Arrhythmia prophylaxis after aorta-coronary bypass

J

THoRAc CARDlOVASC SURG

89:439-443, 1985

Arrhythmia prophylaxis after aorta-coronary bypass The effect of minidose propranolol After aorta-coronary bypass grafting, 164 consecutive patients were randomized to receive propranolol 5 mg every 6 hours oraUy (n = 82) or to serve as control subjects (n = 82). All patients were receiving beta blockers preoperatively. There were no significant differences between the two groups. The incidence of sustained supraventricular (nonsinus) tachyarrhythmias was 23 % in the control group and 9.8 % in the treated group (p = 0.02). The incidence of ventricular arrhythmias was 15 % in tbe control group and 2.4 % in the treated group (p = 0.005). The overaU difference in clinciaUy important arrhythmias was 38 % in the control group and 12.2 % in the treated group (p = 0.0002). We conclude that low-dose oral propranolol in patients who were receiving beta blockers preoperatively is effective in reducing the incidence of clinicaUy important arrhythmias occurring after aorta-coronary bypass grafting.

Murray F. Matangi, M.B., Ch.B., F.R.A.C.P., * John M. Neutze, M.D., F.R.A.c.P., Kenneth J. Graham, M.B., Ch.B., F.R.A.C.S., David G. Hill, M.B., Ch.B., F.R.A.C.S., Alan R. Kerr, M.B., Ch.B., F.R.A.C.S., and Brian G. Barratt-Boyes, M.B., Ch.M., F.R.A.C.S.,

Auckland, New Zealand

Arrhythmias, especially those of supraventricular origin, are an important cause of morbidity after aorta-coronary bypass grafting. I The purpose of this study was to investigate the efficacy of low-dose oral propranolol as a prophylaxis against arrhythmias after aorta-coronary bypass grafting in patients receiving preoperative beta blocking drugs. Patients and methods Two hundred forty-four consecutive patients underwent aorta-coronary bypass grafting during the study period. Seventy-six patients were excluded for the following reasons: ejection fraction of 35% or less (n = 11), asthma (n = 11), no preoperative use of beta blockers (n = 18), pre-randomization supraventricular tachyarrhythmia (n = 10), postoperative inotropic support (n = 6), congestive heart failure (n = 3), postoper-

From the Departments of Cardiology and Cardiothoraeie Surgery, Green Lane Hospital, Auckland, New Zealand. Received for publication Dec. 27, 1983. Accepted for publication April 9, 1984. Address for reprints: John M. Neutze, M.D., F.R.A.C.P., Green Lane Hospital, Green Lane West, Auckland 3, New Zealand. 'Cardiology Research Fellow, University of Minnesota, Box 508 Mayo Memorial Building, 420 Delaware St., S.E., Minneapolis, Minn. 55455.

ative heart block (n = 2), involvement in another study where propranolol was contraindicated (n = 12), sick sinus syndrome (n = 1), known reaction to propranolol (n = 1), and incorrect randomization (n = 1). After informed consent had been obtained, the remaining 168 patients were randomized by the sealed envelope technique to receive propranolol 5 mg every 6 hours orally or to serve as controls. Three patients in each group were receiving digoxin preoperatively (none for arrhythmia), which was discontinued. There were 83 treated patients and 85 control subjects. Initial dosages of propranolol were given as a suspension via a nasogastric tube, after which the drug was continued orally until the time of discharge. The first dose was given 10 ± 7 hours (mean ± standard deviation) after the end of the operation. Preoperatively, the beta blocker was continued until the night before operation for morning cases or the morning of operation for afternoon cases. The endpoints of the study were (I) sustained supraventricular (nonsinus) tachyarrhythmia necessitating antiarrhythmic therapy, with or without the need for elective synchronized direct-current cardioversion; (2) ventricular tachycardia (6 or more consecutive ventricular ectopic beats at a rate> 100 beats/min) necessitating antiarrhythmic therapy with or without direct-current countershock; (3) ventricular fibrillation; (4) frequent ventricular premature beats (more than 6 per minute) 439

440

The Journal of Thoracic and Cardiovascular Surgery

Matangi et al.

Table I. Patient characteristics: Historical data

Age (yr) Sex (male/female) Hypertension Previous MI Angina class (CCS)

54.6 ± 9.3 67/15 41 41

Table

Control (n = fl2)

P Value

55.7 ± 9.9 63/19 44 54

NS NS NS 0.1 NS NS NS NS NS

I II III IV

I 7 37 34

2 6 38 33

V

3

3

Legend: All values arc expressed as the mean ± the standard deviation of the mean. NS, Not significant. M L Myocardial infarction. CCS, Canadian Cardiovascular Society (V = no angina).

Table

n. Patient characteristics: Clinical data Control in = fl2)

Preop. pulse (beats/min) Preop. systolic BP (mm Hg) LVEDP (rnm Hg) Ejection fraction ('Yr) Myocardial score' Pulse rate postop. day 7 (beats/min)

130.5 ± 19.1 12.8 65.2 10.5 80.2

± 5.8 ± 11.2 ± 2.3 ± 10.5

p Value

59.4 ± 9.3

NS

131.0 ± 18.6

NS

13.0 65.0 11.2 93.8

± ± ± ±

6.0 11.1 1.9 12.8

NS NS <0.025 <0.0005

Legend: All values arc expressed as the mean ± the standard deviation of the mean. BP, Resting blood pressure. LVEDP, Left ventricular end-diastolic

pressure.

'''Brandt'' myocardial scoring system for coronary artery disease.

with a normal concentration of serum potassium (2:4.0 mEqjL) necessitating antiarrhythmic therapy. For arrhythmia monitoring, all patients were connected to an oscilloscopiccardiac monitor that was connected to a central monitoring station. The first 24 to 48 hours were spent in the cardiac surgical intensive care unit and a further 24 to 48 hours in a similarly monitored section of the cardiac surgical ward. All important cardiac arrhythmias were recorded with a rhythm strip for review by one of us (M. F. M.). Seventy-five percent of the patients also had a twelvelead electrocardiogram recorded during the arrhythmia. When doubt existed as to the type of arrhythmia (i.e., atrial flutter or atrial tachycardia), simultaneous atrial bipolar electrocardiograms and surface lead electrocardiograms were recorded. Initial management of atrial fibrillation and atrial flutter was with intravenous followed by oral digoxin. Atrial tachycardia was managed with rapid overdrive

m. Patient characteristics: Operative data

Grafts per patient Bypass time (min) Aortic cross-clamp time (min) Second or subsequent ACBG operation Pcrioperative M I Dcaths

Control in = fl2)

P Value

2.8 ± 1.0 . 86.6 ± 26.5 48.4± 17.1

2.9 ± 0.9 89.1 ± 26.0 51.2 ± 14.5

NS NS NS

II

9

NS

5 (6%) I (1.2',fc)

NS NS

4 (5'Yr) I (1.2%)

Legend: All values arc expressed as the mean ± the standard deviation of the mean. ACBG, Aorta-coronary bypass grafting. MI, Myocardial infarction.

atrial pacing followed by digoxin. If these measures failed, then oral treatment with either quinidine or disopyramide was added. If the arrhythmia persisted, cardioversion was performed. Cardioversion was usually carried out 48 to 72 hours after the start of quinidine or disopyramide. When ventricular arrhythmias persisted after correction of the serum potassium, they were managed initially with lidocaine and subsequently with oral procainamide. When hemodynamic embarrassment accompanied ventricular tachycardia, direct-current countershock was employed. For statistical analysis the incidence of phenomena was compared by the chi square test. The unpaired Student's t test was used to determine differences between the means. All values are stated as a mean ± the standard deviation of the mean. A p value of less than 0.05 was considered significant. Results Four patients were withdrawn from the treatment group. Two had a low cardiac output associated with a junctional bradycardia that responded to atrial pacing. One patient had bronchospasm and another had nightmares. Both of these latter patients had been taking atenolol preoperatively. All four patients are still analyzed in accordance with the "intention to treat" principle. Four patients, three in the control group and one in the treatment group, had short untreated episodes of supraventricular tachycardia. In accordance with the predefined endpoints, these patients are not included in the statistical analysis. As a result, 82 patients in each group were analyzed. Analysis of these four patients does not affect the conclusions of the study. In fact, it strengthens the statistical significance between the two groups. Patient characteristics are summarized in Tables I to

Volume 89 Number 3

Propranolol for arrhythmia prophylaxis

March,1985

44 1

Table IV. Patient characteristics: Preoperative drugs

Propranolol (mgjday) Metoprolol (mgjday) Atenolol (mgjday) Nifedipine (mgjday) Isosorbide (mgjday) Diuretics Topical nitrates (mgjday) Digoxin (mgjday)

PROPRANOLOL GROUP

Treatment (n = 82)

Control (n = 82)

Value

218.2 ± 80.8 (37)

230.3 ± 160.4 (35)

NS

192.1 ± 82.0 (19)

228.3 ±

91.5 (23)

NS

104.2 ± 49.8 (12)

95.5 ±

15.1 (11)

NS

53.6 ± 21.0 (53)

57.7 ±

21.6 (43)

NS

00 01 02 03 04 05 06 07(+)

67.7 ± 36.6 (49)

70.2 ±

37.3 (54)

NS

CONTROL GROUP

(22) (32)

(21) (26)

(3)

(3)

p

Patients with

SVT

6 4 2

0..---

Patients

with

SVT

l.egend: Thirteen control patients and 14 treated patients were laking a variety of other beta blockers. Numbers in parentheses represent total numbers of patients taking the medication.

IV. The groups were similar with respect to age, sex, history of hypertension, angina class, ejection fraction, left ventricular end-diastolic pressure, bypass time, aortic cross-clamp time, preoperative myocardial infarction, and perioperative mortality and infarction. The control group had slightly more coronary artery disease as indicated by our myocardial scoring system.' Nineteen patients in the control group (23%) and eight patients in the treatment group (9.8%) experienced sustained supraventricular tachyarrhythmias (chi square = 5.36, P = 0.02). The difference between the two groups was due entirely to the incidence of supraventricular arrhythmias on the second and third postoperative days. After the fourth day the incidence in the two groups was similar (Fig. 1). The types of arrhythmia are shown in Table V. All patients who had supraventricular tachyarrhythmias received intravenous digoxin as the initial treatment for their arrhythmia. However, only one patient in the treatment group required additional antiarrhythmic therapy, as compared to 13 control patients (chi square = 11.25, P = 0.(01). In addition, three control patients required elective synchronized direct-current cardioversion and three further control patients had successful overdriveatrial pacing (two with atrial tachycardia and one with atrial flutter). No patient in the treatment group required cardioversion. The treated patients tended to have better hemodynamics during tachycardia. The ventricular response was considerably slower in treated patients (ventricular rate 119 ± 23 beats/min for the treatment group versus 151 ± 23

Fig. 1. Incidence of

supraventricular

tachyarrhythmias

(SVT) on the days [0" to 0 71+)1 postoperatively for propranolol

and control groups.

Table V. Types of arrhythmias Control (n = 82)

Supraventricular Atrial fibrillation Atrial flutter Atrial tachycardia Total Ventricular Ventricular tachycardia Ventricular fibrillation Frequent VPB Total All arrhythmias

6

2

p Value

12

5

o

2

8 (9.8%)

19 (23',7,)

a

5

0.02

2 I

5

2 (2.4%)

12 (15%)

0.005

10 (12.2%)

31 (38%)

0.0002

Legend: VPB. Ventricular premature beats.

beats/min for the control group, p = 0.(02). The blood pressure response was better, the duration of arrhythmias shorter, and relapses fewer, but these differences were not statistically significant (Table VI). Twelve patients in the control group had ventricular arrhythmias (15%), compared with two (2.4%) in the treatment group (chi square = 7.8, P = 0.(05). No one

44 2

The Journal of Thoracic and Cardiovascular Surgery

M 'atangi et al.

Table VI. Hemodynamic consequences of supraventricular arrhythmias

Ventricular rate (beats/min) Systolic BP (mm Hg) Duration (hr) Relapses

108 ± 18 25 ± 19 2

Control (n = 19)

p Value

151 ± 23

0.002

100 ± 21 67 ± 77

NS NS NS

4

Legend: All values are expressed at the mean ± the standard deviation of the mean. BP, Blood pressure.

inotropic action, 14, 15 a satisfactory antiarrhythmic action,16,17 and a longer effective half-life of 4 to 6 hours. 18 In our study the overall difference in clinically important arrhythmias was 38% in the control group and 12% in the treatment group (p = 0.0002). This is the first study to show a clear benefit for both supraventricular (p = 0.02) and ventricular (p = 0.005) arrhythmias. Other studies have suggested a beneficial trend with regard to prophylaxis against ventricular arrhythmias, but this trend has not achieved statistical signifi-

cance.'" in the treatment group was observed to have ventricular tachycardia (Table V). The overall difference in clinically important arrhythmias was 31 (38%) in the control group to 10 (12%) in the treatment group (chi square = 14.3, P = 0.0002). Discussion There were several reasons for this study. First, arrhythmias have been reported after aorta-coronary bypass in 10% to 40% of patients, 1,3,4 a high enough incidence to be an important cause of postoperative morbidity. Second, although there have been several studies of the effect of intravenous and oral propranolol after aorta-coronary bypass, variation in design of these studies has left a number of questions unanswered. Some studies have included sinall numbers of patients,' a failure to adhere to the "intention to treat" principle," and in one particular study 50% of "control" patients actually received propranolol." However, the results generally favored continued propranolol administration. A recent randomized double blind study in which 80 mg of propranolol was given daily failed to show any benefit.' Third, studies utilizing digoxin under these circumstances still produce conflicting results.v 9 Furthermore, in these studies digoxin has not had any effect on prophylaxis for ventricular arrhythmias. In fact, digoxin may even potentiate ventricular arrhythmias when given prophylactically after aorta-coronary bypass. Finally, we were unconvinced with the total available data on low-dose oral propranolol under these circumstances and decided to investigate the drug in a prospective randomized controlled trial. If low-dose oral propranolol is damping down a hypersensitivity withdrawal phenomenon,'? II as has been suggested by other authors.v II, 12 then the study should be restricted to patients receiving beta blockers before the operation. The low oral dosage was used to be consistent with previous studies. 13 Other attributes of the low-dose and oral route are the avoidance of a negative

Not surprisingly, when supraventricular tachyarrhythmias occurred, the ventricular response was significantly slower in the treatment group, the blood pressure tended to be better maintained, the duration of the tachyarrhythmia tended to be shorter, and relapses fewer. Furthermore, it was easier to achieve normal sinus rhythm in the treated patients without the need for additional antiarrhythmic therapy or cardioversion. Because the study was not double blind, a comparison of "treatment regimens required" in the two groups is open to debate because of the possibility of physician bias. Those attending the patients obviously knew the group to which each patient belonged. We include this information only because it was a striking feature. The question remains: Why does low-dose oral propranolol reduce the incidence of arrhythmias after aorta-coronary bypass grafting? For supraventricular arrhythmias we believe the answer is twofold. First, adrenoreceptor hypersensitivity and a rise in endogenous catecholamines occurs 2 to 6 days after withdrawal of beta blockers. I I This time frame coincides with our peak onset of supraventricular arrhythmias. Second, the autonomic nervous system, especially the sympathetic limb, is hyperactive perioperatively and probably more so after beta blocker withdrawal." It is possible, as other authors have suggested, that institution of low-dose propranolol early after the operation damps down this excitable state. Regarding ventricular arrhythmias, it is known that propranolol even in low doses has an antiarrhythmic effect. It is also known that an excess of catecholamines predisposes patients to ventricular arrhythmias. We proposed that propranolol works through its antiarrhythmic and beta blocking actions. A beta blocking effect was shown in our patients by a marked difference in sinus heart rate on day 7 postoperatively. The mean heart rate in the treatment group was 80.2 ± 10.5 beats/min and in the control group, 93.8 ± 12.2 beats/ min (p = 0.0005). We conclude that low-dose oral propranolol com-

Volume 89 Number 3 March, 1985

menced early after aorta-coronary bypass in patients receiving beta blockers preoperatively is effective in reducing all types of clinically important arrhythmias that occur in the postoperative period. Side effects are few, although four patients were withdrawn from treatment in our study. Studies of postoperative digoxin have also shown a reduction in supraventricular arrhythmias but not of ventricular arrhythmias.' 4. 8. 9 The possibility remains that administration of digoxin may increase the risk of ventricular arrhythmias in some patients early in the postoperative period, when electrolyte disturbances may occur and there is a heightened sensitivity to ventricular arrhythmias. Combined digoxin and propranolol therapy may be useful,2o.21 but for the present we prefer early institution of low-dose oral propranolol as a routine in all previously betablocked patients without a specific contraindication.

2

3

4

5

6

7

8

REFERENCES Wisoff BG, Hartstein ML, Aintablian A, Hamby RI: Risk of coronary surgery. Two hundred consecutive patients with no hospital deaths. J THORAC CARDIOVASC SURG 69:669-673, 1975 Brandt PWT, Partridge JB, Wattie WJ: Coronary arteriography. Method of presentation of the arteriogram report and a scoring system. Clin Radiol 28:361-365, 1977 Johnson LW, Dickstein RA, Fruehan CT, Kane P, Potts JL, Smulyan H, Webb WR, Eich RH: Prophylactic digitalization for coronary artery bypass surgery. Circulation 53:819-822, 1976 O'Kane H, Gre1aA, Bane A, Klerger R, Krone R, Oliver GC: Prophylactic digitalization in aorto-coronary bypass patients. Circulation 45,46:Suppl 2:199, 1975 Salazar C, Frishman W, Friedman S, Patel J, Lin YT, Oka Y, Frater WM, Becker RM: Beta blockade for supraventricular tachycardia post-coronary artery surgery. A propranolol withdrawal syndrome. Angiology 30:816819,1979 Abel RM, Van Gelder HM, Pores IH, Liguori J, Gielchinsky I, Parsonnet V: Continued propranolol administration following coronary bypass surgery. Arch Surg 118:727-731,1983 Ivey MF, Ivey TD, Bailey WW, Williams DB, Hessel EA, Miller DW: Influence of propranolol on supraventricular tachycardia early after coronary artery revascularization. A randomized trial. J THOMC CARDIOYASC SURG 85:214218, 1983 Csicsko JF, Schatzlein MH, King RD: Immediate postoperative digitalization in the prophylaxis of supraventricular arrhythmias following coronary artery bypass. J THOMC CARDIOYASC SURG 81:419-422, 1981

Propranolol for arrhythmia prophylaxis

443

9 Tyras DH, Stothert JC Jr, Kaiser GL, Barner HB, Codd JE, Willman VL: Supraventricular tachyarrhythmias after myocardial revascularization. A randomized trial of prophylactic digitalization. J THORAC CARDlOYASC SURG 77:310-314, 1979 10 Boudoulas H, Lewis RP, Kates RE, Dalamangas G: Hypersensitivity to adrenergic stimulation after propranolol withdrawal in normal subjects. Ann Intern Med 87:433-436, 1977 11 Nattel S, Rangno RE, Van Loon G: Mechanism of propranolol withdrawal phenomena. Circulation 59:11581164, 1979 12 Stephenson LW, MacVaugh H, Tomasello DN, Josephson ME: Propranolol for prevention of postoperative cardiac arrhythmias. A randomized study. Ann Thorac Surg 29:113-116, 1980 13 Mohr R, Smolinsky A, Goor DA: Prevention of supraventricular tachyarrhythmias with low dose propranolol after coronary bypass. J THORAC CARDIOYASC SURG 81:840845, 1981 14 Kirsh MM, Behrendt DM, Jackson AP, Dhadphale P, Alseri W, Brymer J, Orringer MB, Sloan H: Myocardial revascularization in patients receiving long term propranolol therapy. Ann Thorac Surg 25:117-121, 1978 15 Battler A, Ross J Jr, Slutsky R, Pfisterer M, Ashburn W, Froelicher V: Improvement of exercise induced left ventricular dysfunction with oral propranolol in patients with coronary heart disease. Am J Cardiol 44:318-324, 1979 16 Gettes LS, Surawiez B: Long term prevention ofparoxysmal arrhythmias with propranolol therapy. Am J Med Sci 254:257-265, 1967 17 Gianelly R, Griffin JR, Harrison DC: Propranolol in the treatment and prevention of cardiac arrhythmias. Ann Intern Med 66:667-676, 1967 18 Howard EB: American Medical Association Drug Evaluations, ed 2, Acton, Mass., 1973, American Medical Association, p 17 19 Miller RR, Olson HG, Amsterdam EA, Mason DT: Propranolol withdrawal rebound phenomenon. Exacerbation of coronary events after abrupt cessation of antianginal therapy. N Engl J Med 293:416-418,1975. 20 Roffman JA, Fieldman A: Digoxin and propranolol in the prophylaxis of supraventricular tachydysrhythmias after coronary artery bypass surgery. Ann Thorac Surg 31:496501, 1981 21 Parker FB, Greiner-Hayes C, Bove EL, Marvasti MA, Johnson LW, Eich RH: Supraventricular arrhythmias following coronary artery bypass. The effect of preoperative digitalis. J THOMC CARDIOYASC SURG 86:594-600, 1983 22 Campeau L: Grading of angina pectoris. Circulation 54:522-523, 1976