New England Journal of Medicine March 21, 1991

Mortality and morbidity in patients receiving encainide, flecainide, or placebo. The Cardiac Arrhythmia Suppression Trial

D S Echt, P R Liebson, L B Mitchell, R W Peters, D Obias-Manno, A H Barker, D Arensberg, A Baker, L Friedman, H L Greene, et al.

Bottom Line

Suppressing asymptomatic ventricular arrhythmias with the Class IC antiarrhythmics encainide or flecainide after a myocardial infarction significantly increased arrhythmic and all-cause mortality compared to placebo, overturning prior assumptions about antiarrhythmic therapy.

Key Findings

1. Theprimaryendpointofdeathornonfatalcardiacarrestduetoarrhythmiaoccurredin5.7%(43of755)ofpatientsreceivingactivedrugversus2.2%(16of743)ofthosereceivingplacebo(RelativeRisk2.64;P=0.0004)[1.1].
2. Total all-cause mortality or nonfatal cardiac arrest was significantly higher in the active treatment group at 8.3% (63 of 755) compared to 3.5% (26 of 743) in the placebo group (Relative Risk 2.38; P = 0.0001).
3. Among deaths from nonarrhythmic cardiac causes, there was a significant excess in the active drug group compared to placebo (17 vs. 5; P = 0.01), driven largely by acute myocardial infarction complicated by shock.
4. Despite effectively suppressing premature ventricular complexes (PVCs) on Holter monitoring, encainide and flecainide exerted a fatal proarrhythmic effect.
5. Nonlethal adverse events—such as recurrent myocardial infarction, new-onset congestive heart failure, syncope, or the need for pacemakers—were equally distributed between the active drug and placebo arms.

Study Design

Design
Randomized Controlled Trial
Double-Blind
Sample
1,498
Patients
Duration
10 months
Median
Setting
Multicenter
Population Patients 6 days to 2 years post-myocardial infarction with asymptomatic or mildly symptomatic frequent premature ventricular complexes (≥6 per hour) and a left ventricular ejection fraction ≤55%.
Intervention Class IC antiarrhythmic agents (encainide or flecainide) titrated to successfully suppress premature ventricular complexes.
Comparator Matching placebo.
Outcome Death or nonfatal cardiac arrest due to arrhythmia.

Study Limitations

Thetrialarmswerehaltedprematurely(atameanfollow-upof10months)duetounexpectedexcessmortality, precludingtheassessmentoflong-termtrends[1.1].
The results were initially limited to the Class IC agents encainide and flecainide, making it unclear to what extent the proarrhythmic risk generalized to other classes of antiarrhythmic drugs.
The exact pathophysiological mechanism of the excess mortality—specifically why ischemic myocardium reacted fatally to sodium-channel blockade—could not be definitively elucidated by the trial itself.

Clinical Significance

CAST was a massive paradigm shift in cardiology. It definitively proved that Class IC antiarrhythmics are proarrhythmic and contraindicated in patients with structural heart disease and a prior myocardial infarction. Crucially, it demonstrated the danger of relying on surrogate endpoints (like PVC suppression) rather than hard clinical outcomes (like mortality) to guide therapeutic decisions.

Historical Context

Prior to CAST, the medical community widely believed that asymptomatic premature ventricular complexes (PVCs) following a myocardial infarction were harbingers of sudden cardiac death, and that suppressing them with antiarrhythmic drugs would save lives. The unexpected finding that encainide and flecainide actually increased sudden death shattered this dogma and led to a profound shift toward evidence-based medicine emphasizing mortality over surrogate markers in cardiovascular trials.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

How does the mechanism of action of Class IC antiarrhythmics, such as flecainide, explain the paradoxical increase in arrhythmic mortality observed in the CAST trial among post-MI patients?

Key Response

Class IC agents are potent sodium channel blockers that significantly slow cardiac action potential conduction velocity. While they successfully suppress ectopic beats like premature ventricular contractions (PVCs), this slowed conduction can facilitate and stabilize deadly re-entrant circuits within the scarred, heterogeneous post-MI myocardium, leading to sustained ventricular tachycardia or ventricular fibrillation.

Resident
Resident

A 65-year-old patient who had a myocardial infarction 6 months ago presents with frequent palpitations, and a Holter monitor shows frequent asymptomatic PVCs. Based on the legacy of the CAST trial, why are flecainide and propafenone contraindicated in this scenario, and what is the preferred initial management?

Key Response

The CAST trial proved that using Class IC agents to suppress asymptomatic post-MI PVCs actually increases sudden cardiac death due to proarrhythmic effects in structurally abnormal hearts. Consequently, Class IC drugs are absolutely contraindicated in patients with structural heart disease (CAD, prior MI, heart failure). Initial management for asymptomatic or mildly symptomatic PVCs post-MI should prioritize beta-blockers and optimization of underlying ischemia rather than antiarrhythmic suppression.

Fellow
Fellow

The CAST trial highlighted the phenomenon of proarrhythmia in ischemic myocardium. Electrophysiologically, how does the phenomenon of 'use-dependence' with Class IC agents selectively exacerbate the risk of lethal re-entrant arrhythmias during episodes of ischemia or tachycardia in these post-MI patients?

Key Response

Class IC agents exhibit strong use-dependence, meaning their sodium channel blockade intensifies at faster heart rates because the drug binds more effectively during the open or inactivated states of the channel. In post-MI patients, ischemia partially depolarizes the resting membrane potential, slowing the recovery of sodium channels. Combined with the use-dependent kinetics of Class IC agents during sinus tachycardia or rapid ectopic beats, this creates profound, rate-dependent slowing of conduction precisely when conduction is already vulnerable, setting up the perfect milieu for a stable, fatal re-entrant circuit.

Attending
Attending

The CAST trial is widely considered the quintessential cautionary tale against relying on surrogate endpoints in clinical medicine. As an attending teaching evidence-based medicine, how do you use the CAST trial to frame modern discussions about novel cardiovascular therapies that demonstrate excellent biomarker improvements but lack hard outcome data?

Key Response

CAST teaches that 'fixing' a pathophysiologic marker (like PVCs) does not guarantee clinical benefit and can obscure fatal off-target harm. Attendings use CAST to emphasize that while surrogate endpoints (e.g., plaque regression, biomarker reduction, arrhythmia suppression) are valuable for hypothesis generation and early-phase trials, widespread clinical adoption of a therapy must wait for randomized controlled trials powered to detect differences in hard clinical outcomes like mortality, stroke, and myocardial infarction.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The CAST trial was terminated prematurely by its Data and Safety Monitoring Board (DSMB) due to a statistically significant excess of deaths in the active treatment arm. What are the methodological challenges of early trial termination for harm, and how did the investigators utilize asymmetric stopping boundaries to prevent alpha error inflation?

Key Response

Stopping a trial early for harm risks exaggerating the negative effect size due to random high-variance fluctuations (random high). To mitigate this, investigators use predefined, stringent statistical boundaries (like O'Brien-Fleming boundaries). In CAST, an asymmetric boundary was employed: proving benefit required an extremely high threshold of evidence to stop early, but the threshold for harm was set lower (though still rigorous, p < 0.025 for a one-sided boundary) to ethically protect patients while ensuring the observed mortality signal was a genuine biological effect rather than a statistical anomaly.

Journal Editor
Journal Editor

The CAST study design incorporated an open-label dose-titration run-in phase to ensure only patients whose arrhythmias were successfully suppressed by the study drugs proceeded to randomization. As a critical appraiser, how does this enrichment design uniquely impact the internal versus external validity, and what chilling editorial implication does it have for the trial's findings?

Key Response

An enrichment design strengthens internal validity by proving the tested population actually responds to the mechanistic effect of the drug (PVC suppression). However, a tough reviewer would point out that it drastically narrows external validity. The chilling editorial implication is that the massive mortality risk observed was not in a general unselected post-MI population, but specifically in patients who successfully responded to the drug's intended antiarrhythmic surrogate effect, proving the surrogate marker itself was fundamentally flawed.

Guideline Committee
Guideline Committee

Based on the definitive, paradigm-shifting findings of the CAST trial, how do current ACC/AHA/HRS guidelines classify the use of Class IC antiarrhythmic drugs in patients with a history of myocardial infarction, and what is the strength of this recommendation?

Key Response

Current ACC/AHA/HRS guidelines for the management of ventricular arrhythmias give a Class III (Harm) recommendation against the use of Class IC antiarrhythmic drugs (such as flecainide and propafenone) in patients with a history of myocardial infarction or significant structural heart disease. This is supported by Level of Evidence B-R (based on high-quality randomized trials like CAST), which permanently shifted the guideline approach from attempting to suppress post-MI PVCs to explicitly prohibiting the Class IC drugs previously used to do so.

Clinical Landscape

Noteworthy Related Trials

1992

CAST II

n = 1,325 · NEJM

Tested

Moricizine

Population

Post-MI patients with decreased ejection fraction and ventricular premature depolarizations

Comparator

Placebo

Endpoint

Death from any cause

Key result: The trial was stopped early as moricizine increased mortality during the initial dose-titration phase and offered no long-term survival benefit.
1996

SWORD Trial

n = 3,121 · Lancet

Tested

d-Sotalol

Population

Post-MI patients with left ventricular dysfunction

Comparator

Placebo

Endpoint

All-cause mortality

Key result: The trial was terminated prematurely because d-sotalol significantly increased the risk of death compared to placebo.
1997

AVID Trial

n = 1,016 · NEJM

Tested

Implantable Cardioverter-Defibrillator (ICD)

Population

Patients resuscitated from near-fatal ventricular arrhythmias

Comparator

Antiarrhythmic drugs (amiodarone or sotalol)

Endpoint

Overall survival

Key result: Treatment with an ICD resulted in significantly better overall survival compared to antiarrhythmic drug therapy.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis