree of hypothermia, and the total cardioplegic volume used were not significantly different between the groups (Table 2). At the time of unclamping of the aorta, central venous blood temperature, arterial potassium concentration, oxygen tension, carbon dioxide tension, pH, and mean arterial blood pressure were also not significantly different (Table 3). DEMOGRAPHIC AND INTRAOPERATIVE DATA.
REPERFUSION VENTRICULAR FIBRILLATION. The incidence of reperfusion VF after release of the ACC was significantly reduced in the lidocaine group (Fig 1).Also, the number of DC countershocks required to defibrillate the heart was
Table 3. Conditions at Aortic Cross-Clamp Release in Patients Undergoing Coronary Artery Bypass Grafting Variable Temperature (“C) Potassium (mEqlL) PO, (mm Hg) PCOl (mm Hg) PH MAP (mm Hg)
Control Group Lidocaine Group p (n = 24) (n = 26) Value 33.8 f 1.49 4.56 f 0.65 375 f 119 30.3 f 5.1 7.52 f 0.049 91.0 f 14.1
33.9 f 1.8 4.58 f 0.54 404 f 109 28.7 f 2.8 7.54 f 0.041 88.8 f 13.6
NS = not significant; MAP = mean arterial blood pressure; POz = oxygen tension. = carbon dioxide tension;
NS NS
NS NS
NS NS PCO,
BARAKA ET AL LIDOCAINE CRYSTALLOID CARDIOPLEGIA
Ann Thorac Surg
1993:55:1529.33
1531
Table 5. Hernodynamic Data 20 Minutes After lnduction of Anesthesia in Patients Undergoing Valve Replacement
p=o.ooo1 CONTROL GROUP LIDOCAME CROUP
Variable
Control Group Lidocaine Group p (n = 16) (n = 14) Value
HR (beatslmin) MAP (mm Hg) PCWP (mm Hg) CO (L/min) CI (L * min-’ . m-2)
74.7 f 13.7 83.7 f 6.5 23.1 f 18.2 2.91 f 1.35 1.62 2 0.62
86.7 f 23.0 86.3 f 7.85 26.6 f 23.7 3.27 f 0.99 1.90 f 0.59
NS NS NS NS NS
CI = cardiac index; CO = cardiac output; HR = heart rate; MAP = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
VENTRICULAR FIBRILLATION Fig 1. Incidence of reperfusion ventricular fibrillation in the patients undergoing coronary artery bypass grafting. Ventricularfibrillation followed release of the aortic cross-clamp in all 24 control patients (ZOO%), as compared with 11 of the 26 patients of the lidocaine group (42.3% ).
significantly decreased from 2.0 & 0.75 in the control group to 1.27 f 0.44 in the lidocaine group. ATRIOVENTRICULAR BLOCK. Atrioventricular block requiring epicardial pacing occurred after release of the ACC in 9 of 24 in the control group and in 10 of 26 in the lidocaine group. There was no significant difference between the two groups. Among the patients in whom AV block developed, only 3 patients in the control group and 2 patients in the lidocaine group required continued pacing on discharge from the operating room.
HEMODYNAMIC PARAMETERS. The hemodynamic parameters measured 15 minutes after weaning from bypass were similar in the two groups (Table 4).
left ventricular function, and preoperative medications. The mean age was 44.3 f 11.9 years in the control group and 44.4 ? 15.9 years in the lidocaine group. Atrial fibrillation was present in 5 patients of the control group and 6 patients of the lidocaine group. Preoperative digitilization was used in 13 patients of the control group and 11 patients of the lidocaine group. There was no significant difference between the hemodynamic parameters of the two groups determined 20 minutes after induction of anesthesia (Table 5). Bypass time, ACC time, degree of hypothermia, and total cardioplegia volume were not significantly different between the groups (Table 6). At the time of unclamping of the aorta, central venous blood temperature, arterial potassium concentration, oxygen tension, carbon dioxide tension, pH, and mean arterial pressure were also not significantly different (Table 7). The incidence of reperfusion VF after release of the ACC was significantly reduced in the lidocaine group (Fig 2). Also, the number of DC countershocks required to defibrillate the heart was decreased from 2.79 k 1.69 in the control group to 1.26 k 0.54 in the lidocaine group. REPERFUSION VENTRICULAR FIBRILLATION.
The control group and the lidocaine group of patients were similar with respect to age, sex, preoperative atrial fibrillation,
Atrioventricular block requiring epicardial pacing occurred after release of the ACC in 7 of 16 in the control group and in 5 of 14 patients of the lidocaine group. There was no significant difference between the two groups. Among the patients in whom AV
Table 4. Hernodynamic Data 15 Minutes After Weaning Off Bypass in Patients Undergoing Coronary Artery Bypass Grafting
Table 6. lntraoperative Data in Patients Undergoing Valve Replacement
Valve Group
ATRIOVENTRICULAR BLOCK.
DEMOGRAPHIC AND INTRAOPERATIVE DATA.
Control Group Lidocaine Group p (n = 24) (n = 26) Value
Variable HR (beatdmin) MAP (mm Hg) PCWP (mm Hg) CO (Umin) CI (L * mir-’ m-’)
-
92.7 f 14 89.6 f 7.3 14.0 f 6.8 3.4 f 1.4 2.0 0.64
*
86.2 f 9.7 89.1 f 11.7 12.5 f 5.1 3.2 f 0.96 1.76 f 0.53
NS NS NS NS NS
CI = cardiac index; CO = cardiac output; HR = heart rate; M A P = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
Variable Bypass time Win) ACC time (min) Hypothermia (“C) Total cardioplegia volume per weight (mL/kg)
Control Group (n = 16)
Lidocaine Group (n = 14)
p Value
57.4 f 17.4
62.3
18.1
NS
42.3 f 13.0 28.1 f 0.88 16.25 f 6.6
42.9 f 15.7 28.7 f 1.4 13.4 5 3.3
NS NS NS
ACC = aortic cross-clamp;
NS = not Significant.
f
1532
BARAKAETAL LIDOCAINE CRYSTALLOID CARDIOPLECIA
Ann Thorac Surg 1993;55:1529-33
Table 7 . Conditions at Aortic Cross-Clamp Release in Patients Undergoing Valve Replacement
Table 8 . Hernodynamic Data 15 Minutes After Weaning Bypass in Patients Undergoing Valve Replacement
Control Group Lidocaine Group p (n = 16) (n = 14) Value
Variable
Variable
Control Group Lidocaine Group P (n -- 16) (n = 14) Value
~~~~
Temperature (“C) Potassium (mEq/L) PO2 (mm Hg) PCOz (mm Hg) PH Pressure (mm Hg)
35.2 & 1.7 4.13 2 0.56 344 f 135 32.3 f 7.1 7.51 f 0.08 80.6 2 14.8
34.6 f 1.05 4.15 f 0.603 415 2 108 29.5 f 2.3 7.52 f 0.03 81.0 f 13.6
NS NS NS NS NS NS
~~~~
NS = not significant; oxygen tension.
PCO, = carbon dioxide tension;
PO, =
block developed, only 1 patient in each group required continued pacing on discharge from the operating room. HEMODYNAMIC PARAMETERS. The hemodynamic parameters measured 15 minutes after weaning from bypass were similar in the two groups (Table 8).
Comparison of Bypass Grafting and Valve Operation The incidences of reperfusion VF and high-grade AV block were not significantly different between the control group of patients undergoing CABG and those undergoing valve replacement, or between the lidocaine group of patients undergoing CABG and those undergoing valve replacement. In both the valve and the CABG groups, lidocaine cardioplegia significantly decreased the incidence of reperfusion VF after release of the ACC, with no significant change in the incidence of pacemaker insertion.
loo-. 90 -.
93.77.
p=0.002
80-.
w
CONTROL CROUP
70-1
m L I D O C A I N E GROUP
2W 607: a
50- . 40-.
42.0%
30 _. 20-.
-
10 -.
VENTRICULAR FIBRILLATION Fig 2 . Incidence of reperfusion ventricular fibrillation in patients undergoing valve replacement. Ventricular fibrillation followed release of the aortic cross-clamp in 15 of the 16 control patients (93.7%),as compared with 6 of the 14 patients of the lidocaine group (42.8%).
HR (beatdmin) MAP (mm Hg) PCWP (mm Hg) CO (Umin) CI (L * min-’ m-’)
off
102.7 f 14.6 87.3 f 15.0 15.8 2 5.8 3.91 f 1.4 2.23 f 0.78
102.1 f 16.1 87.2 f 6.6 17.8 f 6.0 4.00 f 0.95 2.34 f 0.530
NS NS NS NS NS
CI = cardiac index; CO = cardiac output; HR = heart rate; MAP = mean arterial blood pressure; NS = not significant; PCWP = pulmonary capillary wedge pressure.
Comment Aortic cross-clamping is an important cause of myocardial damage during cardiopulmonary bypass; the magnitude of damage is determined by the duration of ischemia, and the extent of supplyldemand imbalance [9, 101. Suboptimal myocardial protection during ACC may result in a heterogenous population of cells that can be classified as irreversibly damaged, minimally damaged, or stunned [ l l ] . After release of the ACC, the heterogenous recovery of myocardial cells and the subsequent increase in reentry [12] and automaticity [13], as well as the possibility of reperfusion-induced injury [14], can result in a high incidence of reperfusion VF. Ventricular fibrillation has been shown to cause redistribution of coronary blood flow away from the subendocardium, and to result in myocardial ischemia and impaired ventricular performance, in the face of increased myocardial oxygen consumption [5, 61. This will result in intramyocardial acidosis [15] and release of creatine kinase MB postoperatively [16]. In an attempt to decrease the incidence of reperfusion VF after release of the ACC, lidocaine has been added to the cardioplegic solution in patients undergoing CABG (3, 4, 81. The present report demonstrates that lidocaine, 100 mg/L of cardioplegic solution, produces a significant reduction in the incidence of reperfusion VF in patients undergoing CABG as well as in patients undergoing valve replacement. Also, the addition of lidocaine to the cardioplegic solution decreases significantly the number of DC countershocks required to defibrillate the heart in both the CABG and the valve groups. The decrease in number of DC countershocks required to defibrillate is advantageous, because DC countershocks have been shown to cause myocardial necrosis [7], subepicardial contraction abnormalities [171, and marked dehiscence of the intercalated discs between damaged myocytes [18]. Myocardial damage due to direct defibrillation can be avoided by using energy levels lower than 30 Wattseconds, minimizing the number of shocks given, and allowing a few minutes between consecutive shocks [19]. Lidocaine, an amide local anesthetic, belongs to class IB antiarrhythmic drugs. It alters membrane conductance of cations by binding to sodium channels. Lidocaine decreases the slope of phase 4 depolarization and increases
Ann Thorac Surg 1993;55:1529-33
the diastolic electrical current threshold in Purkinje fibers [20, 211. Also, it increases the threshold for ventricular fibrillation, and the increase in threshold is related to plasma lidocaine concentration [22]. These properties can explain our findings showing that the addition of lidocaine to the cardioplegic solution decreases significantly the incidence of reperfusion VF after the release of the ACC, as well as the number of DC countershocks required to defibrillate the heart. Thus, lidocaine cardioplegia will minimize the myocardial damage after release of the ACC by decreasing the incidence of reperfusion VF as well as the energy and number of countershocks required for defibrillation. The increase in high-grade AV block observed by previous investigations when lidocaine, 500 mg/L of cardioplegic solution, has been used [3, 81 is not demonstrated in our study when a lidocaine dose of only 100 mg/L is used. The incidence of AV block in our patients who received lidocaine cardioplegia is not significantly different from that observed in the control group of patients, whether undergoing CABG or valve replacement. The different hemodynamic parameters after weaning from bypass were not significantly different between the lidocaine group and the control group. However, the early postoperative period is a rather unstable period, and hence these findings may not reflect the subsequent hemodynamic performance; recent studies have shown increased levels of creatine kinase MB isoenzymes in patients in whom reperfusion VF develops [16]. In conclusion, the present report demonstrates that lidocaine addition to potassium cardioplegic solution in a dose of 100 mg/L reduces significantly the incidence of reperfusion VF after release of the ACC in patients undergoing CABG as well as in patients undergoing valve replacement. Also, lidocaine cardioplegia decreases in both groups the number of DC countershocks required to defibrillate the heart, without increasing the incidence of high-grade AV block.
References Fall SM, Burton NA, Graeber GM, et al. Prevention of ventricular fibrillation after myocardial revascularization. Ann Thorac Surg 1987;43:1824. Curling PE, Nagle D, Zeidan JR, Kaplan JA. Effects of lidocaine, propranolol and bretylium on reperfusion dysrhythmias. Anesthesiology 1981;55:A52. Tchervenkov CI, Symes JF, Wynands JE, et al. Lidocaine as an adjunct to high potassium cardioplegia: a prospective randomized clinical trial. Surg Forum 1983;34:319-21. Fiore AC, Naunheim KS, Taub J, et al. Myocardial preservation using lidocaine blood cardioplegia. Ann Thorac Surg 1990;50:771-5.
BARAKA ET AL LIDOCAINE CRYSTALLOID CARDIOPLEGIA
1533
5. Hottenrott C, Maloney JV, Buckberg G. Studies of the effects of ventricular fibrillation on the adequacy of regional myocardial flow 111. J Thorac Cardiovasc Surg 1974;68:634-45. 6. Buckberg GD, Hottenrott CE. Ventricular fibrillation: its effect on myocardial flow, distribution and performance. Ann Thorac Surg 1975;20:76-85. 7. Dahl CF, Ewy GA, Warner ED, Thomas ED. Myocardial necrosis from direct current countershock: effect of paddle electrode size and time interval between discharges. Circulation 1974;50:956-61. 8. Tchervenkov CZ, Wynands JE, Symes JF, Malcolm ID, Dobell ARC, Morin JE. Electrical behavior of the heart following high-potassium cardioplegia. Ann Thorac Surg 1983;36: 3149. 9. Iyengar SRK, Ramchand S, Charrette EJP, lyengar CKS, Lynn RB. Anoxic cardiac arrest: an experimental and clinical study of its effects. J Thorac Cardiovasc Surg 1973;66:722-30. 10. Hearse DJ, Garlick PB, Humphrey SM. Ischemic contracture of the myocardium: mechanisms and prevention. Am J Cardiol 1977;39:98&93. 11. Braunwald E, Honer RA. The stunned myocardium: prolonged, postischemic ventricular function. Circulation 1982; 66114G9. 12. Murdock D, Loeb JM, Euler DE, Randall WC. Electrophysiology of coronary reperfusion and mechanism for reperfusion arrhythmias. Circulation 1980;61:175-82. 13. Kaplinsky E, Ogawa S, Michelson EL, Dreifus LS. Instantaneous and delayed ventricular arrhythmias after reperfusion of acutely ischemic myocardium: evidence for multiple mechanisms. Circulation 1981;63:333-40. 14. Hearse DJ. Reperfusion of the ischemic myocardium. J Mol Cell Cardiol 1977;9:605-16. 15. Khuri SF, Maiston WA, Josa M, et al. Observations on 100 patients with continuous intraoperative monitoring of intramyocardial pH: the adverse effects of ventricular fibrillation and reperfusion. J Thorac Cardiovasc Surg 1985;89: 170-82. 16. Lockerman ZS, Rose DM, Cunningham JN, Lichstein E. Reperfusion ventricular fibrillation during coronary artery bypass operations and its association with postoperative enzyme release. J Thorac Cardiovasc Surg 1987;93:247-52. 17. Kerber RE, Martins JB, Gasho ]A, Marcus ML, Grayzel J . Effect of direct current countershocks on regional myocardial contractility and perfusion. Circulation 1981;63:32332. 18. Doherty PW, McLaughlin PR, Billingham M, Kemoff R, Goris ML, Harrison DC. Cardiac damage produced by direct current countershock applied to the heart. Am J Cardiol 1973;43:22532. 19. Lake CL, Sellers TD, Nolan SP, Crosby IK, Wellons HA Jr, Crampton RS. Energy dose and other variables possibly affecting ventricular defibrillation during cardiac surgery. Anesth Analg 1984;63:743-51. 20. Weld FM, Bigga JT Jr. The effect of lidocaine on diastolic transmembrane currents determining pacemaker depolarization in cardiac Purkinje fiber. Circ Res 1976;38:203-8. 21. Carmeliet E, Saikawa T. Shortenings of the action potential and reduction of pacemaker activity by lidocaine, quinidine and procainamide in sheep cardiac purkinge fibers. An effect on Na and K currents? Circ Res 1982;50257-72. 22. Schinittger I, Griffin JC, Hall RJ, Meffin PJ, Winkle RA. Effects of tocainide on ventricular fibrillation threshold. Comparison with lidocaine. Am J Cardiol 1978;42:7@1.