Sudden cardiac arrest (SCA) can be defined as the abrupt cessation of cardiac activity due to an underlying cardiac cause, occurring instantaneously in a previously stable patient and in the absence of non-cardiovascular causes (e.g., trauma, intoxication, drowning, electrocution).1,2 SCA will lead to loss of consciousness within the minute due to insufficient cerebral perfusion. If no immediate action is taken to restore circulation – e.g., defibrillation – SCA will invariably lead to sudden cardiac death (SCD). In this article, we will use SCD as the common term for both SCA and SCD.
A Major Cause of Death with a High Impact on Daily Life
The mechanism underlying SCD is arrhythmic in the majority (80–90 %) of cases3 and over 80 % of the arrhythmic episodes are ventricular arrhythmia (VA): ventricular tachycardia (VT) or ventricular fibrillation (VF).4 As, in many cases, the deadly event is not witnessed (e.g., occurring during sleep and/or out of hospital), no definition of SCD is 100 % accurate.1 This explains why epidemiological data vary from report to report. However, it is clear that SCD is a very frequent cause of mortality worldwide. Incidence estimates from large population studies in the 1990s were as high as 450,000 SCD cases per year in the US,5 400,000 in Europe6 and 3,000,000 worldwide.7 SCD accounts for over 50 % of all cardiac-related deaths in the US8,9 and it is the second leading cause of death after all cancers combined.10 Even though, during the last 20 years, the significant decrease in the incidence of cardiovascular mortality has also been observed for SCD,11 SCD remains a very important health issue. Not only because its incidence remains high, but also because the abrupt and unforeseeable character of the condition leads to very low (less than 10 %) out-of-hospital survival rates11 and loss of life for people who were often free of morbidity up until right before the event.
A Prism of Possible Underlying Causes
The underlying conditions that form the substrate of, and the trigger for, VA leading to SCD are very diverse. In less than 10 % of the patients experiencing SCD, no macroscopic structural heart disease is found.12 This subgroup consists of patients with congenital electrophysiological anomalies, such as the congenital long and short QT syndromes, Brugada syndrome, catecholaminergic polymorphic VT and idiopathic VF. These conditions are a particularly important cause of death in young people. The current knowledge about the origin, detection and treatment of these individual conditions, which is continuously evolving, is beyond the scope of this article.
The majority of SCD cases are accompanied by structural cardiac abnormalities of which SCD is often the first presentation. These abnormalities can be divided between coronary artery disease (myocardial ischaemia and infarction), congestive heart failure, ventricular hypertrophy, arrhythmogenic ventricular cardiomyopathy or a combination of these. Up to 70 % of all SCDs are related to myocardial ischaemia or subsequent infarction and heart failure. SCD in turn is the cause of mortality in up to 50 % of all deaths due to ischaemic heart disease.13 In this article, we will focus on the prevention of SCD in ischaemic cardiomyopathy.
Arrhythmogenic Mechanisms Specific to Ischaemic Cardiomyopathy
SCD is caused in the majority of cases by VA – that is, by VT, VF or VT degenerating into VF. Arrhythmogenesis results from the interaction between a transient initiating event, a cardiac substrate and an arrhythmic mechanism.14 In ischaemic cardiomyopathy, two specific underlying pathophysiological mechanisms leading to VA predominate.
First, acute myocardial ischaemia due to insufficient coronary blood supply and post-ischaemic reperfusion can induce VA through abnormal automaticity.14 This mechanism is responsible for the high incidence of VA during the first 48 hours after acute myocardial infarction (MI). Since the 1980s, strategies for rapid coronary reperfusion, continuous electrocardiogram (ECG) monitoring within specialised coronary care units and early external defibrillation when needed have resulted in a drastic decrease in mortality in the acute phases of acute MI.11 Improved protection against this arrhythmia mechanism is offered by the elimination of behavioural cardiovascular risk factors (e.g., smoking and obesity), optimal reperfusion therapy and the pharmacological prevention of recurrent coronary ischaemia (through, for example, statins, platelet aggregation inhibitors, beta-blockers and angiotensin-converting enzyme [ACE] inhibitors).14
The other main arrhythmic mechanism in ischaemic heart disease is re-entry in myocardium scarred by infarction. It is believed that the conditions for re-entry are established through disruption and reorganisation of the intercellular gap junctions in the border zone of healing infarcts.15 In contrast to abnormal automaticity induced by acute ischaemia, this substrate is not transient but permanent and it is responsible for the ‘late’ presentation of SCD in patients with previous MI. The presence of this substrate for re-entry will put patients at permanent risk of developing life-threatening arrhythmia, in spite of an optimal reduction in recurrent ischaemia risk as discussed above.
Prevention of Sudden Cardiac Death in Ischaemic Cardiomyopathy
In the last 20 years, several options have been developed to prevent and treat VA leading to SCD.
Pharmacological Strategies
The prevention of recurrent ischaemia to prevent SCD in ischaemic cardiomyopathy has already been mentioned. Given that scar tissue is a chronic substrate of re-entry VA, pharmacological therapies limiting myocardial injury and adverse remodelling also reduce the risk of SCD. This was repeatedly proven for ACE inhibitors, making them a cornerstone of post-infarct treatment.16 During the last 20 years, there has been an extensive quest to find an anti-arrhythmic agent that can reduce SCD risk once the arrhythmic substrate is present. Only beta-blockers have been shown to reduce SCD in patients with previous MI.17 This is due to their protective effect against recurrent ischaemia and alleviation of sympathetic tone, but probably also to intrinsic electrophysiological effects.18 The results with other anti-arrhythmic agents have been disappointing. Class I (mexiletine, encainide, flecainide, moricizine), class III (sotalol, dofetilide) and class IV (calcium antagonists) drugs have all failed to reduce, or have even increased, the incidence of SCD after MI.14 Finally, even amiodarone, having the most potent antiarrhytmic effects with almost no proarrhythmic risk (but frequent and important negative side effects), failed to reduce mortality in several large randomised controlled trials (RCTs).19
Implantable Cardioverter-defibrillators
The implantable cardioverter-defibrillator (ICD), a subcutaneous implanted device that detects VA through a lead placed in the right ventricle (see Figure 1) and automatically treats these arrhythmias by delivering anti-tachy pacing or ultimately a high power DC shock, emerged in clinical practice in the early 1980s. This therapy does not prevent VA, but it does prevent SCD following VA. The benefit on mortality was convincingly shown in survivors of a previous sustained VA20,21 and has led to guidelines regarding ICD implantation for secondary prevention that are relatively straightforward and effective (see Table 1). The extremely high mortality of SCD at its very first presentation implied that ICD therapy should not be restricted to survivors of previous VA.11 The first and second Multicenter Automatic Defibrillator Implantation
Trial (MADIT I & II), the Multicenter Unsustained Tachycardia Trial (MUSTT) and the Sudden Cardiac Death in Heart Failure Trial (SCD HeFT) were landmark RCTs all showing significantly improved survival in patients who had suffered an MI and were considered to be at high risk of SCD based on a limited number of parameters – of which depressed left ventricular ejection fraction (LVEF) was the most important (see Table 2).19,22–24 The forthcoming guidelines regarding the primary prevention of SCD were somewhat more complex than those for secondary prevention ICD implantation (see Table 1) but still allowed implementation in daily clinical routine.
Radiofrequency Ablation
Radiofrequency ablation for VA is another recently studied approach. The rationale is straightforward: through the creation of lesions on critical points in the scar tissue, the circuit allowing re-entry is interrupted and the risk of VA lowered. The technique has been shown to significantly reduce VT recurrence in patients with previous MI.25 Unfortunately, it has only proven effective in the limited subset of patients with stable VT. Its impact on SCD-related mortality or the possibility to avoid ICD implantation have currently not been investigated. And it seems less likely to help in patients with the more complex arrhythmic substrates which are more prone to cause SCD.
Risk Stratification for Implantation of a Cardioverter-defibrillator in Primary Prevention of Sudden Cardiac Death
Does the Current Risk Stratification Need to Be Improved?
Although backed by sound scientific evidence, the current identification of candidates for an ICD for the primary prevention of SCD is far from optimal. A meta-analysis of the above-mentioned large RCT showed that, with the current risk stratification, only approximately one in four patients (22.9 %, range 17.8–31.4 %) received appropriate and possibly life-saving intervention in the form of an ICD.26 Combined with the important upfront financial cost, the need for frequent follow-up and the inevitable complications of a device featuring intracardiac leads, this makes the cost-effectiveness of ICD therapy a polemical issue among today’s health economics. Gaining one quality-adjusted life year has been estimated to cost around €30,000 and the cost estimates have been repeatedly challenged, making decisions regarding reimbursement difficult in this era of economic uncertainty.27 Moreover, although the risk stratification that is being used does identify a subpopulation in which the incidence of SCD is higher than in the general population, the absolute number is only a minority compared with the number of SCD cases in a broader population without depressed LVEF or even without ischaemic cardiomyopathy.28
The quasi-monopoly of reduced LVEF in the risk stratification for a primary prevention ICD implantation (see Table 2) has the advantage of simplicity, but also has major drawbacks. The rationale behind its use as a predictor of SCD is that reduced LVEF reflects advanced cardiac remodelling leading to an arrhythmic substrate. Nevertheless, LVEF has relatively low sensitivity and specificity for arrhythmia leading to SCD: the majority of SCD patients do not have low ejection fraction and the majority of patients with low LVEF will never experience SCD.29 Another important issue is the strong predictive value of low LVEF for total mortality, as patients with low LVEF are also at high risk of non-sudden cardiac death. SCD and non-sudden cardiac death risks are to be seen as competing:30 a very high risk of dying from heart failure will prevent an ICD implantation from being useful.31,32 Thus, very low LVEF might rather be a marker of particular ICD-resistant mortality.
The Quest for the Telltale of Specific Arrhythmic Risk
The efficacy of ICD implantation can only be improved if it is possible and easy to identify patients at high risk of arrhythmia but at low risk of non-arrhythmic death. This ‘electrophysiologist’s holy grail’ has urged many to look for new, preferentially non-invasive approaches to detect factors that act as a trigger for, or create a substrate for re-entry leading to, VA and subsequent SCD. The most important pathophysiological factors that have been identified are ventricular ectopy, myocardial scar, slowed ventricular conduction, imbalance in autonomic tone and heterogeneity in ventricular repolarisation. Here we briefly present the rationale and – where already known – the value of these factors (in a systematic structure similar to that used in a consensus document issued by the American Heart Association, American College of Cardiology and Heart Rhythm Society33).
Ventricular Ectopy
Ambulatory ECG (Holter) monitoring can easily detect ventricular premature beats (VPBs) and non-sustained VT (NSVT). The presence of 10 VPBs or more per hour in post-MI patients correlated with higher total mortality: the negative predictive value was over 90 % whereas the positive predictive value ranged from 5–15 %.34 This predictive value became stronger when combined with a reduction of LVEF.33 Holter has proven to be clinically useful to guide ischaemic cardiomyopathy therapy in the subgroup of patients with an LVEF between 35 and 40 %: if NSVT is recorded, an electrophysiological study should be performed. If the latter is positive, ICD implantation has shown to improve survival (see Table 1).
Extent of Myocardial Damage and Scar Formation
More myocardial damage means a higher chance of developing a re-entrant substrate. Lower LVEF is reflecting myocardial damage (its value has been discussed above). Magnetic resonance imaging with delayed contrast enhancement (DE-MRI) is a more direct measurement of the extent of myocardial scar. A large observational study evaluating the correlation between MRI and outcomes measured LVEF and scar size, and showed that, even in patients with near-normal LVEF, significant damage identifies a cohort at high risk of early mortality.34 The extent of scar on DE-MRI has been correlated with ICD interventions for VA in post-MI patients.36 No data on MRI-guided anti-arrhythmic therapy are available.
Slowed Ventricular Conduction
QRS duration can be obtained easily from a standard electrogram. A broadened QRS complex is a marker of high total mortality in ischaemic cardiomyopathy patients with depressed LVEF.37 A broad QRS reflects slow conduction, which has been suggested to be a direct cause of arrhythmogenic dispersion of ventricular recovery,38 but no data warrant the use of QRS duration as a specific marker for SCD.
Signal-averaged ECG (SAECG) is a technique that allows noise reduction and amplification of an ECG signal in order to detect late potentials. These subtle signals at the end of the QRS indicate prolonged activation of small parts of the myocardium, typically in infarct regions, and were shown by some to correlate with re-entry substrate. Abundant clinical data show that abnormal SAECG may identify MI patients at risk of SCD, but no proof of the superiority of SAECG-guided ICD therapy is available.33
Mechanical dispersion of contraction measured by echocardiographic strain imaging is a technique that has been developed only recently. It has been shown to correlate with higher VA risk in post-MI patients.39
Imbalance in Autonomic Tone
The heart rate variability (HRV) is calculated with several different techniques on long-term ambulatory ECG (Holter) registrations and reflects the vagal and sympathetic influences on the heart. The theoretical link between abnormal HRV, autonomic tone and arrhythmogenesis has not been observed in clinical cohorts.
Nevertheless, depressed HRV was clearly related to higher total mortality. Although a limited number of studies have been performed, the same assumptions can currently be made about heart rate turbulence (HRT). This elegant technique measures the slope of return to normal heart rate after the parasympathetic slowing of the heart in answer to a VPB. A higher slope indicates more parasympathetic tone and is believed to correlate with better prognosis.33
Heterogeneity in Ventricular Repolarisation
The QT interval can easily be measured from a surface ECG. In large study populations, a prolonged QT interval correlated with increased total mortality. The benefit of QT interval length in specific SCD risk prediction has only been demonstrated in patients with long QT syndrome.33 The same holds true for QT variability, a parameter defined as the slope of the QT/RR interval relationship. QT dispersion is defined as the maximal difference between QT intervals in the surface ECG. It is unclear whether this parameter represents a spatial heterogeneity in ventricular repolarisation, and studies investigating its relationship with outcomes have been contradictory.33
T-wave alternans (TWA) is beat-to-beat alternating of the amplitude of T-wave, measurable on a microvolt level by spectral analysis techniques. Protocols of both a positive and negative clinical TWA test are displayed in Figure 2. The phenomenon was related to increased clinical VA risk for the first time in 1994.40 Since then, extensive clinical and experimental research has been conducted to understand this phenomenon. Experimental research has shown that repolarisation alternans at cellular level can amplify electrical heterogeneities between cells that can directly cause arrhythmia.41 A meta-analysis of 19 studies investigating the clinical value of TWA showed strong predictive value (relative risk 2.42, 95 % confidence interval 1.3–4.5) for arrhythmic events in ischaemic cardiomyopathy patients.42 Unfortunately, not all recent studies were equally positive43 and no definitive proof that TWA is predicting a benefit from ICD implantation is currently available.
Conclusion
SCD is an important cause of mortality because of its very high incidence and its unexpected way of taking lives with previously little morbidity. Ventricular tachyarrhythmia accounts for the majority of SCD events and can be caused by various heart diseases, the most frequent being ischaemic cardiomyopathy. The most effective ways to reduce SCD risk in ischaemic cardiomyopathy are the optimal prevention of recurrent coronary ischaemia and the use of an ICD in high-risk patients. The cost-effectiveness of the latter, which is an expensive therapy, relies primordially on the identification of patients at high risk of SCD and low risk of non-arrhythmic mortality. The current risk stratification rests mainly on depressed LVEF and needs to be improved. There has been an extensive quest for alternative risk-stratifiers. Despite providing new insights in the mechanisms leading to SCD, this quest has, up to now, failed to provide a clinically useful tool to identify non-invasively patients at high risk of SCD.