Molecular Genetics

Familial disease occurs in approximately 30% to 50% of cases of ARVD, and two patterns of inheritance have been described: autosomal dominant disease with incomplete penetrance (most common) and autosomal recessive disease (rare). Autosomal recessive forms of ARVD include Naxos disease (also known as familial palmoplantar keratosis and “mal de Meleda” disease) and Carvajal syndrome (a disorder of LV cardiomyopathy), both of which are associated with skin and hair abnormalities.

The autosomal dominant nonsyndromic ARVD-1 to ARVD-12 as well as the two rare recessive forms (Naxos disease and Carvajal syndrome) have been mapped to 12 genetic loci, with mutations in eight gene loci identified (Table 29-1). Although some identified gene mutations such as RYR2 may be phenocopies of ARVD, a common theme of desmosomal protein mutations has been emerging. In fact, five of the eight gene loci encode desmosomal proteins (desmoplakin [DSP], plakophilin-2 [PKP2], desmoglein-2 [DSG2], desmocollin-2 [DSC2], junctional plakoglobin [JUP]) and have been found in association with the ARVD phenotype in up to 50% of cases. Among the known genes, mutations in PKP2 appear to be the most common causes of ARVD, accounting for approximately 20% of cases. Mutations in DSG2 and DSP each account for approximately 10% to 15% of cases. Therefore, ARVD is currently considered, at least in a subset, a disease of the cardiac desmosome.^2,3 The majority of these mutations are insertion/deletion or nonsense mutations, which are expected to cause premature termination of the encoded proteins.¹

TABLE 29-1 Known Genetic Loci for Arrhythmogenic Right Ventricular Cardiomyopathy-Dysplasia

Mutations in genes encoding nondesmosomal proteins also can potentially cause ARVD. Alterations of the 5′ and 3′ regulatory elements of TGFB3, encoding transforming growth factor-β₃, have each been reported. More recently, a mutation in TMEM43, encoding a transmembrane protein with ties to an adipogenic transcription factor, was reported as the cause for ARVD-5, a subtype of ARVD with prominent LV involvement. Mutations in RYR2, encoding the cardiac ryanodine receptor (the major calcium release channel of the sarcoplasmic reticulum in cardiomyocytes, also implicated in familial catecholaminergic polymorphic VT), result in a form of arrhythmogenic cardiomyopathy (ARVD-2) characterized by exercise-induced polymorphic VT that does not appear to have a reentrant mechanism, occurring in the absence of significant structural abnormalities. Patients do not develop characteristic features of ARVD on the 12-lead electrocardiogram (ECG) or signal-averaged ECG, and global RV function remains unaffected. ARVD-2 shows a closer resemblance to familial catecholaminergic polymorphic VT in both etiology and phenotype; its inclusion under the umbrella term of ARVD remains controversial.^2–4

Pathogenesis

Cardiac desmosomes are specialized, multiprotein complexes in the intercalated disks that anchor intermediate filaments to the cytoplasmic membrane in adjacent cardiomyocytes, thereby forming a three-dimensional (3-D) scaffolding and providing mechanical strength and cohesion of cardiomyocytes in the beating heart (Fig. 29-1). Cardiac desmosomes are also responsible for regulating transcription of genes involved in adipogenesis and apoptosis as well as maintaining proper electrical conductivity through regulation of gap junctions and calcium homeostasis. Cardiac desmosomes are composed of three groups of proteins: the cadherin family, the Armadillo family, and the plakin family. The cadherin family is composed of three desmocollins and three desmogleins, which are primarily responsible for anchoring the structure to the membrane. The Armadillo family is composed of plakoglobin and three plakophilins, which form the core structure and possess signaling capabilities. The plakin family is composed of DSP, envoplakin, periplakin, plectin, and pinin, which are responsible for the attachment of the desmosomes to intermediary filaments.^1,5

FIGURE 29-1 Molecular structure of the cardiac desmosome. The desmosomal cadherins desmocollin and desmoglein form classic homotypic and heterotypic interactions between cells. These membrane-bound proteins are linked through the Armadillo family members, plakoglobin, and plakophilin to the intermediate filament–binding protein desmoplakin. This complex anchors the desmosome to the cytoskeleton of the cell, and thus indirectly to the sarcomere, nuclear membrane, and the dystrophin-associated glycoproteins.

(From Ellinor PT, MacRae CA, Thierfelder L: Arrhythmogenic right ventricular cardiomyopathy, Heart Fail Clin 6:161-177, 2010.)

The mechanisms by which the affected desmosomes cause myocyte apoptosis, fibrogenesis, adipogenesis, and slow ventricular conduction, thus leading to impaired RV function and increased arrhythmogenicity, remain to be determined. Mutations in desmosomal genes alter the number or integrity of desmosomes and thereby lead to impaired mechanical coupling and failure of cell-to-cell adhesion structures during exposure to physical strain, resulting in cardiomyocyte detachment and degeneration, with subsequent inflammation and replacement by fibrofatty tissue. Fibrofatty replacement is a nonspecific repair process also observed in the muscular dystrophies. The architecture of thin surviving myocardial bundles within the fibrofatty tissue creates lengthened conduction pathways, conduction slowing at pivotal points, and conduction block. All factors contribute to activation delay and can create an electrophysiological (EP) substrate for reentry and VT. Ventricular dysfunction can ensue in the later stages owing to progressive myocyte detachment and death.⁶

Additionally, mutations in desmosomal proteins can impair expression of interacting proteins at the intercalated disk (e.g., gap junction or ion channel proteins), giving rise to impairment of intercellular conductance and promoting ventricular arrhythmogenesis, even in the absence of fibrofatty tissue replacement. Impaired desmosomal structure and function also can affect other cell-to-cell contact structures in the myocardium. In particular, connexin-43 remodeling in the gap junctions, which contributes to delayed conduction and ventricular arrhythmogenesis, is often observed in ARVD patients, and these potential electrical conduction abnormalities may favor arrhythmic events that characterize the disease and predispose patients to high risk of SCD.^2–4

Pathology

Regardless of the mechanism, the patchy replacement of the RV myocardium by fibrofatty tissue provides a substrate for reentrant ventricular arrhythmias. The most striking morphological feature of the disease is the diffuse or segmental loss of RV myocytes, with replacement by fibrofatty tissue and thinning of the RV wall. Patchy inflammatory infiltrates can be present in areas of myocardial damage. Fibrofatty replacement usually begins in the subepicardium or midmural layers and progresses to the subendocardium. Only the endocardium and myocardium of the trabeculae may be spared. The sites of involvement can be localized and in early disease are often confined to the so-called “triangle of dysplasia”—namely, the RV outflow tract (RVOT), RV apex, and inferolateral wall near the tricuspid valve. RV aneurysms (Fig. 29-2), and segmental RV hypokinesia are typical. Diffuse myocardial involvement leads to global RV dilation. However, the fibrofatty pattern of ARVD is limited not only to the RV; the disease also can migrate to the LV free wall (commonly observed in advanced disease), with a predilection for the posteroseptal and posterolateral areas, with relative sparing of the septum.^2–4

FIGURE 29-2 Right ventricular (RV) aneurysm. This cardiac computed tomography angiogram shows thinning and aneurysmal dilation of the RV anterior wall and outflow tract (arrows) in a patient with arrhythmogenic right ventricular dysplasia-cardiomyopathy and ventricular tachycardia. Ao = aorta; LV = left ventricle; PA = pulmonary artery.

(Courtesy of Dr. Nasar Nallamothu, Prairie Cardiovascular Consultants, Springfield, Ill.)

In ARVD, the regions of abnormal myocardium do not always follow the pattern of dense scar surrounded by a ring of abnormal myocardium, often referred to as the scar border zone. Sometimes, abnormal voltages are found alone, without dense scar defining the regions. This is because ARVD is a different process from scarring caused by myocardial infarction (MI). Infarction causes dense scar surrounded by a border zone because of the ischemic penumbra that surrounds the infarcted territory. In ARVD, however, the process is patchy and can cause inhomogeneous scarring in anatomically disparate areas. Nonetheless, previous data have shown that it is still possible to identify well-demarcated borders around these abnormal regions.

Despite the widespread impression that ARVD is universally a progressive disease process, a recent report found that the extent of the endocardial scar as measured by bipolar voltage mapping in patients with ARVD and VT remained relatively stable, despite progressive RV dilation.⁷

A monomorphic VT in the setting of ARVD is associated with a predominantly perivalvular distribution of endocardial electrogram abnormalities and arrhythmia origin. Reentry in areas of abnormal myocardium is the most likely mechanism of VT in ARVD. Most reentrant circuits cluster around the tricuspid annulus and the RVOT. The critical isthmus contained in these reentry circuits often is a narrow path of tissue with abnormal conduction properties, typically bounded by two approximately parallel conduction barriers that consist of scar areas, the tricuspid annulus, or both. Depolarization of the small mass of tissue in the isthmus is usually not detectable on the surface ECG and constitutes the electrical diastole between QRS complexes.

Epidemiology

ARVD occurs in young adults (80% are younger than 40 years) and is more common in men. The mean age at diagnosis of familial ARVD is approximately 31 years. The disease is almost never diagnosed in infancy and rarely before the age of 10.

The incidence and prevalence of ARVD are uncertain and may vary regionally. The prevalence of the disease in the general population is estimated at 0.02% to 0.1% but is dependent on geographic circumstances. In certain regions of Italy (Padua, Venice) and Greece (island of Naxos), an increased prevalence of 0.4% to 0.8% for ARVD has been reported. Others estimate the prevalence in Europe and North America to be 1 per 5000 individuals.

Clinical Presentation

The clinical presentation varies widely because ARVD includes a spectrum of different conditions rather than a single entity. Different pathological processes can manifest a diversity of symptoms. Additionally, ARVD can have a temporal progression and can present differently according to the time of presentation.

Four clinicopathological stages of ARVD can be considered: the early “concealed” phase, followed by the “overt electrical disorder,” the “phase of RV failure,” and finally, the phase of “biventricular failure.” Although ARVD is a genetically transmitted disease, it is associated with a long asymptomatic lead time and individuals in their teens may not have any characteristics of ARVD clinically or on screening tests. Early ARVD is often asymptomatic (the “concealed” phase), occasionally associated with minor ventricular arrhythmias and subtle structural changes. Nonetheless, these patients are still at risk of SCD, especially during intense physical exertion. With disease progression, the “overt electrical disorder” typically causes symptomatic ventricular arrhythmia and more obvious morphological abnormalities detectable by imaging. Further disease extension results in RV dilation and dysfunction (the “phase of RV failure”), precipitating symptoms and signs of right heart failure. Unless SCD occurs, progressive impairment of cardiac function can result in biventricular heart failure late in the evolution of ARVD, usually within 4 to 8 years after typical development of complete right bundle branch block (RBBB). End-stage disease is often indistinguishable from dilated cardiomyopathy and manifests with congestive heart failure, atrial fibrillation, and an increased incidence of thromboembolic events. Overall, judging the accurate position of the patient on the time scale of the spectrum can be difficult, and some patients may remain stable in the same phase of the disease for several decades.⁸

ARVD is one of the most arrhythmogenic human heart diseases known. Electrical instability is present at the very early stages of ARVD. Ventricular arrhythmias range from isolated ectopy to sustained VT or ventricular fibrillation (VF).⁸ Approximately 50% of patients with ARVD present with symptomatic ventricular arrhythmias, most commonly sustained and nonsustained VT, manifested by palpitations, dizziness, and/or syncope. The frequency of ventricular arrhythmias in ARVD varies with the severity of the disease, ranging from 23% in patients with mild disease to almost 100% in patients with severe disease. Ventricular arrhythmias characteristically occur during exercise; up to 50% to 60% of patients with ARVD show monomorphic VT during exercise testing.

One of the unfortunate features of ARVD is SCD, which is the first clinical manifestation of ARVD in 50% of afflicted individuals. In fact, up to 5% of SCDs in young adults in the United States are attributed to ARVD. In northeast Italy, ARVD was found to be responsible for 22.4% of SCDs in young athletes and in 8.2% of SCDs in nonathletes. In most patients, the mechanism of SCD in ARVD is acceleration of VT, with ultimate degeneration into VF. Generally, RV failure and LV dysfunction are independently associated with cardiovascular mortality.

Supraventricular tachycardias are observed in approximately 25% of patients with ARVD referred for treatment of ventricular arrhythmias; less often, they are the only arrhythmia present. In decreasing order of frequency, supraventricular tachycardias in these patients include atrial fibrillation, atrial tachycardia, and atrial flutter.

Some patients are asymptomatic and ARVD is only suspected by the finding of ventricular ectopy and other abnormalities on routine ECG or other testing because of a positive family history. In a review of 37 families, only 17 of 168 patients with ARVD (10%) were healthy carriers. In one report, 9.6% of those initially unaffected subjects developed structural signs of disease on echocardiography during a mean follow-up of 8.5 years; almost 50% had symptomatic ventricular arrhythmias. Progression from mild to moderate disease occurred in 5% of patients, and progression from moderate to severe disease occurred in 8%.

Initial Evaluation

The noninvasive diagnosis of ARVD can be exceedingly difficult. Several factors, including marked phenotypic variation, incomplete and low (30%) penetrance, and age-related disease development and progression contribute to the complexity of the clinical diagnosis. Particularly problematic is recognition of the early stages of ARVD, when overall RV function may be normal, with local or regional wall-motion abnormalities that are difficult to quantify; nonetheless, the absence of clinical features does not necessarily confer low risk.

Definitive diagnosis of ARVD requires histological confirmation. A myocardial biopsy showing myocyte loss (<45% residual myocytes) with fibrosis and fatty infiltration (>40% fibrous tissue and >3% fat) confirms the diagnosis. However, myocardial biopsy lacks sufficient sensitivity (67% in one report) owing to the patchy nature of the disease. For safety reasons, the biopsy is performed mostly in the interventricular septum, which is histopathologically rarely involved in the disease process. Biopsy sampling performed on the RV free wall may improve the ability to diagnose ARVD. Nevertheless, because of the frequently observed wall thinning with aneurysms or diverticula, free wall sampling is associated with risk of perforation, particularly when performed at random sites. Therefore, the role of endomyocardial biopsy in the diagnosis of ARVD remains controversial.

Electroanatomical voltage mapping, which seems to be an effective technique to detect RV low-voltage regions reflecting fibrofatty myocardial atrophy in patients with ARVD, has been shown to improve the diagnostic accuracy of myocardial biopsy by reducing the sampling errors. Endomyocardial biopsies are obtained from low-voltage areas, preferably from the border zone, in order to minimize the risk of perforation.⁹

Additionally, immunohistochemical analysis of conventional biopsy samples to detect a change in the distribution of desmosomal proteins can also improve the sensitivity and specificity of this diagnostic tool. Reduced immunoreactive signal levels of plakoglobin at intercalated disks were found to be a consistent feature in patients with ARVD but not seen in other forms of myocardial disease.¹⁰

Recognition of the problems in diagnosing ARVD and the fact that there is no “gold standard” or single test that is diagnostic of ARVD led to the formation of a task force in 1994 that proposed major and minor criteria to aid in the diagnosis, based on family history as well as structural, histological, functional, arrhythmic, and ECG abnormalities (Table 29-2).¹¹ The diagnosis of ARVD is based on the presence of two major criteria, one major plus two minor criteria, or four minor criteria. It is important to recognize, however, that these criteria were defined retrospectively from aggregated series of referrals to tertiary care institutions dominated by the overt or severe end of the disease spectrum (i.e., symptomatic index cases and SCD victims) and have never been prospectively validated; therefore, their applicability is limited in situations where the prior probability of ARVD is different from the initial derivation set of patients. Consequently, the original task force criteria are highly specific but lacked sensitivity for early and familial disease. Furthermore, this diagnostic approach does not specify the preferred order of imaging techniques to examine and score RV morphology and function.^5,11

TABLE 29-2 Original and Revised Task Force Diagnostic Criteria for Arrhythmogenic Right Ventricular Cardiomyopathy-Dysplasia

By MRI:

By RV angiography:

Pharmacological Therapy

Beta blockers, sotalol, and amiodarone have been used to reduce the risk of recurrence of ventricular arrhythmias in ARVD patients. The usefulness of antiarrhythmic drugs for primary prevention, however, is less well established.

Previous studies, which determined selection of antiarrhythmic therapy based on suppression of VT in the EP laboratory, found that sotalol was more effective than beta blockers and amiodarone in patients with inducible VT and in those with noninducible VT. Thus, sotalol was considered as a first-line antiarrhythmic agent in ARVD both for primary and secondary prevention. However, this was questioned in a recent study of a cohort of rigorously characterized ARVD population treated with empirically prescribed antiarrhythmic agents. In this study, beta blockers were neither protective nor harmful, sotalol was not effective, and amiodarone, although only received by a relatively small number of patients, showed superior efficacy in preventing sustained VT and implantable cardioverter-defibrillator (ICD) therapies.¹⁹

Therefore, among asymptomatic patients and those with mild disease, beta blockers may still be a reasonable recommendation to reduce the possibility of adrenergically induced arrhythmia. Patients with well-tolerated and non–life-threatening ventricular arrhythmias are at relatively low risk for SCD and can be treated with antiarrhythmic drugs, guided either noninvasively with ambulatory monitoring or with EP testing.²⁰ Recent data argue against the empirical use of sotalol therapy, especially when permitting additional episodes of VT/VF is highly undesirable. It appears that, as in other settings, amiodarone is the most effective empirical antiarrhythmic drug therapy in patients with ARVD.¹⁹

Antiarrhythmic drugs can also be given to reduce the frequency of ventricular arrhythmia in patients with unacceptably frequent ICD therapies.

When the disease has progressed to right or biventricular failure, treatment consists of the current therapy for heart failure, including diuretics, beta blockers, and angiotensin-converting enzyme inhibitors. Although angiotensin-converting enzyme inhibitors are well known for slowing progression in other cardiomyopathies, they have not been proven to be helpful in ARVD. Individuals with reduced RV ejection fraction with dyskinetic portions of the RV may benefit from long-term anticoagulation with warfarin to prevent thrombus formation and subsequent pulmonary embolism. For intractable RV failure, cardiac transplantation can be the only remaining alternative.²⁰

Implantable Cardioverter-Defibrillator

An ICD is the most effective prevention against SCD. For secondary prevention, ICD implantation is recommended in patients with prior cardiac arrest and those with poorly tolerated arrhythmias that are not completely suppressed by antiarrhythmic drug therapy. During a mean follow-up of 39 months, nearly one-half of these patients had an appropriate ICD therapy, 16% had an inappropriate therapy, and 14% had a device-related complication.^5,21

ICD implantation is also recommended for primary prevention in patients who are considered to be at high risk for SCD (see above). It is important to recognize, however, that there is not yet clear consensus on the specific risk factors that identify those patients with ARVD in whom the probability of SCD is sufficiently high to warrant an ICD for primary prevention.¹⁸

Importantly, ICD implantation in ARVD patients is associated with several risks. Areas of the RV myocardium in patients with ARVD may be thin and noncontractile and can be penetrated during placement of the RV leads, with subsequent cardiac tamponade. Also, the fibrofatty nature of the RV can lead to difficulty in attaining a location with acceptable R wave amplitude and pacing threshold. Marginal parameters may prevent adequate sensing of arrhythmias, resulting in improper ICD function or failure. A separate pace-sense lead may be necessary, positioned in the RV septum or outflow tract.²⁰

Catheter Ablation

Catheter ablation of monomorphic VT has a 60% to 90% success rate in patients with ARVD. However, given the chance of long-term recurrence, ablation is not considered curative. Therefore, catheter ablation of VT has been considered mainly in patients with recurrent VT causing frequent ICD shocks and refractory to antiarrhythmic medications and those with VT storm or incessant VT refractory to antiarrhythmic medications. Nonetheless, with the recent advances in mapping and ablation techniques and the improved safety and success of the ablation procedures, as well as the lack of highly effective antiarrhythmic drug therapy, some investigators recommend catheter ablation as a first-line approach in patients presenting with recurrent monomorphic VT.^8,22 ICD implantation is recommended as well, because of the unpredictability of recurrences.

Patients with limited endocardial substrate, those with late VT termination with RF delivery, and certainly those with persistent VT despite aggressive endocardial ablation should be considered for an epicardial approach.²³

Participation in Sports

Because of the association between exercise and the induction of ventricular tachyarrhythmias, a diagnosis of ARVD is generally considered incompatible with competitive sports and moderate- to high-intensity-level recreational activities. Furthermore, any activity, competitive or not, that causes symptoms of palpitations, presyncope, or syncope should be avoided.²⁰

Family Screening

First-degree relatives should be screened clinically with 12-lead ECG, echocardiography, and cardiac MR imaging. Given the low penetrance observed in most families, screening should be extended throughout the kindred to at least one generation beyond the last affected individual. Asymptomatic family members with a normal comprehensive evaluation are less likely to have inherited the gene defect, but should undergo follow-up at regular intervals (every 2 to 3 years) until definitive diagnostic tools are available.⁵ When the causative mutation in the index patient is identified, genetic testing is recommended in family members. A negative genetic test result for the familial mutation would obviate the need for repeated follow-up examinations.

ECG during Sinus Rhythm

Approximately 40% to 50% of patients with ARVD have a normal ECG at presentation. However, by 6 years, almost all patients have one or more of several findings on ECG during normal sinus rhythm (NSR), including epsilon wave, T wave inversion, QRS duration equal to or greater than 110 milliseconds in leads V₁ through V₃, and RBBB. Significant differences exist in most ECG features among ARVD patients according to the degree of RV involvement, and they are more prevalent in diffuse ARVD than in the localized form of the disease.^6,8

The hallmark feature of ARVD, the epsilon wave, is a marker of delayed activation of the RV free wall and outflow tract and is considered a major diagnostic criterion for ARVD. The epsilon wave has the appearance of a distinct wave just beyond the QRS, in the ST segment, in the right precordial leads, particularly in lead V₁, and represents low-amplitude potentials caused by delayed activation of some portion of the RV (Fig. 29-3). Although the epsilon wave has been considered highly specific for ARVD, ECG markers that reflect ventricular conduction delay in ARVD (including the epsilon wave) are frequently observed in subjects with a spontaneous or drug-induced type 1 ECG pattern of Brugada syndrome.²⁴ Additionally, this criterion is of limited sensitivity, observed in only 10% to 37% of ARVD patients when evaluated by standard ECG. The use of highly amplified (20 mV) precordial leads and modified limb leads can increase the detection of epsilon potentials to 75% but is rarely utilized.^6,8

FIGURE 29-3 Surface ECG of sinus rhythm in a patient with arrhythmogenic right ventricular dysplasia-cardiomyopathy. Inset, Magnified view of a single complex in V₁ showing epsilon waves (arrow).

Other markers of delayed activation of the RV include prolongation of the QRS duration to at least 110 milliseconds in leads V₁ through V₃, and a ratio of the sum of the QRS duration in leads (V₁ + V₂ + V₃)/(V₄ + V₅ + V₆) ≥ 1.2 (sensitivity of 35% to 98%). In fact, a QRS duration greater than 110 milliseconds in lead V₁ has a sensitivity of 26% to 75% and a specificity of 100% in patients suspected of having ARVD based on historical features. Additionally, incomplete or complete RBBB can be observed in 18% and 15% of patients, respectively. Epicardial mapping suggests that these patterns are usually caused by parietal block, rather than by disease of the bundle branch itself. In the presence of an RBBB pattern, selective prolongation of the QRS duration in leads V₁ to V₃ compared with lead V₆ (>25 milliseconds, parietal block) is an important hallmark of ARVD (sensitivity of 52% to 64%).^6,8,11,12 Furthermore, T wave inversion in leads V₁ through V₃ in the absence of RBBB is considered a minor diagnostic criterion for ARVD and its prevalence in ARVD has been reported as 55% to 94% in different series. The extent of right precordial T wave inversion relates to the degree of RV involvement. Increased QT dispersion (i.e., inter-lead variability of the QT interval) has been observed in ARVD.

Prolongation of the upstroke of the S wave (≥55 milliseconds from the nadir of the S wave to the isoelectric line) in leads V₁ through V₃ in the absence of RBBB, which presumably represents altered depolarization, was found in the original report to be a very sensitive ECG criterion for ARVD (present in all cases with diffuse ARVD and in 90% in the localized form of the disease). However, the sensitivity of this criterion was less than 60% in several other reports. This measurement has the highest diagnostic usefulness for differentiating the localized form of ARVD from idiopathic RVOT VT, correlates with the degree of RV involvement, and is an independent predictor of VT induction. It is recognized that a prolonged S wave upstroke directly relates to QRS width in the right precordial leads; nonetheless, it is superior to localized QRS prolongation in distinguishing the mild form of ARVD from idiopathic RVOT VT.^6,11,12

Another ECG marker of ARVD described recently is the “terminal activation delay.” This parameter is measured from the nadir of the S wave to the end of all depolarization deflections, thereby including not only the S wave upstroke but also both late fractionated signals and epsilon waves, to the completion of the QRS complex. A prolonged terminal activation delay (≥55 milliseconds) was observed in 71% of ARVD patients and only 4% of controls.^6,25

QRS fragmentation in standard ECG leads, defined as deflections at the beginning of the QRS complex, on top of the R wave or in the nadir of the S wave, has a sensitivity of 85% for the diagnosis of ARVD.²⁶

The SAECG is a highly amplified and signal-processed ECG that can detect microvolt-level electrical potentials in the terminal QRS complex, known as late potentials. Late potentials, which reflect the slow conduction in the ventricular myocardium and electrical potentials that extend beyond the activation time of normal myocardium, typically arise from scarred myocardium, an anatomical substrate potentially responsible for reentrant ventricular arrhythmias. The SAECG and its three components (filtered QRS duration, low-amplitude signal duration, and root-mean-square voltage of the last 40 milliseconds of the QRS) are highly associated with the diagnosis of ARVD. Late potentials and fragmented electrical activity can be detected by signal-averaged ECG in 50% to 80% of patients with ARVD. Although there are no guidelines for abnormal SAECG in patients with ARVD, it is common practice to categorize the SAECG as abnormal if two or more of these parameters are abnormal. In a recent report, the sensitivity of using SAECG for diagnosis of ARVD was increased from 47% using two of the three criteria to 69% by using any one of the three criteria, while maintaining a high specificity of 90% to 95%. Abnormal SAECG was strongly associated with dilated RV volumes and decreased RV ejection fraction detected by cardiac MR imaging. SAECG abnormalities did not vary with clinical presentation or reliably predict spontaneous or inducible VT, and had limited correlation with ECG findings.^11,24,27 Additionally, SAECG abnormalities appear to correlate with the presence of low-voltage areas selectively in the RVOT as observed during electroanatomical voltage mapping, whereas surface ECG abnormalities are associated with a more diffuse RV involvement.²⁸ The usefulness of SAECG for predicting inducible VT in ARVD remains uncertain.²⁴

ECG during Ventricular Tachycardia

Ventricular ectopy in ARVD usually arises from the RV and therefore has an LBBB pattern similar to that seen in idiopathic RVOT VT. The ECG morphology of the VT is predominantly LBBB, but RBBB can be observed and does not exclude an RV origin (Fig. 29-4). The QRS axis during VT is typically between –90 and +110 degrees; an extreme rightward direction of the QRS axis is uncommon.²⁹ No correlation has been observed between any specific ECG morphology and a particular activation pattern, although all RBBB morphology VTs exhibit a peritricuspid circuit.

FIGURE 29-4 Surface lead ECGs of ventricular tachycardia (VT) in a patient with arrhythmogenic right ventricular dysplasia-cardiomyopathy. Both have left bundle branch block–type morphology but with very different frontal axes.

Additionally, patients with ARVD frequently demonstrate the presence of more than one VT morphology, either spontaneously (64%) or during an EP study (88%).⁶

Tachycardia Features

Tachycardia is inducible with programmed electrical stimulation in the majority of ARVD patients with clinical VT. Isoproterenol infusion facilitates VT induction in many patients.

Reentry in areas of abnormal myocardium is the most likely mechanism of VT in ARVD. Most reentrant circuits cluster around the tricuspid annulus and the RVOT. Multiple morphologies of inducible VT are common. A variety of different reentry circuit sites can be identified with entrainment mapping. A single region can give rise to multiple VT morphologies. Previous studies showed mean numbers of different VT morphologies per patient ranging from 1.8 to 3.8.⁶

Overdrive pacing during VT at cycle lengths (CLs) 20 to 50 milliseconds shorter than the tachycardia CL typically results in entrainment of the tachycardia. The presence of entrainment should be confirmed by the demonstration of fixed fusion of the paced QRS at any single pacing CL, progressive fusion as the pacing CL decreases (i.e., the surface ECG progressively looks more like the VT QRS than a purely paced QRS), and resumption of the VT following pacing with a nonfused VT QRS at a return cycle equal to the pacing CL. Once the presence of entrainment is confirmed, three parameters should be evaluated—the presence of manifest or concealed fusion, the post-pacing interval (PPI), and the local electrogram-to-QRS interval during VT versus the stimulus-to-QRS (S-QRS) interval during entrainment. The 12-lead ECGs during pacing and during VT are compared to evaluate whether manifest or concealed fusion is present. The PPI is measured from the stimulus artifact to the onset of the local electrogram of the first VT (unpaced) beat in the pacing lead recording. Finally, the local electrogram-to-QRS interval during VT and the S-QRS interval during entrainment are measured on the distal bipolar pair recording of the mapping catheter or with the use of the proximal bipolar pair when the presence of artifacts in the distal pole recording prevents reliable measurement.

Exclusion of Other Arrhythmia Mechanisms

In clinical practice, the disease that most frequently mimics ARVD is idiopathic RVOT VT. Although RVOT VT is associated with a benign prognosis with no familial basis, it can be extremely difficult to distinguish from the concealed phase of ARVD, in which typical ECG and imaging abnormalities are absent.

The resting surface ECG in NSR is typically normal in patients with idiopathic VT, with no epsilon wave, QRS widening or fragmentation, or other markers of delayed activation of the RV typically observed in ARVD patients. However, the ECG is also normal in 40% to 50% of patients with ARVD at presentation, but in almost no patient after 6 years.³⁰

Both idiopathic RVOT VT and VT caused by ARVD generally have LBBB morphology. However, the presence of an LBBB morphology VT with a superior axis is very unlikely in idiopathic RVOT VT. Additionally, VTs in ARVD patients generally have lower QRS amplitude and more fragmentation, but these qualitative differences are not always present.

Patients with idiopathic VT have a normal SAECG when in NSR and normal imaging studies (echocardiography, MR imaging, contrast ventriculography) of RV size and function. Other clues suggesting the presence of an RV cardiomyopathy include the presence of RV dilation, multiple VT morphologies, and a well-defined anatomical substrate.

The response during EP testing is also helpful in distinguishing idiopathic RVOT VT from ARVD. The repetitive initiation of VT by programmed stimulation suggests a reentrant mechanism. In one report, programmed electrical stimulation induced VT in all but 1 of 15 patients with ARVD compared with only 2 patients with idiopathic RVOT VT (93% versus 3%). Additionally, fractionated diastolic electrograms recorded during VT or NSR at the VT site of origin or other RV sites were present in all but one patient with ARVD, but were not seen in any patient with idiopathic RVOT VT. Furthermore, the induction of VTs with different QRS morphologies was seen only in ARVD (73% versus 0%). Although isoproterenol infusion can induce VT in patients with idiopathic RVOT VT, it has a similar effect in ARVD and therefore does not help distinguish between these disorders. Reentry was the mechanism of tachycardia in 80% of the ARVD group, whereas 97% of RVOT VT had features of triggered activity.

Electroanatomical voltage mapping appears to be an effective way of distinguishing early or concealed ARVD from idiopathic VT by detecting RV electroanatomical scars that correlate with the histopathological features pathognomonic of the disease.³¹

Activation Mapping

Most reentrant circuits cluster around the inferolateral tricuspid annulus, the RVOT, and the RV apex, which are the typical areas of fibrous and fatty infiltration in ARVD. Hence, these areas are targeted first by activation mapping. Because the VT circuit is composed of zones of abnormal conduction, characterized by low-amplitude abnormal electrograms, with identifiable exit regions to the surrounding myocardium, endocardial activation mapping during VT typically detects abnormal fragmented or split low-voltage electrograms, with diastolic potentials (Fig. 29-5).⁸

FIGURE 29-5 ECG and intracardiac recordings during normal sinus rhythm (NSR; left) and ventricular tachycardia (VT; right) in a patient with arrhythmogenic right ventricular dysplasia-cardiomyopathy. During NSR, a late potential is present (arrow); the same potential occurs during mid-diastole during VT. RVOT = right ventricular outflow tract.

During activation mapping, particular sites of interest include: (1) sites with abnormal local bipolar electrogram (amplitude, ≤0.5 mV; duration, ≥60 milliseconds); (2) sites at which the local electrogram precedes the QRS complex by at least 50 milliseconds (activation times are taken from the onset of the bipolar electrogram); and (3) sites with the earliest local activation closest to mid-diastole, isolated mid-diastolic potentials, or continuous activity. Once identified, these sites are targeted by entrainment mapping (see later) to establish their relationship to the tachycardia circuit.

Detailed mapping will usually reveal more than one site of presystolic activity. It is therefore essential to demonstrate that the presystolic site recorded is in fact the earliest site. A normal presystolic bipolar electrogram (amplitude, >3 mV; duration, <70 milliseconds) should prompt further search for earlier activity. If, after very detailed mapping, the earliest recorded site is not at least 50 milliseconds presystolic, this suggests that either the map is inadequate (most common) or the VT arises deeper than the subendocardium, in the midmyocardium, or even incorporating the subepicardium. VTs in patients with ARVD exhibiting a focal activation pattern have also been described; however, an epicardial reentrant circuit with a defined endocardial exit may explain the focal activation pattern of the RV endocardium.^8,29

An isolated mid-diastolic potential is defined as a low-amplitude, high-frequency diastolic potential separated from the preceding and subsequent ventricular electrograms by an isoelectric segment. It is likely that isolated mid-diastolic potentials that cannot be dissociated from the VT are generated in segments of the zone of slow conduction or common pathway (isthmus), which are integral components of the reentry circuit. It is important to recognize that not all presystolic activity recorded during VT is related to the mechanism; it could represent delayed activity not related to the tachycardia mechanism, dead-end pathways, or even an artifact. Therefore, it is important to confirm that presystolic activity is related to the reentrant circuit. Demonstration of a fixed relationship of the diastolic electrograms to the subsequent VT QRS, despite spontaneous or induced oscillations of the tachycardia CL, helps exclude the possibility that those electrograms reflect late activation from unrelated dead-end pathways.⁸

Additionally, 3-D electroanatomical mapping can help in the precise description of VT reentrant circuits, sequence of ventricular activation during the VT and rapid visualization of the activation wavefront, and identification of slow-conducting pathways and appropriate sites for entrainment mapping. Local activation times are assigned according to the onset of the bipolar electrogram registered at the tip of the mapping catheter and are color-coded. These systems also help in navigation of the ablation catheter, planning of ablation lines, and maintaining a log of sites of interest (e.g., sites with favorable entrainment or pace mapping findings), which can then be revisited with precision. Additionally, voltage (scar) mapping is a helpful feature of some of the electroanatomical mapping systems (see later).⁸

Entrainment Mapping

Entrainment mapping is used to characterize reentry circuits in ARVD, identify critical isthmuses, and guide ablation. Pacing is performed during VT at a CL 10 to 30 milliseconds shorter than the VT CL, at RV sites identified during activation mapping, demonstrating continuous activity or isolated mid-diastolic potentials. For stable VTs, the following three criteria are used to define the critical isthmus and predict the success of RF application in termination of the VT: (1) entrainment with concealed fusion; (2) PPI equal to the tachycardia CL (within 20 to 30 milliseconds); (3) and S-QRS interval equal to the local electrogram-to-QRS interval (±20 milliseconds). Other predictors of successful ablation sites include S-QRS interval–to–VT CL ratio less than 0.7, long S-QRS interval, and alteration of the tachycardia CL and/or termination by subthreshold or nonpropagated stimuli. However, because of the diffuse nature of the disease in ARVD, isthmuses may sometimes be wide and good entrainment maps can be obtained from a rather large area. In this case, linear lesions across the diastolic pathway may be necessary.⁸

Pace Mapping

Pace mapping can be used to complement activation and entrainment mapping findings and to confirm ablation target sites, although it may not be always necessary, especially when several criteria localizing the VT isthmus have been identified. At isthmus sites defined during VT, pace mapping during NSR is attempted after VT termination, with the mapping catheter being kept at the same location. Pace mapping is preferably performed with unipolar stimuli (10 mA, 2 milliseconds) from the distal electrode of the mapping catheter (cathode) and an electrode in the inferior vena cava (anode).

The resulting 12-lead ECG morphology is compared with that of the tachycardia. ECG recordings should be reviewed at the same gain and filter settings. The greater the degree of concordance between the morphology during pacing and tachycardia, the closer is the catheter to the site of origin of the tachycardia.

At best, a pace map that matches the VT would only identify the exit site to the normal myocardium, and can be distant from the critical sites of the circuit required for ablation. Thus, pace mapping remains only a corroborative method of localizing VT. It can be used to focus initial mapping to regions likely to contain the reentrant circuit exit or abnormal conduction but is not sufficiently specific or sensitive to be the sole guide for ablation.

Pace mapping can also be used in conjunction with substrate mapping when other mapping techniques are not feasible, so that it can provide information on where ablation can be directed. The VT circuit’s exit site is approximated by the site of pace mapping that generates QRS complexes similar to those of VT. From that point, evaluation of the S-QRS interval (measured to the onset of the earliest QRS on the 12-lead ECG) can help trace the course of the VT isthmus. Sites along the VT isthmus are associated with good pace maps but with progressively longer S-QRS intervals.⁸

Electroanatomical Substrate Mapping

Electroanatomical substrate mapping provides a helpful tool in reconstructing the VT circuit in patients with ARVD and hemodynamically stable VTs, in conjunction with activation and entrainment mapping techniques.²⁹ Additionally, in a subset of ARVD patients, conventional mapping during VT can be limited by changing multiple morphologies, nonsustainability, or hemodynamic instability. Substrate-based voltage mapping during sinus rhythm can be used in these cases to identify regions of scar and abnormal myocardium and guide, in conjunction with pace mapping, ablation of linear lesions to connect or encircle the abnormal regions.²²

The dysplastic regions can be identified, quantified, and differentiated from healthy myocardium by the presence of electrograms with low amplitudes and longer durations, reflecting replaced myocardial tissue. Voltage mapping is performed during sinus rhythm, atrial pacing, or ventricular pacing. The peak-to-peak signal amplitude of the bipolar electrogram is measured automatically. Endocardial regions with a bipolar electrogram amplitude greater than 1.5 mV are defined as normal, and dense scar areas are defined as those with an amplitude less than 0.5 mV. Abnormal myocardium is defined as an area of contiguous recordings with a bipolar electrogram amplitude between 0.5 and 1.5 mV.

As noted, in ARVD, the normal myocardium is replaced by fatty and then fibrous tissue, causing thinning and scarring, predominantly in the RV. Studies have found that regions of abnormal voltage suggestive of scar were observed in all ARVD patients. The endocardial bipolar voltage abnormalities tend to be perivalvular and, although affecting predominantly the RV free wall, involve some aspect of the septum in most (76%) patients. This pattern is consistent with pathological studies demonstrating a predilection for scar to occur in these particular areas in ARVD. Whether an extensive perivalvular fibrotic process is specific for patients who present with sustained monomorphic VT in the setting of ARVD is yet to be determined.^22,29 In contrast to the post-MI pattern of dense scar surrounded by a border zone, the regions of abnormal myocardium in ARVD are commonly patchy and inhomogeneous, occurring in anatomically disparate areas. As a consequence, low-voltage identification in ARVD is less useful as a marker for ablation because it is very widespread. Nonetheless, evaluating the existence of different voltage areas within the 0.5-mV scar (i.e., the voltage limits are reset to 0.3 to 0.5 mV as the upper limit and to 0.1 mV as the lower, representing dense scar) can potentially help visualize channels of slow conduction within scar tissue.⁸ The critical isthmus of each of the reentrant circuit typically is bounded by an area of low-voltage scar on one side and an anatomical barrier (tricuspid annulus) on the other side, or by two parallel lines of double potentials.²⁹

A recent report found that identifying ventricular electrograms with an isolated delayed component during sinus rhythm (defined as a distinct electrogram after the QRS separated [by ≥40 milliseconds] by an isoelectric interval or very low-amplitude signal of <0.1 mV) can be used to define the substrate for the VT circuit and, when combined with other mapping maneuvers, can help guide catheter ablation in ARVD patients with noninducible or unmappable VTs.³²

Importantly, the disease in ARVD has been reported to begin epicardially and progress to the endocardium; therefore, endocardial fibrosis, as reflected by the voltage map findings, can be a manifestation of more extensive disease involvement and can be anticipated to be the more common substrate for uniform sustained VT (Fig. 29-6).²³

FIGURE 29-6 Right and left anterior oblique electroanatomical endocardial (top) and epicardial (bottom) voltage maps in a patient with ARVD. Red areas denote low voltage/scar. On the endocardial map, only a small area shows abnormal voltage (near lateral tricuspid valve, red area); however, on the epicardial map, substantially larger areas show abnormal low voltage. LAO = left anterior oblique; RAO = right anterior oblique.

Endocardial unipolar voltage mapping (with a cutoff of 5.5 mV) can potentially provide a more global assessment of myocardial voltage and more closely approximate the degree and location of epicardial bipolar voltage abnormalities that serve as a substrate for VT in ARVD patients with no or limited endocardial bipolar voltage abnormalities. Therefore, endocardial unipolar voltage mapping may help in decision making related to proceeding to epicardial substrate mapping and ablation (Fig. 29-7).³³

FIGURE 29-7 Unipolar endocardial electrograms defining the location and greater extent of epicardial bipolar electrogram abnormalities in a patient with arrhythmogenic right ventricular cardiomyopathy-dysplasia. Left: Endocardial bipolar voltage map demonstrates a paucity of low-voltage regions. Center: Endocardial unipolar voltage mapping reveals a much greater burden of abnormal myocardium (<5.5 mV) extending from the lateral tricuspid valve up to the pulmonic valve region and inferiorly across the right ventricular free wall. Right: Epicardial bipolar voltage map confirms the extensive area of abnormal epicardium. Black dots represent wide, split, and/or late epicardial electrograms and help to identify low-voltage areas consistent with scar versus fat.

(From Polin GM, Haqqani H, Tzou W, et al: Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular cardiomyopathy/dysplasia, Heart Rhythm 8:76-83, 2011.)

Noncontact Mapping

When VT is short-lived, hemodynamically unstable, or cannot be reproducibly initiated, simultaneous multisite data acquisition using a noncontact mapping system (EnSite 3000; St. Jude Medical, St. Paul, Minn.) can help localize the VT site of origin. The system can help identify presystolic endocardial activation sites (potential exits) for reentrant VTs and thus starting points for conventional mapping. It may also help identify VT isthmuses and therefore suitable targets for ablation.

The EnSite 3000 noncontact mapping system consists of a noncontact catheter (9 Fr) with a multielectrode array surrounding a 7.5-mL balloon mounted at the distal end. To create a map, the balloon catheter is positioned in the RV through the left subclavian vein approach with a 260-cm-long 0.035-inch guidewire. The distal tip is placed at the RV apex. The balloon is then deployed, and it can be filled with a mixture of contrast and saline to be visualized fluoroscopically. The balloon is positioned in the center of the RV cavity and does not come into physical contact with the atrial walls being mapped. Systemic anticoagulation is critical to avoid thromboembolic complications. Intravenous heparin is usually given to maintain the activated clotting time (ACT) at 250 to 300 seconds.

A conventional deflectable mapping-ablation catheter is also positioned in the RV and used to collect geometry information. The tricuspid annulus, His bundle, and pulmonic valve are initially tagged and a detailed geometry of the RV and RVOT is reconstructed by moving the mapping catheter around the RV. Subsequently, a detailed geometry of the RV is reconstructed by moving the mapping catheter around the endocardium. Using this information, the computer creates a model, called a convex hull, of the chamber during diastole. The patient may be in NSR or tachycardia during creation of the geometry.

Once chamber geometry has been delineated, VT is induced and mapping can begin. The data acquisition process is performed automatically by the system, and all data for the entire chamber are acquired simultaneously. The system then reconstructs more than 3000 unipolar electrograms simultaneously and superimposes them onto the virtual endocardium, producing color-coded isopotential maps to depict graphically regions that are depolarized. Activation can be tracked on the isopotential map throughout the tachycardia cycle and wavefront propagation can be displayed as a user-controlled 3-D “movie.” The color range represents voltage or timing of onset. Sites of early (presystolic) endocardial activity, which are likely adjacent to reentry circuit exits, are usually identifiable; in some cases, isthmuses can be identified. Diastolic depolarization is defined as activity on the isopotential map that can be continuously tracked back in time from VT exit sites, defined on the map as synchronous with the QRS onset. Diastolic activity and exit sites are then marked on the virtual endocardium, and a mapping catheter is navigated to them by the locator technology. Ablation lesions can be tagged, facilitating performing linear ablation devoid of gaps across the tachycardia critical isthmus.

Unless the multielectrode array position is significantly changed, the accepted geometry can be applicable during the entire procedure, and the same geometry can be used for mapping different VT morphologies as well as for substrate mapping.

Substrate mapping based on scar or diseased tissue, which has been recently introduced to the noncontact mapping technology, can be particularly helpful when the VT is not inducible during the procedure. Dynamic substrate mapping allows the creation of voltage maps from a single cardiac cycle and provides the ability to identify low-voltage areas, as well as fixed and functional block, on the virtual endocardium through noncontact methodology.³⁴

Percutaneous Epicardial Mapping

A recent report found that, in patients with ARVD and VT who failed endocardial ablation, low-voltage areas consistent with scar were more extensive on the epicardium than the endocardium and uniformly included wide, split, and late electrograms consistent with abnormal conduction. The origin of VT defined by activation/entrainment mapping and suggested by the detailed substrate mapping observations and targeted for ablation was also frequently noted beyond the endocardially defined scar. Successful epicardial ablation sites were often opposite endocardial sites with normal voltage or where ablation had been ineffective.²³

Therefore, patients with limited endocardial substrate, those with late VT termination with endocardial RF delivery, and certainly those with persistent VT despite aggressive endocardial ablation should be considered for an epicardial approach. Additionally, surface ECG criteria identifying a QS complex in lead V₂ or leads III and aVF during VT can be helpful in identifying patients who should be considered for an epicardial approach with the initial procedure.²³

The approach to epicardial mapping is essentially the same as for endocardial ablation, including activation mapping, entrainment mapping, substrate mapping, and pace mapping, and is discussed in detail in Chapter 27.

Target of Ablation

For stable VTs, reentry circuit isthmuses are the target of ablation and are defined by activation and entrainment mapping as sites with continuous activity or isolated mid-diastolic potentials, entrainment with concealed fusion, PPI equal to the VT CL, and S-QRS interval less than 70% of the VT CL. Alteration of the tachycardia CL or termination of VT by subthreshold or nonpropagated stimuli, and repeated VT termination occurring with catheter manipulation or pacing at the site, also suggest an isthmus location (Fig. 29-8).²⁹

FIGURE 29-8 Ablation of arrhythmogenic right ventricular dysplasia-cardiomyopathy (ARVD) ventricular tachycardia. Shown is radiofrequency (RF) energy delivery in the basal lateral right ventricle in a patient with ARVD. Note the mid-diastolic potential (red arrow) 130 milliseconds prior to QRS onset and almost immediate cessation of ventricular tachycardia on RF delivery. Blue arrow marks initiation of RF energy delivery.

For hemodynamically unstable VTs, substrate mapping during sinus rhythm followed by brief inductions of VT (with the mapping catheter at the optimal site, as identified by substrate and pace mapping) can allow assessment of the relationship of the abnormal electrograms to the VT circuit. This can also allow entrainment maneuvers to be performed at those sites to help distinguish sites critical to the VT circuit from bystander sites. VT is then quickly terminated before significant hemodynamic compromise ensues. This approach can be facilitated by hemodynamic support with the use of intravenous vasopressors, intraaortic balloon pump, or both. This approach, however, requires being able to induce the same VT reproducibly. The use of noncontact mapping can be of value in these cases.

When the critical isthmuses of the reentrant circuit(s) can be identified, linear ablation lesions are delivered to sever those isthmuses. For unmappable VTs (due to poor inducibility or changing VT morphology), RF ablation can be performed as linear lesions targeting potential exit sites and isthmuses of VT circuits based on the location of the best pace map, location of valvular anatomical boundaries, and the substrate defined by the voltage mapping. Typically, linear lesions are created over target regions by sequential point lesions designed to: (1) connect scar or abnormal myocardium to a valve continuity; (2) connect scar or abnormal myocardium to another scar; (3) extend from the most abnormal endocardium (<0.5 mV) to normal myocardium (>1.5 mV); or (4) encircle the scar or abnormal region, depending on the scar location and size.^22,32

Ablation Technique

Ablation may be performed with a 4-mm-tip (maximal target temperature of 60°C and maximal power of 50 W), an 8-mm-tip (maximal temperature of 65°C and maximal power of 75 W), or an irrigated-tip catheter (maximal temperature of 40-45°C and maximal power of 50 W). Irrigated RF ablation catheters are preferred, because power delivery can be limited in the trabeculated and annular regions where cooling from circulating blood flow is low. However, careful attention to power titration and monitoring impedance is necessary to reduce the risk of perforation when ablating in the free wall of the RV.³⁵ Ablation is usually maintained for 60 to 120 seconds at the site where the tachycardia is terminated (in ablation during stable VT) or at each point along the linear ablation.⁸

Endpoints of Ablation

Lack of inducibility of VT is the endpoint of the ablation procedure. When the VT is scarcely inducible at baseline, the disappearance of all the isolated delayed components of the VT-related area can be used as an endpoint for the substrate-based ablation approach.

Outcome

Ablation of VT in ARVD remains a clinical challenge and should be viewed as a potential palliative procedure in selected patients. Although VT ablation has been reported to successfully eliminate inducible VT acutely in 41% to 88% of patients, long-term success is low, and during average follow-ups of 11 to 24 months, VT recurs in 11% to 83% of patients, even following extensive or repeated ablation attempts.^22,29,36 Recurrences usually manifest as nonsustained VT, a slower monomorphic sustained VT, or a new monomorphic VT.^8,35

Multiple morphologies, hemodynamic instability, or noninducibility can limit the success of VT ablation. Additionally, ARVD can create a large number of potential circuits; an empirical approach may not eliminate all the possible pathways. Achieving adequate energy delivery can also be a problem, especially in regions difficult to access by catheter—in particular, abnormal regions along the tricuspid annulus, which can require the use of 8-mm-tip or irrigation-tip ablation catheters to achieve effective lesions underneath the tricuspid valve. It must be emphasized that caution must be taken when using irrigated-tip ablation in patients with very thin ventricular walls, although thick fibrous tissue is present in many areas. The risk of creating deeper lesions in thin walls of the RV body or apex potentially increases the risk of perforation. Although the magnitude of this risk is not well known, it seems to be infrequent.⁸

Furthermore, recent studies have shown a high prevalence of epicardial isthmuses in ARVD. Epicardial mapping and ablation have yielded a significant improvement in ablation efficacy; thus, it is reasonable to begin mapping endocardially, but be prepared to move to epicardial mapping following pericardial puncture if endocardial mapping and ablation do not quickly yield good results. In a recent report, percutaneous epicardial mapping and ablation were performed in 13 consecutive patients with ARVD after failed endocardial VT ablation. Twenty-seven distinct VTs were targeted for ablation from the epicardium; during 18 ± 13 months of follow-up, 77% of patients were free of VT.²³

The discrepancy between the good acute results and the unfavorable long-term outcome may be explained by the progressive nature of ARVD, which predisposes to the occurrence of new and malignant arrhythmogenic substrates over time. This hypothesis, however, has been questioned by a recent study that demonstrated the absence of progression of the area of the endocardial scar as measured by voltage mapping in patients with ARVD and VT, despite demonstrating progressive RV dilation in these patients.⁷