Clinical Cardiology: New Frontiers

Sudden Cardiac Death

Douglas P. Zipes, MD; Hein J.J. Wellens, MD

S udden cardiac death describes the unexpected natural

death from a cardiac cause within a short time period,
generally #1 hour from the onset of symptoms, in a person
patients in the convalescent phase after a large myocardial
infarction experience sudden cardiac death in the year there-
after, overall they account for a small number of the total
without any prior condition that would appear fatal.1,2 Such a sudden cardiac deaths per year. The use of interventions that
rapid death is often attributed to a cardiac arrhythmia, but limit infarct size, such as thrombolytic agents, has reduced
with the advent of monitoring capabilities from implantable this number still further. These factors have an impact on the
cardioverter-defibrillators (ICDs), it is now well recognized effects of therapeutic interventions because, although it is
that classifications based on clinical circumstances can be relatively easy to identify patients in the small high-risk
misleading and often impossible, because 40% of sudden subgroups and then to possibly prevent or reverse a ventric-
deaths can be unwitnessed.3 Only an ECG or a ventricular ular tachyarrhythmia, the overall impact on the total number
electrogram recorded from an implanted device at the time of of sudden cardiac deaths will be small. It becomes obvious
death can provide definitive information about an arrhythmia. that, to significantly reduce the incidence of sudden cardiac
Prodromal symptoms are often nonspecific, and even those death, more specific markers are needed for the general
taken to indicate ischemia (chest pain), a tachyarrhythmia population to identify large numbers in subgroups that ac-
(palpitations), or congestive heart failure symptoms (dys- count for a bigger percentage of the more than 300 000 who
pnea) can only be considered suggestive. For these reasons, die suddenly. The present risk factors (see below) generally
total mortality, rather than classifications of cardiac and identify the risk of developing the structural heart disease
arrhythmic mortality, should be used as primary objectives underlying sudden cardiac death rather than the proximate
for many outcome studies. precipitator of the event. Because the risk of sudden cardiac
death does not necessarily equate with the risk of developing
Magnitude of the Problem structural heart disease, these risk factors have limited ability
Sudden cardiac death accounts for 300 000 to 400 000 deaths in identifying specific individuals at risk for sudden cardiac
Downloaded from http://ahajournals.org by on June 19, 2025

annually in the United States, depending on the definition death. Nevertheless, their control, with concomitant reduction
used.1,2 When restricted to death ,2 hours from the onset of
in death from coronary artery disease, is probably at least in
symptoms, 12% of all natural deaths were classified as
part responsible for the reduction in overall sudden cardiac
sudden in one study, and 88% of those were due to cardiac
death. Figure 1B shows idealized curves of survival from
disease.1 Sudden cardiac death is the most common and often
sudden cardiac death for a population free of major cardio-
the first manifestation of coronary heart disease and is
vascular events versus a population that has survived a major
responsible for '50% of the mortality from cardiovascular
cardiovascular event. After an initial high attrition rate for the
disease in the United States and other developed countries. In
high-risk group in the first 6 to 18 months, the curves then
less-developed countries, sudden cardiac death rates parallel
become parallel, illustrating the modulating effects of time on
the rates of ischemic heart disease as a whole and therefore
the incidence of sudden cardiac death. Ultimately, risk
are lower. Several population-based studies have documented
stratification will be important only if it can be coupled with
a 15% to 19% decline in the incidence of sudden cardiac
a therapeutic intervention that reduces the risk of dying.
deaths caused by coronary heart disease since the early 1980s.
However, the increasing incidence of congestive heart failure
may halt this decline in the future.4
Risk Factors of Sudden Cardiac Death
Figure 1 places the problem into perspective by expressing Influence of Age, Race, and Sex
the incidence of sudden cardiac death in different subgroups Because up to 80% of individuals who suffer sudden cardiac
at varying risk while indicating the overall number of events death have coronary heart disease, the epidemiology of
per year for each. Thus, if one considers an overall incidence sudden cardiac death to a great extent parallels that of
in the adult population of only 0.1% to 0.2% per year, when coronary heart disease. As such, the incidence of sudden
applied to the entire US population, that accounts for more cardiac death increases with age in both men and women,
than 300 000 events per year. In contrast, although '33% of whites and nonwhites, because the prevalence of ischemic

From the Krannert Institute of Cardiology, Indiana University School of Medicine and the Roudebush Veterans Administration Medical Center,
Indianapolis (D.P.Z.), and the Department of Cardiology, Academic Hospital Maastricht and the Interuniversity Cardiology Institute of the Netherlands,
Utrecht (H.J.J.W.).
Correspondence to Douglas P. Zipes, MD, Krannert Institute of Cardiology, 1111 W 10th St, Indianapolis, IN 46202-4800.
(Circulation. 1998;98:2334-2351.)
© 1998 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org

2334
 Zipes and Wellens November 24, 1998 2335

Figure 2. Mortality rates in deaths per 100 000 for patients with
ischemic heart disease occurring out of hospital or in emer-
gency room (estimate for sudden cardiac death rate) (A) and
occurring in hospital, according to age, sex, and race in 48
states during 1985 (B). Reproduced with permission from Refer-
ence 7.

patients known to have heart disease. Experimentally, it

Figure 1. A, Sudden death incidence and total events for vari- appears that regular exercise in dogs prevents ischemia-
ous population pools, showing inverse relationship between risk induced ventricular fibrillation and death by increasing vagal
Downloaded from http://ahajournals.org by on June 19, 2025

and total number of events. B, Attrition over a 6- to 18-month

interval in a high-risk subgroup compared with a low-risk sub-
activity.8 Thus, it may be that regular exercise decreases
group. After initial falloff, curves become parallel. Reproduced cardiovascular morbidity and mortality, whereas vigorous
with permission from Reference 2, p 744. exercise, particularly in untrained individuals, may have an
adverse effect. The annual incidence of sudden cardiac death
heart disease increases with age (Figure 2). However, among during exercise is 1 per 200 000 to 250 000 healthy young
patients with coronary heart disease, the proportion of coro- people,1 whereas in competitive athletes, sudden cardiac
nary deaths that are sudden decreases with age. Sudden death is very rare, despite the publicity, with only 20 to 25
cardiac death has a much higher incidence in men than sports-related sudden cardiac deaths from cardiac causes
women, reflecting sex differences in the incidence of coro- annually in the United States.9 In young athletes (Figure 3),
nary heart disease as well. Thus, '75% of sudden cardiac sudden cardiac death most often occurs from hypertrophic
deaths occur in men, with an annual incidence 3 to 4 times cardiomyopathy, and in older athletes, from coronary heart
higher than in women. The peaks in incidence of sudden disease.10 Interestingly, in Europe, particularly in northern
cardiac death occur between birth and 6 months of age Italy, arrhythmogenic right ventricular dysplasia, possibly
because of the sudden infant death syndrome, and then again congenital, is the predominant anatomic finding in athletes
between 45 and 75 years of age as a result of coronary artery with sudden cardiac death.11 Commotio cordis, that is, con-
disease. Sudden cardiac death accounts for 19% of sudden cussion of the heart from nonpenetrating blunt trauma to the
deaths in children between 1 and 13 years of age and 30% anterior chest, can lead to fatal cardiac arrest, due to either
between 14 and 21 years of age.5 myocardial trauma or the mechanoelectrical triggering of a
ventricular tachyarrhythmia during the vulnerable period of
Activity the T wave.12 As with some other risk factors, the overall
The impact of physical activity on sudden cardiac death is impact of activity on sudden cardiac death may be small. In
somewhat controversial. Although vigorous exercise can the Maastricht Sudden Death study, 67% of the sudden death
trigger sudden cardiac death and acute myocardial infarction,6 victims were physically inactive at the time of the event.3
in part possibly by increasing platelet adhesiveness and
aggregability, moderate physical activity may be beneficial Anatomy
by decreasing platelet adhesiveness and aggregability.7 In Anatomic findings at autopsy include acute changes in
cardiac rehabilitation programs, cardiac arrests occur at a rate coronary plaque morphology, such as thrombus, plaque
of 1 in 12 000 to 15 000, and during stress testing, cardiac disruption, or both, in .50% of cases of sudden coronary
arrest occurs at a rate of 1 per 2000, at least 6 times greater death, whereas in hearts with myocardial scars and no acute
than the general incidence of sudden cardiac death for infarction, active coronary lesions are identified in 46% of
 2336 Sudden Cardiac Death

Figure 3. Causes of sudden cardiac death in competitive athletes. CM incidates cardiomyopathy; HD, heart disease; MVP, mitral valve
prolapse; and LVH, left ventricular hypertrophy. Reproduced with permission from Maron BJ, Epstein SE, Roberts WC. Causes of sud-
den death in competitive athletes. J Am Coll Cardiol. 1989;7:204 –214.

cases. Erosion of proteoglycan-rich and smooth muscle cell– failure, nonsustained ventricular tachycardia may be an inde-
rich plaques lacking a superficial lipid core, or plaque pendent marker of increased mortality due to sudden cardiac
rupture, is a frequent pathological finding.13 Plaque rupture death.18 According to one study,19 sudden coronary deaths are
appears to be more common in older women.14 Apoptosis less likely to occur at home than nonsudden coronary deaths,
may participate in the genesis and pathophysiology of some whereas individuals who die of sudden coronary death are
cardiac arrhythmias or conduction disturbances responsible more likely to have been current cigarette smokers. However,
for sudden cardiac death.15 However, these anatomic abnor- in the Maastricht study,3 80% of sudden cardiac deaths
malities are not represented by specific clinical risk factors occurred at home. Emotional stress can be an important
Downloaded from http://ahajournals.org by on June 19, 2025

different from those that identify patients with coronary heart trigger for sudden cardiac death, as shown by the Northridge
disease in general. In addition, because mechanisms respon- earthquake that struck the Los Angeles area at 4:31 AM
sible for sudden cardiac death depend in part on anatomic January 17, 1994.20 Depression in a patient in the hospital
substrate, which naturally varies from one individual to
after myocardial infarction is a significant predictor of the
another, the usefulness of risk assessment modalities varies
18-month post–myocardial infarction cardiac mortality, and
from one patient and particular type of anatomic substrate to
another. Furthermore, in addition to interpatient variations,
there may be intrapatient variation due to temporal changes in
specific diseases (Figure 1).

Other Risk Factors

Age, hypertension, left ventricular hypertrophy, intraventric-
ular conduction block, elevated serum cholesterol, glucose
intolerance, decreased vital capacity, smoking, relative
weight, and heart rate identify individuals at risk for sudden
cardiac death (Figure 4). Smoking is an important risk factor.
In the Framingham study, the annual incidence of sudden
cardiac deaths increased from 13 per 1000 in nonsmokers to
almost 2.5 times that for people who smoked .20 cigarettes
per day. Stopping smoking promptly reduced this risk, which
may be mediated by an increase in platelet adhesiveness,
release of catecholamines, and other mechanisms. Elevated
serum cholesterol appears to predispose patients to rupture of
vulnerable plaques, whereas cigarette smoking predisposes
patients to acute thrombosis.16 Female survivors of cardiac
arrest are less likely to have underlying coronary artery Figure 4. Risk of sudden cardiac death by decile of multivariate
disease, even though coronary artery disease status is the risk: 26-year follow-up, Framingham Study. L.V.H. indicates left
most important predictor of cardiac arrest in women; im- ventricular hypertrophy; I-V, intraventricular; and Non-Spec. Abn,
nonspecific abnormality. Reprinted with permission from Kannel
paired left ventricular function appears to be the most WB, Schatzkin A. Sudden death: lessons from subsets in popula-
important predictor in men.17 In patients with severe heart tion studies. J Am Coll Cardiol. 1985:5(suppl):141B–149B.
 Zipes and Wellens November 24, 1998 2337

the risk associated with depression was greatest among

patients with frequent premature ventricular complexes. So-
cioeconomic factors are also important; sudden cardiac death
after myocardial infarction increases 3-fold in men with low
levels of education and complex ventricular ectopy compared
with better educated men who have the same arrhythmias.
History can provide clues to the high-risk patient. For exam-
ple, in patients with ventricular tachycardia after myocardial
infarction, on the basis of clinical history, the following 4
variables identify patients at increased risk of sudden cardiac
death: (1) syncope at the time of the first documented episode
of arrhythmia, (2) NYHA class III or IV, (3) ventricular
tachycardia/fibrillation occurring early after myocardial in-
farction (3 days to 2 months), and (4) history of previous
myocardial infarctions.21 In some patients, family history can Figure 5. Relationship between left ventricular ejection fraction
(EF) and ventricular premature depolarizations (VPD) and sur-
be important.22 vival during 3 years of follow-up after myocardial infarction.
Left ventricular dysfunction is a major independent predic- Reproduced with permission from Bigger JT. Relation between
tor of total and sudden cardiac mortality in patients with left ventricular dysfunction and ventricular arrhythmias after
myocardial infarction. Am J Cardiol. 1986;57:8B.
ischemic and nonischemic cardiomyopathy.23 For example, in
survivors of cardiac arrest who have a left ventricular ejection
fraction ,30%, the risk of sudden cardiac death exceeds 30% influence of central and peripheral neurophysiological ac-
over 1 to 3 years if the patients do not have inducible tions; and the transient effects of toxins such as drugs28 or
ventricular tachycardia, whereas it ranges between 15% and alcohol.1,2 Structural cardiac abnormalities of the myocardi-
50% in those who have inducible ventricular um, coronary arteries, or cardiac nerves provide the substrate
tachyarrhythmias despite therapy with drugs that suppress the on which a transient risk factor operates. Although it is
inducible arrhythmias or with empirical amiodarone.24,25 possible that intense functional changes alone may create
Whether an ICD will reduce total mortality in patients with electrical instability of the normal heart to the degree that a
severe left ventricular dysfunction alone is the subject of ventricular tachyarrhythmia can be provoked, the vast major-
several prospective trials, including the Multicenter Auto- ity of cardiac arrests occur in patients with hearts that have
Downloaded from http://ahajournals.org by on June 19, 2025

matic Defibrillator Implantation Trial (MADIT II) and the structural abnormalities. One group that was identified before
Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT).26 a cardiac arrest with sufficient accuracy to warrant ICD
These patients have competing causes of death, and unless placement was the MADIT population,29 who were post
death is caused primarily by a ventricular tachyarrhythmia myocardial infarction and had spontaneous nonsustained
that can be terminated by prompt defibrillation, the ICD may ventricular tachycardia, inducible sustained ventricular
not have an important impact. Although prompt defibrillation tachycardia not suppressed by intravenous procainamide, and
generally restores sinus rhythm with a very high success rate, an injection fraction ,35%.
it may not be as successful in patients with very advanced As noted earlier, the most common structural abnormality
ventricular dysfunction. is coronary atherosclerosis and its consequences, such as
Certain ECG abnormalities can help identify patients at myocardial infarction. Interestingly, only '20% of patients
increased risk for sudden cardiac death. These include the who survive cardiac arrest develop features of a transmural
presence of AV block or intraventricular conduction defects myocardial infarction, and it is assumed that transient myo-
and QT prolongation, an increase in resting heart rate to .90 cardial ischemia, perhaps caused by coronary spasm or
bpm, and increased QT dispersion in survivors of out-of- unstable platelet thrombi,13,30 plays an important role in
hospital cardiac arrest. A recent study failed to support the precipitating a lethal ventricular tachyarrhythmia. Myocardial
usefulness of QT dispersion in predicting risk in patients after hypertrophy, congestive heart failure, and cardiac dilation,31,32
myocardial infarction.27 The presence of complex ventricular as well as regional autonomic dysfunction,33,34 all may be
arrhythmias, such as nonsustained ventricular tachycardia, is important. Although almost 50% of deaths in heart failure
also a marker1 (Figure 5). patients are sudden, among patients with cardiomyopathies,
those with better-preserved functional capacity (NYHA func-
Transient Risk Factors tional classes I and II) have lower total death rates, but the
Unfortunately, most of these more stable risk factors lack fraction of all deaths that are sudden and unexpected is
sufficient sensitivity, specificity, and predictive accuracy to higher; among class IV patients, total death rates are higher,
pinpoint the patient at risk with a degree of accuracy that but the fraction of sudden deaths is lower; thus, the impact of
would permit using a specific therapeutic intervention before reducing sudden cardiac death in this population will be
an actual event. This probably relates, at least in part, to the influenced by competing causes of other mechanisms of
transient nature of many risk factors, such as myocardial death.1,2 Time of day is also important, with more sudden
ischemia and reperfusion; hemodynamic dysfunction; abnor- cardiac deaths, strokes, and myocardial infarctions occurring
malities in electrolytes, such as hypokalemia and hypomag- in the morning on arising from bed, perhaps related to
nesemia, often due to diuretics; changes in pH or PO2; the increased sympathetic discharge in response to venous pool-
 2338 Sudden Cardiac Death

ing that then triggers increased blood viscosity and platelet both of these parameters judge autonomic modulation at the
aggregation.35 The lack of association between the times of sinus node, which is taken as a surrogate for autonomic
day in almost 600 patients who had at least 2 separate cardiac actions at the ventricular level. Autonomic effects at the sinus
arrests supports the hypothesis that a person’s activity may node and ventricle can easily be dissociated experimentally44
also play a role in triggering the cardiac arrest.36 and may possibly be a cause of false-positive or false-
Thus, if structural factors for the most part only create a negative test results.
substrate on which the transient factors operate to initiate a According to chaos theory, apparently irregularly irregular
ventricular tachyarrhythmia, risk identification requires find- events, such as ventricular ectopy, are nonrandomly distrib-
ing those subjects whose inherent physiological characteris- uted in time, and their clustering can be quantified by fractal
tics make the initiation of electrophysiological instability geometric analysis, which may help identify patients at risk
more likely when these conditions are met. This requires for sudden cardiac death.45 Late potentials, which are electri-
clinically identifiable, genetically based or acquired, individ- cal activity in the microvolt range extending the duration of a
ual differences in the responses of membrane channels, filtered QRS complex and detected by signal-averaged ECG,
receptors, exchangers, and pumps in the susceptible individ- has good negative predictive value but low positive predictive
ual,2 a formidable challenge at present. Patients with the value in patients after myocardial infarction.1 More recently,
congenital long-QT syndrome (see below) serve as the late potentials were not found to be useful in identifying
prototypic example of the interaction between a molecular patients who might benefit from ICD implantation and who
myocardial abnormality, an “ionopathy,” and an inciting were undergoing coronary artery bypass surgery.46 T-wave
event, eg, exercise in LQT1 and sleep/rest in LQT3.37 Some alternans, that is, T-wave changes in alternate beats, can at
patients may have a non– clinically manifest abnormality in times be visible in the scalar ECG and, when present, denote
repolarization, a latent form of long-QT syndrome, that patients with an electrically unstable ventricle. Recently,
becomes provoked only by exposure to certain drugs.37 Thus, T-wave alternans detectable only by computer averaging
molecular abnormalities in the long-QT syndrome, as well as techniques has been used to identify patients at risk for
in conditions such as hypertrophic cardiomyopathy,38 help subsequent ventricular arrhythmias.47 Finally, electrophysio-
provide genetic markers of patients at increased risk. logical studies to induce sustained ventricular arrhythmias
An antiarrhythmic drug can create the abnormality on can be useful to help select appropriate therapy, including
which a transient risk event, such as ischemia, interacts to drug therapy, catheter ablation, surgery, or ICD implantation,
provoke a lethal arrhythmia.39 For example, in the CAST and in identifying high-risk patients such as those suitable for
experience, despite the increased risk of sudden cardiac death treatment with an ICD.29
Downloaded from http://ahajournals.org by on June 19, 2025

established by the presence of complex forms of ventricular

ectopy, particularly in older age groups and in patients post Disease States
myocardial infarction, suppression of those ventricular ar-
rhythmias with encainide and flecainide conferred an in- Coronary Artery Disease
creased risk of death and/or no improvement in survival with As indicated earlier, at least 80% of patients who experience
moricizine.40 Death among those treated with an antiarrhyth- sudden cardiac death have coronary artery disease as the
mic drug may have resulted from an interaction between the underlying anatomic substrate due to atherosclerotic changes
substrate of coronary artery disease, the transient risk factor of the coronary arteries. Nonatherosclerotic coronary artery
of acute myocardial ischemia, and the exacerbation of ische- abnormalities are important in only a very small number of
mia-induced conduction slowing produced by drugs with sudden cardiac deaths and include problems such as coronary
negative dromotropic actions, such as encainide or flecain- arteritis, embolism, dissection, and congenital malformations
ide.41 The results of CAST taught us at least 3 important of anomalous origin of a left coronary artery from the
lessons: (1) that mechanisms responsible for premature ven- pulmonary artery or of a left coronary artery from the right or
tricular complexes, which were suppressed, were different noncoronary aortic sinus of Valsalva, passing between the
from mechanisms that caused sudden cardiac death, presum- aortic and pulmonary artery roots.1,2
ably from a ventricular tachyarrhythmia, which was in- In survivors of cardiac arrest, coronary heart disease with
creased; (2) that proarrhythmia from an antiarrhythmic agent vessels exhibiting more than 75% cross-sectional stenosis are
could occur months after drug initiation and was not always found in '40% to 86% of patients, depending on age and sex
an early event; and (3) that antiarrhythmic drugs could of the population studied. Although ,50% of the patients
become a risk factor when the myocardial substrate changed, resuscitated from ventricular fibrillation evolve evidence of
presumably when ischemia developed. myocardial infarction by elevated cardiac enzymes and
,25% have Q-wave myocardial infarction, autopsy studies
Electrophysiological End Points have reported that a recent occlusive coronary thrombus was
Two tests that reflect autonomic actions on the sinus node can found in 15% to 64% of victims of sudden cardiac death,
also be useful risk stratifiers. Baroreflex sensitivity, reflecting caused by ischemic heart disease, with many hearts showing
a vagal response to acute blood pressure elevation, is reduced plaque fissuring, hemorrhage, and thrombosis.48 There ap-
in patients at risk of sudden cardiac death,42 and heart rate pears to be no specific pattern of distribution of coronary
variability, a measure of beat-to-beat variations of sinus- artery lesions that favors the development of sudden cardiac
initiated RR intervals, with its Fourier-derived parameters, is death. Abrupt changes in regional myocardial blood flow due
also blunted in these patients.43 It is important to stress that to alterations in coronary artery structure and/or function,
 Zipes and Wellens November 24, 1998 2339

such as spasm, platelet thrombi, dissection, plaque rupture, or appears to be an important clinical variable that also identifies
other vasoactive events can provoke acute ischemia.13,30 patients with a higher risk of sudden cardiac death.50
Transition of stable atherosclerotic plaques by fissuring that Arrhythmogenic right ventricular dysplasia is a particular
leads to platelet activation and aggregation followed by kind of cardiomyopathy responsible for sudden death in
thrombosis formation, as well as other biochemical events young individuals and adults, with a gene defect recently
that can have a direct effect on electrophysiological proper- localized to chromosomes 1 and 14 q23-q24.11,52 It occurs as
ties of the heart, may be important in provoking ventricular a familial disorder in '30% of cases, with autosomal domi-
arrhythmias.13,30 Healed infarctions are present in $50% of nant inheritance. Exercise can precipitate ventricular
hearts of sudden cardiac death victims at autopsy and in those tachycardia in these patients, with an annual incidence of
of survivors of cardiac arrest. Interestingly, chronic ischemia sudden death estimated to be '2%. Two pathological pat-
may exert a protective effect by causing the development of terns, fatty and fibrofatty myocardial infiltration, have been
coronary collaterals that can help mitigate the extent of identified. In the fibrofatty variety, myocardial atrophy ap-
ischemia produced by sudden coronary occlusion. Therefore, pears to be the consequence of acquired injury and myocyte
an acute occlusion of a minimally stenosed coronary artery death and repair by fibrofatty replacement, mediated by
can result in a more disastrous outcome than occlusion of a patchy myocarditis. Apoptosis may be important. The left
severely stenosed coronary artery with the jeopardized myo- ventricle and ventricular septum can be involved in 50% to
cardium protected by collaterals. 67% of cases, often later in the disease, confirming a poor
prognosis.52 ECG during sinus rhythm often exhibits T-wave
Cardiomyopathy inversion in V1 to V3 or complete or incomplete right
Cardiomyopathies represent the second largest group of bundle-branch block, and the ventricular tachycardia has a
patients who experience sudden cardiac death. Hypertrophic left bundle-branch block contour, with the frontal-plane axis
cardiomyopathy has a prevalence of '2 in 1000 young adults reflecting the site of origin in 1 of 3 predilection sites for
and an incidence of sudden cardiac death of 2% to 4% per ventricular fatty degeneration: right ventricular inflow and
year in adults and 4% to 6% per year in children and outflow tracts and apex, the so-called “triangle of dysplasia.”
adolescents49 (Figure 3). In patients with hypertrophic cardio- During sinus rhythm, intraventricular conduction may be
myopathy, a history of sudden cardiac death or sustained sufficiently slow as to produce a terminal notch on the QRS
ventricular tachycardia, family history of sudden cardiac complex that Fontaine called an epsilon wave (Figure 6).
death, a diverse genotype, recurrent syncope, multiple epi-
Left Ventricular Hypertrophy
Downloaded from http://ahajournals.org by on June 19, 2025

sodes of nonsustained ventricular tachycardia, and massive

Left ventricular hypertrophy, whether established by ECG or
left ventricular hypertrophy are the strongest risk factors for
by cardiac echo, is a strong independent risk factor for
sudden cardiac death.38,49 Multiple mechanisms may be re-
cardiovascular deaths and, in particular, sudden cardiac death
sponsible, including arrhythmias, abrupt hemodynamic dete-
in patients who also had a history of hypertension. Multiple
rioration, and/or ischemia. Hemodynamic and echocardio-
disease states can result in hypertrophy, including valvular
graphic variables are generally not useful in identifying
heart disease, obstructive and nonobstructive hypertrophic
patients at high risk for sudden cardiac death, and the results
cardiomyopathy, primary pulmonary hypertension with right
of ambulatory ECG monitoring and invasive electrophysio-
ventricular hypertrophy, and various congenital heart disor-
logical study are controversial.49 The presence of mutations in
ders. Although ventricular repolarization (QT interval) is
the a-tropomyosin as well as in the b-myosin heavy chain
prolonged in hypertensive hearts, potentially setting the stage
gene has been associated with sudden cardiac death.38,49 for triggered activity (see below), myocardial ischemia,
Idiopathic dilated cardiomyopathy is a substrate for '10% interstitial fibrosis, and electrolyte disturbances can all con-
of sudden cardiac deaths in the adult population. Mortality in tribute to the genesis of ventricular tachyarrhythmias.1
patients with idiopathic dilated cardiomyopathy ranges from
10% to 50% annually, depending on the severity of the Valvular Disease
disease. In a compilation of 14 studies including 1432 The risk of sudden death in asymptomatic patients with aortic
patients, mean mortality rate after a follow-up of 4 years was valve disease appears to be low. After prosthetic or hetero-
42%, with 28% of deaths classified as sudden.50 The presence graft aortic valve replacements, patients remain at some risk
of nonsustained ventricular tachycardia in this group identi- for sudden cardiac death caused by arrhythmias, prosthetic
fies a population at high risk of sudden death, presumably on valve dysfunction, or coexistent coronary artery disease.
the basis of a ventricular tachyarrhythmia.18 Bundle-branch Hence, sudden cardiac death has been reported to be the
reentry can be an important cause of ventricular tachycardia second most common mode of death after valve replacement
in patients with dilated cardiomyopathy.51 The terminal event surgery, with an incidence of '2% to 4% over a follow-up of
can also be asystole or electromechanical dissociation, par- 7 years, accounting for '20% of the postoperative deaths.1
ticularly in patients with advanced left ventricular dysfunc- Whether mitral valve prolapse causes sudden cardiac death
tion.23 Multiple triggering events in heart failure patients is unresolved. Its prevalence is so high that its presence may
include myocardial stretch, neuroendocrine factors, electro- be just a coincidental finding in victims of sudden cardiac
lyte abnormalities, proarrhythmic effects of antiarrhythmic death and not causally related.1 However, patients with mitral
drugs, and excessive activation of the sympathetic and valve prolapse who have mitral regurgitation and left ventric-
renin-angiotensin systems.31 Syncope in heart failure patients ular dysfunction or myxomatous degeneration of the valve
 2340 Sudden Cardiac Death

Figure 6. Epsilon wave (arrowhead) in a patient with arrhythmogenic right ventricular dysplasia.

are clearly at higher risk for complications, such as infective binations of bundle-branch or fascicular blocks.56 It is impor-
endocarditis, cerebroembolic events, and sudden cardiac tant to remember that the absence of structural abnormalities
death. is established by relatively gross tests, such as cardiac
Downloaded from http://ahajournals.org by on June 19, 2025

catheterization and echocardiography. Other imaging tech-

Congenital Heart Disease niques, for example, those that evaluate sympathetic neural
An increased risk of sudden cardiac death due to arrhythmias function, are often abnormal in these patients.57 With the
has been found predominantly in 4 congenital conditions, development and validation of new diagnostic tools, includ-
including tetralogy of Fallot, transposition of the great arter- ing autonomic imaging by positron emission tomography,
ies, aortic stenosis, and pulmonary vascular obstruction. genetic testing,58 and magnetic resonance imaging, many
Sudden cardiac death has also been described as a late forms of “idiopathic” sudden cardiac death in patients with
complication after surgical repair of complex congenital apparently structurally normal hearts may have to be reclas-
cardiac lesions, such as tetralogy of Fallot and transposition sified, because these patients may become identified as
of the great arteries, and in patients with primary or secondary having a specific structural and/or genetic abnormality. A
pulmonary hypertension. In tetralogy of Fallot, QRS prolon- fascinating recent discovery is that the gene responsible for
gation relates to right ventricular size and predicts patients at the Brugada syndrome, the cardiac sodium channel gene
risk for sudden cardiac death.53 SCN5A on chromosome 3,58 is the same gene, with different
defects, that causes LQT3 syndrome37 (Table 1).
Primary Electrophysiological Abnormalities The idiopathic (congenital) long-QT syndrome is caused
Patients with primary electrophysiological abnormalities rep- by prolongation of repolarization due to abnormal movement
resent a group in whom mechanical function of the myocar- of sodium ions into or potassium ions out of the cardiac
dium is normal and an electrophysiological derangement myocyte, creating prolonged periods of intracellular positiv-
represents the primary cardiac problem. This includes pa- ity.37 Such prolongation of repolarization can lead to the
tients with the congenital long-QT syndrome, Wolff- development of early afterdepolarizations (Figure 7). A
Parkinson-White syndrome, several types of distinctive ven- specific kind of ventricular tachycardia called torsade de
tricular tachycardias, idiopathic ventricular fibrillation54 pointes occurs in patients with the long-QT syndrome,
(including a newly described entity characterized by right whether congenital or acquired (Figure 8). In addition to
bundle-branch block and ST-segment elevation, Brugada’s prolonged repolarization, ECG characteristics of this congen-
syndrome),55 congenital complete AV block, and a variety of ital disorder include abnormal T-wave contours, T-wave
acquired abnormalities, such as the acquired long-QT syn- alternans, a relative sinus bradycardia, and torsade de pointes
drome and acquired diseases of the sinus node, AV node, and that can produce syncope and sudden cardiac death. Genetic
His-Purkinje system, such as Lenegre’s disease or Lev’s heterogeneity (Table 1) may make therapies specifically
disease.2 Isolated cardiac conduction disturbances can be due targeted for the electrophysiological abnormality somewhat
to an autosomal dominant defect that includes various com- difficult. Syndromes LQT1 through LQT5 are inherited as
 Zipes and Wellens November 24, 1998 2341

TABLE 1. Genetics of the Long-QT Syndrome

LQT1 LQT2 LQT3 LQT4 LQT5
Chromosome locus 11p15.5 7q35-36 3p21-24 4q25-27 21q22.1
Gene mutation/protein KvLQT1 HERG SCN5A ? KCNE1*
Current/channel (subunit) IKs (a)† IKr INa ? IKs (b)†
Channel function Reduced function Reduced function Gain of function ? Reduced function
Action potential Delayed phase 3 Delayed phase 3 Prolonged phase 2 ? Delayed phase 3
Clinical syndrome
Heterozygous mutation R-W R-W R-W R-W R-W
Homozygous mutation J-L-N NR NR NR J-L-N
R-W indicates Romano-Ward syndrome; J-L-N, Jervell and Lange-Nielsen syndrome with deafness; and NR, not reported.
*KCNE1 is also referred to as minK.
†The KvLQT1 gene encodes the a-subunit protein of the IKs channel, with the KCNE1 gene encoding the b-subunit protein of this
same channel. The a- and b-subunit proteins combine together to form a cardiac potassium channel expressing the IKs current.
Reproduced with permission from Reference 37.

autosomal dominants, whereas the Jervell-Lange-Nielsen ventricles,59 do not seem to be important clinically, even
syndrome, dominant for the long-QT manifestation, is reces- though the same ionopathy affecting the ventricles should be
sive for the associated deafness and appears to be due to the present in the atria.
presence of both alleles responsible for LQT1. The incidence Patients with the Wolff-Parkinson-White syndrome have a
of cardiac events is higher in LQT1 and LQT2 than in LQT3, risk of sudden cardiac death ,1 per 1000 patient-years of
whereas the lethality of cardiac events is higher in LQT3 than follow-up. Almost all survivors of sudden cardiac death with
in LQT1 and LQT2 patients. 37 Interestingly, atrial Wolff-Parkinson-White syndrome have had symptomatic ar-
tachyarrhythmias, induced in an animal model by mecha- rhythmias before the event, but up to 10% experience sudden
nisms similar to those that cause torsade de pointes in the cardiac death as their first manifestation of the disease.60 The
responsible mechanism most probably is the development of
atrial fibrillation, with rapid conduction to the ventricles over
Downloaded from http://ahajournals.org by on June 19, 2025

the accessory pathway that produces ventricular rates so rapid

that the rhythm degenerates to ventricular fibrillation (Figure
9). The best predictor for development of ventricular fibril-
lation is a rapid ventricular response over the accessory
pathway during atrial fibrillation, with the shortest interval
between ventricular beats conducted over the accessory
pathway #250 ms. Although this response identifies virtually
100% of patients at risk for developing ventricular fibrilla-
tion, its specificity is low, because it may be found in 20% of
asymptomatic patients with Wolff-Parkinson-White conduc-
tion and 50% of those with mild to moderate symptoms due
to atrioventricular reentrant tachycardia.
Idiopathic ventricular tachycardias with monomorphic
contours that occur in patients with apparently structurally
normal hearts include paroxysmal and repetitive forms that
originate from the region of the right ventricular outflow
tract. This ventricular tachyarrhythmia characteristically has a
left bundle-branch block contour and inferior axis and pos-
sesses the unique quality of termination with vagal maneu-
vers such as adenosine infusion.61 Far less common is a
ventricular tachycardia from the left ventricular outflow tract.
A left septal ventricular tachycardia arises in the left posterior
septum and is sometimes called a fascicular tachycardia
because it is often preceded by a fascicular potential. It has a
right bundle-branch block, left-axis-deviation contour. Cal-
cium channel blockers characteristically suppress this ar-
rhythmia. Sudden cardiac death rarely occurs in this
Figure 7. Early afterdepolarizations induced after epinephrine population.
infusion shown in monophasic action potential recording from
left ventricle from patient with congenital long-QT syndrome (ge- Several types of idiopathic polymorphic ventricular
notyped not to be LQT1, LQT2, or LQT3). tachycardias have been described and are associated with a
Sudden Cardiac Death
2342
Downloaded from http://ahajournals.org by on June 19, 2025
 Zipes and Wellens November 24, 1998 2343

Figure 9. Atrial fibrillation with rapid ventricular response due to conduction over an accessory pathway degenerating into ventricular
fibrillation. Reproduced with permission from Zipes DP. Specific arrhythmias: diagnosis and treatment. In: Braunwald E, ed. Heart Dis-
ease: A Textbook of Cardiovascular Medicine. Philadelphia, Pa: WB Saunders; 1997:674.

less favorable outcome than the idiopathic monomorphic (sudden and unexpected death, Japan), and bangungut (to rise
ventricular tachycardias noted above. They may occur as and moan in sleep, Philippines). The cause(s) is unknown.
sporadic or familial forms, frequently precipitated by cate-
cholamine release during physical or emotional stress. Pa- Drugs
Downloaded from http://ahajournals.org by on June 19, 2025

tients with catecholaminergic polymorphic ventricular Antiarrhythmic drugs have long been known to be capable
tachycardia apparently have a favorable response to of provoking ventricular tachyarrhythmias and sudden
b-blockade therapy, whereas those with idiopathic ventricular cardiac death. Nonantiarrhythmic drugs that prolong repo-
fibrillation or short coupled torsade de pointes may not. larization (Table 2), along with class IA antiarrhythmic
Sudden cardiac death can occur in patients with poly-
agents, can cause torsade de pointes. Class IC drugs in the
morphous ventricular tachycardia who have normal QT
CAST study were mentioned earlier. Drug-drug interac-
intervals and normal systolic cardiac function; closely cou-
tions during poly-pharmacy can be dangerous, even with
pled premature complexes can initiate the spontaneous epi-
apparently innocuous medications.28 Phosphodiesterase in-
sodes. 62 Perhaps related is a similar ventricular
hibitors and other positive inotropic agents that increase
tachyarrhythmia characterized by a torsade de pointes con-
intracellular calcium loading have also been demonstrated
tour and normal QT interval and initiated by premature
to exert proarrhythmic actions and increase the risk of
ventricular complexes with extremely short coupling
intervals.63 sudden cardiac death.1 Hypokalemia (in some instances
Sudden cardiac death due to primary ventricular fibrilla- provoked by potassium-wasting diuretics), hypomag-
tion, ie, without apparent evidence of structural heart disease nesemia, and increased intracellular calcium concentration
(see above),54 occurs in '5% of victims of sudden cardiac may be important as primary or triggering events. It is
death. Preliminary data suggest that these patients have a 30% sometimes difficult to determine whether a patient resus-
recurrence rate of ventricular fibrillation, syncope, and car- citated from ventricular fibrillation had the arrhythmia
diac arrest, and it is important to stress that their survival is provoked by hypokalemia, because in the postresuscitation
largely related to potentially controllable or reversible elec- period, serum concentrations of potassium may be reduced
trophysiological disturbances rather than death due to ad- because of the effects of catecholamine release after the
vanced heart disease. ICDs should be particularly useful in cardiac arrest. Thus, unless there is a history of electrolyte
these patients.64 imbalance, drugs known to deplete potassium, special diets
Sudden unexplained nocturnal death can occur in young, such as the liquid protein diet, or documented electrolyte
apparently healthy, males of Southeast Asian origin and has abnormalities when the patient is in a steady state, the
several names, such as lai-tai (sleep death, Laos), pokkuri diagnosis of hypokalemia may be in doubt.

Figure 8. Torsade de pointes. Note typical “twisting about on its points” that is characteristic of this ventricular tachycardia associated
with long-QT syndrome, in this instance in a patient who developed a prolonged QT interval (a) and torsade de pointes (b) after just 50
mg hismanol. c, Reversion of QT interval to normal after drug cessation.
 2344 Sudden Cardiac Death

TABLE 2. Secondary Causes of Acquired QT Prolongation dispersion of refractoriness can occur after ventricular dila-
Antiarrhythmic agents
tion and heart failure. Any process that creates electrical
heterogeneity favors the development of ventricular fibrilla-
Class IA: quinidine, procainamide, N-acetylprocainamide, disopyramide
tion. Recent data indicate that sympathetically denervated
Class III: amiodarone, low risk of torsade de pointes
ventricular myocardium demonstrates abnormal oxygen uti-
Class IV: bepridil, mibefradil lization, which could also affect arrhythmogenesis (G.D.
Antihistamines Hutchins, PhD, unpublished observations, 1998).
Terfenadine Most data suggest that vagal stimulation, profibrillatory for
Astemizole the atria, mitigates the development of ventricular arrhyth-
Antimicrobials mias in a variety of experimental situations. Whether para-
Erythromycin
sympathetic stimulation is protective because of a direct
electrophysiological effect on ventricular myocardium, by
Trimethoprim-sulfamethoxazole
opposing sympathetic actions, or by albeit minimally pro-
Claritromycin
longing refractoriness8 is not known. Because it is difficult to
Cotzimoxazole study the effects of vagal activity on ventricular electrophys-
Azithromycin iological properties noninvasively in humans, the behavior of
Ketoconazole the sinus node has been used as a surrogate for ventricular
Pentamidine actions by measurement of indices of heart rate variability
Chloroquine (reflecting primarily tonic vagal action) and evaluation of
Gastrointestinal
baroreflex sensitivity (as a measure of reflex vagal activity).
Responses can be misleading because vagal actions at the
Cisapride
sinus node and ventricles can easily be dissociated,44 as
Liquid protein diets
mentioned earlier. Reductions in heart rate variability,43 as
Anorexia nervosa well as baroreflex sensitivity,42 identify patients at risk for a
Lipid lowering subsequent cardiac event. An increase in vagal tone in
Probucol animals66 and patients67 has been achieved by scopolamine
Psychotropic agents patches, without a clear benefit on preventing ventricular
Tricyclic and tetracycline antidepressants arrhythmias. In contrast, exercise conditioning in animals,
demonstrated to increase vagal “tone,” has been shown to
Downloaded from http://ahajournals.org by on June 19, 2025

Haloperidol
have a protective effect on preventing ventricular fibrillation
Phenothiazines
in dogs with coronary occlusion.8 Naturally, exercise condi-
Risperidone
tioning does many things that may be unrelated to neural
Selective serotonin reuptake inhibitors function. Vagal stimulation can terminate a specific type of
Other agents ventricular tachycardia originating in the right ventricular
Organophosphates outflow tract,61 but neural innervation at that site may be
Diuretics (reduced K1, Mg21) unique.8
Vasopressin (severe bradycardia)
Chloral hydrate amantadine
Mechanisms
In one sense, sudden cardiac death can be considered an
Electrolyte abnormalities
electrical accident because, although many individuals have
Hypokalemia anatomic and functional substrates conducive to developing a
Hypomagnesemia life-threatening ventricular tachyarrhythmia and many pa-
Hypocalcemia tients have transient events that could predispose to the
initiation of ventricular tachycardia or ventricular fibrillation,
only a relatively small number of patients actually do develop
Autonomic Nervous System sudden cardiac death. It is this interplay between the anatomic
Abnormalities of the autonomic nervous system appear to be
and functional substrates, modulated by the transient events
involved in the genesis of sudden cardiac death.8 Myocardial that perturb the balance, and the impact of all 3 on the
infarction, for example, produces regional cardiac sympa- underlying potential arrhythmia mechanisms possessed by all
thetic and parasympathetic dysfunction not only in the in- hearts that precipitates sudden cardiac death1,2,35 (Figure 11).
farcted area but also in regions apical to the infarct, presum- Understanding this is critical to understanding the pathophys-
ably because of interruption of afferent and efferent nerve iology of sudden cardiac death.
fibers traversing the infarct (Figure 10). Similar changes have The figure also indicates the complexity as well as the
been found in patients after myocardial infarction.8 Dener- potential variations in the inciting factors, because each
vated regions show supersensitivity to catecholamine infu- category in the Venn diagram can interact with the others in
sion, with disproportionate shortening of refractoriness that almost endless permutations and combinations. Most often,
creates autonomic heterogeneity, resulting in dispersion of interaction of a single item in each circle with a single item in
refractoriness and/or conduction, which can be conducive to the other circle (points at which only 2 circles overlap) may
development of ventricular arrhythmias.8 Similar increased normally be insufficient to produce sudden cardiac death,
 Zipes and Wellens November 24, 1998 2345

Figure 10. Autonomic denervation after myo-

cardial infarction. Extent of cardiac
hydroxyephedrine (11C-HED, sympathetic neu-
rotransmitter taken up by prejunctional sympa-
thetic nerve endings) defect exceeds reduction
in myocardial blood flow (imaged by [13N]am-
monia, 13N-NH3), consistent with regional sym-
pathetic denervation after myocardial infarction.
Image obtained with help of Gary Hutchins,
PhD, and Erica D. Engelstein, MD.

unless the single abnormality is extremely severe. For exam- scarred myocardium, and hypokalemia, might now be suffi-
ple, mild electrolyte abnormalities, such as a potassium cient to provoke a ventricular tachyarrhythmia, causing sud-
Downloaded from http://ahajournals.org by on June 19, 2025

concentration of 2.7 mEq/L, alone are usually insufficient to den cardiac death. Changes in the anatomic substrate can alter
cause a problem. Even in a patient with stable coronary artery the susceptibility of the myocardium to the effects of the
disease, that combination may not necessarily be lethal. transient initiating events. For example, experimental studies
However, if the patient had preexisting reentry pathways in indicate that hypertrophied myocardium, as well as myocar-
the ventricular myocardium, perhaps due to an old infarction, dium after a healed myocardial infarction, exhibits a greater
then the combination of the 3, ie, coronary artery disease, arrhythmogenic response than normal tissue to the same

Figure 11. Venn diagram showing interaction

of various anatomic/functional and transient
factors that modulate potential arrhythmogenic
mechanisms capable of causing sudden car-
diac death.
 2346 Sudden Cardiac Death

extent of acute ischemia.2 Catecholamine release can modu- fibrillation. However, even in the normal state, these condi-
late some of the effects of acute coronary occlusion and tions do not exist, because various cell types, eg, ventricular
reperfusion, and reduction in sympathetic action with drugs muscle versus Purkinje fibers, exhibit different action poten-
introduced to the pericardial sac to superfuse sympathetic tial characteristics, refractoriness, and conduction velocities.
nerves68 can prevent ventricular arrhythmias. Conversely, However, when heterogeneity becomes extreme, for instance,
acute ischemia alone, involving a sufficiently large area of if one region of the myocardium exhibits ischemia-induced
myocardium in an otherwise normal ventricle, can precipitate conduction delay and/or block that is different from neigh-
ventricular fibrillation without interplay with other factors, boring regions, or when there is regional sympathetic dys-
although it is interesting to consider the many balloon function8 or unequal stretch70 that can produce regional
angioplasties performed and the infrequent occurrence of electrophysiological alterations, the stage becomes set for
ventricular fibrillation during that procedure. Perhaps the development of ventricular fibrillation. Such alterations can
duration of the ischemia is too short to initiate ventricular be provoked by anatomic/functional substrates and by tran-
fibrillation. Although unquestionably the above logic repre- sient initiating events and can modulate basic arrhythmia
sents a very simplistic synthesis (Figure 11) and actual mechanisms of reentry, automaticity, and triggered activity to
mechanisms are more complex, nevertheless it offers a provoke ventricular arrhythmias (Figure 11). Reentry appears
conceptual framework to understand the interactive forces to be the major mechanism responsible for ventricular ar-
precipitating sudden cardiac death. rhythmias due to acute and chronic coronary disease and must
In the experimental animal, a very definite set of arrhyth- be dependent on heterogeneity. While we know a great deal
mogenic intervals has been described after acute coronary about the electrophysiological alterations that accompany
occlusion, including an arrhythmogenic interval within the acute and chronic ischemia in a variety of experimental
first few minutes after coronary occlusion that begins to abate preparations, the events surrounding the onset of ventricular
after 30 minutes and reappears after several hours. In addi- fibrillation in humans, even after 50 years of study, remain
tion, the initial 30 minutes of arrhythmias can be divided into fairly opaque. The “holy grail” of the electrophysiologist to
the first 10 minutes, presumably directly related to the initial match a particular antiarrhythmic drug that has a specific
ischemic injury, and the second 20 to 30 minutes, related to mechanism of action to an arrhythmia caused by a unique set
either reperfusion or the evolution of different injury patterns of electrophysiological alterations has, to date, still proved
in the epicardial and endocardial muscles and Purkinje elusive. In fact, the only drugs shown to reduce mortality
fibers.1,2 In the ischemic myocardium, a dramatic reduction in from sudden cardiac death are b-blockers and amiodarone
tissue pH to ,6.0, an increase in interstitial potassium levels (by meta-analysis).71 Neither drug has specific and single ion
Downloaded from http://ahajournals.org by on June 19, 2025

to values .15 mmol/L, increases in intracellular calcium channel actions. The reason ICDs are so successful is that a
concentration, and neurohumoral changes all contribute to “dose of electricity” is generic; ie, the mechanism causing the
creating electrophysiological changes characterized by ventricular tachyarrhythmia and the nature of the underlying
slowed conduction, reduced excitability and prolonged re- heart disease, both critical for antiarrhythmic drug effective-
fractoriness, cell-to-cell uncoupling, and the generation of ness, are largely irrelevant.
spontaneous electrical activity.69 Other metabolic changes,
such as accumulation of free fatty acids and their metabolites, Treatment
formation of lysophosphoglycerides, and impaired myocar- Because different electrophysiological mechanisms in the
dial glycolysis, may contribute to the development of elec- presence of different types of cardiac disease can cause
trical instability leading to cardiac arrhythmias.1,2 Although sudden cardiac death and because many of the victims do not
reentry is considered to be a dominant mechanism responsi- have symptoms or signs identifying them as being at high risk
ble for ventricular fibrillation, regional changes in automatic- before the event, a preventive approach to the problem
ity, as well as triggered activity due to afterdepolarizations, becomes complicated. Furthermore, to test the value of
are probably important as well. Reperfusion can also be primary preventive measures such as abstinence from smok-
arrhythmogenic, although the seriousness of this problem ing, exercise, weight reduction, control of high blood pres-
appears to be greater in the experimental animal than sure, and lipid abnormalities in patients without a history of
clinically. cardiac disease, studies have to be performed in communities
Cardiac arrest due to severe bradycardia, asystole, or in which it is possible not only to randomize to preventive
pulseless electrical activity (electromechanical dissociation) versus no preventive measures but also to register all cases of
appears to be more common in severely diseased hearts, sudden deaths accurately, including the unwitnessed ones.
probably representing more global myocardial dysfunction.23 Naturally, this is difficult, if not impossible. However, be-
The outlook for patients exhibiting these disturbances at the cause most cases of sudden cardiac death occur in the
time of attempted resuscitation is worse than for patients who population with coronary artery disease, it is logical that in
exhibit ventricular fibrillation at that time. recent years most attention has been given to secondary
A major, if not the major, electrophysiological feature preventive therapy in patients with proven coronary artery
responsible for the initiation of ventricular fibrillation appears disease and especially to survivors of a myocardial infarction.
to be electrical heterogeneity. A heart that is totally homoge-
neous electrically, that is, all cells are at the same stages of Risk Stratification for Treatment
depolarization and repolarization and conduct normally with- During the past 2 decades, a number of tests have been
out delay or block, very probably cannot develop ventricular developed to stratify cardiac patients as to their risk of dying
 Zipes and Wellens November 24, 1998 2347

TABLE 3. Test Used for Risk Stratification for Sudden Death TABLE 4. Measures to Reduce Sudden Death in Patients
Known to Have Cardiac Disease
Coronary perfusion
Coronary angiography Correcting ischemia
Exercise testing (including imaging) Revascularization
ST-segment changes using ambulatory recordings b-Blocking agent
Pump function Preventing plaque rupture
NYHA functional class Statin
Left ventricular ejection fraction ACE inhibitor
Exercise duration Aspirin
Arrhythmias Stabilizing autonomic balance
Long-term ambulatory recordings b-Blocking agents
Signal-averaged ECG ACE inhibitor
QT-interval duration, dispersion, and dynamic change Improving pump function
T-wave alternans ACE inhibitor
Exercise testing b-Blocking agent
Programmed electrical stimulation Prevention of arrhythmias
Neurohumoral b-Blocking agent
Heart rate variability Amiodarone
Baroreflex sensitivity
Psychosocial amiodarone, whereas a meta-analysis from 13 trials of 6500
Depression patients treated with amiodarone after myocardial infarction
or with heart failure showed a reduction in all-cause mortal-
ity, death from arrhythmia, or sudden death.71 Some data
suddenly. As shown in Table 3, these tests address different
suggest that amiodarone may be more effective when used for
cardiac and noncardiac factors that have been shown to affect
mortality. The relatively low positive predictive accuracy of patients with high (.90 bpm) resting heart rates. Preliminary
these tests adversely affects their usefulness. At best, alone or data also suggest increased effectiveness when amiodarone is
Downloaded from http://ahajournals.org by on June 19, 2025

in combination, the tests reach a positive predictive accuracy combined with a b-blocker. Both of these observations need
of 30%, indicating that if adequate protective treatment were further testing before adoption. The BHAT study showed that
available, 10 patients would have to be treated to save 3. b-blockade with propranolol reduced all-cause mortality by
Although this may not be a problem when low-cost, effective 25% and that the drug was especially useful in patients with
therapy free of adverse effects is possible, unfortunately, this diminished left ventricular function and/or ventricular ar-
is not the case. rhythmias. No evidence indicates that selective b-blockers
are better than nonselective ones. Although b-blocker therapy
Pharmacological Treatment has been shown to be advantageous and should be prescribed
Of the different drugs that have been evaluated, only for most patients after a myocardial infarction unless contra-
b-blockers and amiodarone have reduced sudden death in the indicated, in most countries a fraction of all patients who
myocardial infarction survivor.40 Class I drugs (mexiletine, should receive a b-blocker after a myocardial infarction
encainide, flecainide, moricizine), calcium antagonists, and actually do so. This is especially true for women, diabetics,
class III drugs (d-sotalol, dofetilide) all failed to reduce or and the elderly.77
even increased the incidence of sudden cardiac death after a In view of the complexity of the mechanisms involved in
myocardial infarction.40,41,71,72 A major problem in drug treat-
sudden cardiac death, there has been growing interest in the
ment, as shown by the ESVEM study,73 is that noninvasive
use of measures that may halt or delay progress of cardiac
(Holter) and invasive (programmed stimulation of the heart)
disease or prevent disturbances in the autonomic balance of
tests allow identification of an effective antiarrhythmic drug
the heart, such as the administration of anti-ischemic drugs,
regimen in only a minority of patients with documented
drugs to prevent plaque rupture or thrombus formation, and
life-threatening ventricular arrhythmias. In a study from
South America that included patients with different causes of drugs that stabilize the autonomic balance or improve pump
cardiac disease and diminished left ventricular function (ejec- function (Table 4). With this number and diversity of drugs,
tion fraction, ,35%), empirical amiodarone was shown to it has become increasingly difficult, if not impossible, to
beneficially affect mortality.18 This was not confirmed in a evaluate the individual contribution of each drug to the
multicenter VA trial, CHF-STAT.74 Two recent studies using reduction in sudden cardiac death. Aggressive therapy using
amiodarone in patients with reduced left ventricular function thrombolysis in acute ischemic syndromes or intracoronary
after a myocardial infarction showed a reduction in sudden interventions resulting in reduction of myocardial damage
(presumably arrhythmic) deaths but not in total number of and scar formation and prevention of ventricular remodel-
deaths.75,76 Importantly, these studies showed no increase in ing13,30 will diminish the occurrence of some of the mecha-
mortality compared with placebo for patients treated with nisms that play a role in a fatal arrhythmia. The role of
 2348 Sudden Cardiac Death

Figure 12. Treatment algorithm for patient

resuscitated from cardiac arrest. Cath indicates
catheterization; echo, echocardiography; EP,
electrophysiological study; ABN, abnormality;
and SVT, supraventricular tachycardia.

specific potassium channel blockers like dofetilide and azi- The value of the ICD in patients with markedly diminished
milide needs to be established in the future. ventricular function in the absence of ventricular fibrillation,
VT, or syncope, with or without a previous myocardial
Implantable Cardioverter-Defibrillator infarction, and with or without ambient ventricular ectopy is
For many patients who die suddenly, ventricular fibrillation is currently being evaluated. Trials like MUSST, MADIT-2,
the culprit arrhythmia. Superiority of an ICD over antiar- DEFINITE (Danish Investigations on Arrhythmias and Mor-
rhythmic drug therapy (predominantly amiodarone) has been tality on Dofetilide), ALIVE (azimilide postinfarct survival
Downloaded from http://ahajournals.org by on June 19, 2025

shown recently in the AVID trial64 in this patient population. evaluation), and SCD-HeFT26 all address patients with a 10%
Results consistent with the AVID study were also reported to 30% chance of dying within 2 years.80
from the CIDS78 and the CASH79 studies. At present, it is Again, we should realize that patients included in these
quite clear that an ICD is the initial treatment of choice for trials represent a minority of the total number of patients who
patients resuscitated from documented ventricular fibrillation die suddenly out-of-hospital (Figure 1). Protection against
not related to a reversible or transient cause such as an acute sudden cardiac death by ICD implantation is expensive, as
myocardial infarction, in patients with hemodynamically shown in the MADIT study ($27 000 per life year saved),81 in
poorly tolerated VT, and probably in patients with a history of which 100 devices had to be implanted to prevent 10 patients
unexplained syncope in the presence of impaired ventricular from dying suddenly, and therefore selection of the high-risk
function in whom a sustained ventricular arrhythmia can be population must have very high predictive accuracy.
induced during electrophysiological testing (Figure 12). The
Out-of-Hospital Resuscitation
MADIT study included patients after a myocardial infarction
The majority of sudden cardiac death victims have no
with a left ventricular ejection fraction ,35% and spontane-
symptoms and are not identified as being at high risk before
ously occurring nonsustained VT in whom a sustained mono-
the event.82 This stresses the enormous importance of improv-
morphic VT could be initiated by programmed electrical
ing the outcome of resuscitation attempts outside the hospital
stimulation of the heart and not prevented by procainamide (Figure 13). There is growing awareness that major changes
administration.29 Patients randomized to the defibrillator arm are necessary to reach that goal. The short time frame after
had an 80% reduction in sudden and a 54% reduction in total cardiac arrest during which circulation has to be restored to
mortality compared with patients treated with conventional prevent death or irreversible cerebral damage is essential.83 In
antiarrhythmic drug therapy after 2 years of follow-up, thus the so-called chain of survival, several steps are crucial. The
showing the benefit of prophylactic ICD placement in a first step is to identify and locate the sudden cardiac arrest
high-risk population. In the CABG Patch study, patients with victim. In the Maastricht study, 80% of cardiac arrests
left ventricular ejection fraction of ,36% and abnormal occurred at home, and 40% were unwitnessed.3 Therefore, we
signal-averaged ECG who were undergoing elective coronary must have warning systems able to recognize cardiac arrest,
bypass surgery were randomized to ICD or no antiarrhythmic to raise an alarm, and to transmit the exact location of the
therapy. CABG Patch showed no difference in survival victim to providers of basic and advanced life support. Much
between the two groups and concluded appropriately that if a attention has recently been given to public access defibrilla-
patient’s risk of dying is not from a ventricular tion, allowing nonphysicians to use widely distributed auto-
tachyarrhythmia, then the ICD conferred no benefit.46 mated external defibrillators to defibrillate.84 In fact, it was
 Zipes and Wellens November 24, 1998 2349

Figure 13. Survival before (open bars) and after

(shaded bars) beginning early defibrillation pro-
grams by emergency medical technicians. Repro-
duced with permission from Ornato JP, Om A.
Community experience in treating out-of-hospital
cardiac arrest. In: Akhtar M, Myerburg RJ, Ruskin
JN, eds. Sudden Cardiac Death. Baltimore, Md:
Williams & Wilkins; 1994:450 – 462.

suggested several years ago that external defibrillators be References

made “as common as fire extinguishers,”85 and they may have 1. Engelstein ED, Zipes DP. Sudden cardiac death. In: Alexander RW,
to be, to cover all the places a cardiac arrest can occur.86 This Schlant RC, Fuster V, eds. The Heart, Arteries and Veins. New York,
NY: McGraw-Hill; 1998:1081–1112.
idea has aroused a groundswell of enthusiasm but will 2. Myerburg RJ, Castellanos A. Cardiac arrest and sudden death. In:
obviously have the greatest impact when the sudden cardiac Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular
arrest victim is identified and located as soon as possible. The Medicine. Philadelphia, Pa: WB Saunders; 1997:742–779.
approach must be coupled with education, perhaps during the 3. de Vreede Swagemakers JJM, Gorgels APM, Dubois-Arbouw WI, van
Ree JW, Daemen MJAP, Houben LGE, Wellens HJJ. Out-of-hospital
4 years of high school,85 as well as legislative changes. When cardiac arrest in the 1990’s: a population-based study in the Maastricht
they function properly, devices that document cardiac arrest area on incidence, characteristics and survival. J Am Coll Cardiol. 1997;
and locate the victim might be distributed initially to cohorts 30:1500 –1505.
4. Reddy KS, Yusuf S. Emerging epidemic of cardiovascular disease in
with known cardiac disease and then among asymptomatic
Downloaded from http://ahajournals.org by on June 19, 2025

developing countries. Circulation. 1998;97:596 – 601.

people with risk factors for the development of cardiac 5. Neuspiel DR, Kuller LH. Sudden and unexpected natural death in
disease. It will be more expensive for standard emergency childhood and adolescence. JAMA. 1985;254:1321–1325.
medical systems to carry a defibrillator, but the potential 6. Mittleman MA, Maclure M, Tofler GH, Sherwood JB, Goldberg RJ,
Muller JE, Determinants of Myocardial Infarction Onset Study Investi-
impact on survival should warrant this approach. gators. Triggering of acute myocardial infarction by heavy physical
exertion: protection against triggering by regular exertion. N Engl J Med.
Conclusions 1993;329:1677–1683.
Sudden cardiac death continues to be a major health issue. At 7. Wang JS, Jen CJ, Kung HC, Lin LJ, Hsiue TR, Chen HI. Different effects
of strenuous exercise and moderate exercise on platelet function in men.
present, although insight into mechanisms and circumstances Circulation. 1994;90:2877–2885.
of sudden cardiac death is increasing, our methods for 8. Schwartz PJ, Zipes DP. Autonomic modulation of cardiac arrhythmias
identifying the high-risk candidate and predicting efficacy of and sudden cardiac death. In: Zipes DP, Jalife J, eds. Cardiac Electro-
physiology: From Cell to Bedside. Philadelphia, Pa: WB Saunders. In
measures to prevent sudden cardiac death are still inadequate.
press.
Because many victims are not known to suffer from heart 9. Liberthson RR. Sudden death from cardiac causes in children and young
disease and/or are considered to be at low risk for dying adults. N Engl J Med. 1996;334:1039 –1044.
suddenly, more efforts are needed to improve out-of-hospital 10. Maron BJ, Shirani J, Poliac LC, Mathenge R, Roberts WC, Mueller FO.
Sudden death in young competitive athletes: clinical, demographic, and
resuscitation by better warning systems and widespread pathological profiles. JAMA. 1996;276:199 –204.
availability of automated defibrillation devices. It is likely 11. Corrado D, Basso C, Schiavon M, Thiene G. Screening for hypertrophic
that these measures could increase the number of survivors of cardiomyopathy in young athletes. N Engl J Med. 1998;339:364 –369.
12. Link MS, Wang PJ, Pandian NG, Bharati S, Udelson JE, Lee M, Vecchiotti
cardiac arrest. Implantation of the ICD, in many instances
MA, Vander Brink BA, Mirra G, Maron BJ, Estes NAM. An experimental
probably combined with an antiarrhythmic drug like amiod- model of sudden death due to low-energy chest-wall impact (commotio
arone, would then be used to maintain survival. Until we have cordis). N Engl J Med. 1998;338:1805–1811.
better risk stratifiers and better methods of preventing ven- 13. Theroux P, Fuster V. Acute coronary syndromes: unstable angina and
non–Q-wave myocardial infarction. Circulation. 1998;97:1195–1206.
tricular tachyarrhythmias, the 2 major goals of the cardiolo- 14. Burke AP, Farb A, Malcom GT, Liang Y, Smialek J, Virmani R. Effect
gist/electrophysiologist, that approach should still receive a of risk factors on the mechanism of acute thrombosis and sudden coro-
major emphasis. nary death in women. Circulation. 1998;97:2110 –2116.
15. James TN, St Martin E, Willis PW III, Lohr TO. Apoptosis as a possible
cause of gradual development of complete heart block and fatal arrhyth-
Acknowledgments mias associated with absence of the AV node, sinus node, and internodal
This study was supported in part by the Herman C. Krannert Fund; pathways. Circulation. 1996;93:1424 –1438.
by grant HL-52323 from the National Heart, Lung, and Blood 16. Burke AP, Farb BA, Malcom GT, Liang YH, Smialek J, Virmani R.
Institute of the National Institute of Health, Bethesda, Md; and by the Coronary risk factors and plaque morphology in men with coronary heart
Wijnand N. Pon Foundation. disease who died suddenly. N Engl J Med. 1997;336:1276 –1282.
 2350 Sudden Cardiac Death

17. Albert CM, McGovern BA, Newell JB, Ruskin JN. Sex differences in 38. Marian AJ, Roberts R. Molecular genetic basis of hypertrophic cardio-
cardiac arrest survivors. Circulation. 1996;93:1170 –1176. myopathy: genetic markers for sudden cardiac death. J Cardiovasc Elec-
18. Doval HC, Nul DR, Grancelli HO, Varini SD, Soifer S, Corrado G, trophysiol. 1998;9:88 –99.
Dubner S, Scapin O, Perrone SV, GESICA-GEMA Investigators. Non- 39. Nattel S, Pedersen DH, Zipes DP. Alterations in regional myocardial
sustained ventricular tachycardia in severe heart failure: Independent distribution and arrhythmogenic effects of aprindine produced by coro-
marker of increased mortality due to sudden death. Circulation. 1996;94: nary artery occlusion in the dog. Cardiovasc Res. 1981;15:80 – 85.
3198 –3203. 40. Cardiac Arrhythmia Suppression Trial (CAST) II Investigators. Effect of
19. Escobedo LG, Zack MM. Comparison of sudden and nonsudden coronary the antiarrhythmic agent moricizine on survival after myocardial
deaths in the United States. Circulation. 1996;93:2033–2036. infarction. N Engl J Med. 1992;327:227–233.
20. Leor J, Poole WK, Kloner RA. Sudden cardiac death triggered by an 41. Greenberg HM, Dwyer EM, Hochman JS, Steinberg JS, Echt DS, Peters
earthquake. N Engl J Med. 1996;334:413– 419. RW. Interaction of ischemia and encainide/flecainide treatment: a
21. Brugada P, Talajic M, Smeets J, Mulleneers R, Wellens HJ. The value of proposed mechanism for the increased mortality in CAST I. Br Heart J.
the clinical history to assess prognosis of patients with ventricular 1995;74:631– 655.
tachycardia or ventricular fibrillation after myocardial infarction. Eur 42. La Rovere MT, Bigger JT, Marcus FI, Mortona A, Schwartz PJ.
Heart J. 1989;10:747–752. Baroreflex sensitivity and heart rate variability in mediation of total
22. Friedlander Y, Siscovick DS, Weinmann S, Austin MA, Psaty BM, cardiac mortality after myocardial infarction. Lancet. 1998;351:478 – 484.
Lemaitre RN, Arbogast P, Raghunathan TE, Cobb LA. Family history as 43. Bigger JT. Heart rate variability and sudden cardiac death. In: Zipes DP,
a risk factor for primary cardiac arrest. Circulation. 1998;97:155–160. Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Phila-
23. Stevenson WG, Stevenson LW, Middlekauff HR, Saxon LA. Sudden delphia, Pa: WB Saunders; 1999. In press.
death prevention in patients with advanced ventricular dysfunction. Cir- 44. Chiou C-W, Zipes DP. Selective vagal denervation of the atria eliminates
culation. 1993;88:2953–2961. heart rate variability and baroreflex sensitivity while preserving ventric-
24. Weinberg BA, Miles WM, Klein LS, Bolander JE, Dusman RE, Stanton ular innervation. Circulation. 1998;98:360 –368.
MS, Heger JJ, Langefeld C, Zipes DP. Five-year follow-up of 589 45. Stein KM, Karagounis LA, Anderson JL, Kliegfield P, Lerman BB.
patients treated with amiodarone. Am Heart J. 1993;125:109 –120. Fractal clustering of ventricular ectopy correlates with sympathetic tone
25. Herre JM, Sauve MJ, Malone P, Griffin JC, Helmy I, Langberg JJ, preceding ectopic beats. Circulation. 1995;91:722–727.
Goldberg H, Scheinman MM. Long-term results of amiodarone therapy in 46. Bigger JT. Prophylactic use of implanted cardiac defibrillators in patients
patients with recurrent sustained ventricular tachycardia or ventricular at high risk for ventricular arrhythmias after coronary artery bypass graft
fibrillation. J Am Coll Cardiol. 1989;13:442– 449. surgery. N Engl J Med. 1997;337:1569 –1575.
26. Bardy GH, SCD-HeFT Investigators. Prevention of sudden cardiac death 47. Estes NAM, Michaud G, Zipes DP, El-Sherif N, Vendetti FJ, Rosenbaum
in patients with congestive heart failure. In: Woosley RL, Singh SN, eds. DS, Albrecht P, Wang PJ, Cohen RJ. Electrical alternans during rest and
Clinical Trials on the Treatment of Arrhythmia. New York, NY: Marcel exercise as predictors of vulnerability to ventricular arrhythmias. Am J
Decker, Inc. In press. Cardiol. 1997;80:1314 –1318.
27. Zabel M, Klingenheben T, Franz MR, Hohnloser SH. Assessment of QT 48. Roberts WC, Kragel AH, Bertz D, Roberts LS. Coronary arteries in
dispersion for prediction of mortality or arrhythmic events after myo- unstable angina pectoris, acute myocardial infarction and sudden cardiac
cardial infarction: results of a prospective, long-term follow-up study. death. Am Heart J. 1994;127:1588 –1593.
Circulation. 1998;97:2543–2550. 49. Spirito P, Seidman CE, McKenna WJ, Maron BJ. The management of
Downloaded from http://ahajournals.org by on June 19, 2025

28. Zipes DP. Unwitting exposure to risk. Cardiol Rev. 1993;1:1–3. hypertrophic cardiomyopathy. N Engl J Med. 1997;336:775–785.
29. Moss AJ, Hall WJ, Cannom DS, Daubert JP, Higgins SL, Klein H, Levine 50. Tamburro P, Wilber D. Sudden death in idiopathic dilated cardiomyop-
JH, Saksena S, Waldo AL, Wilber D, Brown MW, Heo M, Multicenter athy. Am Heart J. 1992;124:1035–1045.
Automatic Defibrillator Implantation Trial Investigators. Improved 51. Blanck Z, Dhala A, Deshpande S, Sra J, Jazayeri M, Akhtar M. Bundle
survival with an implanted defibrillator in patients with coronary disease branch reentrant ventricular tachycardia: cumulative experience in 48
at high risk for ventricular arrhythmia. N Engl J Med. 1996;335: patients. J Cardiovasc Electrophysiol. 1993;4:253–262.
1933–1940. 52. Corrado D, Basso C, Thiene G, McKenna WJ, Davies MJ, Fontaliran F,
30. Fuster V, Poon M, Willerson JT. Learning from the transgenic mouse: Nava A, Silvestri F, Blomstrom-Lundqvist C, Wlodarska EK, Fontaine G,
endothelium, adhesive molecules, and neointimal formation. Circulation. Camerini F. Spectrum of clinicopathologic manifestations of arrhyth-
1998;97:16 –18. mogenic right ventricular cardiomyopathy/dysplasia: a multicenter study.
31. Tomaselli GF, Beuckelmann DJ, Calkins HG, Berger RD, Kessler PD, J Am Coll Cardiol. 1997;30:1512–1520.
Lawrence JH, Kass D, Feldman AM, Marban E. Sudden cardiac death in 53. Gatzoulis MA, Till JA, Somerville J, Redington AN. Mechanoelectrical
heart failure: the role of abnormal repolarization. Circulation. 1994;90: interaction in tetralogy of Fallot: QRS prolongation relates to right ven-
2534 –2539. tricular size and predicts malignant ventricular arrhythmias and sudden
32. Kaab S, Nuss HB, Chiamvimonvat N, O’Rourke B, Pak PH, Kass DA, death. Circulation. 1995;92:231–237.
Marban E, Tomaselli GF. Ionic mechanism of action potential prolon- 54. Bardy GH, Bigger JT, Borggrefe M, Camm AJ, Cobb LA, Ewy GA,
gation in ventricular myocytes from dogs with pacing-induced heart Hauer RNW, Kuck HH, Lane RD, Lazarra R, Marcus FI, Muller JE,
failure. Circ Res. 1996;78:262–273. Myerburg RJ, Priori SG, Schwartz PJ, Touboul P, Verrier RL, Wellens
33. Mitrani RD, Klein LS, Miles WM, Hackett FK, Burt RW, Wellman HN, HJJ, Zipes DP. Survivors of out-of-hospital cardiac arrest with apparently
Zipes DP. Regional cardiac sympathetic denervation in patients with normal heart: need for definition and standardized clinical evaluation.
ventricular tachycardia in the absence of coronary artery disease. J Am Circulation. 1997;95:265–272.
Coll Cardiol. 1993;22:1344 –1353. 55. Brugada P, Brugada J. Right bundle branch block, persistent ST segment
34. Calkins H, Allman K, Bolling S, Kirsch M, Wieland D, Morady F, elevation and sudden cardiac death: a distinct clinical and electrocardio-
Schwaiger M. Correlation between scintigraphic evidence of regional graphic syndrome. J Am Coll Cardiol. 1992;20:1391–1396.
sympathetic neuronal dysfunction and ventricular refractoriness in the 56. de Meeus A, Stephan E, Debrus S, Jean MK, Loiselet J, Weissenbach J,
human heart. Circulation. 1993;88:172–179. Demaille J, Bouvagnet P. An isolated cardiac conduction disease maps to
35. Muller JE, Kaufmann PG, Luepker RV, Weisfeldt ML, Deedwania PC, chromosome 19q. Circ Res. 1995;77:735–740.
Willerson JT, for the Mechanisms Precipitating Acute Cardiac Events 57. Englestein ED, Sawada S, Hutchins GD, Straka S, Sehra R, Fain RL,
Participants. Mechanisms precipitating acute cardiac events: review and Amaravadi R, Miles WM, Zipes DP. Idiopathic ventricular tachycardia is
recommendations of an NHLBI workshop. Circulation. 1997;96: not really idiopathic: structural ventricular abnormalities detected by
3233–3239. positron emission tomography. J Am Coll Cardiol. 1998;31:180A.
36. Peckova M, Fahrenbach CE, Cobb LA, Hallstrom AP. Circadian vari- Abstract.
ations in the occurrence of cardiac arrests: initial and repeat episodes. 58. Chen Q, Kirschott GE, Zhang D, Brugada R, Brugada P, Patenzas D, Moya
Circulation. 1998;98:31–39. A, Borggrefe M, Breithardt G, Ortiz-Lopez R, Wang Z, Antzelevich C,
37. Schwartz PJ, Priori SG, Napolitano C. Long QT syndrome. In: Zipes DP, O’Brien RE, Schulze-Bahr E, Keating J, Towbin JA, Wang Q. Genetic basis
Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Orlando, and molecular mechanism for idiopathic ventricular fibrillation. Nature.
Fla: WB Saunders. 1999. In press. 1998;392:293–296.
 Zipes and Wellens November 24, 1998 2351

59. Satoh T, Zipes DP. Cesium-induced atrial tachycardia degenerating into 74. Massie BM, Fisher SG, Radford M, Deedwania PC, Singh BN, Fletcher
atrial fibrillation: atrial torsades de pointes? J Cardiovasc Electrophysiol. RD, Singh SN. Effect of amiodarone on clinical status and left ventricular
1998;9:970 –975. function in patients with congestive heart failure. Circulation. 1996;93:
60. Bromberg BI, Lindsay BD, Cain ME, Cox JL. Impact of clinical history 2128 –2134.
and electrophysiologic characterization of accessory pathways on man- 75. Cairns JA, Connolly SJ, Roberts R, Gent M. Randomized trial of outcome
agement strategies to reduce sudden death among children with Wolff- after myocardial infarction in patients with frequent or repetitive ventric-
Parkinson-White syndrome. J Am Coll Cardiol. 1996;27:690 – 695. ular premature depolarizations: CAMIAT. Lancet. 1997;349:675– 682.
61. Lerman BB, Stein KM, Markowitz SM. Adenosine-sensitive ventricular 76. Julian DG, Camm AJ, Frangin G, Janse MJ, Munoz A, Schwartz PJ,
tachycardia: a conceptual approach. J Cardiovasc Electrophysiol. 1996;
Simon P, European Myocardial Infarct Amiodarone Trial Investigators.
7:559 –569.
Randomised trial of effect of amiodarone on mortality in patients with
62. Viscenberg SJ, Scheinman MM, Bullet MK, Finkbeiner WE, Grifin JC,
left-ventricular dysfunction after recent myocardial infarction: EMIAT.
Eldar M, Franz MR, Gonzalez R, Kadish AH, Lesh MD. Sudden cardiac
death and polymorphous ventricular tachycardia in patient with normal Lancet. 1997;349:667– 674.
QT intervals and normal systolic cardiac function. Am J Cardiol. 1995; 77. Woods KL, Ketley D, Lowy A, Augusti A, Hagn C, Kala R, Karatzas NB,
75:687– 692. Leizorowicz A, Reikvam A, Schilling J, Seabra-Gomes R, Vasiliauskas
63. Leenhardt A, Glaser E, Burguera M, Nurnberg M, Maison-Blanche P, D, Wilhelmsen L, the European Secondary Prevention Study Group.
Coumel P. Short-coupled variant of torsades de pointes: a new electro- Beta-blockers and antithrombotic treatment for secondary prevention
cardiographic entity in a spectrum of idiopathic ventricular after acute myocardial infarction: towards an understanding of factors
tachyarrhythmias. N Engl J Med. 1994;337:1576 –1583. influencing clinical practice. Eur Heart J. 1998;19:74 –779.
64. AVID Investigators. The antiarrhythmics versus implantable defibril- 78. Connolly S. Canadian Implantable Defibrillator Study (CIDS). Presented
lators (AVID) investigators: a comparison of antiarrhythmic drug therapy at the 47th Annual Scientific Session of the American College of Cardi-
with implantable defibrillators in patients resuscitated from near fatal ology, Atlanta, Ga, March 30, 1998.
ventricular arrhythmias. N Engl J Med. 1997;337:1576 –1583. 79. Kuck KH. Cardiac Arrest Study Hamburg (CASH). Presented at the 47th
65. Deleted in proof. Annual Scientific Session of the American College of Cardiology, 1998.
66. Hull SS, Vanoli E, Adamson PB, DeFerrari GM, Foreman RD, Schwartz 80. Naccarelli GV, Wolbrette DL, Dell’Orfano JT, Patel HM, Luck JC. A
PJ. Do increases in markers of vagal activity imply protection from decade of clinical trial developments in post-myocardial infarction, con-
sudden death? The case of scopolamine. Circulation. 1995;91: gestive heart failure and sustained ventricular tachyarrhythmia patients:
2516 –2519.
from CAST to AVID and beyond. J Cardiovasc Electrophysiol. 1998;9:
67. Casadei B, Pipilis A, Sessa F, Conway J, Sleight P. Low doses of
864 – 891.
scopolamine increase cardiac vagal tone in the acute phase of myocardial
81. Mushlin AI, Hall WJ, Zwanziger J, Gajary E, Andrews M, Marron R, Zou
infarction. Circulation. 1993;88:353–357.
KH, Moss AJ. Cost-effectiveness of automatic implantable cardiac defi-
68. Fei L, Baron A, Henry DP, Zipes DP. Intrapericardial delivery of
L-arginine reduces the increased severity of ventricular arrhythmias in
brillators: results from MADIT. Circulation. 1998;97:2129 –2135.
dogs with acute coronary occlusion: nitric oxide modulates sympathetic 82. Kerber RE, Becker LB, Bourland JD, Cummins RO, Hallstrom AP,
effects on ventricular electrophysiological properties. . Circulation. 1997; Michos MB, Nichol G, Oranto JP, Thies WH, White RD, Zuckerman BD.
96:4044 – 4049. Automatic external defibrillators for public access defibrillation: recom-
69. Owens LM, Fralix TA, Murphy E, Cascio WE, Gettes LS. Correlation of mendations for specifying and reporting arrhythmia analysis algorithm
Downloaded from http://ahajournals.org by on June 19, 2025

ischemia-induced extracellular and intracellular ion changes to cell- performance, incorporating new wave forms, and enhancing safety. Cir-
to-cell electrical uncoupling in isolated blood-perfused rabbit hearts. culation. 1997;95:1677–1682.
Circulation. 1996;94:10 –13. 83. Valenzuela TD, Roe DJ, Cretin S, Spaite DW, Larson MP. Estimating
70. Satoh T, Zipes DP. Unequal atrial stretch in dogs increases dispersion of effectiveness of cardiac arrest interventions: a logistic regression survival
refractoriness conducive to developing atrial fibrillation. J Cardiovasc model. Circulation. 1997;96:3308 –3313.
Electrophysiol. 1996;7:833– 842. 84. Nichol G, Hallstrom AP, Kerber R, Moss AJ, Ornato JP, Palmer D,
71. Amiodarone Trials Meta-Analysis Investigators. Effect of prophylactic Riegel B, Smith S, Weisfeldt ML. American Heart Association Report on
amiodarone on mortality after acute myocardial infarction and in con- the Second Public Access Defibrillation Conference, April 17–19, 1997.
gestive heart failure: meta-analysis of individual data from 6500 patients Circulation. 1998;97:1309 –1314.
in randomized trials. Lancet. 1997;350:1417–1424. 85. Zipes DP. Sudden cardiac death: future approaches. Circulation. 1992;
72. Waldo A, Camm JA, DeRuyter H, Friedman PL, MacNeil DJ, Pauls JF,
85(suppl I):I-160 –I-166.
Pitt B, Pratt CM, Schwartz PJ, Veltri EP. Effect of d-sotalol on mortality
86. Becker L, Eisenberg M, Fahrenbruch C, Cobb L. Public locations of
in patients with left ventricular dysfunction after recent and remote
cardiac arrest: implications for public access defibrillators. Circulation.
myocardial infarction. Lancet. 1996;348:7–12.
73. Mason JA. A comparison of electrophysiologic testing with Holter mon- 1998;97:2106 –2109.
itoring to predict antiarrhythmic drug efficacy for ventricular
tachyarrhythmias. N Engl J Med. 1993;329:445– 451. KEY WORDS: death, sudden n mortality n coronary disease