Mortality Following Ventricular

Arrhythmia Suppression by Encainide,

Flecainide, and Moricizine
After Myocardial Infarction
The Original Design Concept of the
Cardiac Arrhythmia Suppression Trial (CAST)
Andrew E. Epstein, MD; Alfred P. Hallstrom, PhD; William J. Rogers, MD; Philip R. Liebson, MD;
A. AllenSeals, MD; Jeffery L. Anderson, MD; Jerome D. Cohen, MD; Robert J. Capone, MD;
D. George Wyse, MD, PhD; for the CAST Investigators

Objective.\p=m-\Totest the hypothesis that in survivors of myocardial infarction, the ALTHOUGH mortality from cardiovas¬
suppression of ventricular premature depolarizations improves survival free of car- cular disease has decreased over the last
diac arrest and arrhythmic death. 10 years, sudden cardiac death remains
a medical problem of epidemic propor¬
Design.\p=m-\International,prospective, multicenter, randomized, placebo-con- tion. It is estimated that over 250000
trolled trial.
persons die suddenly each year in the
Setting.\p=m-\Universityand community hospitals. United States.1·2 Most of these deaths
Patients.\p=m-\Atotal of 3549 patients with myocardial infarction and left ventricular are believed to be a consequence of ven¬
dysfunction. tricular tachyarrhythmias, and only un¬
Intervention.\p=m-\Administrationof encainide, flecainide, moricizine, or placebo to commonly is bradycardia responsible for
suppress ventricular premature depolarizations. sudden death.13 The majority of patients
Main Outcome Measures.\p=m-\Overallsurvival and survival free of cardiac arrest who experience sudden death have coro¬
or arrhythmic death were compared in patients randomized to long-term, active nary artery disease, often with prior
antiarrhythmic drug therapy vs corresponding placebo, using the stratified log rank myocardial infarction.1,2
statistic. The Cardiac Arrhythmia Suppression
Results.\p=m-\At1 year from the time of randomization to blinded therapy, 95% of Trial (CAST) was designed as a multi-
center, randomized, placebo-controlled
placebo-treated patients vs 90% of active drug\p=m-\treatedpatients remained alive trial to test the hypothesis that in pa¬
(P=.0006). Similarly, at 1 year, 96% of placebo-treated patients vs 93% of active tients with prior myocardial infarction,
drug\p=m-\treatedpatients remained free of cardiac arrest or arrhythmic death (P=.003). the suppression of ventricular prema¬
Conclusions.\p=m-\Thesuppression of asymptomatic or mildly symptomatic ven- ture depolarizations improves survival
tricular arrhythmias after myocardial infarction does not improve survival and can free of arrhythmic death.46 Thus, the
increase mortality. Treatment strategies designed solely to suppress these ar- suppression of ventricular arrhythmia
rhythmias should no longer be followed. was the focus of the study, and rather
(JAMA. 1993;270:2451-2455) than being a test of particular agents,
CAST was structured as a trial that
might be generalized to antiarrhythmic
From the Division of Cardiovascular Disease, De- partment of Medicine, University of Rochester (NY) (Dr drugs other than those used in the trial.
partment of Medicine, The University of Alabama at Capone); Division of Cardiology, Department of Medi- The drugs used in CAST were chosen
Birmingham (Drs Epstein and Rogers); CAST Coordi- cine, The University of Calgary (Alberta) (Dr Wyse).
nating Center, University of Washington, Seattle (Dr For a complete list of participants and participating based on theoretical and practical con¬
Hallstrom); Section of Cardiology, Department of Medi- institutions, see The Cardiac Arrhythmia Suppression siderations derived from the Cardiac Ar¬
cine, Rush-Presbyterian-St Luke's Medical Center, Chi- Trial II Investigators. Effect of the antiarrhythmic agent
cago, III (Dr Liebson); Section of Cardiology, Depart- moricizine on survival after myocardial infarction. rhythmia Pilot Study (CAPS).7·8 A Drug
ment of Medicine, University of Florida-Jacksonville N Engl J Med. 1992;327:227-233. Selection Committee considered drugs
(Dr Seals); Cardiology Division, Department of Medi- Reprint requests to CAST Coordinating Center, 1107 in all Vaughan Williams classes (class I
cine, LDS Hospital, Salt Lake City, Utah (Dr Anderson); NE 45th St, Room 505, Seattle, WA 98105 (Ms Margit
Division of Cardiology, Department of Medicine, St Scholz). [sodium channel blockers such as quin-
Louis (Mo) University (Dr Cohen); Cardiology Unit, De- idine, procainamide, disopyramide, mexi-

Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/jama/9825/ by a University of Alberta User on 02/17/2017

 letine, tocainide, phenytoin, encainide,
flecainide, propafenone, moricizine, and
other investigational agents], class II
[ß-blockers], class III [drugs that pro¬
long action potential duration such as
amiodarone and sotalol], and class IV
[calcium channel blockers]).9 In CAPS,
four class I drugs (encainide, flecainide,
moricizine, and imipramine) and a pla¬
cebo were tested to determine the fea¬
sibility of performing a full-scale anti-
arrhythmic drug trial. Other drugs were
excluded for a variety of reasons, the
major ones being anticipated intolerance
during long-term treatment, inefficacy
to suppress ventricular arrhythmias,
toxicity, and limited data for use in the
clinical setting. At the time CAST was
planned, amiodarone was considered a
too toxic and controversial agent to be
included. Furthermore, its long half-life
would have complicated titration with
other drugs had it proved inefficacious
to suppress the index ventricular ar¬
rhythmia.10
As a result of CAPS, CAST was ini¬
tiated using encainide, flecainide, mori¬
cizine, and matching placebos, because
all three active drugs showed a favor¬
able profile for arrhythmia suppression
and patient tolerance. Encainide and
flecainide both reduce the maximal ve¬
locity of phase 0 depolarization and have
minimal effects on action potential du¬
ration (Vaughan Williams subclass IC),
and moricizine shares characteristics
with drugs in all three class I sub¬
classes.11 As data were gathered, it be¬
came evident that encainide and flecai¬
nide worsened survival, and these two
arms of the trial (called CAST-I) were
stopped on recommendation of the Data
and Safety Monitoring Board.4·5 The
moricizine arm of the trial was contin¬
ued as CAST-II. However, when it be¬
came apparent that moricizine could not
improve survival and was in fact detri¬
mental during open-label titration,
CAST-II was terminated on August 1,
1991.6
This report describes the overall sur¬
vival and survival free from cardiac ar¬
rest or arrhythmic death for all patients
enrolled in CAST-I and CAST-II. By
focusing on suppression of ventricular
arrhythmia and not on the specific drugs
used, the report addresses the original
CAST hypothesis that suppression is
beneficial.
METHODS
Patients were screened for participa¬
tion in CAST after documented myo-
cardial infarction. The designs of CAST-I
and CAST-II have been previously de¬
scribed in detail.4"6·10 In CAST-I, patients
were eligible to be screened for enroll¬
ment between 6 days and 2 years fol-
Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/jama/9825/ by a University of Alberta User on 02/17/2017
lowing documented myocardial infarc¬
tion. The arrhythmia requirement for
enrollment was an average of six or more
ventricular premature depolarizations
per hour during an ambulatory electro-
cardiographic recording with a minimum
of 18 hours of analyzable data. To enroll
potentially high-risk patients and main¬
tain adequate statistical power, a left
ventricular ejection fraction of 0.55 or
less was required if the recording was
obtained between 6 and 90 days follow¬
ing the qualifying myocardial infarction,
but 0.40 or less if the recording was
obtained between 90 days and 2 years
following the qualifying myocardial in¬
farction. Patients were excluded from
the trial if they had severely symptom¬
atic ventricular arrhythmias (incapaci¬
tating symptoms, syncope, or presyn-
cope) or if they had 15 or more consecu¬
tive ventricular premature depolariza¬
tions at a rate of 120 or more beats per
minute.
In CAST-I, patients were random¬
ized to receive encainide, flecainide,
moricizine, or a matching placebo, and
in CAST-II, patients were randomized
to receive either moricizine or a match¬
ing placebo. Furthermore, to focus re¬
cruitment on patients at higher risk for
arrhythmic death, and thus potentially
most likely to benefit from antiarrhyth-
mic therapy, entry criteria were changed
in CAST-II. The major changes in
CAST-II were as follows: (1) the left
ventricular ejection fraction was 0.40 or
less in all patients, (2) the enrollment
time window was changed to include
only the period from 4 days to 90 days
following the index myocardial infarc¬
tion, and (3) disqualifying ventricular
tachycardia was defined as symptom¬
atic, or lasting at least 30 seconds at a
rate of at least 120 beats per minute.
Also, since there were no control data
on the initiation of treatment in CAST-I
in which drugs were administered in an
unblinded fashion with open-label titra-
tion, CAST-II was modified to begin
with a 2-week blinded, randomized, pla¬
cebo control to study the early effects of
moricizine treatment.6
Patients were randomized to receive
up to three drugs (encainide, flecainide,
or moricizine) in two doses in CAST-I
and up to three doses of moricizine in
CAST-II, or their matching placebos.
Doses consisted of encainide, 35 and
50 mg three times per day; flecainide,
100 and 150 mg twice a day; and mori¬
cizine, 200 and 250 mg three times per
day in CAST-I and up to 300 mg three
times per day in CAST-II. The lowest
dose of drug was always the first dose
tried. The criteria for arrhythmia sup¬
pression, as measured by 24-hour am¬
bulatory electrocardiographic record-
ings, were an 80% or greater reduction
in the frequency of ventricular prema¬
ture depolarizations and a 90% or greater
reduction in episodes of unsustained ven¬
tricular tachycardia without intolerable
adverse effects. Titration was discon¬
tinued and the patient randomized to
blinded therapy as soon as a drug and
dose were found that were well toler¬
ated and that suppressed the ventricu¬
lar arrhythmia. At that time the patient
was assigned to receive either the ac¬
tive drug and dose that suppressed the
ventricular ectopy or a matching pla¬
cebo. These patients were defined as
having arrhythmia suppression.
Patients with partial arrhythmia sup¬
pression were defined as those who
achieved some but not full suppression
criteria with at least one drug and dose.
If only partial suppression was achieved,
and the drug was well tolerated, the
patient was randomized to placebo or
the best drug and dose and followed up
in a substudy.46 Patients who were still
in the open-label titration phase when
the trials ended, whose arrhythmia in¬
creased or was unchanged during the
open-label titration phase (ie, they failed
to have either suppression or partial sup¬
pression as defined above), or who were
intolerant of drugs, died before random¬
ization, or withdrew for any reason, were
not randomized.
Patients whose arrhythmias were suc¬
cessfully suppressed or partially sup¬
pressed were scheduled for follow-up
visits at 4-month intervals. Nonrandom-
ized patients were followed up by tele¬
phone at 6-month intervals. The primary
end point in both CAST-I and CAST-II
was arrhythmic death (defined as wit¬
nessed instantaneous death in the ab¬
sence of congestive heart failure or shock
or unwitnessed death with no preceding
change in symptoms, and for which no
other cause could be ascribed) or car¬
diac arrest (with resuscitation, provided
that both cardiopulmonary resuscitation
and defibrillation were required).4·5
End-point events were reviewed by an
Events Committee of CAST investiga¬
tors who did not know the patients' as¬
signed treatments. The protocol was
approved by each participating clinical
center's institutional review board. Writ¬
ten, informed consent was obtained from
all patients who participated in the study,
and CAST was monitored independently
by a Data and Safety Monitoring Board.
STATISTICAL ANALYSIS
Overall survival and survival free from
cardiac arrest or arrhythmic death were
compared in patients randomized to long-
term, active antiarrhythmic drug
therapy vs corresponding placebo, us¬
ing the stratified log rank statistic.12 Ac-
 Characteristics of 3549 Patients Who Entered
CAST Titration*
Age, y (meaniSD)
Male, %
Ejection fraction (mean±SD)
VPD/h (mean±SD)
Baseline ECG
PR interval (meaniSD, s)
QRS interval (mean±SD, s)
History before index myocardial
infarction, %
CHF
Angina
Cardiac arrest
Ventricular tachycardia
Prior myocardial infarction
CABGorPTCA
Hypertension
Diabetes
History after index
myocardial
infarction, %
No CHF, or NYHA functional
class I
No angina, or Canadian
angina class I
Thrombolytic therapy
PTCA
CABG
Baseline examination
Sitting HR, bpm (mean±SD)
Systolic BP, mm Hg (mean±SD)
Diastolic BP, mm Hg (mean±SD)
Concurrent drugs, %
ß-Blocker
Calcium channel blocker
ACE inhibitor
Digitalis
Nitrate
Diuretic
"CAST indicates Cardiac Arrhythmia Suppression
Trial; VPD, ventricular premature depolarization; ECG,
electrocardiogram; CHF, congestive heart failure; CABG,
coronary artery bypass grafting; PTCA, percutaneous
transluminal coronary angioplasty; NYHA, New York
Heart Association; HR, heart rate; bpm, beats per minute;
BP, blood pressure; and ACE, anglotensin converting
enzyme.
cordingly, the survival plots are based
on weighted averages of the stratum-
specific Kaplan-Meier survival estimates
and are necessarily restricted to the
shortest follow-up among all strata. The
strata used were based on study design,
namely, CAST-I or CAST-II, ejection
fraction less than 0.30 or 0.30 or greater,
and the particular antiarrhythmic drug
received after randomization: encainide,
flecainide, or moricizine. Patients were
counted only according to their initial
randomization, ie, the 216 patients who
were randomized to moricizine in CAST-
II after having been withdrawn from
randomized therapy in CAST-I did not
have their CAST-II experience included
in the comparison.
Since CAST was designed to examine
the role of arrhythmia suppression and
not drug exposure, survival time was
measured beginning from the time pa¬
tients were randomized to long-term
blinded therapy rather than from the
time they entered into drug titration.
Comparison was made by intention to
treat with censoring occurring accord¬
ing to study design, ie, April 18, 1989,
for patients randomized during CAST-I
to encainide or flecainide or their match¬
ing placebos, and August 1, 1991, for
CAST-I CAST-II
61.8±9.8 2371 Rerandomized to 1178 New Patients
81.3
Randomized to Moricizine From CAST-I Entered After
0.36±0.10
Flecainide or 1593-292 CAST-I Finished
129.6±245.8 Encainide 1498 Suppressed
95 Partially
0.17±0.03 Death/Cardiac Arrest
0.09±0.02 Suppressed
63
Not Randomized
16.4
46.8 Other
3.0
3.4
42.9 Randomized to
19.2 Moricizine 360 216 798
32.4 309 Suppressed 162 Suppressed 684 Suppressed
21.6 51 Partially 54 Partially 114 Partially
Suppressed Suppressed Suppressed
83.8 Fig 1.—Flow diagram of the distribution of patients who entered CAST-I and CAST-II is shown. CAST-I in¬
cluded patients randomized to encainide, flecainide, moricizine, or a matching placebo. CAST-II included
88.2
only patients randomized to moricizine or its matching placebo, specifically, 360 patients who continued to
28.2
receive moricizine or placebo from CAST-I, 216 patients who had been randomized to encainide or flecai¬
18.0
18.0 nide in CAST-I and were rerandomized to moricizine after the encainide and flecainide arms of CAST-I were
terminated, and 798 new patients who entered CAST-II after CAST-I enrollment had ceased. The 216 pa¬
75±13 tients who were randomized to moricizine in CAST-II after having been withdrawn from randomized therapy
125±19 in CAST-I did not have their CAST-II experience included in the comparison. The categories "Randomized
76±10 to Flecainide or Encainide" and "Randomized to Moricizine" refer to the blinded trial arms and include pa¬
tients receiving either active drug or matching placebo. Patients "Not Randomized" failed to have arrhyth¬
29.7 mia suppression, had an adverse effect, or withdrew from the trial electively. Those in the "Other" category
45.0 were in titration at the time the trials were stopped and hence could not be randomized.
4.9
26.7
47.0
38.5
patients randomized during CAST-I or shown in Fig 2. The mean time of follow-
in CAST-II to moricizine or its match¬ up was 399 days. At 1 year from the
ing placebo. time of randomization, 95% of placebo-
treated patients vs 90% of active drug-
RESULTS treated patients remained alive
As of July 31, 1991, when CAST-II (P=.0006). When examined for the pri¬
enrollment ceased, 3549 patients (2371 mary end point of CAST at 1 year (Fig
patients in CAST-I and 1178 patients in 3), 96% of placebo-treated patients re¬
CAST-II) had entered titration. The mained free of cardiac arrest or arrhyth¬
baseline characteristics of these patients mic death vs 93% of active drug-treated
are reviewed in the Table, and their patients (P=.003).
treatment assignments in CAST-I and
CAST-II are shown in Fig 1. Suppres¬ COMMENT
sion of arrhythmias was achieved in 2491 The purpose of CAST was to test the
patients (1498 patients randomized to hypothesis that in survivors of myocar-
encainide/flecainide or matching placebo dial infarction with asymptomatic or
in CAST-I, 309 patients randomized to mildly symptomatic ventricular arrhyth¬
moricizine or matching placebo in CAST- mias and left ventricular dysfunction,
I, and 684 patients randomized to mori¬ the suppression of ventricular prema¬
cizine or matching placebo in CAST-II), ture depolarizations improves surviv¬
and partial suppression was achieved in al.46·10 Since documentation of arrhyth¬
260 patients (95 patients in the encainide/ mia suppression rather than simply the
flecainide CAST-I arm, 51 patients in use of a drug was required to qualify the
the moricizine CAST-I arm, and 114 pa¬ patient for randomization in the study,
tients in CAST-II). Six hundred thirty- it was originally planned to report the
five patients (341 patients in CAST-I outcomes of patients with suppressible
and 294 patients in CAST-II) were not ventricular arrhythmias who were re¬
randomized to long-term, blinded use of ceiving the antiarrhythmic drugs that
a study drug, usually because of failure suppressed the arrhythmias as a com¬
to achieve even partial suppression, or bined group vs the outcome of patients
adverse effects, but occasionally because receiving matching placebos. However,
of their own or their physician's re¬ CAST-I demonstrated that the use of
quest.13 A minority of patients (14 in encainide or flecainide to treat these pa¬
CAST-I and 45 in CAST-II) were in tients increased mortality.4·5 Extending
titration when the study ended and these observations to treatment with
therefore were not randomized. moricizine, CAST-II demonstrated not
Overall survival from the time of ran¬ only that moricizine failed to improve
domization to blinded antiarrhythmic or long-term survival but also that treat¬
placebo therapy in CAST-I and CAST- ment with low-dose moricizine during a
II patients who completed titration is 2-week titration period during drug ini-

Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/jama/9825/ by a University of Alberta User on 02/17/2017

 mie death were decreased by active drug
100- - therapy. These analyses underscore the
negative impact of class I antiarrhyth-
mic drug treatment directed at the
90
suppression of ventricular premature
depolarizations on overall survival and
survival free of cardiac arrest or arrhyth¬
mic death in patients with left ventricu¬
To 80
= .0006
lar dysfunction after myocardial infarc¬
tion. In contrast to the usual pattern of
proarrhythmia that occurs early during
70- Placebo ( =1371) antiarrhythmic drug treatment,14 the
hazard in CAST remained constant
Active ( =1380)
throughout the trial. Finally, antiar¬
60- rhythmic drug treatment intended to
suppress ventricular arrhythmias in
these patients worsened survival regard¬
less of whether suppression was
50 achieved.6·15
10 12 14 16 18 20 22 The 95% survival of patients random¬
Months ized to placebo in CAST is higher than
expected and may raise questions as to
the generalizability of the trial to all
Fig 2.—Overall survival in CAST from initiation of blinded therapy. Actuarial curves that display overall sur¬
vival of all patients who had arrhythmia suppression and were randomized in CAST-I and CAST-II are shown. patients surviving myocardial infarction.
One-year actuarial survival was 95% in the placebo group with 645 patients remaining, and 90% in the ac¬ However, CAST is unique among anti-
tive treatment group with 641 patients remaining.
arrhythmic drug trials since only pa¬
tients with suppressible ventricular ar¬
rhythmias were randomized to receive
100-1 active drug or placebo and included in
the outcome analysis. It should follow
that if the aim of drug therapy for
ventricular ectopy after myocardial in¬
90-
farction is to suppress the "trigger" for
ventricular fibrillation, patients without
P=.003 demonstrable arrhythmia suppression
75 80 should not be expected to benefit from
>
I treatment and, therefore, should not be
treated with an antiarrhythmic drug.
S5 70 Placebo (n=1371) Thus, patients whose ventricular ar¬
Active (n=1380)
rhythmias increased (ie, they had neither
suppression nor partial suppression) dur¬
60 ing drug titration were not randomized
to receive blinded, placebo-controlled
drug therapy. The finding of nonsup-
50 — — —I-
pressibility was in fact a marker for
increased mortality and probably in large
10 12 14 16 18 20 22
Months
part explains the apparently low mor¬
tality in CAST.16"1* Indeed, when the
outcome of the nonrandomized patients
is examined, the population from which
Fig 3.—Survival to cardiac arrest or arrhythmic death in CAST from initiation of blinded therapy. Actuarial
curves that display survival free of cardiac arrest or arrhythmic death are shown for all patients who had ar¬ the randomized patients were derived
rhythmia suppression and were randomized in CAST-I and CAST-II. One-year actuarial survival was 96% is seen to be a typical post-myocardial
in the placebo group with 645 patients remaining, and 93% in the active treatment group with 641 patients infarction group.15
remaining. It is currently unknown whether other
antiarrhythmic drugs decrease mortal¬
tiation increased the risk of death.6 The The present study emphasizes the con¬ ity following myocardial infarction.1922
CAST-I report dealt only with encai¬ clusions of CAST-I and CAST-II and Amiodarone, for example, has the at¬
nide and flecainide, and the CAST-II their broad clinical implications by us¬ tractive attributes of incorporating class
report dealt only with moricizine. Al¬ ing the combined data and by focusing I, II, and IV antiarrhythmic drug action
though the latter study suggested that on arrhythmia suppression by class I with its predominant class III effect to
the treatment of asymptomatic or mildly antiarrhythmic drug action rather than prolong action potential duration. In a
symptomatic premature ventricular on the specific drugs that were used. recent meta-analysis, amiodarone de¬
depolarizations in patients with myo- Figures 2 and 3 show that for patients creased mortality by 44% in patients
cardial infarction and left ventricular who had suppression of ventricular pre¬ with a variety of diagnoses, including
dysfunction could not be supported, mature depolarizations and were ran¬ congestive heart failure, "serious" ven¬
the combined data from CAST-I and domized to active drug or placebo tricular arrhythmias, and myocardial in¬
CAST-II have not been previously re¬ therapy, both overall survival and sur¬ Antiarrhythmic
farction.19 In the Basel
ported. vival free of cardiac arrest or arrhyth- Study of Infarct Survival, patients sur-

Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/jama/9825/ by a University of Alberta User on 02/17/2017

 viving myocardial infarction were ran¬
domized to receive empiric amiodarone,
"individualized antiarrhythmic treat¬
ment" (including ajmaline, disopyramide,
flecainide, propafenone, or sotalol), or
placebo. Survival in patients receiving
amiodarone was greater than survival
in the other two groups.20 However, the
study was limited by crossover of pa¬
tients from the "individualized" arm to
amiodarone, and the study was un-
blinded. Ceremuzynski et al22 reported
the effect of amiodarone on mortality
after myocardial infarction, but the study
had the important limitation of exclud¬
ing patients who could not receive
ß-blockers (defined as those with heart
failure, asthma, treated diabetes, or pe¬
ripheral artery disease with claudica¬
tion). Hence, the enrolled patients may
not be representative of more general
post-myocardial infarction populations.
The Canadian Amiodarone Myocardial
Infarction Arrhythmia Trial may pro¬
vide further insight into the usefulness
References
1. Heart and Stroke Facts. Dallas, Tex: American
Heart Association; 1992:11-17.
2. Kannel WB, Thomas HE. Sudden coronary death:
the Framingham Study. Ann N Y Acad Sci. 1982;
382:3-21.
3. Bay\l=e'\sde Luna A, Coumel P, Leclercq JF. Am-
bulatory sudden cardiac death: mechanisms of pro-
duction offatal arrhythmia on the basis of data from
157 cases. Am Heart J. 1989;117:151-159.
4. The Cardiac Arrhythmia Suppression Trial
(CAST) Investigators. Preliminary report: effect of
encainide and flecainide on mortality in a random-
ized trial of arrhythmia suppression after myocar-
dial infarction. N Engl J Med. 1989;321:406-412.
5. Echt DS, Liebson PR, Mitchell LB, et al. Mor-
tality and morbidity in patients receiving encai-
nide, flecainide, or placebo: the Cardiac Arrhyth-
mia Suppression Trial. N Engl J Med. 1991;324:
781-788.
6. The Cardiac Arrhythmia Suppression Trial II
Investigators. Effect of the antiarrhythmic agent
moricizine on survival after myocardial infarction.
N Engl J Med. 1992;327:227-233.
7. The Cardiac Arrhythmia Pilot Study (CAPS)
Investigators. Effects of encainide, flecainide, imip-
ramine and moricizine on ventricular arrhythmias
during the year after acute myocardial infarction:
the CAPS. Am J Cardiol. 1988;61:501-509.
8. The Cardiac Arrhythmia Pilot Study (CAPS)
Investigators. Recruitment and baseline descrip-
tion of patients in the Cardiac Arrhythmia Pilot
Study. Am J Cardiol. 1988;61:704-713.
9. Vaughan Williams EM. A classification of anti-
arrhythmic drug actions reassessed after a decade
Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/jama/9825/ by a University of Alberta User on 02/17/2017
of amiodarone following myocardial in¬
farction.21
The importance of ß-blocker admin¬
istration following myocardial infarction
cannot be overstated. The risks of both
aiThythmic and nonarrhythmic death can
be decreased with these drugs after myo¬
cardial infarction.23 Furthermore, there
are suggestions that they also have salu¬
tary effects in patients with sustained
ventricular tachyarrhythmias.18 Al¬
though the low incidence of ß-blocker
use in CAST was disappointing, it should
not affect the conclusions, since, as a
consequence of randomization, the fre¬
quency of ß-blocker use was similar in
all subgroups in CAST. Finally, the in¬
frequent use of angiotensin converting
enzyme inhibitors in CAST may be ex¬
plained in part by recognizing that in
the years during which recruitment oc¬
curred, information on the usefulness of
these drugs was not yet available.24
This analysis of combined data from
CAST-I and CAST-II demonstrates that
of new drugs. J Clin Pharmacol. 1984;24:129-147.
10. Greene HL, Roden DM, Katz RJ, et al. The
Cardiac Arrhythmia Suppression Trial: first CAST
I . then CAST-II. J Am Coll Cardiol. 1992;19:
894-898.
. .
11. Clyne CA, Estes MAM, Wang PJ. Moricizine.
N Engl J Med. 1992;327:255-260.
12. Kalbfleisch JP, Prentice RL. The Statistical
Analysis of Heart Failure Time Data. New York,
NY: John Wiley & Sons Inc; 1980.
13. Wyse DG, Hallstrom A, McBride R, et al. Events
in the Cardiac Arrhythmia Suppression Trial
(CAST): mortality in patients surviving open label
titration but not randomized to double-blind therapy.
J Am Coll Cardiol. 1991;18:20-28.
14. Roden DM, Woosley RL, Primm RK. Incidence
and clinical features ofthe quinidine-associated long
QT syndrome: implications for patient care. Am
Heart J. 1986;111:1088-1093.
15. Epstein AE, Bigger JT, Wyse DG, et al. Events
in the Cardiac Arrhythmia Suppression Trial
(CAST): mortality in the entire population enrolled.
J Am Coll Cardiol. 1991;18:14-19.
16. Goldstein S, Pawitan Y, Kennedy H, et al. Ven-
tricular arrhythmia suppression predicts survival
in the Cardiac Arrhythmia Suppression Trial
(CAST). Circulation. 1990;82(suppl III):III-198. Ab-
stract.
17. Hallstrom AP, Greene HL, Huther ML, and
the CAST Investigators. The healthy responder
phenomenon in non-randomized clinical trials. Stat
Med. 1991;10:1621-1631.
18. Steinbeck G, Andresen D, Bach P, et al. A com-
parison of electrophysiologically guided antiarrhyth-
the suppression of asymptomatic or
mildly symptomatic ventricular arrhyth¬
mias after myocardial infarction in itself
does not improve survival. In fact, there
was a significantly higher survival rate
for placebo-treated compared with drug-
treated patients. These results and the
remarkable consistency between the
CAST-I and CAST-II data support the
recommendation that patients with
asymptomatic or mildly symptomatic
ventricular arrhythmias after myocar¬
dial infarction should no longer be
treated with antiarrhythmic drugs, es¬
pecially those with class I (sodium chan¬
nel blocking) action, with the intent to
decrease mortality unless benefit is dem¬
onstrated in a prospective, randomized,
controlled clinical trial. Although pre¬
liminary data regarding amiodarone are
encouraging in this regard, final data
are awaited.
This study was supported by contracts with the
National Heart, Lung, and Blood Institute, Depart¬
ment of Health and Human Services, Bethesda, Md.
mic drug therapy with beta-blocker therapy in pa-
tients with symptomatic, sustained ventricular ta-
chyarrhythmias. N Engl J Med. 1992;327:987-992.
19. Teo K, Yusuf S, Furberg C. Overview of ran-
domized clinical trials of low-dose amiodarone on
mortality in cardiac patients. Circulation. 1992;86
(suppl 1):1-534. Abstract.
20. Burkart F, Pfisterer M, Kiowski W, Follath F,
Burckhardt D, Jordi H. Effect of antiarrhythmic
therapy on mortality in survivors of myocardial
infarction with asymptomatic complex ventricular
arrhythmias: Basel Antiarrhythmic Study of In-
farct Survival (BASIS). J Am Coll Cardiol. 1990;
16:1711-1718.
21. Carins JA, Connolly SJ, Gent M, Roberts R.
Post-myocardial infarction mortality in patients with
ventricular premature depolarizations: Canadian
Amiodarone Myocardial Infarction Arrhythmia Trial
Pilot Study. Circulation. 1991;84:550-557.
22. Ceremuzynski L, Kleczar E, Krzeminska-Pa-
kula M, et al. Effect of amiodarone on mortality
after myocardial infarction: a double-blind, placebo\x=req-\
controlled, pilot study. J Am Coll Cardiol. 1992;
20:1056-1062.
23. Yusef S, Peto R, Lewis J, Collins R, Sleight P.
Beta blockade during and after myocardial infarc-
tion: an overview of the randomized trials. Prog
Cardiovasc Dis. 1985;77:335-371.
24. Pfeffer MA, Braunwald E, Moy\l=e'\LA, et al. Ef-
fect of captopril on mortality and morbidity in pa-
tients with left ventricular dysfunction after myo-
cardial infarction: results of the Survival and Ven-
tricular Enlargement Trial. N Engl J Med. 1992;
327:669-677.