Fibrosis is the most common cause of functional and anatomic block leading to slow conduction and reentry and occurs in both infarct-related and nonischemic cardiomyopathies. The pattern and extent of fibrosis may range from patchy areas in some forms of heart failure to extensive fibrotic scars in MI. In infarct-related cardiomyopathies, fibrosis replaces necrotic regions and may extend into non-infarcted bordering regions. The number of cell-to-cell contacts is reduced with side-to-side connections selectively affected [16, 17]. In healing infarct border zones, increased interstitial collagen disrupts the long transversely oriented gap junctions enhancing anisotropy by increasing the axial resistivity transverse to the long fiber axis [17]. Gap junctions at the ends of myocytes undergo fewer changes. Thus the normal property of uniform anisotropy is converted to non-uniform anisotropy [16, 17].

When there is a large scar such as in hearts with myocardial infarcts, “peninsulas” and bridges of intact myocytes can project into or completely across areas of fibrosis termed “channels”. Myocytes in these bridges sometimes link with normal myocytes on the outer borders of the infarct [18]. Slow conduction can result from circuitous, nonuniform propagation through these myocardial bundles trapped in scar tissue, so-called “zig-zag conduction” [19]. The effects of fibrosis on conduction is often detectable by low amplitude, long duration, fractionated electrograms with delayed activation, the basis for substrate mapping for ablation of VT [20]. The number of fractionated electrograms seen during endocardial substrate mapping has been shown to correlate with the likelihood of inducible monomorphic ventricular tachycardia at the time of electrophysiology study [21]. Fractionated electrograms are also more likely to be seen during endocardial mapping in patients with infarct-related and nonischemic cardiomyopathy who present with VT when compared to patients with no VT, nonsustained ventricular tachycardia (NSVT), or cardiac arrest [22].

Acute ischemia may result in abnormal automaticity or focal discharges from calcium overload and triggered activity in the form of delayed or early after-depolarizations [23, 24]. Acute ischemia activates the ATP-sensitive potassium (K_ATP) channels, causing an increase in extracellular potassium in the cardiac muscle. Minor increases in extracellular potassium depolarize the myocyte’s resting membrane potential, which can increase tissue excitability in early phases of ischemia [25]. Further hyperkalemia causes greater resting depolarization, decreased conduction velocity and tissue excitability, and shortening of the action potential duration, but prolongs the effective refractory period due to post-repolarization refractoriness [25]. These changes provide a substrate for injury current to flow between ischemic and nonischemic cells located in border zones, which can promote focal abnormal automaticity and initiate VT [26, 27].

Prolongation of the action potential has been repeatedly demonstrated in isolated myocytes [28] and intact ventricular preparations [29] from failing hearts independent of the cause. Downregulation of the transient outward current (I_to) is the most consistent ionic current change observed in heart failure [30–32] but downregulation of other potassium channel currents including I_K1, I_Kr, and I_Ks have also been reported in several models of heart failure [30, 33–37]. Another mechanism for altered repolarization is a late sodium current, found in high density in myocytes of animals with chronic heart failure [38].

Alterations in Ca²⁺ handling are also involved in action potential prolongation and a predisposition to ventricular arrhythmias in heart failure. Abnormal intracellular calcium handling is a prominent feature of myocytes from failing hearts [39, 40]. Variable changes in the Na⁺/Ca⁺ exchanger have been reported with the majority of studies reporting an increase in function [41]. Downregulation of SERCA is also a consistent finding across studies [42–44] as is an increased open probability of the ryanodine receptor (the calcium release channel of the SR) [45, 46]. The basis for delayed afterdepolarizations is spontaneous calcium release from the SR which is facilitated by the presence of beta-adrenergic stimulation and the increased open probability of the ryanodine receptor [47]. Several studies have shown that adrenergic stimulation is necessary for the eliciting spontaneous calcium release from the SR and delayed afterdepolarizations [47, 48]. The calcium released is removed from the cell by the electrogenic Na⁺/Ca²⁺ exchanger whose function is upregulated in heart failure. This leads to a transient inward Na⁺ current that causes the delayed after-depolarization [49]. This is facilitated by the higher resting membrane potential created by a decrease in outward potassium current.

As previously mentioned, Purkinje myocytes are commonly the sources of afterdepolarizations associated with triggered arrhythmias and these cells undergo substantial remodeling of both K⁺ and Ca²⁺ currents, prolonging the action potential and leading to labile repolarization in these cells [50]. Prolongation of the action potential duration associated with down regulation of repolarizing currents and an increase in depolarizing currents lead to spatially and temporally labile repolarization that can predispose to after-depolarization-mediated triggered activity and functional reentry. The existence of after-depolarizations and reentry are not mutually exclusive. After-depolarizations that are of sufficient amplitude to illicit an action potential may serve to initiate an arrhythmia while reentry may sustain such arrhythmias [2].

Mechanical abnormalities in patients with heart failure such as increased wall stress and cavitary dilatation can alter the electrophysiologic properties of myocardial tissue, termed mechanoelectric feedback [51, 52]. In the normal heart, acute myocardial stretch exaggerates the normal rate-dependent shortening of refractoriness but does not influence transverse or longitudinal conduction velocity [53]. In contrast, an animal model found that the development of dilated cardiomyopathy resulted in significant prolongation of refractoriness and repolarization that increased further by volume augmentation and was not reversed by pharmacologic load reduction [54]. Stretch has also been shown to cause a reduction of connexin 43 as well as its redistribution to the lateral sarcolemmal membranes resulting in slowing of conduction in the transverse direction, predisposing these cells to reentry [55]. In human studies, direct correlations between left ventricular end-diastolic volume and the prevalence of ventricular arrhythmia have been seen [56]. It has also been reported that the typical QT prolongation seen in patients with advanced heart failure can be significantly reversed by mechanically unloading the left ventricle with a left ventricular assist device [57].

Activation of the adrenergic and renin-angiotensin-aldosterone system (RAAS) is known to be important in the progression of heart failure. Inhibition of these pathways in randomized clinical trials has been associated with reduction in overall mortality as well as SCD mortality [58–64]. Heart failure is associated with enhancement in sympathetic activity and a reduction in parasympathetic activity [65, 66]. This neurohormonal activation likely influences the substrate and triggers for ventricular arrhythmias in heart failure. In ventricular biopsies and autopsy specimens, an increased density and exaggerated spatial heterogeneity of sympathetic nerves have been associated with a previous history of ventricular arrhythmias in cardiomyopathies [67]. In the normal ventricle, sympathetic stimulation shortens the action potential duration and reduces the dispersion of repolarization, both associated with a decrease in arrhythmia potential [68]. However, in heart failure, sympathetic stimulation is a potent stimulus for the generation of arrhythmias perhaps by enhancing the dispersion of repolarization. Beta-adrenergic stimulation is known to have significant effects on calcium handling as noted above [2]. Elevated levels of sympathetic stimulation enhance the rate of spontaneous diastolic depolarization in latent ventricular pacemakers more than the sinus node which can lead to the development of automatic arrhythmias. Increases in intrinsic heart rate in the setting of sympathetic stimulation can also increase the likelihood of occurrence of triggered arrhythmias [6]. Furthermore, catecholamines can alter conduction and refractoriness, which may promote functional block and facilitate reentry.

RAAS signaling has numerous effects on the cardiovascular system that increase heart failure patients’ propensity to develop ventricular arrhythmias. The two major effectors of the RAAS axis, angiotensin II and aldosterone, are upregulated in heart failure and have prominent effects on properties of myocardial cells [2]. Angiotensin II can indirectly promote arrhythmia formation via potassium and magnesium loss in the urine resulting in prolongation of repolarization. It can also potentiate the effects of the sympathetic nervous system. Vasoconstriction caused by RAAS activation alters loading conditions, affecting wall stress and mechanical factors. Angiotensin II and aldosterone also promote generation of fibrosis in the myocardium by myofibroblasts [69]. Angiotensin II has also been shown to influence gap junction coupling in cardiac cells [70] possibly related to abnormal phosphorylation of connexin 43 [71] leading to nonuniform anisotropy and promoting reentry. Angiotensin II can also inhibit a number of K⁺ currents including the transient outward K⁺ current (I_to) and delayed rectifier K⁺ currents (I_Kr) in the myocardium [72–75].

The QRS prolongation that often accompanies progression of myocardial disease can predispose patients to a unique arrhythmia called bundle branch reentrant ventricular tachycardia (BBRVT) [76, 77]. The unique properties of the HPS allow for rapid conduction and long refractory periods ordinarily preventing sustained reentry. However, conditions that result in prolongation of conduction in the HPS such as cardiomyopathy can facilitate sustained reentry [78]. Three circuits have been described in BBRVT. The first utilizes anterograde conduction over the right bundle branch and retrograde conduction over the left bundle branch with the His bundle adjacent to but separate from the circuit. The second form utilizes the left bundle as its anterograde limb and the right bundle for retrograde conduction. In more rare circumstances, intrafasicular reentry can also occur with the separate fascicles of the left bundle branch being used for the reentrant circuit [79]. Most patients that develop BBRVT tend to have advanced structural heart disease and QRS prolongation in the form of bundle branch block or nonspecific conduction delay. Although most have a baseline bundle branch block, the QRS prolongation in these patients is generally due to delayed conduction in one or both of the bundle branches rather than true block as complete block would not allow propagation of wavefronts retrogradely, which is required to facilitate reentry. This delay tends to be a result of the severity of the underlying heart disease with cardiac enlargement and heart failure being common.

Sustained monomorphic VT occurs most frequently in the setting of a healed MI and may occur in the subacute phase or long after the acute ischemic injury [80]. The extent of myocardial necrosis and subsequent fibrosis as well as the degree of left ventricular dysfunction are important determinants of arrhythmia risk following an MI. The overall incidence of sustained VT following an MI was classically reported to be between 3 and 5 percent but has been estimated to have declined to 1 % in recent years due to advances in revascularization resulting in smaller infarct scars [80]. Although most VTs in infarct-related cardiomyopathy are sustained through a reentrant mechanism, focal activation by abnormal automaticity or triggered activity from the ischemic border zone may serve to initiate the tachycardia, especially in the setting of ischemia. Reentry is the mechanism underlying VT associated with healed or healing MIs in more than 95 % of cases [80].

The 12-lead ECG during VT can provide important information regarding the presence or absence of prior myocardial infarction and localization of the VT origin. In the setting of prior MI, VT tends to arise subendocardially in the region of myocardial scarring. This results in qR or QR patterns of the QRS complexes overlying areas of infarction reflecting terminal activation of the subepicardium. A QS pattern can also occur in the setting of a prior full thickness MI. However, in this setting, the QS pattern represents a cavity potential and is of no value in localizing the VT origin. In the absence of a prior MI, such as in nonischemic cardiomyopathy, a QS pattern of the QRS complex may suggest an epicardial origin of the VT (Fig. 15.2).

Although several possible anatomic abnormalities allowing for reentry in the setting of MI have been postulated, it is now accepted that reentry in the presence of MI utilizes surviving bundles of myocardium within the scar, separated by connective tissue, fibrosis, and disordered intercellular coupling [81] (Fig. 15.3). Evidence for this hypothesis includes the fact that fixed areas of slow conduction can be mapped during sinus rhythm in patients with VT and ablation at these sites can effectively eliminate VT [80]. The triggers for initiation of VT in this setting are increased in situations such as acute heart failure decompensation, which lead to surges in autonomic tone, electrolyte imbalances, and acute ischemia [24, 80].

The pathophysiologic basis for ventricular arrhythmias in dilated cardiomyopathies is not well understood although several possible mechanisms have been postulated [80]. Extensive myocardial abnormalities, fibrosis, and the loss of cell-to-cell coupling in patients with dilated cardiomyopathy may provide the substrate for reentry. Increased levels of fibrosis in the endocardium as well as the epicardium lead to increase in tissue anisotropy, particularly from epicardium to endocardium [82]. Most endocardial scars in dilated cardiomyopathy tend to be adjacent to the valve annulus while epicardial scars in these patients are often greater in extent than on the endocardium [83]. These epicardial scars may facilitate reentry due to development of conduction block in lines parallel to the epicardial fiber orientation between the epicardium and endocardium [82]. Focal mechanisms may also contribute to the genesis of ventricular arrhythmias in these patients. Spontaneous and induced VT in patients with dilated cardiomyopathy have been found to arise in the subendocardium through triggered activity from early or delayed after-depolarizations without evidence of reentry [84]. Patients with dilated cardiomyopathy have been shown to have nonhomogeneous distribution of sympathetic fibers. Regions with dense fibrosis show myocardial denervation similar to what is seen in a scar due to MI while other regions may show an increase in sympathetic innervations [85]. It is postulated that denervation hypersensitivity could lead to a regional increase in sensitivity to catecholamines. Patients with severe heart failure requiring transplantation have been shown to have regional increases in sympathetic nerves around diseased myocardium and blood vessels and this was shown to be more pronounced in patients with a history of VT [67]. Autoantibodies against beta-1 adrenergic receptors have also been detected in up to 50 percent of patients with dilated cardiomyopathy, some subgroups of which can exert sympathomimetic activity [86]. These autoantibodies have been associated with increased rates of ventricular arrhythmias in heart failure patients as well [86, 87]. It is important to note that patients with dilated cardiomyopathy are ten times more likely to develop cardiac arrest associated with polymorphic VT or ventricular fibrillation (VF) than to develop hemodynamically tolerated sustained monomorphic VT and this is likely an underestimate as this experience only includes survivors or cardiac arrest [80].

Multiple variations of hypertrophic cardiomyopathy (HCM) exist with different clinical manifestations depending on the site and extent of hypertrophy. The hemodynamic consequences of myocardial hypertrophy in these patients include left ventricular outflow tract obstruction, mitral regurgitation, diastolic dysfunction, and myocardial ischemia. These patients have an increased likelihood of developing ventricular arrhythmias. Similar to all other etiologies of heart failure, it is thought that myocardial fibrosis, potentially related to ischemic damage, plays an important role in the arrhythmogenic substrate of HCM [88]. Late gadolinium enhancement (LGE) on cardiovascular magnetic resonance (CMR) imaging is a common finding in HCM patients, likely reflecting collagen deposition and fibrosis [89–91]. The extent of LGE in HCM patients has been correlated with SCD risk factors [90] as well as frequency of ventricular premature depolarizations (VPDs) and NSVT on ambulatory monitoring [92]. Although sustained monomorphic VT may occur in the presence of HCM, similar to dilated cardiomyopathies, these patients more commonly have cardiac arrest due to polymorphic VT or VF [80].

Arrhythmogenic right ventricular cardiomyopathy (ARVC) is characterized by fibrous or fibro-fatty replacement of the right ventricular myocardium in the inflow tract, outflow tract, or apex of the right ventricle. The right ventricular myocardial scarring initially produces regional wall motion abnormalities but as the disease progresses, it may involve the free wall and become global with right ventricular dilatation [93]. In some cases the left ventricle may also be involved [94]. These areas of scarring tend to be the focus of ventricular arrhythmias in patients with ARVC who often present with sustained or nonsustained monomorphic VT originating in the right ventricle. VTs originate from areas where scarring develops, most commonly the inflow tract, apex, or outflow tract of the right ventricle. As in the case with dilated cardiomyopathy, the fibrosis and scarring in this setting tends to be more pronounced in the epicardium than the endocardium. For individuals with advanced fibrosis, VT tends to be due to macroreentry in areas of epicardial scar. However, in patients with earlier stages of disease, VT may be due to triggered or automatic activity in the diseased myocardium, often related to adrenergic stimulation. In such instances, VT and SCD are often exercise-induced [95]. Exercise is thought to increase the risk of sudden death in ARVC patients, potentially due to catecholamine effects on triggered or automatic activity or due to the increased stress placed on the right ventricle during exercise [96]. Animal models of ARVC have suggested that exercise increases right ventricular dilatation and worsens manifestations of the disease [97]. In regards to the relationship of catecholamines and VT in patients with ARVC, a higher frequency of induced ventricular arrhythmias in patients with ARVC on isoproterenol infusion has been reported when compared to control patients [98]. Onset of VT has also been shown to be associated with gradual rises in sinus rates and shortening of coupling intervals prior to initiation suggesting an increase in sympathetic tone may play a role in initiation [99].

Ventricular arrhythmias are of greatest concern in heart failure patients because of the risks of developing a fatal arrhythmia leading to SCD. Although heart failure patients represent a high-risk group of patients for SCD, only a minority of patients with heart failure develops fatal ventricular arrhythmias. Knowing this, antiarrhythmic medications and ICDs that prevent or abort SCD will benefit some patients but most high-risk patients will never require these interventions [129]. Interventions that prevent or abort SCD carry risks. Multiple large randomized controlled trials have demonstrated that antiarrhythmic drugs have significant and often fatal proarrhythmic effects [104, 130]. ICDs have risks as well including infectious complications, inappropriate ICD discharges, and device malfunctions [131]. Furthermore, ICDs may have proarrhythmic effects [132]. Further advances in risk stratification of heart failure patients will be an important role of future research.

The role of AADs for ventricular arrhythmias in patients with heart failure is limited. The use of AADs for suppression of ventricular arrhythmias fell out of favor after the CAST trial demonstrated that AADs had significant, often proarrhythmic effects [104]. This was later confirmed in large randomized trials, suggesting AADs might harm more patients than they benefit. Beyond their proarrhythmic effects, most antiarrhythmic drugs also have some negative inotropic activity, which can produce heart failure in patients with ventricular dysfunction [133]. Low cardiac output and the associated renal dysfunction that is common in heart failure can impair elimination of these drugs, increasing the risk of drug toxicity. Some of the AADs, particularly the class I and class III agents exert a strong proarrhythmic effect. Heart failure is a risk factor for the development of torsades de pointes in patients receiving the class III agents ibutilide, sotalol, and dofetilide. Amiodarone is generally considered to be less proarrhythmic than other AADs and is often the preferred drug for treatment of ventricular arrhythmias in heart failure. Amiodarone trials have shown mixed results on mortality and SCD [120, 134–136]. The 2005 ACC/AHA guidelines on the management of chronic heart failure do not recommend amiodarone for the prevention of SCD but concluded that the evidence was in favor of efficacy for its use for reducing the frequency of ICD shocks in patients with recurrent ventricular arrhythmias [137].