J-Wave syndromes expert consensus conference report: Emerging concepts and gaps in knowledge ^{☆ ☆☆}

Abstract

1 Preamble The J-wave syndromes (JWSs), consisting of the Brugada syndrome (BrS) and early repolarization syndrome (ERS), have captured the interest of the cardiology community over the past 2 decades following the identification of BrS as a new clinical entity by Pedro and Josep Brugada in 1992 [1]. The clinical impact of ERS was not fully appreciated until 2008 [2], [3], [4]. Consensus conferences dedicated to BrS were held in 2000 and 2004 [5], [6], but a consensus conference specifically focused on ERS has not previously been convened other than that dealing with terminology, and guidelines for both syndromes were last considered in 2013 [7]. A great deal of new information has emerged since. The present forum was organized to evaluate new information and highlight emerging concepts with respect to differential diagnosis, prognosis, cellular and ionic mechanisms, and approaches to therapy of the JWSs. Leading experts, including members of the Heart Rhythm Society (HRS), the European Heart Rhythm Association (EHRA), and the Asian-Pacific Heart Rhythm Society (APHRS), met in Shanghai, China, in April 2015. The Task Force was charged with a review of emerging concepts and assessment of new evidence for or against particular diagnostic procedures and treatments. Every effort was made to avoid any actual, potential, or perceived conflict of interest that might arise as a result of outside relationships or personal interest. This consensus report is intended to assist health care providers in clinical decision-making. The ultimate judgment regarding care of a particular patient, however, must be made by the health care provider based on all of the facts and circumstances presented by the patient. Members of this Task Force were selected to represent professionals involved with the medical care of patients with the JWSs, as well as those involved in research into the mechanisms underlying these syndromes. These selected experts in the field undertook a comprehensive review of the literature. Critical evaluation of methods of diagnosis, risk stratification, approaches to therapy, and mechanistic insights was performed, including assessment of the risk- to-benefit ratio. The level of evidence and the strength of the recommendation of particular management options were weighed and graded. Recommendations with class designations are taken from HRS, EHRA, APHRS, and/or European Society of Cardiology (ESC) consensus statements or guidelines [8], [9]. Recommendations without class designations are derived from unanimous consensus of the authors. The consensus recommendations in this document use the commonly used Class I, IIa, IIb, and III classifications and the corresponding language: “is recommended” for a Class I consensus recommendation; “can be useful” or “is reasonable” for a Class IIa consensus recommendation; “may be considered” for a Class IIb consensus recommendation; and “is not recommended” for a Class III consensus recommendation. 2 Introduction The appearance of prominent J waves in the electrocardiogram (ECG) have long been reported in cases of hypothermia [10], [11], [12] and hypercalcemia [13], [14]. More recently, accentuation of the J wave has been associated with life-threatening ventricular arrhythmias [15]. Under these circumstances, the accentuated J wave typically may be so broad and tall as to appear as an ST-segment elevation, as in cases of BrS. In humans, the normal J wave often appears as a J-point elevation, with part of the J wave buried inside the QRS. An early repolarization pattern (ERP) in the ECG, consisting of a distinct J-wave or J-point elevation, or a notch or slur of the terminal part of the QRS with and without an ST-segment elevation, has traditionally been viewed as benign [16], [17]. The benign nature of an ERP was challenged in 2000 [18] based on experimental data showing that this ECG manifestation predisposes to the development of polymorphic ventricular tachycardia (VT) and ventricular fibrillation (VF) in coronary-perfused wedge preparations [15], [18], [19], [20]. Validation of this hypothesis was provided 8 years later by Haissaguerre et al. [2], Nam et al. [3] and Rosso et al. [4]. These seminal studies together with numerous additional case-control and population-based studies have provided clinical evidence for an increased risk for development of life-threatening arrhythmic events and sudden cardiac death (SCD) among patients presenting with an ERP, particularly in the inferior and inferolateral leads. The lack of agreement regarding the terminology relative to early repolarization (ER) has led to a great deal of confusion and inconsistency in reporting [21], [22], [23]. A recent expert consensus report that focused on the terminology of ER recommends that the peak of an end QRS notch and/or the onset of an end QRS slur be designated as Jp and that Jp should exceed 0.1 mV in ≥2 contiguous inferior and/or lateral leads of a standard 12-lead ECG for ER to be present [24]. It was further recommended that the start of the end QRS notch or J wave be designated as Jo and the termination as Jt. ERS and BrS are thought to represent 2 manifestations of the JWSs. Both syndromes are associated with vulnerability to development of polymorphic VT and VF leading to SCD [1], [2], [3], [15] in young adults with no apparent structural heart disease and occasionally to sudden infant death syndrome [25], [26], [27]. The region generally most affected in BrS is the anterior right ventricular outflow tract (RVOT); in ERS, it is the inferior region of the left ventricle (LV) [2], [4], [28], [29], [30], [31], [32]. As a consequence, BrS is characterized by accentuated J waves appearing as a coved-type ST-segment elevation in the right precordial leads V 1−V 3, whereas ERS is characterized by J waves, Jo elevation, notch or slur of the terminal part of the QRS, and ST segment or Jt elevation in the lateral (type 1), inferolateral (type 2), or inferolateral þ anterior or right ventricular (RV) leads (type III) [15]. An ERP is often encountered in ostensibly healthy individuals, particularly in young males, black individuals, and athletes. ERP is also observed in acquired conditions, including hypothermia and ischemia [15], [33], [34]. When associated with VT/VF in the absence of organic heart disease, ERP is referred to as ERS. The prevalence of BrS with a type 1 ECG in adults is higher in Asian countries, such as Japan (0.15–0.27%) [35], [36] and the Philippines (0.18%) [37], and among Japanese-Americans in North America (0.15%) [38] than in western countries, including Europe (0%–0.017%) [39], [40], [41] and North America (0.005–0.1%) [42], [43]. In contrast, the prevalence of an ERP in the inferior and/or lateral leads with a J-point elevation ≥0.1 mV ranges between 1% and 24% and for J-point elevation ≥0.2 mV ranges between 0.6% and 6.4% [44], [45], [46]. No significant regional differences in the prevalence of an ERP have been reported [47]. However, ERP is significantly more common in blacks than in Caucasians. Little in the way of regional differences in the manifestation of ERS has been reported. ERP appears to be more common in Aboriginal Australians than in Caucasian Australians [48]. 3 Updates on the diagnosis of BrS According to the 2013 consensus statement on inherited cardiac arrhythmias [8] and the 2015 guidelines for the management of patients with ventricular arrhythmias and prevention of SCD [9]: “BrS is diagnosed in patients with ST- segment elevation with type 1 morphology ≥2 mm in ≥1 lead among the right precordial leads V1, V2, positioned in the 2nd, 3rd or 4th intercostal space occurring either spontaneously or after provocative drug test with intravenous administration of Class I antiarrhythmic drugs. BrS is diagnosed in patients with type 2 or type 3 ST-segment elevation in ≥1 lead among the right precordial leads V1, V2 positioned in the 2nd, 3rd or 4th intercostal space when a provocative drug test with intravenous administration of Class I antiarrhythmic drugs induces a type I ECG morphology.” The present Task Force is concerned that this could result in overdiagnosis of BrS, particularly in patients displaying a type 1 ECG only after a drug challenge. Data suggest the latter population is at very low risk and that the presumed false- positive rate of pharmacologic challenge is not trivial [49]. Although a rigorous process was undertaken to establish the preceding guidelines, there remains no gold standard for establishing a diagnosis, particularly in patients with weak evidence of disease. Accordingly, we recommend adoption of the following diagnostic criteria and score system for BrS. Consistent with the recommendation of the 2013 and 2015 guidelines, only a type 1 (“coved-type”) ST-segment elevation is considered diagnostic of BrS (Fig. 1), and BrS is characterized by ST-segment elevation ≥2 mm (0.2 mV) in ≥1 right precordial leads (V 1−V 3) positioned in the 4th, 3rd, or 2nd intercostal space. However, as a departure from the guidelines, this consensus report recommends that when a type 1 ST-segment elevation is unmasked using a sodium channel blocker (Table 1), diagnosis of BrS should require that the patient also present with 1 of the following: documented VF or polymorphic VT, syncope of probable arrhythmic cause, a family history of SCD at o45 years old with negative autopsy, coved-type ECGs in family members, or nocturnal agonal respiration. Inducibility of VT/VF with 1 or 2 premature beats supports the diagnosis of BrS under these circumstances [50]. A type 2 (“saddle-back type”) or type 3 ST-segment elevation cannot substitute for a type 1, unless converted to type 1 with fever or sodium drug challenge. A drug challenge–induced type 1 can be used to diagnose BrS only if accompanied by 1 of the criteria specified above. Type 2 is characterized by ST-segment elevation ≥0.5 mm (generally ≥2 mm in V 2) in ≥1 right precordial lead (V 1−V 3), followed by a convex ST. The ST segment is followed by a positive T wave in V2 and variable morphology V 1. Type 3 is characterized by either a saddleback or coved appearance with an ST-segment elevation o1 mm. Placement of the right precordial leads in more cranial positions (in the 3rd or 2nd intercostal space) in a 12-lead resting ECG or 12-lead Holter ECG increases the sensitivity of ECG [51], [52], [53]. It is recommended that ECG recordings be obtained in the standard and superior positions for the V 1 and V 2 leads. Veltman et al. [54]. showed that RVOT localization using magnetic resonance imaging (MRI) correlates with type 1 ST-segment elevation in BrS and that lead positioning according to RVOT location improves the diagnosis of BrS. Interestingly, in most cases a type I pattern was found in the 3rd intercostal space in the sternal and left parasternal positions [54]. In reviewing ECGs of a large cohort of BrS patients, Richter et al [55]. concluded that lead V 3 does not yield diagnostic information in BrS. A proposed diagnostic score system for BrS, referred to as the Proposed Shanghai BrS Score, is presented in Table 2. These recommendations are based on the available literature and the clinical experience of the Task Force members [8], [56], [57], [58], [59], [60]. Weighting of variables is based on expert opinion informed by cohort studies that typically do not include all variables presented. Thus, rigorous, objectively weighted coefficients were not derived from large-scale risk factor and outcome- informed datasets. Nonetheless, the authors believed that some inferential weighting would be of benefit when applied to patients. As with all such recommendations, they will need to undergo initial and ongoing validation in future studies. 4 Pharmacologic tests and other diagnostic tools When there is clinical suspicion of BrS in the absence of spontaneous type 1 ST-segment elevation, a pharmacologic challenge using a sodium channel blocker is recommended. A list of agents used for this purpose is presented in Table 1 (also see www.brugadadrugs.org). The test is considered positive only if a type 1 ECG pattern is obtained, and it should be discontinued in case of frequent ventricular extrasystoles or other arrhythmias, or widening of the QRS 4130% over the baseline value [6]. As an alternative, the “full stomach test” has been proposed for diagnosing BrS [61]. In this case, ECGs are performed before and after a large meal. The use of “high electrodes” increases the sensitivity for recognizing spontaneous type I ST-segment elevation at night or after heavy meals [62]. A type 1 ST-segment elevation recorded using a Holter is a spontaneous type 1, and it is reasonable to assume that a spontaneous type 1 recorded by Holter at night or after a large meal has more value—both diagnostic and prognostic—than a drug-induced type 1. Drug challenge is not indicated in asymptomatic patients displaying the type 1 ECG under baseline conditions because of the lack of the additional diagnostic value. These provocative drug tests are also not recommended in cases in which fever has been documented to induce a type I ECG, other than for research purposes. Much debate has centered around the definition of a false-positive sodium channel block challenge [63]. The consensus is that a false-positive is difficult to define because of the lack of a gold standard. The development of a type 1 ST-segment elevation in response to sodium block challenge should be considered as probabilistic, rather than binary, in nature. As will be discussed later, a similar approach is recommended in evaluating the ability of genetic variants to promote the BrS phenotype. Asymptomatic patients with a family history of BrS or SCD should be informed of the availability of a sodium channel blocker challenge test to provide a more definitive diagnosis of BrS. However, patients should be advised that no therapy may be recommended regardless of the outcome because the long- term risk of patients with BrS diagnosed by this test is significantly lower than the risk of patients with spontaneous type 1. Patients also should be informed about the risk of the test and about the emotional consequences of having a positive test not followed by definitive therapy. The decision as to whether to undergo the drug challenge ultimately should be left up to the well-informed patient. Performing an ajmaline test in children is problematic for 2 reasons. First, the test is apparently less sensitive in children than in adults. In fact, in 1 study, a repeat ajmaline challenge performed after puberty unmasked BrS in 23% of relatives with a previously negative drug test performed during childhood [64]. Second, the test is associated with greater risk than in adults. In 1 series, 10% of children undergoing the ajmaline test, including 3% of the asymptomatic subgroup, developed sustained VT [64], [65]. Caution also should be exercised when performing a sodium blocker challenge in adults with a known pathogenic sodium channel mutation or in patients with prolonged PR intervals, pointing to a carrier of such a mutation [66]. 5 Differential diagnosis Other causes of ST-segment elevation should be excluded before establishing the diagnosis of BrS (Table 3). Artifacts secondary to low-pass filtering should be ruled out [67]. Circumstances that produce a type 1 Brugada-like ECG include right bundle branch block (RBBB), pectus excavatum, arrhythmogenic right ventricular cardiomyopathy (ARVC), and occlusion of the left anterior descendent artery or the conus branch of the right coronary artery, which supplies the RVOT (Table 3A). Discrimination between BrS and ARVC is particularly challenging. Although debate continues as to the extent to which structural abnormalities are present in BrS, most investigators consider BrS to be a channelopathy. Concealed structural abnormalities, such as histologic myocardial fibrosis of the RVOT, which may not become evident using conventional imaging techniques, have been proposed to account for or contribute to delayed conduction and ventricular arrhythmias in BrS. MRI and electron beam computed tomographic studies of BrS patients consistently show subtle abnormalities, including wall motion abnormalities and reduced contractile function of the RV and, to a lesser extent, of the LV, and dilation of the RVOT [68], [69], [70], [71]. In the only study that discriminated between patients with and those without SCN5A mutations, no difference was observed in RVOT dimensions or RV ejection fraction between these patients. Slightly greater depressions of LV dimensions and ejection fraction were observed in patients with SCN5A mutations. Significant differences were observed in RV and LV dimensions and ejection fraction compared to healthy controls [72]. Cardiac dilation and reduced contractility in all of these studies were attributed to structural changes (fibrosis, fatty degeneration). However, as noted by van Hoorn et al. [72] virtually no signs of fibrosis or fatty degeneration could be detected, perhaps because the spatial resolution of the imaging used was too low to detect such subtle changes. Antzelevitch and colleagues have long suggested an alternative explanation [31], [73], [74]. Loss of the action potential (AP), which has been shown in experimental models to create the arrhythmogenic substrate in BrS, leads to contractile changes that could explain the wall motion abnormalities observed. The all-or-none repolarization at the end of phase 1 of the epicardial AP responsible for loss of the dome causes the calcium channel to inactivate very soon after it activates. As a consequence, calcium channel current is dramatically reduced, the cell becomes depleted of calcium, and contractile function ceases in those cells. This is expected to lead to wall motion abnormalities, particularly in the RVOT, dilation of the RVOT region, and reduced ejection fraction observed in patients with BrS. It has also been proposed that the loss of the AP dome, because it creates a hibernation-like state, may, over long periods of time, lead to mild structural changes, including intracellular lipid accumulation, vacuolization, and connexin 43 redistribution. These structural changes may, in turn, contribute to the arrhythmogenic substrate of BrS, although they are very different from those encountered in arrhythmogenic right ventricular cardiomyopathy/dysplasia (ARVC/D) [31], [75]. This hypothesis would predict that some of the changes observed by recent studies may be the result of, rather than the cause of, the BrS phenotype [76]. In a recent study, Nademanee et al. [76]. reported additional evidence pointing to pathologic changes in the RVOT of patients with BrS that have proved undetectable by echocardiography or MRI. In contrast, imaging techniques in ARVC clearly display morphologic and functional changes (e.g., dilation, bulging/ aneurysms, wall motion abnormalities). ARVC is an inherited cardiac disease resulting from genetically defective desmosomal (DS) proteins [77], [78], characterized by fibrofatty myocardial replacement predisposing to scar-related ventricular arrhythmias that may lead to SCD, mostly in young people and athletes [79]. Life-threatening ventricular arrhythmias may occur early, during the “concealed phase” of the disease, before overt structural changes [77], [78], [80]. Recent experimental studies demonstrated that loss of expression of DS proteins may induce electrical ventricular instability by causing sodium channel dysfunction and current reduction as a consequence of the cross-talk between these molecules at the intercalated discs, which predisposes to sodium current-dependent lethal arrhythmias, similar to those leading to SCD in patients with J-wave syndromes [80], [81], [82]. Further evidence of the overlap between phenotypic manifestation of ARVC and BrS comes from (1) clinicopathologic studies showing that a subset of ARVC patients may share ECG changes and patterns of ventricular arrhythmias with BrS [83]; and (2) genotype–phenotype correlation studies demonstrating that PKP2 mutation may cause a Brugada phenotype in the human heart by reducing sodium current [84]. These findings support the concept that specific DS gene mutations involved in the pathogenesis of ARVC can lead to a decreased depolarization reserve that manifests as J-wave/BrSs. Thus, ARVC and J wave syndromes are not completely different conditions but are the ends of a spectrum of structural myocardial abnormalities and sodium current deficiency that share a common origin as diseases of the connexome [84]. The ECG abnormalities in ARVC are not dynamic and display a constant T-wave inversion, epsilon waves, and, in the progressive stage, reduction of the R amplitude. End-stage ARVC is usually associated with monomorphic VT with left bundle branch morphology and is precipitated by catecholamines [85], whereas BrS is associated with polymorphic VT predominantly during sleep or rest [86]. A positive ajmaline challenge has been reported in 16% of patients with ARVC [87], [88]. 6 Modulating factors Sympathovagal balance, hormones, metabolic factors, and pharmacologic agents are thought to modulate not only ECG morphology but also explain the development of ventricular arrhythmias under certain conditions [89]. Any of these modulating factors, if present, should be promptly corrected (Table 3B). 7 Acquired Brugada pattern and phenocopies The Brugada ECG is often concealed and can be unmasked with a wide variety of drugs and conditions, including a febrile state, vagotonic agents and maneuvers, α-adrenergic agonists, β-adrenergic blockers, Class IC antiarrhythmic drugs, tricyclic or tetracyclic antidepressants, hyperkalemia, hypokalemia, hypercalcemia, and alcohol and cocaine toxicity [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100]. Preexcitation of RV can unmask the BrS phenotype in cases of RBBB [101]. An up-to-date list of agents known to unmask the Brugada ECG that should be avoided by patients with BrS can be found at www.brugadadrugs.org [89]. Environmental factors leading to the appearance of an ECG similar or identical to a type 1 BrS pattern in the absence of any apparent genetic dysfunction has been suggested to represent a Brugada ECG phenocopy [102]. Features of the Brugada phenocopies include (1) Brugada- like ECG pattern; (2) presence of an identifiable underlying condition; (3) disappearance of the ECG pattern after resolution of the condition; (4) absence of family history of sudden death in relatively young first-degree relatives (≤45 years) or of type 1 BrS pattern; (5) absence of symptoms such as syncope, seizures, or nocturnal agonal respiration; and (6) a negative sodium channel blocker challenge test. Debate continues as to the appropriateness of this terminology given that it is very difficult to rule out a genetic predisposition, which is a prerequisite for designating the ECG manifestation as a phenocopy. Designation of these conditions as acquired forms of Brugada ECG pattern or BrS may be more appropriate and better aligned with the terminology used in the long QT syndrome. 8 Update on the diagnosis of ERS ERS is generally diagnosed in patients who display ER in the inferior and/or lateral leads presenting with aborted cardiac arrest, documented VF, or polymorphic VT. Consistent with the recent consensus report on ERP [24], ER is recognized if (1) there is an end QRS notch (J wave) or slur on the downslope of a prominent R wave with and without ST-segment elevation; (2) the peak of the notch or J wave (J p) ≥0.1 mV in ≥2 contiguous leads of the 12-lead ECG, excluding leads V 1−V 3; and (3) QRS duration (measured in leads in which a notch or slur is absent) o120 ms. Table 4 lists the exclusion criteria in the differential diagnosis of ERS. A proposed diagnostic score system for ERS, referred to as the Proposed Shanghai ERS Score, is presented in Table 5. The scoring system is based on evidence available in the literature to date. As in BrS, weighting of variables is based on expert opinion informed by cohort studies that do not include all variables presented. Thus, rigorous, objectively weighted coefficients were not derived from large-scale risk factor- and outcome-informed datasets. Nonetheless, the authors believed that some inferential weighting would be of benefit when applied to patients. As with all such recommendations, they will need to undergo initial and ongoing validation in future studies. 9 Similarities and difference between BrS and ERS BrS and ERS display several clinical similarities, suggesting similar pathophysiology (Table 6) [19], [21], [103], [104], [105]. Males predominate in both syndromes, with BrS presenting in 71–80% among Caucasians and 94–96% among Japanese [106], [107]. In the setting of ERP, VF occurred mainly in males (72%) when studied in an international cohort [2] but in a much higher percentage in a report by Japanese investigators [108]. BrS and ERS patients may be totally asymptomatic until they present with cardiac arrest. In both syndromes, the highest incidence of VF or SCD occurs in the third decade of life, perhaps related to testosterone levels in males [109]. In both syndromes, the appearance of accentuated J waves and ST-segment elevation is generally associated with bradycardia or pauses [110], [111]. This can explain why VF in both syndromes often occurs during sleep or during a low level of physical activities [108], [112]. The QT interval is relatively short in patients with ERS [2], [113], and BrS who carry mutations in calcium channel genes [114]. As will be discussed in more detail later, ERS and BrS also share similarities with respect to the response to pharmacologic therapy. In both, electrical storms and associated J-wave manifestations can be suppressed using β- adrenergic agonists [115], [116], [117], [118]. Chronic oral pharmacologic therapy using quinidine [119], [120], bepridil [117], denopamine [115], [121], and cilostazol [115], [117], [121], [122], [123], [124], [125] is reported to suppress the development of VT/VF in both ERS and BrS secondary to inhibition of Ito, augmentation of ICa, or both [3], [122], [126]. Differences between the 2 syndromes include (1) the region of the heart most affected (RVOT vs inferior LV); (2) the presence of (discrete) structural abnormalities in BrS but not in ERS; (3) the incidence of late potentials in signal- averaged ECGs (BrS 60% 4 ERS 7%) [108]; and (4) greater elevation of Jo, Jp, or Jt (ST-segment elevation) in response to sodium channel blockers in BrS vs ERS and higher prevalence of atrial fibrillation in BrS vs ERS [127]. Early studies suggested a different pathophysiologic basis for ERS and BrS based on the observation that sodium channel blockers unmask or accentuate J-wave manifestation in BrS but reduces the amplitude in ERS [108]. However, the recent study by Nakagawa et al. [357]. showed that J waves recorded using unipolar LV epicardial leads introduced into the left lateral coronary vein in ERS patients are indeed augmented, even though J waves recorded in the lateral precordial leads are diminished, due principally to engulfment of the surface J wave by the widened QRS [29], [108]. The case report of Nakagawa et al. has recently been supplemented with additional cases in which this technique was used; 2 of these 3 cases showed pilsicainide-induced accentuation of the J waves in electrograms recorded from the epicardial surface of the LV (H. Morita, unpublished observations). Also in support of the thesis that these ECG patterns and syndromes are closely related are reports of cases in which ERS transitions into ERS plus BrS [105], [128]. The principal difference between BrS and ERS is related to the region of the ventricle most affected. Epicardial mapping studies in BrS patients report accentuated J waves and fragmented and/or late potentials in the epicardial region of the RVOT [129], [130], [131], whereas in ERS only accentuated J waves, particularly in the inferior wall of LV, are observed [29]. Fractionated electrogram activity and late potentials have been observed in experimental models of ERS [30] but have not yet been reported clinically. Noninvasive mapping electroanatomic studies have reported very steep localized repolarization gradients across the inferior/lateral regions of LV of ERS patients, preceded by normal ventricular activation [132], whereas in BrS both slow discontinuous conduction and steep dispersion of repolarization are present in the RVOT [133]. Another presumed difference is the presence of structural abnormalities in BrS, which have not yet been described in ERS [76]. Although J waves are accentuated or induced by both hypothermia and fever [33], [34], [134], [135], [136], [137], [138], [139], the development of arrhythmias in ERS is much more sensitive to hypothermia, and arrhythmogenesis in BrS appears to be promoted only by fever [33], [34], [138], [139]. Hypothermia has been reported to increase the risk of VF in ERS [33], [34], [134], [135], [140], and fever is well recognized as a major risk factor in BrS [138], [139]. It is noteworthy that hypothermia can diminish the manifestation of a BrS ECG when already present [141], [142]. An ERP is associated with an increased risk for VF in patients with acute myocardial infarction[143] and hypothermia [33], [144]. A concomitant ERP in the inferolateral leads has also been reported to be associated with an increased risk of arrhythmic events in patients with BrS. Kawata et al. [145]. reported that the prevalence of ER in inferolateral leads was high (63%) in BrS patients with documented VF. 10 Genetics BrS has been associated with variants in 18 genes (Table 7). To date, more than 300 BrS-related variants in SCN5A have been described [21], [146], [147], [148] Fig. 2 shows the overlap syndromes attributable to genetic defects in SCN5A. Loss-of- function mutations in SCN5A contribute to the development of both BrS and ERS, as well as to a variety of conduction diseases, Lenegre disease, and sick sinus syndrome. The available evidence suggests that the presence of a prominent Ito determines whether loss-of-function mutations resulting in a reduction in INa will manifest as BrS/ERS or as conduction disease [59], [149], [150], [151]. Variants in CACNA1C (Cav1.2), CACNB2b (Cavβ2b), and CACNA2D1 (Cavα2δ) have been reported in up to 13% of probands [152], [153], [154], [155]. Mutations in glycerol-3-phophate dehydrogenase 1-like enzyme gene (GPD1L), SCN1B (β1 subunit of Na channel), KCNE3 (MiRP2), SCN3B (β3 subunit of Na channel), KCNJ8 (Kir6.1), KCND3 (Kv4.3), RANGRF (MOG1), SLMAP, ABCC9 (SUR2A), (Navβ2), PKP2 (plakophillin-2), FGF12 (FHAF1), HEY2, and SEMA3A (semaphorin) are relatively rare [156], [157], [158], [159], [160], [161], [162], [163], [164], [165], [166], [167], [168], [169], [170], [171], [172], [173], [174], [175], [176]. An association of BrS with SCN10A, a neuronal sodium channel, was recently reported [167], [177], [178]. A wide range of yields of variants was reported by the 2 studies that examined the prevalence of pathogenic SCN10A mutations and rare variants (5–16.7%) [177], [178], [179]. Mutations in these genes lead to loss of function in sodium (I Na) and calcium (I Ca) channel currents, as well as to a gain of function in transient outward potassium current (I to) or ATP-sensitive potassium current (I K-ATP) [178]. New susceptibility genes recently proposed and awaiting confirmation include the transient receptor potential melastatin protein-4 gene (TRPM4) [180] and the KCND2 gene. The mutation uncovered in KCND2 in a single patient was shown to cause a gain of function in Ito when heterologously expressed [181]. Variants in KCNH2, KCNE5, and SEMA3A, although not causative, have been identified as capable of modulating the substrate for the development of BrS [182], [183], [184], [185]. Loss-of-function mutations in HCN4 causing a reduction in the pacemaker current I f can unmask BrS by reducing heart rate [186]. An ERP in the ECG has been shown to be familial [187], [188], [189]. ERP and ERS have been associated with variants in 7 genes. Consistent with the findings that IK-ATP activation can generate an ERP in canine ventricular wedge preparations, variants in KCNJ8 and ABCC9, responsible for the pore- forming and ATP-sensing subunits of the IK-ATP channel, have been reported in patients with ERS [156], [158], [190]. Loss-of- function variations in the α1, β2, and α2δ subunits of the cardiac L-type calcium channel (CACNA1C, CACNB2, CACNA2D1) and the α1 subunit of NaV1.5 and NaV1.8 (SCN5A, SCN10A) have been reported in patients with ERS [113], [152], [177]. It is important to point out that only a small fraction of identified genetic variants in the genes associated with BrS and ERS have been examined using functional expression studies to ascertain causality and establish a plausible contribution to pathogenesis. Only a handful have been studied in genetically engineered animal models, and very few have been studied in native cardiac cells or in induced pluripotent stem cell-derived cardiac myocytes isolated from ERS and BrS patients. Computational strategies developed to predict the functional consequences of mutations are helpful, but these methods have not been rigorously tested. The lack of functional or biologic validation of mutation effects remains the most severe limitation of genetic test interpretation, as recently highlighted by Schwartz et al. [191]. Recent technological advances have resulted in expansion of disease-specific panels [192]. Large public databases of genetic variation from next-generation sequencing programs such as the 1000 Genomes Project, the National Heart Lung and Blood Institute Grand Opportunity Exome Sequencing Project (GO-ESP), and the Exome Aggregation Consortium (ExAC), have challenged drastically our understanding of the “normal” burden and extent of background genetic variation within cardiac channelopathy susceptibility genes [193], [194], [195]. Although SCN5A variants account for 18–28% of BrS [196], SCN5A genetic testing is complicated by an approximately 3–5% “benign” variant frequency in the general population [194]. Therefore, even in the most common genetic cause of BrS, 1 in 10 “positive” tests could be a “false- positive” even if found in an individual with a robust BrS phenotype. To date, there are more than 20 JWS susceptibility genes [146], [195], [197]. However, these additional genes have only magnified the issues of interpretation by adding to the overall “genetic noise” without significantly increasing the true mutation yield [178], [198], [199], [200]. In fact, 1 study revealed that 1:23 individuals in the GO-ESP population possess a previously published BrS-associated variant that would prompt a “positive” genetic test had it been identified in a patient [201]. These issues reinforce the necessity to interpret JWS genetic test results as strictly probabilistic, rather than binary/deterministic, in nature. Additional lines of evidence [202] can be amassed to aid in the probabilistic interpretation of variants in JWS susceptibility genes, such as case phenotype [203], segregation, functional studies [204], in silico predictions [205], [206], [207], [208], variant type and location [194], and variant frequency in cases and control databases [193]. Despite these aids, a large number of variants remain in “genetic purgatory,” and this number will only increase as exome/genome sequencing becomes more utilized. This then demands the development and utilization of a uniform variant repository that would include clinical assertions and evidence for variant classification. Even with these issues, the emergence of exome/genome sequencing holds promise for the opportunity to study genetic variation like never before, holding the promise of improvements in diagnostic, prognostic, and therapeutics for the JWSs and the other heritable cardiac channelopathies. Kapplinger et al. [209]. recently reported the synergistic use of up to 7 in silico tools to help promote or demote a variant׳s pathogenic status and alter its relegation to genetic purgatory. It is noteworthy that in a recent study, Le Scouarnec et al. [199]. estimated the burden of rare coding variation in arrhythmia susceptibility genes among 167 BrS index patients and compared that with 167 individuals ≥65 years old with no history of cardiac arrhythmia. The authors concluded that, except for SCN5A, rare coding variations in all previously reported BrS susceptibility genes do not contribute significantly to the occurrence of BrS in a population of European ancestry, emphasizing that caution should be taken when interpreting genetic variations in these other genes because rare coding variants are observed to a similar extent in both cases and controls [199]. Similar data were obtained and a similar conclusion was reached by Kapplinger et al. [209]. by analyzing the prevalence of rare variants in the BrS susceptibility genes in the publicly available ExAC exomes. Collectively, these data suggest the possibility that. in the individual patient, BrS and the susceptibility to VF and SCD may not be due to a single mutation (classic mendelian view) but rather to inheritance of multiple BrS susceptibility variants (oligogenic) acting in concert through one or more mechanistic pathways [167]. This also fits with the findings of Probst et al. [210] that in 5 of 13 large families with a putative SCN5A mutation, the genotype did not co-segregate with the phenotype. In addition to the multifactorial nature of the genetics, expressivity of the syndrome may be multifactorial in that the genetic predisposition can be modulated by hormonal (testosterone [211], [212], thyroxine [213]) and other environmental factors, as well as morphologic changes (fibrosis) [76]. 11 Update on the ionic and cellular mechanisms underlying BrS and ERS The JWSs are so named because they involve accentuation of the ECG J wave. Experimental evidence indicates that the J wave is inscribed as a consequence of a transmural voltage gradient caused by the manifestation of an AP notch in epicardium but not endocardium due to a heterogeneous transmural distribution of Ito [104]. An end of QRS notch, resembling a J wave, has been proposed to be due to intraventricular conduction delays. The 2 ECG manifestations can be distinguished based on their response to rate, with the latter showing accentuation at faster rates [24], [59]. The cellular mechanisms underlying JWS have long been a matter of debate [214], [215]. In the case of BrS, 2 principal hypotheses have been advanced. (1) The repolarization hypothesis asserts that an outward shift in the balance of currents in RV epicardium can lead to repolarization abnormalities resulting in the development of phase 2 reentry, which generates closely coupled premature beats capable of precipitating VT/VF. (2) The depolarization hypothesis suggests that slow conduction in the RVOT, secondary to fibrosis and reduced Cx43 leading to discontinuities in indeterminate conduction, plays a primary role in the development of the ECG and arrhythmic manifestations of the syndrome. Conduction slowing is not necessarily limited to the RVOT area. Some investigators have postulated that changes in ion channel current responsible for BrS (i.e., loss of function INa and ICa and gain of function of Ito) can alter AP morphology so as to reduce the safety of conduction at high-resistance junctions, such as regions of extensive fibrosis [216], [217]. Others have argued that this is highly unlikely because conduction at critical junctions of current-to-load mismatch is exquisitely sensitive to changes in rate. The typical behavior of patients with BrS to acceleration of rate is diminution of ST-segment elevation, opposite to that expected at a site of discontinuous conduction. The diminution of ST-segment elevation is consistent with the reduced availability of Ito at the faster rate due to slow recovery of the current from inactivation [59], [214]. The repolarization and depolarization theories are not necessarily mutually exclusive and may indeed be synergistic. The most compelling apparent evidence in support of the depolarization hypothesis derives from the seminal studies of Nademanee et al. [129]. showing that radiofrequency ablation (RFA) of epicardial sites displaying late potentials and fractionated bipolar electrograms in the RVOT of patients with BrS significantly reduced the arrhythmia vulnerability as well as the ECG manifestation of the syndrome. Similar results were reported by Brugada et al. [131]. and by Sacher et al. [130] who also observed in an isolated case that accentuation of the Brugada ECG by ajmaline was associated with an increased area of low-voltage and fragmented electrogram activity. A wider area of low-voltage activity was associated with a more prominent ST-segment elevation [131]. These authors concluded that the late potentials and fractionated electrogram activity are due to conduction delays within the RVOT/RV anterior wall and that ablation of the sites of slow conduction is the basis for the ameliorative effect of ablation therapy [129], [130], [131]. In a direct test of this hypothesis, Szel and Antzelevitch [150] provided evidence for an alternative mechanism using an experimental model of BrS. The low-voltage fractionated electrogram activity was shown to develop as a result of regional desynchronization in the appearance of the second AP upstroke, secondary to accentuation of the epicardial AP notch, and high-frequency late potentials to develop in the RV epicardium secondary to concealed phase 2 reentry. Delayed conduction of the primary beat was never observed in a wide variety of BrS models created by exposing canine RV wedge preparations to drugs mimicking the different genetic defects known to give rise to BrS [150]. In more recent studies, ablation of the RV epicardium was shown to diminish the manifestation of J waves and ST-segment elevation and to abolish all arrhythmic activity by destroying the cells with the most prominent AP notch, thus eliminating the cells responsible for the repolarization abnormalities that give rise to phase 2 reentry and VT/VF [15], [218]. Confirmation of all of these results in in vivo animal models is desirable. In an attempt to create such a model, Park et al. [149]. recently genetically engineered Yucatan minipigs to heterozygously express a nonsense mutation in SCN5A (E558X) originally identified in a child with BrS. Patch clamp analysis of atrial myocytes isolated from the SCN5AE558X/þ pigs showed a loss of function of INa. Conduction abnormalities consisting of prolongation of P wave, QRS complex, and PR interval were observed, but a BrS phenotype was not observed, not even after administration of flecainide. These observations are expected because of the lack of Ito in the pig, which is a prerequisite for the development of the repolarization abnormalities associated with BrS. Some have argued that the absence of a BrS phenotype is due to the young age of the minipigs (22 months) [219]. However, it is difficult to reconcile why the minipigs manifest major conduction delays at this age but not a BrS phenotype, if indeed the latter depends on the former. Finally, it is noteworthy that monophasic APs recorded from the epicardial and endocardial surfaces of the RVOT of a patient with BrS are nearly identical to transmembrane APs recorded from the epicardial and endocardial surfaces of the wedge model of BrS [220], [221]. These differences were not observed in an isolated heart explanted from a BrS patient after transplantation of a new heart. However, the epicardium of this heart was very depressed, perhaps as a result of the 129 shocks delivered by the implantable cardioverter-defibrillator (ICD) in an attempt to control the multiple electrical storms [32]. Zhang et al. [133]. recently performed noninvasive electrocardiographic imaging (ECGI) on 25 BrS and 6 RBBB patients. The authors concluded that both slow discontinuous conduction and steep dispersion of repolarization are present in the RVOT of patients with BrS. ECGI was able to differentiate between BrS and RBBB. Unlike BrS, RBBB showed delayed activation in the entire RV, without ST- segment elevation, fractionation, or repolarization abnormalities showing on the electrograms. Importantly, the response to an increase in rate was studied in 6 BrS patients. Increasing rate increased fractionation of the electrogram but reduced ST-segment elevation, indicating that the conduction impairment was not the principal cause of the BrS ECG. The congruence between BrS and ERS with respect to clinical manifestations and response to therapy lends further support to the repolarization hypothesis. Using an experimental model of ERS, Koncz et al. [30]. recently provided evidence in support of the hypothesis that, similar to the mechanism operative in BrS, an accentuation of transmural gradients in the LV wall are responsible for the repolarization abnormalities underlying ERS, giving rise to J-point elevation, distinct J waves, or slurring of the terminal part of the QRS. The repolarization defect is accentuated by cholinergic agonists and reduced by quinidine, isoproterenol, cilostazol, and milrinone, accounting for the ability of these agents to reverse the repolarization abnormalities responsible for ERS [30], [222]. Higher intrinsic levels of Ito in the inferior LV were also shown to underlie the greater vulnerability of the inferior LV wall to VT/VF [30]. The advent and implementation of ECGI by Rudy and colleagues provided additional evidence for repolarization abnormalities by identifying abnormally short activation–recovery intervals in the inferior and lateral regions of LV and a marked dispersion of repolarization [132]. More recent studies involving ECGI mapping in an ERS patient during VF have demonstrated VF rotors anchored in the inferior lateral LV wall [22]. Conduction delay is known to give rise to notching of the QRS complex. When it occurs on the rising phase of the R wave, it is due to a conduction defect within the ventricle. When it occurs at the terminal portion of the QRS, thus masquerading as a J wave, it may be due to either a conduction defect or a repolarization defect [21], [223]. The response to prematurity or to an increase in rate can differentiate between the two [59]. Delayed conduction invariably becomes more exaggerated at faster rates or during premature beats, thus leading to accentuation of the QRS notch, whereas repolarization defects usually are mitigated, resulting in diminution of the J wave at faster rates. Although typical J waves usually are accentuated with bradycardia or long pauses, the opposite has also been described [224], [225]. J waves are often seen in young males with no apparent structural heart diseases, whereas intraventricular conduction delay is often observed in older individuals or those with a history of myocardial infarction or cardiomyopathy [223], [224]. The prognostic value of a fragmented QRS has been demonstrated in BrS [49], [226], although fragmentation of the QRS is not associated with increased risk in the absence of cardiac disease [227]. Factors that may aid in the differential diagnosis of J wave vs intraventricular conduction delay (IVCD)-mediated syndromes are summarized in Table 8. 12 Risk stratification 12.1 J-wave syndromes A great deal of attention has been devoted to risk assessment for the development of life-threatening arrhythmias in BrS and ERS [1], [228]. The incidental discovery of a J wave on routine screening should not be interpreted as a marker of “high risk” for SCD because the odds for this fatal disease are approximately 1:10,000 [229]. Rosso et al. indicated that the presence of a J wave on the ECG increases the probability of VF from 3.4:100,000 to 11:100,000 [4], [230]. However, careful attention needs to be paid to subjects with “high-risk” ER or J waves. Fig. 3 illustrates the various ECG manifestations of ER. Fig. 4 shows a graphic representation of the prevalence and arrhythmic risk associated with the appearance of ECG J waves and clinical manifestations of BrS and ERS. Table 9, Table 10, Table 11 list the available data from studies designed to identify patients at high risk for BrS and ERS. Among these risk stratifiers, some are highly predictive, including (1) history of cardiac events or syncope likely due to VT/VF and (2) prominent J waves in global leads including type 1 ST-segment elevation in the right precordial leads (Fig. 5). 13 Early repolarization syndrome The majority of the studies using the criteria of Haissaguerre et al. [2]. for diagnosing the ERP have shown that ER, especially in the inferior ECG leads, predicts cardiac and arrhythmic death. Negative studies are few and may be attributable to the exclusion criteria used (e.g., atrial fibrillation, flutter, acute coronary syndrome), a relatively short follow-up period [231], [232], or different definitions of ERP [233]. The recent consensus paper by Macfarlane et al. [24]. dealing with the terminology of J-wave–related phenomena in the setting of ER should enable us to avoid such confusion in the future. The inclusion of Africans or African-Americans, in whom ER is prevalent but apparently not associated with high risk, may alter outcomes as well [234]. Huikuri and colleagues reported in a series of seminal papers the results of a population-based study in Finland involving long-term prognosis of subjects with an ERP in the ECG [44]. Tikkanen et al. [44]. showed that J-point elevation ≥0.1 mV was present in 5.8% of the population and that only 0.3% of the population had significant J-point elevation ≥0.2 mV. J-point elevation ≥0.1 mV in the inferior leads was associated with cardiovascular death (relative risk [RR] 1.28) and arrhythmic death (RR 1.43), and J-point elevation ≥0.2 mV had a markedly elevated risk of death from cardiac causes (RR 2.92) and from arrhythmia (RR 2.92). Subsequent studies confirmed the association of J wave or ER with death from all causes, death from cardiovascular disease, sudden/unexpected death, and death from arrhythmias [44], [45], [46], [127], [235], [236], [237], [238]. A horizontal or descending ST segment is associated with a worse prognosis than is an ascending ST segment (Fig. 3) [237], [239]. Individuals with a high-amplitude J wave ≥0.2 mV followed by a horizontal or descending ST segment in the inferior/inferolateral leads have a higher risk of lethal arrhythmias than do those with a lower-amplitude J wave, especially those with a rapidly ascending ST segment following the J wave. The appearance of J wave or ER is now recognized to predispose to the development of arrhythmogenesis when associated with other cardiac disorders, such as ischemia, heart failure, and hypothermia. The J wave might predict prognosis of cardiac events in various heart diseases, and the appearance of a new J wave during acute ischemia seems to be a messenger of VF [240], [241]. Family history of sudden death in subjects with ERP has been identified as a risk factor [189], [242]. The presence of coexisting Brugada ECG pattern (J waves in V 1−V 3) or short QT intervals in subjects with ER also suggests a more malignant nature [243], [244]. ER is commonly observed in the young, especially in fit and highly trained athletes, with a prevalence ranging up to 40%. In the majority of cases, the ensuing ST segment is rapidly ascending, suggesting that this is a benign ECG manifestation [237], [245]. Mahida et al. [246]. recently reported that electrophysiologic (EP) study using programmed stimulation protocols does not enhance risk stratification in ERS. 14 Brugada syndrome Numerous studies consistently show that clinical presentation is the strongest predictor of risk in BrS, overshadowing all other risk factors. The risk of recurrent VF among patients presenting with cardiac arrest is considerable: ≈35% at 4 years [247], [248], 44% at 7 years [249], and 48% at 10 years [250]. Fortunately, only a minority of patients with BrS (6% in Europe [247] but 18% in Japan [248]) diagnosed today have a history of cardiac arrest. Approximately one-third of contemporary BrS cohorts present with syncope [247]. Their risk of arrhythmic events during follow-up is intermediate: approximately 4 times higher than the risk of asymptomatic patients [49], [247], [251] but 4 times lower than that of patients diagnosed after cardiac arrest [247]. One explanation for this observation is that the syncope population consists of 2 different groups, one with arrhythmic syncope and bad prognosis, and a second with vagal syncope and good prognosis. Although a detailed clinical history may be of great value in differentiating between these 2 groups, it is not infallible [252]. In reviewing the records of 342 BrS patients, Olde Nordkamp et al. [60]. concluded that arrhythmic and nonarrhythmic syncope can be distinguished by clinical characteristics, including the absence of prodromes and specific triggers. Compared to patients with suspected nonarrhythmic syncope, patients presenting with presumed arrhythmic syncope were more likely to be male (RR 2.1) and to have urinary incontinence (RR 4.6) and were less likely to report prodromes. They also were older at first event (45 vs 20 years), and their syncope was never triggered by hot/crowded surroundings, pain or other emotional stress, sight of blood, or prolonged standing as in the case of nonarrhythmic syncope. During follow-up, all of the spontaneous arrhythmic events occurred in patients who originally presented with presumed arrhythmic syncope; patients with benign syncope had an excellent long-term prognosis. Today asymptomatic patients represent a majority (~63%) of newly diagnosed Brugada patients [247], [248]. Their risk of developing symptoms is relatively low (0.5% per year) [247], [248]. Unfortunately, for most the first symptom is cardiac arrest or SCD. Therefore, risk stratification of asymptomatic patients is of utmost importance, and strategies for doing so are discussed later. In cardiac arrest patients or patients with presumed arrhythmic syncope, these strategies are of little benefit because these patients are recognized to be at high risk. 14.1 Age and gender Mean age at the time of cardiac arrest in Brugada patients is 39–48 years, and the vast majority develop symptoms between 20 and 65 years of age [247], [248], [249]. Asymptomatic elderly patients with BrS are thought to be at relatively low risk for future cardiac events [253]. BrS in children is very rare, but sudden death in this population has been described [65], [254], [255]. As in adults, cardiac arrest survivors are at high risk for recurrence, but data regarding risk stratification of asymptomatic children are limited. In a large series, 64%–94% of patients with BrS who presented with cardiac arrest were male [247], [248], [250]. Males are also at increased risk for displaying a spontaneous type I Brugada ECG and for having inducible VF during EP studies [106]. Nevertheless, because the majority of asymptomatic patients are also male, gender is not an independent predictor of arrhythmic events [106], [248]. 14.2 Familial and genetic background Neither family history of SCD nor the presence of a mutation (of any type) in the SCN5A gene has consistently been demonstrated to be of value in risk stratification [247], [248], [249]. However, 1 study has shown that SCN5A mutations resulting in protein truncation do confer greater risk [256]. Certain rare variants and polymorphisms in SCN5A and in other genes have also been associated with prognosis [167], [257], [258], [259], [260], [261]. Nevertheless, the data are limited, and genetic testing is not generally used for risk stratification at this time. 14.3 Spontaneous vs drug-induced type I Brugada pattern A consistent finding in nearly all BrS series is that patients with spontaneous type I ECG at the time of diagnosis have a greater risk of arrhythmic events than do patients who develop such an ECG pattern only when challenged with a sodium channel blocker [247], [262]. This observation is true for asymptomatic patients [247] as well as for patients who present with syncope [49], [247], [251], [263] and remains an independent predictor of arrhythmic events in multivariate analysis. The problem is that only a minority of patients have a consistent spontaneous type 1 pattern when repeated ECGs are analyzed [264], [265]. Therefore, caution is needed when using a single ECG for risk stratification. Ventricular arrhythmias are rarely induced during a sodium channel block challenge [65], [262], [266], [267] and are likely to be dose dependent (i.e., will not occur when the infusion is stopped for safety reasons before the full dose is reached). Their long-term prognostic significance remains unclear. Extra caution should be exercised when administering sodium channel blockers to patients with significant conduction disease. Placing the right precordial leads over the 2nd and 3rd intercostal spaces in addition to the standard 4th intercostal space increases the sensitivity for detecting the coved ST- segment elevation of the type 1 Brugada pattern. Available data suggest that this increased detection of spontaneous type 1 pattern does not affect the value of this parameter for predicting VF [52], [268], [269]. 14.4 EP studies with programmed ventricular stimulation The role of programmed ventricular stimulation during EP studies continues to be passionately debated [270], [271], [272]. The question is not whether VF inducibility correlates with arrhythmic risk. In all series, the VF inducibility rate is highest for survivors of cardiac arrest, intermediate for those with syncope, and lowest for those who were asymptomatic at presentation [273]. The central question is whether the prognostic information provided by VF inducibility is robust enough for clinical decision-making. Some studies suggest that this is the case [262], [274], [275], but others do not [248], [249]. One central issue suggested as an explanation for the discrepancy between studies is the extrastimulation protocol used. Specifically, the prognostic impact of site (RV apex vs RV apex þ RVOT) [49], [270], [276] and the number of extrastimuli (2 vs 3) [49], [276] has been analyzed but with inconsistent results. The only prospective study specifically designed to examine the prognostic yield of EP studies in BrS was the PRELUDE registry [49]. It did not show that sustained VF induction identifies high-risk patients but did demonstrate that a short effective ventricular refractory period (o200 ms) is a risk marker. 14.5 Other risk markers Several ECG markers have been associated with risk in BrS. These include (1) fragmentation of the QRS [226], [277], [278], (2) the concomitant finding of a type 1 Brugada pattern and an ERP in inferolateral leads [145], [248], [279], [280], [281], and (3) dynamic changes in the manifestation of prominent J waves or ST-segment elevation [282]. Other markers associated with increased risk but with limited or inconsistent data include (1) late potentials recorded using signal-averaged ECG [283], [284], [285], (2) microscopic T-wave alternans [286], (3) macroscopic T-wave alternans during a sodium blocker challenge test [285], [287], [288], (4) increased QRS width [49], [248], [277], [289], [290], (5) prominent R wave in aVR [289], [290], [291], and (6) augmented ST-segment elevation of a type 1 Brugada pattern during the recovery phase of an exercise test [292]. Prolonged Tpeak-Tend [293], [294], [295], [296] and relatively steep QT/RR slope have been associated with higher risk in cases of BrS [297], [298]. Combining several risk factors (e.g. fragmented QRS þ ERP [277]) appears to confer an additive risk; however, data supporting this are limited. 15 Update on approaches to therapy for BrS and ERS Fig. 6, Fig. 7 graphically present recommendations for the management of BrS and ERS as modified from the 2013 HRS/EHRA/APHRS expert consensus statement on the diagnosis and management of patients with inherited primary arrhythmia syndromes and the 2015 ESC guidelines for the management of patients with ventricular arrhythmias and the prevention of SCD [8], [9]. Those recommendations are based on the available literature and on the clinical experience of the Task Force members. As with all such recommendations, they will need to undergo continuous validation in future studies. Education and lifestyle changes for the prevention of arrhythmias are critical in BrS. Patients should be informed of the various modulators and precipitating factors that could cause malignant arrhythmias. A prominent S-wave in Lead I has recently been identified as an ECG marker of BrS [298A]. Fever should be treated aggressively with antipyretics, and contraindicated substances should be avoided (see www.brugadadrugs.org) [89]. Referral for ECG is recommended during high fever. Family members may be referred for cardiopulmonary resuscitation training and advised to consider purchasing an automatic external defibrillator for home use. Because malignant ventricular arrhythmias are infrequent in asymptomatic patients with BrS [247] or ERP [44] and usually unrelated to physical activity, the presence of these patterns does not contraindicate participation in sports. It is noteworthy, however, that the Brugada pattern is accentuated immediately after exercise, presumably because of an increase in vagal tone [292], [299], [300]. In reviewing 98 case of BrS studies dealing with exercise, Masrur et al. [299]. concluded that there are insufficient data on the risks of exercise in BrS to make recommendations for exercise. 16 Implantable cardioverter-defibrillator The only proven effective therapeutic strategy for the prevention of SCD in high-risk BrS and ERS patients is an ICD [301], [302]. It is important to recognize that ICDs are associated with complications, especially in young active individuals [249], [303]. At 10 years postimplantation, the rates of inappropriate shock and lead failure are 37% and 29%, respectively. Remote monitoring can identify lead failure and prevent inappropriate shocks [304]. Subcutaneous ICDs are thought to represent the future for this indication because they are expected to be associated with fewer complications over a lifetime [305]. Implantation of an ICD is first-line therapy for JWS patients presenting with aborted SCD or documented VT/VF with or without syncope (Class I recommendation) [301], [306]. ICDs can be useful (Class IIa) in symptomatic BrS patients with type 1 pattern, in whom syncope was likely caused by VT/VF. The HRS/EHRA/APHRS expert consensus states that the ICD may be considered (Class IIb) in asymptomatic patients with inducible VF during programmed electrical stimulation (PES) [8]. Some studies suggest that the predictive value of EP studies may be improved by limiting the PES protocol to 2 extrastimuli [50], [276], but that observation is not supported by other studies [49], [307]. Similarly, some studies advocate that PES should be limited to the RVA and credit this limited PES strategy for a very high positive predictive value found is some series [275]. Again, that observation is not confirmed by other studies [276]. The current Task Force proposes that ICDs are reasonable (Class IIa) in symptomatic BrS patients with type 1 pattern but that implantation be considered on a case-by-case basis by an electrophysiologist experienced in BrS, taking into consideration age, gender, clinical presentation, ECG characteristics (QRS fragmentation, Jp amplitude), and patient preference. The current Task Force also proposes that EP study may be considered in asymptomatic individuals with spontaneous type 1 Brugada pattern. If VT/VF is inducible, an ICD should be considered [7]. More recent studies argue in favor of using ≤2 extrastimuli to induce VT/VF [50], [276]. ICDs are not indicated in asymptomatic patients without any of these characteristics. At present, there is no clear role for PES in patients with ERS. 17 Pacemaker therapy Arrhythmic events and SCD in both BrS and ERS generally occur during sleep or at rest and are associated with slow heart rates. These observations notwithstanding, a potential therapeutic role for cardiac pacing remains largely unexplored [308]. A few case reports are available [309], [310]. 18 RFA therapy Nademanee et al. [129]. showed that RFA of epicardial sites displaying late potentials and fractionated bipolar electrograms in the RVOT of BrS patients can significantly reduce arrhythmia vulnerability and the ECG manifestation of the disease. Ablation at these sites was reported to render VT/VF noninducible and to normalize the Brugada ECG pattern in the vast majority of patients over a period of weeks or months. Long-term follow-up (20–6 months) showed no recurrent VT/VF, with only 1 patient on medical therapy with amiodarone. Case reports in support of these effects have been published [311]. Additional evidence in support of the effectiveness of epicardial substrate ablation was provided by Sacher et al. [130]. and Shah et al. [312]. More recently, Brugada et al. [131]. used flecainide to identify the full extent of low-voltage electrogram activity in the anterior RV and RVOT and targeted this region for RFA. In all 14 BrS patients, RFA eliminated abnormal bipolar electrograms, normalized ST-segment elevation on right precordial leads of ECG, and VT/VF was no longer inducible. Ablation therapy can be lifesaving in otherwise uncontrollable cases. RF ablation may be considered (Class IIb recommendation) in BrS patients with frequent appropriate ICD shocks due to recurrent electrical storms [7]. There are no clinical reports of ablation of the LV substrate in patients with ERS. In patients in whom BrS combines with ERS, ablation of the anterior RV epicardium (including the RVOT) is not ameliorative. 19 Pharmacologic approach to therapy 19.1 Brugada syndrome ICD implantation may be problematic in infants or young children because of the high complication rate. ICDs are also economically out of reach for patients in some regions of the world. A pharmacologic approach to therapy, based on a rebalancing of currents active during the early phases of the epicardial AP in the RV so as to reduce the magnitude of the AP notch and/or restore the AP dome, has been a focus of basic and clinical research in recent years. Antiarrhythmic agents such as amiodarone and beta-blockers have been shown to be ineffective [313]. Class IC antiarrhythmic drugs (e.g., flecainide, propafenone) and Class IA agents (e.g., procainamide) are contraindicated because of their effects of unmasking BrS and inducing arrhythmogenesis. Disopyramide is a Class IA antiarrhythmic that has been demonstrated to normalize ST-segment elevation in some Brugada patients but to unmask the syndrome in others [314]. Because the presence of a prominent Ito is a prerequisite for the development of both BrS and ERS, partial inhibition of this current is thought to be effective regardless of the ionic or genetic basis for the disease. Unfortunately, cardioselective and Ito-specific blockers are not available. The only agent with significant Ito-blocking properties available in the United States and around the world is quinidine [19], [73]. Experimental studies have shown that quinidine is effective in restoring the epicardial AP dome, thus normalizing the ST segment and preventing phase 2 reentry and polymorphic VT in a variety of different experimental models of BrS [19], [150], [315], [316], [317]. A recent experimental study suggests that quinidine, because of its effect of blocking Ito, can also exert a protective effect against hypothermia- induced VT/VF in a JWS model [144]. It is noteworthy that, historically, quinidine was used to prevent VF in patients who required hypothermia for surgical procedures [317]. Clinical evidence for the effectiveness of quinidine in normalizing ST-segment elevation and/or preventing arrhythmic events in patients with BrS has been reported in numerous studies and case reports [117], [119], [120], [124], [318], [319], [320], [321], [322], [323], [324], [325], [326], [327], [328], [329], [330], [331]. Hermida et al. [119]. reported 76% efficacy in prevention of VF-induced by PES. Belhassen et al. [332]. recently reported a 90% efficacy in prevention of VF induction after treatment with quinidine despite the use of very aggressive protocols of extrastimulation. Furthermore, no arrhythmic events occurred among BrS patients treated with quinidine during a mean follow-up period of 10 years. In a recent trial conducted at 2 French centers, 44 asymptomatic BrS patients with inducible VT/VF were enrolled (47–10 years, 95% male) [333]. Of these patients, 34 (77%) were no longer inducible while treated with 600 mg/day hydroquinidine for 6.2–3 years. Among the 10 other patients (22%) who remained inducible and received ICD (group PVSþ), none received appropriate therapy during mean follow-up of 7.7–2 years. A prospective registry of empiric quinidine for asymptomatic BrS has been established. The study appears at the National Institutes of Health website (ClinicalTrials.gov) and can be accessed at http://clinicaltrials.gov/ct2/show/ NCT00789165?term_brugada&rank_2. Doses between 600 and 900 mg were recommended, if tolerated [322]. Quinidine may be considered (Class IIb indication) in BrS patients presenting with electrical storms and in patients implanted with an ICD who are experiencing repeated appropriate shocks. Quinidine can also be useful in asymptomatic BrS patients displaying a spontaneous type I ECG, if they qualify for an ICD and the device is refused or is contraindicated (Class IIa recommendation). Agents that augment the L-type calcium channel current, such as beta-adrenergic agents (e.g., isoproterenol, denopamine, orciprenaline) are useful as well [19], [117], [121], [327], [334], [335]. Isoproterenol, at times in combination with quinidine, has been used successfully to control VF storms and normalize ST elevation, particularly in children [93], [115], [116], [117], [118], [279], [319], [320], [325], [331], [336], [337], [338], [339], [340], [341], [342], [343]. Spontaneous VF in patients with BrS is often related to increases in vagal tone and is amenable to treatment by an increase of sympathetic tone via isopro- terenol administration. Administration of isoproterenol is a Class IIa recommendation for BrS patients presenting with electrical storms [7]. Another promising pharmacologic approach for BrS is the administration of the phosphodiesterase III inhibitor cilostazol [117], [121], [123], which normalizes the ST segment, most likely by augmenting calcium current (ICa) as well as by reducing Ito secondary to an increase in cAMP and heart rate [344]. Other effects of cilostazol may contribute to its actions (e.g., adenosine, NO, mitochondrial IKATP [345]). Its efficacy in combination with bepridil in preventing VF episodes was recently reported by Shinohara et al. [125]. The failure of cilostazol in the treatment of BrS has been described in a single case report [346]. Milrinone is another phosphodiesterase III inhibitor recently identified as a more potent alternative to cilostazol in suppressing ST elevation and arrhythmogenesis in an experimental model of BrS [150], [347]. No clinical reports have yet been published. Wenxin Keli, a traditional Chinese medicine, has recently been shown to inhibit Ito and thus to suppress polymorphic VT in experimental models of BrS when combined with low concentrations of quinidine (5 μM) [316]. Agents that augment peak and late INa, including bepridil and dimethyl lithospermate B, are suggested to be of value in BrS. Bepridil has been reported to suppress VT/VF in several studies of patients with BrS [117], [297], [298], [348]. The drug׳s action are thought to be mediated by (1) inhibition of Ito; (2) augmentation of INa via up-regulation of the sodium channels [349]; and (3) prolongation of QT interval at slow rates thus increasing the QT/RR slope [297], [298]. Dimethyl lithospermate B, an extract of Danshen, a traditional Chinese herbal remedy, has been reported to slow inactivation of INa, thus increasing INa during the early phases of the AP and suppressing arrhythmogenesis in experimental models of BrS [350]. Because malignant ventricular arrhythmias are infrequent in asymptomatic patients with BrS [247] or ERP [44] and usually unrelated to physical activity, the presence of these patterns does not contraindicate participation in sports, although, as previously discussed, insufficient data are currently available to make definitive recommendations for participation in sports. 19.2 Early repolarization syndrome It is not surprising that the approach to therapy of ERS is similar to that of BrS, because the mechanisms underlying the 2 syndromes are potentially similar. Quinidine, phosphodiesterase III inhibitors, and isoproterenol have all been shown to exert an ameliorative effect in preventing or quieting arrhythmias associated with ERS. Isoproterenol has been shown to be effective in quieting electrical storms developing in patients with either BrS [117], [338] or ERS [190]. Isoproterenol has been shown to act by reversing the repolarization abnormalities responsible for the disease phenotype secondary to restoration of the epicardial AP dome in experimental models of both BrS [19], [315] and ERS [30]. This action of the beta-adrenergic agonist is expected because of its actions to potently increase ICa. The phosphodiesterase III inhibitor cilostazol has been reported to reduce the ECG and arrhythmic manifestations of ERS [122]. Phosphodiesterase inhibitors are known to activate ICa secondary to an increase in cAMP [121], [344], [351], [352], [353], [354], [355]. The augmentation of ICa is thought to prevent arrhythmias associated with JWS by reversing the repolarization defects and restoring electrical homogeneity across the ventricular wall secondary to restoration of the epicardial AP dome in both BrS [347] and ERS [144]. Cilostazol has been hypothesized to also block Ito. Augmentation of ICa together with inhibition of Ito are expected to produce an inward shift in the balance of currents active during the early phases of the epicardial AP that should be especially effective in suppressing J-wave activity. The effectiveness of bepridil in ERS has been reported in a single patient thus far [356]. No clinical data are available regarding the effectiveness of RFA in the setting of ERS, despite the fact that low- voltage fractionated electrogram activity and high-frequency late potentials are observed in the LV of patients with ERS [357] and in experimental models of ERS (Yoon and Antzelevitch, unpublished data). Nakagawa et al. [357]. reported the results of a study in which they recorded epicardial electrograms directly from the LV of patients diagnosed with ERS by introducing a multipolar catheter into the left lateral (marginal) coronary vein, anterior interventricular vein, and middle cardiac vein via the coronary sinus. The authors reported late potentials in the bipolar electrograms recorded from the LV epicardium of the ERS patients [357].