|Home | About | Journals | Submit | Contact Us | Français|
Sudden cardiac death (SCD), which is usually defined as death from cardiac causes within an hour of symptom onset, affects more than 3 million people annually worldwide.1 In the United States, estimates of annual sudden deaths range from 200,000 to 450,000, but are typically cited as accounting for 300,000 annually.2 With an estimated population of 300 million people in the United States, approximately 1/1,000 will die suddenly each year.
Clinical factors used to predict sudden death include a prior history of aborted sudden death, left ventricular dysfunction, and the diagnosis of inherited syndromes associated with arrhythmias such as hypertrophic cardiomyopathy, arrhythmogenic right ventricular cardiomyopathy, long QT syndrome, and Brugada syndrome.3, 4 Diagnostic studies such as programmed electrical stimulation and T-wave alternans have proved disappointing, and the majority of SCD occurs in people without overt risk factors.5 Pharmacological therapy, with the exception of β-adrenergic blockade, does not prevent SCD.6 Implantable cardioverter defibrillators (ICDs) are effective therapy for those at risk for SCD, but their use is associated with complications during implantation, device and lead failure, inappropriate shocks, limitations to quality of life, and cost.7 In addition, when used as primary prevention for SCD in cardiomyopathy, as many as 10 ICDs must be placed to prevent one sudden death.
Our laboratory and others are searching for novel biomarkers to identify heart failure patients at highest risk for sudden death.8 In this chapter, we will discuss the development and use of genomic predictors to define the population at risk for sudden cardiac death.
Prolonged action potential duration (APD) and downregulation of the repolarizing K+ currents Ito and IK1 are present in tissue and cardiac myocytes isolated from patients and animal models with heart failure.9–12 This delayed repolarization, along with enhanced dispersion of repolarization, may contribute to arrhythmias and sudden death.13–15 Mechanisms leading to arrhythmias may include triggered activity, such as early afterdepolarizations (EADs, resulting from recovery from inactivation of inward calcium channels) and delayed afterdepolarizations (DADs, resulting from Ca2+ release from overloaded internal stores). In addition, reentry leading to the rotors and scroll waves that cause ventricular tachycardia and fibrillation may be enhanced by abnormalities in intercellular communication with slow conduction, anatomical abnormalities such as fibrosis and scars, and heterogeneities in ion channel distribution in the myopathic heart.16–19
Changes in Ca2+ handling are also well documented in animal models and humans with heart failure, and may contribute to arrhythmias.18, 20–22 In most cases, sarcoplasmic reticulum calcium ATPase 2a (SERCA2a) is decreased, the ratio of SERCA2a to phospholamban (an inhibitor of SERCA2a) is decreased, and the Na-Ca exchanger is increased, leading to smaller but prolonged Ca2+ transients and decreased myocyte contractile function. As noted above, these changes may contribute to afterdepolarizations. In addition, abnormal Ca2+ handling appears to facilitate reentrant arrhythmias in both ischemic and nonischemic cardiomyopathies.23 Beta adrenergic receptor downregulation and changes in protein phosphorylation further decrease contractile reserve in heart failure and may contribute to arrhythmias.24–26
During the 1970’s, ventricular tachycardia and fibrillation were identified as the primary cause of SCD in the context of coronary artery disease27, 28 and the benefits of external defibrillation were shown.29, 30 Because that the majority of SCD occurs out of the hospital and away from facilities with external defibrillators,31 investigators developed the implantable cardioverter defibrillator.32 ICDs were initially used in the 1980’s with the primary indication being the treatment of malignant, refractory ventricular arrhrythmias.33 Randomized controlled trials subsequently showed that ICDs improved mortality in survivors of aborted SCD (secondary prevention), and in subjects with ischemic or non-ischemic cardiomyopathies (primary prevention).34–37
In patients with severe left ventricular dysfunction, SCD rates generally range from 4–6% annually.38,39 Aggressive implementation of “standard of care” medical therapy including β-blockers and angiotensin converting enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs)/aldosterone inhibitors may decrease SCD rates40–42, although the delay in mortality from pump failure might lead to an increase in the overall incidence of SCD. Antiarrhythmic medications other than β-blockers can cause sudden death (e.g. flecainide), or at best not alter mortality (e.g. amiodarone).6, 43 Current ACC/AHA/HRS guidelines recommend ICD placement for subjects with NYHA Class II or III heart failure and EF <35% due to a nonischemic cardiomyopathy or a prior MI at least 40 days previously, and for subjects who are NYHA Class I with an EF <30% from a prior MI at least 40 days previously.44
It is estimated that only 1 in 11 people who receive an ICD for primary prevention in cardiomyopathy will experience a SCD event for which the ICD would deliver an appropriate shock.5, 7 There are complications associated with ICD implantation, including infection and bleeding at the time of placement, device and lead failure at later times, inappropriate shocks for atrial tachycardias, and limitations to type of employment, quality of life and leisure activities. In addition, the high cost of ICD therapy raises the questions regarding whether the health system can afford ICD therapy in everyone who currently has an indication for a device. As such, clinical tools beyond EF have been sought to predict the risk of SCD. Programmed electrical stimulation, signal averaged ECG, and T-wave alternans have proven disappointing in their ability to identify high-risk patient subsets.45 The identification of genomic predictors of SCD risk could help to tailor ICD therapy to a highest risk subgroup of patients.
A number of studies support the presence of a significant heritable component to sudden cardiac death.8 In the Family History and Primary Cardiac Arrest study from Seattle, 235 out-of-hospital SCD subjects (with no known history of coronary artery disease, congenital heart disease, or other severe illnesses) were compared to matched controls to assess the importance of a family history of SCD and myocardial infarction (MI).46 A history of SCD in a parent or a first degree relative was associated with a greater than 2-fold increased risk of cardiac arrest for age <65, independent of a family history of MI. In the Paris Prospective Study I, more than 7,700 subjects with known ischemic heart disease were prospectively followed for 5 years for SCD. After correcting for known risk factors and parental history of MI, subjects having a parent with a history of SCD had a 1.8-fold higher risk of SCD.47 More recently, a retrospective case-control study showed that the incidence of SCD was 1.6-fold higher in relatives of subjects with SCD than in relatives of subjects with an MI but no SCD, and 2.2-fold higher in relatives of SCD than in relatives of controls.48 In addition, a case-control study comparing ST elevation MI patients with or without primary VF showed a 2.7-fold higher incidence of familial sudden death in cases versus controls.49
Thus, there is convincing evidence for a genetic component to the risk of SCD in the setting of acute MI. The genetic basis for this risk and its extension to sudden death in the setting of heart failure remain to be clarified.
During the last 15 years, the molecular determinants of a number of inherited arrhythmia syndromes associated with sudden cardiac death (long QT syndrome, short QT syndrome, Brugada syndrome, catecholaminergic polymorphic ventricular tachycardia) have been determined by positional cloning and the candidate gene approach (Table 1).50–52 The majority of mutations identified to date are in ion channels or ion-channel related genes. While it makes sense that ion channel genes should cause these syndromes, it is worth noting that investigators preferentially search for mutations in ion channel genes and other genetic causes may be underestimated.
Previously unidentified rare mutations in these ion channel genes are not responsible for a significant fraction of sudden cardiac death. In the Oregon Sudden Unexpected Death Study (Ore-SUDS), amino acid-altering variants of SCN5A did not contribute to the risk of sudden death in 67 SCD victims.53 Similarly, no pathological mutations were identified in SCN5A or KvLQT1 in 59 Australian SCD victims.54 In contrast, sequencing of SCN5A, KCNQ1, KCNH2, KCNE1, and KCNE2 in 113 SCD cases from the Nurses’ Health Study and the Health Professional Follow-Up Study identified 5 rare SCN5A variants in 6 women, a frequency greater than in controls.55 While similar studies have not been done for heart failure patients, rare ion channel mutations are unlikely to play a major role.
A number of relatively common single nucleotide polymorphisms (SNPs) that lead to amino acid changes have been identified in cardiac ion channel genes, along with many more in introns and non-coding regions that could affect RNA and protein expression (Table 2). If rare mutations in ion channel genes can cause sudden death in patients with structurally normal hearts, it seems logical to hypothesize that the SNPs that cause relatively minor changes in ion channel function or numbers could predispose individuals with structural heart disease and heart failure to arrhythmias and sudden death. In support of this hypothesis, the S1103Y polymorphism in the cardiac Na+ channel, which is relatively common only in African Americans (allele frequency ~7%) and causes subtle QT prolongation in vitro, is associated with unexplained arrhythmias, sudden death and sudden infant death syndrome.56–58 In addition, the SCN5A H558R polymorphism, which is common in Caucasians (allele frequency ~20%), has been shown to a) alter current expression in conjunction with an SCN5A splice variant in vitro, b) rescue Brugada syndrome mutations whether on the same or the other chromosome in vitro, and c) potentially explains some cases of incomplete penetrance.59–61 It remains to be proven whether any of these polymorphisms are associated with arrhythmia susceptibility and/or sudden death in heart failure patients.
β-adrenergic receptor polymorphisms have been associated with differences in heart failure progression and altered pharmacogenetic responses to β-adrenergic blockade.62, 63 Of note, homozygotes for the glutamine allele of the β2-AR Q27E polymorphism had an ~1.6 fold increased risk of sudden death in both the Cardiovascular Health Study and the Cardiac Arrest Blood Study.64 Similarly, we recently reported an ~1.7-fold increased appropriate shock frequency for glutamine homozygotes in patients with heart failure and ICDs enrolled in the Genetic Risk Assessment of Defibrillator Events (GRADE) study.65
Genome-wide association studies (GWAS) have been used to identify genes associated with QTc prolongation on the surface electrocardiogram, an intermediate phenotype expected to correlate with sudden death. Multiple studies have identified several SNPs in the nitric oxide synthase 1 adaptor protein (NOS1AP) responsible for 3–5 ms of QTc prolongation per allele.66, 67 SNPs in the same region of NOS1AP were associated with an ~1.3 fold per allele increased risk of sudden cardiac death in subjects in the Atherosclerosis Risk In Communities Study and the Cardiovascular Health Study.68 It remains to be seen whether similar findings are true in heart failure populations.
We do not currently know all of the genes that are associated with arrhythmias and sudden death. Genes associated with inflammation, redox state, and omega-3 fatty acid metabolism are other potential candidates.69, 70 GWAS studies will identify additional genes associated with QTc prolongation that are candidate genes for sudden death, and may be used to directly identify sudden death genes in the future.
The identification of genetic predictors of sudden death in heart failure is in its earliest stages. Mutations in ion channels have been shown to cause inherited forms of sudden death; there is, however, little evidence that mutations or rare SNPs in those genes are important causes of the common forms of sudden death.8 Other common variants in ion channels and related genes are associated with sudden death in the setting of acute MI and/or heart failure. Unfortunately, the mechanisms of sudden death in heart failure differ from those in the setting of acute MI. It is likely that multiple genetic variants will interact to control risk, and that different variants will be relevant for different underlying disease states. In addition, potential interactions with race, gender, environment, and other cardiovascular risk factors may further complicate the analysis.
Ultimately, we hope to identify a handful of SNPs that modulate the risk of sudden death in heart failure and develop an algorithm to predict risk based on genotype. This will require prospective testing in large heart failure cohorts, most likely using appropriate ICD shocks as a surrogate for sudden cardiac death. Of note, however, appropriate ICD shock rates significantly exceeds the rates of sudden death.71
A significant question is whether genetic testing can or should be used to restrict ICD placement. If successful, genetic predictors would be used in conjunction with other clinical predictors of sudden death risk. Currently, the beneficial effects of ICDs in preventing sudden death in heart failure patients are counteracted by the relatively low likelihood of sudden death, the morbidity associated with device placement, ongoing complications from the device, effects on quality of life, and cost. As a result, many physicians are reluctant to strongly advise ICD placement in heart failure patients and only a fraction of patients with indications for ICDs accept the therapy. Better predictors of sudden death risk could identify a group of patients with the highest risk who would benefit most from ICD placement, and might identify patients with less severe left ventricular dysfunction at sufficient risk to warrant ICD placement. In addition, delineation of a group of patients at low risk for sudden death could shift the focus on that cohort from devices to optimal medical management.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.