for R&fusions in Multidose Coldbrdioplegia?

Transcription

1 Is Hi& Potassium Solution Necessarv for R&fusions in Multidose Coldbrdioplegia? A Randomized Prospective Study Using Computerized Holter System Michael Dewar, M.D., Michael D. Rosengarten, M.D., Roger Samson, C.P., Carin Wittnich, D.V.M., Peter E. Blundell, M.D., and Ray C.-J. Chiu, M.D., Ph.D. ABSTRACT Multidose potassium cardioplegia is a common method of myocardial preservation. Although initial potassium arrest conserves high-energy phosphates, there is conflicting evidence that repeat high potassium boluses augment this protection. Fifty-six patients were prospectively randomized to receive multidose cold high potassium cardioplegia (27 meq of KCl/L) both in the initial and subsequent infusions (Group ) or an initial cold high potassium (27 meqll) cardioplegia followed by boluses of cold low potassium (7 meq, of KCVL) solution (Group 2). The two groups were compared in terms of postoperative myocardial electrical stability and hemodynamic performance. Electrocardiograms were recorded by continuous Holter monitor, and the data were analyzed by computer. The duration of aortic cross-clamping and cardiopulmonary bypass did not differ between groups. Group, who received more total KCl than Group 2 (p <.005), experienced more high-grade ventricular ectopia during both reperfusion (p <.00) and the immediate postoperative period (p <.00), and required more lidocaine hydrochloride (p <.00) for arrhythmias. There was no significant difference in hemodynamic performance between the two groups. This study fails to show an advantage to multidose high potassium cardioplegia and found a significant increase in ventricular ectopia associated with its use. We advocate using low potassium solutions after initial cold high potassium arrest. The two principles of myocardial preservation that have become generally accepted in recent years are hypothermia and the chemical induction of diastolic arrest. There is abundant proof that the initial bolus of high potassium solution causes rapid diastolic arrest of the myocardium and conserves high-energy phosphates. Although it is believed that multidose cardioplegia using the same high potassium solution augments this conservation, From the Divisions of Cardiovascular and Thoracic Surgery and of Cardiology, Montreal General Hospital and McCill University, Montreal, Que, Canada. Accepted for publication July 5, 986. Address reprint requests to Dr. Chiu, The Montreal General Hospital, 650 Cedar Ave, Room 97, Montreal, Que, Canada H3G A. several investigators have obtained contradictory results. Although some have claimed benefits of better ultrastructural preservation [ and improved postoperative performance [2], others have found no benefit [3] and have even documented detrimental effects of this regimen on the conducting system [, 5, as well as an increased incidence of dysrhythmia [6] and decreased levels of adenosine triphosphate (ATP) after reperfusion [7. The literature fails to show whether such supposed benefits of multidose cardioplegia were derived from the high potassium solution, maintenance of hypothermia, or simply better washout of metabolites of anaerobic glycolysis. Theoretically, once the heart has been rapidly arrested and cooled by the initial infusion of hypothermic high potassium solution, there is no obvious advantage in repeatedly infusing high concentrations of potassium, because adequate myocardial hypothermia alone can maintain the electromechanical arrest of the heart. To test this hypothesis, we prospectively randomized a series of 56 patients into two groups. One group received multidose high potassium cardioplegia, and the other group was given an initial single dose of high potassium cardioplegia followed by subsequent doses of cold low potassium crystalloid solution. The two groups were evaluated for cardiac hemodynamic performance and electrical activity. The latter was recorded continuously using a Holter monitor with computerized data analyses. Material and Methods The study comprised 56 adult patients, 36 having coronary artery bypass grafting (CABG), 3 having CABG with valve replacement, 2 having CABG with aneurysmectomy, undergoing valve replacements, and 2 having repair of an atrial septa defect. The patients were randomized prospectively into two groups. Group received hypothermic ( C) high potassium (27 meq of KCVL) cardioplegic solution immediately after aortic cross-clamping and received additional boluses of the same high potassium solution to maintain adequate myocardial hypothermia as described in detail later. Group 2 received high potassium (27 m Eq of KCVL) hypothermic cardioplegia only as the initial bolus to arrest the heart. If any further cooling was required, C 09 Ann Thorac Surg 3:09-5, Apr 987

2 0 The Annals of Thoracic Surgery Vol 3 No April 987 Ionosol (7 meq of KCVL) was infused. The two groups were randomized by assigning each patient a cipher in a statistical random number table, odd numbers going into Group and even numbers into Group 2. All the members of the surgical team, the postoperative intensive care unit team, and the electrocardiogram (ECG) technician who interpreted the ECG data were ignorant of the randomization. Only the perfusionist (R. S.) knew to which group the patient had been assigned. Eight patients were excluded from the study. Five were eliminated because of various malfunctions of the ECG Holter monitor which precluded obtaining precise data. One patient (Group 2) died 2 hours after operation, which made the data collected inadequate. Two patients were excluded because they required repeat crossclamping of the aorta and reinjection of high potassium cardioplegia. The remaining 8 patients were analyzed in terms of preoperative status, intraoperative conditions, and postoperative performance. The criteria evaluated include New York Heart Association (NYHA) Functional Class, gender, age, clinical or ECG evidence of previous myocardial infarction, left ventricular hypertrophy, preexisting dysrhythmia, and regular use of beta-blocking or calcium channel-blocking medications. Beta-blocking medications were given the morning of the operation if the patient had these drugs normally prescribed. Calcium channel-blocking drugs were stopped the night before operation. The anesthetic protocol was identical to one described previously [S]. The operation was performed using a midline sternotomy, cardiopulmonary bypass (CPB) with a Shiley bubble oxygenator and systemic hypothermia (28" to 30 C). The heart was arrested with an injection of C crystalloid high potassium (27 meq of KCl/L) cardioplegic solution into the coronary root or coronary ostia. The detailed composition of this crystalloid cardioplegic solution has been previously described [9]. Myocardial temperatures were recorded with a thermistor needle probe inserted into the middle zone of the myocardial mass, often in the territory of the coronary artery with high-grade obstruction, because this is the segment most likely to remain warmer [9]. Supplementary topical cooling is routinely used. Repeat doses of high potassium cardioplegia (Group ) or Ionosol (Group 2) were injected into the coronary artery bypass grafts after the completion of the distal anastomoses or into the native coronary circulation to maintain intramyocardial temperatures at or less than 5 C. Intraoperatively, the period of aortic cross-clamping and total CPB, the total amount of cardioplegic solution given, and the total milliequivalents of KC given were recorded, as were the intramyocardial temperature, the requirements for vasopressor agents, and the need for electrical defibrillation of the heart following myocardial reperfusion. Postoperatively, patients were followed for an average of 6 hours. Cardiac output at hour and 6 hours postoperatively was measured for comparison. Requirements for inotropic and antiarrhythmic agents were recorded. The surgeons and anesthesiologists who prescribed these medications were blinded concerning the group to which the patients had been randomized. Blood samples were taken for fraction analysis of the myocardial-specific isoenzyme of creatine kinase (CK- MB) using the Gelman CK isoenzyme electrophoresis system 6 hours following operation. A 2-lead electrocardiogram was obtained 8 hours after operation and compared with one made the day before operation. On the morning of operation, all patients had a Holter monitor applied, which recorded all cardiac electrical activity for 2 hours. Our electrode system to enable continuous recording during and after cardiac operation and the method of analyzing the massive quantity of data obtained have been reported previously [8]. The clock channel of the Holter system allowed us to precisely correlate data concerning cardiac rhythms to the events in the perioperative period. The periods of interest in this study were divided into the period of aortic cross-clamping or CPB, especially that of the reperfusion period, the elapsed time from release of the aortic crossclamp until a stable rhythm was obtained, and the immediate postoperative period of 6 hours' duration. The results were recorded on a standard Holter cassette and were analyzed by our Holter computer system. This system is composed of a tape playback (Oxford Medilog), which replays the tape at sixty times the speed it was recorded, and the detector unit (Oxford), which is interfaced to a computer. The computer was programmed by the Biomedical Engineering Department of McGill University. As described previously [8], computer numerical counts of ectopic activity and the isoprobability heart rate trend were used as guides to the physician interpreting the tapes (Figure). Serious dysrhythmias were defined for this study as supraventricular tachycardia, paroxysmal or sustained, that the patient had not demonstrated preoperatively, and onset of ventricular tachycardia, ventricular fibrillation, or high-grade premature ventricular contractions, Lown Grade IV [lo,. The randomization was carried out prior to the operation. Consequently, if a patient received only one dose of cardioplegia, he remained randomized to the assigned group, regardless of the fact that retrospectively the groups appear indistinguisable from one another because no reinfusion took place. This is required by the prospective randomized nature of the experimental protocol. The groups were compared to ascertain their similarities with respect to preoperative, intraoperative, and postoperative variables. The data were compared statistically by chi-square or t test as appropriate. Results Preoperative Status The two groups did not differ in regard to their stratification according to NYHA Functional Class (Table ). More Group patients received beta-blocking medications, but the groups did not differ in the number of patients taking calcium channel blockers. Group 2 had

3 Dewar, Rosengarten, Samson, et al: High KCL Solution in Multidose Cold Cardioplegia minut- inbualm x 20-5X 00-!30x 88)( 80- BBf B (A) Computer analysis of continuous ECG monitor: heart rhythm ove~ the 2-hour period of the operative day. The horizontal lines are isoprobability heart rate lines showing the percentage of beats that fall below a particular number of beats per minute. For example, the 5% line at 0600 shows 95% of beats were slower than a rate of 20 beats per minute, whereas at 0700, 95% of beats were slower than a rate of 00 beats per minute. Convergence of the isoprobability lines indicates a regular heart rhythm; divergence indicates irregularity. The arrow at 0600 to 0700 indicates an obvious change. ECG printout was obtained. The vertical strokes between 300 and 700 are artifacts introduced by electrocautery during operation. The gap in the tracing from 00 to 530 is the period of electromechanical arrest during aortic cross-clamping. Numerical readouts of the heart rhythm for any particular period can also be obtained from the computer, which will enumerate the number of couplets, salvos, and tachycardic episodes. (B) ECG printout from computer concerning period in question at This reading at 060 shows premature ventricular contractions, Lown Grade IV 0, Ill. significantly more patients with an existing dysrhythmia, but was not distinct from Group in the number of patients with evidence of a previous myocardial infarction or left ventricular hypertrophy. lntraoperative Data The two groups differed in that more patients undergoing valve replacement were randomized to Group 2 (Table 2). There were no significant differences regarding the duration of aortic cross-clamping or CPB, the number of patients with atrial activity during aortic cross-clamping, or the need for repeat boluses of cardioplegic solution. Group did not receive a greater volume of cardioplegic solution than Group 2, but did receive more KC (p <.005). There was no difference concerning elapsed time to stable rhythm, though Group 2 did require defibrillation more frequently than Group ( p <.00). Only patients who did not require defibrillation were included in the calculation of elapsed time to stable rhythm during myocardial reperfusion. There were more patients in Group who were difficult to wean from CPB (i.e., more than two attempts to discontinue CPB). There was no difference in the need for intraoperative pressor agents. Table I. Preoperative Status Group Group 2 Variable (N = 28) (N = 20) pvalue NYHA Class I 3 I 0 7 I 3 0 IV 2 2 Medications Beta blockers Calcium channel blockers 8 Electrocardiogram Dysrhythmia Myocardial infarction 2 6 Left ventricular hypertrophy 3 5 NYHA = New York Heart Association; = not significant.

4 2 The Annals of Thoracic Surgery Vol 3 No April 987 Table 2. lntraoperative Data Group Group 2 Variable (N = 28) (N = 20) p Value Procedure CABG CABG + valve replacement Valve replacement Miscellaneous Aortic cross-clamping (min) CPB (min) Lowest temperature ( C) Total amount of cardioplegia (ml) Total KCI (meq) Multidose cardioplegic solution required Requirement for defibrillation after DC aortic cross-clamping Elapsed time from DC aortic cross-clamping to stable condition Difficulty weaning from CPB Requirement for intraoperative pressor agents Atrial activity during aortic cross-clamping s ? c ? c s s c 9.7 0,00,005,00.02 CABG = coronary artery bypass grafting; = not significant; CPB = cardiopulmonary bypass; DC = direct-current. Table 3. Results after CPB and after Operation Group Group 2 Variable (N = 28) (N = 20) p Value Ventricular performance Cardiac output hour (Wmin) Cardiac output 6 hours (Wmin) Requirements for inotropic agents Requirement for antiarrhythmic agents Electrical stability Reperfusion period 3.9 c t c.3 6 5,00 SVT VTNF PVC Postoperative period SVT VTNF PVC ,00.00 Evidence of periop infarction Significant change in postop electrocardiogram 2 3 CK-MB (6 hr postop) 56.0? CPB = Cardiopulmonary bypass; = not significant; SVT = supravenmcular tachycardia; VTNF = ventricular tachycardialventricular fibrillation; PVL = premature ventricular contraction; CK-MB = myocardial-specific isoenzyme of creatine kinase. Results after CPB and after Operation Although Group 2 received less inotropic medication postoperatively and had slightly lower cardiac outputs at and 6 hours after operation, these differences were not significant (Table 3). Group, however, received more lidocaine hydrochloride postoperatively than the Ionosol group (p <.00). Group exhibited more ventricular ectopia than Group 2 in the reperfusion period (p <.00). The increased incidence of supraventricular tachycardia or of ventricular tachycardia or ventricular fibrillation, however, was not significant. The increased incidence of electrical instability in Group continued in the immediate postoperative period though only the rate of premature ventricular contractions achieved statistical significance. Postoperatively, neither group exhibited any difference in evidence (electrocardiogram and CK- MB) of having sustained a perioperative myocardial infarction or ischemic injury.

5 3 Dewar, Rosengarten, Samson, et al: High KCL Solution in Multidose Cold Cardioplegia Comment The quest for the optimal technique of myocardial preservation continues to be of great importance. The advantages of a bolus of high-concentration potassium to induce cardioplegia have been frequently documented since Gay and Ebert [2] revived the abandoned technique of Melrose by combining it with hypothermia. The proven advantages are decreased oxygen consumption [ 3, rapid cessation of electrical activity [2], prevention of ischemic contracture [], decreased tissue impedance indicative of wekpreserved myocardium [ 5, conservation of high-energy phosphate stores even at normal temperatures [2], and synergistic myocardial protection when combined with hypothermia [6]. The rationale for the benefit of the initial dose of potassium seems well grounded. Multidose high potassium cardioplegia, a quite common practice, however, is more controversial. For example, claims that multidose potassium cardioplegia is superior to simple hypothermia [l] fail to distinguish whether the washout effect, the hypothermia, or the multidose potassium was responsible for the added protection. Likewise, although many investigators [l, 7-2] published studies showing that multidose potassium cardioplegia is superior to single-dose cardioplegia, it is also difficult to determine whether this improvement is due to the greater amount of potassium given, better maintenance of hypothermia, or the washout of metabolites [22]. Indeed, Tucker and associates [3] questioned the benefit of high potassium ion concentration and attributed most of the benefit to hypothermia, a point of view supported by Shumway [23], Griepp [2], Jacocks [25], and their co-workers. Engelman and colleagues [26] initially failed to show any major benefit to cardioplegic arrest, though subsequently he and his associates [27] have used increasing amounts of potassium cardioplegia. The recent study by Kao and associates [7] is of particular interest because they showed both single- and double-dose cold potassium cardioplegia to be superior to cold multidose potassium as far as preservation of ATP stores after the reperfusion period was concerned. There have been few adequate clinical studies to compare the use of an initial cold hyperkalemic arrest followed by reinfusions of cold isotonic solution against the commonly used multidose high potassium cardioplegia. In our randomized prospective study, we scrutinized carefully the electrical behavior of the myocardium to determine whether the addition of potassium to cardioplegia reinfusions subsequent to the initial diastolic arrest is not, in reality, the addition of an ingredient of uncertain benefit that only complicates an already complicated problem. Routine indexes of myocardial hemodynamic performance showed that our Group required as much inotropic support as Group 2 and recorded no significant difference in cardiac output and no decrease in the incidence of perioperative infarction. More patients in Group 2 did receive direct-current defibrillation after removal of the aortic cross-clamp. However, with our criteria we were unable to document that this early defibrillation (0 J) on reperfusion caused any severe damage to the myocardium. Ferguson and associates [28] were also unable to determine whether the early return of low-level electrical activity, which was often the reason for the application of direct-current countershock, was detrimental to the postoperative course. The effect of hyperkalemic cardioplegia on the conduction system has always been suspect, especially in the first few hours after exposure. It causes important atrioventricular block [, 29, 30, nodal tachycardia [3], and bundle-branch blocks [32]. The prolonged atrioventricular block [, 3, 33 that can occur has led some to abandon potassium in secondary cardioplegia [3] or in multidose cardioplegia [5, 35. Smith [36], Magilligan [37], and their colleagues maintained that potassium cardioplegia does not protect the conducting system, and others [28,38-0 noticed that its paralytic effect is shortlived if adequate hypothermia is not maintained. Our study of the myocardial electrical stability of these patients exposed to multidose high potassium cardioplegia focused on the intraoperative reperfusion and early postoperative periods. Compared with the usual studies of opertive dysrhythmias, the Holter monitor technique, which we adopted for perioperative monitoring, is extremely sensitive to ventricular activity and is made very specific by computer analysis corroborated by an electrocardiogram technician. Even when observer vigilance is maximum, a human observer cannot compete with the use of a beat-by-beat monitoring device that gives a quantitative and reproducible hard copy of the electrical history of the ventricle [8, 37, -3]. Despite the prospective randomization in this study because of the relatively small sample size, our analysis suggests that there were several factors that should have biased the study in favor of Group. There were fewer patients having valve replacement in that group. Such patients are known to have an increased incidence of dysrhythmias [8,, 5. Group also had a higher incidence of preoperative beta blockade, which should have contributed to electrical stability [6-9. Finally, Group patients had fewer preoperative dysrhythmias. Despite these advantages, Group demonstrated significantly more ventricular ectopia and required more frequent use of intravenous administration of lidocaine during the critical immediate postoperative period. It would have been of some interest to correlate the serum potassium levels with the occurrence of arrhythmias. However, the design of this study was such that dysrhythmias were analyzed only after the fact, and this made the simultaneous serum potassium sampling impossible. Although the exact mechanism is uncertain, the increased ectopia may be related to the accumulation of potassium in marginal ischemic tissue [50] or the sequestration of inordinate amounts of calcium intracellularly, a process augmented by increased extracellular potassium [5], thereby allowing for the possibility of ventricular reentry rhythms (33, 52. To dismiss these ventricular dysrhythmias as trivial could be dangerous

6 The Annals of Thoracic Surgery Vol 3 No April 987 because the report of Kron and associates [53] showed the ominous prognosis for patients with malignant dysrhythmias after CABG. There is a small but known risk of hyperkalemia in some patients [5] given high potassium cardioplegic solutions. There are also studies that indicate that hyperkalemic solutions can accelerate venous graft thrombosis [55]. Because hyperkalemic cardioplegia fails to protect and may, in fact, impair the conducting system, it appears imperative that some advantage be clearly shown to justify the continued use of multidose high potassium cardioplegia. Our study failed to show any functional difference between our groups and demonstrated that hearts that received multidose high potassium cardioplegia had increased ventricular ectopia and required more antiarrhythmic medications. We advocate, therefore, using high-concentration potassium only in the initial bolus of cardioplegia or when obvious electromechanical activity of the heart has returned prematurely during aortic cross-clamping. This work was supported by a grant from the Quebec Heart Foundation. References. Lucas SK, Elmer EB, Flaherty JT: Effects of multidose potassium cardioplegia on myocardial ischemia, return of ventricular function and ultrastructural preservation. J Thorac Cardiovasc Surg 80:02, Conti VR, Bertranou EG, Blackstone EH, et al: Cold cardioplegia versus hypothermia for myocardial protection. J Thorac Cardiovasc Surg 76:577, Tucker WY, Ellis RJ, Mangano DT, et al: Questionable importance of high potassium concentrations in cardioplegic solutions. J Thorac Cardiovasc Surg 7783, 979. Ellis RJ, Mavroudis C, Gardner C: Relationship between atrioventricular arrhythmias and the concentration of K+ ion in cardioplegc solution. J Thorac Cardiovasc Surg 80:57, Kopman EA, Ramirez-Inawat RC: Persistent electromechanical cardiac arrest following administration of cardioplegic and glucose-insulin-potassium solutions. Anesth Analg (Cleve) 59:69, Engelman RM, Rousou JH, Vertrees RA, et al: Safety of prolonged ischemic arrest using hypothermic cardioplegia. J Thorac Cardiovasc Surg 79:705, Kao RL, McGovern GJ, Chen JY, et al: Prevention of reperfusional damage from ischemic myocardium. J Thorac Cardiovasc surg 9:06, Dewar ML, Rosengarten MD, Blundell PE, et al: Perioperative Holter monitoring and computer analysis of dysrhythmias in cardiac surgery. Chest 87593, Chiu RCJ, Blundell PE, Scott HJ, Cain S: The importance of monitoring intramyocardial temperature during hypothermic myocardial protection. Ann Thorac Surg 28:37, Lown B, Graboys T: Management of patients with ventricular arrhythmia. Am J Cardiol 39:90, 977. Lown 8, Wolf M: Approaches to sudden death from coronary heart disease. Circulation :30, Gay WA, Ebert PA: Functional, metabolic and morphologic effects of potassium-induced cardioplegia. Surgery 7:28, Brandt B, Richardson JV, OBryan P, Ehrenhaft JL: Intramyocardial electrical and metabolic activity during hypothermia and potassium cardioplegia. Ann Thorac Surg 3:7, 98. Freedman BM, Pasque MK, Pellom GL, et al: Effects of delay in administration of potassium cardioplegia to the isolated rat heart. Ann Thorac Surg 37:309, Garrido H, Sueiro J, Rivas J, et al: Bioelectrical tissue resistance during various methods of myocardial preservation. Ann Thorac Surg 36:3, Tyers GFO, Manley NJ, Williams EH: Preliminary clinical experience with isotonic hypothermic potassium-induced arrest. J Thorac Cardiovasc Surg 7:67, Nelson RL, Fey KH, Follette DM, et al: The critical importance of intermittent infusion of cardioplegic solution during aortic cross clamping. Surg Forum 272, Engelman RM, Avril J, ODonoghue MJ, et al: The significance of multidose cardioplegia and hypothermia in myocardial preservation during ischemic arrest. J Thorac Cardiovasc Surg 75:555, Flaherty JT, Schaffy HV, Goldman RA, et al: Metabolic and functional effects of progressive degrees of hypothermia during global ischemia. Am J Physiol 236:H839, Fremes SE, Weisel RD, Mickle DAG: Myocardial metabolism and ventricular function following cold potassium cardioplegia. J Thorac Cardiovasc Surg 89:53, Molina JE, Feiber W, Sisk A, et al: Cardioplegia without fibrillation or defibrillation in cardiac surgery. Surgery 8:69, Grover FL, Fewel JG, Ghidoni JJ, et al: Is potassium a necessary component of cardioplegic solution? J Surg Res 29:62, Shumway NE, Lower RR, Stofer RC: Selective hypothermia of the heart in anoxic cardiac arrest. Surg Gynecol Obstet 09:750, Griepp RB, Stinson EB, Shumway NE: Profound local hypothermia for myocardial protection during open heart surgery. J Thorac Cardiovasc Surg 66:73, Jacocks MA, Fowler BN, Chaffin JS, et al: Hypothermic ischemic arrest versus hypothermic potassium cardioplegia in human beings. Ann Thorac Surg 3:57, Engelman RM, Baumann G, Boyd AD, Kaplan F: Myocardial injury associated with potassium arrest. Ann Thorac Surg 22:557, Engelman F M, Rousou JH, Vertrees RA, et al: High volume crystalloid cardioplegia. J Thorac Cardiovasc Surg 86937, Ferguson TB, Smith PK, Buhrman WC: Studies on the physiology of the conducting system during hyperkalemic, hypothermic cardioplegic arrest. Surg Forum 3:302, Fisch C, Greenspan K, Edmonds RE: Complete atrioventricular block due to potassium. Circ Res 9:373, Fisch C, Feigenbaum H, Bowers JA: The effect of potassium on atrioventricular conduction of normal dogs. Am J Cardiol :87, Fisch C, Feigenbaum H, Bowers J: Nonparoxysmal AV nodal tachycardia due to potassium. Am J Cardiol :357, OConnell JB, Wallis D, Johnson SA, et al: Transient bundle branch block following use of hypothermic cardioplegia in coronary artery bypass surgery: high incidence without perioperative myocardial infarction. Am Heart J 00:85, 982

7 5 Dewar, Rosengarten, Samson, et al: High KCL Solution in Multidose Cold Cardioplegia 33. Cohen HC, Gozo EG, Pick A: The nature and type of arrhythmias in acute experimental hyperkalemia in the intact dog. Am Heart J 82:777, Roberts AJ, Far0 RS, Knauf DG: Coronary artery bypass graft surgery: relative efficacy of 2 C cardioplegia. J Surg Res 37, Weber DO, Yamoz MD: Hyperkalemia complicating cardiopulmonary bypass: analysis of risk factors. Ann Thorac Surg 3:39, Smith PK, Buhrman WC, Ferguson TB: Conduction block after cardioplegic arrest: prevention by augmented atrial hypothermia. Circulation 68:, Magdligan DJ Jr, Vij D, Peper W, et al: Failure of standard cardioplegic techniques to protect the conducting system. Ann Thorac Surg 39:03, Novick RJ, Stefaniszyn HJ, Morin JE, et al: Atrial electrical activity and its suppression during cardioplegic arrest in pigs. J Thorac Cardiovasc Surg 86:235, Novick RJ, Stefaniszyn HJ, Malcolm ID, et al: Differential electrical activity of the atria during cardioplegic arrest in pigs. Ann Thorac Surg 37:5, Tchervenkov CI, Wynands JE, Symes JF, et al: Electrical behavior of the heart following high-potassium cardioplegia. Ann Thorac Surg 36:3, 983. Michelson E, Morganroth J, MacVaugh H: Post-operative arrhythmias after coronary artery and cardiac valvular surgery detected by long-term electrocardiographic monitoring. Am Heart J 972, Berman ND, Walker BM: Frequency and significance of unrecognized ventricular arrhythmias after coronary artery bypass grafting. J Clin Invest Med 5:28, Cairns J, Kumana C, Rambihar P, et al: Insensitivity of conventional monitoring in the detection of ventricular arrhythmias in patients with acute myocardial infarction. J Clin Invest Med 5:28, 982. Smith R, Grossman W, Johnson L, et al: Arrhythmias following cardiac valve replacement. Circulation 5:08, Angelini P, Feldman M, Lufschanowski R, et al: Cardiac arrhythmia during and after heart surgery: diagnosis and management. Prog Cardiovasc Dis 6;69, Hammon JW Jr, Wood AJJ, Prager RL, et al: Perioperative beta blockade with propranolol: reduction in myocardial oxygen demands and incidence of atrial and ventricular arrhythmias. Ann Thorac Surg 38:363, Roffman JA, Fieldman A: Digoxin and propranolol in the prophylaxis of supraventricular tachydysrhythmias after coronary artery bypass surgery. Ann Thorac Surg 3:96, Stephenson LW, MacVaugh H, Tomasello DN, Josephson ME: Propranolol for prevention of postoperative cardiac arrhythmias: a randomized study. Ann Thorac Surg 29:3, Silverman NA, Wright R, Levitsky S: Efficacy of low-dose propranolol in preventing postoperative supraventricular tachyarrhythmia. Ann Surg 96:9, Kleber AG: Extracellular potassium accumulation in acute myocardial ischemia. J Mol Cell Cardiol 6:389, Fenton RA, Bruttig SP, Rubio R, et al: Effect of adenosine on calcium uptake by intact and cultured vascular smooth muscle. Am J Physiol 22:H797, Janse MJ, van Capelle FJL, Morsink H, et al: Flow of injury current and patterns of excitation during early ventricular arrhythmias in acute regional myocardial ischemia in isolated porcine and canine hearts: evidence for two different arrhythmogenic mechanisms. Circ Res 7:5, Kron IL, DiMarco JP, Harman PK, et al: Unanticipated postoperative ventricular tachyarrhythmias. Ann Thorac Surg 38:37, Kopman EA, Ferguson TB: Scavenging of cardioplegic solution from right heart to prevent hyperkalemia. J Thorac Cardiovasc Surg 86:53, DeCaterina R, Weksler BB, Alonso DR, et al: Functional endothelial damage by high potassium cardioplegic solutions to saphenous vein bypass grafts. Surgery 98:65,985