Current issues of ACP Journal Club are published in Annals of Internal Medicine


Encainide and flecainide increased mortality in patients with asymptomatic or mildly symptomatic ventricular arrhythmias

ACP J Club. 1991 July-Aug;115:14. doi:10.7326/ACPJC-1991-115-1-014

Source Citation

Echt DS, Liebson PR, Mitchell LB, et al. Mortality and morbidity in patients receiving encainide, flecainide, or placebo. The Cardiac Arrhythmia Suppression Trial. N Engl J Med. 1991 March 21;324:781-8.



To determine the effectiveness of the antiarrhythmic drugs encainide and flecainide in reducing mortality from ventricular arrhythmias after myocardial infarction.


Randomized controlled trial.


27 sites in North America and Sweden.


Eligibility criteria were 6 days to 2 years after myocardial infarction, average of ≥ 6 ventricular premature depolarizations (VPD) per hour during ≥ 18 hours of ambulatory monitoring, no runs of ventricular tachycardia (VT) of ≥ 15 beats at a rate of ≥ 120 beats/min, ejection fraction ≥ 90 days.


Patients responsive to encainide (35 or 50 mg, 3 times daily) or flecainide, 100 or 150 mg, 2 times daily, were randomized to active drug or placebo. Responsiveness was defined as suppression of 80% of VPDs and ≥ 90% of VT. Flecainide was not given to patients with ejection fractions < 0.30.

Main outcome measures

Arrhythmia-related death or arrhythmia-related cardiac arrest with resuscitation. 9 secondary end points were also defined.

Main results

Mean follow-up for the 1498 patients was 10 months. 60 (8%) cardiac deaths and cardiac arrests occurred among patients on active drug, and 21 events (3%) among patients on placebo (P < 0.001) {This absolute risk difference of 5% means that 1 additional cardiac death or cardiac arrest occurred for every 20 patients receiving active drug (compared with placebo), 95% CI 13 to 34; the relative risk increase was 181%, CI 74% to 356%}*. There were more cardiac deaths and cardiac arrests both related and unrelated to arrhythmia among patients on active drug (43 and 17, respectively) than on placebo (16 and 5, respectively) (P < 0.001 for those related to arrhythmia) {absolute risk difference 4%; 1 additional patient had an arrhythmia - related cardiac death or cardiac arrest for every 28 patients treated with active drug (compared with placebo), CI 18 to 62; relative risk increase 164%, CI 52% to 362%}*, (P = 0.01 for those unrelated to arrhythmia) {ARD 1.6%; 1 additional patient had an arrhythmia - unrelated cardiac death or cardiac arrest for every 63 patients treated with active drug (compared wtih placebo), CI 34 to 252; relative risk increase 235%, CI 29% to 771%}*. 3 noncardiac deaths occurred in the active group and 5 in the placebo group (P > 0.2). The relative risk for death or cardiac arrest by all causes was 2.38 (95% CI 1.59 to 3.57) and death or cardiac arrest caused by arrhythmia was 2.64 (CI 1.60 to 4.36). The relative risks for arrhythmia-related death and cardiac arrest for patients with ejection fractions < 0.30 and ≥ 0.30 were 1.97 and 3.38, respectively. The active drug and placebo groups were similar in the incidence of nonlethal cardiac secondary end points, the incidence of adverse effects requiring discontinuation of study drug, and the concomitant use of cardioactive medications. Compliance (pill count) was > 90% for 70% of patients.


The use of encainide or flecainide in patients who have asymptomatic or mildly symptomatic ventricular arrhythmias after myocardial infarction resulted in more deaths and cardiac arrests compared with placebo.

Source of funding: National Heart, Lung, and Blood Institute.

Address for article reprint: CAST Coordinating Center, 1107 NE 45th, Room 505, Seattle, WA 98105, USA.

*Numbers calculated from data in article.


This trial provides the landmark finding that survivors of acute myocardial infarction with ventricular ectopy who are treated with class 1C agents (encainide and flecainide) are at increased risk for arrhythmia-related death and cardiac mortality. This finding invalidates the hypothesis that suppression of ectopy leads to improved survival because arrhythmia-related deaths increased despite effective suppression of ectopy by the drugs. Encainide and flecainide, which work primarily by profoundly slowing conduction velocity, are not only VPD "killers" but can also induce life-threatening arrhythmias. It is likely, although not proven, that the increase in arrhythmia-related deaths was caused by proarrhythmic drug effects.

An alternative explanation for the increased risk is ongoing myocardial ischemia. Although symptoms preceding the arrhythmia-related deaths and evidence for the presence or absence of residual ischemia in study patients are not described, the possibility of ongoing ischemia cannot be ruled out. Because most of the deaths unrelated to arrhythmias were attributed to acute myocardial ischemia and recurrent infarction, it is possible that similar episodes preceded arrhythmia-related events in those who died suddenly. The combination of ischemia-induced conduction delay and drug-induced effects on conduction velocity may have created a favorable milieu for lethal ventricular arrhythmias. This hypothesis needs further examination.

Although the precise mechanism responsible for arrhythmia-related deaths is unclear, it is reasonable to conclude, based on the findings of the Cardiac Arrhythmia Suppression Trial (CAST) and the subsequent report (1), that routine use of antiarrhythmic therapy is not advisable for survivors of acute myocardial infarction with asymptomatic or mildly symptomatic ventricular ectopy. Perhaps it can be said that just because messengers (VPDs) bear bad news (risk for sudden death), killing the messengers might not be the best solution.

Prakash C. Deedwania, MD
University of California at San FranciscoFresno, California, USA


1. Epstein AE, Hallstrom AP, Rodgers WJ, et al. Mortality followed ventricular arrhythmia suppression by encainide, flecainide, and moricizine after myocardial infarction. The original design concept of the Cardiac Arrhythmia Suppression Trial (CAST). Jama. 1993;270:2451-5.