Sudden Death Post Myocardial Infarction

Abstract & Commentary

By John P. DiMarco, MD, PhD, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville. Dr. DiMarco does research for Medtronic, is a consultant for Medtronic, Novartis, and St. Jude, and is a speaker for Boston Scientific.

Source: Ye S, et al. Circumstances and outcomes of sudden unexpected death in patients with high-risk myocardial infarction: Implications for prevention. Circulation 2011;123:2674-2680.

In this paper, the authors present data from the Valsartan in Acute Myocardial Infarction Trial (VALIANT) on the circumstances and outcomes of cardiac arrest after myocardial infarction (MI). VALIANT was a double-blind, randomized, controlled trial that assessed treatment with valsartan, captopril, or both in more than 14,000 patients after an acute MI that had been complicated by heart failure and/or left ventricular systolic dysfunction. All deaths in VALIANT were reviewed by a central-blinded adjudication committee and patients having episodes of cardiac arrest with successful resuscitation or sudden cardiac death were included in this trial. Patients who underwent ICD implantation during the acute hospitalization (n = 94) were excluded.

Among the 14,609 patients enrolled in VALIANT, 1067 experienced sudden death or resuscitated cardiac arrest (903 deaths and 164 successful resuscitations). In addition, there were 1486 nonsudden cardiovascular deaths and 385 noncardiovascular deaths. Risk factors for sudden death or cardiac arrest included the following: advanced age; higher systolic and diastolic baseline blood pressures; higher baseline heart rates; higher Killip class; lower left ventricular ejection fractions; higher rates of diabetes, hypertension, and prior MI; and a lower likelihood of treatment with percutaneous intervention, thrombolytics, beta adrenergic blockers, or amiodarone. Among the 1067 patients who had sudden death or cardiac arrest, 251 (24%) had an event within the first 40 days of their index MI, 140 patients had their event between 41 and 90 days after index MI, and 671 had an event more than 90 days after index MI. The initial ECG rhythm at the time of cardiac arrest was available for 283 of the 1062 patients and included ventricular tachycardia or ventricular fibrillation in 189 patients (67%), asystole in 59 patients (21%), and pulseless electric activity in 17 patients (6%). Symptoms were reported to precede the arrest in 46% of patients in whom the data were available. Most sudden deaths occurred at home. Among the 978 sudden death events where the arrest location was known, 645 (66%) occurred at home, 204 (21%) occurred in hospital, and 129 (13%) occurred outside the home but not in hospital. In-hospital sudden death was more common within the first 40 days after MI with most of the later events occurring in the home. Among those patients with sudden death events at home, activity was known for 269 of 645 patients. Of these, 139 (52%) were asleep at the time of the event and 130 (48%) were awake when the event occurred. Events during sleep were witnessed in only 19% of cases compared to 70% of events in awake patients. In the subgroup of patients who were resuscitated from a cardiac arrest, ICD implantation was associated with improved survival.

The authors conclude that sudden death after complicated MI is relatively common (7% at about 2 years) and it occurs most frequently at home with preceding symptoms or during sleep.


Sudden death in the early phases after an acute MI remains a major clinical problem. Revascularization when possible and various pharmacologic strategies have been shown to be effective and should be routinely employed. Early, within 40 days post infarct, ICD therapy has been tested in several trials (IRIS and DINAMIT) and has been shown not to improve survival. As a result, early ICD implantation after infarction is specifically not allowed under Medicare guidelines in the United States. At-home automatic external defibrillator use also has been tested in a randomized trial (HAT), and this too was shown not to be effective. The data presented here about the epidemiology of post infarction sudden death are interesting and should cause cardiologists to reexamine their practices with post MI patients. About half of the cardiac deaths are sudden. The majority of the sudden deaths occur more than 40 days after infarct and the mechanisms of death appear to be amenable to intervention with an ICD. If these patients had been routinely screened for persistent risk at the 40 day time point, it is likely that the sudden death mortality could have been reduced substantially.