Implantable cardioverter defibrillator compared with antiarrhythmic drug treatment in cardiac arrest survivors (the Cardiac Arrest Study Hamburg)

SESSION

VI. CLINICAL

EXPERIENCE

WITH ICD THERAPY

Implantable cardioverter defibrillator compared with antiarrhythmic drug treatment in cardiac arrest survivors (the Cardiac Arrest Study Hamburg) In 1987, the Cardiac Arrest Study Hamburg (CASH), a prospective, multicenter, randomized controlled study, was started in survivors of sudden cardiac death resulting from documented ventricular tachyarrhythmias. Through December 1991, 230 survivors (48 women, 184 men; mean age 57 + 11 years) of cardiac arrest caused by ventricular tachyarrhythmias were randomly assigned to receive either oral propafenone (58 patients), amiodarone (58 patients), or metoprolol (59 patients) or to have an implantable defibrillator (59 patients) without concomitant antiarrhythmic drugs. The primary endpoint of the study was total mortality. In March 1992, the propafenone arm of CASH was stopped because of excess mortality compared with the implantable defibrillator group. This article presents preliminary results of the comparison of implantable defibrillator therapy with propafenone therapy. A significantly higher incidence of total mortality, sudden death (12%), and cardiac arrest recurrence or sudden death (23%) was found in the propafenone group compared with the implantable defibrillator-treated patients (O%, p < 0.05). It was concluded that, in survivors of cardiac arrest, propafenone treatment is less effective than implantable defibrillator treatment. (AM HEART J 1994;127:1139-44.)

Jiirgen Siehels, MD, Karl-Heinz Hamburg, Germany

Kuck, MD, and the CASH Investigabors

Despite major advances in experimental and clinical electrophysiology in the last 20 years, sudden cardiac death remains a complex pathophysiologic entity. Ninety thousand to 100,000 sudden deaths per year are estimated to occur in Germany,lT3 accounting for approximately 50 % of all cardiovascular deaths. For those patients who survive an episode of sudden cardiac arrest, the risk of recurrence in the first 2 years if no therapy or only empiric therapy is offered may be as high as 40 % .3-5In the absence of myocardial ischemia, monomorphic ventricular tachycardia degenerating into polymorphic tachycardia, ventricular fibrillation, or both, has been reported in 70% of documented episodes of sudden cardiac death.6,7 Three major options must be considered to prevent recurrences of sudden cardiac death: first, antiarrhythmic therapy guided by either programmed electrical stimulation or Holter monitoring; second, From

the Department

of Cardiology,

Reprint requests: Jiirgen Medical Clinic, Department burg, Germany. Copyright 14 1994 0002.8703/94/$3.00

University

Siebels, MD, of Cardiology,

by Moshy-Year + 0 4/O/53002

Book,

Hospital

Eppendorf.

University Hospital Eppendorf, Martinistrasse 52,20246, HamInc.

empiric amiodarone therapy; and third, implantation of a cardioverter defibrillator (ICD). Until now, no study has been available to demonstrate superiority of one of these options over the others. Programmed electrical stimulation, however, is limited by the fact that a sustained ventricular arrhythmia is inducible in only 50 % of survivors resuscitated from ventricular fibrillation.4, 8-11Of these, only 10 % to 30 % can be suppressed by antiarrhythmic drug therapy, if class I drugs are used. After 2 years, the sudden death rate is 12 % ; in contrast, it is 30 % to 40 % in patients in whom arrhythmias remain inducible despite antiarrhythmic drug therapy. 4lg-12 Empiric amiodarone applied in patients who either survived sudden cardiac death or had a sustained ventricular tachycardia leads also to a sudden death rate of 12 % after 2 years. In contrast, ICD therapy has been shown to be associated with a sudden death rate of only 4% after 4 years and a total mortality ranging from 8 % at 1 year to 20% at 4 years. l3 The mean total mortality after 2 years is far below the sudden death rate after either amiodarone or class I antiarrhythmic treatment guided by programmed electrical stimulation. However, these data derive from different studies. No 1139

1140

Siebels et al.

randomized data comparing antiarrhythmic drug and defibrillator efficacy are available at present. In 1987, a prospective, multicenter, randomized controlled study was started in survivors of sudden cardiac death resulting from documented ventricular tachycardia or ventricular fibrillation or both, to compare the incidence of sudden death, cardiac mortality, and total mortality among four treatment groups: amiodarone, propafenone, metoprolol, and the ICD. After an average of 11 months of follow-up, the propafenone arm was prematurely stopped by the safety board of the study, because total mortality and recurrences of cardiac arrest differed significantly from those of ICD-treated patients. The study is still continuing for the other limbs: amiodarone, metoprolol, and the ICD. METHODS

In 1987, the Cardiac Arrest Study Hamburg (CASH) was started, involving several centers (Appendix). CASH is a randomized controlled study of survivors of suddencardiac death caused by documented ventricular tachycardia, fibrillation, or both, unrelated to myocardial infarction. The aim of the study is to compare the incidence of recurrence of cardiac arrest, suddencardiac death, cardiac mortality, and total mortality amongpatients treated with antiarrhythmic drugs or with an ICD. The projected enrollment includes400 patients, whoseadmissionat the various centers needsto occur within 3 months after the cardiac arrest event. All patients undergo complete physical examination, standard laboratory analysis, chest x-ray examination, 12-leadsurface ECG, echocardiogram,exercise testing, 24-hour Holter monitoring, and coronary and left ventricular angiography. Myocardial thallium-201 scintigraphy completesthe exercise testing when myocardial ischemiaor exercise-inducedarrhythmias or both have been documented. If no structural heart diseaseemergesfrom these examinations, right heart angiography, ergonovine test during coronary angiography, right heart biopsy, and magnetic resonanceimaging are additionally performed. Programmed electrical stimulation is performed in all patients before and after randomization. Patients with severe ischemia undergo revascularization, with exercise testing, 24-hour Holter monitoring, and programmedelectrical stimulation repeated before and after revascularization. During programmed electrical stimulation with use of the extrastimulus technique (0.5 msecpulse width, twice diastolic threshold), up to three extrastimuli are given at two right ventricular sites (apex and outflow tract) during sinusrhythm and at three basicpacing drives of eight beats (640,510, and 440 msec).The study protocol wasapproved by the ethical committee of the University of Hamburg. Before randomization, written informed consent is given by all patients. After electrophysiologictesting, all patients are randomized independent of clinical findings or testing results to receive either amiodarone(loading dose1000mg/day for 7

American

April 1994 Heart Journal

days and 400 to 600 mg/day from day 8 on), metoprolol (starting at a low initial doseof 12.5 to 25 mg/day; if tolerated, the doseis advanced during 10 to 20 days up to the maximally tolerated dose or 200 mg/day), propafenone (starting at a low initial doseof 450mg/day; if tolerated, the doseis advanced during 8 to 14 days up to the maximally tolerated doseor 900 mg/day), or primary implantation of an ICD with no concomitant antiarrhythmic treatment. Cardiac Pacemakers,Inc. (St. Paul, Minn.) devicesare being used in the present study (AID-B, Ventak 1500-1520, Ventak P 1600 and Ventak PRx 1700-1705).By protocol design,during doseranging, investigators assess tolerance and safety, not efficacy of antiarrhythmic drugs, by dayto-day review of adverse experiences,concomitant medications, and ECG analysison the maximally tolerated dosage. After randomization, exercise testing, 24-hour Holter monitoring, and programmedelectrical stimulation are repeatedin all patients in a steady state. Resultsof thesetests are not allowed to influence the treatment but may ultimately provide information about the predictive power of the tests. In all programmableICDs, the cut-off rate is set between 170 and 200 beats/min. Shock delivery is chosenas a primary interventional option in all patients, independent of the type of tachyarrhythmia responsible for the first recurrence. If available, antitachycardia modalities can be activated during repeated episodesof documented ventricular tachycardia if shock delivery proves to be poorly tolerated by the patient. Complete clinical and instrumental evaluation (ECG, exercise testing, 24-hour Holter monitoring) is performed in eachpatient during follow-up; in patients randomized to the antiarrhythmic drug arms, this occurs at months 2, 4, 6, 12, 18, and 24; in patients in the ICD arm, a full additional system check-up test is scheduledevery 2 months. The primary endpoint of the study istotal mortality. The secondary endpoints of the study are (1) recurrences of cardiac arrest requiring cardiopulmonary resuscitation; (2) recurrences of hemodynamically unstable ventricular tachycardia; (3) incidence of drug withdrawal becauseof side effects or other reasons;(4) incidence of heart transplantation requirement. In patients with an ICD, thosedischarges occurring during syncopeare counted asventricular fibrillation recurrences,and those occurring during presyncope, documented ventricular tachycardia, or both are counted asventricular tachycardia recurrences;dischargesof which the nature remains unknown are not classified. For statistical analysis,continuous data are expressedas means + 1 SD. Comparisonsbetween groupsare madeby unpaired t-tests or chi-square analysis, as appropriate. A value of p < 0.05 is consideredsignificant. Kaplan-Meier analysis and traditional two-sided log-rank tests are used to comparetime-dependent outcome of the patients on an intention-to-treat basis.Subanalysisamongthe four treat-

ment groups is allowed only if the multiple group analysis showsa significant difference (p < 0.05) amongall groups. Multivariate Cox regressionanalysis is used to determine predictors of outcome. Variables included in this multivariate analysis are listed in Table I.

Volume 127, Number 4, Part 2 American Heart Journal

Table

I. Demographic

Siebels et al.

data of 230 cardiac

arrest survivors ZCD

59

n M/F

Age (yr) (mean f SD) LVEF (5%) (mean k SD) Heart disease Coronary artery disease Coronary spasm Dilated cardiomyopathy Hypertrophic cardiomyopathy Arrhythmogenic dysplasia Valvular disease Mitral valve prolapse No organic heart disease LVEF,

Left

ventricular

ejection

randomly

assigned

Amiodarone

to either treatment Metoprolol

Propafenone

59

56 46/10 58 + 10 38 * 15

44 k 17

56 45/11 57 * 13 43 ? 18

41 2

45

50

46

0

0

7 1

4

1

1 1

0 0

0 0 1

0 1 2

0 7

0

47112 56 r 11 42 f 16

1 0 2

46/13 57 k

2

0

5

1141

5

11

5

fraction.

RESULTS

During the annual meeting of the CASH Safety Monitoring Board in March 1992, all data collected through December 15, 1991, were reviewed. This represented an average follow-up of 11 months for all study patients. Analysis of the primary endpoint revealed no significant differences, as of this date, among the three populations treated with amiodarone, metoprolol, and the ICD. Conversely, a significantly higher incidence of total mortality, sudden death, and cardiac arrest recurrence was found in the propafenone arm when compared with the ICD arm. The results of the other study groups are not presented because the Safety Monitoring Board and the ethical committee recommended the study to be continued with amiodarone, metoprolol, and ICD treatment. Through December 1991,230 patients (46 women, 184 men; mean age 57 -t 11 years, range 19 to 78 years) were included in the study. The underlying heart disease in each group is given in Table I. Patients in each group did not differ in age, sex, left ventricular ejection fraction, and underlying heart disease. Coronary artery disease prevailed in all groups (about 80 % ), whereas in almost 10 % of cases in each group, full clinical and instrumental evaluation did not reveal any overt heart disease. We present here only a cursory review of the ICD and propafenone data, as thorough statistical analysisis presently being performed. In detail, 56 patients comprised the propafenone arm and 59 the ICD arm. Coronary revascularization was performed in 14 patients at implantation of the cardioverter defibrillator and in 11 patients before propafenone treatment. Programmed electrical stimulation in the off-drug status resulted in inducible sustained monomorphic ventricular tachycardia (>30 seconds) in 16/56 pa-

Table

II. All events leading

to study endpoint

within

24

months in 115 of 230 of suddencardiac death survivors ZCD (n = 59)

Total mortality Sudden death Cardiaclnonsudden Noncardiac Discontinuation of therapy Recurrent cardiac arrest Ventricular tachycardia Other arrhythmias Side effects Cardiac transplantation ICD explantation (infection) *Occurrence ICensored

after discontinuation only for natural history

Propafenone (n = 56)

9

(+7)* 6 (+4)* 3 (+2)*

0 (+l)*

@)t

(9)t m 0 0 2

of propafenone. comparison.

tients taking propafenone and in 13159 patients with an ICD; inducible polymorphic ventricular tachycardia or fibrillation was observed in 2156 and 4159 additional patients, respectively. After propafenone administration, ventricular tachycardia remained inducible in 10 patients; in 3 of the remaining 8 patients, the second electrophysiologic test was not performed because of spontaneous occurrence of ventricular tachyarrhythmias in 2 and refusal of a second test in 1. Therefore, 5 of 18 patients (28%) were apparently protected by drug administration. Of the 37 patients in whom ventricular tachycardia was not inducible before randomization, 5 had inducible sustained ventricular tachycardia after propafenone. All primary and secondary endpoints in both study groups are given in Table II. Sudden death occurred in none of the patients with an ICD and in six of the patients taking propafenone (p < 0.05) (Fig. 1). In

1142

Siebels et al.

American

l

April 1994 Heart Journal

I -

1, I 1-w--1-1

I L--I---,,,,,,

z 0.92 4 z g 0.80

-

ICD (n = 59)

- - -

Propafenone (n = 56) p < 0.05

I,

0.7 0

2

I ,I,,,,,,,,,,,,,,,,,,, 4 6 8 10 12 14 16 months after randomization

Fig. 1. Sudden death survival among 115 cardiac arrest

survivors

, 18 randomly

20

22 assigned

24 to ICD

or

propafenone treatment. Calculations were basedon all causesof sudden or unwitnesseddeaths. Nominal p value was basedon a traditional two-sided log-rank test.

this latter group, five additional patients had a recurrence of cardiac arrest during follow-up and were successfully resuscitated. Furthermore, the drug had to be discontinued in 12 more cases; in five patients this was caused by sustained, monomorphic ventricular tachycardia and in two by frequent episodes of nonsustained ventricular tachycardia. Of the five patients taking propafenone who were protected because of stimulation results, two (40%) had arrhythmia occurrences, one had cardiac arrest, and one had ventricular tachycardia. Of the 10 patients who had tachycardia still inducible and who were taking propafenone, two (20%) died suddenly. Among the 27 patients without inducible arrhythmias in both electrophysiologic studies, five (19%) arrhythmic events occurred (two sudden deaths, two cardiac arrests, and one ventricular tachycardia) and two patients died nonsuddenly. None of the five patients taking propafenone with proarrhythmic results in the electrophysiologic study had any arrhythmic event or died. Of five patients without inducible arrhythmia in the off-drug status, in whom no electrophysiologic study while taking propafenone could be performed, four had arrhythmic events (two sudden deaths, one cardiac arrest, and one ventricular tachycardia) during follow-up while they were receiving propafenone treatment. Recurrences of ventricular fibrillation (eight patients) and ventricular tachycardia (nine patients) in

the patients with an ICD were censored only to calculate the “natural history” reported in Fig. 2. Three of eight patients with true-positive ICD discharges caused by ventricular fibrillation died within 2 months after this event. No sudden death or cardiac arrest occurred in the ICD group compared with 12% sudden deaths and 23% sudden deaths or cardiac arrests in the propafenone group (p < 0.05; Figs. 1 and 2). The incidence of true-positive discharges during syncope (natural history) was similar in the ICD group (23 % ) to the risk of sudden death or cardiac arrest in the propafenone-treated patients (Fig. 2). DISCUSSION

The incidence of recurrent cardiac arrest alone is reduced by 23 % with the ICD compared with antiarrhythmic treatment with the class IC drug propafenone. Though previous reports13 suggested a similar conclusion, this is the first contribution where the benefit of primary implantation of an ICD was assessedin the absence of any concomitant antiarrhythmic drug therapy in a prospective, randomized fashion. The incidence of arrhythmic events in the propafenone arm is comparable to that in other investigations of survivors of cardiac arrest who have undergone serial electrophysiologic testing.4T 5,g-12 In this present study, the results of the electrophysiologic study with propafenone, including the

Volume 127, Number 4, Part 2 American Heart Journal

Siebels et al.

LI,

099-

7i E .z o 0,8$ 3

9 L - m-4---I------.--, 1 --1-m-

&0,7-

l ,--,,-----I’-

-

ICD (n = 59)

- - 11-1.

Propafenone (n = 56)

1143

-.m---mm--. I L III

“natural history” (ICD - syncope) p < 0.05

096

III

0

I

2

4

I

I

6

I

I

I

I

I

I

I

I

I

I

10 12 14 16 8 months after randomization

I

I

18

I

I

20

I

I

22

I

I

I

24

Fig. 2. Sudden death survival and recurrence of cardiac arrest among 115 cardiac arrest survivors randomly assignedto ICD or propafenone treatment. Calculations were basedon all causesof suddenor unwitnesseddeathsand additional recurrencesof cardiac arrest in both groups.Natural history curve indicates all first ICD shocksduring syncopeor documented ventricular fibrillation. Nominal p value was basedon a traditional two-sided log-rank test.

noninducibility of previously sustained arrhythmias or proarrhythmic effects, did not influence the incidence of arrhythmic events or total mortality. These data appear different from those in previous studies4>5,g, lo in which a high number of antiarrhythmic drugs were tested but are in agreement with recent observationsl’, l2 in which a low predictive value for recurrences of ventricular tachycardia or fibrillation was shown even in patients with suppressed arrhythmias. Such differences may be explained in the definition of sustained arrhythmias, stimulation protocols, and the definition of successful antiarrhythmic treatment according to electrophysiologic testing. In addition, the differences might be caused by the relatively small number of patients in each treatment arm or by inadequate specificity of the electrophysiologic study. Recently reported data on the multicenter Electrophysiologic Study Versus Electrocardiographic Monitoring study14, l5 may indicate that noninducibility of ventricular arrhythmias is lesspredictive than suggested by studies reported in the early 1980s. Studies with class I antiarrhythmic treatment14-l7 in different patient populations indicate that these drugs have no benefit or may even increase the risk of fatal events despite effectiveness according to 24-hour Holter monitoring or electrophysiologic study.14, l5

Conclusion. Our data suggest that in survivors of cardiac arrest, long-term treatment with oral propafenone is associated with a significantly higher incidence of death as compared with ICD treatment. Therefore, in this high-risk population, propafenone should not be recommended.

REFERENCES

5.

6. 7. 8.

9.

10.

Gillum RF. Sudden coronary death in the United States, 1980-1985. Circulation 1989;79:756-65. Kannel WB, Schatzkin A. Sudden death: lessons from subsets in population studies. J Am Co11 Cardiol 1985;5:141B-9B. Rapaport E. Sudden cardiac death. Am J Cardiol 1988;62:3161. Wilber DJ, Garan H, Finkelstein D, Kelly E, Newell J, McGovern B, Ruskin JN. Out-of-hospital cardiac arrest: use of electrophysiologic testing in the prediction of long-term outcome. N Engl J Med 1988;318,1:19-24. Freedman R, Swerdlow CD, Soderholm-Diffatte V, Mason JW. Prognostic significance of arrhythmia inducibility or non-inducibility at initial electrophysiologic study in survivors of cardiac arrest. Am J Cardiol 1988,61:578-82. Kempf FC, Josephson ME. Cardiac arrest recorded on ambulatory electrocardiograms. Am J Cardiol 1984;53:1577-82. Fozzard HA. Electromechanical dissociation and its possible role in sudden cardiac death. Am J Cardiol 1985;5,6:31b-4b. Swerdlow CD, Winkle RA, Mason JW. Determinants of survival in patients with ventricular tachyarrhythmias. N Engl J Med 1983;308:1436-42. Ruskin JN, DiMarco JP, Garan H. Out-of-hospital cardiac arrest: electrophysiologic observations and selection of longterm antiarrhythmic therapy. N Engl J Med 1980;303:607-12. Roy D, Waxman HL, Kienzle MG, Buxton AE, Marchlinski

1144

11.

12.

13.

14.

15.

16.

17.

Siebels et al.

FE, Josephson ME. Clinical characteristics and long-term follow-up in 119 survivors of cardiac arrest: relation toinducibilitv at electrouhvsioloaic testinn. Am J Cardiol1983:52:969-74. Morady F, Scheinman MM, Hess DS, Sung RJ, Shen E, Shapiro W. Electrophysiologic testing in the management of survivors of out-of-hospital cardiac arrest. Am J Cardiol 1983; 51:85-9. Poole EP, Mathisen TL, Kudenchuk PJ, McAnulty JH, Swerdlow CD, Bardy GH, Greene HL. Long-term outcome in patients who survive out of hospital ventricular fibrillation and undergo electrophysiologic studies: evaluation by electrophysiologic subgroups: J Am Co11 Cardiol 1990;16:657-65. Winkle RA. Mead RH. Ruder MA. Gaudiani VA. Smith NA. Buch WS, Schmidt P,‘Shipman T: Long-term outcome with the automatic implantable cardioverter-defibrillator. J Am Co11 Cardiol 1989;13:1353-61. The ESVEM Investigators. The ESVEM trial. Electrophysiologic Study Versus Electrocardiographic Monitoring for selection of antiarrhythmic therapy of ventricular tachyarrhythmias. Circulation 1989;79:1354-60. Mason JW for the Electrophysiologic Study Versus Electrocardiographic Monitoring investigators. A comparison of electrophysiologic testing with Holter monitoring to predict antiarrhythmic-drug efficacy for ventricular tachyarrhythmias. N Engl J Med 1993;329:445-51. The Cardiac Arrhythmia Suppression Trial (CAST) Investigators. Preliminary report: effect of encainide and flecainids on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;321:406-12. The Cardiac Arrhythmia Suppression Trial II Investigators. Effect of the antiarrhythmic agent moricizine on survival after myocardial infarction. N Engl J Med 1992;327:227-33.

American

April 1994 Hearl Journal

APPENDIX: PRIMARY INVESTIGATORS OF THE CARDIAC ARREST STUDY IN HAMBURG

Prof. Dr. K. H. Kuck, Dr. J. Siebels,Dr. R. Riippel, Dr. M. A. E. Schneider, Universitiitskrankenhaus Eppendorf, Hamburg Prof. Dr. P. Kalmar, Dr. Kalkowski, Universittitskrankenhaus Eppendorf, Hamburg Prof. Dr. H. Greten, Dr. Kreymann, Universitiitskrankenhaus Eppendorf, Hamburg Prof. Dr. Spiller, Dr. Heihn, Allgemeines Krankenhaus Barmbek, Hamburg Prof. Dr. Ostendorf, Dr. Kleinebenne, Marienkrankenhaus, Hamburg Prof. Dr. von Ohlshausen,Allgemeines Krankenhaus Altona, Hamburg Prof. Dr. Sill, AllgemeinesKrankenhaus Wandsbek, Hamburg Prof. Dr. Pop, Dr. Kleinert, Allgemeines Krankenhaus Harburg, Hamburg Prof. Dr. T. Meinertz, Dr. Volkmann, AllgemeinesKrankenhaus St. Georg, Hamburg Prof. Dr. Hanrath, Dr. C. Stellbrink, RTWH Aachen, Aachen