Efficacy and safety of timolol for prevention of supraventricular tachyarrhythmias after coronary artery bypass surgery

Transcription

1 THERAPY AND PREVENTION Il-ADRENERGIC BLOCKADE Efficacy and safety of timolol for prevention of supraventricular tachyarrhythmias after coronary artery bypass surgery HARVEY D. WHITE, M.B., CH.B., ELLIOTT M. ANTMAN, M.D., MARY ANN GLYNN, R.N., JOHN J. COLLINS, M.D., LAWRENCE H. COHN, M.D., RICHARD J. SHEMIN, M.D., AND PETER L. FRIEDMAN, M.D., PH.D. Downloaded from by on September 18, 2018 ABSTRACT Forty-one patients undergoing coronary artery bypass grafting were randomly assigned to receive prophylactic timolol or placebo, given in a double-blind fashion. /3-Adrenoceptor-blocking therapy was stopped at least one half-life before surgery. Three to 7 hr after surgery ( min), 0.5 mg of timolol or placebo was given intravenously twice daily in a double-blind manner. When oral medications were resumed postoperatively, 10 mg of timolol twice daily or placebo was continued orally. Continuous electrocardiograms were recorded for 24 hr before and for 7 days after surgery with a standard cassette recorder. No patient received digoxin. Both groups were comparable for frequency of preoperative supraventricular arrhythmias, left ventricular ejection fraction, duration of cardiopulmonary bypass, aortic cross-clamp time, number of bypass grafts, and total duration of monitoring. Analysis of arrhythmias was done by hand counts, and supraventricular arrhythmias were divided into supraventricular tachycardia and atrial fibrillation and/or flutter. Timolol decreased the frequency of supraventricular tachycardia (581 episodes placebo vs 84 timolol; p <.05) and of atrial fibrillation and/ or flutter (291 episodes placebo vs five timolol; p <.05). Timolol decreased the number of patients with severe (heart rate > 200 beats/min, duration > 50 beats) episodes of supraventricular tachycardia (four placebo vs 0 timolol; p <.05) and also decreased the number of episodes of severe (heart rate > 200 beats/min, duration > 5 min) atrial fibrillation and/or flutter (16 placebo vs one timolol; p <.005). There were no differences in the durations of supraventricular arrhythmias. All 1 1 episodes of supraventricular tachycardia and 11 episodes of atrial fibrillation and/or flutter with heart rates over 200 beats/min occurred in the placebo group (p <.005). Four patients on placebo and one patient on timolol required treatment for symptomatic supraventricular arrhythmias. Two patients in the timolol group were nauseated and one patient developed wheezing while on placebo. The results of this study demonstrate that prophylactic use of timolol after coronary bypass surgery is effective and safe therapy for decreasing the frequency and severity of supraventricular arrhythmias. Circulation 70, No. 3, , SUPRAVENTRICULAR arrhythmias are common after coronary artery bypass surgery, occurring in 20% to 54% of patients These arrhythmias may be associated with excessively rapid ventricular rates that jeopardize hemodynamic stability; in such cases urgent treatment is required to slow the heart rate or restore sinus rhythm. On occasion, when antiarrhythmic drug therapy is ineffective, it is even necessary to From the Departments of Medicine and Surgery, Brigham and Women's Hospital and Harvard Medical School, Boston. Address for correspondence: Peter L. Friedman, M.D., Ph.D., Cardiovascular Division, Brigham and Women's Hospital, 75 Francis St., Boston, MA Received Dec. 9, 1983; revision accepted May 10, Dr. White is supported by the Odlin Research Fellowship of the Royal Australasian College of Physicians. Vol. 70, No. 3, September 1984 resort to electrical cardioversion. A number of studies have examined the efficacy of prophylactic treatment for prevention of postoperative supraventricular arrhythmias. Prophylaxis with digoxin has had equivocal results." 4-- Propranolol appears to be effective when given prophylactically, although interpretation of the data reported previously has been hampered by suboptimal study design. 68, 9 Timolol is a potent /3- adrenoceptor blocker that does not have membranestabilizing activity. It is less lipophilic than propranolol and would be expected to cause fewer side effects in the central nervous system of the type associated with propranolol. This study was designed to evaluate the efficacy and safety of timolol as compared with place- 479

2 Downloaded from by on September 18, 2018 WHIITE et al. bo in the prophylactic treatment of supraventricular tachyarrhythmias after coronary artery surgery. Methods Patients. Forty-one patients undergoing elective coronary artery bypass surgery were randomly assigned to receive timolol (21 patients) or placebo (20 patients) in a double-blind manner for 7 days after surgery. Patients with the following contraindications to taking /3-adrenoceptor-blocking agents were not eligible for inclusion: (1) second- or third-degree atrioventricular block, (2) resting sinus bradycardia less than 56 beats/min. (3) diabetes mellitus requiring insulin or oral hypoglycemic agents, (4) history of attacks of spontaneous hypoglycemia, (5) allergic rhinitis, (6) bronchospasm of any cause, or (7) chronic obstructive pulmonary disease. Patients who were being treated with digitalis glycosides were also excluded from the study. A low left ventricular ejection fraction, by itself, was not considered to be a contraindication to participation in the study. However, no patient enrolled had a left ventricular ejection fraction less than 41 %, presumably because those patients requiring the inotropic support of digitalis had already been excluded from participation. There were 34 men and seven women, ages 35 to 72 years (mean 55). /3-Adrenoceptor-blocking therapy was stopped at least 12 hr before surgery for those patients taking propranolol (23 patients), metoprolol (four patients), atenolol (three patients), or timolol (two patients) and at least 20 hr before surgery for the seven patients taking nadolol. Two patients in the placebo group were taking verapamil. Study design. All patients had a 24 hr, two-channel electrocardiographic recording for the 24 hr immediately before surgery. After surgery, double-blind therapy with either placebo (10 ml of normal saline) or timolol (0.5 mg diluted in 10 ml of saline) was given intravenously over 1 min twice daily. Medication was given when the patients had returned to the recovery room and their conditions were judged to be stable (mean min). Long-term electrocardiographic recording on cassette was begun at the time of the intravenous injection and was continued without interruption for the ensuing 7 days. When patients were able to take oral medications, 10 mg of timolol twice daily or matching placebo was given. The total treatment period was 7 days. If during this period a supraventricular tachyarrhythmia or a ventricular arrhythmia occurred that was judged to require treatment (arrhythmia associated with intolerable symptoms or a decline in systolic blood pressure to 100 mg or less), then appropriate therapy was begun and the patient was excluded from further participation in the study. No patient received digoxin during the course of the study. Electrocardiographic monitoring, data acquisition, and analysis. Two-channel, long-term electrocardiographic monitoring, with either a Picker International Model 7300 or Oxford 424 cassette recorder, was performed for 7 days after the first intravenous administration of drug or placebo. Tapes were scanned on either a Camscan ECG Scanner (Picker International) or Heartscreen (Delmar Avionics) and were analyzed to detect and quantitate supraventricular tachyarrhythmias. Because of the inferior accuracy of the available automated analytical systems for detecting supraventricular tachyarrhythmias as compared with ventricular arrhythmias, all tapes were analyzed by hand. This was accomplished by freezing the monitoring scope when a supraventricular tachyarrhythmia was detected during scanning of the tape. Details of the arrhythmia such as P wave morphology, rate, length of run, and abruptness of onset/ offset were noted either directly from the frozen image on the scope or by review of a hard-copy electrocardiographic printout. Supraventricular tachyarrhythmias were divided into two 480 broad categories: supraventricular tachycardia and atrial fibrillation and/or flutter. Criteria for the diagnosis of supraventricular tachycardia included an atrial rate in excess of 100 beats! min, abrupt onset and termination of tachycardia, and a P wave morphology that distinctly differed from the sinus P wave. The episode of supraventricular tachycardia also had to have a cycle length that was at least 15% faster than the basic sinus rhythm. All of the above criteria had to be fulfilled to minimize the chance that sinus tachycardia would be included in the data. The rate and duration of each episode of arrhythmia were determined. Arrhythmias were classified as mild, moderate, or severe according to both heart rate and duration (table 1). For an arrhythmia to be classified as severe, both criteria for rate and duration had to be fulfilled. If one criterion was moderate and one severe, the grading became moderate; if one criterion was mild and one severe, the grading became moderate; and if one was mild and one moderate, the grading became mild. Perioperative myocardial infarction was diagnosed by the development of new Q waves and elevation of cardiac-specific isoenzymes. The study protocol was approved by the institutional Committee for the Protection of Human Subjects, and written informed consent was obtain from each patient. Statistical analysis. Data are reported as mean + SD. Comparisons between groups were made with unpaired t tests or the Mann-Whitney U test as appropriate. The incidence of phenomena was compared by chi-square analysis. Results The patient characteristics for the placebo- and timolol-treated groups are listed in table 2. There were no differences in age, sex, left ventricular ejection fraction, extent of coronary artery disease, duration of cardiopulmonary bypass, aortic cross-clamp time, or number of grafts applied. Four patients in the placebo group had a total of five episodes of supraventricular tachycardia in the preoperative 24 hr monitoring period. Seven patients in the timolol group had a total of 10 episodes of supraventricular tachycardia. This difference was not statistically significant. No patient had preoperative atrial fibrillation and/or flutter. Frequency and grade of severity of postoperative supraventricular tachyarrhythmias. The duration of electro- TABLE 1 Grading of severity of supraventricular arrhythmias Grade Duration RateA Supraventricular tachycardia Mild <10 beats <150/min Moderate beats /min Severe >50 beats >200/min Atrial fibrillation and/or flutter Mild <30 sec <150/min Moderate 30 sec-s min /min Severe >S min >200/min AAtrial and ventricular rate for supraventricular tachycardia; ventricular rate for atrial fibrillation and/or flutter. No instances of supraventricular tachycardia with atrioventricular block were encountered. CIRCULATION

3 THERAPY AND PREVENTION-f-ADRENERGIc BLOCKADE TABLE 2 Patient characteristics SUPRAVENTRICULAR TACHYCARQIA ATRIL FIBILLATION OANDO d FLUTTER. Downloaded from by on September 18, 2018 Placebo Timolol p (n - 20) (n = 21) value Age (yr)a 56± NS Sex (F/M) 3/17 4/17 NS Ejection fraction (%)A ± 11 NS Coronary artery disease (n) 3-vessel disease NS 2-vessel disease 6 8 NS 1-vessel disease 0 2 NS Episodes of preoperative SVA 5 10 NS Patients with preoperative SVA 4 7 NS Cross-clamp time (min)a 33 ± ± 12 NS Duration of bypass (min)a 64 ± ± 24 NS No. of graftsa 3.1± ±1.2 NS Preoperative fl-blocker (n) Propranolol NS Atenolol 2 1 NS Metoprolol 1 3 NS Timolol 1 1 NS Nadolol 3 4 NS None 2 0 NS SVA = supraventricular arrhythmia. AValues are mean ± SD. cardiographic monitoring in the group given placebo was hr, compared with hr in the patients treated with timolol. This difference was not statistically significant. For both groups these values fell short of the expected 168 hr (7 days) of monitoring. Part of this difference was due to appearance of possible drug-induced side effects in three patients (one in the placebo group, two in the timolol group) that led to their removal from the remainder of the study before completion of the expected 7 days of monitoring. Five patients who required active intervention for severe arrhythmias before 7 days of monitoring had elapsed were also excluded from subsequent rhythm analysis. In one patient an ambulatory electrocardiographic monitor was inadvertently disconnected prematurely. There were no episodes of tape malfunction during the study. Patients' in the timolol-treated group had a lower mean maximum sinus rate per day over the total time monitored compared with patients given placebo (114 ± 19 beats/min placebo vs beats/min timolol; p <.001). This difference suggested that the dose of timolol administered did result in a significant degree of,3-adrenoceptor blockade. All patients had episodes of supraventricular tachyarrhythmias in the postoperative period. However, as illustrated in figure 1, timolol decreased the number of episodes of supraventricular tachycardia (581 placebo vs 84 timolol; p <.05) Vol. 70, No. 3, September 1984 Number of PLACEBO M T IMOLOL pc 0.05 p X 0.05 FIGURE 1. Frequency of postoperative supraventricular tachyarrhythmias. and of atrial fibrillation and/or flutter (291 placebo vs five timolol; p <.05). As shown in table 3, four patients in the placebo group had severe episodes of supraventricular tachycardia (heart rate > 200'beats/min and duration > 50 beats) compared with no patients in the timolol-treated group (p <.05). Conversely, more timolol patients had mild episodes of supraventricular tachycardia. There was no significant difference between the groups for the numbers of patients with atrial fibrillation and/or flutter. Although four patients in the placebo group had severe episodes compared with one patient in the timolol group, this did not achieve statistical significance. There were five episodes of severe supraventricular tachycardia in the placebo group and none in the timolol group (p - NS). As shown in'figure 2, there was a TABLE 3 Maximum grade of supraventricular arrhythmiaa Placebo Timolol (No. of (No. of p Grade patients) patients) value Supraventricular tachycardia None 1 1 NS Mild 7 16 <.05 Moderate 8 4 NS Severe 4 0 <.05 Overall test <.05 Atrial fibrillation andior flutter None NS Mild 2 0 NS Moderate 1 2 NS Severe 4 1 NS Overall test NS AMaximum grade for a patient is the level of the most severe episode over the time the patient was monitored. The overall test (an extension of the Fischer exact test) is a test of whether the four category distributions in the timolol group differed from the distributions in the placebo group. 481

4 Downloaded from by on September 18, 2018 WHITE et al Number 5 of Episodes 10 NS E PLACEBO M TIMOLOL ATRIAL FI3RILLATION AND/OR FLUTTER 77,, 0 1 A FIGURE 2. Frequency of severe arrhythmias. significant difference for the number of severe episodes of atrial fibrillation and/or flutter (16 episodes placebo vs one episode timolol; p <.005). Duration of episodes and rates of supraventricular arrhythmia. Although the frequency of episodes of supraventricular tachycardia and atrial fibrillation and/or flutter was reduced in the group of patients treated with timolol compared with those treated with placebo, there was not a striking difference in the duration of episodes of arrhythmia between the two groups. Episodes of moderate duration (10 to 50 beats for supraventricular tachycardia, 30 sec to 5 min for atrial fibrillation and/or flutter) occurred with equal frequency in the two groups (table 4). The same was true for longlasting episodes of supraventricular tachycardia and atrial fibrillation and/or flutter (table 4). In contrast, brief paroxysms of supraventricular tachycardia (< 10 beats) and brief paroxysms of atrial fibrillation and/or flutter (<30 sec) occurred less frequently in patients treated with timolol (table 4). There were significant differences between the atrial and ventricular rates during arrhythmia in the two groups. There were 11 episodes of supraventricular tachycardia with atrial rates in excess of 200 beats/min in the placebo-treated group compared with none in the timolol-treated group (p <.005). There were also 11 TABLE 4 Duration of arrhythmia Grade Placebo Timolol p value Supraventricular tachyarrhythmia Mild (<10 beats) 566A 69 <.05 Moderate (10-50 beats) NS Severe (>50 beats) 5 0 NS Atrial fibrillation and/or atrial flutter Mild (<30 see) <.001 Moderate (30 sec-5 min) 6 1 NS Severe (> 5 min) 12 4 NS SUPRAVENTRICULAR TACHYCftRIA AData expressed as number of episodes. --I episodes of atrial fibrillation and/or flutter and ventricular rates in excess of 200 beats/min in the placebotreated group compared with none in the timolol-treated group (p <.005). The mean maximum heart rate during episodes of supraventricular tachyarrhythmia for each patient was not significantly different (175 ± 30 beats/min placebo vs beats/min timolol; p.059). The mean maximum heart rate during atrial fibrillation and/or flutter, however, was lower in the group treated with timolol ( beats/min placebo vs beats/min timolol; p <.05). Adverse effects. One patient in the placebo group developed wheezing. Two patients in the timolol group developed nausea, possibly caused by timolol. One patient in the timolol group developed electrocardiographic signs of perioperative myocardial infarction 12 hr after surgery. Several hours later he developed rapid atrial fibrillation requiring active intervention. At this point the patient was removed from further participation in the study. Subsequently he developed additional signs of infarction and died from electromechanical dissociation 15 hr after receiving timolol. There was no clinical evidence that timolol caused his myocardial infarction. Discussion Previous studies have examined the frequency of supraventricular arrhythmias after coronary artery bypass surgery and have reported an incidence ranging between 15% and 54%.l 10 These figures are almost certainly underestimates. Studies to date have relied on frequency data derived from nurses' observations of monitor screens and patients' reports of symptoms. The inaccuracy of such methods is highlighted by a report'0 based on long-term electrocardiographic monitoring on the first and fifth days after discharge from intensive care; these results showed that 53% of postoperative supraventricular tachyarrhythmias went unnoticed by the patients and the physicians or nurses caring for them. Our study is important in that 100% of patients had supraventricular arrhythmias after surgery, an incidence much greater than that previously reported. It is unique in that continuous 24 hr, longterm electrocardiographic recording was performed for the first 7 consecutive days after surgery. In all likelihood, the higher frequency of arrhythmia we observed is due in part to the improved method of detection. Prevention of postoperative supraventricular arrhythmias. Although supraventricular arrhythmias after coronary artery bypass surgery are not often life threatening, they can jeopardize hemodynamic stability or lead to intolerable symptoms. This is particularly true for CIRCULATION

5 THERAPY AND PREVENTION-jr-ADRENERGIC BLOCKADE Downloaded from by on September 18, 2018 patients with diffuse coronary artery disease in whom complete revascularization may not be possible. A number of previous studies have examined the efficacy of antiarrhythmic drug therapy for prevention of postoperative supraventricular arrhythmias. Unfortunately, interpretation of many of these studies has been hampered by inconclusive or conflicting results and inadequate study design. Prophylactic use of digoxin, for example, has been found to be beneficial by some investigators" 1 but not by others.4 Considerable attention has been focused on the prophylactic use of /3-adrenoceptor-blocking agents. This interest stems from evidence that a rebound phenomenon after withdrawal of such agents at the time of surgery may contribute to the appearance of arrhythmia in the postoperative period.3' 6, Propranolol may be effective in reducing the incidence of postoperative supraventricular arrhythmia.2 68, 9 In an unblinded study, Mohr et al.8 gave propranolol to patients via nasogastric tube 6 hr after surgery and found that the frequency of arrhythmias, based on their appearance on a bedside monitor, was reduced from 45% to 5%. Stephenson et al.,5 using an open-label study design, showed that oral propranolol given to patients after discharge from the intensive care unit reduced the incidence of supraventricular tachycardia from 18% to 8% as compared with control data. In that study, however, the groups were of different sizes and the time of initiation of therapy was not uniform. A third unblinded study9 in which oral propranolol was administered for 48 hr beginning the morning after surgery showed a reduction in the incidence of supraventricular tachycardia, detected by observation of electrocardiographic monitors, from 28% to 6%. A similar reduction in postoperative supraventricular tachyarrhythmias was reported by Oka et al.6 in a small group of patients given intravenous propranolol in a unblinded fashion. It is important to reemphasize that our study is the first one to be carried out in a double-blinded, placebocontrolled fashion with prolonged continuous electrocardiographic recording as a measure of drug efficacy. Efficacy of prophylactic timolol. Timolol is a nonselective /3-adrenoceptor-blocking agent that lacks intrinsic sympathomimetic or membrane-stabilizing activity. It is unlike other /B-adrenoceptor-blocking agents in that it has two heterocyclic rings, one containing sulfur and nitrogen and one with nitrogen and oxygen. Part of its antiarrhythmic properties may stem from its ability to antagonize the effects of catecholamines on cardiac automaticity and conduction. The dose of timolol administered in the present study would be expected to result in a significant degree of /3-adrenoceptor block- Vol. 70, No. 3, September 1984 ade. This was supported by the finding that maximum daily sinus rates in patients treated with timolol were significantly lower than those in the group given placebo. Timolol, given intravenously immediately after coronary artery bypass surgery and ensuring rapid predictable onset of action without variable absorption, was effective in preventing supraventricular tachyarrhythmias during the first postoperative week. This antiarrhythmic effect was demonstrated in two different ways. First, timolol decreased the frequency of supraventricular tachycardia as well as atrial fibrillation and/or flutter. Second, and perhaps more importantly, timolol decreased the severity of supraventricular arrhythmias according to a grading system developed to reflect the clinical assessment of the importance of a supraventricular arrhythmia. In reducing the severity of postoperative supraventricular arrhythmias, the dominant effect of timolol seemed to be on rate rather than duration. All 22 episodes of arrhythmias faster than 200 beats/min occurred in the placebo group. The maximum heart rates of patients with episodes of atrial fibrillation and/or flutter were also significantly lower in the timolol group. It is also worth noting that fewer patients given prophylactic timolol required treatment for symptomatic arrhythmia compared with patients given placebo, although this difference did not achieve statistical significance. Comparability of groups. The fact that most patients in this study had had long-term treatment with a variety of /3-adrenoceptor-blocking agents before surgery is a potentially confounding feature. Ideally, one would prefer to study groups of patients in whom such drugs had been discontinued many days previously, ensuring complete elimination of drug from both tissue and plasma. Since our patients were admitted to the hospital only 1 day before surgery, such an approach was not possible. Those patients taking nadolol before surgery, for example, almost certainly had some degree of persistent,3-adrenergic blockade postoperatively due to the long pharmacokinetic half-life of this agent and to the fact that it was discontinued only 20 hr before surgery. However, this cannot account for the differences in incidence or severity of supraventricular arrhythmias between the placebo- and timolol-treated groups, since virtually the same number of patients had taken nadolol before surgery in each group (table 2). Patients receiving long-term treatment with other f3- adrenoceptor drugs with shorter pharmacokinetic halflives had these drugs discontinued at least and often more than 12 hr before surgery, a period equivalent to several elimination half-times. These patients would be expected to have had very little residual /,3-adreno- 483

6 Downloaded from by on September 18, 2018 WHITE et al. ceptor blockade postoperatively, particularly in view of the enhanced clearance of these drugs that occurs in paiients during cardiopulmonary bypass.6 It may be argued that a rebound phenomenon occurred in these patients and contributed to the rather high incidence of postoperative supraventricular arrhythmias encountered in the placebo group. Conversely, the lower incidence of supraventricular arrhythmias in patients treated with timolol may, in part, have been caused by the absence of a rebound phenomenon in that group. Of interest is the fact that two patients in the placebo group had not taken any /8-adrenoceptor-blocking drugs before surgery and might have been expected to have had fewer supraventricular arrhythmias after surgery because of the absence of a rebound phenomenon. Whatever the mechanism of timolol's action, the reduction in postoperative arrhythmias observed in the timolol-treated patients may therefore actually be a minimum estimate of drug efficacy. Finally, it should be noted that differences in postoperative arrhythmias in the placebo- and timolol-treated groups could not be accounted for by other factors, such as preoperative ejection fraction, number of bypass grafts, aortic cross-clamp time, duration of cardiopulmonary bypass, or presence of preoperative supraventricular arrhythmia. Safety. In this study, none of the patients treated with prophylactic timolol developed significant adverse effects that could be attributed to the drug. It should be noted, however, that high-risk patients were not included in the study. Patients with second- or thirddegree atrioventricular block on their preoperative electrocardiograms were excluded, and no patient in the study had a left ventricular ejection fraction less than 41%. Use of,3-adrenoceptor-blocking agents may be hazardous in patients with jeopardized atrioventricular conduction or severely compromised ventricular function. Our results demonstrate that in the absence of these relative contraindications, timolol given prophylactically is both safe and effective for prevention of supraventricular arrhythmias after coronary bypass surgery. We thank Pamela Alvarez, Tori Then, and Susan Barry, R.N., for technical assistance, the Statistics Department of the Harvard School of Public Health for statistical assistance, and the nursing staff of the Thoracic Intensive Care Unit of the Brigham and Women's Hospital for the meticulous attention to detail and excellent care they provided. References 1. Johnson LW, Dickstein RA, Fruehan CT, Kane P, Potts JL, Smulyan H,. Watts R, Webb WR, Eich RH: Prophylactic digitalization for coronary artery bypass surgery. Circulation 53: 819, Boudoulas H, Lewis RP, Snyder GL, Karayannacos P, Vasko JS: Beneficial effect of continuation of propranolol through coronary bypass surgery., Clin Cardiol 2: 87, Salazar C, Frishman W, Friedman S, Patel J, Lin YT, Oka Y, Frater RWM, Becker RM: /3-Blockade therapy for supraventricular tachyarrhythmias after coronary surgery: a propranolol withdrawal syndrome? Angiology 30: 816, Tyras DH, Stothert JC, Kaiser GC, Barner HB, Codd JE, William VL: Supraventricular tachyarrhythmias after myocardial revascularisation: a randomized trial of prophylactic digitalization. J Thorac Cardiovasc Surg 77: 310, Stephenson LW, MacVaugh H, Tomasello DN, Josephson ME: Propranolol for prevention of postoperative cardiac arrhythmias: a randomised study. Ann Thorac Surg 29: 113, Oka Y,,Frishman W, Becker RN, Kadish A, Strom J, Matsumoto M, Orkin L, Frater R: Clinical pharmacology of the new beta adrenergic blocking drugs. 10. Beta adrenoceptor blockade and coronary artery surgery. Am Heart J 99: 255, Csicsko JF, Schatzlein MH, King RD: Immediate postoperative digitalization in the prophylaxis of supraventricular arrhythmi,as following coronary artery bypass. Ann Thorac Surg 81: 419, Mohr R, Smolinsky A, Goor DA: Prevention of supraventricular tachyarrhythmias with low-dose propranolol after coronary bypass. J Thoracic Cardiovasc Surg 81: 840, Silverman NA, Wright R, Levitsky S: Efficacy of low dose propranolol in preventing postoperative supraventricular tachyarrhythmias. Ann Surg 196: 194, 1982, 10. Michelson EL, Morganroth J, MacVaugh H: Postoperative arrhythmias after coronary artery and cardiac valvular surgery detected by long-term electrocardiographic monitoring. Am Heart J 97: 442, Chee TP, Prakash S, Desser KB, Benchimol A: Postoperative supraventricular arrhythmias and the role of prophylactic digoxin in cardiac surgery. Am Heart J 104: 974, Boudoulas H, Lewis RP, Kates RE, Dalamangas G: Hypersensitivity to adrenergic stimulation after propranolol withdrawal in normal subjects. Ann Intern Med 87: 433, CIRCULATION